Gray's Anatomy Special Introduction for the Web Welcome to the website for the 39th edition of Gray's Anatomy. I want this site to become an extension of the book - it is my intention that it should grow with the book. How the site develops will therefore depend partly on you and how you interact with it. One of the major aims of the website is to keep you informed, via PubMed abstracts, updates and commentaries, about what's new in anatomy - and that includes topographical anatomy, embryology, neuroanatomy and histology. Weekly PubMed links presented as Abstracts will offer you the capability to view papers on specific topics. Simply click on the link to be taken to PubMed where you can read the abstract. Updates will "flesh out" selected papers or topics. Commentaries are invited articles from experts in the field. I also want to encourage you to interact with the site. Through the Ask the Author option, you can ask me questions or suggest topics that you would like us to explore in the update section. Tell us about the variations in normal anatomy that you see in your clinics - we aim to establish a database that will support the text for the 40th edition. Air your views about the place of anatomy in a modern medical curriculum, the use of dissection as a teaching tool ... I look forward with great interest to hearing from you.
Copyright © 2008 Elsevier Inc. All rights reserved. Read our Terms and Conditions of Use and our Privacy Policy . For problems or suggestions concerning this service, please contact:
[email protected] Gray's Anatomy Preface When I was a student at Guy's Hospital Medical School, London, in the mid sixties, anatomy was regarded as one of the cornerstones of basic medical science. I remember using the 33rd edition of Gray's Anatomy , not as a course book, but as a source of additional information and detailed illustration: it became an old friend. I still have my copy and often refer to it - indeed it was used when preparing some of the illustrations in this new edition. Almost four decades later, anatomy occupies a less prominent place within an overcrowded undergraduate medical curriculum. Paradoxically, over the same period, the need for detailed anatomical knowledge at the postgraduate level has increased dramatically, fuelled particularly by developments in imaging and computer-assisted three-dimensional reconstruction (both macro- and microscopically); anaesthetics; endoscopic surgery, and the miniaturization of instruments. Anatomists and clinicians have learned to look with a fresh eye at familiar structures revealed in new ways, e.g. the arthroscopic appearance of joint cavities; high-resolution CT images of the petrous bone; images of the coronary circulation during MRI-guided cardiovascular catheter-based interventions; radionuclide imaging of the thyroid gland. As I write this, I have opened in front of me on my desk an English translation of Wilhelm Braune's Atlas of Topographical Anatomy, which was published by J and A Churchill in 1877: it contains detailed woodcuts of plane sections of frozen bodies, and displays a level of detail comparable with that seen in the best modern atlases of sectional anatomy. The following passage from the translator's preface to Braune's Atlas (published in Liepsig [sic] in 1874) speaks about the fundamental place of anatomy in medicine. By means of the sections found in this Atlas the exact position and relations of the structures which must be divided or avoided in the course of an operation are indicated: and the track of a bullet or puncture wound suggested. At the same time they afford an absolutely correct representation of the intimate relations of the viscera of the thorax and abdomen. For clinicians who work at the cutting edge (quite literally) of medicine, the message has not changed over the intervening years - anatomical knowledge remains an essential item in their armamentarium. The 39th edition of Gray's Anatomy is radically different from earlier editions because the body is described in regions rather than in systems. In an ideal world, an anatomical reference book should contain both systematic and regional anatomy. In the real world, the editorial team for the 39th edition decided that a book which would be of the greatest benefit to practising clinicians should mirror their daily practice and describe anatomy in the way in which they use it, i.e. regionally. Talking to colleagues around the world, this view has been the one that we have heard most frequently. We have responded to these comments by updating and clarifying the text, and have also paid particular attention to issues of navigability and clinical relevance. The members of the editorial team who have worked with me in preparing the 39th edition brought a wide range of experience as academic anatomists and clinicians: I am indebted to them all for their dedication and enthusiasm. The Lead Editors - Harold Ellis, Jeremiah Healy, David Johnson and Andrew Williams - have been responsible with me for overseeing the revision of specific parts of the book, initially by drawing together relevant material from the various sections of the 38th edition, and subsequently by guiding and advising the Editors of the sections and chapters. They also helped me to take strategic decisions about the overall content and organization of the 39th edition. Thus, for example, we have included descriptions of the blood supply to the skin and muscles, on the grounds that they have surgical relevance when raising flaps for reconstructive surgery, and we have made extensive use of new imaging modalities. The Theme Editors, Caroline Wigley and Pat Collins, worked closely with all members of the editorial team to update microstructure and embryology respectively throughout the book. The work of drilling down into the existing text, updating it and setting it in a clinical context, was undertaken by the Editors of the sections and chapters, a group of clinicians and anatomists (sometimes both) with a wealth of experience of teaching applied anatomy and neuroanatomy to medical and dental undergraduates and postgraduates. Editors and Specialist Contributors have provided new insights into topics such as the anatomy of the pelvic floor, inner ear, peritoneum, preimplantation embryology, assisted fertilization, spread of infection via fascial planes in the head and neck, smooth and cardiac muscle, wrist kinematics and kinetics, and the temporomandibular joint. Neuroanatomy has been comprehensively revised and now focuses on the human nervous system. The manuscript has been submitted to rigorous scrutiny by Specialist Reviewers (who commented on specific chapters), and by General Reviewers (who were able to comment on the text at first proof stage): their comments have been incorporated into the text. I am grateful to them all for their encouragement and suggestions. As far as possible, the orientation of diagrams and photographs throughout the book has been standardized to show the left side of the body, irrespective of whether a lateral or medial view is presented; transverse sections are viewed from below to facilitate comparison with clinical images. Clinicopathological examples have been selected where the pathology is either a direct result of, or a
consequence of, the anatomy, or where the anatomical features are instrumental in the diagnosis/treatment/management of the condition. Wherever possible, the new photomicrographs illustrate human histology and embryology. I recognize that re-orientating a great many of the illustrations from previous editions has created an enormous amount of work for the artwork staff, who were asked to reposition thousands of leader lines and their associated names. I am very grateful to Dr Michael Hutchinson for checking the accuracy of these changes. However, I have no doubt that eagle-eyed readers will alert me to any errors that may have slipped undetected through several iterations of proof reading of both text and figures. The Lead Editorial team took the view that Gray's Anatomy is not, and should not attempt to be, a source book for molecular biology, pathology, neuroscience, physiology or operative procedures. Moreover, space considerations mean that it is not an appropriate vehicle in which to rehearse experimental data, and preclude the level of descriptive detail that is found in highly specialized dedicated texts. Mindful of these caveats, reference lists have been provided at the end of all but a few very short chapters to guide further reading: the references have been annotated when the content of the paper or book is not evident from its title. A list of general texts and references covering material presented in more than one chapter, e.g. the distribution of angiosomes, or other basic medical sciences, is included on page xv. I offer my thanks to the production team at Elsevier, initially under the leadership of Richard Furn (19992002) and latterly of Inta Ozols (2003- 2004), for their guidance, professionalism, good humour and unfailing support. In what has truly been a team effort, it is difficult to single out individuals for especial mention. I would like to place on record my heartfelt thanks to Alison Whitehouse, Colin Arthur, Martin Mellor and Lesley Small, for being at the end of a phone or e-mail whenever I needed advice. I am grateful to Guy Standring, my long-suffering husband, for his patience and tolerance throughout the last four years.
August 2004 Copyright © 2008 Elsevier Inc. All rights reserved. Read our Terms and Conditions of Use and our Privacy Policy . For problems or suggestions concerning this service, please contact:
[email protected] Bibliography The following references contain information relevant to numerous chapters in this edition. They are therefore cited here rather than at the end of individual chapters. Terminology Basic Sciences Imaging and Radiology Clinical Clinical Examination Bibliography of 38th Edition
TERMINOLOGY
Federative Committee on Anatomical Terminology 1998 Terminologia Anatomica. International Anatomical Nomenclature. Stuttgart: Thieme. Dorland, 2003, Dorland's Illustrated Medical Dictionary, 30th edn. Philadelphia: W B Saunders. BASIC SCIENCES
Abrahams PH, Marks SC Jr, Hutchings R 2002 McMinn's Colour Atlas of Human Anatomy, 5th edn. London: Churchill Livingstone. Alberts B, Johnson A, Lewis J, Raff M, Roberts K, Walter P 2002 Molecular Biology of the Cell, 4th edn. New York: Garland Science Publishing. Berkovitz BKB, Kirsh C, Moxham BJ, Alusi G, Cheeseman T 2002 Interactive Head and Neck. London: Primal Pictures. Boron W, Boulpaep E 2002 Medical Physiology. Philadelphia: W B Saunders. Crossman AR, Neary D 2000 Neuroanatomy, 2nd edn. Edinburgh: Churchill Livingstone. Fitzgerald MJT, Folan-Curran J 2001 Clinical Neuroanatomy and Related Science, 4th edn. Edinburgh: Churchill Livingstone. Guyton AC, Hall JE 1996 Human Physiology and Mechanisms of Disease, 6th edn. Philadelphia: W B Saunders. Kerr JB 1999 Atlas of Functional Histology. London: Mosby. Kierszenbaum AL 2002 Histology and Cell Biology: An Introduction to Pathology. St Louis: Mosby. Moore KL, Persaud TVN 2003 Before We Are Born: Essentials of Embryology and Birth Defects, 6th edn. Philadelphia: W B Saunders. Pollard TD, Earnshaw WC 2002 Cell Biology. Philadelphia: W B Saunders. Roitt I, Brostoff J, Male D 2001 Immunology, 6th edn. London: Mosby. Salmon M 1994 Anatomic Studies: Book 1 Arteries of the Muscles of the Extremities and the Trunk, Book 2 Arterial Anastomotic Pathways of the Extremities. Ed. by Taylor G, Razaboni RM. St Louis: Quality Medical Publishing Inc. Stevens A, Lowe JS 1996 Human Histology, 2nd edn. London: Mosby. Young B, Heath JW 2000 Wheater's Functional Histology: A Text and Colour Atlas. Edinburgh: Churchill Livingstone. Back to Top IMAGING AND RADIOLOGY - RADIOLOGICAL ANATOMY
Butler P, Mitchell AWM, Ellis H 1999 Applied Radiological Anatomy. New York: Cambridge University Press. Ellis H, Dixon A, Logan BM 1999 Human Sectional Anatomy: Atlas of Body Sections, CT and MRI Images, 2nd edn. Oxford: Oxford University Press. Haaga JR, Lanzieri CF, Gilkeson RC 2002 CT and MR Imaging of the Whole Body, 4th edn. St Louis: Mosby. Lasjaunias P, Berenstein A, ter Brugge K 2001 Surgical Neuroangiography, vol 1. Clinical Vascular Anatomy and Variations, 2nd edn. Berlin, New York: Springer. Meyers MA 1994 Dynamic Radiology of the Abdomen: Normal and Pathologic Anatomy, 4th edn. New York: Springer.
Pomeranz SJ 1992 MRI Total Body Atlas. Cincinnati: MRI-EFI Publications. Sutton D 2002 Textbook of Radiology and Imaging, 7th edn. Edinburgh: Churchill Livingstone. Weir J, Abrahams PH 2003 Imaging Atlas of Human Anatomy, 3rd edn. London: Mosby. Wicke L 1998 Atlas of Radiologic Anatomy, 6th edn. Baltimore: Williams and Wilkins. Whaites E 2002 Essential of Dental Radiography and Radiology, 3rd edn. Edinburgh: Churchill Livingstone. Back to Top CLINICAL
Birch R, Bonney G, Wynn Parry CB 1998 Surgical Disorders of the Peripheral Nerves. Edinburgh: Churchill Livingstone. Bogduk N 1997 Clinical Anatomy of the Lumbar Spine and Sacrum, 3rd edn. Edinburgh: Churchill Livingstone. Borges AF 1984 Relaxed skin tension lines (RSTL) versus other skin lines. Plast Reconstr Surg 73: 144-50. Burnand K, Young A, Rowlands B, Scholefield J 2004 The New Aird's Companion in Surgical Studies, 3rd edn. Edinburgh: Churchill Livingstone. Canale ST 2003 Campbell's Operative Orthopaedics, 10th edn. Philadelphia: Mosby. Cormack GC, Lamberty BGH 1994 The Arterial Anatomy of Skin Flaps. Edinburgh: Churchill Livingstone. Dyck PJ, Thomas PK 2004 Peripheral Neuropathy, 4th edn. Philadelphia: W B Saunders. Ellis H 2002 Clinical Anatomy: A Revision and Applied Anatomy for Clinical Students, 10th edn. Maldem MA: Blackwell Scientific Publications. Ellis H, Feldman S, Harrop-Griffiths W 2003 Anatomy for Anaesthetists, 8th edn. Oxford: Blackwell Science. Rosai J 2004 Rosai and Ackerman's Surgical Pathology, 2-volume set with CD-ROM, 9th edn. London: Mosby Shah J 2003 Head and Neck Surgery and Oncology, 3rd edn. Edinburgh; New York: Mosby Taylor GI, Palmer JH, McManamy D 1987 The vascular territories of the body (angiosomes): experimental study and their clinical applications. In McCarthy JG, 1990, Plastic Surgery, vol I General Principles. Philadelphia: W B Saunders. Zancolli E A, Cozzi E P 1992 Atlas of Surgical Anatomy of the Hand, Edinburgh: Churchill Livingstone. Back to Top CLINICAL EXAMINATION
Aids to the Examination of the Peripheral Nervous System, 4th Edition. London: W B Saunders, 2000. Lumley JSP 2002 Surface Anatomy: The Anatomical Basis of Clinical Examination, 3rd edn. Edinburgh: Churchill Livingstone. Back to Top Copyright © 2008 Elsevier Inc. All rights reserved. Read our Terms and Conditions of Use and our Privacy Policy . For problems or suggestions concerning this service, please contact:
[email protected] Gray's Anatomy Special Introduction for the Web Welcome to the website for the 39th edition of Gray's Anatomy. I want this site to become an extension of the book - it is my intention that it should grow with the book. How the site develops will therefore depend partly on you and how you interact with it. One of the major aims of the website is to keep you informed, via PubMed abstracts, updates and commentaries, about what's new in anatomy - and that includes topographical anatomy, embryology, neuroanatomy and histology. Weekly PubMed links presented as Abstracts will offer you the capability to view papers on specific topics. Simply click on the link to be taken to PubMed where you can read the abstract. Updates will "flesh out" selected papers or topics. Commentaries are invited articles from experts in the field. I also want to encourage you to interact with the site. Through the Ask the Author option, you can ask me questions or suggest topics that you would like us to explore in the update section. Tell us about the variations in normal anatomy that you see in your clinics - we aim to establish a database that will support the text for the 40th edition. Air your views about the place of anatomy in a modern medical curriculum, the use of dissection as a teaching tool ... I look forward with great interest to hearing from you.
Copyright © 2008 Elsevier Inc. All rights reserved. Read our Terms and Conditions of Use and our Privacy Policy . For problems or suggestions concerning this service, please contact:
[email protected] Editors EDITOR-IN-CHIEF
Susan Standring PhD DSc Professor of Experimental Neurobiology and Head, Division of Anatomy, Cell and Human Biology, Guy's, King's and St Thomas' School of Biomedical Sciences, King's College London, London, UK LEAD EDITORS
Harold Ellis CBE MCh FRCS Emeritus Professor of Surgery of the former Westminster Hospital Medical School, Clinical Anatomist, Division of Anatomy, Cell and Human Biology, Guy's, King's and St Thomas' School of Biomedical Sciences, King's College London, London, UK Jeremiah C Healy MA MB BChir MRCP FRCR Consultant Radiologist, Chelsea and Westminster Hospital, London, UK David Johnson MA BM BCh DM FRCS(Eng) Specialist Registrar in Plastic and Reconstructive Surgery, Department of Plastic and Reconstructive Surgery, Radcliffe Infirmary, Oxford, UK Andrew Williams MB BS FRCS FRCS(Orth) Consultant Orthopaedic Surgeon, Trauma and Orthopaedics Department, Chelsea and Westminster Hospital, London, UK THEME EDITORS
Patricia Collins PhD Associate Professor of Anatomy, Anglo-European College of Chiropractic, Bournemouth, UK Caroline Wigley BSc PhD Honorary University Teaching Fellow, The Peninsula Medical School, Exeter, UK EDITORS
Barry KB Berkovitz BDS FDSRCS(Eng) MSc PhD (Chapters 25, 27, 29-33, 35, 36 & 38) Reader, Division of Anatomy, Cell and Human Biology, Guy's, King's and St Thomas' School of Biomedical Sciences, King's College London, London, UK Neil R Borley FRCS FRCS(Ed) MS (Chapters 66-89 & 108) Consultant Colorectal Surgeon, Department of Gastrointestinal Surgery, Cheltenham General Hospital, Gloucestershire Hospitals NHS Trust, Cheltenham, UK Alan R Crossman BSc PhD DSc (Chapters 12, 13, 15-19 & 21-24) Professor of Anatomy, The University of Manchester, Manchester, UK Mark S Davies MB BS FRCS FRCS(Orth) (Chapter 115) Consultant Orthopaedic Surgeon, Guy's and St Thomas' Hospitals, and The London Foot and Ankle Centre, The Hospital of St John and St Elizabeth, London, UK MJ Turlough FitzGerald MD, PhD, DSc, MRIA (Chapter 20) Emeritus Professor of Anatomy, National University of Ireland, Galway, Ireland Jonathan Glass MB BS FRCS(Urol) (Chapters 91-101) Consultant Urologist, Guy's and St Thomas' Hospitals, London, UK Carole M Hackney PhD (Chapter 39) Professor of Auditory Neuroscience, MacKay Institute of Communication and Neuroscience, School of Life Sciences, Keele University, Keele, UK Thomas Ind MD MRCOG (Chapters 102-107) Consultant Gynaecological Surgeon, Royal Marsden and St George's Hospitals, London, UK Anthony R Mundy MS FRCP FRCS (Chapters 91-101 & 108) Professor of Urology, Institute of Urology and Nephrology, London, UK Richard LM Newell BSc MB BS FRCS (Chapters 44-46 & 110-114) Clinical Anatomist, Cardiff School of Biosciences, Cardiff University, Wales, and Honorary Consultant Orthopaedic Surgeon, Royal Devon and Exeter Healthcare NHS Trust, Exeter, UK The late Gordon L Ruskell PhD
(Chapters 41 & 42) Emeritus Professor, Department of Optometry and Visual Science, City University, London, UK Pallav Shah MD FRCP (Chapters 56-60 & 62-64) Consultant Physician, Royal Brompton Hospital, London, and Department of Respiratory Medicine, Chelsea and Westminster Hospital, London, UK Copyright © 2008 Elsevier Inc. All rights reserved. Read our Terms and Conditions of Use and our Privacy Policy . For problems or suggestions concerning this service, please contact:
[email protected] Copyright Notices ELSEVIER CHURCHILL LIVINGSTONE © Elsevier Ltd 2005 No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without either the prior permission of the publishers or a licence permitting restricted copying in the United Kingdom issued by the Copyright Licensing Agency, 90 Tottenham Court Road, London W1T 4LP. Permissions may be sought directly from Elsevier's Health Sciences Rights Department in Philadelphia, USA: phone: (+1) 215 238 7869, fax: (+1) 215 238 2239, e-mail:
[email protected]. You may also complete your request on-line via the Elsevier homepage (http://www.elsevier.com), by selecting 'Customer Support' and then 'Obtaining Permissions'. First edition JW Parker & Son 1858 Second edition 1860 Third to eleventh editions, Longman, Green & Co. 1864-1887 Twenty-fifth edition Longman, Green & Co. 1932 Twenty-sixth edition Longman, Green & Co. 1935 Twenty-seventh edition Longman, Green and Co. 1938 Twenty-eighth edition Longman, Green & Co. 1942 Twenty-ninth edition Longman, Green & Co. 1947 Thirtieth edition Longman, Green & Co. 1949 Thirty-first edition Longman, Green & Co. 1954 Thirty-second (centenary) edition Longman, Green & Co. 1958 Thirty-third edition Longman 1964 Thirty-fourth edition Longman 1967 Thirty-fifth edition Longman 1973 Thirty-sixth edition Churchill Livingstone 1980 Thirty-seventh edition Churchill Livingstone 1989 Thirty-eighth edition Churchill Livingstone 1999 ISBN: Main Edition 0 443 07168 3 International Student Edition 0 443 07169 1 e-dition™ 0 443 06676 0 e-dition™ online access only 0 443 06675 2 British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library Library of Congress Cataloging in Publication Data A catalog record for this book is available from the Library of Congress Note No responsibility is assumed by the Publishers, Editors or Contributors for any loss or injury and/or damage to persons or property arising out of or related to any use of the material contained in this book. It is the responsibility of the treating practitioner, relying upon independent expertise and knowledge of the patient, to determine the best treatment and method of application for the patient.
Every effort has been made to contact holders of copyright to obtain permission to reproduce copyright material. However, if any have been inadvertently overlooked, the Publisher will be pleased to make the necessary arrangements at the first opportunity. The Publisher
For Elsevier Commissioning Editors: Inta Ozols, Richard Furn Senior Project Development Manager: Alison Whitehouse Project Development Managers: Martin Mellor, with Jim Killgore and Gus Gomes Head of Project Management: Colin Arthur Project Manager: Lesley W Small Illustration Manager: Bruce Hogarth Illustrators (39th edition): Robert Britton, Graeme Chambers, Michael Courtney, Peter Cox, Ethan Danielson, Brian Evans, Sandie Hill, Bruce Hogarth, Gillian Lee, Gillian Oliver, Richard Tibbitts (Antbits), Philip Wilson Illustrators (recent editions): Andrew Bezear, Marks Creative, Patrick Elliot, Jenny Halstead, Dr AA van Horssen, Peter Jack, Peter Lamb, Paul Richardson, Lesley J Skeates, Denise Smith Photographers: Sarah-Jane Smith (38th and 39th editions), Kevin Fitzpatrick (38th edition) Designers: Sarah Russell, Keith Kail Copyeditors: Carolyn Holleyman, with Lewis Derrick, Sue Lowry and Ruth Swan Proofreaders: Isobel Black, Christian Simpson, Ian Ross Index: Merrall-Ross International Ltd Printed in Spain Copyright © 2008 Elsevier Inc. All rights reserved. Read our Terms and Conditions of Use and our Privacy Policy . For problems or suggestions concerning this service, please contact:
[email protected] Acknowledgements We are indebted to Dr Michael Hutchinson for checking the labelling and placement of labels in anatomical illustrations across the book. NEW PHOTOGRAPHY COMMISSIONED FOR THIS EDITION We thank Sarah-Jane Smith for the following photos Microstructure: 3.4, 3.10, 3.15, 3.17, 3.19, 5.2, 5.4A- F, 5.6, 5.8, 5.12, 5.13, 5.14, 5.18, 6.6, 6.15, 6.17, 6.19A, 6.20, 6.28, 6.29, 6.37, 6.42, 6.48, 7.15, 7.17, 8.4, 8.8, 8.11, 8.13, 8.14, 8.16, 8.17, 9.1, 9.2, 9.3, 17.1, 21.14, 21.15, 31.31, 33.40, 33.41, 58.3, 71.16A, 71.17, 72.1, 75.3, 75.4, 85.16, 86.5, 88.6, 88.7, 89.4, 89.6, 96.3, 104.6. Osteology: 45.11, 45.17, 45.18, 45.19, 45.21, 45.22, 45.23, 45.24, 45.26, 45.31, 45.32, 45.33A, 45.34, 49.1A,B, 49.3, 49.4, 49.5, 49.6, 49.8A,B, 49.9A,B, 49.10A,B, 49.11, 52.5, 52.6, 52.7, 52.8, 52.12, 57.4A,B, 57.7, 57.8, 57.9A,B, 57.10A,B, 57.11A,B, 111.4A,B, 111.5A,B, 111.6, 111.15A,B, 111.16A,B, 111.17, 111.18, 111.19, 113.5, 113.6, 113.13, 114.1A,B, 114.2A,B, 114.5. Surface anatomy: 25.1, 25.2, 25.3, 25.4, 25.5, 25.6, 25.7, 25.8, 44.1, 44.2, 44.3, 48.14, 48.15, 48.16AC, 48.17, 56.3, 110.12, 110.13, 110.14, 110.15, 110.16, 110.17, 110.18, 110.19, 110.20, 110.21, 110.22, 110.24, 110.25, 110.26, 110.27. Diagrammatic overlays for figures: 25.1, 25.2, 25.3, 25.4, 25.5, 25.6, 25.7, 25.8, 110.12, 110.13, 110.14, 110.15, 110.16, 110.17, 110.18, 110.20, 110.21, 110.22, 110.24. PREVIOUSLY PUBLISHED ILLUSTRATIONS Within individual figure captions, we have acknowledged all figures kindly loaned from other sources. However, we would particularly like to thank the following authors who have generously loaned so many figures from other books published by Elsevier: Microstructure: Kerr JB 1999 Atlas of Functional Histology. London: Mosby. Figures 3.8, 3.9, 3.14, 4.23, 6.11, 7.5, 7.18, 7.21, 33.12B, 58.2B, 59.23, 71.18, 72.5, 85.14, 85.15, 87.7, 88.5, 91.17, 97.5A, 102.8, 103.5. Kierszenbaum AL 2002 Histology and Cell Biology: An Introduction to Pathology. St Louis: Mosby. Figures 3.6, 3.11, 4.16, 6.25, 8.3, 8.5, 32.5, 33.42, 59.17, 62.2, 62.3, 71.15, 96.4A,B, 97.7, 104.5. Stevens A, Lowe JS 1996 Human Histology, 2nd edn. London: Mosby. Figures 2.9, 2.17, 4.17, 31.30, 32.6, 62.5, 91.22, 92.4, 102.6, 102.9. Young B, Heath JW 2000 Wheater's Functional Histology: A Text and Colour Atlas, 4th edn. Edinburgh: Churchill Livingstone Figures 2.4, 2.5B, 2.12, 2.15, 2.16A,B, 2.18B, 2.24, 3.2, 3.3, 3.13, 3.16, 4.1, 4.10B,C, 5.11, 5.17, 6.41, 7.10, 7.20, 33.10, 33.11, 35.7A, 42.15, 58.5, 62.7, 62.8, 62.9, 72.3, 85.18, 91.19, 91.20, 97.4, 102.10. Head and neck: Berkovitz BKB, Holland GR, Moxham BJ 2002 Oral Anatomy, Embryology and Histology, 3rd edn. Edinburgh: Mosby. Figures 33.2, 33.12A, 33.18B- E, 33.25, 33.26, 33.28, 33.29, 33.30, 33.32, 33.34, 34.10D, 35.12, Table 33.1. Berkovitz BKB, Moxham BJ 1994 Color Atlas of the Skull. London: Mosby-Wolfe. Figures 27.1, 27.3, 27.4, 27.5, 27.7, 27.8, 27.9, 27.10, 27.11, 27.13, 27.14, 27.16, 27.17, 27.19, 27.20, 27.21, 27.22, 27.23, 27.24, 27.25, 27.27, 27.28, 32.1, 32.2, 33.16. Surface anatomy: Lumley JSP 2002 Surface Anatomy: the Anatomical Basis of Clinical Examination, 2nd edn. Edinburgh: Churchill Livingstone. Figures 27.2, 27.6, 41.15, 41.16, 48.13A,B, 48.18B, 56.1, 56.2, 56.4, 66.1, 66.2, 66.3, 66.4, 110.23. Diagrammatic overlays for figures 25.1, 25.2, 25.3, 25.4, 25.5, 25.6, 25.7, 25.8, 110.12, 110.13, 110.14, 110.15, 110.16, 110.17, 110.18, 110.20, 110.21, 110.22, 110.24. NEW ILLUSTRATIONS COMMISSIONED FOR THIS EDITION We thank the illustrators for their valuable contribution to the new edition Antbits: 1.1, 16.3, 16.11, 19.16, 19.18, 41.13, 59.1. All of surface anatomy and bones overlays. Robert Britton: 2.1, 2.3, 2.5A, 2.6, 2.11, 2.18A, 2.22, 2.23, 3.1, 3.12, 4.11, 4.12, 4.27, 6.39, 7.3, 8.5,
8.12, 8.20, 8.22, 12.40, 21.3, 30.8, 30.11, 33.39, 39.22, 42.17, 42.19, 42.23, 59.18, 62.1, 62.6, 65.5, 71.14, 85.17, 87.6, 87.8, 102.5, 105.8, 110.29, 110.30, 110.31. Graeme Chambers: 53.32. Michael Courtney: 17.6, 17.7, 17.13, 17.15, 17.16, 45.50, 45.51, 45.52, 45.53, 45.55, 45.56, 45.57, 48.5, 49.16, 49.18, 49.21, 49.23, 49.24, 50.1, 50.2, 50.3, 50.4, 51.1, 51.9, 51.11, 51.12, 52.1, 52.2, 52.3, 52.9, 52.15, 52.16, 52.17, 52.18, 52.19, 52.20, 52.21, 52.22, 52.23, 53.2, 53.3, 53.4, 53.12, 53.34, 53.36, 53.37, 53.38, 53.40, 53.42, 53.43, 53.51, 53.52, 53.53, 53.55, 53.56, 53.58, 53.59, 63.5, 63.13, 64.3, 113.19, 115.1, 115.7, 115.17, 115.37, 115.42. Peter Cox: 13.3, 13.15, 15.6, 15.10, 15.12, 16.2, 23.4, 23.5, 23.9, 23.12, 23.15, 23.16, 24.4, 24.5. Ethan Danielson: 4.19, 6.49, 10.22, 10.23, 10.24, 11.1, 11.4, 11.5, 11.6B, 12.2, 14.5, 14.15, 14.26, 14.27, 14.28, 15.3, 18.10, 18.11, 18.12, 19.1, 19.17, 20.4, 20.6, 20.7, 20.8, 20.9, 20.10, 20.11, 20.12, 20.13, 20.14, 20.15, 22.30, 23.14, 26.6, 29.10, 31.18, 34.5, 34.8, 38.11, 45.1, 45.2, 45.5, 46.12, 47.6, 47.8, 47.9, 47.10, 48.1, 48.4, 49.27, 49.29, 54.2, 54.3, 57.1, 57.2, 61.3, 65.1, 72.4, 85.3, 89.5, 90.3, 90.5, 90.6, 91.12, 91.15, 97.6, 105.3, 109.1, 109.2, 109.3, 109.4, 109.5, 109.6, 109.7, 109.11, 109.12, 109.13, 109.14, 109.15, 109.16, 109.18, 109.19, 109.20, 110.4, 110.5, 110.9. Brian Evans: 15.1, 15.3, 29.3, 29.15, 31.1, 31.4, 31.9, 31.26, 33.7, 41.4, 41.6, 76.1, 76.10, 76.11, 76.14, 76.16, 76.17, 82.1, 82.2, 83.8, 83.9, 108.2, 108.3, 108.9, 108.11, 108.12. Sandie Hill: 35.2, 35.11A, 36.1, 36.2, 36.3, 36.4, 36.6, 36.7, 36.8, 36.9, 36.10, 36.11, 36.13, 36.14, 36.15, 67.4, 67.5B, 67.6B, 67.12, 67.13, 67.15, 67.16, 71.2, 71.11A,B, 73.1, 84.1, 84.3, 84.5B, 84.6A, 84.7, 85.1, 85.2, 85.3, 85.5, 85.12, 86.1, 87.2, 87.3, 87.4, 87.8. Bruce Hogarth: 8.28, 16.5,16.6, 16.9, 18.23, 19.13, 19.14, 19.23, 22.1, 22.3, 22.6, 29.2, 44.4, 45.3, 45.4, 48.6, 48.8, 53.34, 71.1, 71.3, 71.5, 71.8, 71.10, 71.12, 110.3, 110.28. Gillian Lee: 6.33, 6.34, 6.35, 6.36, 31.2, 31.16, 32.3, 32.8, 33.43, 41.3, 41.22, 60.17, 63.1, 63.11, 63.34C, 69.2, 69.7, 92.2. Gillian Oliver: 71.4. Philip Wilson: 30.10, 44.6, 44.8, 45.5, 45.6, 45.10, 45.13, 45.14, 45.38, 45.41, 45.43, 45.44, 46.8, 46.11, 46.13, 49.9, 57.21, 58.4, 67.7, 68.1, 68.3, 68.4, 68.8, 68.12, 68.15, 76.15, 80.1, 92.3, 111.21. COVER ILLUSTRATION An oblique pcoronal shaded volume-rendition slab image of the head and neck superimposed on a photograph of a young male subject, posterior view. The image was derived from a coronally acquired T1-weighted volumetric magnetic resonance dataset: it was produced using a Voxar 3D workstation (Voxar Ltd, Edinburgh, Scotland). Image supplied by Dr RJS Chinn, Consultant Radiologist, Chelsea and Westminster Hospital, London, UK. Copyright © 2008 Elsevier Inc. All rights reserved. Read our Terms and Conditions of Use and our Privacy Policy . For problems or suggestions concerning this service, please contact:
[email protected] SECTION 1 INTRODUCTION AND SYSTEMIC OVERVIEW Susan Standring (Lead Editor) Caroline Wigley (Microstructure) Patricia Collins (Embryology, Growth and Development) Andrew Williams (Biomechanics) Critical reviewers: Peter Braude (chapter 10), Robert Brooks (2 & 3), Fred Cody (4 & 6), Mike Hall (11), Jonathan C Kentish (7), Birgit Lane (2 & 8), Susan Pickering (10), Mary Ritter (5), Anthea Rowlerson (4 & 6), Jeremy PT Ward (7) page 1 page 1 page 2
© 2008 Elsevier
1 INTRODUCTION Anatomical nomenclature Anatomy is the study of the structure of the body, from the submicroscopic to the macroscopic. It is conventionally divided into topographical or gross anatomy (including surface, endoscopic and radiological anatomy), histology, embryology and neuroanatomy. Anatomical language is one of the fundamental languages of medicine. The unambiguous description of thousands of structures is impossible without an extensive and often highly specialized vocabulary. Ideally, these terms, which are often derived from Latin or Greek, should be used to the exclusion of any other, throughout the world. In reality, many terms are vernacularized. The Terminologia Anatomica, drawn up by the Federative Committee on Anatomical Terminology (FCAT) in 1998, has served as our guide in preparing the 39th Edition of Gray's Anatomy. Where we have anglicized some of the Latin terms, we have given the official form, at least once, in parentheses. We have also included eponyms, since these are often used, possibly more so by clinicians than anatomists. Indeed, certain eponyms are so firmly entrenched in the language of the clinician that to avoid them could lead to confusion: the eponymous term is often the only way to describe a particular structure, because there is no simple alternative anatomical term.
© 2008 Elsevier
PLANES, DIRECTIONS AND RELATIONSHIPS To avoid ambiguity, all anatomical descriptions assume that the body is in the conventional 'anatomical position', i.e. standing erect and facing forwards, upper limbs by the side with the palms facing forwards, and lower limbs together with the feet facing forwards (Fig. 1.1). Descriptions are based on four imaginary planes, median, sagittal, coronal and horizontal, applied to a body in the anatomical position. The median plane passes longitudinally through the body and divides it into right and left halves. The sagittal plane is parallel to the median plane. The coronal plane is orthogonal to the median plane and divides the body into anterior (front) and posterior (back). The horizontal (transverse) plane is orthogonal to both median and sagittal planes. Radiologists refer to transverse planes as (trans)axial. Convention dictates that axial anatomy is viewed as though looking from the feet towards the head. Structures nearer the head are superior; cranial may be used when talking about the head. Structures closer to the feet are inferior; caudal is frequently used in embryology to refer to the tail end of the embryo. Medial and lateral indicate closeness to the midline, medial being nearer to the midline than lateral. External (outer) and internal (inner) refer to the distance from the centre of an organ or cavity, e.g. the layers of the body wall. Various degrees of obliquity are acknowledged using compound terms, e.g. posterolateral. When referring to structures in the trunk and upper limb we have used freely the synonyms anterior, ventral, flexor, palmar, volar, and posterior, dorsal and extensor. We recognize that these synonyms are not always satisfactory, e.g. the extensor aspect of the leg is anterior with respect to the knee and ankle joints, and superior in the foot and digits; the plantar (flexor) aspect of the foot is inferior. Dorsal (dorsum) and ventral are terms used particularly by embryologists and neuroanatomists: they therefore feature most often in Sections 1 and 2. Distal and proximal are used to describe structures in the limbs, taking the datum point as the attachment of the limb to the trunk, such that a proximal structure is closer to the attachment of the limb than a distal structure; they are also used in describing branching structures, e.g. bronchi. Superficial and deep are used to describe the relationships between adjacent structures. Ipsilateral refers to the same side (of the body, organ or structure), bilateral to both sides, and contralateral to the opposite side.
© 2008 Elsevier
MOVEMENTS Movements at joints, e.g. flexion, extension, adduction and abduction, and the many possible combinations of these 'pure' movements, are described in Chapter 6. Specialized movements, such as those that occur at the elbow, wrist, ankle and temporomandibular joints, are described in the appropriate chapters. page 3 page 4
Figure 1.1 The terminology widely used in descriptive anatomy. Abbreviations shown on arrows: AD, adduction; AB, abduction; FLEX, flexion (of the thigh at the hip joint); EXT, extension (of the leg at the knee joint).
© 2008 Elsevier
2 Basic structure and function of cells CELL STRUCTURE GENERAL CHARACTERISTICS OF CELLS Most cells lie within the size range 5-50 µm in diameter: e.g. resting lymphocytes are c.6 µm across, red blood cells c.7.5 µm and columnar epithelial cells are c.20 µm tall and 10 µm wide. Some cells are much larger than this: e.g. megakaryocytes of the bone marrow are more than 200 µm in diameter. Large neurones and skeletal muscle cells have relatively enormous volumes because of their extended shapes, some of the former being over 1 metre in length. Cell size is limited by rates of diffusion, either that of material entering or leaving cells, or of diffusion within them. Diffusion can be much accelerated by processes of active transport across membranes and also directed by transport mechanisms within the cell. Motility is a characteristic of most cells, in the form of movements of cytoplasm or specific organelles from one part of the cell to another. It also includes: the extension of parts of the cell surface such as pseudopodia, membrane ruffles, filopodia and microvilli; locomotion of entire cells as in the amoeboid migration of tissue macrophages; the beating of flagella or cilia to move the cell (e.g. in spermatozoa) or fluids overlying it (e.g. in respiratory epithelium); cell division and muscle contraction. Cell movements are also involved in the uptake of materials from their environment (endocytosis, phagocytosis) and the passage of large molecular complexes out of cells (exocytosis, secretion). The shapes of cells vary widely depending on their interactions with each other, their extracellular environment and internal structures. Their surfaces are often highly folded when absorptive or transport functions take place across their boundaries. According to the location of absorptive or transport functions, apical microvilli or basolateral infoldings create a large surface area for transport or diffusion. Cells rarely operate independently of each other and commonly form aggregates by adhesion, often assisted by specialized intercellular junctions. They may also communicate with each other either by releasing and detecting molecular signals that diffuse across intercellular spaces, or more rapidly by membrane contact, which often involves small, transient, transmembrane channels. Cohesive groups of cells constitute tissues and more complex assemblies of tissues form functional systems or organs.
CELLULAR ORGANIZATION Cytoplasm is contained within a limiting plasma membrane. All cells except mature red blood cells also contain a nucleus that is surrounded by a nuclear membrane (Figs 2.1, 2.2). The nucleus includes the genome of the cell contained within the chromosomes, and the nucleolus. The cytoplasm contains several systems of organelles. These include a series of membrane-bound structures that form separate compartments within the cytoplasm, such as rough and smooth endoplasmic reticulum, Golgi apparatus, lysosomes, peroxisomes, mitochondria and vesicles for transport, secretion and storage of cellular components. There are also structures that lie free in the non-membranous, cytosolic compartment. They include ribosomes and several filamentous protein structures known collectively as the cytoskeleton. The cytoskeleton determines general cell shape and supports specialized extensions of the cell surface (microvilli, cilia, flagella). It is involved in the assembly of new filamentous organelles (e.g. centrioles) and controls internal movements of the cytoplasm and cytoplasmic vesicles. The cytosol contains many soluble proteins, ions and metabolites.
Cell domains
Figure 2.1 The main structural components and internal organization of a generalized cell.
In polarized cells, particularly in epithelia, the cell is generally subdivided into domains that reflect the polarization of activities within the cell. The free surface, e.g. that facing the intestinal lumen or airway, is the apical surface, and its adjacent cytoplasm is the apical cell domain. This is where the cell interfaces with a specific body compartment (or, in the case of the epidermis, with the outside world). The apical surface is specialized to act as a barrier, restricting access of substances from this compartment to the rest of the body. Specific components are selectively absorbed from, or added to, the external compartment by the active processes, respectively, of active transport and endocytosis inwardly or exocytosis and secretion outwardly. The surface of the cell opposite to the apical surface is the basal surface, with its associated basal cell domain. In a single-layered epithelium, this surface is apposed to the basal lamina. The remaining surfaces are known as the lateral cell surfaces. In many instances the lateral and basal surfaces perform similar functions and the cellular domain is termed the basolateral domain. Cells actively transport substances, such as digested nutrients from the intestinal lumen or endocrine secretions, across their basal (or basolateral) surfaces into the subjacent connective tissue matrix and the blood capillaries within it. Dissolved non-polar gases (oxygen and carbon dioxide) diffuse freely between the cell and the bloodstream across the basolateral surface. page 5
page 6
Figure 2.2 A protein-synthesizing cell (immunoglobulin-secreting plasma cell) in connective tissue. The main ultrastructural features are a nucleus surrounded by a nuclear envelope and containing peripheral heterochromatin, central euchromatin and a nucleolus; and cytoplasm containing rough endoplasmic reticulum, mitochondria, Golgi apparatus, transport vesicles and small lysosomes.
Plasma membrane Cells are bounded by a distinct plasma membrane, which shares features with the system of internal membranes that compartmentalize the cytoplasm and surround the nucleus. They are all composed of lipids (mainly phospholipids, cholesterol and glycolipids) and proteins, in approximately equal ratios. Plasma membrane lipids form a layer two molecules thick, the lipid bilayer. The hydrophobic ends of each lipid molecule face the interior of the membrane and the hydrophilic ends face outwards. Most proteins are embedded within, or float in, the lipid bilayer as a fluid mosaic. Some proteins, because of extensive hydrophobic regions of their polypeptide chains, span the entire width of the membrane (transmembrane proteins), whereas others are only superficially attached to the bilayer by lipid groups. Both are integral (intrinsic) membrane proteins, as distinct from peripheral (extrinsic) membrane proteins, which are membrane-bound only through their association with other proteins. Carbohydrates in the form of oligosaccharides and polysaccharides are bound either to proteins (glycoproteins) or to lipids (glycolipids), and project mainly into the extracellular domain. Combinations of biochemical, biophysical and biological techniques have revealed that lipids are not homogenously distributed in membranes, but that some are organized into microdomains in the bilayer, called 'detergent-resistant membranes' or lipid 'rafts', rich in sphingomyelin and cholesterol (Morris et al 2003). The ability of select subsets of proteins to partition into different lipid microdomains has profound effects on their function, e.g. in T-cell receptor and neurotrophin signalling. The highly organized environment of the domains provides a signalling, trafficking and membrane fusion environment very different from that found in the disorganized fluid mosaic membrane.
Figure 2.3 The molecular organization of the plasma membrane, according to the fluid mosaic model of membrane structure. Intrinsic or integral membrane proteins include diffusion or transport channel complexes, receptor proteins and adhesion molecules. These may span the thickness of the membrane (transmembrane proteins) and can have both extracellular and cytoplasmic domains. Transmembrane proteins have hydrophobic zones, which cross the phospholipid bilayer and allow the protein to 'float' in the plane of the membrane. Some proteins are restricted in their freedom of movement where their cytoplasmic domains are tethered to the cytoskeleton.
page 6
page 7
Figure 2.4 The plasma membrane covering microvilli on absorptive epithelial cells in the small intestine. The lipid bilayer is clearly seen at this high magnification, as is the cell coat or glycocalyx projecting into extracellular space (the gut lumen, right) as an outer fuzzy layer. (By permission from Young B, Heath JW 2000 Wheater's Functional Histology. Edinburgh: Churchill Livingstone.)
In the electron microscope, membranes fixed and contrasted by heavy metals such as osmium appear in section as two densely stained layers separated by an electron-translucent zone - the classic unit membrane (Figs 2.3, 2.4). The total thickness is c.5 nm. Freeze-fracture cleavage planes usually pass along the midline of each membrane, where the hydrophobic tails of phospholipids meet. This technique has also demonstrated intramembranous particles embedded in the lipid bilayer; these are in the 5-15 nm range and in most cases represent large transmembrane protein molecules or complexes of molecules. Intramembranous particles are distributed asymmetrically between the two halfmembranes, usually adhering more to one face than to the other. In plasma membranes, the inner or protoplasmic (cytoplasmic) half-membrane carries most particles, seen on its surface facing the exterior (the P face). Where they have been identified, particles usually represent channels for the transmembrane passage of ions or molecules. Biophysical measurements show the lipid bilayer to be highly fluid, allowing diffusion in the plane of the membrane. Thus proteins are able to move freely in such planes unless anchored from within the cell. Membranes in general, and the plasma membrane in particular, form boundaries selectively limiting diffusion and creating physiologically distinct compartments. Lipid bilayers are impermeable to hydrophilic solutes and ions and so membranes actively control the passage of ions and small organic molecules such as nutrients, through the activity of membrane transport proteins. However, lipid-soluble substances can pass directly through the membrane so that, for example, steroid hormones enter the cytoplasm freely. Their receptor proteins are either cytosolic or nuclear, rather than being located on the cell surface. Plasma membranes are able to generate electrochemical gradients and potential differences by selective ion transport, and actively take up or export small molecules by energy dependent processes. They also provide surfaces for the attachment of enzymes, sites for the receptors of external signals, including hormones and other ligands, and sites for the recognition and attachment of other cells. Internally, plasma membranes can act as points of attachment for intracellular structures, in particular those concerned with motility and other cytoskeletal functions. Cell membranes are synthesized by the rough endoplasmic reticulum in conjunction with the Golgi apparatus.
THE CELL COAT (GLYCOCALYX) The plasma membrane differs structurally from internal membranes in that it possesses an external, diffuse, carbohydrate-rich coat, the cell coat or glycocalyx. The cell coat forms an integral part of the plasma membrane, projecting as a diffusely filamentous layer 2-20 nm or more from the lipoprotein surface (Fig. 2.4). The overall thickness of the plasma membrane is therefore variable, but is typically 8-10 nm. The cell coat is composed of the carbohydrate portions of glycoproteins and glycolipids embedded in the plasma membrane (Fig. 2.3). The precise composition of the glycocalyx varies with cell type: many tissue and cell type-specific antigens are located in the coat, including the major histocompatibility antigen systems and, in the case of erythrocytes, blood group antigens. It also contains adhesion molecules, which enable cells to adhere selectively to other cells or to the extracellular matrix. They have important roles in maintaining the integrity of tissues and in a wide range of dynamic cellular
processes, e.g. the formation of intercommunicating neural networks in the developing nervous system and the extravasation of leukocytes. Cells tend to repel each other because of the predominance of negatively charged carbohydrates at cell surfaces. There is consequently a distance of at least 20 nm between the plasma membranes of adjacent cells, other than at specialized junctions.
CELL SURFACE CONTACTS The plasma membrane is the surface which establishes contact with other cells and with structural components of extracellular matrices. These contacts may have a predominantly adhesive role, or initiate instructive signals within and between cells, or both; they frequently affect the behaviour of cells. Structurally, there are two main classes of contact, both associated with cell adhesion molecules. One class is associated with specializations at discrete regions of the cell surface that are ultrastructurally distinct. These are described on page 7. The second, general, class of adhesive contact has no obvious associated ultrastructural features.
GENERAL ADHESIVE CONTACTS One class of transmembrane or membrane-anchored glycoproteins that project externally from the plasma membrane, and which form adhesive contacts, are the cell adhesion molecules. There are a number of molecular subgroups, which are broadly divisible on the basis of their calcium dependence. Calcium-dependent adhesion molecules
Cadherins, selectins and integrins are calcium-dependent adhesion molecules. Cadherins are transmembrane proteins, with five heavily glycosylated external domains. They are responsible for strong general intercellular adhesion, as well as being components of some specialized adhesive contacts, and are attached by linker proteins (catenins) at their cytoplasmic ends to underlying cytoskeletal fibres (either actin or intermediate filaments). Different cell types possess different members of the cadherin family, e.g. N-cadherins in nervous tissue, E-cadherins in epithelia, and P-cadherins in the placenta. These molecules bind to those of the same type in other cells (homophilic binding), so that cells of the same class adhere to each other preferentially, forming tissue aggregates or layers, as in epithelia. Selectins are found on leukocytes, platelets and vascular endothelial cells. They are transmembrane lectin glycoproteins that can bind with low affinity to the carbohydrate groups on other cell surfaces to permit movement between the two, e.g. the rolling adhesion of leukocytes on the walls of blood vessels (p. 146). They function cooperatively in sequence with integrins, which strengthen the selectin adhesion. Integrins are glycoproteins that typically mediate adhesion between cells and extracellular matrix components such as fibronectin, collagen, laminin. They integrate interactions between the matrix and the cell cytoskeleton to which they are linked, and so facilitate cell migration within the matrix. An integrin molecule is formed of two subunits (! and "), each of which has several subtypes. Combinations of alternative subunits provide more than 20 integrin heterodimers, each one directed to a particular extracellular molecule, although there is considerable overlap in specificity. Some integrins depend for their binding on magnesium, rather than calcium. Calcium-independent adhesion molecules
The best known calcium-independent adhesion molecules are glycoproteins that have external domains related to immunoglobulin molecules. Most are transmembrane proteins. Some are entirely external, either attached to the
plasma membrane by a glycosylphosphatidylinositol anchor, or secreted as soluble components of the extracellular matrix. Different types are expressed in different tissues. Neural cell adhesion molecules are found on a number of cell types, but are expressed widely by neural cells. Intercellular adhesion molecules are expressed on vascular endothelial cells. Cell adhesion molecule binding is predominantly homophilic, although some, e.g. intercellular adhesion molecules, use a heterophilic mechanism and can bind to integrins. For further information on all aspects of cell adhesion molecules and intercellular contacts, see Alberts et al (2002).
SPECIALIZED ADHESIVE CONTACTS Specialized adhesive contacts, some of which mediate activities other than simple mechanical cohesion, are localized regions of the cell surface with particular ultrastructural characteristics. Three major classes exist: occluding, adhesive and communicating junctions (Fig. 2.5). Occluding junctions (tight junctions, zonula occludens)
Occluding junctions create diffusion barriers in continuous layers of cells, including epithelia, mesothelia and endothelia, and prevent the passage of materials across the cellular layer through intercellular spaces. They form a continuous belt (zonula) around the cell perimeter, near the apical surface in cuboidal or columnar epithelial cells. At a tight junction, the membranes of the adjacent cells come into contact, so that the gap between them is obliterated. Freeze-fracture electron microscopy shows that the contacts between the membranes lie along branching and anastomosing ridges formed by the incorporation of chains of intramembranous protein particles on the P face of the lipid bilayer (Fig. 2.5C). page 7 page 8
Figure 2.5 Intercellular junctions. A, The position of the apical junctional complex of epithelial cells (top) and the structures of the three elements of a junctional complex and of a gap junction (below). B, The ultrastructural appearance of an epithelial junctional complex; C, a freeze-fractured preparation showing the anastomotic network of contacts between adjacent cell membranes forming a tight junction; D, a freeze-fractured preparation showing the structure of a gap junction, with numerous channels (pores within connexons) clustered to form a plaque-like junctional region between adjacent plasma membranes; E, hemidesmosomes in the basal layer of the epidermis, contacting the underlying basal lamina. Dermal collagen fibrils are sectioned transversely below. (B, by permission from Young B, Heath JW 2000 Wheater's Functional Histology. Edinburgh: Churchill Livingstone; C, by kind permission from Dr Andrew Kent, King's College London; D, by kind permission from Professor Dieter Hülser, University of Stuttgart; E, by permission from the Company of Biologists Ltd, Cambridge, UK, and Frye M, Gardner C, Li ER, Arnold I, Watt FM 2003 Evidence that Myc activation depletes the epidermal stem cell compartment by modulating adhesive interactions with the local microenvironment. Development 130: 2793-2808.)
This arrangement ensures that substances can only pass through the layer of cells by diffusion or transport through their apical membranes and cytoplasm. The cells thus selectively modify the environment on either side of the layer. Occluding junctions also create regional differences in the plasma membranes of the cells in which they are found. For example, in epithelia, the composition of the apical plasma membrane differs from that of the basolateral regions, and this allows these regions to engage in functions such as directional transport of ions and uptake of macromolecules. Because tight junctions have high concentrations of fixed transmembrane proteins, they act as barriers to lateral diffusion of lipid and protein within membranes. The integrity of tight junctions is calcium-dependent. Cells can transiently alter the permeability of their tight junctions to increase passive paracellular transport in some circumstances. Adhesive junctions
Adhesive junctions include intercellular and cell-extracellular matrix contacts, where cells adhere strongly to each other or to adjacent matrix components. Intercellular contacts can be subdivided according to the extent and location of the contact. They all display a high concentration of cell adhesion molecules, which externally bind adjacent cells, and internally link to the cytoskeleton via intermediary proteins. Zonula adherens (intermediate junction) page 8 page 9
A zonula adherens is a continuous, belt-like zone of adhesion around the apical perimeters of epithelial, mesothelial and endothelial cells, parallel and just basal to the tight junction in epithelia. High concentrations of cadherins occur in this zone; their cytoplasmic ends are anchored via the proteins vinculin and !-actinin
to a layer of actin microfilaments. These junctions help to reinforce the intercellular attachment of the tight junction and prevent its mechanical disruption. The gap between cell surfaces is c.20 nm. Usually, no electron-dense material is observed within this intercellular space. Fascia adherens
A fascia adherens is similar to a zonula adherens, but is more limited in extent and forms a strip or patch of adhesion, e.g. between smooth muscle cells, in the intercalated discs of cardiac muscle cells (p. 152) and between glial cells and neurones. The junctions involve cadherins attached indirectly to actin filaments on the inner side of the membrane. Desmosomes (macula adherens)
Desmosomes are limited, plaque-like areas of particularly strong intercellular contact. They can be located anywhere on the cell surface. In epithelial cells, there may be a circumferential row of desmosomes parallel to the tight and intermediate junctional zones, an arrangement that forms the third, most basally situated, component of the epithelial apical junctional complex (Fig. 2.5A,B and p. 29). The intercellular gap is c.25 nm, is filled with electron-dense filamentous material running transversely across it and is also marked by a series of densely staining bands running parallel to the cell surfaces. Adhesion is mediated by calcium-dependent cadherins, desmoglein and desmocollin. Within the cells on either side, a cytoplasmic density underlies the plasma membrane and includes the anchor proteins desmoplakin and plakoglobin, into which the ends of intermediate filaments are inserted. The type of intermediate filament depends on cell type, e.g. cytokeratins are found in epithelia and desmin filaments in cardiac muscle cells. Desmosomes form strong anchorage points, likened to spot-welds, between cells subject to mechanical stress, e.g. in the spinous layer of the epidermis, where they are extremely numerous and large (p. 157). Hemidesmosomes
Hemidesmosomes are best known as anchoring junctions between the bases of epithelial cells and the basal lamina. Ultrastructurally, they resemble a singlesided desmosome, anchored on one side to the plasma membrane, and on the other to the basal lamina and adjacent collagen fibrils (Fig. 2.5E). On the cytoplasmic side of the membrane there is a dense coat into which cytokeratin filaments are inserted. Hemidesmosomes use integrins as their adhesion molecules, whereas desmosomes use cadherins. Less highly structured attachments with a similar arrangement exist between many other cell types and their surrounding matrix, e.g. between smooth muscle cells and their matrix fibrils, and between the ends of skeletal muscle cells and tendon fibres. The smaller, punctate adhesions resemble focal adhesion plaques. Focal adhesion plaques
Focal adhesion plaques are regions of local attachment between cells and the extracellular matrix. They are typically situated at or near the ends of actin filament bundles (stress fibres), which are anchored through intermediary proteins to the cytoplasmic domains of integrins. In turn, these are attached at their external ends to collagen or other filamentous structures in the extracellular matrix. They are usually short-lived: their formation and subsequent disruption are part of the motile behaviour of migratory cells. Gap junctions (communicating junctions)
Gap junctions resemble tight junctions in transverse section, but the two apposed lipid bilayers are separated by an apparent gap of 3 nm which is bridged by numerous transmembrane channels (connexons). Connexons are formed by a ring of six connexin proteins in each membrane. Their external surfaces meet
those of the adjacent cell in the middle. A minute central pore links one cell to the next (Fig. 2.5A). These channels may exist in small numbers, and this makes them difficult to detect structurally. However, they lower the transcellular electrical resistance and so can be detected by microelectrodes. Larger assemblies of many thousands of channels are often packed in hexagonal arrays (Fig. 2.5D). Such junctions form limited attachment plaques rather than continuous zones, and so allow free passage of substances within the adjacent intercellular space, unlike tight junctions. They occur in numerous tissues including the liver, epidermis, pancreatic islet cells, connective tissues, cardiac muscle and smooth muscle, and are also common in embryonic tissues. In the central nervous system, they are found in the ependyma and between neuroglial cells, and they form electrical synapses between some types of neurone, although this is rare in humans. Although gap junctions form diffusion channels between cells, the size of their apertures limits diffusion to small molecules and ions (up to a molecular weight of about 1000 kDa). Thus they admit sodium, potassium and calcium ions, various second messenger components, and a number of metabolites, but they exclude messenger RNA and other macromolecules. In some excitable tissues (e.g. cardiac and smooth muscle), one cell can activate another electrically by current flow through gap junctions. Communicating junctions probably permit metabolic cooperation between groups of adjacent cells; the significance of this activity in embryogenesis, normal tissue function, homeostasis and repair is only beginning to be understood. Other types of junction
Chemical synapses and neuromuscular junctions are specialized areas of intercellular adhesion where neurotransmitters secreted from a neuronal terminal gain access to specialized receptor molecules on a recipient cell surface. They are described on pages 44 and 64, respectively.
CELL SIGNALLING Cellular systems in the body communicate with each other to coordinate and integrate their functions. This occurs through a variety of processes known collectively as cell signalling, in which a signalling molecule produced by one cell is detected by another, almost always by means of a specific receptor protein molecule. The recipient cell transduces the signal, which it most usually detects at the plasma membrane, into intracellular chemical messages that change cell behaviour. The signal may act over a long distance, as in endocrine signalling through the release of hormones into the bloodstream or neuronal synaptic signalling via electrical impulse transmission along axons (p. 44) and subsequent release of chemical transmitters of the signal at synapses (p. 44) or neuromuscular junctions (p. 64). A specialized variation of endocrine signalling (neurocrine signalling) occurs when neurones or paraneurones (e.g. chromaffin cells of the suprarenal medulla) secrete a hormone into the bloodstream. Alternatively, signalling may occur at short range through a paracrine mechanism, in which cells of one type release molecules into the interstitial fluid of the local environment, to be detected by nearby cells of a different type that express the specific receptor protein. Cells may generate and respond to the same signal. This is autocrine signalling, a phenomenon that reinforces the coordinated activities of a group of like cells, which respond together to a high concentration of a local signalling molecule. The most extreme form of short-distance signalling is contact-dependent signalling, where one cell responds to transmembrane proteins of an adjacent cell that bind to surface receptors in the responding cell membrane. This type of signalling is important during development and in immune
responses. These different types of signalling mechanism are illustrated in Fig. 2.6. For further reading, see Alberts et al (2002) and Pollard and Earnshaw (2002).
SIGNALLING MOLECULES AND THEIR RECEPTORS page 9 page 10
Figure 2.6 The different modes of cell-cell signalling.
The majority of signalling molecules (ligands) are hydrophilic. They cannot cross the plasma membrane of a recipient cell to effect changes intracellularly unless they first bind to a plasma membrane receptor protein. Ligands are mainly proteins, polypeptides or highly charged biogenic amines. They include: classic peptide hormones of the endocrine system (Ch. 9); cytokines, which are mainly of haemopoietic cell origin and involved in inflammatory responses and tissue remodelling, e.g. the interferons, interleukins, tumour necrosis factor, leukaemia inhibitory factor; polypeptide growth factors, e.g. the epidermal growth factor superfamily, nerve growth factor, platelet-derived growth factor, the fibroblast growth factor family, transforming growth factor beta and the insulin-like growth factors. Polypeptide growth factors are multifunctional molecules with more widespread actions and cellular sources than their names suggest. They and their receptors are commonly mutated or aberrantly expressed in certain cancers. The cancer-causing gene variant is termed a transforming oncogene and the normal (wild-type) version of the gene is a cellular oncogene or proto-oncogene. The activated receptor acts as a transducer to generate intracellular signals, which are either small diffusible second messengers (e.g. calcium, cyclic adenosine monophosphate or the plasma membrane lipid-soluble diacylglycerol), or larger protein complexes that amplify and relay the signal to target control systems. For further reading on growth factors and other signalling molecules, see Epstein (2003). Some signals are hydrophobic and able to cross the plasma membrane freely. Classic examples are the steroid hormones, thyroid hormones, retinoids and vitamin D. Steroids, for instance, enter cells non-selectively, but elicit a specific response only in those target cells which express specific cytoplasmic or nuclear receptors. Light stimuli also cross the plasma membranes of photoreceptor cells and interact intracellularly, at least in rod cells, with membrane-bound photosensitive receptor proteins. Hydrophobic ligands are transported in the bloodstream or interstitial fluids, generally bound to carrier proteins, and they often have a longer half-life and longer-lasting effects on their targets than do water-soluble ligands.
A separate group of signalling molecules that are able to cross the plasma membrane freely is typified by the gas, nitric oxide . The principal target of shortrange nitric oxide signalling is smooth muscle, which relaxes in response. Nitric oxide is released from vascular endothelium as a result of the action of autonomic nerves that supply the vessel wall. It causes local relaxation of smooth muscle and dilation of vessels. In the penis, this mechanism is responsible for penile erection. Nitric oxide is unusual among signalling molecules in having no specific receptor protein; instead, it acts directly on intracellular enzymes of the response pathway. Receptor proteins
There are c.20 different families of receptor proteins, each with several isoforms responding to different ligands. The great majority of these receptors are transmembrane proteins. Members of each family share structural features that indicate either shared ligand-binding characteristics in the extracellular domain or shared signal transduction properties in the cytoplasmic domain, or both. There is little relationship either between the nature of a ligand and the family of receptor proteins to which it binds and activates, or the signal transduction strategies by which an intracellular response is achieved. The same ligand may activate fundamentally different types of receptor in different cell types. Cell surface receptor proteins are generally grouped according to their linkage to one of three intracellular systems: ion channel-linked receptors; G-proteincoupled receptors; receptors that link to enzyme systems. Other receptors do not fit neatly into any of these categories. All the known G-protein-coupled receptors belong to a structural group of proteins that pass through the membrane seven times in a series of serpentine loops. These receptors are thus known as sevenpass transmembrane receptors or, because the transmembrane regions are formed from !-helical domains, as seven-helix receptors. The most well-known of this large group of phylogenetically ancient receptors are the odorant-binding proteins of the olfactory system, the light-sensitive receptor protein, rhodopsin, and many of the receptors for clinically useful drugs. A comprehensive list of receptor proteins, their activating ligands and examples of the resultant biological function, is given in Pollard and Earnshaw (2002).
TRANSPORT ACROSS CELL MEMBRANES Lipid bilayers are increasingly impermeable to molecules as they increase in size or hydrophilicity. Transport mechanisms are therefore required to carry essential polar molecules, including ions, nutrients, nucleotides and metabolites of various kinds, across the plasma membrane and into or out of membrane-bound intracellular compartments. Transport is facilitated by a variety of membrane transport proteins, each with specificity for a particular class of molecule, e.g. sugars. Transport proteins fall mainly into two major classes, channel proteins and carrier proteins. page 10 page 11
Channel proteins form aqueous pores in the membrane, which open and close under the regulation of intracellular signals, e.g. G-proteins, to allow the flux of solutes (usually inorganic ions) of specific size and charge. Transport through ion channels is always passive and ion flow through an open channel depends only on the ion concentration gradient and its electronic charge, and the potential difference across the membrane. These factors combine to produce an electrochemical gradient, which governs ion flux. Channel proteins are utilized most effectively by the excitable plasma membranes of nerve cells, where the resting membrane potential can change transiently from about -70 mV (negative inside the cell) to +50 mV (positive inside the cell) when stimulated by a neurotransmitter (as a result of the opening and subsequent closure of channels selectively permeable to sodium and potassium).
Carrier proteins bind their specific solutes, such as amino-acids, and transport them across the membrane through a series of conformational changes. This latter process is slower than ion transport through membrane channels. Transport by carrier proteins can occur either passively by simple diffusion, or actively against the electrochemical gradient of the solute. Active transport must therefore be coupled to a source of energy, such as ATP generation, or energy released by the coordinate movement of an ion down its electrochemical gradient. Linked transport can be in the same direction as the solute, in which case the carrier protein is described as a symporter, or in the opposite direction, when the carrier acts as an antiporter.
TRANSLOCATION OF PROTEINS ACROSS INTRACELLULAR MEMBRANES Proteins are generally synthesized on ribosomes in the cytosol or on the rough endoplasmic reticulum. A few are made on mitochondrial ribosomes. Once synthesized, many proteins remain in the cytosol, where they carry out their functions. Others, such as integral membrane proteins or proteins for secretion, are translocated across intracellular membranes for post-translational modification and targeting to their destinations. This is achieved by the signal sequence, an addressing system contained within the protein sequence of amino-acids, which is recognized by receptors or translocators in the appropriate membrane. Proteins are thus sorted by their signal sequence (or set of sequences that become spatially grouped as a signal patch when the protein folds into its tertiary configuration), so that they are recognized by and enter the correct intracellular membrane compartment.
EXOCYTOSIS AND ENDOCYTOSIS Secreted proteins, lipids, mucins, small molecules such as amines and other cellular products destined for export from the cell are transported to the plasma membrane in small vesicles released from the trans face of the Golgi apparatus. This pathway is either constitutive, in which transport and secretion occur more or less continuously, or it is regulated by external signals, as in the control of salivary secretion by autonomic neural stimulation. In regulated secretion, the secretory product is stored temporarily in membrane-bound secretory granules or vesicles. Exocytosis is achieved by fusion of the secretory vesicular membrane with the plasma membrane and release of the vesicle contents into the extracellular domain. In polarized cells, e.g. most epithelia, exocytosis occurs at the apical plasma membrane and the cells secrete into a duct lumen or onto a free surface such as the lining of the stomach. In hepatocytes, bile is secreted across a very small area of plasma membrane forming the wall of the bile canaliculus (p. 1222). This region is defined as the apical plasma membrane, and is the site of exocrine secretion (p. 34), whereas secretion of hepatocyte plasma proteins into the bloodstream is targeted to the basolateral surfaces facing the sinusoids. Packaging of different secretory products into appropriate vesicles takes place in the trans-Golgi network. Delivery of secretory vesicles to their correct plasma membrane domains is achieved by sorting sequences in the cytoplasmic tails of vesicular membrane proteins.
Figure 2.7 Transcytotic vesicles (arrows) shuttle in both directions between blood plasma and extracellular fluid, across the cytoplasm of these endothelial cells in the wall of a capillary.
There are other mechanisms in which initial delivery of secretory products is less selective, but is followed by selective retention (or degradation) or reprocessing and redistribution by endosomes. Ultimately, secretory vesicles undergo docking, priming (to prepare the vesicle for a regulatory signal, where secretion is regulation-dependent) and fusion with the plasma membrane to release their contents. The process of exocytosis also delivers integral membrane components to the cell surface in the normal turnover and recycling of the plasma membrane. However, excess plasma membrane generated by vesicle fusion during exocytosis is rapidly removed by concurrent endocytosis. The process of endocytosis involves the internalization of vesicles derived from the plasma membrane. The vesicles may contain: engulfed fluids and solutes from the extracellular interstitial fluid (pinocytosis); larger macromolecules, often bound to surface receptors (receptor-mediated endocytosis); particulate matter, including microorganisms or cellular debris (phagocytosis). Pinocytosis generally involves small fluid-filled vesicles and is a marked property of capillary endothelium, e.g. where vesicles containing nutrients and oxygen dissolved in blood plasma are transported from the vascular lumen to the endothelial basal plasma membrane (Fig. 2.7). Interstitial fluid containing dissolved carbon dioxide is also taken up by pinocytosis for simultaneous transportation across the endothelial cell wall in the opposite direction, for release into the bloodstream by exocytosis. This shuttling of pinocytotic vesicles is also termed transcytosis. Larger volumes of fluid are engulfed by dendritic cells, e.g. in the process of sampling interstitial fluids by macropinocytosis in immune surveillance for antigens (p. 81). Interstitial fluid is inevitably taken up during receptor-mediated endocytosis when ligands are internalized. Receptor-mediated endocytosis, also known as clathrin-dependent endocytosis, is initiated at specialized regions of the plasma membrane known as clathrin-coated pits. Clathrin is a protein that cross-links adjacent adaptor protein (adaptin)
complexes to form a basket-like structure, bending the membrane inwards into a hemisphere. Much, but not all, fluid-phase pinocytosis also utilizes clathrin-coated pits. Ligands such as the iron-transporting protein, transferrin, and the cholesterol-transporting low-density lipoprotein bind to their receptors, which cluster in clathrin-coated pits through an interaction with adaptins. The pits then invaginate and pinch off from the plasma membrane, internalizing both receptor and ligand. The processing of endocytic vesicles and their contents is described on page 14. For further details of the molecular mechanisms of endocytosis, see Alberts et al (2002) or Pollard and Earnshaw (2002).
PHAGOCYTOSIS page 11 page 12
Phagocytosis is a property of many cell types, but is most efficient in cells specialized for this activity. The professional phagocytes of the body belong to the monocyte lineage of haemopoietic cells, in particular the tissue macrophages (p. 80). Other effective phagocytes are neutrophil granulocytes and most dendritic cells (p. 81), which are also of haemopoietic origin. Phagocytosis plays an important part in the immune defence system of the body, in which the amoeboid process of ingestion of organisms for nutrition has evolved into a mechanism for the clearance of microorganisms invading the body. Macrophages also ingest particulate material including inorganic matter, such as inhaled dust particles, in addition to debris from dead cells and protein aggregates such as immune complexes in the blood, airways, interstitial spaces and connective tissue matrices. Phagocytosis is a triggered process, initiated when a phagocytic cell binds to a particle or organism, often through a process of molecular recognition. Typically, a pathogenic microorganism may first be coated by antibodies, which are bound in turn by receptors for the Fc portion of the antibody molecule expressed by macrophages and neutrophils; in this way the microorganism is attached to the cell. This triggers the production of large pseudopodia, which engulf the organism and internalize it, as their pseudopod tips fuse together. The process appears to depend on actin-myosin-based cellular motility and, unlike receptor-mediated endocytosis, it is energy dependent. Phagosomes thus formed are as large as the body they engulf and can be a considerable proportion of the volume of the phagocytic cell. Inside the cell, the phagosome fuses with lysosomes, which degrade its contents.
Cytoplasm ENDOPLASMIC RETICULUM Endoplasmic reticulum is a system of interconnecting membrane-lined channels within the cytoplasm (Fig. 2.8). These channels take various forms, including cisternae (flattened sacs), tubules and vesicles. The membranes divide the cytoplasm into two major compartments. The intramembranous compartment includes the space where secretory products are stored or transported to the Golgi complex and cell exterior. The extramembranous cytosol is made up of the colloidal proteins such as enzymes, carbohydrates and small protein molecules, together with ribosomes and ribonucleic acids, and elements of the cytoskeleton. Structurally, the channel system can be divided into rough or granular endoplasmic reticulum, which has ribosomes attached to its outer cytosolic surface, and smooth or agranular endoplasmic reticulum, which lacks ribosomes.
ROUGH ENDOPLASMIC RETICULUM The rough endoplasmic reticulum, studded with ribosomes, is a site of protein synthesis (Fig. 2.8A). Most proteins pass through its membranes and accumulate within its cisternae, although some integral membrane proteins, e.g. plasma
membrane receptors, are inserted into the rough endoplasmic reticulum membrane, where they remain. After passage from the rough endoplasmic reticulum, proteins remain in membrane-bound cytoplasmic organelles such as lysosomes, become incorporated into new plasma membrane, or are secreted by the cell. Some carbohydrates are also synthesized by enzymes within the cavities of the rough endoplasmic reticulum and may be attached to newly formed protein (glycosylation). Vesicles are budded off from the rough endoplasmic reticulum for transport to the Golgi as part of the protein-targeting mechanism of the cell.
Figure 2.8 The endoplasmic reticulum. A, Rough endoplasmic reticulum with attached ribosomes; B, smooth endoplasmic reticulum with associated vesicles. The dense particles are glycogen granules. (By kind permission from Rose Watson, Cancer Research UK.)
SMOOTH ENDOPLASMIC RETICULUM The smooth endoplasmic reticulum (Fig. 2.8B) is associated with carbohydrate metabolism and many other metabolic processes, including detoxification and synthesis of lipids, cholesterol and other steroids. The membranes of the smooth endoplasmic reticulum serve as surfaces for the attachment of many enzyme systems, e.g. the enzyme cytochrome P450, which is involved in important detoxification mechanisms and is thus accessible to its substrates, which are generally lipophilic. They also cooperate with the rough endoplasmic reticulum and the Golgi apparatus to synthesize new membranes; the protein, carbohydrate and lipid components are added in different structural compartments. Highly specialized types of endoplasmic reticulum are present in some cells. For example, in skeletal muscle cells, the smooth endoplasmic reticulum (sarcoplasmic reticulum) stores calcium ions, which are released into the cytosol
to initiate contraction after stimulation initiated by a motor neurone at the neuromuscular junction (p. 64).
RIBOSOMES Ribosomes are macromolecular machines that catalyse the synthesis of proteins from amino-acids. They are granules c.15 nm in diameter, composed of ribosomal RNA (rRNA) molecules assembled into two unequal subunits. A large number of proteins, mostly small and basic, are applied mainly to the surfaces of the subunit cores of RNA. The subunits can be separated by their sedimentation coefficients (S) in an ultracentrifuge, into larger 60S and smaller 40S components. These are associated with 73 different proteins (40 in the large subunit and 33 in the small), which have structural and enzymatic functions. Three small, highly convoluted rRNA strands (28S, 5.8S and 5S) make up the large subunit, and one strand (18S) is in the small subunit. Their synthesis and assembly into subunits takes place in the nucleolus, and includes association with ribosomal proteins translocated from their site of synthesis in the cytoplasm. The individual subunits are then transported into the cytoplasm, where they remain separate from each other when not actively synthesizing proteins. A typical cell contains millions of ribosomes. They may be solitary, relatively inactive structures, or may form groups (polyribosomes or polysomes) attached to messenger RNA (mRNA), which they translate during protein synthesis (Fig. 2.9). Polysomes may be attached to the membranes of the rough endoplasmic reticulum or may lie free in the cytosol, where they synthesize proteins for use outside the system of membrane compartments, including enzymes of the cytosol and cytoskeletal proteins. Some of the cytosolic products include proteins that can be inserted directly into (or through) membranes of selected organelles, such as mitochondria and peroxisomes. page 12 page 13
Figure 2.9 Ribosomes, distributed either singly, clustered as polyribosomes (polysomes), or attached to the rough endoplasmic reticulum (right). (By permission from Stevens A, Lowe JS 1996 Human Histology, 2nd edn. London: Mosby.)
In a mature polysome, all the attachment sites of the mRNA are occupied as ribosomes move along it, synthesizing protein according to its nucleic acid sequence. Consequently, the number of ribosomes in a polysome indicates the
length of the mRNA molecule and hence the size of the protein being made. The two subunits have separate roles in protein synthesis. The 40S subunit is the site of attachment and translation of mRNA. The 60S subunit is responsible for the release of the new protein and, where appropriate, attachment to the endoplasmic reticulum via an intermediate docking protein that directs the newly synthesized protein through the membrane into the cisternal space.
GOLGI APPARATUS (GOLGI COMPLEX) (Figs 2.10, 2.11)
Figure 2.10 Golgi apparatus in a fibroblast. Several Golgi stacks are present, each with convex cis- and concave trans-Golgi surfaces, and associated transport vesicles. The edge of the nucleus appears on the left. (By kind permission from Rose Watson, Cancer Research UK.)
page 13 page 14
Figure 2.11 The Golgi apparatus and its functional relationships with associated structures.
The Golgi apparatus is a distinct cytoplasmic region near the nucleus, and is particularly prominent in secretory cells when stained with silver or other metallic salts. The Golgi apparatus forms part of the pathway by which proteins synthesized in the rough endoplasmic reticulum undergo post-translational modification and are targeted to the cell surface for secretion or for storage in membranous vesicles. Ultra-structurally, the Golgi apparatus is a membranous organelle consisting of a stack of several flattened membranous cisternae, together with clusters of vesicles surrounding its surfaces. Seen in vertical section, it is often cup-shaped. Small transport vesicles from the rough endoplasmic reticulum, generated by a process of budding and pinching off, are received at one face of the Golgi stack, the convex cis-face (entry or forming surface). Here, they deliver their contents to the first cisterna in the series by membrane fusion. From the edges of this cisterna, the protein is transported to the next cisterna by vesicular budding and then fusion, and this process is repeated until the final cisterna at the concave trans face (exit or condensing surface) is reached. Here, larger vesicles are formed for delivery to other parts of the cell. In addition to these cisternae, there are other membranous structures that form an integral part of the Golgi apparatus, termed the cis-Golgi and trans-Golgi networks. The cis-Golgi network is a region of complex membranous channels interposed between the rough endoplasmic reticulum and the Golgi cis face (Golgi-rough endoplasmic reticulum complex), which receives and transmits vesicles in both directions. Its function is to select appropriate proteins synthesized on the rough endoplasmic reticulum for delivery by vesicles to the Golgi stack, while inappropriate proteins are shuttled back to the rough
endoplasmic reticulum. The trans-Golgi network, at the other side of the Golgi stack, is also a region of interconnected membrane channels engaged in protein sorting. Here, modified proteins processed in the Golgi cisternae are packaged selectively into vesicles and dispatched to different parts of the cell. The packaging depends on the detection, by the trans-Golgi network, of particular amino-acid signal sequences, leading to their enclosure in membranes of appropriate composition that will further modify their contents, e.g. by extracting water to concentrate them or by pumping in protons to acidify their contents. The membranes contain specific signal proteins, which may allocate them to microtubule-based transport pathways and allow them to dock with appropriate targets elsewhere in the cell, e.g. the plasma membrane in the case of secretory vesicles. Vesicle formation and budding at the trans-Golgi network involves the addition of clathrin on their external surface, to form coated pits. Within the Golgi stack proper, proteins undergo a series of sequential chemical modifications that started in the rough endoplasmic reticulum. These include: changes in glycosyl groups, e.g. removal of mannose, addition of N-acetyl glucosamine and sialic acid; sulphation of attached glycosaminoglycans; protein phosphorylation. Lipids formed in the endoplasmic reticulum are also routed for incorporation into vesicles. The role of the Golgi apparatus in the synthesis of primary lysosomes is a major activity in cells with abundant lysosomes, such as those with phagocytic roles. In glandular cells with an apical secretory zone, the Golgi apparatus lies between the secretory surface and the nucleus. In fibroblasts, there are two or more groups of Golgi stacks; up to 50 groups are found in liver cells. The Golgi apparatus is often closely associated with the centrosome (a region of the cell containing a centriole pair and related microtubules), reflecting a link with the microtubule-mediated vesicle transport system.
ENDOSOMES, LYSOSOMES AND PEROXISOMES The endosome system of vesicles originates in small endocytic vesicles or larger phagosomes taken up by the cell from the exterior. The system is linked functionally to a second series of membranous structures, the lysosomes. Lysosomes contain acid hydrolases, which process or degrade exogenous materials (heterophagy), and intracellular organelles that are exhausted, damaged or no longer required (autophagy). There is a continual exchange of vesicles between this system and the Golgi-rough endoplasmic reticulum complex, so that the endosomal/lysosomal system is provided with hydrolytic enzymes and the Golgi receives depleted vesicles for recharging. Once internalized, endocytic vesicles shed their coat of adaptin and clathrin, and fuse with a tubular cisterna termed an early endosome, where the receptor molecules release their bound ligands. Membrane and receptors from the early endosomes can be recycled to the cell surface as exocytic vesicles.
LATE ENDOSOMES After a brief period in the early endosomes, materials can be passed on to late endosomes, which are a more deeply placed set of tubules, vesicles or cisternae. Late endosomes receive lysosomal enzymes via vesicles (small lysosomes) transported from the Golgi apparatus. The pH of late endosomes is low (about 5.0) and this activates lysosomal acid hydrolases to degrade the endosomal contents. The products of hydrolysis are either passed through the membrane into the cytosol, or may be retained in the endosome. Late endosomes may grow considerably in size by vesicle fusion to form multivesicular bodies (Fig. 2.12), and the enzyme concentration may increase greatly to form the large, dense classic lysosomes described by de Duve. However, such large organelles do not
appear in all cells, perhaps because late endosomes often deal very rapidly with endocytosed material.
LYSOSOMES Lysosomes are dense, spheroidal, membrane-bound bodies 80-800 nm in diameter (Fig. 2.12), often with complex inclusions of material undergoing hydrolysis (secondary lysosomes). They contain acid hydrolases able to degrade a wide variety of substances. To date, more than 40 lysosomal enzymes have been described, including proteases, lipases, carbohydrases, esterases and nucleases. The enzymes are heavily glycosylated, and are maintained at a low pH by proton pumps in the lysosomal membranes.
page 14 page 15
Figure 2.12 The typical features of primary and secondary lysosomes in the cytoplasm of a liver cell. Primary lysosomes (Ly1) are homogeneous membrane-bound bodies, whereas secondary lysosomes (Ly2) are typically variable in density and content, and often difficult to distinguish from later-stage residual bodies. Note their size relative to mitochondria (M). A number of late endosomes (multivesicular bodies, MB) are also shown. (By permission from Young B, Heath JW 2000 Wheater's Functional Histology. Edinburgh: Churchill Livingstone.)
Lysosomes are numerous in actively phagocytic cells, e.g. macrophages and neutrophil granulocytes, in which lysosomes are responsible for destroying phagocytosed bacteria. In these cells, the phagosome containing the bacterium may fuse with several lysosomes. Lysosomes are also frequent in cells with a high turnover of organelles, e.g. exocrine gland cells and neurones. Effete organelles are targeted for demolition by a process that is not fully understood, but which results in engulfment of areas of cytoplasm, including entire organelles, in a membranous cisterna. The structure then fuses with lysosomes and the contents are rapidly degraded. Material that has been hydrolysed within late endosomes and lysosomes may be completely degraded to soluble products, e.g. amino-acids, which are recycled through metabolic pathways. However degradation is usually incomplete, and
some debris remains. A debris-laden vesicle is called a residual body, and may be passed to the cell surface, where it is ejected by exocytosis; alternatively, it may persist inside the cell as an inert residual body. Considerable numbers of residual bodies can accumulate in long-lived cells, often fusing to form larger dense vacuoles with complex lamellar inclusions. As their contents are often darkly pigmented, this may change the colour of the tissue, e.g. in neurones the end-product of lysosomal digestion, lipofuscin (neuromelanin or senility pigment), gives ageing brains a brownish-yellow colouration. Lysosomal enzymes may also be secreted - often as part of a process to alter the extracellular matrix, as in osteoclast erosion of bone (p. 92). Abnormal release of enzymes can cause tissue damage, as in certain types of arthritis. Some drugs, e.g. cortisone, can stabilize lysosomal membranes and may therefore inhibit many lysosomal activities, including the secretion of enzymes, and their fusion with phagocytic vesicles. Lysosomal storage diseases
If any of the lysosomal enzymes are defective because of gene mutations, the materials that they normally degrade will accumulate within late endosomes and lysosomes. Many such lysosomal storage diseases are known, e.g. Tay-Sachs disease, in which a faulty gangliosidase leads to the accumulation of glycolipid in neurones, causing death during childhood. In Hurler's syndrome, failure to metabolize certain glycosaminoglycans causes the accumulation of large amounts of matrix within connective tissue, which distorts growth of many parts of the body.
PEROXISOMES Peroxisomes are membrane-bound vacuoles c.0.5-0.15 µm across, present in all nucleated cell types. They often contain dense cores or crystalline interiors composed mainly of high concentrations of the enzyme urate oxidase. Large (0.5 µm) peroxisomes are particularly numerous in hepatocytes and kidney tubule cells. Peroxisomes are important in the oxidative detoxification of various substances taken into or produced within cells, including ethanol and formaldehyde . Oxidation is carried out by a number of enzymes, including Damino-acid oxidase and urate oxidase, which generate hydrogen peroxide as a source of molecular oxygen. Excess amounts of hydrogen peroxide are broken down by the enzyme, catalase. Peroxisomes also oxidize fatty acid chains by "oxidation. The formation of peroxisomes is unusual in that they appear to be derived by the growth and fission of previously existing peroxisomes. Their internal proteins are passed from the cytosol directly through channels in their membranes, rather than by packaging from the rough endoplasmic reticulum and Golgi apparatus. These features are also found in mitochondria, although peroxisomal proteins are coded for entirely in the nucleus. A genetic abnormality in the translocation of proteins into peroxisomes, leading to peroxisomal enzyme deficiencies, is seen in Zellweger syndrome, caused by a gene mutation in an integral membrane protein (peroxisome assembly factor-1). In homozygotes, this is usually fatal shortly after birth.
MITOCHONDRIA
Figure 2.13 A mitochondrion. The folded cristae project into the matrix from the inner mitochondrial membrane.
The mitochondrion is a membrane-bound organelle (Fig. 2.13). It is the principal source of chemical energy in most cells. Mitochondria are the site of the citric acid (Kreb's, tricarboxylic acid) cycle and the electron transport (cytochrome) pathway by which complex organic molecules are finally oxidized to carbon dioxide and water. This process provides the energy to drive the production of ATP from ADP and inorganic phosphate (oxidative phosphorylation). The various enzymes of the citric acid cycle are located in the mitochondrial matrix, whereas those of the cytochrome system and oxidative phosphorylation are localized chiefly in the inner mitochondrial membrane. The numbers of mitochondria in a particular cell reflect its general energy requirements; e.g. in hepatocytes there may be as many as 2000, whereas in resting lymphocytes there are usually very few. Mature erythrocytes lack mitochondria altogether. Cells with few mitochondria generally rely largely on glycolysis for their energy supplies. These include some very active cells, e.g. fast twitch skeletal muscle fibres, which are able to work rapidly, but for only a limited duration. Mitochondria appear in the light microscope as long thin threads, or alternatively as spherical or ellipsoid bodies in the cytoplasm of most cells, particularly those with a high metabolic rate, e.g. secretory cells in exocrine glands. In living cells, mitochondria constantly change shape and intracellular position; they multiply by growth and fission and may undergo fusion. In the electron microscope, mitochondria usually appear as elliptical bodies 0.52.0 µm long. Each mitochondrion is lined by an outer and an inner unit membrane, separated by a variable gap termed the intermembrane space. The lumen is surrounded by the inner membrane and contains the mitochondrial matrix. The outer membrane is smooth and sometimes attached to other organelles, particularly microtubules. The inner membrane is deeply folded to form incomplete transverse or longitudinal tubular invaginations, cristae, which create a relatively large surface area of membrane. Mitochondrial shape, and the shape and organization of the cristae, vary with the cell type. Cristae are most numerous and complex in cells with a high metabolic rate, e.g. cardiac muscle cells. The permeabilities of the two mitochondrial membranes differ considerably: the outer membrane is freely permeable to many substances because of the presence of large non-specific channels formed by proteins (porins), whereas the inner membrane is permeable to only a narrow range of molecules. The presence of cardiolipin, a phospholipid, in the inner membrane may contribute to this relative impermeability.
The mitochondrial matrix is an aqueous environment. It contains a variety of enzymes, and strands of mitochondrial DNA with the capacity for transcription and translation of a unique set of mitochondrial genes (mitochondrial mRNAs and transfer RNAs, mitochondrial ribosomes with rRNAs). The DNA forms a closed loop, c.5 µm across; several identical copies are present in each mitochondrion. The ratio between its bases differs from that of nuclear DNA, and the RNA sequences also differ in the precise genetic code used in protein synthesis. At least 13 respiratory chain enzymes of the matrix and inner membrane are encoded by the small number of genes along the mitochondrial DNA. The great majority of mitochondrial proteins are encoded by nuclear genes and made in the cytosol, then inserted through special channels in the mitochondrial membranes to reach their destinations. Their membrane lipids are synthesized in the endoplasmic reticulum. page 15 page 16
Mitochondrial ribosomes are smaller and quite distinct from those of the rest of the cell. Mitochondrial ribosomes and nucleic acids resemble those of bacteria. This similarity underpins the theory that mitochondrial ancestors were oxygenutilizing bacteria that existed in a symbiotic relationship with eukaryotic cells unable to metabolize the oxygen produced by early plants. As mitochondria are formed only from previously existing ones, it follows that all mitochondria in the body are descended from those in the cytoplasm of the fertilized ovum. It has also been shown that mitochondria are of maternal origin because the mitochondria of the sperm are not generally incorporated into the ovum at fertilization. Thus mitochondria (and mitochondrial genetic variations and mutations) are passed only through the female line. Mitochondria are distributed within a cell according to regional energy requirements, e.g. near the bases of cilia in ciliated epithelia, in the basal domain of the cells of proximal convoluted tubules in the renal cortex (where considerable active transport occurs) and around the proximal end of the flagellum in spermatozoa. They may be involved with tissue-specific metabolic reactions, e.g. various urea-forming enzymes in liver cell mitochondria. Moreover, a number of genetic diseases of mitochondria affect particular tissues exclusively, e.g. mitochondrial myopathies (skeletal muscle) and mitochondrial neuropathies (nervous tissue). For further information see Graff et al (2002).
CYTOSOLIC ORGANELLES The aqueous cytosol surrounds the membranous organelles described above. It also contains various non-membranous organelles, including free ribosomes, a system of filamentous proteins known as the cytoskeleton, and other inclusions, such as storage granules (e.g. glycogen) and lipid vacuoles.
LIPID VACUOLES Lipid vacuoles are spherical bodies of various sizes found within many cells, but are especially prominent in the adipocytes (lipocytes) of adipose connective tissue. They do not belong to the Golgi-related vacuolar system of the cell. They are not membrane bound, but are droplets of lipid suspended in the cytosol. In cells specialized for lipid storage the vacuoles reach 80 µm or more in diameter. Lipid vacuoles are often surrounded by cytoskeletal filaments that help to stabilize them within cells and to prevent their fusion with the membranes of other organelles, including the plasma membrane. They function as stores of chemical energy, thermal insulators and mechanical shock absorbers in adipocytes. In many cells, they may represent end-products of other metabolic pathways, e.g. in steroid-synthesizing cells, where they are a prominent feature of the cytoplasm. They may also be secreted, as in the alveolar epithelium of the lactating breast.
CYTOSKELETON The cytoskeleton is a system of filamentous intracellular proteins of different shapes and sizes that form a complex, often interconnected, network throughout the cytoplasm. It provides mechanical support, maintains cell shape and rigidity, and enables cells to adopt highly asymmetric or irregular profiles, e.g. in neurones. The cytoskeleton plays an important part in establishing structural polarity and different functional domains within a cell. It also provides mechanical support for projections from the cell surface such as microvilli and cilia, and anchors them into the cytoplasm. The cytoskeleton restricts specific organelles to particular cellular locations, e.g. the Golgi apparatus is near the nucleus and endoplasmic reticulum, and mitochondria are near sites of energy requirement. Most specifically, the cytoskeleton is concerned with motility, either within the cell (e.g. shuttling vesicles and macromolecules between cytoplasmic sites, or the movement of chromosomes during mitosis), or of the entire cell (e.g. in embryonic morphogenesis or the chemotactic migration of leukocytes). One of the most highly developed and specialized functions of the cytoskeleton is seen in the contractility of muscle cells.
Figure 2.14 Actin microfilaments present at high density in the cytoplasm of a smooth muscle cell. Cytoplasmic dense bodies (arrows) are points of attachment for the actin filaments.
The catalogue of cytoskeletal structural proteins is extensive and still increasing. The major filamentous structures found in non-muscle cells are microfilaments (actin), microtubules (tubulin), and intermediate filaments (varieties of cell specific intermediate filament proteins). Other important components are generally smaller proteins that bind to the principal filamentous types to link them together or to generate movement. These include actin-binding proteins such as myosin, which in some cells can assemble into thick filaments, and microtubule-associated proteins. Actin filaments (microfilaments)
Actin filaments are well-defined, fine filaments with a width of 6-8 nm (Fig. 2.14), and a solid cross-section. Within most cell types, actin constitutes the most abundant protein and in some motile cells its concentration may exceed 200 µM (10 mg protein per ml cytoplasm). The filaments are formed by the ATPdependent polymerization of actin monomer into a characteristic linear form in which the subunits are arranged in a single tight helix with a distance of 13 subunits between turns. The polymerized form is termed F-actin (fibrillar actin) and the unpolymerized form is G-actin (globular actin), with a molecular mass of 43 kDa. Each monomer has an asymmetric structure. When monomers polymerize, they confer a defined polarity on the filament: the plus end favours monomer addition, and the minus end favours monomer dissociation. Myosins bind to filamentous actin at an angle to give the appearance of a series of
arrowheads pointing towards the minus end of the filament, and the barbs point towards the plus end. There is a dynamic equilibrium between G-actin and Factin: in most cells c.50% of the actin is estimated to be in the polymerized state. Actin-binding proteins
A wide variety of actin-binding proteins exist that are capable of modulating the form of actin within the cell. These interactions are fundamental to the organization of cytoplasm and to cell shape. Actin-binding proteins can be divided into bundling proteins, gel-forming proteins and filament severing proteins. Bundling proteins tie actin filaments together in longitudinal arrays to form cables or core structures. The bundles may be closely spaced, e.g. in microvilli, microspikes and filopodia, where parallel filaments are tied tightly together to form stiff bundles orientated in the same direction. Proteins with this function include fimbrin and villin (also classified as a severing protein). Other actin-bundling proteins form rather looser bundles of filaments that run anti-parallel to each other with respect to their plus and minus ends. They include !-actinin and myosin II, which can form cross-links with ATP-dependent motor activity, and cause adjacent actin filaments to slide on each other, and either change the shape of cells or (if the actin bundles are anchored into the cell membrane at both ends), maintain a degree of active rigidity. Gel-forming proteins, such as filamin, interconnect adjacent actin filaments to produce loose filamentous meshworks (gels) composed of randomly orientated Factin. These networks are frequently found in the outer cortical regions of cells, e.g. fibroblasts. They form a semi-rigid zone from which most other organelles are excluded. Severing proteins, such as gelsolin and severin, bind to F-actin filaments and sever them, which produces profound changes within the actin cytoskeleton and in its coupling to the cell surface. page 16 page 17
Other classes of actin-binding proteins link the actin cytoskeleton to the plasma membrane either directly or indirectly through a variety of membrane-associated proteins. The latter may also create links via transmembrane proteins to the extracellular matrix. Best known of these is the family of spectrin-like molecules, which can bind to actin and also to each other and various membrane-associated proteins to create supportive networks beneath the plasma membrane. Spectrin is found in erythrocytes, and closely related molecules are present in many other cells; for instance, fodrin is found in nerve cells, and dystrophin occurs in muscle cells, linking the contractile apparatus with the extracellular matrix via integral membrane proteins. Proteins such as ankyrin (which also binds actin directly), vinculin, talin, zyxin and paxillin connect actin-binding proteins to integral plasma membrane proteins such as integrins (directly or indirectly), and thence to focal adhesions. Myosin I and other unconventional myosins connect actin filaments to membranous structures, including the plasma membrane and transport vesicle membranes. Tropomyosin, an important regulatory protein of muscle fibres, is also present in non-muscle cells, where its function may be primarily to stabilize actin filaments against depolymerization. For further reading see Pollard and Earnshaw (2002). Myosins - the motor proteins
The myosin family of microfilaments is often classified within a distinct category of motor proteins. Myosin proteins have a globular head region consisting of a heavy and a light chain. The heavy chain bears an !-helical tail of varying length. The head has an ATPase activity and can bind to and move along actin filaments - the basis for myosin function as a motor protein. The best-known class is myosin II, which occurs in muscle and in many non-muscle cells. Its molecules have two heads and two tails, intertwined to form a long rod. The rods can bind to each other to form long, thick filaments, as seen in striated and smooth muscle
fibres, myoepithelial cells and myofibroblasts. Myosin II molecules can also assemble into smaller groups, especially dimers, which can cross-link individual actin microfilaments in stress fibres and other F-actin arrays. The ATP-dependent sliding of myosin on actin forms the basis for muscle contraction and the extension of microfilament bundles, as seen in cellular motility or in the contraction of the ring of actin and myosin around the cleavage furrow of dividing cells. There are a number of known subtypes of myosin II: they assemble in different ways and have different dynamic properties. In skeletal muscle the myosin molecules form filaments c.15 nm thick, reversing their direction of assembly at the midpoint, which is bare of head regions, to produce a symmetric arrangement of subunits. In smooth muscle the molecules form thicker, flattened ribbons and are orientated in different directions on either face of the ribbon. These arrangements have important consequences for the contractile force characteristics of the different types of muscle cell. Related molecules are known as unconventional myosins. They include the myosin I subfamily of single-headed molecules with tails of varying length. These molecules are associated with membranes to which their tails can attach, and are implicated in the movements of membranes on actin filaments. So, for example, vesicles track along F-actin in a similar manner to kinesin and dynein-related movements along microtubules. Other functions of myosin I are the movements of membranes in endocytosis, microspike formation in neuronal growth cones, actinactin sliding and attachment of actin to membranes, e.g. of microvilli. Other myosins have been isolated; the significance of their diversity is not fully understood. Other thin filaments
A heterogeneous group of filamentous structures with diameters of 2-4 nm occur in various cells. The two most widely studied forms, titin and nebulin, constitute c.13% of the total protein of skeletal muscle. They are amongst the largest known molecules, and have subunit weights of around 106; native molecules are c.1 µm in length. Their elastic properties are important for the effective functioning of muscle, and possibly for other cells. Intermediate filaments
Intermediate filaments are c.10 nm thick and formed by a heterogeneous group of filamentous proteins. They are found in different cell types and are often present in large numbers, either where structural strength is needed (Fig. 2.15), or to provide scaffolding for the attachment of other structures. Intermediate filaments of different molecular classes are characteristic of particular tissues or states of maturity. They are therefore important indicators of the origins of cells or levels of differentiation, and are of considerable value in histopathology.
Figure 2.15 Keratin filaments (tonofilaments) in two adjacent keratinocytes (K) of the epidermis. Keratin filaments also insert into the desmosome contacts (D) between cells. (By permission from Young B, Heath JW 2000 Wheater's Functional Histology. Edinburgh: Churchill Livingstone.)
Of the different classes of intermediate filaments, keratin (cytokeratin) proteins are found in epithelia, where keratin filaments are always composed of equal ratios of types I (acidic) and II (basic to neutral) keratins. About 20 types of each of the acidic and basic/neutral keratin proteins are known. Within the epidermis, expression of keratin heterodimer combinations changes as keratinocytes mature during their transition from basal to superficial layers. Genetic abnormalities of keratins are known to affect the mechanical stability of epithelia. For example, the disease epidermolysis bullosa simplex causes lysis of epidermal basal cells and blistering of the skin after mechanical trauma. It is caused by defects in genes encoding keratins 5 and 14, which produce cytoskeletal instability and thus cellular fragility in the basal cells. When keratins 1 and 10 are affected, cells in the spinous layer of the epidermis lyse, and this produces the intraepidermal blistering of epidermolytic hyperkeratosis. For a recent review, see Porter and Lane (2003). Vimentins occur in mesenchyme-derived cells of connective tissue, desmins in muscle cells, glial fibrillary acidic protein in glial cells, and peripherin in peripheral axons. Neurofilaments are a major cytoskeletal element in neurones, particularly in axons (Fig. 2.16), where they are the dominant protein. They are heteropolymers of low, medium and high molecular weight neurofilament proteins; the low molecular weight form is always present in combination with either the medium- or the high-molecular weight neurofilament. Abnormal accumulations of neurofilaments (neurofibrillary tangles) are characteristic features of a number of neuropathological conditions. Other intermediate filament proteins include nestin, a molecule resembling neurofilament protein which forms intermediate filaments in neurectodermal stem cells in particular. Nuclear lamins form intermediate filaments that line the inner surface of the nuclear envelope of all nucleated cells. They provide a mechanical
framework for the nucleus and act as attachment sites for chromosomes. They are unusual in that they form a square lattice of regularly spaced crossing filaments rather than bundles, reflecting their unusual molecular composition. page 17 page 18
Figure 2.16 Microtubules (MT) in axons in peripheral nerve. A, In transverse section, microtubules appear as hollow spherical structures. The axon is ensheathed by a Schwann cell (S) and its associated basal lamina. B, In longitudinal section, microtubules appear as straight, unbranched tubes. Neurofilaments (intermediate filaments of neurones; NF) appear in transverse section as solid punctate structures, smaller than microtubules. (By permission from Young B, Heath JW 2000 Wheater's Functional Histology. Edinburgh: Churchill Livingstone.)
The exact manner in which intermediate filament proteins polymerize to form linear filaments is much more complex than that of tubulin or actin, and has not been fully determined. The individual intermediate filament proteins are chains
with a middle !-helical region flanked on either side by non-helical domains. The proteins associate as coiled coil dimers that form short rods c.48 nm long. These assemble in pairs in a staggered antiparallel formation to form soluble tetramers, eight of which pack together laterally and twist into the rope-like 10 nm intermediate filament. The 32 !-helices in parallel give the filaments their tensile strength. However, unlike actin and myosin, the antiparallel arrangement of the dimers produces a filamentous protein with no intrinsic polarity. The non-coiled regions of the subunits project outwards as side arms that can link intermediate filaments into bundles or attach them to other structures. The existence of different combinations of subunit proteins within one filament is the basis of their functional diversity. In the living cell they have been shown to be quite dynamic structures, possibly as a result of reversible phosphorylation. Microtubules
Microtubules are polymers of tubulin with the form of hollow, relatively rigid cylinders, c.25 nm in diameter and of varying length (up to 70 µm in spermatozoan flagella) (Fig. 2.16). They are present in most cell types, and are particularly abundant in neurones, leukocytes, blood platelets and the mitotic spindles of dividing cells. They also form part of the structure of cilia, flagella and centrioles. There are two major classes of tubulin: !- and "-tubulins. Before microtubule assembly, tubulins are associated as dimers with a combined molecular mass of 100 kDa (50 kDa each). Each protein subunit is c.5 nm across and arranged along the long axis in straight rows of alternating and "-tubulins, forming protofilaments. About 13 protofilaments (the number varies between 11 and 16), associate in a ring to form the wall of a hollow cylindrical microtubule. Each longitudinal row is slightly out of alignment with its neighbour, so that a spiral pattern of alternating ! and " subunits appears when the microtubule is viewed from the side. There is a dynamic equilibrium between the dimers and assembled microtubules: dimeric asymmetry creates polarity (!-tubulins are all orientated towards the minus end, "-tubulins towards the plus end). Tubulin is added preferentially to the plus end; the minus end is relatively slow growing. When tubulin is added at one end while being removed at the other (treadmilling), the microtubule gradually shifts its position longitudinally. Polymerization requires phosphorylation of tubulins by guanosine triphosphate, and either a nucleation site (e.g. the end of a pre-existing microtubule), or a microtubule-organizing centre (e.g. those surrounding, and including, centrioles), around which spindle microtubules polymerize during cell division and from which cilia can grow. page 18 page 19
Various drugs (e.g. colcemid, vinblastine, griseofulvin , nocodazole) cause microtubule depolymerization by binding the soluble tubulin dimers and so shifting the equilibrium towards the unpolymerized state. Microtubule demolition causes a wide variety of effects, including the inhibition of cell division by disruption of the mitotic spindle. Conversely, the drug taxol stabilizes microtubules and promotes abnormal microtubule assembly which causes a peripheral neuropathy. Different microtubules possess varying degrees of stability, e.g. microtubules in cilia are generally unaffected by many drugs that cause microtubular demolition. There are also differences between tissues, e.g. neurones have a special tubulin subclass. Tubulins associated with microtubule organizing centres include #tubulin. Microtubule-associated proteins
Various small proteins that can bind to assembled tubulins may be concerned with structural properties or associated with motility. Structural microtubuleassociated proteins (MAPs) form cross-bridges between adjacent microtubules or between microtubules and other structures such as intermediate filaments,
mitochondria and the plasma membrane. Microtubule-associated proteins found in neurones include: MAPs 1A and 1B, which are present in neuronal dendrites and axons; MAPs 2A and 2B, found chiefly in dendrites; and tau, found only in axons. MAP 4 is the major microtubule-associated protein in many other cell types. Structural microtubule-associated proteins are implicated in microtubule formation, maintenance and demolition, and are therefore of considerable significance in cell morphogenesis, mitotic division, and the maintenance and modulation of cell shape. Motility-associated microtubule-associated proteins are found in situations in which movement occurs over the surfaces of microtubules, e.g. the transport of cytoplasmic vesicles, bending of cilia and flagella, and some movements of mitotic spindles. They include a large family of motor proteins, the best known of which are the dyneins and kinesins. Another protein, dynamin, is involved in endocytosis. The kinetochore proteins assemble at the centromere during mitosis and meiosis. They attach to spindle microtubules; some of the kinetochore proteins are responsible for chromosomal movements in mitotic and meiotic anaphase. All of these microtubule-associated proteins bind to microtubules and actively slide along their surfaces. Kinesins and dyneins can simultaneously attach to membranes such as transport vesicles and convey them along microtubules for considerable distances, thus enabling selective targeting of materials within the cell. Such movements occur in both directions along microtubules. Kinesindependent motion is mostly towards the plus ends of microtubules, e.g. from the cell body towards the axon terminals in neurones, and away from the centrosome in other cells. Conversely, dynein-related movements are in the opposite direction, i.e. to the minus ends of microtubules. Dyneins also form the arms of peripheral microtubules in cilia and flagella, where they make dynamic crossbridges to adjacent microtubule pairs. The resulting shearing forces cause the axonemal array of microtubules to bend, generating ciliary and flagellar beating movements. A number of related motor proteins can also interact with microtubules, e.g. kinesin-related proteins cross-link mitotic spindle microtubules to push the two centriolar poles apart during mitotic prophase. Centrioles, centrosomes and basal bodies
Centrioles are microtubular cylinders c.0.2 µm in diameter and 0.4 µm long (Fig. 2.17). They are formed by a ring of nine microtubule triplets linked by a number of other proteins. At least two centrioles occur in all cells that are capable of mitotic division. They usually lie close together, at right angles or, most usually, at an oblique angle to each other (an arrangement often termed a diplosome), within the centrosome, a densely filamentous region of cytoplasm at the centre of the cell. The centrosome is the major microtubule-organizing centre of the cell; it is the site at which new microtubules are formed and the mitotic spindle is generated during cell division. Before cell division, a new centriole forms at right angles to each one of the existing centrosomal pair; the resulting new pairs are passed on to the daughter cells. The proximity of the centrosome to the Golgi apparatus provides a means of targeting Golgi vesicular products to different parts of the cell. The microtubule-organizing centre contains complexes of #-tubulin that bind the minus ends of microtubules. Basal bodies are microtubule-organizing centres that are closely related to centrioles, and are believed to be derived from them. They are located at the bases of cilia and flagella, which they anchor to the cell surface. The outer microtubule doublets of cilia and flagella originate from two of the microtubules in each triplet of the basal body.
Figure 2.17 In electron microscopic preparations, one centriole is usually visible in cross-section, revealing the circular arrangement of tubules, while its partner is cut either longitudinally or slightly obliquely (above). (By permission from Stevens A, Lowe JS 1996 Human Histology, 2nd edn. London: Mosby.)
CELL SURFACE PROJECTIONS The surfaces of many different types of cell are specialized to form structures that project from the surface. These projections may permit movement of the cell itself (flagella), or of fluids across the apical cell surface (cilia), or increase the surface area available for absorption (microvilli). Infoldings of the basolateral plasma membrane also increase the area for transport across this surface of the cell.
CILIA AND FLAGELLA Cilia and flagella are motile, hair-like projections of the cell surface which create currents in the surrounding fluid, movements of the cell to which they are attached, or both. Cilia occur on many internal surfaces of the body, in particular: the epithelia of most of the respiratory tract; parts of the male (p. 1307) and female (p. 1329) reproductive tracts; the ependyma that line the central canal of the spinal cord and ventricles of the brain (p. 53). They also occur at the endings of olfactory receptors and vestibular hair cells, and, in modified form, as portions of the rods and cones of the retina (p. 710). A single cell may bear many cilia, e.g. in bronchial epithelium, or only one or two. Each male gamete possesses a single flagellum c.70 µm long. A cilium or flagellum consists of a shaft (c.0.25 µm diameter) constituting most of its length, a tapering tip and a basal body at its base, which lies within the surface cytoplasm of the cell (Fig. 2.18). Other than at its base, the entire structure is covered by plasma membrane. The core of the cilium is the axoneme, a cylinder of nine microtubule doublets that surrounds a central pair of single microtubules. Several filamentous structures are associated with the microtubules in the shaft, e.g. radial spokes extend inwards from the outer microtubules towards the central pair. The outer doublet microtubules bear two rows of tangential dynein arms attached to the A subfibre of the doublet, which point towards the B subfibre of the adjacent doublet. Adjacent doublets are also linked by thin filaments. Other
filaments partially encircle the central pair of microtubules, which are also united by ladder-like spokes. The '9 + 2' pattern of microtubules imparts a plane of symmetry that passes perpendicular to a line joining the central pair and corresponds to the direction of bending. page 19 page 20
Figure 2.18 A, Structure of a cilium shown in (left) transverse and (right) longitudinal section. B, Apical region of respiratory epithelial cells, showing the proximal parts of three cilia sectioned longitudinally, anchored into the cytoplasm by basal bodies (BB). Other cilia project out of the plane of section and are cut transversely, showing the '9 + 2' arrangement of microtubules. (A, redrawn by permission from Sleigh MA 1977 The nature and action of respiratory tract cilia. In: Brain JD, Proctor DF, Reid LM (eds) Respiratory Defense Mechanisms. Part 1. New York: Dekker; pp247-288. B, by permission from Young B, Heath JW 2000 Wheater's Functional Histology. Edinburgh: Churchill Livingstone.)
Movements of cilia and flagella are broadly similar. Flagella move by rapid undulation, which passes from the attached to the free end. In human spermatozoa there is an additional helical component to this motion. In cilia, the beating is planar, but asymmetric. In the effective stroke, the cilium remains stiff except at the base, where it bends to produce an oar-like stroke. The recovery stroke follows, during which the bend passes from base to tip, returning the cilium to its initial position for the next cycle. The activity of groups of cilia is usually coordinated so that the bending of one is rapidly followed by the bending of the next and so on, resulting in long travelling waves of metachronal synchrony. These pass over the tissue surface in the same direction as the effective stroke. When a cilium bends, the microtubules do not change in length, but slide on one another. The dynein arms of peripheral doublets slant towards the base of the cilium from their attached ends. Dynein has an ATPase activity, which is stimulated by magnesium ions, and causes mutual sliding of adjacent doublets by initially attaching sideways to the next pair, then swinging upwards towards the tip of the cilium. There is a group of genetic diseases in which cilia beat either ineffectively or not at all, e.g. Kartagener's immotile cilium syndrome. Affected cilia exhibit various ultrastructural defects in their internal structure, such as a lack of dynein arms or missing spokes. Patients with this syndrome suffer various respiratory problems caused by the accumulation of particles in the lungs; males are typically sterile because of the loss of sperm motility, and c.50% have an alimentary tract that is a mirror image of the usual pattern (situs inversus) - i.e. it rotates in the opposite direction during early development (p. 1257).
MICROVILLI Microvilli are finger-like cell surface extensions usually c.0.1 µm in diameter and up to 2 µm long (Fig. 2.19). When arranged in a regular parallel series, they constitute a striated border, as typified by the absorptive surfaces of the epithelial enterocytes of the small intestine. When they are less regular, as in the gallbladder epithelium and proximal kidney tubules, the term brush border is used.
Figure 2.19 Microvilli sectioned longitudinally in the striated border of two adjacent intestinal absorptive cells with interlocking lateral plasma membranes (below, centre). Actin filaments fill the cores of the microvilli and insert into a terminal web of actin filaments in the apical cytoplasm.
Microvilli are covered by plasma membrane, and supported internally by closely packed bundles of actin microfilaments linked by cross-bridges of the actinbundling proteins, villin and fimbrin. Other bridges composed of myosin I and calmodulin connect the microfilaments to the plasma membrane. The microfilament bundles of microvilli are embedded in the apical cytoplasm amongst a meshwork of transversely running microfilaments linked by spectrin to form the terminal web (Fig. 2.19). The web is anchored laterally to the zonula adherens. Myosin is also found in the terminal web, where it is believed to bind to the actin and so stiffen this part of the cell. At the apex of each microvillus, the free ends of microfilaments are inserted into a dense mass that includes the protein, !-actinin. page 20 page 21
Microvilli greatly increase the area of cell surface (up to 40 times), particularly at sites of active absorption. In the small intestine, they have a very thick cell coat or glycocalyx, which reflects the presence of integral membrane glycoproteins, including enzymes concerned with digestion and absorption. Irregular microvilli, filopodia, are also found on the surfaces of many types of cell, particularly of free macrophages and fibroblasts, where they may be associated with phagocytosis and cell motility. Long, regular microvilli are called stereocilia, an early misnomer, as they are not motile and lack microtubules. They are found on cochlear and vestibular receptor cells (p. 663), where they act as sensory transducers, and also in the absorptive epithelium of the epididymis (p. 1307).
Nucleus The nucleus (Figs 2.1, 2.2) is generally the largest intracellular structure and is usually spherical or ellipsoid in shape, with a diameter of 3-10 µm. Histological stains used to identify nuclei in tissue sections mainly detect the acidic molecules of deoxyribonucleic acid (DNA), which are largely confined to the nucleus.
NUCLEAR MEMBRANE The nucleus is surrounded by two layers of membrane, each of which is a lipid bilayer, and which together form the nuclear membrane or envelope. The outer membrane layer and the lumen between the two layers are continuous with the rough endoplasmic reticulum. Like the rough endoplasmic reticulum, the outer membrane of the nuclear envelope is studded with ribosomes that are active in protein synthesis; the newly synthesized proteins pass into the perinuclear space between the two membrane layers. Intermediate filaments are associated with both the inner (nuclear) and outer (cytoplasmic) surfaces of the nuclear membrane. Within the nucleus they form a dense shell beneath the membrane, the nuclear lamina, consisting of specialized nuclear intermediate filaments called nuclear lamins. These cross each other at right angles to create a meshwork that covers the interior surface of the nuclear membrane. In so doing, they reinforce the nuclear membrane mechanically, determine the shape of the nucleus and anchor the ends of chromosomes. Condensed chromatin (heterochromatin) also tends to aggregate near the nuclear membrane during interphase. At the end of mitotic and meiotic prophase (p. 23), the lamin filaments disassemble, causing the nuclear membranes to vesiculate. At the end of anaphase, the lamins reattach to the chromosomes and create a new nuclear compartment around which the nuclear membranes reform. A network of filamentous proteins, the nuclear matrix, is also present throughout the nucleus. It is associated with newly replicated DNA and with genes that are being actively transcribed, and incorporates enzymes of the replication machinery. The transport of molecules between the nucleus and the cytoplasm is achieved by specialized nuclear pore structures that perforate the nuclear membrane (Fig.
2.20). They act as highly selective directional molecular filters, permitting proteins such as histones and gene regulatory proteins (which are synthesized in the cytoplasm but function in the nucleus) to enter the nucleus, and molecules that are synthesized in the nucleus but destined for the cytoplasm (e.g. ribosomal subunits, transfer RNAs and messenger RNAs), to leave the nucleus. Ultrastructurally, nuclear pores appear as disc-like structures with an outer diameter of c.130 nm and an inner pore with an effective diameter for free diffusion of 9 nm (Fig. 2.20B). The nuclear membrane of an active cell is bridged by up to 4000 such pores. The nuclear pore complex has an octagonal symmetry and is formed by an assembly of more than 50 proteins, the nucleoporins. The inner and outer nuclear membranes fuse around the pore complex (Fig. 2.20A). Transfer of lipids and proteins between the two is prevented, possibly by the luminal subunits of the pore. Nuclear pores are freely permeable to small molecules, ions and proteins up to about 17 kDa. Proteins of up to 60 kDa seem to be able to equilibrate slowly between the nucleus and cytoplasm through the pore, but larger proteins are normally excluded. However, certain proteins are selectively transported into the nucleus and some of these, such as the DNA polymerases, are very large. Proteins that are selectively transported into the nucleus possess a nuclear localization signal within their amino-acid sequence. This is recognized by cytoplasmic proteins that facilitate the docking of the proteins to be transported with the cytoplasmic surface of the pore. Subsequent translocation into the nucleus is energy dependent and requires the hydrolysis of GTP. The nuclear pore can open to a maximum of c.25 nm to permit the entry or exit of large, actively transported molecules. Steroid hormone receptors, which are gene regulatory proteins, associate with the cytoskeleton until they bind their ligands, when they dissociate from the cytoskeleton and are transported into the nucleus. Transport also occurs from the nucleus to the cytoplasm, e.g. RNAs synthesized in the nucleus are transported through nuclear pores into the cytoplasm.
CHROMATIN
page 21 page 22
Figure 2.20 A, Nuclear envelope with a nuclear pore (centre field) in transverse section, showing the continuity between the inner and outer phospholipid layers of the envelope on either side of the pore. The fine 'membrane' spanning the pore is formed by proteins of the pore complex. B, Nuclear pores seen 'en face' in a tangential section through the nuclear membrane. (A, by kind permission from Rose Watson, Cancer Research UK.)
DNA is organized within the nucleus in a DNA-protein complex known as chromatin. The protein constituents of chromatin are the histones and the nonhistone proteins. Non-histone proteins are an extremely heterogeneous group that includes DNA and RNA polymerases and gene regulatory proteins. Histones are the most abundant group of proteins in chromatin, primarily responsible for the packaging of chromosomal DNA into its primary level of organization, the nucleosome. There are five histone proteins: H1, H2A, H2B, H3 and H4; the last four combine in equal ratios to form a compact octameric nucleosome core. The DNA molecule (one per chromosome) winds 1.65 times around each nucleosome core, taking up 146 nucleotide pairs. This packaging organizes the DNA into a chromatin fibre 11 nm in diameter, and imparts to this form of chromatin the electron microscopic appearance of beads on a string, in which each bead is separated by a variable length of DNA, c.50 nucleotide pairs long. The nucleosome core region and one of the linker regions constitute the nucleosome proper, which is thus c.200 nucleotide pairs in length. However, chromatin rarely exists in this simple form and is usually packaged further into a 30 nm thick fibre, involving a single H1 histone per nucleosome, which interacts with both DNA and protein to impose a higher order of nucleosome packing. Usually, 30 nm fibres are further folded into loop-like domains, but individual loops are believed to decondense and extend during active transcription. In a typical interphase nucleus, euchromatin (nuclear regions that appear pale in appropriately stained tissue sections, or relatively electron-lucent in electron micrographs; Fig. 2.2) is likely to consist mainly of 30 nm fibres and loops, and contains the transcriptionally active genes. Transcriptionally active cells, such as most neurones, have nuclei that are predominantly euchromatic and often described as 'open face' nuclei. Heterochromatin (nuclear regions that appear dark in appropriately stained tissue sections or electron-dense in electron micrographs) is characteristically located mainly around the periphery of the nucleus, except over the nuclear pores, and around the nucleolus (Fig. 2.2). It is a highly compacted form of chromatin, containing additional proteins; its higher order packaging is poorly understood. Heterochromatin includes non-coding regions of DNA, such as centromeric and telomeric regions, which are known as constitutive heterochromatin. DNA that is inactivated (becoming resistant to transcription) in some cells as they differentiate during development or cell maturation contributes to heterochromatin, and is known as facultative heterochromatin. The inactive X chromosome in females is an example of facultative heterochromatin and can be identified in the light microscope as the deeply staining Barr body (drumstick chromosome) that
projects from the nuclear periphery. In transcriptionally inactive cells, chromatin is predominantly in the condensed, heterochromatic state, and may comprise as much as 90% of the total. Examples of such cells are mature neutrophil leukocytes (in which the condensation of chromatin induces the formation of a multilobed, densely staining nucleus), and the highly condensed nuclei of orthochromatic erythroblasts (late-stage erythrocyte precursors). In most mature cells, a mixture of the two occurs, indicating that only a proportion of the DNA is being transcribed. A particular instance of this is seen in the mature B lymphocyte (plasma cell), in which much of the chromatin is in the condensed condition and is arranged in regular masses around the perimeter of the nucleus, producing the so-called 'clock-face' nucleus (Fig. 2.2). Although this cell is actively transcribing, much of its protein synthesis is of a single immunoglobulin type, and consequently much of its genome is in an inactive state. During mitosis, the chromatin is further condensed to form the much shortened chromosomes characteristic of metaphase. This shortening is achieved through further levels of close packing of the chromatin, and is an energy dependent process involving proteins known as condensins. Progressive folding of the chromosomal DNA by interactions with specific proteins can reduce c.5 cm of chromosomal DNA by 10,000 fold, to a length of c.5 µm in the mitotic chromosome.
CHROMOSOMES AND KARYOTYPES The nuclear DNA of eukaryotic cells is organized into linear units called chromosomes. The DNA in a normal human diploid cell contains 6 $ 109 nucleotide pairs organized in the form of 46 chromosomes (44 autosomes and 2 sex chromosomes). The largest human chromosome (number 1) contains c.2.5 $ 108 nucleotide pairs, and the smallest (the Y chromosome) c.5 $ 107 nucleotide pairs. Each chromosomal DNA molecule contains a number of specialized nucleotide sequences that are associated with its maintenance. One is the centromere. During mitosis, a disc-shaped structure composed of a complex array of proteins, the kinetochore, associates with the centromeric region of DNA in order to attach it to the microtubular spindle. Another sequence, the telomere, defines the end of each chromosomal DNA molecule. Telomeres consist of tandem repeats of a short sequence enriched in guanosine nucleotides. They are not replicated by the same DNA polymerase as the rest of the chromosome, but by a specific enzyme called telomerase. The number of tandem repeats of the telomeric DNA sequence varies. It appears to shorten with successive cell divisions, because telomerase activity reduces or is absent in differentiated cells with a finite lifespan. It is believed that this mechanism regulates cell senescence and protects against proliferative disorders, including cancer.
CLASSIFICATION OF HUMAN CHROMOSOMES A number of genetic abnormalities can be directly related to the chromosomal pattern. The characterization or karyotyping of chromosome number and structure is therefore of considerable diagnostic importance. The identifying features of individual chromosomes are most easily seen during metaphase, although prophase chromosomes can be used for more detailed analyses. Lymphocytes separated from blood samples, or cells taken from other tissues, are used as a source of chromosomes. Diagnosis of fetal chromosome patterns is generally carried out on samples of amniotic fluid containing fetal cells aspirated from the uterus by amniocentesis, or on a small piece of chorionic villus tissue removed from the placenta. Whatever their origin, the cells are cultured in vitro
and stimulated to divide by treatment with agents that stimulate cell division. Mitosis is interrupted at metaphase with spindle inhibitors. The chromosomes are dispersed by first causing the cells to swell in a hypotonic solution, then the cells are gently fixed and mechanically ruptured on a slide to spread the chromosomes. They are subsequently stained in various ways to allow the identification of individual chromosomes by size, shape and distribution of stain (Fig. 2.21). General techniques show the obvious landmarks, e.g. lengths of arms and positions of constrictions. Banding techniques demonstrate differential staining patterns, characteristic for each chromosome type. Fluorescence staining with quinacrine mustard and related compounds produces Q bands, and Giemsa staining (after treatment that partially denatures the chromatin) gives G bands (Fig. 2.21A). Other less widely used methods include: reverse-Giemsa staining, in which the light and dark areas are reversed (R bands); the staining of constitutive heterochromatin with silver salts (C-banding); T-banding to stain the ends (telomeres) of chromosomes. Collectively, these methods permit the classification of chromosomes into numbered autosomal pairs in order of decreasing size, from 1 to 22 plus the sex chromosomes. Group 1-3 (A) 4-5 (B) 6-12 + X (C) 13-15 (D) 16-18 (E) 19-20 (F) 21-22 + Y (G)
Features Large metacentric chromosomes Large submetacentric chromosomes Metacentrics of medium size Medium-sized acrocentrics with satellites Shorter metacentrics (16) or submetacentrics (17,18) Shortest metacentrics Short acrocentrics; 21, 22 with satellites, Y without
A summary of the major classes of chromosomes is given below: Methodological advances in banding techniques improved the recognition of abnormal chromosome patterns. The use of in-situ hybridization with fluorescent DNA probes specific for each chromosome (Fig. 2.21B) permits the identification of even very small abnormalities.
NUCLEOLUS page 22 page 23
Figure 2.21 Chromosomes from normal males, arranged as karyotypes. A, G-banded preparation; B, preparation stained by multiplex fluorescence in-situ hybridization to identify each chromosome. (By kind permission from Dr Denise Sheer, Cancer Research UK.)
Nucleoli are a prominent feature of an interphase nucleus (Fig. 2.2). They are the site of most of the synthesis of rRNA and assembly of ribosome subunits. Ultrastructurally, the nucleolus appears as a pale fibrillar region (non-transcribed DNA), containing dense fibrillar cores (sites of rRNA gene transcription) and granular regions (sites of ribosome subunit assembly) within a diffuse nucleolar matrix. Five pairs of chromosomes carry rRNA genes organized in clusters of tandemly repeated units on each chromosome. Each rRNA unit is transcribed individually and encodes the 28S, 18S and 5.8S rRNA molecules. During mitosis the nucleolus breaks down. It reforms after telophase, in a process initiated by the onset of transcription in nucleolar organizing centres on each chromosome. The 28S, 18S and 5.8S rRNA molecules are assembled into their ribosomal subunits in the granular region of the nucleolus together with the 5S rRNA, which is not synthesized in the nucleolus. The newly formed ribosomal subunits are then translocated to the cytoplasm through the nuclear pores.
© 2008 Elsevier
CELL STRUCTURE GENERAL CHARACTERISTICS OF CELLS Most cells lie within the size range 5-50 µm in diameter: e.g. resting lymphocytes are c.6 µm across, red blood cells c.7.5 µm and columnar epithelial cells are c.20 µm tall and 10 µm wide. Some cells are much larger than this: e.g. megakaryocytes of the bone marrow are more than 200 µm in diameter. Large neurones and skeletal muscle cells have relatively enormous volumes because of their extended shapes, some of the former being over 1 metre in length. Cell size is limited by rates of diffusion, either that of material entering or leaving cells, or of diffusion within them. Diffusion can be much accelerated by processes of active transport across membranes and also directed by transport mechanisms within the cell. Motility is a characteristic of most cells, in the form of movements of cytoplasm or specific organelles from one part of the cell to another. It also includes: the extension of parts of the cell surface such as pseudopodia, membrane ruffles, filopodia and microvilli; locomotion of entire cells as in the amoeboid migration of tissue macrophages; the beating of flagella or cilia to move the cell (e.g. in spermatozoa) or fluids overlying it (e.g. in respiratory epithelium); cell division and muscle contraction. Cell movements are also involved in the uptake of materials from their environment (endocytosis, phagocytosis) and the passage of large molecular complexes out of cells (exocytosis, secretion). The shapes of cells vary widely depending on their interactions with each other, their extracellular environment and internal structures. Their surfaces are often highly folded when absorptive or transport functions take place across their boundaries. According to the location of absorptive or transport functions, apical microvilli or basolateral infoldings create a large surface area for transport or diffusion. Cells rarely operate independently of each other and commonly form aggregates by adhesion, often assisted by specialized intercellular junctions. They may also communicate with each other either by releasing and detecting molecular signals that diffuse across intercellular spaces, or more rapidly by membrane contact, which often involves small, transient, transmembrane channels. Cohesive groups of cells constitute tissues and more complex assemblies of tissues form functional systems or organs.
CELLULAR ORGANIZATION Cytoplasm is contained within a limiting plasma membrane. All cells except mature red blood cells also contain a nucleus that is surrounded by a nuclear membrane (Figs 2.1, 2.2). The nucleus includes the genome of the cell contained within the chromosomes, and the nucleolus. The cytoplasm contains several systems of organelles. These include a series of membrane-bound structures that form separate compartments within the cytoplasm, such as rough and smooth endoplasmic reticulum, Golgi apparatus, lysosomes, peroxisomes, mitochondria and vesicles for transport, secretion and storage of cellular components. There are also structures that lie free in the non-membranous, cytosolic compartment. They include ribosomes and several filamentous protein structures known collectively as the cytoskeleton. The cytoskeleton determines general cell shape and supports specialized extensions of the cell surface (microvilli, cilia, flagella). It is involved in the assembly of new filamentous organelles (e.g. centrioles) and controls internal movements of the cytoplasm and cytoplasmic vesicles. The cytosol contains many soluble proteins, ions and metabolites. Cell domains
Figure 2.1 The main structural components and internal organization of a generalized cell.
In polarized cells, particularly in epithelia, the cell is generally subdivided into domains that reflect the polarization of activities within the cell. The free surface, e.g. that facing the intestinal lumen or airway, is the apical surface, and its adjacent cytoplasm is the apical cell domain. This is where the cell interfaces with a specific body compartment (or, in the case of the epidermis, with the outside world). The apical surface is specialized to act as a barrier, restricting access of substances from this compartment to the rest of the body. Specific components are selectively absorbed from, or added to, the external compartment by the active processes, respectively, of active transport and endocytosis inwardly or exocytosis and secretion outwardly. The surface of the cell opposite to the apical surface is the basal surface, with its associated basal cell domain. In a single-layered epithelium, this surface is apposed to the basal lamina. The remaining surfaces are known as the lateral cell surfaces. In many instances the lateral and basal surfaces perform similar functions and the cellular domain is termed the basolateral domain. Cells actively transport substances, such as digested nutrients from the intestinal lumen or endocrine secretions, across their basal (or basolateral) surfaces into the subjacent connective tissue matrix and the blood capillaries within it. Dissolved non-polar gases (oxygen and carbon dioxide) diffuse freely between the cell and the bloodstream across the basolateral surface. page 5 page 6
Figure 2.2 A protein-synthesizing cell (immunoglobulin-secreting plasma cell) in connective tissue. The main ultrastructural features are a nucleus surrounded by a nuclear envelope and containing peripheral heterochromatin, central euchromatin and a nucleolus; and cytoplasm containing rough endoplasmic reticulum, mitochondria, Golgi apparatus, transport vesicles and small lysosomes.
Plasma membrane Cells are bounded by a distinct plasma membrane, which shares features with the system of internal membranes that compartmentalize the cytoplasm and surround the nucleus. They are all composed of lipids (mainly phospholipids, cholesterol and glycolipids) and proteins, in approximately equal ratios. Plasma membrane lipids form a layer two molecules thick, the lipid bilayer. The hydrophobic ends of each lipid molecule face the interior of the membrane and the hydrophilic ends face outwards. Most proteins are embedded within, or float in, the lipid bilayer as a fluid mosaic. Some proteins, because of extensive hydrophobic regions of their polypeptide chains, span the entire width of the membrane (transmembrane proteins), whereas others are only superficially attached to the bilayer by lipid groups. Both are integral (intrinsic) membrane proteins, as distinct from peripheral (extrinsic) membrane proteins, which are membrane-bound only through their association with other proteins. Carbohydrates in the form of oligosaccharides and polysaccharides are bound either to proteins (glycoproteins) or to lipids (glycolipids), and project mainly into the extracellular domain. Combinations of biochemical, biophysical and biological techniques have revealed that lipids are not homogenously distributed in membranes, but that some are organized into microdomains in the bilayer, called 'detergent-resistant membranes' or lipid 'rafts', rich in sphingomyelin and cholesterol (Morris et al 2003). The ability of select subsets of proteins to partition into different lipid microdomains has profound effects on their function, e.g. in T-cell receptor and neurotrophin signalling. The highly organized environment of the domains provides a signalling, trafficking and membrane fusion environment very different from that found in the disorganized fluid mosaic membrane.
Figure 2.3 The molecular organization of the plasma membrane, according to the fluid mosaic model of membrane structure. Intrinsic or integral membrane proteins include diffusion or transport channel complexes, receptor proteins and adhesion molecules. These may span the thickness of the membrane (transmembrane proteins) and can have both extracellular and cytoplasmic domains. Transmembrane proteins have hydrophobic zones, which cross the phospholipid bilayer and allow the protein to 'float' in the plane of the membrane. Some proteins are restricted in their freedom of movement where their cytoplasmic domains are tethered to the cytoskeleton.
page 6
page 7
Figure 2.4 The plasma membrane covering microvilli on absorptive epithelial cells in the small intestine. The lipid bilayer is clearly seen at this high magnification, as is the cell coat or glycocalyx projecting into extracellular space (the gut lumen, right) as an outer fuzzy layer. (By permission from Young B, Heath JW 2000 Wheater's Functional Histology. Edinburgh: Churchill Livingstone.)
In the electron microscope, membranes fixed and contrasted by heavy metals such as osmium appear in section as two densely stained layers separated by an electron-translucent zone - the classic unit membrane (Figs 2.3, 2.4). The total thickness is c.5 nm. Freeze-fracture cleavage planes usually pass along the midline of each membrane, where the hydrophobic tails of phospholipids meet. This technique has also demonstrated intramembranous particles embedded in the lipid bilayer; these are in the 5-15 nm range and in most cases represent large transmembrane protein molecules or complexes of molecules. Intramembranous particles are distributed asymmetrically between the two halfmembranes, usually adhering more to one face than to the other. In plasma membranes, the inner or protoplasmic (cytoplasmic) half-membrane carries most particles, seen on its surface facing the exterior (the P face). Where they have been identified, particles usually represent channels for the transmembrane passage of ions or molecules. Biophysical measurements show the lipid bilayer to be highly fluid, allowing diffusion in the plane of the membrane. Thus proteins are able to move freely in such planes unless anchored from within the cell. Membranes in general, and the plasma membrane in particular, form boundaries selectively limiting diffusion and creating physiologically distinct compartments. Lipid bilayers are impermeable to hydrophilic solutes and ions and so membranes actively control the passage of ions and small organic molecules such as nutrients, through the activity of membrane transport proteins. However, lipid-soluble substances can pass directly through the membrane so that, for example, steroid hormones enter the cytoplasm freely. Their receptor proteins are either cytosolic or nuclear, rather than being located on the cell surface. Plasma membranes are able to generate electrochemical gradients and potential differences by selective ion transport, and actively take up or export small molecules by energy dependent processes. They also provide surfaces for the attachment of enzymes, sites for the receptors of external signals, including hormones and other ligands, and sites for the recognition and attachment of other cells. Internally, plasma membranes can act as points of attachment for intracellular structures, in particular those concerned with motility and other cytoskeletal functions. Cell membranes are synthesized by the rough endoplasmic reticulum in conjunction with the Golgi apparatus.
THE CELL COAT (GLYCOCALYX) The plasma membrane differs structurally from internal membranes in that it possesses an external, diffuse, carbohydrate-rich coat, the cell coat or glycocalyx. The cell coat forms an integral part of the plasma membrane, projecting as a diffusely filamentous layer 2-20 nm or more from the lipoprotein surface (Fig. 2.4). The overall thickness of the plasma membrane is therefore variable, but is typically 8-10 nm. The cell coat is composed of the carbohydrate portions of glycoproteins and glycolipids embedded in the plasma membrane (Fig. 2.3). The precise composition of the glycocalyx varies with cell type: many tissue and cell type-specific antigens are located in the coat, including the major histocompatibility antigen systems and, in the case of erythrocytes, blood group antigens. It also contains adhesion molecules, which enable cells to adhere selectively to other cells or to the extracellular matrix. They have important roles in maintaining the integrity of tissues and in a wide range of dynamic cellular
processes, e.g. the formation of intercommunicating neural networks in the developing nervous system and the extravasation of leukocytes. Cells tend to repel each other because of the predominance of negatively charged carbohydrates at cell surfaces. There is consequently a distance of at least 20 nm between the plasma membranes of adjacent cells, other than at specialized junctions.
CELL SURFACE CONTACTS The plasma membrane is the surface which establishes contact with other cells and with structural components of extracellular matrices. These contacts may have a predominantly adhesive role, or initiate instructive signals within and between cells, or both; they frequently affect the behaviour of cells. Structurally, there are two main classes of contact, both associated with cell adhesion molecules. One class is associated with specializations at discrete regions of the cell surface that are ultrastructurally distinct. These are described on page 7. The second, general, class of adhesive contact has no obvious associated ultrastructural features.
GENERAL ADHESIVE CONTACTS One class of transmembrane or membrane-anchored glycoproteins that project externally from the plasma membrane, and which form adhesive contacts, are the cell adhesion molecules. There are a number of molecular subgroups, which are broadly divisible on the basis of their calcium dependence. Calcium-dependent adhesion molecules
Cadherins, selectins and integrins are calcium-dependent adhesion molecules. Cadherins are transmembrane proteins, with five heavily glycosylated external domains. They are responsible for strong general intercellular adhesion, as well as being components of some specialized adhesive contacts, and are attached by linker proteins (catenins) at their cytoplasmic ends to underlying cytoskeletal fibres (either actin or intermediate filaments). Different cell types possess different members of the cadherin family, e.g. N-cadherins in nervous tissue, E-cadherins in epithelia, and P-cadherins in the placenta. These molecules bind to those of the same type in other cells (homophilic binding), so that cells of the same class adhere to each other preferentially, forming tissue aggregates or layers, as in epithelia. Selectins are found on leukocytes, platelets and vascular endothelial cells. They are transmembrane lectin glycoproteins that can bind with low affinity to the carbohydrate groups on other cell surfaces to permit movement between the two, e.g. the rolling adhesion of leukocytes on the walls of blood vessels (p. 146). They function cooperatively in sequence with integrins, which strengthen the selectin adhesion. Integrins are glycoproteins that typically mediate adhesion between cells and extracellular matrix components such as fibronectin, collagen, laminin. They integrate interactions between the matrix and the cell cytoskeleton to which they are linked, and so facilitate cell migration within the matrix. An integrin molecule is formed of two subunits (! and "), each of which has several subtypes. Combinations of alternative subunits provide more than 20 integrin heterodimers, each one directed to a particular extracellular molecule, although there is considerable overlap in specificity. Some integrins depend for their binding on magnesium, rather than calcium. Calcium-independent adhesion molecules
The best known calcium-independent adhesion molecules are glycoproteins that have external domains related to immunoglobulin molecules. Most are transmembrane proteins. Some are entirely external, either attached to the
plasma membrane by a glycosylphosphatidylinositol anchor, or secreted as soluble components of the extracellular matrix. Different types are expressed in different tissues. Neural cell adhesion molecules are found on a number of cell types, but are expressed widely by neural cells. Intercellular adhesion molecules are expressed on vascular endothelial cells. Cell adhesion molecule binding is predominantly homophilic, although some, e.g. intercellular adhesion molecules, use a heterophilic mechanism and can bind to integrins. For further information on all aspects of cell adhesion molecules and intercellular contacts, see Alberts et al (2002).
SPECIALIZED ADHESIVE CONTACTS Specialized adhesive contacts, some of which mediate activities other than simple mechanical cohesion, are localized regions of the cell surface with particular ultrastructural characteristics. Three major classes exist: occluding, adhesive and communicating junctions (Fig. 2.5). Occluding junctions (tight junctions, zonula occludens)
Occluding junctions create diffusion barriers in continuous layers of cells, including epithelia, mesothelia and endothelia, and prevent the passage of materials across the cellular layer through intercellular spaces. They form a continuous belt (zonula) around the cell perimeter, near the apical surface in cuboidal or columnar epithelial cells. At a tight junction, the membranes of the adjacent cells come into contact, so that the gap between them is obliterated. Freeze-fracture electron microscopy shows that the contacts between the membranes lie along branching and anastomosing ridges formed by the incorporation of chains of intramembranous protein particles on the P face of the lipid bilayer (Fig. 2.5C). page 7 page 8
Figure 2.5 Intercellular junctions. A, The position of the apical junctional complex of epithelial cells (top) and the structures of the three elements of a junctional complex and of a gap junction (below). B, The ultrastructural appearance of an epithelial junctional complex; C, a freeze-fractured preparation showing the anastomotic network of contacts between adjacent cell membranes forming a tight junction; D, a freeze-fractured preparation showing the structure of a gap junction, with numerous channels (pores within connexons) clustered to form a plaque-like junctional region between adjacent plasma membranes; E, hemidesmosomes in the basal layer of the epidermis, contacting the underlying basal lamina. Dermal collagen fibrils are sectioned transversely below. (B, by permission from Young B, Heath JW 2000 Wheater's Functional Histology. Edinburgh: Churchill Livingstone; C, by kind permission from Dr Andrew Kent, King's College London; D, by kind permission from Professor Dieter Hülser, University of Stuttgart; E, by permission from the Company of Biologists Ltd, Cambridge, UK, and Frye M, Gardner C, Li ER, Arnold I, Watt FM 2003 Evidence that Myc activation depletes the epidermal stem cell compartment by modulating adhesive interactions with the local microenvironment. Development 130: 2793-2808.)
This arrangement ensures that substances can only pass through the layer of cells by diffusion or transport through their apical membranes and cytoplasm. The cells thus selectively modify the environment on either side of the layer. Occluding junctions also create regional differences in the plasma membranes of the cells in which they are found. For example, in epithelia, the composition of the apical plasma membrane differs from that of the basolateral regions, and this allows these regions to engage in functions such as directional transport of ions and uptake of macromolecules. Because tight junctions have high concentrations of fixed transmembrane proteins, they act as barriers to lateral diffusion of lipid and protein within membranes. The integrity of tight junctions is calcium-dependent. Cells can transiently alter the permeability of their tight junctions to increase passive paracellular transport in some circumstances. Adhesive junctions
Adhesive junctions include intercellular and cell-extracellular matrix contacts, where cells adhere strongly to each other or to adjacent matrix components. Intercellular contacts can be subdivided according to the extent and location of the contact. They all display a high concentration of cell adhesion molecules, which externally bind adjacent cells, and internally link to the cytoskeleton via intermediary proteins. Zonula adherens (intermediate junction) page 8 page 9
A zonula adherens is a continuous, belt-like zone of adhesion around the apical perimeters of epithelial, mesothelial and endothelial cells, parallel and just basal to the tight junction in epithelia. High concentrations of cadherins occur in this zone; their cytoplasmic ends are anchored via the proteins vinculin and !-actinin
to a layer of actin microfilaments. These junctions help to reinforce the intercellular attachment of the tight junction and prevent its mechanical disruption. The gap between cell surfaces is c.20 nm. Usually, no electron-dense material is observed within this intercellular space. Fascia adherens
A fascia adherens is similar to a zonula adherens, but is more limited in extent and forms a strip or patch of adhesion, e.g. between smooth muscle cells, in the intercalated discs of cardiac muscle cells (p. 152) and between glial cells and neurones. The junctions involve cadherins attached indirectly to actin filaments on the inner side of the membrane. Desmosomes (macula adherens)
Desmosomes are limited, plaque-like areas of particularly strong intercellular contact. They can be located anywhere on the cell surface. In epithelial cells, there may be a circumferential row of desmosomes parallel to the tight and intermediate junctional zones, an arrangement that forms the third, most basally situated, component of the epithelial apical junctional complex (Fig. 2.5A,B and p. 29). The intercellular gap is c.25 nm, is filled with electron-dense filamentous material running transversely across it and is also marked by a series of densely staining bands running parallel to the cell surfaces. Adhesion is mediated by calcium-dependent cadherins, desmoglein and desmocollin. Within the cells on either side, a cytoplasmic density underlies the plasma membrane and includes the anchor proteins desmoplakin and plakoglobin, into which the ends of intermediate filaments are inserted. The type of intermediate filament depends on cell type, e.g. cytokeratins are found in epithelia and desmin filaments in cardiac muscle cells. Desmosomes form strong anchorage points, likened to spot-welds, between cells subject to mechanical stress, e.g. in the spinous layer of the epidermis, where they are extremely numerous and large (p. 157). Hemidesmosomes
Hemidesmosomes are best known as anchoring junctions between the bases of epithelial cells and the basal lamina. Ultrastructurally, they resemble a singlesided desmosome, anchored on one side to the plasma membrane, and on the other to the basal lamina and adjacent collagen fibrils (Fig. 2.5E). On the cytoplasmic side of the membrane there is a dense coat into which cytokeratin filaments are inserted. Hemidesmosomes use integrins as their adhesion molecules, whereas desmosomes use cadherins. Less highly structured attachments with a similar arrangement exist between many other cell types and their surrounding matrix, e.g. between smooth muscle cells and their matrix fibrils, and between the ends of skeletal muscle cells and tendon fibres. The smaller, punctate adhesions resemble focal adhesion plaques. Focal adhesion plaques
Focal adhesion plaques are regions of local attachment between cells and the extracellular matrix. They are typically situated at or near the ends of actin filament bundles (stress fibres), which are anchored through intermediary proteins to the cytoplasmic domains of integrins. In turn, these are attached at their external ends to collagen or other filamentous structures in the extracellular matrix. They are usually short-lived: their formation and subsequent disruption are part of the motile behaviour of migratory cells. Gap junctions (communicating junctions)
Gap junctions resemble tight junctions in transverse section, but the two apposed lipid bilayers are separated by an apparent gap of 3 nm which is bridged by numerous transmembrane channels (connexons). Connexons are formed by a ring of six connexin proteins in each membrane. Their external surfaces meet
those of the adjacent cell in the middle. A minute central pore links one cell to the next (Fig. 2.5A). These channels may exist in small numbers, and this makes them difficult to detect structurally. However, they lower the transcellular electrical resistance and so can be detected by microelectrodes. Larger assemblies of many thousands of channels are often packed in hexagonal arrays (Fig. 2.5D). Such junctions form limited attachment plaques rather than continuous zones, and so allow free passage of substances within the adjacent intercellular space, unlike tight junctions. They occur in numerous tissues including the liver, epidermis, pancreatic islet cells, connective tissues, cardiac muscle and smooth muscle, and are also common in embryonic tissues. In the central nervous system, they are found in the ependyma and between neuroglial cells, and they form electrical synapses between some types of neurone, although this is rare in humans. Although gap junctions form diffusion channels between cells, the size of their apertures limits diffusion to small molecules and ions (up to a molecular weight of about 1000 kDa). Thus they admit sodium, potassium and calcium ions, various second messenger components, and a number of metabolites, but they exclude messenger RNA and other macromolecules. In some excitable tissues (e.g. cardiac and smooth muscle), one cell can activate another electrically by current flow through gap junctions. Communicating junctions probably permit metabolic cooperation between groups of adjacent cells; the significance of this activity in embryogenesis, normal tissue function, homeostasis and repair is only beginning to be understood. Other types of junction
Chemical synapses and neuromuscular junctions are specialized areas of intercellular adhesion where neurotransmitters secreted from a neuronal terminal gain access to specialized receptor molecules on a recipient cell surface. They are described on pages 44 and 64, respectively.
CELL SIGNALLING Cellular systems in the body communicate with each other to coordinate and integrate their functions. This occurs through a variety of processes known collectively as cell signalling, in which a signalling molecule produced by one cell is detected by another, almost always by means of a specific receptor protein molecule. The recipient cell transduces the signal, which it most usually detects at the plasma membrane, into intracellular chemical messages that change cell behaviour. The signal may act over a long distance, as in endocrine signalling through the release of hormones into the bloodstream or neuronal synaptic signalling via electrical impulse transmission along axons (p. 44) and subsequent release of chemical transmitters of the signal at synapses (p. 44) or neuromuscular junctions (p. 64). A specialized variation of endocrine signalling (neurocrine signalling) occurs when neurones or paraneurones (e.g. chromaffin cells of the suprarenal medulla) secrete a hormone into the bloodstream. Alternatively, signalling may occur at short range through a paracrine mechanism, in which cells of one type release molecules into the interstitial fluid of the local environment, to be detected by nearby cells of a different type that express the specific receptor protein. Cells may generate and respond to the same signal. This is autocrine signalling, a phenomenon that reinforces the coordinated activities of a group of like cells, which respond together to a high concentration of a local signalling molecule. The most extreme form of short-distance signalling is contact-dependent signalling, where one cell responds to transmembrane proteins of an adjacent cell that bind to surface receptors in the responding cell membrane. This type of signalling is important during development and in immune
responses. These different types of signalling mechanism are illustrated in Fig. 2.6. For further reading, see Alberts et al (2002) and Pollard and Earnshaw (2002).
SIGNALLING MOLECULES AND THEIR RECEPTORS page 9 page 10
Figure 2.6 The different modes of cell-cell signalling.
The majority of signalling molecules (ligands) are hydrophilic. They cannot cross the plasma membrane of a recipient cell to effect changes intracellularly unless they first bind to a plasma membrane receptor protein. Ligands are mainly proteins, polypeptides or highly charged biogenic amines. They include: classic peptide hormones of the endocrine system (Ch. 9); cytokines, which are mainly of haemopoietic cell origin and involved in inflammatory responses and tissue remodelling, e.g. the interferons, interleukins, tumour necrosis factor, leukaemia inhibitory factor; polypeptide growth factors, e.g. the epidermal growth factor superfamily, nerve growth factor, platelet-derived growth factor, the fibroblast growth factor family, transforming growth factor beta and the insulin-like growth factors. Polypeptide growth factors are multifunctional molecules with more widespread actions and cellular sources than their names suggest. They and their receptors are commonly mutated or aberrantly expressed in certain cancers. The cancer-causing gene variant is termed a transforming oncogene and the normal (wild-type) version of the gene is a cellular oncogene or proto-oncogene. The activated receptor acts as a transducer to generate intracellular signals, which are either small diffusible second messengers (e.g. calcium, cyclic adenosine monophosphate or the plasma membrane lipid-soluble diacylglycerol), or larger protein complexes that amplify and relay the signal to target control systems. For further reading on growth factors and other signalling molecules, see Epstein (2003). Some signals are hydrophobic and able to cross the plasma membrane freely. Classic examples are the steroid hormones, thyroid hormones, retinoids and vitamin D. Steroids, for instance, enter cells non-selectively, but elicit a specific response only in those target cells which express specific cytoplasmic or nuclear receptors. Light stimuli also cross the plasma membranes of photoreceptor cells and interact intracellularly, at least in rod cells, with membrane-bound photosensitive receptor proteins. Hydrophobic ligands are transported in the bloodstream or interstitial fluids, generally bound to carrier proteins, and they often have a longer half-life and longer-lasting effects on their targets than do water-soluble ligands.
A separate group of signalling molecules that are able to cross the plasma membrane freely is typified by the gas, nitric oxide . The principal target of shortrange nitric oxide signalling is smooth muscle, which relaxes in response. Nitric oxide is released from vascular endothelium as a result of the action of autonomic nerves that supply the vessel wall. It causes local relaxation of smooth muscle and dilation of vessels. In the penis, this mechanism is responsible for penile erection. Nitric oxide is unusual among signalling molecules in having no specific receptor protein; instead, it acts directly on intracellular enzymes of the response pathway. Receptor proteins
There are c.20 different families of receptor proteins, each with several isoforms responding to different ligands. The great majority of these receptors are transmembrane proteins. Members of each family share structural features that indicate either shared ligand-binding characteristics in the extracellular domain or shared signal transduction properties in the cytoplasmic domain, or both. There is little relationship either between the nature of a ligand and the family of receptor proteins to which it binds and activates, or the signal transduction strategies by which an intracellular response is achieved. The same ligand may activate fundamentally different types of receptor in different cell types. Cell surface receptor proteins are generally grouped according to their linkage to one of three intracellular systems: ion channel-linked receptors; G-proteincoupled receptors; receptors that link to enzyme systems. Other receptors do not fit neatly into any of these categories. All the known G-protein-coupled receptors belong to a structural group of proteins that pass through the membrane seven times in a series of serpentine loops. These receptors are thus known as sevenpass transmembrane receptors or, because the transmembrane regions are formed from !-helical domains, as seven-helix receptors. The most well-known of this large group of phylogenetically ancient receptors are the odorant-binding proteins of the olfactory system, the light-sensitive receptor protein, rhodopsin, and many of the receptors for clinically useful drugs. A comprehensive list of receptor proteins, their activating ligands and examples of the resultant biological function, is given in Pollard and Earnshaw (2002).
TRANSPORT ACROSS CELL MEMBRANES Lipid bilayers are increasingly impermeable to molecules as they increase in size or hydrophilicity. Transport mechanisms are therefore required to carry essential polar molecules, including ions, nutrients, nucleotides and metabolites of various kinds, across the plasma membrane and into or out of membrane-bound intracellular compartments. Transport is facilitated by a variety of membrane transport proteins, each with specificity for a particular class of molecule, e.g. sugars. Transport proteins fall mainly into two major classes, channel proteins and carrier proteins. page 10 page 11
Channel proteins form aqueous pores in the membrane, which open and close under the regulation of intracellular signals, e.g. G-proteins, to allow the flux of solutes (usually inorganic ions) of specific size and charge. Transport through ion channels is always passive and ion flow through an open channel depends only on the ion concentration gradient and its electronic charge, and the potential difference across the membrane. These factors combine to produce an electrochemical gradient, which governs ion flux. Channel proteins are utilized most effectively by the excitable plasma membranes of nerve cells, where the resting membrane potential can change transiently from about -70 mV (negative inside the cell) to +50 mV (positive inside the cell) when stimulated by a neurotransmitter (as a result of the opening and subsequent closure of channels selectively permeable to sodium and potassium).
Carrier proteins bind their specific solutes, such as amino-acids, and transport them across the membrane through a series of conformational changes. This latter process is slower than ion transport through membrane channels. Transport by carrier proteins can occur either passively by simple diffusion, or actively against the electrochemical gradient of the solute. Active transport must therefore be coupled to a source of energy, such as ATP generation, or energy released by the coordinate movement of an ion down its electrochemical gradient. Linked transport can be in the same direction as the solute, in which case the carrier protein is described as a symporter, or in the opposite direction, when the carrier acts as an antiporter.
TRANSLOCATION OF PROTEINS ACROSS INTRACELLULAR MEMBRANES Proteins are generally synthesized on ribosomes in the cytosol or on the rough endoplasmic reticulum. A few are made on mitochondrial ribosomes. Once synthesized, many proteins remain in the cytosol, where they carry out their functions. Others, such as integral membrane proteins or proteins for secretion, are translocated across intracellular membranes for post-translational modification and targeting to their destinations. This is achieved by the signal sequence, an addressing system contained within the protein sequence of amino-acids, which is recognized by receptors or translocators in the appropriate membrane. Proteins are thus sorted by their signal sequence (or set of sequences that become spatially grouped as a signal patch when the protein folds into its tertiary configuration), so that they are recognized by and enter the correct intracellular membrane compartment.
EXOCYTOSIS AND ENDOCYTOSIS Secreted proteins, lipids, mucins, small molecules such as amines and other cellular products destined for export from the cell are transported to the plasma membrane in small vesicles released from the trans face of the Golgi apparatus. This pathway is either constitutive, in which transport and secretion occur more or less continuously, or it is regulated by external signals, as in the control of salivary secretion by autonomic neural stimulation. In regulated secretion, the secretory product is stored temporarily in membrane-bound secretory granules or vesicles. Exocytosis is achieved by fusion of the secretory vesicular membrane with the plasma membrane and release of the vesicle contents into the extracellular domain. In polarized cells, e.g. most epithelia, exocytosis occurs at the apical plasma membrane and the cells secrete into a duct lumen or onto a free surface such as the lining of the stomach. In hepatocytes, bile is secreted across a very small area of plasma membrane forming the wall of the bile canaliculus (p. 1222). This region is defined as the apical plasma membrane, and is the site of exocrine secretion (p. 34), whereas secretion of hepatocyte plasma proteins into the bloodstream is targeted to the basolateral surfaces facing the sinusoids. Packaging of different secretory products into appropriate vesicles takes place in the trans-Golgi network. Delivery of secretory vesicles to their correct plasma membrane domains is achieved by sorting sequences in the cytoplasmic tails of vesicular membrane proteins.
Figure 2.7 Transcytotic vesicles (arrows) shuttle in both directions between blood plasma and extracellular fluid, across the cytoplasm of these endothelial cells in the wall of a capillary.
There are other mechanisms in which initial delivery of secretory products is less selective, but is followed by selective retention (or degradation) or reprocessing and redistribution by endosomes. Ultimately, secretory vesicles undergo docking, priming (to prepare the vesicle for a regulatory signal, where secretion is regulation-dependent) and fusion with the plasma membrane to release their contents. The process of exocytosis also delivers integral membrane components to the cell surface in the normal turnover and recycling of the plasma membrane. However, excess plasma membrane generated by vesicle fusion during exocytosis is rapidly removed by concurrent endocytosis. The process of endocytosis involves the internalization of vesicles derived from the plasma membrane. The vesicles may contain: engulfed fluids and solutes from the extracellular interstitial fluid (pinocytosis); larger macromolecules, often bound to surface receptors (receptor-mediated endocytosis); particulate matter, including microorganisms or cellular debris (phagocytosis). Pinocytosis generally involves small fluid-filled vesicles and is a marked property of capillary endothelium, e.g. where vesicles containing nutrients and oxygen dissolved in blood plasma are transported from the vascular lumen to the endothelial basal plasma membrane (Fig. 2.7). Interstitial fluid containing dissolved carbon dioxide is also taken up by pinocytosis for simultaneous transportation across the endothelial cell wall in the opposite direction, for release into the bloodstream by exocytosis. This shuttling of pinocytotic vesicles is also termed transcytosis. Larger volumes of fluid are engulfed by dendritic cells, e.g. in the process of sampling interstitial fluids by macropinocytosis in immune surveillance for antigens (p. 81). Interstitial fluid is inevitably taken up during receptor-mediated endocytosis when ligands are internalized. Receptor-mediated endocytosis, also known as clathrin-dependent endocytosis, is initiated at specialized regions of the plasma membrane known as clathrin-coated pits. Clathrin is a protein that cross-links adjacent adaptor protein (adaptin)
complexes to form a basket-like structure, bending the membrane inwards into a hemisphere. Much, but not all, fluid-phase pinocytosis also utilizes clathrin-coated pits. Ligands such as the iron-transporting protein, transferrin, and the cholesterol-transporting low-density lipoprotein bind to their receptors, which cluster in clathrin-coated pits through an interaction with adaptins. The pits then invaginate and pinch off from the plasma membrane, internalizing both receptor and ligand. The processing of endocytic vesicles and their contents is described on page 14. For further details of the molecular mechanisms of endocytosis, see Alberts et al (2002) or Pollard and Earnshaw (2002).
PHAGOCYTOSIS page 11 page 12
Phagocytosis is a property of many cell types, but is most efficient in cells specialized for this activity. The professional phagocytes of the body belong to the monocyte lineage of haemopoietic cells, in particular the tissue macrophages (p. 80). Other effective phagocytes are neutrophil granulocytes and most dendritic cells (p. 81), which are also of haemopoietic origin. Phagocytosis plays an important part in the immune defence system of the body, in which the amoeboid process of ingestion of organisms for nutrition has evolved into a mechanism for the clearance of microorganisms invading the body. Macrophages also ingest particulate material including inorganic matter, such as inhaled dust particles, in addition to debris from dead cells and protein aggregates such as immune complexes in the blood, airways, interstitial spaces and connective tissue matrices. Phagocytosis is a triggered process, initiated when a phagocytic cell binds to a particle or organism, often through a process of molecular recognition. Typically, a pathogenic microorganism may first be coated by antibodies, which are bound in turn by receptors for the Fc portion of the antibody molecule expressed by macrophages and neutrophils; in this way the microorganism is attached to the cell. This triggers the production of large pseudopodia, which engulf the organism and internalize it, as their pseudopod tips fuse together. The process appears to depend on actin-myosin-based cellular motility and, unlike receptor-mediated endocytosis, it is energy dependent. Phagosomes thus formed are as large as the body they engulf and can be a considerable proportion of the volume of the phagocytic cell. Inside the cell, the phagosome fuses with lysosomes, which degrade its contents.
Cytoplasm ENDOPLASMIC RETICULUM Endoplasmic reticulum is a system of interconnecting membrane-lined channels within the cytoplasm (Fig. 2.8). These channels take various forms, including cisternae (flattened sacs), tubules and vesicles. The membranes divide the cytoplasm into two major compartments. The intramembranous compartment includes the space where secretory products are stored or transported to the Golgi complex and cell exterior. The extramembranous cytosol is made up of the colloidal proteins such as enzymes, carbohydrates and small protein molecules, together with ribosomes and ribonucleic acids, and elements of the cytoskeleton. Structurally, the channel system can be divided into rough or granular endoplasmic reticulum, which has ribosomes attached to its outer cytosolic surface, and smooth or agranular endoplasmic reticulum, which lacks ribosomes.
ROUGH ENDOPLASMIC RETICULUM The rough endoplasmic reticulum, studded with ribosomes, is a site of protein synthesis (Fig. 2.8A). Most proteins pass through its membranes and accumulate within its cisternae, although some integral membrane proteins, e.g. plasma
membrane receptors, are inserted into the rough endoplasmic reticulum membrane, where they remain. After passage from the rough endoplasmic reticulum, proteins remain in membrane-bound cytoplasmic organelles such as lysosomes, become incorporated into new plasma membrane, or are secreted by the cell. Some carbohydrates are also synthesized by enzymes within the cavities of the rough endoplasmic reticulum and may be attached to newly formed protein (glycosylation). Vesicles are budded off from the rough endoplasmic reticulum for transport to the Golgi as part of the protein-targeting mechanism of the cell.
Figure 2.8 The endoplasmic reticulum. A, Rough endoplasmic reticulum with attached ribosomes; B, smooth endoplasmic reticulum with associated vesicles. The dense particles are glycogen granules. (By kind permission from Rose Watson, Cancer Research UK.)
SMOOTH ENDOPLASMIC RETICULUM The smooth endoplasmic reticulum (Fig. 2.8B) is associated with carbohydrate metabolism and many other metabolic processes, including detoxification and synthesis of lipids, cholesterol and other steroids. The membranes of the smooth endoplasmic reticulum serve as surfaces for the attachment of many enzyme systems, e.g. the enzyme cytochrome P450, which is involved in important detoxification mechanisms and is thus accessible to its substrates, which are generally lipophilic. They also cooperate with the rough endoplasmic reticulum and the Golgi apparatus to synthesize new membranes; the protein, carbohydrate and lipid components are added in different structural compartments. Highly specialized types of endoplasmic reticulum are present in some cells. For example, in skeletal muscle cells, the smooth endoplasmic reticulum (sarcoplasmic reticulum) stores calcium ions, which are released into the cytosol
to initiate contraction after stimulation initiated by a motor neurone at the neuromuscular junction (p. 64).
RIBOSOMES Ribosomes are macromolecular machines that catalyse the synthesis of proteins from amino-acids. They are granules c.15 nm in diameter, composed of ribosomal RNA (rRNA) molecules assembled into two unequal subunits. A large number of proteins, mostly small and basic, are applied mainly to the surfaces of the subunit cores of RNA. The subunits can be separated by their sedimentation coefficients (S) in an ultracentrifuge, into larger 60S and smaller 40S components. These are associated with 73 different proteins (40 in the large subunit and 33 in the small), which have structural and enzymatic functions. Three small, highly convoluted rRNA strands (28S, 5.8S and 5S) make up the large subunit, and one strand (18S) is in the small subunit. Their synthesis and assembly into subunits takes place in the nucleolus, and includes association with ribosomal proteins translocated from their site of synthesis in the cytoplasm. The individual subunits are then transported into the cytoplasm, where they remain separate from each other when not actively synthesizing proteins. A typical cell contains millions of ribosomes. They may be solitary, relatively inactive structures, or may form groups (polyribosomes or polysomes) attached to messenger RNA (mRNA), which they translate during protein synthesis (Fig. 2.9). Polysomes may be attached to the membranes of the rough endoplasmic reticulum or may lie free in the cytosol, where they synthesize proteins for use outside the system of membrane compartments, including enzymes of the cytosol and cytoskeletal proteins. Some of the cytosolic products include proteins that can be inserted directly into (or through) membranes of selected organelles, such as mitochondria and peroxisomes. page 12 page 13
Figure 2.9 Ribosomes, distributed either singly, clustered as polyribosomes (polysomes), or attached to the rough endoplasmic reticulum (right). (By permission from Stevens A, Lowe JS 1996 Human Histology, 2nd edn. London: Mosby.)
In a mature polysome, all the attachment sites of the mRNA are occupied as ribosomes move along it, synthesizing protein according to its nucleic acid sequence. Consequently, the number of ribosomes in a polysome indicates the
length of the mRNA molecule and hence the size of the protein being made. The two subunits have separate roles in protein synthesis. The 40S subunit is the site of attachment and translation of mRNA. The 60S subunit is responsible for the release of the new protein and, where appropriate, attachment to the endoplasmic reticulum via an intermediate docking protein that directs the newly synthesized protein through the membrane into the cisternal space.
GOLGI APPARATUS (GOLGI COMPLEX) (Figs 2.10, 2.11)
Figure 2.10 Golgi apparatus in a fibroblast. Several Golgi stacks are present, each with convex cis- and concave trans-Golgi surfaces, and associated transport vesicles. The edge of the nucleus appears on the left. (By kind permission from Rose Watson, Cancer Research UK.)
page 13 page 14
Figure 2.11 The Golgi apparatus and its functional relationships with associated structures.
The Golgi apparatus is a distinct cytoplasmic region near the nucleus, and is particularly prominent in secretory cells when stained with silver or other metallic salts. The Golgi apparatus forms part of the pathway by which proteins synthesized in the rough endoplasmic reticulum undergo post-translational modification and are targeted to the cell surface for secretion or for storage in membranous vesicles. Ultra-structurally, the Golgi apparatus is a membranous organelle consisting of a stack of several flattened membranous cisternae, together with clusters of vesicles surrounding its surfaces. Seen in vertical section, it is often cup-shaped. Small transport vesicles from the rough endoplasmic reticulum, generated by a process of budding and pinching off, are received at one face of the Golgi stack, the convex cis-face (entry or forming surface). Here, they deliver their contents to the first cisterna in the series by membrane fusion. From the edges of this cisterna, the protein is transported to the next cisterna by vesicular budding and then fusion, and this process is repeated until the final cisterna at the concave trans face (exit or condensing surface) is reached. Here, larger vesicles are formed for delivery to other parts of the cell. In addition to these cisternae, there are other membranous structures that form an integral part of the Golgi apparatus, termed the cis-Golgi and trans-Golgi networks. The cis-Golgi network is a region of complex membranous channels interposed between the rough endoplasmic reticulum and the Golgi cis face (Golgi-rough endoplasmic reticulum complex), which receives and transmits vesicles in both directions. Its function is to select appropriate proteins synthesized on the rough endoplasmic reticulum for delivery by vesicles to the Golgi stack, while inappropriate proteins are shuttled back to the rough
endoplasmic reticulum. The trans-Golgi network, at the other side of the Golgi stack, is also a region of interconnected membrane channels engaged in protein sorting. Here, modified proteins processed in the Golgi cisternae are packaged selectively into vesicles and dispatched to different parts of the cell. The packaging depends on the detection, by the trans-Golgi network, of particular amino-acid signal sequences, leading to their enclosure in membranes of appropriate composition that will further modify their contents, e.g. by extracting water to concentrate them or by pumping in protons to acidify their contents. The membranes contain specific signal proteins, which may allocate them to microtubule-based transport pathways and allow them to dock with appropriate targets elsewhere in the cell, e.g. the plasma membrane in the case of secretory vesicles. Vesicle formation and budding at the trans-Golgi network involves the addition of clathrin on their external surface, to form coated pits. Within the Golgi stack proper, proteins undergo a series of sequential chemical modifications that started in the rough endoplasmic reticulum. These include: changes in glycosyl groups, e.g. removal of mannose, addition of N-acetyl glucosamine and sialic acid; sulphation of attached glycosaminoglycans; protein phosphorylation. Lipids formed in the endoplasmic reticulum are also routed for incorporation into vesicles. The role of the Golgi apparatus in the synthesis of primary lysosomes is a major activity in cells with abundant lysosomes, such as those with phagocytic roles. In glandular cells with an apical secretory zone, the Golgi apparatus lies between the secretory surface and the nucleus. In fibroblasts, there are two or more groups of Golgi stacks; up to 50 groups are found in liver cells. The Golgi apparatus is often closely associated with the centrosome (a region of the cell containing a centriole pair and related microtubules), reflecting a link with the microtubule-mediated vesicle transport system.
ENDOSOMES, LYSOSOMES AND PEROXISOMES The endosome system of vesicles originates in small endocytic vesicles or larger phagosomes taken up by the cell from the exterior. The system is linked functionally to a second series of membranous structures, the lysosomes. Lysosomes contain acid hydrolases, which process or degrade exogenous materials (heterophagy), and intracellular organelles that are exhausted, damaged or no longer required (autophagy). There is a continual exchange of vesicles between this system and the Golgi-rough endoplasmic reticulum complex, so that the endosomal/lysosomal system is provided with hydrolytic enzymes and the Golgi receives depleted vesicles for recharging. Once internalized, endocytic vesicles shed their coat of adaptin and clathrin, and fuse with a tubular cisterna termed an early endosome, where the receptor molecules release their bound ligands. Membrane and receptors from the early endosomes can be recycled to the cell surface as exocytic vesicles.
LATE ENDOSOMES After a brief period in the early endosomes, materials can be passed on to late endosomes, which are a more deeply placed set of tubules, vesicles or cisternae. Late endosomes receive lysosomal enzymes via vesicles (small lysosomes) transported from the Golgi apparatus. The pH of late endosomes is low (about 5.0) and this activates lysosomal acid hydrolases to degrade the endosomal contents. The products of hydrolysis are either passed through the membrane into the cytosol, or may be retained in the endosome. Late endosomes may grow considerably in size by vesicle fusion to form multivesicular bodies (Fig. 2.12), and the enzyme concentration may increase greatly to form the large, dense classic lysosomes described by de Duve. However, such large organelles do not
appear in all cells, perhaps because late endosomes often deal very rapidly with endocytosed material.
LYSOSOMES Lysosomes are dense, spheroidal, membrane-bound bodies 80-800 nm in diameter (Fig. 2.12), often with complex inclusions of material undergoing hydrolysis (secondary lysosomes). They contain acid hydrolases able to degrade a wide variety of substances. To date, more than 40 lysosomal enzymes have been described, including proteases, lipases, carbohydrases, esterases and nucleases. The enzymes are heavily glycosylated, and are maintained at a low pH by proton pumps in the lysosomal membranes.
page 14 page 15
Figure 2.12 The typical features of primary and secondary lysosomes in the cytoplasm of a liver cell. Primary lysosomes (Ly1) are homogeneous membrane-bound bodies, whereas secondary lysosomes (Ly2) are typically variable in density and content, and often difficult to distinguish from later-stage residual bodies. Note their size relative to mitochondria (M). A number of late endosomes (multivesicular bodies, MB) are also shown. (By permission from Young B, Heath JW 2000 Wheater's Functional Histology. Edinburgh: Churchill Livingstone.)
Lysosomes are numerous in actively phagocytic cells, e.g. macrophages and neutrophil granulocytes, in which lysosomes are responsible for destroying phagocytosed bacteria. In these cells, the phagosome containing the bacterium may fuse with several lysosomes. Lysosomes are also frequent in cells with a high turnover of organelles, e.g. exocrine gland cells and neurones. Effete organelles are targeted for demolition by a process that is not fully understood, but which results in engulfment of areas of cytoplasm, including entire organelles, in a membranous cisterna. The structure then fuses with lysosomes and the contents are rapidly degraded. Material that has been hydrolysed within late endosomes and lysosomes may be completely degraded to soluble products, e.g. amino-acids, which are recycled through metabolic pathways. However degradation is usually incomplete, and
some debris remains. A debris-laden vesicle is called a residual body, and may be passed to the cell surface, where it is ejected by exocytosis; alternatively, it may persist inside the cell as an inert residual body. Considerable numbers of residual bodies can accumulate in long-lived cells, often fusing to form larger dense vacuoles with complex lamellar inclusions. As their contents are often darkly pigmented, this may change the colour of the tissue, e.g. in neurones the end-product of lysosomal digestion, lipofuscin (neuromelanin or senility pigment), gives ageing brains a brownish-yellow colouration. Lysosomal enzymes may also be secreted - often as part of a process to alter the extracellular matrix, as in osteoclast erosion of bone (p. 92). Abnormal release of enzymes can cause tissue damage, as in certain types of arthritis. Some drugs, e.g. cortisone, can stabilize lysosomal membranes and may therefore inhibit many lysosomal activities, including the secretion of enzymes, and their fusion with phagocytic vesicles. Lysosomal storage diseases
If any of the lysosomal enzymes are defective because of gene mutations, the materials that they normally degrade will accumulate within late endosomes and lysosomes. Many such lysosomal storage diseases are known, e.g. Tay-Sachs disease, in which a faulty gangliosidase leads to the accumulation of glycolipid in neurones, causing death during childhood. In Hurler's syndrome, failure to metabolize certain glycosaminoglycans causes the accumulation of large amounts of matrix within connective tissue, which distorts growth of many parts of the body.
PEROXISOMES Peroxisomes are membrane-bound vacuoles c.0.5-0.15 µm across, present in all nucleated cell types. They often contain dense cores or crystalline interiors composed mainly of high concentrations of the enzyme urate oxidase. Large (0.5 µm) peroxisomes are particularly numerous in hepatocytes and kidney tubule cells. Peroxisomes are important in the oxidative detoxification of various substances taken into or produced within cells, including ethanol and formaldehyde . Oxidation is carried out by a number of enzymes, including Damino-acid oxidase and urate oxidase, which generate hydrogen peroxide as a source of molecular oxygen. Excess amounts of hydrogen peroxide are broken down by the enzyme, catalase. Peroxisomes also oxidize fatty acid chains by "oxidation. The formation of peroxisomes is unusual in that they appear to be derived by the growth and fission of previously existing peroxisomes. Their internal proteins are passed from the cytosol directly through channels in their membranes, rather than by packaging from the rough endoplasmic reticulum and Golgi apparatus. These features are also found in mitochondria, although peroxisomal proteins are coded for entirely in the nucleus. A genetic abnormality in the translocation of proteins into peroxisomes, leading to peroxisomal enzyme deficiencies, is seen in Zellweger syndrome, caused by a gene mutation in an integral membrane protein (peroxisome assembly factor-1). In homozygotes, this is usually fatal shortly after birth.
MITOCHONDRIA
Figure 2.13 A mitochondrion. The folded cristae project into the matrix from the inner mitochondrial membrane.
The mitochondrion is a membrane-bound organelle (Fig. 2.13). It is the principal source of chemical energy in most cells. Mitochondria are the site of the citric acid (Kreb's, tricarboxylic acid) cycle and the electron transport (cytochrome) pathway by which complex organic molecules are finally oxidized to carbon dioxide and water. This process provides the energy to drive the production of ATP from ADP and inorganic phosphate (oxidative phosphorylation). The various enzymes of the citric acid cycle are located in the mitochondrial matrix, whereas those of the cytochrome system and oxidative phosphorylation are localized chiefly in the inner mitochondrial membrane. The numbers of mitochondria in a particular cell reflect its general energy requirements; e.g. in hepatocytes there may be as many as 2000, whereas in resting lymphocytes there are usually very few. Mature erythrocytes lack mitochondria altogether. Cells with few mitochondria generally rely largely on glycolysis for their energy supplies. These include some very active cells, e.g. fast twitch skeletal muscle fibres, which are able to work rapidly, but for only a limited duration. Mitochondria appear in the light microscope as long thin threads, or alternatively as spherical or ellipsoid bodies in the cytoplasm of most cells, particularly those with a high metabolic rate, e.g. secretory cells in exocrine glands. In living cells, mitochondria constantly change shape and intracellular position; they multiply by growth and fission and may undergo fusion. In the electron microscope, mitochondria usually appear as elliptical bodies 0.52.0 µm long. Each mitochondrion is lined by an outer and an inner unit membrane, separated by a variable gap termed the intermembrane space. The lumen is surrounded by the inner membrane and contains the mitochondrial matrix. The outer membrane is smooth and sometimes attached to other organelles, particularly microtubules. The inner membrane is deeply folded to form incomplete transverse or longitudinal tubular invaginations, cristae, which create a relatively large surface area of membrane. Mitochondrial shape, and the shape and organization of the cristae, vary with the cell type. Cristae are most numerous and complex in cells with a high metabolic rate, e.g. cardiac muscle cells. The permeabilities of the two mitochondrial membranes differ considerably: the outer membrane is freely permeable to many substances because of the presence of large non-specific channels formed by proteins (porins), whereas the inner membrane is permeable to only a narrow range of molecules. The presence of cardiolipin, a phospholipid, in the inner membrane may contribute to this relative impermeability.
The mitochondrial matrix is an aqueous environment. It contains a variety of enzymes, and strands of mitochondrial DNA with the capacity for transcription and translation of a unique set of mitochondrial genes (mitochondrial mRNAs and transfer RNAs, mitochondrial ribosomes with rRNAs). The DNA forms a closed loop, c.5 µm across; several identical copies are present in each mitochondrion. The ratio between its bases differs from that of nuclear DNA, and the RNA sequences also differ in the precise genetic code used in protein synthesis. At least 13 respiratory chain enzymes of the matrix and inner membrane are encoded by the small number of genes along the mitochondrial DNA. The great majority of mitochondrial proteins are encoded by nuclear genes and made in the cytosol, then inserted through special channels in the mitochondrial membranes to reach their destinations. Their membrane lipids are synthesized in the endoplasmic reticulum. page 15 page 16
Mitochondrial ribosomes are smaller and quite distinct from those of the rest of the cell. Mitochondrial ribosomes and nucleic acids resemble those of bacteria. This similarity underpins the theory that mitochondrial ancestors were oxygenutilizing bacteria that existed in a symbiotic relationship with eukaryotic cells unable to metabolize the oxygen produced by early plants. As mitochondria are formed only from previously existing ones, it follows that all mitochondria in the body are descended from those in the cytoplasm of the fertilized ovum. It has also been shown that mitochondria are of maternal origin because the mitochondria of the sperm are not generally incorporated into the ovum at fertilization. Thus mitochondria (and mitochondrial genetic variations and mutations) are passed only through the female line. Mitochondria are distributed within a cell according to regional energy requirements, e.g. near the bases of cilia in ciliated epithelia, in the basal domain of the cells of proximal convoluted tubules in the renal cortex (where considerable active transport occurs) and around the proximal end of the flagellum in spermatozoa. They may be involved with tissue-specific metabolic reactions, e.g. various urea-forming enzymes in liver cell mitochondria. Moreover, a number of genetic diseases of mitochondria affect particular tissues exclusively, e.g. mitochondrial myopathies (skeletal muscle) and mitochondrial neuropathies (nervous tissue). For further information see Graff et al (2002).
CYTOSOLIC ORGANELLES The aqueous cytosol surrounds the membranous organelles described above. It also contains various non-membranous organelles, including free ribosomes, a system of filamentous proteins known as the cytoskeleton, and other inclusions, such as storage granules (e.g. glycogen) and lipid vacuoles.
LIPID VACUOLES Lipid vacuoles are spherical bodies of various sizes found within many cells, but are especially prominent in the adipocytes (lipocytes) of adipose connective tissue. They do not belong to the Golgi-related vacuolar system of the cell. They are not membrane bound, but are droplets of lipid suspended in the cytosol. In cells specialized for lipid storage the vacuoles reach 80 µm or more in diameter. Lipid vacuoles are often surrounded by cytoskeletal filaments that help to stabilize them within cells and to prevent their fusion with the membranes of other organelles, including the plasma membrane. They function as stores of chemical energy, thermal insulators and mechanical shock absorbers in adipocytes. In many cells, they may represent end-products of other metabolic pathways, e.g. in steroid-synthesizing cells, where they are a prominent feature of the cytoplasm. They may also be secreted, as in the alveolar epithelium of the lactating breast.
CYTOSKELETON The cytoskeleton is a system of filamentous intracellular proteins of different shapes and sizes that form a complex, often interconnected, network throughout the cytoplasm. It provides mechanical support, maintains cell shape and rigidity, and enables cells to adopt highly asymmetric or irregular profiles, e.g. in neurones. The cytoskeleton plays an important part in establishing structural polarity and different functional domains within a cell. It also provides mechanical support for projections from the cell surface such as microvilli and cilia, and anchors them into the cytoplasm. The cytoskeleton restricts specific organelles to particular cellular locations, e.g. the Golgi apparatus is near the nucleus and endoplasmic reticulum, and mitochondria are near sites of energy requirement. Most specifically, the cytoskeleton is concerned with motility, either within the cell (e.g. shuttling vesicles and macromolecules between cytoplasmic sites, or the movement of chromosomes during mitosis), or of the entire cell (e.g. in embryonic morphogenesis or the chemotactic migration of leukocytes). One of the most highly developed and specialized functions of the cytoskeleton is seen in the contractility of muscle cells.
Figure 2.14 Actin microfilaments present at high density in the cytoplasm of a smooth muscle cell. Cytoplasmic dense bodies (arrows) are points of attachment for the actin filaments.
The catalogue of cytoskeletal structural proteins is extensive and still increasing. The major filamentous structures found in non-muscle cells are microfilaments (actin), microtubules (tubulin), and intermediate filaments (varieties of cell specific intermediate filament proteins). Other important components are generally smaller proteins that bind to the principal filamentous types to link them together or to generate movement. These include actin-binding proteins such as myosin, which in some cells can assemble into thick filaments, and microtubule-associated proteins. Actin filaments (microfilaments)
Actin filaments are well-defined, fine filaments with a width of 6-8 nm (Fig. 2.14), and a solid cross-section. Within most cell types, actin constitutes the most abundant protein and in some motile cells its concentration may exceed 200 µM (10 mg protein per ml cytoplasm). The filaments are formed by the ATPdependent polymerization of actin monomer into a characteristic linear form in which the subunits are arranged in a single tight helix with a distance of 13 subunits between turns. The polymerized form is termed F-actin (fibrillar actin) and the unpolymerized form is G-actin (globular actin), with a molecular mass of 43 kDa. Each monomer has an asymmetric structure. When monomers polymerize, they confer a defined polarity on the filament: the plus end favours monomer addition, and the minus end favours monomer dissociation. Myosins bind to filamentous actin at an angle to give the appearance of a series of
arrowheads pointing towards the minus end of the filament, and the barbs point towards the plus end. There is a dynamic equilibrium between G-actin and Factin: in most cells c.50% of the actin is estimated to be in the polymerized state. Actin-binding proteins
A wide variety of actin-binding proteins exist that are capable of modulating the form of actin within the cell. These interactions are fundamental to the organization of cytoplasm and to cell shape. Actin-binding proteins can be divided into bundling proteins, gel-forming proteins and filament severing proteins. Bundling proteins tie actin filaments together in longitudinal arrays to form cables or core structures. The bundles may be closely spaced, e.g. in microvilli, microspikes and filopodia, where parallel filaments are tied tightly together to form stiff bundles orientated in the same direction. Proteins with this function include fimbrin and villin (also classified as a severing protein). Other actin-bundling proteins form rather looser bundles of filaments that run anti-parallel to each other with respect to their plus and minus ends. They include !-actinin and myosin II, which can form cross-links with ATP-dependent motor activity, and cause adjacent actin filaments to slide on each other, and either change the shape of cells or (if the actin bundles are anchored into the cell membrane at both ends), maintain a degree of active rigidity. Gel-forming proteins, such as filamin, interconnect adjacent actin filaments to produce loose filamentous meshworks (gels) composed of randomly orientated Factin. These networks are frequently found in the outer cortical regions of cells, e.g. fibroblasts. They form a semi-rigid zone from which most other organelles are excluded. Severing proteins, such as gelsolin and severin, bind to F-actin filaments and sever them, which produces profound changes within the actin cytoskeleton and in its coupling to the cell surface. page 16 page 17
Other classes of actin-binding proteins link the actin cytoskeleton to the plasma membrane either directly or indirectly through a variety of membrane-associated proteins. The latter may also create links via transmembrane proteins to the extracellular matrix. Best known of these is the family of spectrin-like molecules, which can bind to actin and also to each other and various membrane-associated proteins to create supportive networks beneath the plasma membrane. Spectrin is found in erythrocytes, and closely related molecules are present in many other cells; for instance, fodrin is found in nerve cells, and dystrophin occurs in muscle cells, linking the contractile apparatus with the extracellular matrix via integral membrane proteins. Proteins such as ankyrin (which also binds actin directly), vinculin, talin, zyxin and paxillin connect actin-binding proteins to integral plasma membrane proteins such as integrins (directly or indirectly), and thence to focal adhesions. Myosin I and other unconventional myosins connect actin filaments to membranous structures, including the plasma membrane and transport vesicle membranes. Tropomyosin, an important regulatory protein of muscle fibres, is also present in non-muscle cells, where its function may be primarily to stabilize actin filaments against depolymerization. For further reading see Pollard and Earnshaw (2002). Myosins - the motor proteins
The myosin family of microfilaments is often classified within a distinct category of motor proteins. Myosin proteins have a globular head region consisting of a heavy and a light chain. The heavy chain bears an !-helical tail of varying length. The head has an ATPase activity and can bind to and move along actin filaments - the basis for myosin function as a motor protein. The best-known class is myosin II, which occurs in muscle and in many non-muscle cells. Its molecules have two heads and two tails, intertwined to form a long rod. The rods can bind to each other to form long, thick filaments, as seen in striated and smooth muscle
fibres, myoepithelial cells and myofibroblasts. Myosin II molecules can also assemble into smaller groups, especially dimers, which can cross-link individual actin microfilaments in stress fibres and other F-actin arrays. The ATP-dependent sliding of myosin on actin forms the basis for muscle contraction and the extension of microfilament bundles, as seen in cellular motility or in the contraction of the ring of actin and myosin around the cleavage furrow of dividing cells. There are a number of known subtypes of myosin II: they assemble in different ways and have different dynamic properties. In skeletal muscle the myosin molecules form filaments c.15 nm thick, reversing their direction of assembly at the midpoint, which is bare of head regions, to produce a symmetric arrangement of subunits. In smooth muscle the molecules form thicker, flattened ribbons and are orientated in different directions on either face of the ribbon. These arrangements have important consequences for the contractile force characteristics of the different types of muscle cell. Related molecules are known as unconventional myosins. They include the myosin I subfamily of single-headed molecules with tails of varying length. These molecules are associated with membranes to which their tails can attach, and are implicated in the movements of membranes on actin filaments. So, for example, vesicles track along F-actin in a similar manner to kinesin and dynein-related movements along microtubules. Other functions of myosin I are the movements of membranes in endocytosis, microspike formation in neuronal growth cones, actinactin sliding and attachment of actin to membranes, e.g. of microvilli. Other myosins have been isolated; the significance of their diversity is not fully understood. Other thin filaments
A heterogeneous group of filamentous structures with diameters of 2-4 nm occur in various cells. The two most widely studied forms, titin and nebulin, constitute c.13% of the total protein of skeletal muscle. They are amongst the largest known molecules, and have subunit weights of around 106; native molecules are c.1 µm in length. Their elastic properties are important for the effective functioning of muscle, and possibly for other cells. Intermediate filaments
Intermediate filaments are c.10 nm thick and formed by a heterogeneous group of filamentous proteins. They are found in different cell types and are often present in large numbers, either where structural strength is needed (Fig. 2.15), or to provide scaffolding for the attachment of other structures. Intermediate filaments of different molecular classes are characteristic of particular tissues or states of maturity. They are therefore important indicators of the origins of cells or levels of differentiation, and are of considerable value in histopathology.
Figure 2.15 Keratin filaments (tonofilaments) in two adjacent keratinocytes (K) of the epidermis. Keratin filaments also insert into the desmosome contacts (D) between cells. (By permission from Young B, Heath JW 2000 Wheater's Functional Histology. Edinburgh: Churchill Livingstone.)
Of the different classes of intermediate filaments, keratin (cytokeratin) proteins are found in epithelia, where keratin filaments are always composed of equal ratios of types I (acidic) and II (basic to neutral) keratins. About 20 types of each of the acidic and basic/neutral keratin proteins are known. Within the epidermis, expression of keratin heterodimer combinations changes as keratinocytes mature during their transition from basal to superficial layers. Genetic abnormalities of keratins are known to affect the mechanical stability of epithelia. For example, the disease epidermolysis bullosa simplex causes lysis of epidermal basal cells and blistering of the skin after mechanical trauma. It is caused by defects in genes encoding keratins 5 and 14, which produce cytoskeletal instability and thus cellular fragility in the basal cells. When keratins 1 and 10 are affected, cells in the spinous layer of the epidermis lyse, and this produces the intraepidermal blistering of epidermolytic hyperkeratosis. For a recent review, see Porter and Lane (2003). Vimentins occur in mesenchyme-derived cells of connective tissue, desmins in muscle cells, glial fibrillary acidic protein in glial cells, and peripherin in peripheral axons. Neurofilaments are a major cytoskeletal element in neurones, particularly in axons (Fig. 2.16), where they are the dominant protein. They are heteropolymers of low, medium and high molecular weight neurofilament proteins; the low molecular weight form is always present in combination with either the medium- or the high-molecular weight neurofilament. Abnormal accumulations of neurofilaments (neurofibrillary tangles) are characteristic features of a number of neuropathological conditions. Other intermediate filament proteins include nestin, a molecule resembling neurofilament protein which forms intermediate filaments in neurectodermal stem cells in particular. Nuclear lamins form intermediate filaments that line the inner surface of the nuclear envelope of all nucleated cells. They provide a mechanical
framework for the nucleus and act as attachment sites for chromosomes. They are unusual in that they form a square lattice of regularly spaced crossing filaments rather than bundles, reflecting their unusual molecular composition. page 17 page 18
Figure 2.16 Microtubules (MT) in axons in peripheral nerve. A, In transverse section, microtubules appear as hollow spherical structures. The axon is ensheathed by a Schwann cell (S) and its associated basal lamina. B, In longitudinal section, microtubules appear as straight, unbranched tubes. Neurofilaments (intermediate filaments of neurones; NF) appear in transverse section as solid punctate structures, smaller than microtubules. (By permission from Young B, Heath JW 2000 Wheater's Functional Histology. Edinburgh: Churchill Livingstone.)
The exact manner in which intermediate filament proteins polymerize to form linear filaments is much more complex than that of tubulin or actin, and has not been fully determined. The individual intermediate filament proteins are chains
with a middle !-helical region flanked on either side by non-helical domains. The proteins associate as coiled coil dimers that form short rods c.48 nm long. These assemble in pairs in a staggered antiparallel formation to form soluble tetramers, eight of which pack together laterally and twist into the rope-like 10 nm intermediate filament. The 32 !-helices in parallel give the filaments their tensile strength. However, unlike actin and myosin, the antiparallel arrangement of the dimers produces a filamentous protein with no intrinsic polarity. The non-coiled regions of the subunits project outwards as side arms that can link intermediate filaments into bundles or attach them to other structures. The existence of different combinations of subunit proteins within one filament is the basis of their functional diversity. In the living cell they have been shown to be quite dynamic structures, possibly as a result of reversible phosphorylation. Microtubules
Microtubules are polymers of tubulin with the form of hollow, relatively rigid cylinders, c.25 nm in diameter and of varying length (up to 70 µm in spermatozoan flagella) (Fig. 2.16). They are present in most cell types, and are particularly abundant in neurones, leukocytes, blood platelets and the mitotic spindles of dividing cells. They also form part of the structure of cilia, flagella and centrioles. There are two major classes of tubulin: !- and "-tubulins. Before microtubule assembly, tubulins are associated as dimers with a combined molecular mass of 100 kDa (50 kDa each). Each protein subunit is c.5 nm across and arranged along the long axis in straight rows of alternating and "-tubulins, forming protofilaments. About 13 protofilaments (the number varies between 11 and 16), associate in a ring to form the wall of a hollow cylindrical microtubule. Each longitudinal row is slightly out of alignment with its neighbour, so that a spiral pattern of alternating ! and " subunits appears when the microtubule is viewed from the side. There is a dynamic equilibrium between the dimers and assembled microtubules: dimeric asymmetry creates polarity (!-tubulins are all orientated towards the minus end, "-tubulins towards the plus end). Tubulin is added preferentially to the plus end; the minus end is relatively slow growing. When tubulin is added at one end while being removed at the other (treadmilling), the microtubule gradually shifts its position longitudinally. Polymerization requires phosphorylation of tubulins by guanosine triphosphate, and either a nucleation site (e.g. the end of a pre-existing microtubule), or a microtubule-organizing centre (e.g. those surrounding, and including, centrioles), around which spindle microtubules polymerize during cell division and from which cilia can grow. page 18 page 19
Various drugs (e.g. colcemid, vinblastine, griseofulvin , nocodazole) cause microtubule depolymerization by binding the soluble tubulin dimers and so shifting the equilibrium towards the unpolymerized state. Microtubule demolition causes a wide variety of effects, including the inhibition of cell division by disruption of the mitotic spindle. Conversely, the drug taxol stabilizes microtubules and promotes abnormal microtubule assembly which causes a peripheral neuropathy. Different microtubules possess varying degrees of stability, e.g. microtubules in cilia are generally unaffected by many drugs that cause microtubular demolition. There are also differences between tissues, e.g. neurones have a special tubulin subclass. Tubulins associated with microtubule organizing centres include #tubulin. Microtubule-associated proteins
Various small proteins that can bind to assembled tubulins may be concerned with structural properties or associated with motility. Structural microtubuleassociated proteins (MAPs) form cross-bridges between adjacent microtubules or between microtubules and other structures such as intermediate filaments,
mitochondria and the plasma membrane. Microtubule-associated proteins found in neurones include: MAPs 1A and 1B, which are present in neuronal dendrites and axons; MAPs 2A and 2B, found chiefly in dendrites; and tau, found only in axons. MAP 4 is the major microtubule-associated protein in many other cell types. Structural microtubule-associated proteins are implicated in microtubule formation, maintenance and demolition, and are therefore of considerable significance in cell morphogenesis, mitotic division, and the maintenance and modulation of cell shape. Motility-associated microtubule-associated proteins are found in situations in which movement occurs over the surfaces of microtubules, e.g. the transport of cytoplasmic vesicles, bending of cilia and flagella, and some movements of mitotic spindles. They include a large family of motor proteins, the best known of which are the dyneins and kinesins. Another protein, dynamin, is involved in endocytosis. The kinetochore proteins assemble at the centromere during mitosis and meiosis. They attach to spindle microtubules; some of the kinetochore proteins are responsible for chromosomal movements in mitotic and meiotic anaphase. All of these microtubule-associated proteins bind to microtubules and actively slide along their surfaces. Kinesins and dyneins can simultaneously attach to membranes such as transport vesicles and convey them along microtubules for considerable distances, thus enabling selective targeting of materials within the cell. Such movements occur in both directions along microtubules. Kinesindependent motion is mostly towards the plus ends of microtubules, e.g. from the cell body towards the axon terminals in neurones, and away from the centrosome in other cells. Conversely, dynein-related movements are in the opposite direction, i.e. to the minus ends of microtubules. Dyneins also form the arms of peripheral microtubules in cilia and flagella, where they make dynamic crossbridges to adjacent microtubule pairs. The resulting shearing forces cause the axonemal array of microtubules to bend, generating ciliary and flagellar beating movements. A number of related motor proteins can also interact with microtubules, e.g. kinesin-related proteins cross-link mitotic spindle microtubules to push the two centriolar poles apart during mitotic prophase. Centrioles, centrosomes and basal bodies
Centrioles are microtubular cylinders c.0.2 µm in diameter and 0.4 µm long (Fig. 2.17). They are formed by a ring of nine microtubule triplets linked by a number of other proteins. At least two centrioles occur in all cells that are capable of mitotic division. They usually lie close together, at right angles or, most usually, at an oblique angle to each other (an arrangement often termed a diplosome), within the centrosome, a densely filamentous region of cytoplasm at the centre of the cell. The centrosome is the major microtubule-organizing centre of the cell; it is the site at which new microtubules are formed and the mitotic spindle is generated during cell division. Before cell division, a new centriole forms at right angles to each one of the existing centrosomal pair; the resulting new pairs are passed on to the daughter cells. The proximity of the centrosome to the Golgi apparatus provides a means of targeting Golgi vesicular products to different parts of the cell. The microtubule-organizing centre contains complexes of #-tubulin that bind the minus ends of microtubules. Basal bodies are microtubule-organizing centres that are closely related to centrioles, and are believed to be derived from them. They are located at the bases of cilia and flagella, which they anchor to the cell surface. The outer microtubule doublets of cilia and flagella originate from two of the microtubules in each triplet of the basal body.
Figure 2.17 In electron microscopic preparations, one centriole is usually visible in cross-section, revealing the circular arrangement of tubules, while its partner is cut either longitudinally or slightly obliquely (above). (By permission from Stevens A, Lowe JS 1996 Human Histology, 2nd edn. London: Mosby.)
CELL SURFACE PROJECTIONS The surfaces of many different types of cell are specialized to form structures that project from the surface. These projections may permit movement of the cell itself (flagella), or of fluids across the apical cell surface (cilia), or increase the surface area available for absorption (microvilli). Infoldings of the basolateral plasma membrane also increase the area for transport across this surface of the cell.
CILIA AND FLAGELLA Cilia and flagella are motile, hair-like projections of the cell surface which create currents in the surrounding fluid, movements of the cell to which they are attached, or both. Cilia occur on many internal surfaces of the body, in particular: the epithelia of most of the respiratory tract; parts of the male (p. 1307) and female (p. 1329) reproductive tracts; the ependyma that line the central canal of the spinal cord and ventricles of the brain (p. 53). They also occur at the endings of olfactory receptors and vestibular hair cells, and, in modified form, as portions of the rods and cones of the retina (p. 710). A single cell may bear many cilia, e.g. in bronchial epithelium, or only one or two. Each male gamete possesses a single flagellum c.70 µm long. A cilium or flagellum consists of a shaft (c.0.25 µm diameter) constituting most of its length, a tapering tip and a basal body at its base, which lies within the surface cytoplasm of the cell (Fig. 2.18). Other than at its base, the entire structure is covered by plasma membrane. The core of the cilium is the axoneme, a cylinder of nine microtubule doublets that surrounds a central pair of single microtubules. Several filamentous structures are associated with the microtubules in the shaft, e.g. radial spokes extend inwards from the outer microtubules towards the central pair. The outer doublet microtubules bear two rows of tangential dynein arms attached to the A subfibre of the doublet, which point towards the B subfibre of the adjacent doublet. Adjacent doublets are also linked by thin filaments. Other
filaments partially encircle the central pair of microtubules, which are also united by ladder-like spokes. The '9 + 2' pattern of microtubules imparts a plane of symmetry that passes perpendicular to a line joining the central pair and corresponds to the direction of bending. page 19 page 20
Figure 2.18 A, Structure of a cilium shown in (left) transverse and (right) longitudinal section. B, Apical region of respiratory epithelial cells, showing the proximal parts of three cilia sectioned longitudinally, anchored into the cytoplasm by basal bodies (BB). Other cilia project out of the plane of section and are cut transversely, showing the '9 + 2' arrangement of microtubules. (A, redrawn by permission from Sleigh MA 1977 The nature and action of respiratory tract cilia. In: Brain JD, Proctor DF, Reid LM (eds) Respiratory Defense Mechanisms. Part 1. New York: Dekker; pp247-288. B, by permission from Young B, Heath JW 2000 Wheater's Functional Histology. Edinburgh: Churchill Livingstone.)
Movements of cilia and flagella are broadly similar. Flagella move by rapid undulation, which passes from the attached to the free end. In human spermatozoa there is an additional helical component to this motion. In cilia, the beating is planar, but asymmetric. In the effective stroke, the cilium remains stiff except at the base, where it bends to produce an oar-like stroke. The recovery stroke follows, during which the bend passes from base to tip, returning the cilium to its initial position for the next cycle. The activity of groups of cilia is usually coordinated so that the bending of one is rapidly followed by the bending of the next and so on, resulting in long travelling waves of metachronal synchrony. These pass over the tissue surface in the same direction as the effective stroke. When a cilium bends, the microtubules do not change in length, but slide on one another. The dynein arms of peripheral doublets slant towards the base of the cilium from their attached ends. Dynein has an ATPase activity, which is stimulated by magnesium ions, and causes mutual sliding of adjacent doublets by initially attaching sideways to the next pair, then swinging upwards towards the tip of the cilium. There is a group of genetic diseases in which cilia beat either ineffectively or not at all, e.g. Kartagener's immotile cilium syndrome. Affected cilia exhibit various ultrastructural defects in their internal structure, such as a lack of dynein arms or missing spokes. Patients with this syndrome suffer various respiratory problems caused by the accumulation of particles in the lungs; males are typically sterile because of the loss of sperm motility, and c.50% have an alimentary tract that is a mirror image of the usual pattern (situs inversus) - i.e. it rotates in the opposite direction during early development (p. 1257).
MICROVILLI Microvilli are finger-like cell surface extensions usually c.0.1 µm in diameter and up to 2 µm long (Fig. 2.19). When arranged in a regular parallel series, they constitute a striated border, as typified by the absorptive surfaces of the epithelial enterocytes of the small intestine. When they are less regular, as in the gallbladder epithelium and proximal kidney tubules, the term brush border is used.
Figure 2.19 Microvilli sectioned longitudinally in the striated border of two adjacent intestinal absorptive cells with interlocking lateral plasma membranes (below, centre). Actin filaments fill the cores of the microvilli and insert into a terminal web of actin filaments in the apical cytoplasm.
Microvilli are covered by plasma membrane, and supported internally by closely packed bundles of actin microfilaments linked by cross-bridges of the actinbundling proteins, villin and fimbrin. Other bridges composed of myosin I and calmodulin connect the microfilaments to the plasma membrane. The microfilament bundles of microvilli are embedded in the apical cytoplasm amongst a meshwork of transversely running microfilaments linked by spectrin to form the terminal web (Fig. 2.19). The web is anchored laterally to the zonula adherens. Myosin is also found in the terminal web, where it is believed to bind to the actin and so stiffen this part of the cell. At the apex of each microvillus, the free ends of microfilaments are inserted into a dense mass that includes the protein, !-actinin. page 20 page 21
Microvilli greatly increase the area of cell surface (up to 40 times), particularly at sites of active absorption. In the small intestine, they have a very thick cell coat or glycocalyx, which reflects the presence of integral membrane glycoproteins, including enzymes concerned with digestion and absorption. Irregular microvilli, filopodia, are also found on the surfaces of many types of cell, particularly of free macrophages and fibroblasts, where they may be associated with phagocytosis and cell motility. Long, regular microvilli are called stereocilia, an early misnomer, as they are not motile and lack microtubules. They are found on cochlear and vestibular receptor cells (p. 663), where they act as sensory transducers, and also in the absorptive epithelium of the epididymis (p. 1307).
Nucleus The nucleus (Figs 2.1, 2.2) is generally the largest intracellular structure and is usually spherical or ellipsoid in shape, with a diameter of 3-10 µm. Histological stains used to identify nuclei in tissue sections mainly detect the acidic molecules of deoxyribonucleic acid (DNA), which are largely confined to the nucleus.
NUCLEAR MEMBRANE The nucleus is surrounded by two layers of membrane, each of which is a lipid bilayer, and which together form the nuclear membrane or envelope. The outer membrane layer and the lumen between the two layers are continuous with the rough endoplasmic reticulum. Like the rough endoplasmic reticulum, the outer membrane of the nuclear envelope is studded with ribosomes that are active in protein synthesis; the newly synthesized proteins pass into the perinuclear space between the two membrane layers. Intermediate filaments are associated with both the inner (nuclear) and outer (cytoplasmic) surfaces of the nuclear membrane. Within the nucleus they form a dense shell beneath the membrane, the nuclear lamina, consisting of specialized nuclear intermediate filaments called nuclear lamins. These cross each other at right angles to create a meshwork that covers the interior surface of the nuclear membrane. In so doing, they reinforce the nuclear membrane mechanically, determine the shape of the nucleus and anchor the ends of chromosomes. Condensed chromatin (heterochromatin) also tends to aggregate near the nuclear membrane during interphase. At the end of mitotic and meiotic prophase (p. 23), the lamin filaments disassemble, causing the nuclear membranes to vesiculate. At the end of anaphase, the lamins reattach to the chromosomes and create a new nuclear compartment around which the nuclear membranes reform. A network of filamentous proteins, the nuclear matrix, is also present throughout the nucleus. It is associated with newly replicated DNA and with genes that are being actively transcribed, and incorporates enzymes of the replication machinery. The transport of molecules between the nucleus and the cytoplasm is achieved by specialized nuclear pore structures that perforate the nuclear membrane (Fig.
2.20). They act as highly selective directional molecular filters, permitting proteins such as histones and gene regulatory proteins (which are synthesized in the cytoplasm but function in the nucleus) to enter the nucleus, and molecules that are synthesized in the nucleus but destined for the cytoplasm (e.g. ribosomal subunits, transfer RNAs and messenger RNAs), to leave the nucleus. Ultrastructurally, nuclear pores appear as disc-like structures with an outer diameter of c.130 nm and an inner pore with an effective diameter for free diffusion of 9 nm (Fig. 2.20B). The nuclear membrane of an active cell is bridged by up to 4000 such pores. The nuclear pore complex has an octagonal symmetry and is formed by an assembly of more than 50 proteins, the nucleoporins. The inner and outer nuclear membranes fuse around the pore complex (Fig. 2.20A). Transfer of lipids and proteins between the two is prevented, possibly by the luminal subunits of the pore. Nuclear pores are freely permeable to small molecules, ions and proteins up to about 17 kDa. Proteins of up to 60 kDa seem to be able to equilibrate slowly between the nucleus and cytoplasm through the pore, but larger proteins are normally excluded. However, certain proteins are selectively transported into the nucleus and some of these, such as the DNA polymerases, are very large. Proteins that are selectively transported into the nucleus possess a nuclear localization signal within their amino-acid sequence. This is recognized by cytoplasmic proteins that facilitate the docking of the proteins to be transported with the cytoplasmic surface of the pore. Subsequent translocation into the nucleus is energy dependent and requires the hydrolysis of GTP. The nuclear pore can open to a maximum of c.25 nm to permit the entry or exit of large, actively transported molecules. Steroid hormone receptors, which are gene regulatory proteins, associate with the cytoskeleton until they bind their ligands, when they dissociate from the cytoskeleton and are transported into the nucleus. Transport also occurs from the nucleus to the cytoplasm, e.g. RNAs synthesized in the nucleus are transported through nuclear pores into the cytoplasm.
CHROMATIN
page 21 page 22
Figure 2.20 A, Nuclear envelope with a nuclear pore (centre field) in transverse section, showing the continuity between the inner and outer phospholipid layers of the envelope on either side of the pore. The fine 'membrane' spanning the pore is formed by proteins of the pore complex. B, Nuclear pores seen 'en face' in a tangential section through the nuclear membrane. (A, by kind permission from Rose Watson, Cancer Research UK.)
DNA is organized within the nucleus in a DNA-protein complex known as chromatin. The protein constituents of chromatin are the histones and the nonhistone proteins. Non-histone proteins are an extremely heterogeneous group that includes DNA and RNA polymerases and gene regulatory proteins. Histones are the most abundant group of proteins in chromatin, primarily responsible for the packaging of chromosomal DNA into its primary level of organization, the nucleosome. There are five histone proteins: H1, H2A, H2B, H3 and H4; the last four combine in equal ratios to form a compact octameric nucleosome core. The DNA molecule (one per chromosome) winds 1.65 times around each nucleosome core, taking up 146 nucleotide pairs. This packaging organizes the DNA into a chromatin fibre 11 nm in diameter, and imparts to this form of chromatin the electron microscopic appearance of beads on a string, in which each bead is separated by a variable length of DNA, c.50 nucleotide pairs long. The nucleosome core region and one of the linker regions constitute the nucleosome proper, which is thus c.200 nucleotide pairs in length. However, chromatin rarely exists in this simple form and is usually packaged further into a 30 nm thick fibre, involving a single H1 histone per nucleosome, which interacts with both DNA and protein to impose a higher order of nucleosome packing. Usually, 30 nm fibres are further folded into loop-like domains, but individual loops are believed to decondense and extend during active transcription. In a typical interphase nucleus, euchromatin (nuclear regions that appear pale in appropriately stained tissue sections, or relatively electron-lucent in electron micrographs; Fig. 2.2) is likely to consist mainly of 30 nm fibres and loops, and contains the transcriptionally active genes. Transcriptionally active cells, such as most neurones, have nuclei that are predominantly euchromatic and often described as 'open face' nuclei. Heterochromatin (nuclear regions that appear dark in appropriately stained tissue sections or electron-dense in electron micrographs) is characteristically located mainly around the periphery of the nucleus, except over the nuclear pores, and around the nucleolus (Fig. 2.2). It is a highly compacted form of chromatin, containing additional proteins; its higher order packaging is poorly understood. Heterochromatin includes non-coding regions of DNA, such as centromeric and telomeric regions, which are known as constitutive heterochromatin. DNA that is inactivated (becoming resistant to transcription) in some cells as they differentiate during development or cell maturation contributes to heterochromatin, and is known as facultative heterochromatin. The inactive X chromosome in females is an example of facultative heterochromatin and can be identified in the light microscope as the deeply staining Barr body (drumstick chromosome) that
projects from the nuclear periphery. In transcriptionally inactive cells, chromatin is predominantly in the condensed, heterochromatic state, and may comprise as much as 90% of the total. Examples of such cells are mature neutrophil leukocytes (in which the condensation of chromatin induces the formation of a multilobed, densely staining nucleus), and the highly condensed nuclei of orthochromatic erythroblasts (late-stage erythrocyte precursors). In most mature cells, a mixture of the two occurs, indicating that only a proportion of the DNA is being transcribed. A particular instance of this is seen in the mature B lymphocyte (plasma cell), in which much of the chromatin is in the condensed condition and is arranged in regular masses around the perimeter of the nucleus, producing the so-called 'clock-face' nucleus (Fig. 2.2). Although this cell is actively transcribing, much of its protein synthesis is of a single immunoglobulin type, and consequently much of its genome is in an inactive state. During mitosis, the chromatin is further condensed to form the much shortened chromosomes characteristic of metaphase. This shortening is achieved through further levels of close packing of the chromatin, and is an energy dependent process involving proteins known as condensins. Progressive folding of the chromosomal DNA by interactions with specific proteins can reduce c.5 cm of chromosomal DNA by 10,000 fold, to a length of c.5 µm in the mitotic chromosome.
CHROMOSOMES AND KARYOTYPES The nuclear DNA of eukaryotic cells is organized into linear units called chromosomes. The DNA in a normal human diploid cell contains 6 $ 109 nucleotide pairs organized in the form of 46 chromosomes (44 autosomes and 2 sex chromosomes). The largest human chromosome (number 1) contains c.2.5 $ 108 nucleotide pairs, and the smallest (the Y chromosome) c.5 $ 107 nucleotide pairs. Each chromosomal DNA molecule contains a number of specialized nucleotide sequences that are associated with its maintenance. One is the centromere. During mitosis, a disc-shaped structure composed of a complex array of proteins, the kinetochore, associates with the centromeric region of DNA in order to attach it to the microtubular spindle. Another sequence, the telomere, defines the end of each chromosomal DNA molecule. Telomeres consist of tandem repeats of a short sequence enriched in guanosine nucleotides. They are not replicated by the same DNA polymerase as the rest of the chromosome, but by a specific enzyme called telomerase. The number of tandem repeats of the telomeric DNA sequence varies. It appears to shorten with successive cell divisions, because telomerase activity reduces or is absent in differentiated cells with a finite lifespan. It is believed that this mechanism regulates cell senescence and protects against proliferative disorders, including cancer.
CLASSIFICATION OF HUMAN CHROMOSOMES A number of genetic abnormalities can be directly related to the chromosomal pattern. The characterization or karyotyping of chromosome number and structure is therefore of considerable diagnostic importance. The identifying features of individual chromosomes are most easily seen during metaphase, although prophase chromosomes can be used for more detailed analyses. Lymphocytes separated from blood samples, or cells taken from other tissues, are used as a source of chromosomes. Diagnosis of fetal chromosome patterns is generally carried out on samples of amniotic fluid containing fetal cells aspirated from the uterus by amniocentesis, or on a small piece of chorionic villus tissue removed from the placenta. Whatever their origin, the cells are cultured in vitro
and stimulated to divide by treatment with agents that stimulate cell division. Mitosis is interrupted at metaphase with spindle inhibitors. The chromosomes are dispersed by first causing the cells to swell in a hypotonic solution, then the cells are gently fixed and mechanically ruptured on a slide to spread the chromosomes. They are subsequently stained in various ways to allow the identification of individual chromosomes by size, shape and distribution of stain (Fig. 2.21). General techniques show the obvious landmarks, e.g. lengths of arms and positions of constrictions. Banding techniques demonstrate differential staining patterns, characteristic for each chromosome type. Fluorescence staining with quinacrine mustard and related compounds produces Q bands, and Giemsa staining (after treatment that partially denatures the chromatin) gives G bands (Fig. 2.21A). Other less widely used methods include: reverse-Giemsa staining, in which the light and dark areas are reversed (R bands); the staining of constitutive heterochromatin with silver salts (C-banding); T-banding to stain the ends (telomeres) of chromosomes. Collectively, these methods permit the classification of chromosomes into numbered autosomal pairs in order of decreasing size, from 1 to 22 plus the sex chromosomes. Group 1-3 (A) 4-5 (B) 6-12 + X (C) 13-15 (D) 16-18 (E) 19-20 (F) 21-22 + Y (G)
Features Large metacentric chromosomes Large submetacentric chromosomes Metacentrics of medium size Medium-sized acrocentrics with satellites Shorter metacentrics (16) or submetacentrics (17,18) Shortest metacentrics Short acrocentrics; 21, 22 with satellites, Y without
A summary of the major classes of chromosomes is given below: Methodological advances in banding techniques improved the recognition of abnormal chromosome patterns. The use of in-situ hybridization with fluorescent DNA probes specific for each chromosome (Fig. 2.21B) permits the identification of even very small abnormalities.
NUCLEOLUS page 22 page 23
Figure 2.21 Chromosomes from normal males, arranged as karyotypes. A, G-banded preparation; B, preparation stained by multiplex fluorescence in-situ hybridization to identify each chromosome. (By kind permission from Dr Denise Sheer, Cancer Research UK.)
Nucleoli are a prominent feature of an interphase nucleus (Fig. 2.2). They are the site of most of the synthesis of rRNA and assembly of ribosome subunits. Ultrastructurally, the nucleolus appears as a pale fibrillar region (non-transcribed DNA), containing dense fibrillar cores (sites of rRNA gene transcription) and granular regions (sites of ribosome subunit assembly) within a diffuse nucleolar matrix. Five pairs of chromosomes carry rRNA genes organized in clusters of tandemly repeated units on each chromosome. Each rRNA unit is transcribed individually and encodes the 28S, 18S and 5.8S rRNA molecules. During mitosis the nucleolus breaks down. It reforms after telophase, in a process initiated by the onset of transcription in nucleolar organizing centres on each chromosome. The 28S, 18S and 5.8S rRNA molecules are assembled into their ribosomal subunits in the granular region of the nucleolus together with the 5S rRNA, which is not synthesized in the nucleolus. The newly formed ribosomal subunits are then translocated to the cytoplasm through the nuclear pores.
© 2008 Elsevier
CELL DIVISION AND THE CELL CYCLE During prenatal development, most cells undergo repeated division as the body grows in size and complexity. As cells mature, they differentiate structurally and functionally. Some cells, such as neurones, lose the ability to divide. Others may persist throughout the lifetime of the individual as replication-competent stem cells, e.g. cells in the haemopoietic tissue of bone marrow. Many stem cells divide infrequently, but give rise to daughter cells that undergo repeated cycles of mitotic division as transit (or transient) amplifying cells. Their divisions may occur in rapid succession, as in cell lineages with a short lifespan and similarly fast turnover and replacement time. Transit amplifying cells are all destined to differentiate and ultimately to die and be replaced, unlike the population of parental stem cells, which self-renews. Patterns and rates of cell division within tissues vary considerably. In many epithelia, such as the crypts between intestinal villi, the replacement of damaged or effete cells by division of stem cells can be rapid. Rates of cell division may also vary according to demand, as occurs in the healing of wounded skin, in which cell proliferation increases to a peak and then returns to the normal replacement level. The rate of cell division is tightly coupled to the demand for growth and replacement. Where this coupling is faulty, tissues either fail to grow or replace their cells, or they can overgrow, producing neoplasms. The cell cycle is the period of time between the birth of a cell and its own division to produce two daughter cells. It lasts a minimum of 12 hours, but in most adult tissues is considerably longer, and is divided into four distinct phases, which are known as G1, S, G2 and M. The combination of G1, S and G2 phases is known as interphase. M is the mitotic phase. G1 is the period when cells respond to growth factors directing the cell to initiate another cycle; once made, this decision is irreversible. It is also the phase in which most of the molecular machinery required to complete another cell cycle is generated. Cells that retain the capacity for proliferation, but which are no longer dividing, have entered a phase called G0 and are described as quiescent. Growth factors can stimulate quiescent cells to leave G0 and re-enter the cell cycle, whereas the proteins encoded by certain tumour suppressor genes (e.g. the gene mutated in retinoblastoma, Rb) block the cycle in G1. DNA replication occurs during S phase, at the end of which the DNA content of the cell has doubled. During G2, the cell prepares for division; this period ends with the breakdown of the nuclear membrane and the onset of chromosome condensation. The times taken for S, G2 and M are similar for most cell types, and occupy c.6-8, 2-4 and 1-2 hours respectively. In contrast, the duration of G1 shows considerable variation, sometimes ranging from less than 2 hours in rapidly dividing cells, to more than 100 hours within the same tissue. The regulation of the transitions between the cell cycle phases is now becoming understood at the molecular level. At the G1-S and G2-M transitions, members of a family of proteins called cyclins attain their maximum abundance in the cell. The G1 cyclins progressively accumulate during G1. The M phase cyclins accumulate during late S phase and throughout G2. High concentrations of cyclin proteins activate a family of cyclin-dependent protein kinase enzymes (CDKs), which are present in constant concentrations during the cell cycle, although their state of activation varies. The activation of different cyclin-CDK complexes regulates the G1-S and G2-M transitions. The activities of these enzymes and their cyclin activators are themselves subject to complex regulation, beyond the scope of this text. There are important checkpoints in the cell cycle at which progress will be arrested if, for instance, DNA replication or mitotic spindle assembly and chromosome attachment are incomplete. Negative regulation systems also operate to delay cell cycle progression when DNA has been damaged by radiation or chemical mutagens. Cells with checkpoint defects, such as loss of the protein p53 which is a major negative control element in the division cycle of all cells, are commonly associated with the development of malignancy. Cells lacking one of the critical checkpoint functions are then able to progress through the cycle carrying defects, which increases the probability that further abnormalities will accumulate in their progeny. The p53 gene is an example of a tumour suppressor gene. For further reading on cell cycle regulation, see Alberts et al (2002).
Mitosis and meiosis (Figs 2.22, 2.23) Mitosis occurs in most somatic cells. It results in the distribution of identical copies of the parent cell genome to the two daughter cells. In meiosis, the divisions immediately before the final production of gametes halve the number of chromosomes to the haploid number, so that at fertilization the diploid number is restored. Moreover, meiosis includes a phase in which exchange of genetic material occurs between homologous chromosomes. This allows a reassortment of genes to take place, which means that the daughter cells differ from the parental cell in both their precise genetic sequence and their haploid state. Mitosis and meiosis are alike in many respects, and differ principally in chromosomal behaviour during the early stages of cell division. In meiosis, two divisions occur in quick succession. Meiosis I is unlike mitosis, whereas meiosis II is more like mitosis.
MITOSIS page 23 page 24
Figure 2.22 The stages in mitosis, including the appearance and distribution of the chromosomes.
New DNA is synthesized during the S phase of the cell cycle interphase. This means that the amount of DNA in diploid cells has doubled to the tetraploid value by the onset of mitosis, although the chromosome number is still diploid. During mitosis, this amount is halved between the two daughter cells, so that DNA
quantity and chromosome number are diploid in both cells. The nuclear changes that achieve this distribution are conventionally divided into four phases called prophase, metaphase, anaphase and telophase (Figs 2.22, 2.24).
PROPHASE During prophase, the strands of chromatin, which are highly extended during interphase, shorten, thicken and resolve themselves into recognizable chromosomes. Each chromosome is made up of duplicate chromatids joined at their centromeres. Outside the nucleus, the two centriole pairs begin to separate, and move towards opposite poles of the cell. Parallel microtubules are assembled between them to create the mitotic spindle, and others radiate to form the asters, which come to lie at the spindle poles. As prophase proceeds, the nucleoli disappear, and the nuclear membrane suddenly disintegrates into small vesicles to release the chromosomes, an event that marks the end of prophase.
PROMETAPHASE-METAPHASE As the nuclear membrane disappears, the spindle microtubules extend into the central region of the cell, attaching to the chromosomes which move towards the equator of the spindle (prometaphase). This plane is called the metaphase or equatorial plate. The chromosomes, attached at their centromeres, appear to be arranged in a ring when viewed from either pole of the cell, or to lie linearly across this plane when viewed from above. Cytoplasmic movements during late metaphase effect the approximately equal distribution of mitochondria and other organelles around the cell periphery.
ANAPHASE The centromere in metaphase is a double structure (one per sister chromatid). During anaphase its halves separate, each carrying an attached chromatid. Each original chromosome appears therefore to split lengthwise into two new chromosomes, which move apart, one towards each pole. At the end of anaphase the chromosomes are grouped at either end of the cell, and both clusters are diploid in number. An infolding of the cell equator begins, and deepens during telophase as the cleavage furrow.
TELOPHASE During telophase the chromosomes decondense. Each nuclear membrane forms, beginning as membranous vesicles at the ends of the chromosomes, and the nucleoli appear. At the same time, cytoplasmic division, which usually begins in early anaphase, continues until the new cells separate, each with its derived nucleus. The spindle remnant now disintegrates. While the cleavage furrow is active, a peripheral band or belt of actin and myosin appears in the constricting zone: contraction of this band is responsible for furrow formation. Failure of disjunction of chromatids, so that paired chromatids pass to the same pole, may sometimes occur. Of the two new cells, one will have more, and the other fewer, chromosomes than the diploid number. Exposure to ionizing radiation promotes non-disjunction and may, by chromosomal damage, inhibit mitosis altogether. A typical symptom of radiation exposure is the failure of rapidly dividing epithelia to replace lost cells, with consequent ulceration of the skin and mucous membranes. Mitosis can also be disrupted by chemical agents, particularly colchicine and its derivatives. These compounds inhibit or reverse spindle microtubule formation, so that mitosis is arrested in metaphase. This underpins the rationale for many types of cytotoxic drugs used in cancer therapy.
MEIOSIS There are two cell divisions during meiosis. Details of this process differ at a cellular level for male and female lineages.
MEIOSIS I Prophase I
Prophase I is a long and complex phase that differs considerably from mitotic prophase and is customarily divided into five substages, called leptotene, zygotene, pachytene, diplotene and diakinesis. Leptotene stage
page 24 page 25
Figure 2.23 The stages in meiosis, depicted by two pairs of maternal and paternal homologues (dark and pale colours). DNA and chromosome complement changes and exchange of genetic information between homologues are indicated.
Chromosomes appear as individual threads that are attached at one end to the nuclear membrane. They show characteristic beading throughout their length. Their DNA has been replicated in the preceding S phase. Zygotene stage Chromosomes lie together side by side in homologous pairs, a process which may be initiated during the previous mitotic division. The homologous chromosomes pair point for point progressively, beginning at their attachment to the nuclear membrane, so that corresponding regions lie in contact. This process is known as synapsis, conjugation or pairing, and each pair is now a bivalent. In the case of the unequal X and Y sex chromosomes, only limited pairing segments are homologous and these pair end to end. Homologous chromosomes are held together by a highly structured fibrillar band, the synaptonemal complex. Pachytene stage As shortening and thickening of each chromosome progress, its two chromatids, which are joined at the centromere, become visible. Each bivalent pair therefore consists of four chromatids, forming a tetrad. Two chromatids, one from each bivalent chromosome, partially coil round each other, and during this stage, exchange of DNA (crossing over or decussation) occurs by breaking and rejoining of strands. Diplotene stage
Homologous pairs, now much shortened, separate except where crossing over has occurred (chiasmata). At least one chiasma forms between each homologous pair and up to five have been observed. In the ovaries, primary oocytes become diplotene by the fifth month in utero and each remains at this stage until the period before ovulation (up to 50 years). Diakinesis The chromosomes, still as bivalents, become even shorter and thicker. They subsequently disperse, as bivalents, to lie against the nuclear membrane. During prophase, the nucleoli disappear and the spindle and asters form as they do in mitosis. At the end of prophase the nuclear membrane disappears and bivalent chromosomes move towards the equatorial plate (prometaphase). Metaphase I
Metaphase I resembles mitotic metaphase, except that the bodies attaching to the spindle microtubules are bivalents, not single chromosomes. These become arranged so that the homologous pairs lie parallel to the equatorial plate, with one on either side. Anaphase and telophase I
Chiasmata finally disappear. Anaphase and telophase I also occur as in mitosis, except that in anaphase the centromeres do not split. Instead of paired chromatids separating to move towards the poles, entire homologous chromosomes (made up of two joined chromatids) move to opposite poles. As positioning of bivalent pairs is random, assortment of maternal and paternal chromosomes in each telophase nucleus is also random. page 25 page 26
Figure 2.24 Stages in mitosis, seen in immature blood cells in a smear preparation of human bone marrow. (By permission from Young B, Heath JW 2000 Wheater's Functional Histology. Edinburgh: Churchill Livingstone.)
During meiosis I, cytoplasmic division occurs as it does in mitosis, to produce two new cells.
MEIOSIS II Meiosis II commences after only a short interval during which no DNA synthesis occurs. This second division is more like mitosis, in that chromatids separate during anaphase, but, unlike mitosis, the separating chromatids are genetically different. Cytoplasmic division also occurs and thus, in the male, four haploid cells result from meiosis I and II.
© 2008 Elsevier
CELL DIFFERENTIATION page 26 page 27
As the embryo develops, its cells pass through a series of changes in gene expression, reflected in alterations of cell structure and behaviour. They begin to diversify, separating first into two main tissue arrangements, epithelium and embryonic mesenchyme, then into more restricted subtypes of tissue, until finally they mature into cells of their particular adult lineage. In this process, and in the maturation of functioning cells of the different lineages from their stem cells, there is a sequential pattern of gene expression that changes and limits the cell to a particular specialized range of activities. Such changes involve alterations in cell structure and biochemical characteristics, particularly in the types of proteins that are synthesized. At the genetic level, differentiation is based on a change in the pattern of repression and activation of the DNA sequences encoding proteins specific to that stage of development. A cell may be committed to a particular differentiated fate without manifesting its commitment until later. Once switched in this way, cells are not usually able to revert to an earlier stage of differentiation, so that an irreversible repression of some gene sequences must have occurred. Differentiation signals include interactions between cells that are mediated by diffusible signalling molecules elaborated by one cell and detected by another, and by contact-mediated signalling (such as Delta-Notch signalling). The latter is particularly important in establishing boundaries between different cell populations in development. Differentiation may also depend in some instances on a temporal sequence such as the number of previous cell divisions. In mature tissues in which cell turnover occurs, similar mechanisms appear to ensure the final differentiation to a functional end cell. This may be linked to the presence of a physiological stimulus, e.g. B lymphocytes respond to exposure to an antigen by differentiating into plasma cells that secrete a neutralizing antibody. In other cases, particularly where a cell is part of a highly organized tissue system, more subtle mechanisms exists to ensure a balance between cell proliferation, differentiation and programmed cell death (apoptosis) (p. 203).
© 2008 Elsevier
APOPTOSIS Cells die as a result of either tissue injury (necrosis) or the internal activation of a 'suicide' programme (apoptosis). Apoptosis (programmed cell death, regulated cell suicide) is a central mechanism controlling multicellular development. During morphogenesis, apoptosis mediates activities such as the separation of the developing digits, and has an important role in regulating the number of neurones in the nervous system (the majority of neurones die during development). Apoptosis also ensures that inappropriate or inefficient cells of the acquired immune system are eliminated. The morphological changes exhibited by necrotic cells are very different from those seen in apoptotic cells. Necrotic cells swell and subsequently rupture; the resulting debris may induce an inflammatory response. Apoptotic cells shrink, their nuclei and chromosomes fragment, forming apoptotic bodies, and their plasma membranes undergo conformational changes that act as a signal to local phagocytes. The dead cells are removed rapidly, and as their intracellular contents are not released into the extracellular environment, inflammatory reactions are avoided. Apoptosis and cell proliferation are intimately coupled: several cell cycle regulators can influence both cell division and apoptosis. The signals that trigger apoptosis include withdrawal of survival factors or exposure to inappropriate proliferative stimuli. The current model of the intracellular pathway(s) that lead to apoptosis implicates permeabilization of the mitochondrial membrane, the release of cytochrome c (from the space between the inner and outer mitochondrial membranes) into the cytosol, and subsequent activation of a family of cysteine proteases known as caspases. Caspases are the intracellular mediators of apoptosis: when activated, they initiate a cascade of degradative processes targeting, in particular, proteins of the nuclear lamina and cytoskeleton. Subversion of the apoptotic response is a key characteristic of many cancer cells. Thus the tumour suppressor gene p53 (which functions in cell-cycle control, regulation of apoptosis and the maintenance of genetic stability), is mutated in about 50% of all human cancers. For further details, see Alberts et al (2002). REFERENCES Alberts B, Johnson A, Lewis J, Raff M, Roberts K, Walter P 2002 Molecular Biology of the Cell, 4th edn. New York: Garland Press. Epstein RJ 2003. Human Molecular Biology. An Introduction to the Molecular Basis of Health and Disease. Cambridge: Cambridge University Press. Reviews signalling molecules and their receptors in the context of human disease. Graff C, Bui T-H, Larsson N-G 2002. Mitochondrial diseases. Best Pract Res Clin Obstet Gynaecol 16: 715-28. Reviews clinical conditions related to the inheritance of maternal mitochondrial gene mutations. Medline Similar articles Full article Morris R, Cox H, Mombelli E, Quinn P 2004. Rafts, little caves and large potholes: how lipid structure interacts with membrane proteins to create functionally diverse membrane environments. In: Quinn PJ (ed) Membrane Dynamics and Domains. London: Kluwer Academic/Plenum Publishers: (in press). Reviews how lipids partition proteins into different environments within membranes, and the benefits that accrue to the proteins as a result. Pollard TD, Earnshaw WC 2002. Cell Biology. Philadelphia: Saunders. Reviews cytoskeletal and motor proteins.
Porter RM, Lane EB 2003. Phenotypes, genotypes and their contribution to understanding keratin function. Trends Genet 19: 278-85. Reviews the largest family of intermediate filament proteins, providing evidence for the functional roles of this diverse group, and addresses inherited tissue fragility disorders resulting from keratin gene mutations. Medline Similar articles Full article page 27 page 28
© 2008 Elsevier
3 Integrating cells into tissues Although some cells in the body are essentially migratory, most exist as cellular aggregates in which individual cells carry out similar or closely related functions in a coordinated manner. These aggregates are termed tissues, and can be classified into a fairly small number of broad categories on the basis of their structure, function and molecular properties. On the basis of their structure, most tissues are divided into four major types: epithelia, connective or supporting tissue, muscle and nervous tissue. Epithelia are continuous layers of cells with little intercellular space, which cover or line surfaces, or have been so derived. In connective tissues, the cells are embedded in an intercellular matrix which, typically, forms a substantial and important component of the tissue. Muscle consists largely of specialized contractile cells. Nervous tissue consists of cells specialized for conducting and transmitting electrical and chemical signals and the cells that support this activity. There is molecular evidence that this structure-based scheme of classification has validity. Thus the intermediate filament proteins (p. 17) characteristic of all epithelia are keratins; those of connective tissue are vimentins; those of muscle are desmins; and those of nervous tissue are neurofilament and glial fibrillary acidic proteins. However, cells such as myofibroblasts, neuroepithelial sensory receptors and ependymal cells of the central nervous system have features of more than one tissue type. Despite its anomalies, the scheme is useful for descriptive purposes and widely used, and will be adopted here. In this section, two of the major tissue categories, epithelia and general connective and supporting tissues, will be described. Specialized skeletal connective tissues, i.e. cartilage and bone, together with skeletal muscle, are described in detail in Chapter 6 as part of the musculoskeletal system overview. Smooth muscle and cardiac muscle are described in Chapter 7. Nervous system tissues are described in Chapter 4. Specialized defensive cells, which also form a migrant population within the general connective tissues, are considered in more detail in Chapter 5, with blood, lymphoid tissues and haemopoiesis.
© 2008 Elsevier
EPITHELIA The term epithelium is applied to the layer or layers of cells that cover the body surfaces or line the body cavities that open on to it. Developmentally, epithelia are derived from all three layers of the early embryo (p. 206). The ectoderm gives rise to the epidermis, glandular tissue of the breast, cornea and the junctional zones of the buccal cavity and anal canal. The endoderm forms the epithelial lining of the alimentary canal and its glands, most of the respiratory tract and the distal parts of the urogenital tract. Mesodermal derivatives include the epithelia of the kidney, the suprarenal (adrenal) cortex and endocrine cells of the ovary and testis. These endocrine cells are atypical epithelia in that they differentiate from embryonic mesenchyme (p. 207) and, in common with endocrine cells in general, they lack a free surface that communicates with the exterior. This atypical category also includes endothelia that line blood vessels and lymphatics (p. 146), and the epithelium-like cell layers of mesodermal (and mesenchymal) origin that line internal cavities of the body and are usually classified separately as mesothelia (p. 41): they line the pericardial, pleural and peritoneal cavities. Epithelia function generally as selective barriers that facilitate, or inhibit, the passage of substances across the surfaces they cover. In addition, they may: protect underlying tissues against dehydration, chemical or mechanical damage; synthesize and secrete products into the spaces that they line; function as sensory surfaces. In this respect, many features of nervous tissue can be regarded as those of a modified epithelium and the two tissue types share an origin (p. 207) in embryonic ectoderm. Epithelia (Fig. 3.1) are predominantly cellular and the little extracellular material they possess is limited to the basal lamina. Intercellular junctions, which are usually numerous, maintain the mechanical cohesiveness of the epithelial sheet and contribute to its barrier functions. A series of three intercellular junctions forms a typical epithelial junctional complex: in sequence from the apical surface, this consists of a tight junctional zone, an adherent (intermediate) junctional zone and a region of discrete desmosome junctions (p. 7). Epithelial cell shape is most usually polygonal and partly determined by cytoplasmic features such as secretory granules. The basal surface of an epithelium lies in contact with a thin layer of filamentous protein and proteoglycan termed the basal lamina, which is synthesized predominantly by the epithelial cells. The basal lamina is described in the section on extracellular matrix (p. 38). Epithelia can usually regenerate when injured. Indeed, many epithelia continuously replace their cells to offset cell loss caused by mechanical abrasion. Blood vessels do not penetrate typical epithelia and so cells receive their nutrition by diffusion from capillaries of neighbouring connective tissues. This arrangement limits the maximum thickness of living epithelial cell layers. Epithelia, together with their supporting connective tissue, can often be removed surgically as one layer, which is collectively known as a membrane. Where the surface of a membrane is moistened by mucous glands it is called a mucous membrane or mucosa (p. 41), whereas a similar layer of connective tissue covered by mesothelium is called a serous membrane or serosa (p. 41).
Classification Epithelia can be classified as unilaminar (single-layered, simple), in which a single layer of cells rests on a basal lamina, or multilaminar, in which the layer is more than one cell thick. The latter includes: stratified squamous epithelia, in which superficial cells are constantly replaced from the basal layers; urothelium (transitional epithelium), which serves special functions in the urinary tract; and other multilaminar epithelia which, like urothelium, are replaced only very slowly under normal conditions. Seminiferous epithelium is a specialized multilaminar tissue found only in the testis.
UNILAMINAR (SIMPLE) EPITHELIA Unilaminar epithelia are further classified according to the shape of their cells, into squamous, cuboidal, columnar and pseudostratified types. Cell shape is largely related to cell volume. Where little cytoplasm is present, there are generally few organelles and therefore low metabolic activity and cells are squamous or low cuboidal. Highly active cells, e.g. secretory epithelia, contain abundant mitochondria and endoplasmic reticulum and are typically tall cuboidal or columnar. Unilaminar epithelia can also be subdivided into those which have special functions, such as those with cilia, numerous microvilli, secretory vacuoles
(in mucous and serous glandular cells), or sensory features. Myoepithelial cells, which are contractile, are found as isolated cells associated with glandular structures, e.g. salivary and mammary glands.
SQUAMOUS EPITHELIUM page 29 page 30
Figure 3.1 Classification of epithelial tissues and cells.
Simple squamous epithelium is composed of flattened, tightly apposed, polygonal cells (squames). This type of epithelium is described as tessellated when the cells have complex, interlocking borders rather than straight boundaries. The cytoplasm may in places be only 0.1 µm thick and the nucleus usually bulges into the overlying space (Fig. 3.2). These cells line the alveoli of the lungs and form the outer capsular wall of renal corpuscles, the thin segments of the renal tubules and various parts of the inner ear. Because it is so thin, simple squamous epithelium allows rapid diffusion of gases and water; it may also engage in active transport, as indicated by the presence of numerous endocytic vesicles in these cells. Tight junctions between adjacent cells ensure that materials pass primarily through cells, rather than between them.
CUBOIDAL AND COLUMNAR EPITHELIA Cuboidal and columnar epithelia consist of regular rows of cylindrical cells (Figs 3.3, 3.4). Cuboidal cells are approximately square in vertical section, whereas columnar cells are taller than their diameter, and both are polygonal when sectioned horizontally. Commonly, microvilli are found on their free surfaces, which considerably increases the absorptive area, e.g. in the epithelia of the small
intestine (columnar cells with a striated border of very regular microvilli), the gallbladder (columnar cells with a brush border) and proximal and distal convoluted tubules of the kidney (large cuboidal to low columnar cells with brush borders). Ciliated columnar epithelium lines: most of the respiratory tract (except for the lower pharynx and vocal folds) where it is pseudostratified (Fig. 3.5) as far as the terminal bronchioles; some of the tympanic cavity and auditory tube; the uterine tube; the efferent ductules of the testis. Mucous glands also line much of the respiratory tract and cilia sweep a layer of mucus and trapped dust particles etc., from the lung towards the pharynx in the mucociliary rejection current, which clears the respiratory passages of inhaled particles. Cilia in the uterine tube assist the passage of oocytes and fertilized ova to the uterus (p. 1329). Some columnar cells are glandular, and their apical domains (p. 5) contain mucus- or protein-filled (zymogen) vesicles, e.g. mucin-secreting and chief cells of the gastric epithelium. Where mucous cells lie among non-secretory cells, e.g. in the intestinal epithelium, their apical cytoplasm and its secretory contents often expand to produce a characteristic cell shape, and they are known as goblet cells (Fig. 3.5). For further details of glandular tissue, see page 34 and for the characteristics of mucus, see page 41.
PSEUDOSTRATIFIED EPITHELIUM page 30 page 31
Figure 3.2 Simple squamous epithelium lining the outer parietal layer of Bowman's capsule in the renal corpuscle. Oval nuclei project into the urinary space, covered by a highly attenuated cytoplasm. The underlying basement membrane is stained blue in this azan preparation. (By permission from Young B, Heath JW 2000 Wheater's Functional Histology. Edinburgh: Churchill Livingstone.)
Figure 3.3 Simple cuboidal epithelium lining a collecting tubule in the renal medulla. Azan preparation; the basement membrane is stained blue. (By permission from Young B, Heath JW 2000 Wheater's Functional Histology. Edinburgh: Churchill Livingstone.)
Figure 3.4 Simple columnar epithelium covering the tip (right) of a villus in the small intestine. Tall, columnar absorptive cells bear a striated border of microvilli, just visible as a deeper-stained apical fringe. A few interspersed goblet cells are present, with pale apical secretory granules and dark, elongated nuclei. (Photograph by Sarah-Jane Smith.)
Figure 3.5 Ciliated columnar epithelium lining a bronchus in the respiratory tract, stained with alcian blue to show the goblet cells with blue-stained mucinogen granules in their apical cytoplasm.
page 31 page 32
Figure 3.6 Pseudostratified columnar epithelium, with ciliated cells, goblet cells with their apical cytoplasm distended by mucinogen granules, and basal cells. This type of epithelium is found almost exclusively in the larger airways of the respiratory system and is thus also known as respiratory epithelium. (By permission from Kierszenbaum AL 2002 Histology and Cell Biology. St Louis: Mosby.)
Pseudostratified epithelium is a single-layered (simple) columnar epithelium in which nuclei lie at different levels in a vertical section (Figs 3.5, 3.6). All cells are in contact with the basal lamina throughout their lifespan, but not all cells extend through the entire thickness of the epithelium. Some constitute an immature basal cell layer of smaller cells, which are often mitotic and able to replace damaged mature cells. Migrating lymphocytes and mast cells within columnar epithelia may also give a pseudostratified appearance because their nuclei are found at different depths. Much of the ciliated lining of the respiratory tract is of the pseudostratified type, and so is the sensory epithelium of the olfactory area.
SENSORY EPITHELIA Sensory epithelia are found in special sense organs of the olfactory, gustatory and vestibulocochlear receptor systems. All of these contain sensory cells surrounded by supportive, non-receptor cells. Olfactory receptors are modified neurones, and their axons pass directly to the brain, but the other types are specialized epithelial cells that synapse with terminals of afferent (and sometimes efferent) nerve fibres.
MYOEPITHELIAL CELLS Myoepithelial cells, which are also sometimes termed basket cells, are fusiform or stellate in shape (Fig. 3.7), contain actin and myosin filaments, and contract when stimulated by nervous or endocrine signals. They surround the secretory portions and ducts of some glands, e.g. mammary, lacrimal, salivary and sweat glands, and lie between the basal lamina and the glandular or ductal epithelium. Their contraction assists the initial flow of secretion into larger conduits. Myoepithelial cells are ultrastructurally similar to smooth muscle cells in the arrangement of their actin and myosin, but differ from them because they originate from
embryonic ectoderm or endoderm. They can be identified immunohistochemically on the basis of the co-localization of myofilament proteins (which signify their contractile function), and keratin intermediate filaments (which accords with their epithelial lineage).
MULTILAMINAR (STRATIFIED) EPITHELIA Multilaminar epithelia are found at surfaces subjected to mechanical damage or other potentially harmful conditions. They can be divided into those which continue to replace their surface cells from deeper layers, designated stratified squamous epithelia, and others in which replacement is extremely slow except after injury.
STRATIFIED SQUAMOUS EPITHELIA
Figure 3.7 Stellate myoepithelial cells wrapped around secretory acini in the lactating mammary gland, seen in the scanning electron microscope after enzymatic removal of extracellular matrix. Rodent tissue. (Reproduced from Cell Tissue Res 209: 1-10. A scanning electron microscope study of myoepithelial cells in exocrine glands. Nagato T et al, 1980 © Springer-Verlag.)
Stratified squamous epithelia are multilayered tissues in which the formation, maturation and loss of cells is continuous, although the rates of these processes can change, e.g. after injury. New cells are formed in the most basal layers by the mitotic division (p. 23) of stem cells and transit (or transient) amplifying cells. The daughter cells move more superficially, changing gradually from a cuboidal shape to a more flattened form and are eventually shed from the surface as a highly flattened squame. Typically, the cells are held together by numerous desmosomes to form strong, contiguous cellular sheets that provide protection to the underlying tissues against mechanical, microbial and chemical damage. Stratified squamous epithelia may be broadly subdivided into keratinized and non-keratinized types. Keratinized epithelium
Keratinized epithelium (Fig. 3.8) is found at surfaces that are subject to drying or mechanical stresses, or are exposed to high levels of abrasion. These include the entire epidermis and the mucocutaneous junctions of the lips, nostrils, distal anal canal, outer surface of the tympanic membrane and parts of the oral lining (gingivae, hard palate and filiform papillae on the anterior part of the dorsal surface of the tongue). Their cells, keratinocytes, are described in more detail on page 158. A distinguishing feature of keratinized epithelia is that cells of the superficial layer, the stratum corneum, are anucleate, dead, flattened squames that eventually flake off from the surface. In addition, the tough keratin intermediate filaments become firmly embedded in a matrix protein. This unusual combination of strongly coherent layers of living cells and more superficial strata made of plates of inert, mechanically robust protein complexes, interleaved with
water-resistant lipid, makes this type of epithelium an efficient barrier against different types of injury and water loss. Non-keratinized epithelium
Non-keratinized epithelium is present at surfaces that are subject to abrasion but protected from drying (Fig. 3.9). These include: the buccal cavity (except for the areas noted above); oropharynx and laryngopharynx; oesophagus; part of the anal canal; vagina; distal uterine cervix; distal urethra; conjunctiva and cornea; inner surfaces of the eyelids; the vestibule of the nasal cavities. Cells go through the same transitions in general shape as are seen in the keratinized type, but they do not fill completely with keratin or secrete glycolipid, and they retain their nuclei until they desquamate at the surface. In sites where considerable abrasion occurs, e.g. parts of the buccal cavity, the epithelium is thicker and its most superficial cells may partly keratinize, so that it is referred to as parakeratinized, in contrast to the orthokeratinized state of fully keratinized epithelium. Diets deficient in vitamin A may induce keratinization of such epithelia, and excessive doses may lead to its transformation into mucus-secreting epithelium.
page 32 page 33
Figure 3.8 Keratinized stratified squamous epithelium from the epidermis (E) of the lateral surface of a toe, showing thick skin with a prominent cornified layer (C) of dead keratinized squames. The granular layer (G) is prominent in thick skin. (By permission from Dr JB Kerr, Monash University, from Kerr JB 1999 Atlas of Functional Histology. London: Mosby.)
Figure 3.9 The luminal surface of the oesophagus is nonkeratinized, stratified squamous epithelium (SSE), similar to much of the epithelial lining of the oral cavity. The most superficial layers of stratum corneum are only several cells thick, and the nuclei (N) are retained with little or no transformation into plaques of keratin. Individual lymphocytes are noted in the epithelium (arrow) and at intervals in the subjacent lamina propria (LP). (By permission from Dr JB Kerr, Monash University, from Kerr JB 1999 Atlas of Functional Histology. London: Mosby.)
STRATIFIED CUBOIDAL AND COLUMNAR EPITHELIA Two or more layers of cuboidal or low columnar cells (Fig. 3.10) are typical of the walls of the larger ducts of some exocrine glands, e.g. the pancreas, salivary glands and the ducts of sweat glands and they presumably provide more strength than a single layer. Parts of the male urethra are also lined by stratified columnar epithelium. The layers are not continually replaced by basal mitoses and there is no progression of form from base to surface, but they can repair themselves if damaged.
UROTHELIUM (URINARY OR TRANSITIONAL EPITHELIUM) Urothelium (Fig. 3.11) is a specialized epithelium that lines much of the urinary tract and prevents its rather toxic contents from damaging surrounding structures. It extends from the ends of the collecting ducts of the kidneys, through the ureters (p. 1277) and bladder (p. 1292), to the proximal portion of the urethra. In males it covers the urethra as far as the ejaculatory ducts, then becomes intermittent and is finally replaced by stratified columnar epithelium in the membranous urethra. In females it extends as far as the urogenital membrane. During development, part of it is derived from mesoderm and part from ectoderm and endoderm.
Figure 3.10 Stratified cuboidal epithelium lining a large interlobular collecting duct of the parotid salivary gland. (Photograph by Sarah-Jane Smith.)
Figure 3.11 Urothelium (transitional epithelium) lining the urinary bladder. The most superficial cells have a thickened plasma membrane as a result of the presence of intramembranous plaques, which give an eosinophilic appearance to the luminal surface. (By permission from Kierszenbaum AL 2002 Histology and Cell Biology. St Louis: Mosby.)
The epithelium appears to be four to six cells thick, and lines organs that undergo considerable distension and contraction. It can therefore stretch greatly without losing its integrity. In stretching, the cells become flattened, without altering their positions relative to each other, as they are firmly connected by numerous desmosomes. However, the urothelium appears to be reduced to two to three cells thick. The epithelium is called transitional because of the apparent transition between a stratified cuboidal epithelium and a stratified squamous epithelium, which occurs as it is stretched to accommodate urine, particularly in the bladder. The basal cells are basophilic, with many ribosomes, and are cuboidal, uninucleate (diploid) when relaxed. More apically, they form large binucleate, or, more often, polyploid uninucleate cells. The surface cells are largest and may even be octaploid: in the relaxed state, they typically bulge into the lumen as dome-shaped cells with a thickened, eosinophilic glycocalyx or cell coat (p. 6). Their luminal surfaces are covered by a specialized plasma membrane in which plaques of intramembranous glycoprotein particles are embedded. These plaques stiffen the membrane. When the epithelium is in the relaxed state, and the surface area of the cells is reduced, the plaques are partially internalized by the hinge-like action of the more flexible interplaque membrane regions. They re-emerge onto the surface when it is stretched. Normally, cell turnover is very slow; cell division is infrequent and is restricted to the basal layer. However, when damaged, the epithelium regenerates quite rapidly.
SEMINIFEROUS EPITHELIUM Seminiferous epithelium is a highly specialized, complex stratified epithelium. It consists of a heterogeneous population of cells that form the lineage of the spermatozoa (spermatogonia, spermatocytes, spermatids), together with supporting cells (Sertoli cells). It is described in detail in Chapter 97 (p. 1307).
© 2008 Elsevier
GLANDS page 33 page 34
One of the features of many epithelia is their ability to alter the environment facing their free surfaces by the directed transport of ions, water or macromolecules. This is particularly well demonstrated in glandular tissue, in which the metabolism and structural organization of the cells is specialized for the synthesis and secretion of macromolecules, usually from the apical surface. Such cells may exist in isolation amongst other non-secretory cells of an epithelium, e.g. goblet cells in the absorptive lining of the small intestine, or may form highly coherent sheets of epithelium with a common secretory function, e.g. the mucous lining of the stomach and, in a highly invaginated structure, the complex salivary glands. Glands may be subdivided into exocrine glands and endocrine glands. Exocrine glands secrete, often via a duct, onto surfaces that are continuous with the exterior of the body, including the alimentary tract, respiratory system, urinary and genital ducts and their derivatives, and the skin. Endocrine glands are ductless and secrete hormones directly into the circulatory system, which then conveys them throughout the body to affect the activities of other cells (p. 179). Paracrine glandular cells are similar to endocrine cells, but their secretions diffuse locally to cellular targets in the immediate vicinity. In addition to strictly epithelial glands, some tissues derived from the nervous system, including the suprarenal medulla (p. 1247) and neurohypophysis (p. 380), are neurosecretory. Modes of signalling by secretory cells are illustrated in Fig. 2.6.
Exocrine glands TYPES OF SECRETORY PROCESSES (Fig. 3.12) The mechanism of secretion varies considerably. If the secretions are initially packaged into vesicles, these are conveyed to the cell surface (p. 11), where they are discharged in a number of different ways. In merocrine secretion, which is by far the most common mechanism, vesicle membranes fuse with the plasma membrane to release their contents to the exterior. Specialized transmembrane molecules in the secretory vacuole wall recognize marker proteins on the cytoplasmic side of the plasma membrane and bind to them. This initiates interactions with other proteins that cause the fusion of the two membranes and the consequent release of the vesicle contents. The stimulus for secretion varies with the type of cell, but often appears to involve a rise in intracellular calcium. Glands such as the simple sweat glands of the skin, where ions and water are actively transported from plasma, were once classified as eccrine glands. They are now known to synthesize and secrete small amounts of protein by a merocrine mechanism, and are thus reclassified as merocrine glands. In apocrine glands, some of the apical cytoplasm is pinched off with the contained secretions, which are stored in the cell as membrane-free droplets. The best understood example of this is the secretion of milk fat by mammary gland cells (p. 972), in which a small amount of cytoplasm is incorporated into the plasma
membrane-bound lipid globule as it is released from the cell. Larger amounts of cytoplasm are included in secretions by specialized apocrine sweat glands in the axilla and genitoanal regions of the body. In some tissues there is a combination of different types of secretion, e.g. mammary gland cells secrete milk fat by apocrine secretion and milk protein, casein, by merocrine secretion. In holocrine glands, e.g. sebaceous glands in the skin, the cells first fill with secretory products (lipid droplets or sebum, in this instance) and then the entire cell disintegrates to liberate the accumulated mass of secretion into the duct or hair follicle.
STRUCTURAL AND FUNCTIONAL CLASSIFICATION Exocrine glands are either unicellular or multicellular. The latter may be in the form of simple sheets of secretory cells, e.g. at the surface of the stomach, or may be structurally more complex and invaginated to a variable degree. Such glands (Fig. 3.12) may be single units or their connection to the surface may be branched. Simple unbranched tubular glands exist in the walls of many of the hollow viscera, e.g. the small intestine and uterus, whereas some single glands have expanded, flask-like ends (acini or alveoli). Such glands may consist entirely of secretory cells, or may have a blind-ending secretory portion that leads through a non-secretory duct to the surface, in which case the ducts may modify the secretions as they pass along them. Glands with ducts may be branched (compound), and sometimes form elaborate ductal trees. Such glands generally have acinar or alveolar secretory lobules, as in the exocrine pancreas, but the secretory units may alternatively be tubular or mixed tubulo-acinar. More than one type of secretory cell may occur within a particular secretory unit, or individual units may be specialized to just one type of secretion (e.g. serous acini of salivary glands). Exocrine glands are also classified by their secretory products. Secretory cells in mucus-secreting or mucous glands have frothy cytoplasm and basal, flattened nuclei. They stain deeply with metachromatic stains and periodic acid-Schiff (PAS) methods that detect carbohydrate residues. However, in general (i.e. nonspecific) histological preparations they are weakly stained because much of their content of water-rich mucin is extracted by the processing procedures. Secretory cells in serous glands have centrally placed nuclei and eosinophilic secretory storage granules in their cytoplasm. They secrete mainly glycoproteins (including lysozyme and digestive enzymes). Some glands are almost entirely mucous (e.g. the sublingual salivary gland), whereas others are mainly serous (e.g. the parotid salivary gland). The submandibular gland is mixed, in that some lobules are predominantly mucous and others serous. In some regions, mucous acini share a lumen with clusters of serous cells (seen in routine preparations as serous demilunes). Although this simple approach to classification is useful for general descriptive purposes, the diversity of molecules synthesized and secreted by glands is such that complex mixtures often exist within the same cell.
Endocrine glands Endocrine glands secrete directly into the circulation. Their cells are grouped around beds of capillaries or sinusoids (p. 143) which typically are lined by fenestrated endothelia to allow the rapid passage of macromolecules through their walls. Endocrine cells may be arranged in clusters around vascular networks, in cords between parallel vascular channels or as hollow structures (follicles) surrounding their stored secretions. Isolated endocrine cells also exist scattered amongst other tissues as part of the dispersed neuroendocrine system (p. 180), e.g. throughout the alimentary and respiratory tracts.
Control of glandular secretion Glandular activity may be controlled directly by autonomic secretomotor fibres, which may either form synapses on the bases of gland cells (e.g. in the suprarenal medulla) or release neuromediators in the vicinity of the glands, to reach them by diffusion. Alternatively, the autonomic nervous system may act indirectly on gland cells, e.g. via histamine released neurogenically from another cell, as occurs in the gastric lining. Paracrine activities of neuroendocrine cells are important in the alimentary and bronchial glands. Circulating hormones from the adenohypophysis stimulate synthesis and secretion by target cells in many endocrine glands. Such signals, mostly detected by receptors at the cell surface and mediated by second messenger systems, may increase the synthetic activity of gland cells, and may cause them to discharge their secretions by exocytosis. Secretions already released into ducts are expressed rapidly from certain glands by the contraction of associated myoepithelial cells that enclose the secretory units and smaller ducts. These may be under direct neural control, as in the salivary glands, or they may respond to circulating hormones, e.g. cells in the mammary gland respond to the concentration of circulating oxytocin .
© 2008 Elsevier
BASEMENT MEMBRANE AND BASAL LAMINA There is a narrow layer of extracellular matrix (p. 38), which stains strongly for carbohydrates at the interface between connective and other tissues, e.g. between epithelia and their supporting connective tissues. In early histological texts this layer was termed the basement membrane. As almost all of its components are synthesized by the epithelium or other tissue (e.g. muscle), rather than the adjacent connective tissue, it will be discussed here. page 34 page 35
Figure 3.12 Classification of the different types of epithelial gland.
Electron microscopy revealed that the basement membrane is composed of two distinct components. A thin, finely fibrillar layer, the basal lamina, is associated closely with the cell surface (Fig. 3.13). A variable reticular lamina of larger fibrils and glycosaminoglycans of the extracellular matrix underlies this layer and is continuous with the connective tissue proper, although it is much reduced or largely absent in some tissues, e.g. surrounding muscle fibres, Schwann cells and capillary endothelia. In other tissues, the basal lamina separates two layers of cells and there are no intervening typical connective tissue elements. This occurs in the thick basal lamina of the renal glomerular filter (p. 1277), the basal lamina of the thin portions of the lung interalveolar septa across which gases exchange between blood and air p. 1060), and the anterior limiting (Descemet's) lamina in the cornea (p. 703). page 35 page 36
Figure 3.13 The basal lamina, underlying an epithelium (top). The finely fibrillar dense layer corresponds to the lamina densa and spherical collagen fibrils sectioned transversely lie in the subjacent connective tissue. These contribute to the appearance of the basement membrane in light microscope preparations stained for carbohydrate-rich structures. (By permission from Young B, Heath JW 2000 Wheater's Functional Histology. Edinburgh: Churchill Livingstone.)
The basal lamina is usually c.80 nm thick, varying between 40 and 120 nm, and consists of a sheet-like fibrillar layer, the lamina densa (20-50 nm wide), separated from the plasma membrane of the cell it supports by a narrow electronlucent zone, the lamina lucida. The lamina lucida is absent from tissues prepared by rapid freezing and so may be an artefact. In many tissues this zone is crossed by integral plasma membrane proteins, e.g. keratinocyte hemidesmosomes are anchored into the lamina densa in the basal lamina of the epidermis. The basal lamina is a delicate felt-like network composed largely of two glycoprotein polymers, laminin and type IV collagen, which self-assemble into twodimensional sheets interwoven with each other. Early embryonic basal lamina is formed only of the laminin polymer. Two other molecules cross-link and stabilize the network: entactin (nidogen) and perlecan (a large heparan sulphate proteoglycan). Although all basal laminae have a similar form, their thickness and precise molecular composition vary between tissues and even within a tissue, e.g. between the crypts and villi of the small intestine. The isoforms of laminin and collagen type IV differ in various tissues, thus Schwann cells and muscle cells express laminin-2 (merosin) rather than the prototypical laminin-1. Laminin-5, although not itself a basal lamina component, is found in the hemidesmosomes of the basal epidermis and links the basal lamina with epidermal transmembrane proteins, !6"4 integrin and collagen type XVII (formerly known as bullous pemphigoid antigen, the target of the autoimmune blistering skin disease, bullous pemphigoid). The particular isoform of collagen type IV in the basal lamina of different tissues is reflected in tissue-specific disease patterns. Mutations in a collagen expressed by muscle and kidney glomeruli cause Allport syndrome, a
form of renal failure. Renal failure also occurs in Goodpasture syndrome, in which renal basal lamina collagen is targeted by autoantibodies. In Descemet's membrane in the cornea, collagen type VIII replaces collagen type IV in the much thickened endothelial basal lamina. The basal lamina of the neuromuscular junction (p. 64) contains agrin, a heparan sulphate proteoglycan, which plays a part in the clustering of muscle acetylcholine receptors in the plasma membrane at these junctions.
RETICULAR LAMINA The reticular lamina consists of a dense extracellular matrix that contains collagen. In skin, it contains fibrils of type VII collagen (anchoring fibrils), which bind the lamina densa to the adjacent connective tissue. The high concentration of proteoglycans in the reticular lamina is responsible for the positive reaction of the entire basement membrane to stains for carbohydrates, seen in sections prepared for light microscopy.
Functions of basal lamina Basal laminae perform a number of important roles. They form selectively permeable barriers between adjacent tissues, e.g. in the glomerular filter of the kidney; they anchor epithelial and connective tissues and so stabilize and orient the tissue layers; they may exert instructive effects on adjacent tissues, and so determine their polarity, rate of cell division, cell survival, etc. In addition, they may act as pathways for the migration and pathfinding of growing cell processes, both in development and in tissue repair, e.g. in guiding the outgrowth of axons and the re-establishment of neuromuscular junctions during regeneration after injury in the peripheral nervous system. Changes in basal lamina thickness are often associated with pathological conditions, e.g. the thickening of the glomerular membrane in glomerulonephritis and diabetes.
© 2008 Elsevier
CONNECTIVE AND SUPPORTING TISSUES The connective tissues are defined as those composed predominantly of intercellular material, the extracellular matrix, which is secreted mainly by the connective tissue cells. The cells are therefore usually widely separated by their matrix, which is composed of fibrous proteins and a relatively amorphous ground substance. Many of the special properties of connective tissues are determined by the composition of the matrix, and their classification is also largely based on its characteristics. In some types of connective tissue, the cellular component eventually dominates the tissue, even though the tissue originally has a high matrix:cell ratio, e.g. adipose tissue. Connective tissues are derived from embryonic mesoderm or, in the head region, largely from neural crest. Connective tissues have several essential roles in the body. These may be subdivided into structural roles, which largely reflect the special mechanical properties of the extracellular matrix components, and defensive roles, in which the cellular component has the dominant role. Connective tissues often also play important trophic and morphogenetic parts in organizing and influencing the growth and differentiation of surrounding tissues, e.g. in the development of glands from an epithelial surface. Structural connective tissues are divided into ordinary (or general) types, which are widely distributed, and special skeletal types, i.e. cartilage and bone, which are described in Chapter 6. A third type, haemolymphoid tissues, consists of peripheral blood cells, lymphoid tissues and their precursors; these tissues are described in Chapter 5. They are often grouped with other types of connective tissue, because of their similar mesenchymal origins and because the various defensive cells of the blood also form part of a typical connective tissue cell population. They reach connective tissues via the blood circulation and migrate into them through the endothelial walls of vessels.
Cells of general connective tissues Cells of general connective tissue can be separated into the resident cell population (fibroblasts, adipocytes, mesenchymal stem cells, etc.) and a population of migrant cells with various defensive functions (macrophages, lymphocytes, mast cells, neutrophils and eosinophils), which may change in number and moderate their activities according to demand. Embryologically, fibroblasts and adipocytes arise from mesenchymal stem cells, some of which may remain in the tissues to provide a source of replacement cells postnatally. As noted above, the cells of haemopoietic origin migrate into the tissue from bone marrow and lymphoid tissue.
RESIDENT CELLS FIBROBLASTS Fibroblasts are usually the most numerous resident cells. They are flattened and irregular in outline, with extended processes, and in profile they appear fusiform or spindle-shaped (Figs 3.14, 3.16). Fibroblasts synthesize most of the extracellular matrix of connective tissue; accordingly they have all the features typical of cells active in the synthesis and secretion of proteins. Their nuclei are relatively large and euchromatic and possess prominent nucleoli. In young, highly active cells, the cytoplasm is abundant and basophilic (reflecting the high concentration of rough endoplasmic reticulum), mitochondria are abundant and several sets of Golgi apparatus are present. In old and relatively inactive fibroblasts (often termed fibrocytes) the cytoplasmic volume is reduced, the endoplasmic reticulum is sparse and the nucleus is flattened and
heterochromatic. Fibroblasts are usually adherent to the fibres of the matrix (collagen and elastin), which they lay down. In some highly cellular structures, e.g. liver, kidney and spleen, and in most lymphoid tissue, fibroblasts and delicate collagenous fibres (type III collagen; reticular fibres) form fibrocellular networks which are often called reticular tissue. The fibroblasts may then be termed reticular cells. page 36 page 37
Figure 3.14 Fibroblasts in connective tissue, surrounded by bundles of finely banded collagen fibrils which they secrete. (From Dr B Oakes, Anatomy and Cell Biology, Monash University, and by permission from Kerr JB 1999 Atlas of Functional Histology. London: Mosby.)
Fibroblasts are particularly active during wound repair, when they proliferate and lay down a fibrous matrix that becomes invaded by numerous blood vessels (granulation tissue). Contraction of wounds is at least in part caused by the shortening of specialized contractile fibroblasts (myofibroblasts) that arise in such areas. Fibroblast activity is influenced by various factors such as steroid hormone concentrations, dietary content and prevalent mechanical stresses. In vitamin C deficiency, there is an impairment of collagen formation.
ADIPOCYTES (LIPOCYTES, FAT CELLS) Adipocytes occur singly or in groups in many, but not all, connective tissues. They are numerous in adipose tissue (Fig. 3.15). Individually, the cells are oval or spherical in shape, but when packed together they are polygonal. They vary in diameter, averaging c.50 µm. Each cell consists of a peripheral rim of cytoplasm, in which the nucleus is embedded, surrounding a single large central globule of fat, which consists of glycerol esters of oleic, palmitic and stearic acids. There is a small accumulation of cytoplasm around the nucleus, which is oval in shape and compressed against the cell membrane by the lipid droplet, as is the Golgi complex. Many cytoskeletal filaments, some endoplasmic reticulum and a few mitochondria lie around the lipid droplet, which is in direct contact with the surrounding cytoplasm and not enclosed within a membrane. In sections of tissue not specially treated to preserve lipids, the lipid droplet is usually dissolved out by the solvents used in routine preparations, so that only the nucleus and the peripheral rim of cytoplasm surrounding a central empty space remain.
Figure 3.15 Adipose tissue in the fibrous pericardium of the heart. Adipocytes are distended polygonal cells filled with lipid, which is extracted by the tissue processing. This leaves only the plasma membranes with scant cytoplasm and nuclei, occasionally visible compressed against the cell periphery. An autonomic nerve fibre is shown sectioned longitudinally on its course through the adipose tissue. (Photograph by Sarah-Jane Smith.)
In neonates, a special form of adipose tissue known as brown fat is present in the interscapular region and may be more widespread. Brown fat is characterized by the presence of a large cell type in which the fat is present as several separate droplets and not as a single globule, and the mitochondria have unusually large and numerous cristae. These deposits of fat are concerned with heat production, mediated by mitochondria. The mobilization of fat is under nervous or hormonal control: noradrenaline (norepinephrine) released at sympathetic nerve endings in adipose tissue is particularly important in this respect.
MESENCHYMAL STEM CELLS Mesenchymal stem cells are normally inconspicuous cells in connective tissues. They are derived from embryonic mesenchyme (p. 207) and are able to differentiate into the mature cells of connective tissue during normal growth and development, in the turnover of cells throughout life and, most conspicuously, in the repair of damaged tissues in wound healing. There is evidence that, even in mature tissues, mesenchymal stem cells remain pluripotent and able to give rise to all the resident cells of connective tissues in response to local signals and cues.
MIGRANT CELLS MACROPHAGES Macrophages are typically numerous in connective tissues, where they are either attached to matrix fibres or are motile and migratory. They are relatively large cells, c.15-20 µm in diameter, with indented and relatively heterochromatic nuclei and a prominent nucleolus. Their cytoplasm is slightly basophilic, contains many lysosomes (p. 14) and typically has a foamy appearance under the light microscope. Macrophages are important phagocytes, and form part of the mononuclear phagocyte system (p. 80). They can engulf and digest particulate organic materials, such as bacteria, and are also able to clear dead or damaged cells from a tissue. They are also the source of a number of secreted cytokines that have profound effects on many other cell types. Macrophages are able to proliferate in connective tissues to a limited extent, but are derived primarily from haemopoietic stem cells (p. 80) in the bone marrow, which circulate in the blood as monocytes before migrating through vessel walls into connective tissues.
Many properties of macrophages in general connective tissue (Fig. 3.16) are similar to those of related cells in other sites. These include: circulating monocytes, from which they are derived; alveolar macrophages in the lungs; phagocytic cells in the lymph nodes, spleen and bone marrow; Kupffer cells of the liver sinusoids; microglial cells of the central nervous system.
LYMPHOCYTES Lymphocytes are typically present in small numbers (Fig. 3.16), and are numerous in general connective tissue only in pathological states, when they migrate in from adjacent lymphoid tissue or from the circulation. The majority are small cells (6-8 µm) with highly heterochromatic nuclei, but they enlarge when stimulated. Two major functional classes exist, termed B and T lymphocytes (p. 80). B lymphocytes originate in the bone marrow, then migrate to various lymphoid tissues, where they proliferate. When antigenically stimulated, they undergo further mitotic divisions, then enlarge as they mature, commonly in general connective tissues, to form plasma cells that synthesize and secrete antibodies (immunoglobulins). Mature plasma cells are rounded or ovoid, up to 15 µm across, and have an extensive rough endoplasmic reticulum. Their nuclei are spherical and have a characteristic 'clock-face' configuration of heterochromatin that is regularly distributed in peripheral clumps. The prominent Golgi complex is visible with a light microscope as a pale region to one side of the nucleus and the remaining cytoplasm is deeply basophilic because of the abundant endoplasmic reticulum. Mature plasma cells do not divide. T lymphocytes originate from precursors in bone marrow haemopoietic tissue, but later migrate to the thymus, where they develop T cell identity, before passing into the peripheral lymphoid system where they continue to multiply. When antigenically stimulated, T cells enlarge and their cytoplasm becomes filled with free polysome clusters. The functions of T lymphocytes are numerous: different subsets recognize and destroy virus-infected cells, tissue and organ grafts, or interact with B lymphocytes and several other defensive cell types (p. 73). page 37 page 38
Figure 3.16 Macrophages (M) in loose connective tissue. Their actively phagocytic properties are indicated by the brown material visible in their cytoplasm; these are residual bodies from engulfed particles. Other leukocytes are shown, including lymphocytes (L), eosinophils (Eo), neutrophils (N) and plasma cells (P). Elongated nuclei of fibroblasts (F) and erythrocytes (Er) in small vessels are also visible. (By permission from Young B, Heath JW 2000 Wheater's Functional Histology. Edinburgh: Churchill Livingstone.)
MAST CELLS Mast cells are important defensive cells which occur particularly in loose connective tissues and often in the fibrous capsules of certain organs such as the liver. They are characteristically numerous around blood vessels and nerves. Mast cells are round or oval, c.12 µm in diameter, with many filopodia extending from the cell surface. The nucleus is centrally placed and relatively small, and is surrounded by large numbers of prominent vesicles and a well developed Golgi apparatus, but scant endoplasmic reticulum. The vesicles have a high content of glycosaminoglycans and show a strongly positive reaction with the periodic acidSchiff stain for carbohydrates. The membrane-bounded vesicles vary in size and shape (mean diameter c.0.5 µm) and have a rather heterogeneous content of dense, lipid-containing material, which may be finely granular, lamellar or in the form of membranous whorls. The major granule components, many of them associated with inflammation, are the proteoglycan heparin, histamine, tryptase, superoxide dismutase, aryl sulphatase, !-hexosaminidase and various other enzymes, together with chemotactic factors for neutrophil and eosinophil granulocytes. Mast cells may be disrupted to release some or all of their contents, either by direct mechanical or chemical trauma, or after contact with particular antigens to which the body has previously been exposed. The consequences of granule release include alteration of capillary permeability, smooth muscle contraction, and activation and attraction to the locality of various other defensive cells. Responses to mast cell
degranulation may be localized, e.g. urticaria, or there may occasionally be a generalized response to the release of large amounts of histamine into the circulation (anaphylactic shock). Mast cells closely resemble basophil granulocytes of the general circulation but are now known to be derived from a separate lineage precursor. It is believed that they are generated in the bone marrow and circulate to the tissues as basophil-like cells, migrating through the capillary and venule walls to their final destination.
GRANULOCYTES (POLYMORPHONUCLEAR LEUKOCYTES) Neutrophil and eosinophil granulocytes are immigrant cells from the circulation. Relatively infrequent in normal connective tissues, their numbers may increase dramatically in infected tissues, where they are important components of cellular defence. Neutrophils are highly phagocytic, especially towards bacteria. The functions of eosinophils are less well understood. These cells are described further in Chapter 5.
Cells of specialized connective tissues Skeletal tissues, namely cartilage and bone, are generally classified with the connective tissues, but their structure and functions are highly specialized; they are described in Chapter 6. As with the general connective tissues, these specialized types are characterized by their extracellular matrix, which forms the major component of the tissues and is responsible for their properties. The resident cells are different from those in general connective tissues. Cartilage is populated by chondroblasts, which synthesize the matrix, and by mature chondrocytes. Bone matrix is elaborated by osteoblasts. Their mature progeny, osteocytes, are embedded within the matrix, which they help to mineralize, turn over and maintain. A third cell type, the osteoclasts, have a different lineage origin and are derived from haemopoietic tissue; they are responsible for bone degradation and remodelling in collaboration with osteoblasts.
Extracellular matrix The term extracellular matrix is applied collectively to the extracellular components of connective and supporting tissues. Essentially it consists of a system of insoluble protein fibres, adhesive glycoproteins and soluble complexes composed of carbohydrate polymers linked to protein molecules (proteoglycans and glycosaminoglycans), which bind water. The extracellular matrix distributes the mechanical stresses on tissues and also provides the structural environment of the cells embedded in it, forming a framework to which they adhere and on which they can move. With the exception of bone matrix, it provides a highly hydrated medium, through which metabolites, gases and nutrients can diffuse freely between cells and the blood vessels traversing it or, in the case of cartilage, passing nearby. There are many complex interactions between connective tissue cells and the extracellular matrix. The cells continually synthesize, secrete, modify and degrade extracellular matrix components, and themselves respond to contact with the matrix in the regulation of cell metabolism, proliferation and motility. page 38 page 39
Figure 3.17 Elastic fibres, seen as fine, dark, relatively straight fibres in a whole-mount preparation of mesentery, stained for elastin. The wavy pink bands are collagen bundles and oval grey nuclei are mainly of fibroblasts. (Photograph by Sarah-Jane Smith.)
The insoluble fibres are mainly of two types of structural protein: members of the collagen family, and elastin (Fig. 3.17). The interfibrillar matrix (ground substance) includes a number of adhesive glycoproteins that perform a variety of functions in connective tissues, including cell-matrix adhesion and matrix-cell signalling. These glycoproteins include fibronectin, laminin, tenascin and vitronectin, in addition to a number of other less well characterized proteins. The glycosaminoglycans of the interfibrillar matrix are, with one notable exception, post-translationally modified proteoglycan molecules in which long polysaccharide side chains are added to short core proteins during transit through the secretory pathway between the rough endoplasmic reticulum and the trans-Golgi network. The exception, the polymeric disaccharide, hyaluronan , has no protein core and is synthesized entirely by cell surface enzymes. For further reading on extracellular matrix molecules, see Pollard and Earnshaw (2002). Functional attributes of connective tissues vary and depend on the abundance of its different components. Collagen fibres resist tension, whereas elastin provides a measure of resilience to deformation by stretching. The highly hydrated, soluble polymers of the interfibrillar material (proteoglycans and glycosaminoglycans, mainly hyaluronan ) generally form a stiff gel resisting compressive forces. Thus tissues that are specialized to resist tensile forces (e.g. tendons) are rich in collagen fibrils, tissues that accommodate changes in shape and volume (e.g. mesenteries) are rich in elastic fibres and those that absorb compressive forces (e.g. cartilages) are rich in glycosaminoglycans and proteoglycans. In bone, mineral crystals take the place of most of the soluble polymers, and endow the tissue with incompressible rigidity.
FIBRILLAR MATRIX: COLLAGENS Collagens make up a very large proportion (c.30%) of all the proteins of the body. They consist of a wide range of related molecules that have various roles in the organization and properties of connective (and some other) tissues. The first collagen to be characterized was type I, the most abundant of all the collagens and a constituent of the dermis, fasciae, bone, tendon, ligaments, blood vessels and the sclera of the eyeball. The characteristic collagen of cartilage and the vitreous body of the eye, with a slightly different chemical composition, is type II, whereas type III is present in several tissues, including the dermis and blood
vessels, and type IV is in basal lamina. The other types are widely distributed in various tissues. Five of the collagens, types I, II, III, V and XI, form fibrils; types IV, VIII and X form sheets or meshworks; types VI, VII, IX, XII, XIV and XVIII have an anchoring or linking role; types XIII and XVII are transmembrane proteins. Biochemically, all collagens have a number of features in common. Unlike most other proteins, they contain high levels of hydroxyproline and all are composed of three polypeptides that form triple helices and are substantially modified posttranslationally. After secretion, individual molecules are further cross-linked to form stable polymers. Functionally, collagens are structural proteins with considerable mechanical strength. Just a few of their distinguishing structural features are described below. For further reading on the molecular structure and functions of the collagens, see Pollard and Earnshaw (2002). Type I collagen
Type I collagen is very widely distributed. It forms inextensible fibrils in which collagen molecules (triple helices) are aligned side by side in a staggered fashion, with three-quarters of the length of each molecule in contact with neighbouring molecules. The fibril has well-marked bands of charged and uncharged aminoacids arranged across it (these stain with heavy metals in a banding pattern that repeats every 65 nm in longitudinal sections viewed in the electron microscope (Fig. 3.14). Fibril diameters vary between tissues and with age. Developing tissues often have thinner fibrils than mature tissues. Corneal stroma fibrils are of uniform and thin diameter, whereas tendon fibrils may be up to 20 times thicker and quite variable. Tissues in which the fibrils are subject to high tensile loading tend to have thicker fibrils. Thick fibrils are composites of uniform thin fibrils with a diameter of 8-12 nm. The fibrils themselves are relatively flexible, but when mineralized (as in bone) or surrounded by high concentrations of proteoglycan (as in cartilage), the resulting fibre-reinforced composite materials are rigid. Fresh type I collagen fibres are white and glistening. They form bundles of various sizes that are generally visible at the light microscope level. The component fibres may leave one bundle and interweave with others. In many situations, collagen fibrils are laid down in precise geometrical patterns, in which successive layers alternate in direction, e.g. corneal stroma, where the high degree of order is essential for transparency. Tendons, aponeuroses and ligaments are also highly ordered tissues. Types II, III, V and XI collagens
Types II, III, V and XI collagens can also aggregate to form linear fibrils. Type II collagen occurs in extremely thin (10 nm) short fibrils in the vitreous humour and in very thick fibrils in ageing human cartilage. The amino-acid sequence and banding pattern are very similar to those of type I collagen, as are the posttranslational modifications of the triple helical protein molecule. The fine fibrils in the vitreous may fuse into thicker aggregates in older tissue. Type III collagen is very widely distributed, particularly in young and repairing tissues. It usually co-localizes with type I collagen, and covalent links between Type I and Type III collagen have been demonstrated. In skin, many fibrils are probably composites of Types I and III collagens. Reticular fibres
Fine branching and anastomosing reticular fibres form the supporting mesh framework of many glands, the kidney and lymphoreticular tissue (lymph nodes, spleen, etc.). Classically, these fibres stained intensely with silver salts, although
they are poorly stained using conventional histological techniques. They associate with basal laminae and are often found in the neighbourhood of collagen fibre bundles. Reticular fibres are formed principally of type III collagen. Elastin
Elastin is a 70 kDa protein, rich in the hydrophobic amino-acids valine and alanine. Elastic fibrils, which also contain fibrillin, are highly cross-linked via two elastin-specific amino-acids, desmosine and iso-desmosine, which are formed extracellularly from lysine residues. They are less widely distributed than collagen, yellowish in colour, typically cross-linked and are usually thinner (10-20 nm) than collagen fibrils. They can be thick, e.g. in the ligamenta flava and ligamentum nuchae. Unlike collagen type I, they show no banding pattern in the electron microscope. They stain poorly with routine histological stains, but are stained with orcein-containing preparations. They sometimes appear as sheets, as in the fenestrated elastic lamellae of the aortic wall. Elastin-rich structures stretch easily with almost perfect recoil, although they tend to calcify with age and lose elasticity. Elastin is highly resistant to attack by acid and alkali, even at high temperatures.
INTERFIBRILLAR MATRIX: GLYCOSAMINOGLYCANS page 39 page 40
The structural soluble polymers characteristic of the extracellular matrix are the acidic glycosaminoglycans, which are unbranched chains of repeating disaccharide units, each unit carrying one or more negatively charged groups (carboxylate or sulphate esters, or both). The anionic charge is balanced by cations (Na + , K + , etc.) in the interstitial fluid. Their polyanionic character endows the glycosaminoglycans with high osmotic activity, which helps keep the fibrils apart, confers stiffness on the porous gel that they collectively create and gives the tissue a varying degree of basophilia. Glycosaminoglycans are named according to the tissues in which they were first found, e.g. hyaluronan (vitreous body), chondroitins (cartilage), dermatan (skin), keratan (cornea), heparan (liver). This terminology is no longer relevant, as most glycosaminoglycans are very widely distributed, whereas, conversely, some corneas contain little or no keratan sulphate. Of the glycosaminoglycans, all except hyaluronan have short protein cores and are highly variable in their carbohydrate side chain structure. Hyaluronan
Hyaluronan was formerly called hyaluronic acid (or hyaluronate, as only the salt exists at physiological pH). It is a very large, highly hydrated molecule (25,000 kDa). Hyaluronan is found in all extracellular matrices and in most tissues and is a prominent component of embryonic and developing tissues. Hyaluronan is important in the aggregation of proteoglycans and link proteins that possess specific hyaluronan binding sites (e.g. laminin). Indeed, the very large aggregates that are formed may be the essential compression-resisting units in cartilage. Hyaluronan also forms very viscous solutions, which are probably the major lubricants in synovial joints. Because of its ability to bind water, it is often present in semi-rigid structures (e.g. vitreous humour in the eye), where it cooperates with sparse but regular meshworks of thin collagen fibrils. Proteoglycans
Proteoglycans have been classified according to the size of their protein core: their nomenclature is under review. The same core protein can bear different glycosaminoglycan side chains in different tissues. The functions of many proteoglycans are poorly understood. Some of the better known proteoglycans are: aggrecan in cartilage; perlecan in basal laminae; decorin associated with
fibroblasts in collagen fibril assembly; syndecan in embryonic tissues. Adhesive glycoproteins
These proteins include molecules that mediate adhesion between cells and the extracellular matrix, often in association with collagens, proteoglycans or other matrix components. All of them are glycosylated and they are, therefore, glycoproteins. General connective tissue contains the well known families of fibronectins (and osteonectin in bone), laminins and tenascins; there is a rapidly growing list of other glycoproteins associated with extracellular adhesion (Pollard & Earnshaw 2002). They possess binding sites for other extracellular matrix molecules and for cell adhesion molecules (p. 7), especially the integrins; in this way they enable cells selectively to adhere to appropriate matrix structures (e.g. the basal lamina). They also function as signalling molecules, which are detected by cell surface receptors and initiate changes within the cytoplasm (e.g. to promote the formation of hemidesmosomes or other areas of strong adhesion; reorganize the cytoskeleton; promote or inhibit locomotion and cell division). Fibronectin Fibronectin is a large glycoprotein consisting of a dimer joined by disulphide links. Each subunit is composed of a string of large repetitive domains linked by flexible regions. Fibronectin subunits have binding sites for collagen, heparin and cell surface receptors, especially integrins, and so can promote adhesion between all these elements. In connective tissues, the molecules are able to bind to cell surfaces in an orderly fashion, to form short fibronectin filaments. The liver secretes a related protein, plasma fibronectin, into the circulation. The selective adhesion of different cell types to the matrix during development and in postnatal life is mediated by numerous isoforms of fibronectin generated by alternative splicing. Isoforms found in embryonic tissues are also expressed during wound repair, when they facilitate tissue proliferation and cell movements; the adult form is re-expressed once repair is complete. Laminin Laminin is a large (850 kDa) flexible molecule composed of three polypeptide chains (designated ", !, and #). There are many isoforms of the different chains, and at least 18 types of laminin. The prototypical molecule has a cruciform shape, in which the terminal two-thirds are wound round each other to form the stem of a cross, and the shorter free ends form the upright and transverse members. Laminin bears binding sites for other extracellular matrix molecules such as heparan sulphate, type IV collagen and entactin and also for laminin receptor molecules situated in cell plasma membranes. Laminin molecules can assemble themselves into flat regular meshworks, e.g. in the basal lamina. Tenascin Tenascin is large glycoprotein composed of six subunits that are joined at one end to form a structure that resembles the spokes of a wheel. There is a family of tenascin molecules, generated by alternative splicing of the tenascin gene. Tenascin is abundant in embryonic tissues, but its distribution is restricted in the adult. It appears to be important in guiding cell migration and axonal growth in early development: it may either promote or inhibit these activities depending on the cell type and tenascin isoform.
Classification of connective tissues Connective and supporting tissues differ considerably in appearance, consistency and composition in different regions. These differences reflect local functional requirements and are related to the predominance of the cell types, the concentration, arrangement and types of fibre and the characteristics of the interfibrillar matrix. On these bases, general connective tissues can be classified
into irregular and regular types, according to the degree of orientation of their fibrous components.
IRREGULAR CONNECTIVE TISSUES Irregular connective tissues can be further subdivided into loose, dense and adipose connective tissue.
LOOSE (AREOLAR) CONNECTIVE TISSUE Loose connective tissue is the most generalized form (Fig. 3.18) and is extensively distributed. Its chief function is to bind structures together, while still allowing a considerable amount of movement to take place. It constitutes the submucosa in the digestive tract and other viscera lined by mucosae, and the subcutaneous tissue in regions where this is devoid of fat (e.g. eyelids, penis, scrotum and labia), and it surrounds muscles, vessels and nerves, connecting them with surrounding structures. It is present in the interior of organs, where it binds together the lobes and lobules of glands, forms the supporting layer (lamina propria) of mucosal epithelia and vascular endothelia, and lies within and between fascicles of muscle and nerve fibres. Loose connective tissue consists of a meshwork of thin collagen and elastin fibres interlacing in all directions to give a measure of both elasticity and tensile strength. The large meshes contain the soft, semi-fluid interfibrillar matrix or ground substance, and different connective tissue cells, which are scattered along the fibres or in the meshes. It also contains adipocytes, usually in small groups, and particularly around blood vessels. A variant of loose connective tissue occurs in the choroid and the sclera of the eye, where large numbers of pigment cells (melanocytes) are also present.
page 40 page 41
Figure 3.18 General connective tissue supporting a respiratory epithelium, with fibroblasts and other cells of connective tissue, bundles of collagen (red) and small blood vessels.
DENSE IRREGULAR CONNECTIVE TISSUE
Dense irregular connective tissue is found in regions that are under considerable mechanical stress and where protection is given to ensheathed organs. The matrix is relatively acellular and contains a high proportion of collagen fibres organized into thick bundles interweaving in three dimensions and imparting considerable strength. There are few active fibroblasts, which are usually flattened with heterochromatic nuclei. Dense irregular connective tissue occurs in: the reticular layer of the dermis; the connective tissue sheaths of muscle and nerves and the adventitia of large blood vessels; the capsules of various glands and organs (e.g. testis, sclera of the eye, periostea and perichondria).
ADIPOSE TISSUE A few adipocytes occur in loose connective tissue in most parts of the body. However, they constitute the principal component of adipose tissue, where they are embedded in a vascular loose connective tissue, usually divided into lobules by stronger fibrous septa carrying the larger blood vessels (Fig. 3.15). Adipose tissue only occurs in certain regions. In particular it is found in: subcutaneous tissue; the mesenteries and omenta; the female breast; bone marrow; as retroorbital fat behind the eyeball; around the kidneys; deep to the plantar skin of the foot; as localized pads in the synovial membrane of many joints. Its distribution in subcutaneous tissue shows characteristic age and sex differences. Fat deposits serve as energy stores, sources of metabolic lipids, thermal insulation (subcutaneous fat) and mechanical shock-absorbers (e.g. soles of the feet, palms of the hands, gluteal region and synovial membranes).
REGULAR CONNECTIVE TISSUES (Fig. 3.19) Regular connective tissues include highly fibrous tissues in which fibres are regularly orientated, either to form sheets such as fasciae and aponeuroses, or as thicker bundles such as ligaments or tendons. The direction of the fibres within these structures is related to the stresses which they undergo, moreover, fibrous bundles display considerable interweaving, even within tendons, which increases their structural stability and resilience. The fibroblasts that secrete the fibres may eventually become trapped within the fibrous structure, where they become compressed, relatively inactive cells with stellate profiles and small heterochromatic nuclei. Fibroblasts on the external surface may be active in continued fibre formation and they constitute a pool of cells available for repair of injured tissue. Although regular connective tissue is predominantly collagenous, some ligaments contain significant amounts of elastin, e.g. the ligamenta flava of the vertebral laminae and the vocal folds. The collagen fibres may form precise geometrical patterns, as in the cornea (p. 702).
Figure 3.19 Dense regular connective tissue in a tendon. Thick parallel bundles of type 1 collagen (pink) give tendon its white colour in life. The elongated nuclei of fibroblasts are visible between collagen bundles. (Photograph by Sarah-Jane Smith.)
MUCOID TISSUE Mucoid tissue is a fetal or embryonic type of connective tissue, found chiefly as a stage in the development of connective tissue from mesenchyme. It exists in Wharton's jelly, which forms the bulk of the umbilical cord, and consists substantially of extracellular matrix, largely made up of hydrated mucoid material and a fine meshwork of collagen fibres, in which nucleated, fibroblast-like cells with branching process are found. Fibres are usually rare in typical mucoid tissue, although the full-term umbilical cord contains perivascular collagen fibres. Postnatally, mucoid tissue is seen in the pulp of a developing tooth, the vitreous body of the eye (a persistent form of mucoid tissue that contains few fibres or cells) and the nucleus pulposus of the intervertebral disc.
© 2008 Elsevier
MUCOSA (MUCOUS MEMBRANE) A mucosa or mucous membrane lines many internal hollow organs of which the inner surfaces are moistened by mucus, such as the intestines, conducting portions of the airway, and the genital and urinary tracts. The mucosa proper consists of an epithelial lining, which may have mucosal glands opening on to its surface, the underlying loose connective tissue, the lamina propria, and a thin layer of smooth muscle, the muscularis mucosae. This last layer is either absent from some mucosae, or is replaced by a layer of elastic fibres. The term mucous membrane reflects the fact that these tissues can all be peeled away as a sheet or membrane from underlying structures; the plane of separation occurs along the muscularis mucosae. Submucosa is a layer of supporting connective tissue, which usually lies below the muscularis mucosae. It may contain mucous or seromucous submucosal glands, which convey their secretions through ducts to the mucosal surface. Most mucosae are also supported by one or more layers of smooth muscle, the muscularis externa. Contraction of this muscle may constrict the mucosal lumen (e.g. in the airway) or, where there are two or more muscle layers orientated in opposing directions (e.g. in the intestines), cause peristaltic movement of the viscus and the contents of its lumen. The outer surface of the muscle may be covered by a serosa or, where the structure is retroperitoneal or passes through the pelvic floor, by a connective tissue adventitia.
MUCUS Mucus is a viscous suspension of complex glycoproteins (mucins) of various kinds, and is secreted by scattered individual epithelial (goblet) cells, a secretory surface epithelium (e.g. the stomach) or mucous and sero-mucous glands. The precise composition of the mucus varies with the tissue and secretory cells that produce it. All mucins consist of filamentous core proteins to which are attached carbohydrate chains, usually branched; salivary mucus contains nearly 600 chains. Carbohydrate residues include glucose , fucose, galactose and Nacetylglucosamine (sialic acid). The terminals of some carbohydrate chains are identical with the blood group antigens of the ABO group in the majority of the population (secretors, bearing the secretor gene S e), and can be detected in salivary mucus by means of appropriate clinical tests. The long polymeric carbohydrate chains bind water and protect surfaces against drying; they also provide good lubricating properties. In concentrated form, mucins form viscous layers that protect the underlying tissues against damage. Synthesis of mucus starts in the rough endoplasmic reticulum. It is then passed to the Golgi complex, where it is conjugated with sulphated carbohydrates to form the glycoprotein, mucinogen, and this is exported in small dense vesicles, which swell as they approach the cell surface, with which they fuse before releasing their contents.
© 2008 Elsevier
SEROSA (SEROUS MEMBRANE) page 41 page 42
Serosa consists of a single layer of squamous mesothelial cells supported by an underlying layer of loose connective tissue that contains numerous blood and lymphatic vessels. Serosal cells are derived from embryonic mesenchyme and so share a common lineage with connective tissue cells. However, structurally they resemble squamous epithelia and they express keratin intermediate filaments. Serosa lines the pleural, pericardial and peritoneal cavities, covers the external surfaces of organs lying within those cavities and, in the abdomen, the mesenteries that envelop them. A potential space, filled with a small amount of protein-containing serous fluid (largely an exudate of interstitial fluid), exists between the outer parietal and the inner visceral layer of the serosa.
© 2008 Elsevier
FASCIA Fascia is a term applied to masses of connective tissue large enough to be visible to the unaided eye. Its structure is highly variable but, in general, collagen fibres in fascia tend to be interwoven and seldom show the compact, parallel orientation seen in tendons and aponeuroses. Fascia that is organized into condensations on the surfaces of muscles and other tissues is termed investing fascia, but this may not be its sole function. Between muscles that move extensively, it takes the form of loose areolar connective tissue and provides a degree of mechanical isolation. It constitutes the loose packing of connective tissue around peripheral nerves, blood and lymph vessels as they pass between other structures and often links them together as neurovascular bundles. It forms a dense connective tissue layer investing some large vessels, e.g. the common carotid and femoral arteries, and its presence here may be functionally significant, aiding venous return by approximating large veins to pulsating arteries.
SUPERFICIAL FASCIA Superficial fascia is a layer of loose connective tissue of variable thickness that merges with the deep aspect of the dermis. It is often adipose, particularly between muscle and skin. It allows increased mobility of skin, and the adipose component contributes to thermal insulation and constitutes a store of energy for metabolic use. Subcutaneous nerves, vessels and lymphatics travel in the superficial fascia; their main trunks lie in its deepest layer, where adipose tissue is sparse. In the head and neck, superficial fascia also contains a group of striated muscles (collectively termed the muscles of facial expression), which are a remnant of more extensive sheets of skin-associated musculature found in other mammals. The quantity and distribution of subcutaneous fat differs in the sexes. It is generally more abundant and widely distributed in females. In males, it diminishes from the trunk to the extremities; this distribution becomes more obvious in middle age, when the total amount increases in both sexes. There is an association with climate (rather than race), and superficial fat is more abundant in colder geographical regions. Superficial fascia is most distinct on the lower anterior abdominal wall, where it contains much elastic tissue and appears many-layered as it passes through the inguinal regions into the thighs. It is well differentiated in the limbs and the perineum, but is thin where it passes over the dorsal aspects of the hands and feet, the sides of the neck and face, around the anus and over the penis and scrotum, and is almost absent from the external ears. Superficial fascia is particularly dense in the scalp, palms and soles, where it is permeated by numerous strong connective tissue bands that bind the superficial fascia and skin to underlying structures.
DEEP FASCIA Deep fascia is also composed mainly of collagenous fibres, but these are compacted and in many cases arranged so regularly that the deep fascia may be
indistinguishable from aponeurotic tissue. In limbs, where deep fascia is well developed, the collagen fibres are longitudinal or transverse, and condense into tough, inelastic sheaths around the musculature. REFERENCES Pollard TD, Earnshaw WC 2002 Cell Biology. Philadelphia: Saunders. A cell biology text with extensive coverage of extracellular matrix components.
© 2008 Elsevier
4 SYSTEMIC OVERVIEW Nervous system The nervous system has two major divisions, the central nervous system (CNS) and the peripheral nervous system (PNS). The CNS consists of the brain and spinal cord and contains the majority of neuronal cell bodies. The PNS includes all nervous tissue outside the central nervous system and is subdivided into the cranial and spinal nerves, autonomic nervous system (ANS) (including the enteric nervous system of the gut wall, ENS) and special senses (taste, olfaction, vision, hearing and balance). It is composed mainly of the axons of sensory and motor neurones which pass between the CNS and the body. However, the ENS contains as many intrinsic neurones in its ganglia as the entire spinal cord, is not connected directly to the CNS, and may be considered separately as a third division of the nervous system. The CNS is derived from the neural tube (p. 207). The cell bodies of neurones are often grouped together in areas termed nuclei, or they may form more extensive layers or masses of cells collectively called grey matter. Neuronal dendrites and synaptic activity are mostly confined to areas of grey matter, and they form part of its meshwork of neuronal and glial processes which is collectively termed the neuropil (Fig. 4.1). Their axons pass into bundles of nerve fibres which tend to be grouped separately to form tracts. In the spinal cord, cerebellum, cerebral cortices (Chs 20, 22) and some other areas, concentrations of tracts constitute the white matter, so called because the axons are often ensheathed in myelin which is white when fresh (Figs 18.1, 18.4).
Figure 4.1 Neuronal somata, dendrites and axons in the CNS neuropil; their cytoskeleton has been stained using a gold method. The toluidine blue counterstaining reveals the nuclei of surrounding glial cells. (By permission from Young B, Heath JW 2000 Wheater's Functional Histology. Edinburgh: Churchill Livingstone.)
The PNS is composed of the axons of motor neurones situated inside the CNS, and the cell bodies of sensory neurones (grouped together as ganglia) and their processes. Sensory cells in dorsal root ganglia give off both centrally and peripherally directed processes: there are no synapses on their cell bodies. Ganglionic neurones of the ANS receive synaptic contacts from various sources. Neuronal cell bodies in peripheral ganglia are all derived embryologically from cells which migrate from the neural crest (p. 207). When the neural tube is formed during prenatal development its walls thicken greatly but do not completely obliterate the cavity within. The latter remains in the spinal cord as the narrow central canal, and in the brain it becomes greatly expanded to form a series of interconnected cavities called the ventricular system. In the fore- and hindbrains, parts of the neural tube roof do not generate nerve cells but become thin folded sheets of secretory tissue which are invaded by blood vessels and are called the choroid plexuses. The plexuses secrete cerebrospinal fluid which fills the ventricles and subarachnoid spaces (p. 282), and penetrates the intercellular spaces of the brain and spinal cord to create their
interstitial fluid. The CNS has a rich blood supply, which is essential to sustain its high metabolic rate. The blood-brain barrier places considerable restrictions on the substances which can diffuse from the bloodstream into the nervous tissue. Neurones encode information, conduct it over considerable distances, and then transmit it to other neurones or to various non-neural cells. The movement of this information within the nervous system depends on the rapid conduction of transient electrical impulses along neuronal plasma membranes. Transmission to other cells is mediated by secretion of neurotransmitters at special junctions either with other neurones (synapses), or with cells outside the nervous system, e.g. muscle cells (neuromuscular junctions), gland cells, adipose tissue, etc. and this causes changes in their behaviour. The nervous system contains large populations of non-neuronal cells, neuroglia or glia, which, whilst not electrically active in the same way, are responsible for creating and maintaining an appropriate environment in which neurones can operate efficiently. In the CNS, glia outnumber neurones by 10-50 times and consist of microglia and macroglia. Macroglia are further subdivided into three main types, oligodendrocytes, astrocytes and ependymal cells. The principal glial cell of the PNS is the Schwann cell. Satellite cells surround each neuronal soma in ganglia.
© 2008 Elsevier
NEURONES Most of the neurones in the CNS are either clustered into nuclei, columns or layers, or dispersed within grey matter. Neurones of the PNS are confined to ganglia. Irrespective of location, neurones share many general features, which are discussed here in the context of central neurones. Special characteristics of ganglionic neurones and their adjacent tissues are discussed on page 58. Neurones exhibit great variability in their size (cell bodies range from 5 to 100 µm diameter) and shapes. Their surface areas are extensive because most neurones display numerous narrow branched cell processes. They usually have a rounded or polygonal cell body (perikaryon or soma). This is a central mass of cytoplasm which encloses a nucleus and gives off long, branched extensions, with which most intercellular contacts are made. Typically, one of these processes, the axon, is much longer than the others, the dendrites (Fig. 4.2). Dendrites conduct electrical impulses towards a soma whereas axons conduct impulses away from it. page 43 page 44
Figure 4.2 A typical neurone (here, a motor neurone), showing the soma, part of the dendritic tree with dendritic spines and synaptic contacts, and an axon myelinated by oligodendrocytes and (in the PNS) by Schwann cells and ending at a neuromuscular junction.
Neurones can be classified according to the number and arrangement of their processes. Multipolar neurones (Figs 4.3, 22.9) are common: they have an extensive dendritic tree which arises from either a single primary dendrite or directly from the soma, and a single axon. Bipolar neurones, which typify neurones of the special sensory systems, e.g. retina (p. 712), have only a single dendrite which emerges from the soma opposite the axonal pole. Unipolar neurones which transmit general sensation, e.g. dorsal root ganglion neurones, have a single short process which bifurcates into peripheral and central processes (p. 58), an arrangement which arises by the fusion of the proximal axonal and dendritic processes of a bipolar neurone during development. Neurones are postmitotic cells and, with few exceptions, they are not replaced when lost.
Figure 4.3 Section through the cerebral cortex (mouse) stained by the Golgi method which demonstrates only a small proportion of the total neuronal population. (Specimen prepared by Martin Sadler, Division of Anatomy and Cell Biology, GKT School of Medicine, London.)
Figure 4.4 Large multipolar neuronal perikarya in the magnocellular part of the feline red nucleus, showing prominent Nissl granules, bases of dendrites and axon hillocks. The nuclei are euchromatic and vesicular, with prominent nucleoli. The small nuclei scattered in the surrounding neuropil are characteristic of the various categories of neuroglial cell. (Photograph by Kevin Fitzpatrick on behalf of GKT School of Medicine, London.)
SOMA The plasma membrane of the soma is unmyelinated and contacted by inhibitory and excitatory axosomatic synapses (p. 44): very occasionally, somasomatic and dendrosomatic contacts may be made. The non-synaptic surface is covered by either astrocytic or satellite oligodendrocyte processes. page 44 page 45
The cytoplasm of a typical soma (Fig. 4.2) is rich in rough and smooth endoplasmic reticulum and free polyribosomes, which reveals a high level of protein synthetic activity. Free polyribosomes often congregate in large groups associated with the rough endoplasmic reticulum. These aggregates of RNA-rich structures are visible by light microscopy as basophilic Nissl (chromatin) bodies or granules (Fig. 4.4). They are more obvious in large, highly active cells, such as spinal motor neurones, which contain large stacks of rough endoplasmic reticulum and polyribosome aggregates. Maintenance and turnover of cytoplasmic and membranous components are necessary in all cells: the huge total volume of cytoplasm within the soma and processes of many neurones requires a considerable commitment of protein synthetic machinery. Neurones synthesize other proteins (enzyme systems, etc.) which are involved in the production of neurotransmitters and in the reception and transduction of incoming stimuli. Various transmembrane channel proteins and enzymes are located at the surfaces of neurones where they are associated with ion transport. The apparatus for protein synthesis (including RNA and ribosomes) occupies the soma and dendrites, but is usually absent from axons. The nucleus is characteristically large, round and euchromatic, with one or more prominent nucleoli, as is typical of all cells engaged in substantial levels of protein synthesis. The cytoplasm contains many mitochondria and moderate numbers of lysosomes. Golgi complexes are typically seen close to the nucleus, near the bases of the main dendrites and opposite the axon hillock. The neuronal cytoskeleton is a prominent feature of its cytoplasm, and it gives shape, strength and rigidity to the dendrites and axons. Neurofilaments (the intermediate filaments of neurones) and microtubules are abundant: they occur in the soma and extend along dendrites and axons, in proportions which vary with the type of neurone and cell process. Bundles of neurofilaments constitute neurofibrils which can be seen by light microscopy in silver stained sections. Neurofilaments are heteropolymers of proteins assembled from three polypeptide subunits, NF-L (68 kDa), NF-M (160 kDa) and NF-H (200 kDa). NF-M and NF-H have long C-terminal domains which project as side arms from the assembled neurofilament and bind to neighbouring filaments. They can be heavily phosphorylated, particularly in the highly stable neurofilaments of mature axons, and are thought to give axons their tensile strength. Some axons are almost filled by neurofilaments. Dendrites usually have more microtubules than axons. Microtubules are important in axonal transport. Centrioles persist in mature postmitotic neurones, where they are concerned with the generation of
microtubules rather than cell division. Centrioles are associated with cilia on the surfaces of developing neuroblasts. Their significance, other than at some sensory endings (e.g. the olfactory mucosa, p. 568), is not known. Pigment granules appear in certain regions, e.g. neurones of the substantia nigra contain neuromelanin, probably a waste product of catecholamine synthesis (p. 344). In the locus coeruleus a similar pigment, rich in copper, gives a bluish colour to the neurones. Some neurones are unusually rich in certain metals, which may form a component of enzyme systems, e.g. zinc in the hippocampus and iron in the oculomotor nucleus. Ageing neurones especially in spinal ganglia accumulate granules of lipofuscin (senility pigment). They represent residual bodies, which are lysosomes packed with partially degraded lipoprotein material (corpora amylaceae).
DENDRITES Dendrites are highly branched, usually short processes which project from the soma (Fig. 4.2). The branching patterns of many dendritic arrays are probably established by random adhesive interactions between dendritic growth cones and afferent axons which occur during development. There is an overproduction of dendrites in early development, which is pruned in response to functional demand as the individual matures and information is processed through the dendritic tree. There is evidence that dendritic trees may be plastic structures throughout adult life, expanding and contracting as the traffic of synaptic activity varies through afferent axodendritic contacts (for review see Berry 1991). Groups of neurones with similar functions have a similar stereotypic tree structure (Fig. 4.5), suggesting that the branching patterns of dendrites are important determinants of the integration of afferent inputs which converge on the tree. Dendrites differ from axons in many respects. They represent the afferent rather than the efferent system of the neurone, and receive both excitatory and inhibitory axodendritic contacts. They may also make dendrodendritic and dendrosomatic connections (see Fig. 4.8), some of which are reciprocal. Synapses occur either on small projections called dendritic spines or on the smooth dendritic surface. Dendrites contain ribosomes, smooth endoplasmic reticulum, microtubules, neurofilaments, actin filaments and Golgi complexes. The neurofilament proteins of dendrites are poorly phosphorylated. Dendrite microtubules express the microtubule-associated protein (MAP-2) almost exclusively in comparison with axons.
Figure 4.5 Purkinje neurone from the cerebellum of a rat stained by the Golgi-Cox method, showing the extensive two-dimensional array of dendrites. (Provided by Martin Sadler and M Berry, Division of Anatomy and Cell Biology, GKT School of Medicine, London.)
Dendritic spine shapes range from simple protrusions to structures with a slender stalk and expanded distal end. Most spines are not more than 2 µm long, and have one or more terminal expansions, but they can also be short and stubby, branched or bulbous. Free ribosomes and polyribosomes are concentrated at the base of the spine. Ribosomal accumulations near synaptic sites provide a mechanism for activity-dependent synaptic plasticity through the local regulation of protein synthesis.
AXONS The axon originates either from the soma or from the proximal segment of a dendrite, at a specialized region, the axon hillock (Fig. 4.2), which is free of Nissl granules. Action potentials are initiated here. The axonal plasma membrane (axolemma) is undercoated at the hillock by a concentration of cytoskeletal molecules, including spectrin and actin fibrils, which are thought to be important in
anchoring numerous voltage sensitive channels to the membrane. The axon hillock is unmyelinated and often participates in inhibitory axo-axonal synapses. This region of the axon is unique because it contains ribosomal aggregates immediately below the postsynaptic membrane. page 45 page 46
When present, myelin begins at the distal end of the axon hillock. Myelin thickness and internodal segment lengths are positively correlated with axon diameter. In the PNS unmyelinated axons are embedded in Schwann cell cytoplasm, in the CNS they lie free in the neuropil. Nodes of Ranvier are specialized constricted regions of myelin-free axolemma where action potentials are generated and where an axon may branch. The density of sodium channels in the axolemma is highest at nodes of Ranvier, and very low along internodal membranes. In contrast, sodium channels are spread more evenly within the axolemma of unmyelinated axons. Fast potassium channels are also present in the paranodal regions of myelinated axons. Fine processes of glial cytoplasm (astrocyte in the CNS, Schwann cell in the PNS) surround the nodal axolemma. The terminals of an axon are unmyelinated. They expand into presynaptic boutons which may form connections with axons, dendrites, neuronal somata or, in the periphery, muscle fibres, glands and lymphoid tissue. They may themselves be contacted by other axons, forming axoaxonal presynaptic inhibitory circuits. Further details of neuronal microcircuitry are given in Kandel & Schwartz (2000). Axons contain microtubules, neurofilaments, mitochondria, membrane vesicles, cisternae, and lysosomes: they do not usually contain ribosomes or Golgi complexes, except at the axon hillock. However, ribosomes are found in the neurosecretory fibres of hypothalamo-hypophyseal neurones which contain the mRNA of neuropeptides. Organelles are differentially distributed along axons, e.g. there is a greater density of mitochondria and membrane vesicles in the axon hillock, at nodes, and in presynaptic endings. Axonal microtubules are interconnected by cross-linking microtubule-associated proteins (MAPs) of which tau is the most abundant. Microtubules have an intrinsic polarity (p. 18): in axons all microtubules are uniformly orientated with their rapidly growing ends directed away from the soma towards the axon terminal. Neurofilament proteins ranging from high to low molecular weights are highly phosphorylated in mature axons, whereas growing and regenerating axons express a calmodulin-binding membrane-associated phosphoprotein, growth-associated protein-43 (GAP-43), as well as poorly phosphorylated neurofilaments. Axons respond differently to injury, depending on whether the damage occurs in the CNS or PNS. The glial microenvironment of a damaged central axon does not facilitate regrowth, and reconnection with original synaptic targets does not normally occur. In the PNS, the glial microenvironment is capable of facilitating axonal regrowth, however the functional outcome of clinical repair of a large mixed peripheral nerve, especially if the injury occurs some distance from the target organ, or produces a long defect in the damaged nerve, is frequently unsatisfactory. Axoplasmic flow
Neuronal organelles and cytoplasm are in continual motion. Bidirectional streaming of vesicles along axons results in a net transport of materials from the soma to the terminals, with more limited movement in the opposite direction. Two major types of transport occur, one slow, and one relatively fast. Slow axonal transport is a bulk flow of axoplasm only in the anterograde direction, carrying cytoskeletal proteins and soluble, non-membrane bound proteins at a rate of c.0.1-3 mm a day. In contrast, fast axonal transport carries vesicular material at c.200 mm a day in the retrograde direction and c.40 mm per day anterogradely. Rapid flow depends on microtubules. Vesicles with side projections line up along microtubules and are transported along them by their side-arms. Two microtubule-based motor proteins with ATPase activity are involved in fast transport. Kinesin family proteins are responsible for the fast component of anterograde transport, and cytoplasmic dynein is responsible for retrograde transport. Fast anterograde transport carries vesicles, including synaptic vesicles containing neurotransmitters, from the soma to the axon terminals. Retrograde axonal transport accounts for the flow of mitochondria, endosomes and lysosomal autophagic vacuoles from the axonal terminals into the soma. Retrograde transport mediates the movement of neurotrophic viruses, e.g. herpes zoster, rabies and polio, from peripheral terminals, and their subsequent concentration in the neuronal soma.
SYNAPSES Transmission of impulses across specialized junctions (synapses) between two neurones is largely chemical. It depends on the release of neurotransmitters from the presynaptic side: this causes a change in the electrical state of the postsynaptic neuronal membrane, resulting in either its depolarization or hyperpolarization. The patterns of axonal termination vary considerably. A single axon may synapse with one neurone, e.g. climbing fibres ending on cerebellar Purkinje neurones, or more often with many, e.g. cerebellar parallel fibres, which provide an extreme example of this phenomenon (p. 357). In synaptic glomeruli, e.g. in the olfactory bulb, and synaptic cartridges, groups of synapses between two or many neurones form interactive units encapsulated by neuroglia (Fig. 4.6). Electrical synapses (direct communication via gap junctions) are rare in the human CNS and are confined largely to groups of neurones with tightly coupled activity, e.g. the inspiratory centre in the medulla. They will not be discussed further here. Classification of chemical synapses
Chemical synapses have an asymmetric structural organization (see Figs 4.7, 4.8) in keeping with the unidirectional nature of their transmission. Typical chemical synapses share a number of important features. They all display an area of presynaptic membrane apposed to a corresponding postsynaptic membrane: the two are separated by a narrow (20-30 nm) gap, the synaptic cleft. Synaptic vesicles containing neurotransmitter lie on the presynaptic side, clustered near an area of dense material on the cytoplasmic aspect of the
presynaptic membrane. A corresponding region of submembrane density is present on the postsynaptic side. Together these define the active zone, the area of the synapse where neurotransmission takes place. Chemical synapses can be classified according to a number of different parameters, including the neuronal regions forming the synapse; their ultrastructural characteristics; the chemical nature of their neurotransmitter(s); their effects on the electrical state of the postsynaptic neurone. The following classification is limited to associations between neurones. Neuromuscular junctions share many (though not all) of these parameters, and are often referred to as peripheral synapses. They are described separately on page 64.
page 46 page 47
Figure 4.6 The arrangement of complex synaptic units. A, Synaptic glomerulus with excitatory ('+') and inhibitory ('-') synapses grouped around a central dendritic terminal
expansion. The directions of transmission are shown by the arrows. B, Synaptic cartridge with a group of synapses surrounding a dendritic segment. Each complex unit is enclosed within a glial capsule (green).
Figure 4.7 Electron micrographs demonstrating various types of synapse. A, A pale cross-section of a dendrite upon which two synaptic boutons end. The upper bouton contains round vesicles, and the lower bouton contains flattened vesicles of the small type. A number of pre- and postsynaptic thickenings mark the specialized zones of contact. B, A type I synapse containing both small, round, clear vesicles and also large dense-cored vesicles of the neurosecretory type. C, A large terminal bouton of an optic nerve afferent fibre, which is making contact with a number of postsynaptic processes, in the dorsal lateral geniculate nucleus of the rat. One of the postsynaptic processes (*) also receives a synaptic contact from a bouton containing flattened vesicles (right). D, Reciprocal synapses between two neuronal processes in the olfactory bulb. (A, B, D and E, provided by AR Lieberman, Department of Anatomy, University College, London.)
Synapses can occur between almost any surface regions of the participating neurones. The most common type occurs between an axon and either a dendrite or a soma, when the axon is expanded as a small bulb or bouton (Figs 4.7, 4.8). This may be a terminal of an axonal branch (terminal bouton) or one of a row of bead-like endings, when the axon makes contact at several points, often with more than one neurone (bouton de passage). Boutons may synapse with dendrites, including dendritic spines or the flat surface of a dendritic shaft; a soma, usually on its flat surface, but occasionally on spines; the axon hillock; the terminal boutons of other axons. The connection is classified according to the direction of transmission, with the incoming terminal region named first. Most common are axodendritic synapses, although axosomatic connections are frequent. All other possible combinations are found, but they are less common, i.e. axoaxonic, dendroaxonic, dendrodendritic, somatodendritic or somatosomatic. Axodendritic and axosomatic synapses occur in all regions of the CNS and in autonomic ganglia including those of the ENS. The other types appear restricted to regions of complex interaction between larger sensory neurones and microneurones, e.g. in the thalamus. Ultrastructurally, synaptic vesicles may be internally clear or dense, and of different size (loosely categorized as small or large) and shape (round, flat or pleomorphic, i.e. irregularly shaped). The submembranous densities may be thicker on the postsynaptic than on the presynaptic side (asymmetric synapses), or equivalent in thickness (symmetrical synapses). Synaptic ribbons are found at
sites of neurotransmission in the retina and inner ear. They have a distinctive morphology, in that the synaptic vesicles are grouped around a ribbon- or rod-like density orientated perpendicular to the cell membrane (Fig. 4.8). page 47 page 48
Figure 4.8 The structural arrangements of different types of synaptic contact. A, The gap junction (B) and the desmosome (E) are without synaptic significance. Excitatory synaptic boutons are shown (C, G) containing small spherical translucent vesicles. D: a bouton with dense-cored, catecholamine-containing vesicles; F: an inhibitory synapse containing small flattened vesicles; H: a reciprocal synaptic structure between two dendritic profiles, inhibitory towards dendrite A and excitatory in the opposite direction; I: an inhibitory synapse containing large flattened vesicles. J and K: two serial synapses; J is excitatory to the dendrite; K is inhibitory to J; L: a neurosecretory ending adjacent to a vascular channel (M), surrounded by a fenestrated endothelium. All the boutons in this diagram are of the terminal type, except G which is a bouton de passage. B, Axosomatic and axoinitial segment synapses: RA, asymmetrical synapses with rounded vesicles; FS, symmetrical synapses with flattened vesicles. C, Ribbon synapse: triad at base of retinal rod.
Synaptic boutons make obvious close contacts with postsynaptic structures, but many other terminals lack specialized contact zones. Areas of transmitter release occur in the varicosities of unmyelinated axons, where effects are sometimes diffuse, e.g. the aminergic pathways of the basal ganglia (p. 428) and in autonomic fibres in the periphery (p. 65). In some instances, such axons may ramify widely throughout extensive areas of the brain and affect the behaviour of very large populations of neurones, e.g. the diffuse cholinergic innervation of the cerebral cortices. Pathological degeneration of these pathways can therefore cause widespread disturbances in neural function.
Neurones express a variety of neurotransmitters, either as one class of neurotransmitter per cell or more often as several. Good correlations exist between some types of transmitter and specialized structural features of synapses. In general, asymmetric synapses with relatively small spherical vesicles are associated with acetylcholine (ACh), glutamate, serotonin (5hydroxytryptamine, 5-HT), and some amines; those with dense-core vesicles include many peptidergic synapses and other amines (e.g. noradrenaline (norepinephrine), adrenaline (epinephrine ), dopamine). Symmetrical synapses with flattened or pleomorphic vesicles have been shown to contain either !aminobutyric acid (GABA) or glycine . Neurosecretory endings found in various parts of the brain and in neuroendocrine glands have many features in common with presynaptic boutons. They all contain peptides or glycoproteins within dense- core vesicles of characteristic size and appearance, which are often ellipsoidal or irregular in shape, and are relatively large, e.g. oxytocin and vasopressin vesicles in the neurohypophysis may be up to 200 nm across. Synapses may cause depolarization or hyperpolarization of the postsynaptic membrane, depending on the neurotransmitter released and the classes of receptor molecule in the postsynaptic membrane. Depolarization of the postsynaptic membrane results in excitation of the postsynaptic neurone, whereas hyperpolarization has the effect of transiently inhibiting electrical activity. Subtle variations in these responses may also occur at synapses where mixtures of neuromediators are present and their effects are integrated. Type I and II synapses
There are two broad categories of synapse: type I synapses, in which the subsynaptic zone of dense cytoplasm is thicker than on the presynaptic side, and type II synapses, in which the two zones are more symmetrical but thinner. Other differences include the widths of the synaptic clefts, which are c.30 nm in type I and c.20 nm in type II synapses, and their vesicle content. Type I boutons contain a predominance of small spherical vesicles c.50 nm in diameter, and type II boutons contain a variety of flat forms. The general principle found to apply in broad outline throughout the CNS classifies type I synapses as excitatory and type II as inhibitory. In a few instances types I and II synapses are found in close proximity, orientated in opposite directions across the synaptic cleft (a reciprocal synapse). Mechanisms of synaptic activity
Synaptic activation begins with arrival of one or more action potentials at the presynaptic bouton, which causes the opening of voltage-sensitive calcium channels in the presynaptic membrane. The response time in typical fast-acting synapses is then very rapid; classic neurotransmitter (e.g. ACh) is released in less than a millisecond, which is faster than the activation time of a classic second messenger system on the presynaptic side. The influx of calcium activates Ca 2+-dependent protein kinases. This uncouples synaptic vesicles from a spectrin-actin meshwork within the presynaptic ending, to which they are bound via synapsins I and II. The vesicles dock with the presynaptic membrane, through
processes not yet fully understood, and their membranes fuse to open a pore through which neurotransmitter diffuses into the synaptic cleft. Once the vesicle has discharged its contents, its membrane is incorporated into the presynaptic plasma membrane and is then more slowly recycled back into the bouton by endocytosis around the edges of the active site. The time between endocytosis and re-release may be c.30 seconds; newly recycled vesicles compete randomly with previously stored vesicles for the next cycle of neurotransmitter release. The fusion of vesicles with the presynaptic membrane is responsible for the observed quantal behaviour of neurotransmitter release, both during neural activation and spontaneously, in the slightly leaky resting condition. page 48 page 49
Postsynaptic events vary greatly, depending on the receptor molecules and their related molecular complexes. Receptors are generally classed as either ionotropic or metabotropic. Ionotropic receptors function as ion channels, so that conformational changes induced in the receptor protein when it binds the neurotransmitter cause the opening of an ion channel within the same protein assembly, thus causing a voltage change within the postsynaptic cell. Examples are the nicotinic ACh receptor and the N-methyl-D-aspartate (NMDA) glutamate receptor. Alternatively, the receptor and ion channel may be separate molecules, coupled by G-proteins, some via a complex cascade of chemical interactions (a second messenger system), e.g. the adenylate cyclase pathway (p. 9). Postsynaptic effects are generally rapid and short-lived, because the transmitter is quickly inactivated either by an extracellular enzyme (e.g. acetylcholinesterase, AChE), or by uptake by neuroglial cells. Examples of such metabotropic receptors are the muscarinic Ach receptor and 5-HT receptor. Neurohormones
Neurohormones are included in the range of transmitter activities. They are synthesized in neurones and released into the blood circulation by exocytosis at synaptic terminal-like structures. As with classic endocrine gland hormones (Ch. 9, they may act at great distances from their site of secretion. Neurones secrete into the cerebrospinal fluid or local interstitial fluid to affect other cells, either diffusely or at a distance. To encompass this wide range of phenomena the general term neuromediation has been used, and the chemicals involved are called neuromediators. Neuromodulators Some neuromediators do not appear to affect the postsynaptic membrane directly, but they can affect its responses to other neuromediators, either enhancing their activity (increasing the immediate response in size, or causing a prolongation), or perhaps limiting or inhibiting their action. These substances are called neuromodulators. A single synaptic terminal may contain one or more neuromodulators in addition to a neurotransmitter, usually (though not always) in separate vesicles. Neuropeptides (see below and p. 180) are nearly all neuromodulators, at least in some of their actions. They are stored within dense granular synaptic vesicles of various sizes and appearances.
Development and plasticity of synapses
Embryonic synapses first appear as inconspicuous dense zones flanking synaptic clefts. Immature synapses often appear after birth, suggesting that they may be labile, and are reinforced if transmission is functionally effective, or withdrawn if redundant. This is implicit in some theories of memory, which postulate that synapses are modifiable by frequency of use, to establish preferential conduction pathways. Evidence from hippocampal neurones suggests that even brief synaptic activity can increase the strength and sensitivity of the synapse for some hours or longer (long-term potentiation, LTP). During early postnatal life, the normal developmental increase in numbers and sizes of synapses and dendritic spines depends on the degree of neural activity and is impaired in areas of damage or functional deprivation. Neurotransmitters
Until recently the molecules known to be involved in chemical synapses were limited to a fairly small group of classic neurotransmitters, e.g. ACh, noradrenaline, adrenaline, dopamine and histamine, all of which had well-defined rapid effects on other neurones, muscle cells or glands. However, many synaptic interactions cannot be explained on the basis of classic neurotransmitters, and it now appears that other substances, particularly some amino acids such as glutamate, glycine , aspartate, GABA and the monoamine, serotonin, also function as transmitters. Substances first identified as hypophyseal hormones or as part of the dispersed neuroendocrine system of the alimentary tract, can be detected widely throughout the CNS and PNS, often associated with functionally integrated systems. Many of these are peptides: more than 50 (together with other candidates), function mainly as neuromodulators and influence the activities of classic transmitters. Acetylcholine
Acetylcholine (ACh) is perhaps the most extensively studied neurotransmitter of the classic type. Its precursor, choline, is synthesized in the neuronal soma and transported to the axon terminals where it is acetylated by the enzyme choline acetyl transferase (ChAT), and stored in clear spherical vesicles c.50 nm in diameter. ACh is synthesized by motor neurones and released at all their motor terminals on skeletal muscle and at synapses in parasympathetic and sympathetic ganglia. Many parasympathetic, and some sympathetic, ganglionic neurones are also cholinergic. In some sites ACh is also associated with the degradative extracellular enzyme acetyl cholinesterase (AChE), e.g. at neuromuscular junctions. The effects of ACh on nicotinic receptors (i.e. those in which nicotine is an agonist) are rapid and excitatory. In the peripheral ANS, the slower, more sustained excitatory effects of cholinergic autonomic endings are mediated by muscarinic receptors via a second messenger system. Monoamines
Monoamines include the catecholamines (noradrenaline, adrenaline and dopamine), the indoleamine serotonin (5-hydroxytryptamine, 5-HT) and histamine.
Neurones which synthesize the monoamines include sympathetic ganglia and their homologues, the chromaffin cells of the suprarenal medulla (pp. 180, 1247) and paraganglia (p. 181). Within the CNS, their somata lie chiefly in the brainstem, although their axons spread and ramify widely into all parts of the nervous system. Monoaminergic cells are also present in the retina (p. 710). Noradrenaline is the chief transmitter present in sympathetic ganglionic neurones with endings in various tissues, notably smooth muscle and glands, and in other sites including adipose and haemopoietic tissues, and the corneal epithelium. It is also found at widely distributed synaptic endings within the CNS, many of them terminals of neuronal somata situated in the locus coeruleus in the medullary floor. The actions of noradrenaline depend on its site of action, and vary with the type of postsynaptic receptor. In some cases, e.g. the neurones of the submucosal plexus of the intestine and of the locus coeruleus, it is strongly inhibitory via actions on the "2-adrenergic receptor, whereas the #-receptors, e.g. of vascular smooth muscle, mediate depolarization and therefore vasoconstriction. Adrenaline is present in central and peripheral nervous pathways and occurs with noradrenaline in the suprarenal medulla. Both of these monoamines are found in dense-cored synaptic vesicles c.50 nm diameter. Dopamine is a neuromediator of considerable clinical importance, present mainly in the CNS, where it is found in neurones with cell bodies in the telencephalon, diencephalon and mesencephalon. A major dopaminergic neuronal population in the midbrain constitutes the substantia nigra, so called because its cells contain neuromelanin, a black granular byproduct of dopamine synthesis. Dopaminergic endings are particularly numerous in the corpus striatum, limbic system and cerebral cortex. Pathological reduction in dopaminergic activity has widespread effects on motor control, affective behaviour and other neural activities, as seen in Parkinson's syndrome. Structurally, dopaminergic synapses contain numerous dense-cored vesicles resembling those of noradrenaline. Serotonin and histamine are found in neurones mainly in the CNS. Serotonin is synthesized chiefly in small median neuronal clusters of the brainstem, mainly in the raphe nuclei, whose axons spread and branch extensively throughout the entire brain and spinal cord. Synaptic terminals contain rounded, clear vesicles c.50 nm diameter and are of the asymmetrical type. Histaminergic neurones appear to be relatively sparse, and are restricted largely to the hypothalamus. Amino acids
The best understood amino acid is GABA, which is a major inhibitory transmitter released at the terminals of local circuit neurones within the brainstem and spinal cord (e.g. the recurrent inhibitory Renshaw loop; p. 307), cerebellum (as the main transmitter of Purkinje neurones) and elsewhere. It is stored in flattened or pleomorphic vesicles within symmetrical synapses: it may be inhibitory to the postsynaptic neurone, or it may mediate either presynaptic inhibition or facilitation, depending on the synaptic arrangement. Glutamate and aspartate are major excitatory transmitters present widely within the CNS, including the major projection pathways from the cortex to the thalamus,
tectum, substantia nigra and pontine nuclei. They are found in the central terminals of the auditory and trigeminal nerves, and glutamate is present in the terminals of parallel fibres ending on Purkinje cells in the cerebellum. Structurally, they are associated with asymmetrical synapses containing small (c.30 nm) round, clear synaptic vesicles. Glycine is a well-established inhibitory transmitter of the CNS, particularly the lower brainstem and spinal cord, where it is mainly found in local circuit neurones. page 49 page 50
Nitric oxide
Nitric oxide (NO) is of considerable importance at autonomic and enteric synapses, where it mediates smooth muscle relaxation. NO has been implicated in the mechanism of long-term potentiation. The gas is able to diffuse freely through cell membranes, and so is not under such tight quantal control as vesiclemediated neurotransmission. Neuropeptides
Many neuropeptides coexist with other neuromediators in the same synaptic terminals. As many as three peptides often share a particular ending with a wellestablished neurotransmitter, in some cases within the same synaptic vesicles. Some peptides occur both in the CNS and PNS, particularly in the ganglion cells and peripheral terminals of the ANS, whilst others are entirely restricted to the CNS. Only a few examples are given here. Most of the neuropeptides are classified according to the site where they were first discovered; for example, the gastrointestinal peptides were initially found in the gut wall, and a group first associated with the pituitary gland includes releasing hormones, adenohypophyseal and neurohypophyseal hormones. Some of these peptides are closely related to each other in their chemistry, because they are derived from the same gene products (e.g. the pro-opiomelanocortin group), which are cleaved to produce smaller peptides. Substance P (SP) was the first of the peptides to be characterized as a gastrointestinal neuromediator. It consists of 11 amino acid residues and is a major neuromediator in the brain and spinal cord. It occurs in c.20% of dorsal root and trigeminal ganglion cells, in particular in small nociceptive neurones. It is also present in some fibres of the facial, glossopharyngeal and vagal nerves. Within the CNS, SP is present in several apparently unrelated major pathways. It is contained within large granular synaptic vesicles. Its known action is prolonged postsynaptic excitation. Vasoactive intestinal polypeptide (VIP), another gastrointestinal peptide, is widely present in the CNS, where it is probably an excitatory neurotransmitter or neuromodulator. Its distribution includes distinctive bipolar neurones of the cerebral cortex, small dorsal root ganglion cells, particularly of the sacral region, the median eminence of the hypothalamus, where it may be involved in endocrine regulation, intramural ganglion cells of the gut wall and sympathetic ganglia. Somatostatin (ST, somatotropin release inhibiting factor) has a broad distribution
within the nervous system, and may be a central neurotransmitter or neuromodulator. It occurs in small dorsal root ganglion cells. #-Endorphin, leu- and metenkephalins and the dynorphins belong to a group of peptides (naturally occurring opiates) which have aroused much interest because of their analgesic properties. They bind to opiate receptors in the brain where, in general, their action seems to be inhibitory. The enkephalins have been localized in many areas of the brain, particularly the septal nuclei, amygdaloid complex, basal ganglia and hypothalamus, from which it has been inferred that they are important mediators in the limbic system and in the control of endocrine function. They have been implicated strongly in the central control of pain pathways, because they are found in the periaqueductal grey matter of the midbrain, a number of reticular raphe nuclei, the spinal nucleus of the trigeminal nerve and the substantia gelatinosa of the spinal cord. The enkephalinergic pathways exert an important presynaptic inhibitory action on nociceptive afferents in the spinal cord and brainstem. Like many other neuromediators, the enkephalins also occur widely in other parts of the brain in lower concentrations.
© 2008 Elsevier
CENTRAL GLIA Glial (neuroglial) cells vary considerably in type and number in different regions of the CNS. There are two major groups which are classified according to origin. Macroglia arise within the neural plate, in parallel with neurones, and constitute the great majority of glial cells. Microglia are smaller cells, generally considered to be monocytic in origin, and are derived from haemopoietic tissue (Fig. 4.9).
ASTROCYTES
page 50 page 51
Figure 4.9 The different types of non-neuronal cell in the CNS and their structural organization and interrelationships with each other and with neurones.
Astrocytes are star-shaped glia whose processes ramify through the entire central neuropil (Fig. 4.9). Their processes are functionally coupled at gap junctions and form an interconnected network which ensheathes all neurones, except at synapses and along the myelinated segments of axons. Astrocyte processes terminate as end-feet at the basal lamina of blood vessels and where they form the glia limitans (glial limiting membrane) at the pial surface (p. 284). Ultrastructurally, astrocytes typically have a pale nucleus with a narrow rim of heterochromatin, although this is variable. They have a pale cytoplasm containing glycogen, lysosomes, Golgi complexes and bundles of glial intermediate filaments within their processes (these last are found particularly in fibrous astrocytes, which occur predominantly in white matter). Glial intermediate filaments are formed from glial fibrillary acidic protein (GFAP): its presence can be used clinically to identify tumour cells of glial origin. A second morphological type of astrocyte, the protoplasmic astrocyte, is found mainly in grey matter. The significance of these subtypes is unclear: there are few known functional differences between fibrous and protoplasmic astrocytes. Astrocytes are thought to provide a network of communication in the brain via interconnecting low resistance gap junctional complexes (p. 7). They signal to each other using intracellular calcium wave propagation, triggered by synaptically released glutamate. Functionally, this may coordinate astrocyte activities, e.g. ion (particularly potassium) buffering; neurotransmitter uptake and metabolism (e.g. of excess glutamate, which is excitotoxic); membrane transport; the secretion of
peptides, amino acids , trophic factors etc., essential for efficient neuronal activity. Injury to the CNS induces astrogliosis, which is seen as a local increase in the number and size of cells expressing GFAP and in the extent of their meshwork of processes which form a glial scar. It is thought that the local glial scar environment, which may include oligodendrocytes and myelin debris, inhibits regeneration of CNS axons and/or fails to provide the necessary stimuli for axonal regrowth. Pituicytes are glial cells found in the neural parts of the pituitary gland, the infundibulum and neurohypophysis. They resemble astrocytes, but their processes end mostly on endothelial cells in the neurohypophysis and tuber cinereum. Blood-brain barrier
Proteins circulating in the blood enter most tissues of the body except those of the brain, spinal cord or peripheral nerves. This concept of a blood-brain barrier (and blood-nerve barrier) covers many substances, some of which are actively transported across the blood-brain barrier, whereas others are actively excluded. The blood-brain barrier is located at the capillary endothelium within the brain. It depends upon the presence of tight junctions between endothelial cells and a relative lack of transcytotic vesicular transport (pp. 11, 146). The tightness of the barrier depends upon the close apposition of astrocytes to blood capillaries (Figs 4.10C, 4.11). The blood-brain barrier develops during embryonic life but may not be fully completed by birth. Moreover, there are certain areas of the adult brain in which the endothelial cells do not have tight junctions and a free exchange of molecules occurs between blood and adjacent brain. Most of these areas are situated close to the ventricles and are known as circumventricular organs (p. 53). Otherwise unrestricted diffusion through the blood-brain barrier is only possible for substances which can cross biological membranes because of their lipophilic character. Lipophilic molecules may be actively re-exported by the brain endothelium. Breakdown of the blood-brain barrier occurs following brain damage caused by ischaemia or infection, and this permits an influx of fluid, ions, protein and other substances into the brain. It is also associated with primary and metastatic cerebral tumours. CT and MRI scans can demonstrate such breakdown of the blood-brain barrier clinically. A similar breakdown of the blood-brain barrier may be seen at postmortem in patients who are jaundiced. Normally brain, spinal cord and peripheral nerves remain unstained by the bile, except for the choroid plexus, which is often stained a deep yellow. However, areas of recent infarction (1-3 days), will be stained by bile pigment as a result of localized breakdown of the blood-brain barrier.
OLIGODENDROCYTES
Figure 4.10 Astrocytes. A, Immunofluorescent technique, human cerebral cortex, showing astrocytes immunopositive for glial fibrillary acidic protein (GFAP). B, Classic heavy metal impregnation technique (Cajal method). C, Immunoperoxidase technique, GFAP. Note perivascular end-feet embracing the capillary, C. (A, preparation by Jonathan Carlisle, Division of Anatomy and Cell Biology, GKT School of Medicine, London.) (B and C, by permission from Young B, Heath JW 2000 Wheater's Functional Histology. Edinburgh: Churchill Livingstone.)
Oligodendrocytes myelinate CNS axons and are most commonly seen as intrafascicular cells in myelinated tracts (Figs 4.12, 4.13). They usually have round nuclei and their cytoplasm contains numerous mitochondria, microtubules and glycogen. They display a spectrum of morphological variation, from large euchromatic nuclei and pale cytoplasm, to heterochromatic nuclei and dense cytoplasm. Oligodendrocytes may enclose up to 50 axons in separate myelin sheaths: the largest calibre axons are usually ensheathed on a 1:1 basis. Some oligo dendrocytes are not associated with axons, and are either precursor cells or perineuronal (satellite) oligodendrocytes whose processes ramify around neuronal somata. Within tracts, interfascicular oligodendrocytes are arranged in long rows in which single astrocytes intervene at regular intervals. Groups of oligodendrocytes myelinate the surrounding axons: their processes are radially aligned to the axis of each row. Myelinated tracts therefore consist of cables of axons, which are predominantly myelinated by a row of oligodendrocytes running down the axis of each cable.
Oligodendrocytes originate from the ventricular neurectoderm and the subependymal layer in the fetus (p. 262), and continue to be generated from the subependymal plate postnatally. Stem cells migrate and seed into white and grey matter to form a pool of adult progenitor cells which may later differentiate to replenish lost oligodendrocytes, and possibly remyelinate pathologically demyelinated regions. Nodes of Ranvier and incisures of Schmidt-Lanterman page 51 page 52
Figure 4.11 The relationship between the glia limitans, perivascular cells and blood vessels within the brain, in longitudinal and transverse section. A sheath of astrocytic end-feet wraps around the vessel and, in vessels larger than capillaries, its investment of pial meninges. Vascular endothelial cells are joined by tight junctions and supported by pericytes; perivascular macrophages lie outside the endothelial basal lamina.
The territory ensheathed by an oligodendrocyte process defines an internode, the interval between internodes is called a node of Ranvier and the territory immediately adjacent to the nodal gap is a paranode, where loops of oligodendrocyte cytoplasm abut the axolemma. Nodal axolemma is contacted by the end-feet of perinodal cells which have been shown in animal studies to have a presumptive adult oligodendrocyte progenitor phenotype: their function is unknown (Butt & Berry 2000). Schmidt-Lanterman incisures are helical decompactions of internodal myelin where the major dense line of the myelin sheath splits to enclose a spiral of oligodendrocyte cytoplasm. Their function is unknown: their structure suggests that they may play a role in the transport of molecules across the myelin sheath.
MYELIN AND MYELINATION Myelin is secreted by oligodendrocytes (CNS) and Schwann cells (PNS). A single oligodendrocyte may ensheathe up to 50 separate axons, depending upon calibre, whereas myelinating Schwann cells ensheathe axons on a 1:1 basis. In general, myelin is laid down around axons above 2 µm diameter. However, the critical minimal axon diameter for myelination is smaller and more variable in the CNS than in the PNS and is c.0.2 µm (compared with 1-2 µm in the PNS). Since there is considerable overlap between the size of the smallest myelinated and the
largest unmyelinated axons, axonal calibre is unlikely to be the only factor in determining myelination. Additionally, the first axons to become ensheathed ultimately reach larger diameters than later ones. There is a reasonable linear relationship between axon diameter and internodal length and myelin sheath thickness. As the sheath thickens from a few lamellae to up to 200, the axon may also grow from 1 to 15 µm in diameter. Internodal lengths increase about tenfold during the same time.
Figure 4.12 The ensheathment of a number of axons by the processes of an oligodendrocyte. The oligodendrocyte soma is shown in the centre and its myelin sheaths are unfolded to varying degrees to show their extensive surface area. (Modified from Morell and Norton 1980 by Raine 1984, by permission.)
It is not known how myelin is formed in either PNS or CNS. The ultrastructural appearance of myelin (Fig. 4.14) is usually explained in terms of the spiral wrapping of a flat glial process around an axon, and the subsequent extrusion of cytoplasm from the sheath at all points other than incisures and paranodes (p. 56). In this way, the compacted external surfaces of the plasma membrane of the ensheathing glial cell are thought to produce the minor dense lines, and the compacted inner cytoplasmic surfaces, the major dense lines, of the mature myelin sheath (Fig. 4.15). These correspond to the intraperiod and period lines respectively defined in X-ray studies of myelin. The inner and outer zones of occlusion of the spiral process are continuous with the minor dense line and are called the inner and outer mesaxons. There are significant differences between central and peripheral myelin, reflecting the fact that oligodendrocytes and Schwann cells express different proteins during myelinogenesis. The basic dimensions of the myelin membrane are different. CNS myelin has a period repeat thickness of 15.7 nm whereas PNS myelin has a period to period line thickness of 18.5 nm. The major dense line space is c.1.7 nm in CNS myelin, compared with 2.5 nm in PNS myelin. Myelin membrane contains protein, lipid and water, which forms at least 20% of the wet weight. It is a relatively lipid-rich membrane and contains 70-80% lipid. All classes of lipid have been found: the precise lipid composition of PNS and CNS myelin is different. The major lipid species are cholesterol (the commonest single molecule), phospholipids and glycosphingolipids. Minor lipid species include
galactosylglycerides, phosphoinositides and gangliosides. The major glycolipids are galactocerebroside and its sulphate ester, sulphatide: these lipids are not unique to myelin, but they are present in characteristically high concentrations. CNS and PNS myelin also contain low concentrations of acidic glycolipids, which constitute important antigens in some inflammatory demyelinating states. Gangliosides, which are glycosphingolipids characterized by the presence of sialic acid (N-acetylneuraminic acid), account for less than 1% of the lipid. page 52 page 53
Figure 4.13 A, An oligodendrocyte enwrapping several axons with myelin, demonstrated in a whole-mounted rat anterior medullary velum, immunolabelled with antibody to an oligodendrocyte membrane antigen. B, C, Confocal micrographs of a mature myelin forming oligodendrocyte (B) and astrocyte (C) iontophoretically filled in the adult rat optic nerve with an immunofluorescent dye by intracellular microinjection. (A, provided by Fiona Ruge; B and C, prepared by Dr A Butt and Kate Colquhoun, Division of Physiology, GKT School of Medicine, London. Photograph by Sarah-Jane Smith using the pseudocolour technique, Division of Anatomy and Cell Biology, GKT School of Medicine, London.)
A relatively small number of protein species account for the majority of myelin protein. Some of these proteins are common to both PNS and CNS myelin, but others are different. Proteolipid protein (PLP) and its splice variant DM20 are found only in CNS myelin, whereas myelin basic protein (MBP) and myelin associated glycoprotein (MAG) occur in both. MAG is a member of the immunoglobulin supergene family, and is localized specifically at those regions of
the myelin segment where compaction starts, namely, the mesaxons and inner periaxonal membranes, paranodal loops and incisures, in both CNS and PNS sheaths. It is thought to have a functional role in membrane adhesion. In the developing CNS, axonal outgrowth precedes the migration of oligodendrocyte precursors, and oligodendrocytes associate with and myelinate axons after their phase of elongation: oligodendrocyte myelin gene expression is not dependent on axon-association. In marked contrast, Schwann cells in the developing PNS are associated with axons during the entire phase of outgrowth from CNS to target organ. Myelination does not occur simultaneously in all parts of the body in late fetal and early postnatal development. White matter tracts and nerves in the periphery have their own specific temporal patterns, which relate to their degree of functional maturity. Mutations of the major myelin structural proteins have now been recognized in a number of inherited human neurological diseases. As would be expected, these mutations produce defects in myelination, and in the stability of nodal and paranodal architecture, which are consistent with the suggested functional roles of the relevant proteins in maintaining the integrity of the myelin sheath. The molecular organization of myelinated axons is described in Scherer & Arroyo (2002).
Figure 4.14 Transverse section of sciatic nerve showing a myelinated axon and several non-myelinated axons (A), ensheathed by Schwann cells (S). (Provided by Professor Susan Standring, GKT School of Medicine, London.)
EPENDYMA Ependymal cells line the ventricles and central canal of the spinal cord (Fig. 4.16). They form a single-layered epithelium which varies from squamous to columnar in form. At the ventricular surface, cells are joined by gap junctions and occasional desmosomes. Their apical surfaces have numerous microvilli and cilia, which contribute to the flow of cerebrospinal fluid (p. 292). There is considerable regional variation in the ependymal lining of the ventricles, but four major types have been described. These are general ependyma which overlies grey matter; general ependyma which overlies white matter; specialized areas of ependyma in the third and fourth ventricles; choroidal epithelium. The ependymal cells overlying areas of grey matter are cuboidal; each cell bears c.20 central apical cilia, surrounded by short microvilli. The cells are joined by gap junctions and desmosomes and do not have a basal lamina. Beneath them there may be a subependymal zone, from two to three cells deep, which consists of cells which generally resemble ependymal cells. The capillaries beneath them
have no fenestrations and few transcytotic vesicles, which is typical of the CNS. Where the ependyma overlies myelinated tracts of white matter, the cells are much flatter and few are ciliated. There are gap junctions and desmosomes between cells, but their lateral margins interdigitate, unlike those overlying grey matter. No subependymal zone is present. Specialized areas of ependymal cells are found in four areas around the margins of the third ventricle which are called the circumventricular organs. These are the lining of the median eminence of the hypothalamus; the subcommissural organ; the subfornical organ and the vascular organ of the lamina terminalis (p. 375). The area postrema, at the inferoposterior limit of the fourth ventricle, has a similar structure. In all of these sites the ependymal cells are only rarely ciliated and their ventricular surfaces bear many microvilli and apical blebs. They have numerous mitochondria, well-formed Golgi complexes and a rather flattened basal nucleus. They are joined laterally by tight junctions which form a barrier to the passage of materials across the ependyma, and desmosomes. Many of the cells are tanycytes (ependymal astrocytes) and have basal processes which project into the perivascular space surrounding the underlying capillaries. Significantly these capillaries are fenestrated and therefore do not form a blood-brain barrier. It is believed that neuropeptides can pass from nervous tissue into the cerebrospinal fluid (CSF) by active transport through the ependymal cells in these specialized areas, and in this way access a wide population of neurones via the permeable ependymal lining of the rest of the ventricle. The ependyma is highly modified where it lies adjacent to the vascular layer of the choroid plexuses (p. 292). page 53 page 54
Figure 4.15 Stages in myelination of a peripheral axon.
Choroid plexus
Figure 4.16 Ciliated cuboidal ependymal cells lining the central canal of the spinal cord. Similar cells line most of the ventricular system of the brain. (By permission from Kierszenbaum AL 2002 Histology and Cell Biology. St Louis: Mosby, and by kind permission from Dr Wan-hua Amy Yu.)
Figure 4.17 Choroid plexus within a ventricle. Frond-like projections of vascular stroma derived from the pial meninges are covered with a low columnar epithelium which secretes cerebrospinal fluid. (By permission from Stevens A, Lowe JS 1996 Human Histology, 2nd edn. London: Mosby.)
The ependymal cells in the choroid plexuses resemble those of the circumventricular organs, except that they do not have basal processes, but form a cuboidal epithelium which rests on a basal lamina adjacent to the enclosed fold of pia mater (p. 284) and its capillaries (Figs 4.17, 4.18). Capillaries of the choroid plexuses are lined by a fenestrated endothelium. Cells have numerous long microvilli with only a few cilia interspersed between them. They also have many mitochondria, large Golgi complexes and basal nuclei, which is consistent with their secretory activity: they produce most components of the CSF. They are linked by tight junctions which form a transepithelial barrier (a component of the blood-CSF barrier), and by desmosomes. Their lateral margins are highly folded. The choroid plexus has a villous structure where the stroma is composed of pial meningeal cells, and contains fine bundles of collagen and blood vessels. During fetal life, erythropoiesis occurs in the stroma, which is then occupied by bone marrow-like cells. In adult life the stroma contains phagocytic cells, and these,
together with the cells of the choroid plexus epithelium, phagocytose particles and proteins from the ventricular lumen. page 54 page 55
Figure 4.18 Schematic representation of the arrangement of tissues forming the choroid plexus. (By permission from Nolte J 2002 The Human Brain, 5th edn. London: Mosby.)
Age-related changes occur in the choroid plexus which can be detected on imaging the brain. Calcification of the choroid plexus can be detected by X-ray or CT scan in 0.5% of individuals in the first decade of life and in 86% in the eighth decade. There is a sharp rise in the incidence of calcification, from 35% of CT scans in the fifth decade, to 75% in the sixth decade. The visible calcification is usually restricted to the glomus region of the choroid plexus, i.e. the vascular bulge in the choroid plexus as it curves to follow the anterior wall of the lateral ventricle into the temporal horn (p. 287).
MICROGLIA Microglia are small dendritic cells found throughout the CNS (Fig. 4.19) including the retina (p. 712). Evidence largely supports the view that they are derived from fetal monocytes, or their precursors, which invade the developing nervous system. An alternative hypothesis holds that microglia share a lineage with ependymal cells and are thus neural tube derivatives. According to the monocyte theory, haematogenous cells pass through the walls of neural blood vessels and invade CNS tissue prenatally as amoeboid cells. Later they lose their motility and transform into typical microglia, bearing branched processes which ramify in nonoverlapping territories within the brain. All microglial domains, defined by their dendritic fields, are equivalent in size, and form a regular mosaic throughout the brain. The expression of microglia-specific antigens changes with age: many are downregulated as microglia attain the mature dendritic form.
Figure 4.19 Micrograph showing activated microglial cells in the central nervous system, in a biopsy from a patient with Rasmussen's encephalitis, visualized using MHC class II antigen immunohistochemistry. (By kind permission from Dr Norman Gregson, Division of Neurology, GKT School of Medicine, London.)
Microglia have elongated nuclei with peripheral heterochromatin. The scant cytoplasm is pale staining, and contains granules, scattered cisternae of rough endoplasmic reticulum and Golgi complexes at both poles. Two or three primary processes stem from opposite poles of the cell body and branch repeatedly to form short terminal processes. The function of microglia in the normal brain is obscure. Like astrocytes, microglia are activated by traumatic and ischaemic injury. In many diseases including Parkinson's disease, Alzheimer's disease, multiple sclerosis, acquired immunodeficiency syndrome (AIDS), amyotrophic lateral sclerosis (motor neurone disease) and paraneoplastic encephalitis, they become phagocytic and are actively involved in synaptic stripping and clearance of neuronal debris. Some transform into amoeboid, motile cells.
ENTRY OF INFLAMMATORY CELLS INTO THE BRAIN Although the CNS has long been considered to be an immunologically privileged site, lymphocyte surveillance of the brain may be a normal, low-grade activity, which is enhanced in disease. Lymphocytes are able to enter the brain in response to virus infections and as part of the autoimmune response in multiple sclerosis. Activated, but not resting, lymphocytes pass through the endothelium of small venules, a process that requires the expression of recognition and adhesion molecules, which are induced following cytokine activation. They subsequently migrate into the brain parenchyma. Within the CNS, microglia and astrocytes can be induced by T-cell cytokines to act as efficient antigen-presenting cells. Lymphocytes probably drain along lymphatic pathways to regional cervical lymph nodes. Polymorphonuclear leukocyte entry into the CNS is less common than lymphocyte entry, but is seen in the early stages of infarction and autoimmune disease and, in particular, in pyogenic infections. These cells probably enter the nervous system following expression of adhesion molecules on endothelium and pass through the endothelial layer. In the later stages of inflammation, monocytes may follow similar pathways. Within the subarachnoid space, polymorphonuclear leukocytes and lymphocytes pass through the endothelium of large veins into the CSF during the inflammatory phase of meningitis.
© 2008 Elsevier
PERIPHERAL NERVES Afferent nerve fibres connect peripheral receptors to the CNS: their neuronal somata are located either in special sense organs (e.g. the olfactory epithelium) or in the sensory ganglia of craniospinal nerves. Efferent nerve fibres connect the CNS to the effector cells and tissues: they are the peripheral axons of neurones with somata in the central grey matter. page 55 page 56
Figure 4.20 Transverse section through a peripheral nerve, showing the arrangement of its connective tissue sheaths. Individual axons, myelinated and unmyelinated, are arranged in a small fascicle bounded by a perineurium. Abbreviations: P, perineurium; Ep, epineurium; E, endoneurium. (Provided by Professor Susan Standring, GKT School of Medicine, London.)
Peripheral nerve fibres are grouped in widely variable numbers into bundles (fasciculi). The size, number and pattern of fasciculi (Fig. 4.20) vary in different nerves and at different levels along their paths. Their number increases and their size decreases some distance proximal to a point of branching. Where nerves are subjected to pressure, e.g. deep to a retinaculum, fasciculi are increased in number but reduced in size, and the amount of associated connective tissue and degree of vascularity also increase. At these points, nerves may occasionally show a pink, fusiform dilatation, sometimes termed a pseudoganglion or gangliform enlargement.
PERIPHERAL NERVE FIBRES The classification of peripheral nerve fibres is based on various parameters such as conduction velocity, function and fibre diameter. Of two classifications in common use, the first divides fibres into three major classes, designated A, B and C, corresponding to peaks in the distribution of their conduction velocities. In man, group A fibres are subdivided into !, " and # subgroups: B group fibres are preganglionic autonomic efferents, and C fibres are unmyelinated. Fibre diameter and conduction velocity are proportional in most fibres. Group A! fibres are the largest and conduct most rapidly, and C fibres are the smallest and slowest. The largest afferent axons (A! fibres) innervate encapsulated cutaneous, joint and muscle receptors, and some large alimentary enteroceptors. A" fibres innervate thermoreceptors and nociceptors, including those in dental pulp, skin and connective tissue. C fibres have thermoreceptive, nociceptive and interoceptive functions. The largest somatic efferent fibres (A!) are up to 20 µm in diameter. They innervate extrafusal muscle fibres exclusively and conduct at a maximum of 120 m/s. Fibres to fast twitch muscles are larger than those to slow twitch muscle. A$ fibres are restricted to collaterals of A! fibres, and form plaque endings on some intrafusal muscle fibres. A# fibres are exclusively fusimotor to plate and trail endings on intrafusal muscle fibres. C fibres are postganglionic sympathetic and parasympathetic axons. This scheme can be applied to all fibres of spinal and cranial nerves except perhaps those of the olfactory nerve, whose fibres form a uniquely small and slow group.
A different classification, used for afferent fibres of somatic muscles, divides myelinated fibres into groups I, II and III. Group I fibres are large (12-22 µm), and include primary sensory fibres of muscle spindles (Group Ia) and smaller fibres of Golgi tendon organs (Group Ib). Group II fibres are the secondary sensory terminals of muscle spindles, with diameters of 6-12 µm. Group III fibres, 1-6 µm in diameter, have free sensory endings in the connective tissue sheaths around and within muscles, and are believed to be nociceptive, relaying pressure pain in externally stimulated muscles. Paciniform (encapsulated) endings of muscle sheaths may also contribute fibres to this class. Group IV fibres are unmyelinated, with diameters below 1.5 µm: they include free endings in muscles, and are primarily nociceptive.
CONNECTIVE TISSUE SHEATHS Nerve trunks, whether uni- or multifascicular, are surrounded by an epineurium. Individual fasciculi are enclosed by a multilayered perineurium, which in turn surrounds the endoneurium or intrafascicular connective tissue (Fig. 4.20). Epineurium
Epineurium is a condensation of loose (areolar) connective tissue, and is derived from mesoderm. As a general rule, the more fasciculi present in a peripheral nerve, the thicker the epineurium. Epineurium contains fibroblasts, collagen (types I and III), and variable amounts of fat, and it cushions the nerve it surrounds. Loss of this protective layer may be associated with pressure palsies seen in wasted, bedridden patients. The epineurium also contains lymphatics (which probably pass to regional lymph nodes) and blood vessels, the vasa nervorum, which pass across the perineurium to communicate with a network of fine vessels within the endoneurium. Perineurium
Perineurium extends from the CNS-PNS transition zone to the periphery, where it is continuous with the capsules of muscle spindles and encapsulated sensory endings. At unencapsulated endings and neuromuscular junctions the perineurium ends openly. It consists of alternating layers of flattened polygonal cells which are thought to be derived from fibroblasts, and collagen. It can often contain 15-20 layers of such cells, each layer enclosed by a basal lamina up to 0.5 µm thick. Cells within each layer interdigitate along extensive tight junctions and their cytoplasm contains numerous pinocytotic vesicles and often, bundles of microfilaments. These features indicate that the perineurium functions as a metabolically active diffusion barrier, and together with the blood-nerve barrier (p. 58), probably plays an essential role in maintaining the osmotic milieu and fluid pressure within the endoneurium. Endoneurium
Strictly speaking, the term endoneurium is restricted to interfascicular connective tissue excluding the perineurial partitions within fascicles. Endoneurium consists of a fibrous matrix composed predominantly of type I collagen fibres, which are mainly organized in fine bundles lying parallel to the long axis of the nerve, and condensed around individual Schwann cell-axon units and endoneurial vessels. The fibrous and cellular components of the endoneurium are bathed in endoneurial fluid at a slightly higher pressure than that outside in the surrounding epineurium. The major cellular constituents of the endoneurium are Schwann cells, associated with axons, and endothelial cells. Schwann cell-axon units and endothelial cells are enclosed within individual basal laminae. Other cells which are always present within the endoneurium are fibroblasts (constituting c.4% of the total endoneurial cell population), resident macrophages and mast cells. Endoneurial arterioles have a poorly developed smooth muscle layer, and do not autoregulate well. In sharp contrast, epineurial and perineurial vessels have a dense perivascular plexus of peptidergic, serotoninergic and adrenergic nerves.
SCHWANN CELLS Schwann cells are the major glial type in the PNS. In vitro they are fusiform in appearance. Both in vitro and in vivo they ensheathe peripheral axons, and myelinate those greater than 2 µm diameter. In a mature peripheral nerve fibre, they are distributed along the axons in longitudinal chains. The precise geometry of their association depends on whether the axon is myelinated or unmyelinated.
In myelinated axons the territory of a Schwann cell defines an internode. The molecular phenotype of mature myelin-forming Schwann cells is different from that of the mature non-myelinating Schwann cell. Adult myelin-forming Schwann cells are characterized by the presence of several myelin proteins, some, but not all, of which are shared with oligodendrocytes and central myelin. In contrast, expression of the low affinity neurotrophin receptor (p75 NTR) and GFAP intermediate filament protein (which differs from the CNS form in its posttranslational modification), characterize adult non-myelin forming Schwann cells. page 56 page 57
Schwann cells arise during development from multipotent cells of the very early migrating neural crest (p. 244) which also give rise to peripheral neurones. Axonassociated signals are critical in controlling the proliferation of developing Schwann cells and their precursors. Neurones may also regulate the developmentally programmed death of Schwann cell precursors, as a mechanism for matching numbers of axons and glia within each peripheral nerve bundle. Neuronal signals appear to control the production of basal laminae by Schwann cells; the induction and maintenance of myelination; and, in the mature nerve, Schwann cell survival (few Schwann cells persist in chronically denervated nerves). Schwann cell signals may influence axonal calibre, and they are of crucial importance in the repair of damaged peripheral nerves. The acute Schwann cell response to axonal injury and degeneration involves mitotic division and the elaboration of signals which promote the regrowth of axons. Unmyelinated axons
Unmyelinated axons are commonly 500µm), suggests that a precise topographical organization is unlikely within the pallidum. Striatopallidal fibres are of two main types. They project either to the lateral or the medial pallidal segment. Those projecting to the lateral segment constitute the beginning of the so-called 'indirect pathway'. They utilize GABA as their primary transmitter and also contain enkephalin. Efferent axons from neurones in the lateral segment pass through the internal capsule in the subthalamic fasciculus, and travel to the subthalamic nucleus (Fig. 21.17). Striatopallidal axons destined for the medial pallidum constitute the so-called 'direct pathway'. Like the indirect projection, these also utilize GABA as their primary transmitter but they also contain substance P and dynorphin, rather than enkephalin. Efferent axons from the medial pallidal segment project through the ansa lenticularis and fasciculus lenticularis (Figs 21.17, 23.12). The former runs round the anterior border of the internal capsule and the latter penetrates the capsule directly. Having traversed the internal capsule, both pathways unite in the subthalamic region, where they follow a horizontal hairpin trajectory, and turn upwards to enter the thalamus as the thalamic fasciculus. The trajectory circumnavigates the zona incerta and creates the so-called 'H' fields of Forel (Figs 21.17, 23.5, 23.12). Within the thalamus, pallidothalamic fibres end in the ventral anterior and ventral lateral nuclei and in the intralaminar centromedian nucleus. These in turn project excitatory (presumed glutamatergic) fibres primarily to the frontal cortex, including the primary and supplementary motor areas. The medial pallidum also projects fibres caudally to the pedunculopontine nucleus (Fig. 23.14). This lies at the junction of the midbrain and the pons, close to the superior cerebellar peduncle, and corresponds approximately to the physiologically identified 'mesencephalic locomotor region'.
© 2008 Elsevier
SUBSTANTIA NIGRA page 427 page 428
The substantia nigra is a nuclear complex deep to the crus cerebri in each cerebral peduncle of the midbrain. It consists of a pars compacta and a pars reticulata (Figs 23.10, 23.11). The pars compacta, together with the smaller pars lateralis, corresponds to the dopaminergic cell group A9. With the retrorubral nucleus (A8), it makes up most of the dopaminergic neurone population of the midbrain and is the source of the mesostriatal dopamine system that projects to the striatum. The pars compacta of each side is continuous with its opposite counterpart through the ventral tegmental dopamine group A10, which is sometimes known as the paranigral nucleus. This is the source of the mesolimbic dopamine system, which supplies the ventral striatum and neighbouring parts of the dorsal striatum, as well as the prefrontal and anterior cingulate cortices. The dopaminergic neurones of the pars compacta (A9) and paranigral nucleus (ventral tegmental group A10) also contain cholecystokinin (CCK) or somatostatin. The pars reticulata contains large multipolar cells, which are very similar to those of the pallidum. Together they constitute the output neurones of the basal ganglia system. Their disc-like dendritic trees, like those of the pallidum, are orientated at right angles to afferents from the striatum, probably making en-passant contacts. Like the striatopallidal axons, of which they may be collaterals, striatonigral axons utilize GABA and substance P (SP) or enkephalin. They distribute differentially in the pars reticulata, such that the enkephalinergic axons terminate in the medial part, whereas substance P axons terminate throughout. The efferent neurones of the pars reticulata are GABAergic. They project to the deep (polysensory) layers of the superior colliculus and to the brain stem reticular formation, including the pedunculopontine nucleus. The pathway from the striatum to the superior colliculus, via the substantia nigra pars reticulata, is thought to function in the control of gaze in a manner analogous to the pathway that initiates general body movement via the pallidum, thalamus and supplementary motor cortex. The uncontrolled or fixed-gaze disturbances of advanced Parkinson's disease, progressive supranuclear palsy (PSP) and Huntington's disease tend to support this.
© 2008 Elsevier
SUBTHALAMIC NUCLEUS The subthalamic nucleus is a biconvex, lens-shaped nucleus in the subthalamus of the diencephalon. It lies medial to the internal capsule, immediately rostral to the level at which the latter becomes continuous with the crus cerebri of the midbrain (Figs 23.2, 23.5, 23.8). Within its substance, small interneurones intermingle with large multipolar cells with dendrites, which extend for about onetenth the diameter of the nucleus. It is encapsulated dorsally by axons, many of which are derived from the subthalamic fasciculus, and which carry a major GABAergic projection from the lateral segment of the globus pallidus as part of the indirect pathway. It also receives afferents from the cerebral cortex. The subthalamic nucleus is unique in the intrinsic circuitry of the basal ganglia in that its cells are glutamatergic. They project excitatory axons to both the globus pallidus and substantia nigra pars reticulata. Within the pallidum, subthalamic efferent fibres end predominantly in the medial segment but many also end in the lateral segment. The subthalamic nucleus plays a central role in the normal function of the basal ganglia and in the pathophysiology of basal ganglia-related disorders. Destruction of the nucleus, which occurs rarely as a result of stroke, results in the appearance of violent, uncontrolled involuntary movements, known as ballism (ballismus). The subthalamic nucleus is also crucially involved in the pathophysiology of Parkinson's disease and is a target for functional neurosurgical therapy of the condition.
© 2008 Elsevier
PATHOPHYSIOLOGY OF BASAL GANGLIA DISORDERS The normal functions of the basal ganglia are difficult to summarize succinctly. As far as their role in movement control is concerned, however, a reasonable definition is that they function to promote and support patterns of behaviour and movement that are appropriate in the prevailing circumstances and to inhibit unwanted or inappropriate behaviour and movements. This is exemplified by disorders of the basal ganglia, which are characterized, depending on the underlying pathology, by an inability to initiate and execute wanted movements (as in Parkinson's disease) or an inability to prevent unwanted movements (as in Huntington's disease). Parkinson's disease is the most common pathological condition affecting the basal ganglia. This is characterized by akinesia, muscular rigidity and tremor. It is due to degeneration of the dopaminergic neurones of the substantia nigra pars compacta. As a consequence, dopamine levels in the striatum are depleted. This has been amply confirmed by post-mortem studies. Furthermore, in parkinsonian patients, positron emission tomography (PET) reveals a deficit of dopamine storage and reuptake, due to loss of nigrostriatal terminals, but intact dopamine receptors, which are located upon the medium spiny neurones, the target of the nigrostriatal pathway. Dopamine appears to have a dual action on medium spiny striatal neurones. It inhibits those of the indirect pathway and excites those of the direct pathway. Consequently, when dopamine is lost from the striatum, the indirect pathway becomes overactive and the direct pathway becomes underactive (Fig. 23.15). Overactivity of the striatal projection to the lateral pallidum results in inhibition of pallidosubthalamic neurones and, consequently, overactivity of the subthalamic nucleus. Subthalamic efferents mediate excessive excitatory drive to the medial globus pallidus and substantia nigra pars reticulata. This is exacerbated by underactivity of the GABAergic, inhibitory direct pathway. Overactivity of basal ganglia output then inhibits the motor thalamus and its excitatory thalamocortical connections. While this description is little more than a first approximation of the underlying pathophysiology, this model of the basis of parkinsonian symptoms has led to the introduction of new neurosurgical approaches to the treatment of Parkinson's disease, based upon lesioning and deep brain stimulation of the medial globus pallidus and subthalamic nucleus (see below). UPDATE Date Added: 21 June 2005 Abstract: The cerebral oscillatory network of parkinsonian resting tremor. Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=12477707&query_hl=14 The cerebral oscillatory network of parkinsonian resting tremor.
page 428 page 429
Figure 23.15 Pathophysiology of Parkinson's disease. Dotted lines indicate dysfunctional pathways. (By permission from Crossman AR, Neary D 2000 Neuroanatomy, 2nd edn. Edinburgh: Churchill Livingstone.)
The current medical treatment for Parkinson's disease relies upon levodopa (L-DOPA; L-dihydroxyphenylalanine), the immediate metabolic precursor of dopamine, or dopamine agonists. Whilst these usually provide good symptomatic relief for many years, they eventually lead to the development of side-effects, including dyskinesias. The involuntary movements that occur as a consequence of long-term treatment of Parkinson's disease resemble those seen in Huntington's disease, tardive dyskinesia and ballism. Experimental evidence suggests that these may share a common neural mechanism (Fig. 23.16). Thus, the indirect pathway becomes underactive, e.g. due to the effects of dopaminergic drugs in Parkinson's disease or the degeneration of the striatopallidal projection to the lateral pallidum in Huntington's disease. This leads to physiological inhibition of the subthalamic nucleus by overactive pallidosubthalamic neurones. The involvement of the subthalamic nucleus explains why the dyskinetic movements of levodopa-induced dyskinesia and Huntington's disease resemble those of
ballism produced by lesion of the subthalamic nucleus. Underactivity of the subthalamic nucleus removes the excitatory drive from medial pallidal neurones, which are known to be underactive in dyskinesias. Once again this is an oversimplification. Whilst it is true that underactivity of the medial globus pallidus is associated with dyskinesias, it is also known that lesions of the globus pallidus alleviate them. This suggests that the dynamic aspects of pallidal and nigral efferent activity are important factors in the generation of dyskinesia. Another manifestation of basal ganglia dysfunction is dystonia, which is characterized by increased muscle tone and abnormal postures. This may occur as a consequence of levodopa treatment in Parkinson's disease or inherited disease (e.g. idiopathic torsion, or Oppenheim's dystonia). The pathophysiological basis of dystonia is unclear. Like dyskinesias it is probably caused by underactivity of basal ganglia output, and so deep-brain stimulation of the globus pallidus may be beneficial.
Figure 23.16 Pathophysiology of dyskinesias. Dotted lines indicate dysfuntional pathways. (By permission from Crossman AR, Neary D 2000 Neuroanatomy, 2nd edn. Edinburgh: Churchill Livingstone.)
There is evidence that dysfunction of the basal ganglia is involved in other
There is evidence that dysfunction of the basal ganglia is involved in other complex, less well-understood, behavioural disorders. In animal experiments, lesions of the basal ganglia, especially of the caudate nucleus, induce uncontrollable hyperactivity (e.g. obstinate progression, incessant pacing and other constantly repeated behaviour). This and other evidence has led to the notion that the corpus striatum enables the individual to make motor choices and to avoid 'stimulus-bound' behaviour. PET scanning studies in man have shown that sufferers from obsessive-compulsive disorder (OCD), which is characterized by repeated ritualistic motor behaviour and intrusive thoughts, exhibit abnormal activity in the prefrontal cortex and caudate nuclei. There are similar suggestive observations in the hyperactive child syndrome. In this respect it may be significant that the basal ganglia, besides receiving connections from the frontal lobe and limbic cortices, also have an ascending influence on the prefrontal and cingulate cortices through the substantia nigra pars reticulata and dorsomedial and ventromedial thalamus (Fig. 23.13B, C, D). Before the advent of levodopa , neurosurgery for Parkinson's disease was commonplace. The globus pallidus and thalamus were favoured targets for chemical or thermal lesions. Pallidotomy and thalamotomy often improved rigidity and tremor, but they produced little consistent beneficial effect upon akinesia. With the arrival of levodopa therapy, which had a profound effect upon akinesia, the surgical treatment of Parkinson's disease underwent progressive decline. However, it soon became clear that long-term use of levodopa was associated with a number of side-effects such as dyskinesias, 'wearing-off', and the 'on-off' phenomenon. More recent advances in understanding the pathophysiology of movement disorders, and in particular Parkinson's disease, have stimulated a renaissance in the use of neurosurgery to treat movement disorders. In primates that had been made parkinsonian experimentally with the neurotoxin MPTP, lesioning the subthalamic nucleus alleviated tremor, rigidity and bradykinesia. This finding raised the possibility that the subthalamic nucleus could be used as a clinical target. Indeed, lesions of the subthalamic nucleus in humans exert a powerful effect in alleviating tremor, rigidity and bradykinesia. However, the likelihood of side-effects is not trivial (the subthalamic nucleus is a small structure wrapped by fibres of passage and close to the hypothalamus and internal capsule), and relatively few centres perform this procedure. In 1992, Laitinen et al reintroduced pallidotomy for the treatment of end-stage Parkinson's disease, but confined the lesions to the posteroventral part of the internal pallidal segment. These lesions were found to be extremely reliable in abolishing contralateral rigidity and drug-induced dyskinesias, with slightly less efficacy on tremor and bradykinesia. Implantation of deep-brain electrodes through which high-frequency pulses generated by a pacemaker could inhibit cells in the vicinity has been a concept since the early 1970s, but did not become a widespread reality until the late 1980s, as a result of technological advances. The introduction of this technique, which avoids making permanent lesions, made bilateral surgery safer. There have been numerous reports of the effectiveness of both bilateral pallidal and subthalamic nucleus stimulation (Figs 23.17, 23.18). Subthalamic nucleus stimulation is favoured by most groups because, unlike pallidal stimulation, it allows patients to reduce their anti-parkinsonian medication. Serendipity also has a role in such surgery. Clinically, parkinsonian patients can develop painful dystonic posturing of their limbs which responds dramatically to
bilateral pallidal stimulation. This has led to preliminary studies of bilateral pallidal stimulation for dystonia with very promising results. Since it is held that in dystonia the pallidal neurones fire at rates below normal, this presents quite a puzzle as it is open to question how stimulation works. It also would appear that the neural mechanism that underlies this therapeutic effect on dystonia may differ from that in Parkinson's disease and tremor, because in dystonia the improvement may take weeks to emerge, whereas it is immediate in the case of Parkinson's disease. page 429 page 430
Figure 23.17 MRI image showing placement of deep brain stimulating electrodes bilaterally in the globus pallidus of a patient with Parkinson's disease. (By kind permission from Professor TZ Aziz, Radcliffe Infirmary, Oxford and Charing Cross Hospital, London.)
Figure 23.18 MRI image showing placement of deep brain stimulating electrodes bilaterally in the subthalamic nucleus of a patient with Parkinson's disease. (By kind permission from Professor AM Lozano, Toronto Western Hospital.)
REFERENCES Alexander GE, DeLong MR, Strick PL 1986 Parallel organization of functionally segregated circuits linking basal ganglia and cortex. Ann Rev Neurosci 9: 357-82. A landmark publication setting out a conceptual framework for the way in which the basal ganglia and cerebral cortex process different types of information through largely distinct parallel circuits based upon known anatomical connectivity. Medline Similar articles Crossmann AR 1990 A hypothesis on the pathophysiological mechanisms that underlie levodopa-or dopamine agonist-induced dyskinesia in Parkinson's disease: implications for future strategies in treatment. Mov Disord 5: 100-108. Medline Similar articles Krack P, Batir A, Van Blercom N et al 2003 Five-year follow-up of bilateral stimulation of the subthalamic nucleus in advanced Parkinson's disease. N Engl J Med 349: 1925-1934. Reviews the long-term outcome of deep brain stimulation of the subthalamic nucleus in Parkinson's disease. Medline Similar articles Full article Laitinen LV, Bergenheim AT, Hariz MI 1992 Ventroposterolateral pallidotomy can abolish all Parkinsonian symptoms. Stereotact Funct Neurosurg 58: 14-21. A key paper which ignited widespread interest in functional neurosurgery for Parkinson's disease. Medline Similar articles Full article Penney JB Jr, Young AB 1986 Striatal inhomogeneities and basal ganglia function. Mov Disord 1: 3-15. A landmark publication, introducing some of the basic concepts behind current models of the pathophysiology of Parkinson's disease and Huntington's disease. Medline Similar articles Full article
© 2008 Elsevier
24 Special senses The special senses of olfaction, vision, taste, hearing and balance are conveyed to the brain in cranial nerves. In each case, highly specialized peripheral receptors respond to stimuli in the external environment or our relationship to it. The olfactory system has an ancient lineage, reflected by the fact that afferent olfactory pathways proceed directly to the cerebral cortex and bypass the thalamus. Its terminal fields are, likewise, primitive cortical areas in a phylogenetic sense and are considered to be parts of the limbic system. All other special senses have a thalamic representation which projects to specialized regions of the neocortex.
© 2008 Elsevier
OLFACTION Olfactory pathways subserving the sense of smell are described in this section. Details of the relationship between the olfactory pathways and the limbic system are shown in Fig. 22.7. The olfactory nerves arise from olfactory receptor neurones in the olfactory mucosa. The axons collect into c.20 bundles and enter the anterior cranial fossa by passing through the foramina in the cribriform plate. They attach to the inferior surface of the olfactory bulb, which is situated at the anterior end of the olfactory sulcus on the orbital surface of the frontal lobe, and terminate in the bulb. Apparently unique in the nervous system, olfactory receptor neurones are continually replaced throughout life by differentiation of stem cells in the olfactory mucosa. The olfactory bulb is continuous posteriorly with the olfactory tract, through which the output of the bulb passes directly to the olfactory cortex. There is a clear laminar structure in the olfactory bulb (Fig. 24.1). From the surface inwards the laminae are the olfactory nerve layer, glomerular layer, external plexiform layer, mitral cell layer, internal plexiform layer and granule cell layer. The olfactory nerve layer consists of unmyelinated axons of the olfactory neurones. The continuous turnover of receptor cells means that axons in this layer are at different stages of growth, maturity or degeneration. The glomerular layer consists of a thin sheet of glomeruli where the incoming olfactory axons divide and synapse on terminal dendrites of secondary olfactory neurones, i.e. mitral, tufted and periglomerular cells. The external plexiform layer contains the principal and secondary dendrites of mitral and tufted cells. The mitral cell layer is a thin sheet composed of the cell bodies of mitral cells, each of which sends a single principal dendrite to a glomerulus, secondary dendrites to the external plexiform layer, and a single axon to the olfactory tract.
page 431 page 432
Figure 24.1 Organization of the olfactory bulb. The radial organization of the bulb into 'layers', with their principal neurone types, and an indication of their main connections is shown. Red: mitral and tufted neurones and their processes; light blue: internal granule neurones; dark blue: dopaminergic periglomerular neurones; black: olfactory receptor neurones and their processes. The olfactory tract consists of (1) centripetal
axons of mitral and tufted cells, some of which synapse with neurones in the anterior olfactory nucleus and (2) centrifugal axons (yellow) which terminate in the different zones indicated.
It also contains a few granule cell bodies. The internal plexiform layer contains axons, recurrent and deep collaterals of mitral and tufted cells, and granule cell bodies. The granule cell layer contains the majority of the granule cells and their superficial and deep processes, together with numerous centripetal and centrifugal nerve fibres which pass through the layer. The principal neurones in the olfactory bulb are the mitral and tufted cells: their axons form its output via the olfactory tract. These cells are morphologically similar and most use an excitatory amino acid, probably glutamate or aspartate, as their neurotransmitter. The mitral cell spans the layers of the bulb, and receives the sensory input superficially at its glomerular tuft. The axons of mitral and tufted cells appear to be parallel output pathways from the olfactory bulb. It is not known whether they receive inputs from different olfactory sensory neurones. The main types of intrinsic neurone in the olfactory bulb are periglomerular cells and granule cells. The majority of periglomerular cells are dopaminergic (cell group A15); some are GABAergic. Their axons are distributed laterally and terminate within extraglomerular regions. Granule cells are similar in size to periglomerular cells. Their most characteristic feature is the absence of an axon, and they therefore resemble amacrine cells in the retina. Granule cells have two principal spine-bearing dendrites which pass radially in the bulb, to ramify and terminate in the external plexiform layer. They appear to be GABAergic. The granule cell is likely to be a powerful inhibitory influence on the output neurones of the olfactory bulb. Centrifugal inputs to the olfactory bulb arise from a variety of central sites. Neurones of the anterior olfactory nucleus and collaterals of pyramidal neurones in the olfactory cortex project to the granule cells of the olfactory bulb. Cholinergic neurones in the horizontal limb nucleus of the diagonal band of Broca, part of the basal forebrain cholinergic system, project to the granule cell layer and also to the glomerular layer. Other afferents to the granule cell layer and the glomeruli arise from the pontine locus coeruleus and the mesencephalic raphe nucleus. The olfactory tract leaves the posterior pole of the olfactory bulb to run along the olfactory sulcus on the orbital surface of the frontal lobe (Fig. 22.7). The granule cell layer of the bulb is extended into the olfactory tract as scattered mediumsized multipolar neurones which constitute the anterior olfactory nucleus. They continue into the olfactory striae and trigone to the grey matter of the prepiriform cortex, the anterior perforated substance and precommissural septal areas. Many centripetal axons from mitral and tufted cells relay in, or give collaterals to, the anterior olfactory nucleus; the axons from the nucleus continue with the remaining direct fibres from the bulb into the olfactory striae. As the olfactory tract approaches the anterior perforated substance it flattens and splays out as the olfactory trigone. Fibres of the tract continue from the caudal angles of the trigone as diverging medial and lateral olfactory striae, which border the anterior perforated substance. An intermediate stria sometimes passes from the centre of the trigone to end in a small olfactory tubercle. The lateral olfactory stria follows the anterolateral margin of the anterior perforated substance to the limen insulae, where it bends posteromedially to merge with an elevated region, the gyrus semilunaris, at the rostral margin of the uncus in the temporal lobe (Fig. 22.7). The lateral olfactory gyrus forms a tenuous grey layer covering the lateral olfactory stria: it merges laterally with the gyrus ambiens, part of the limen insulae. The lateral olfactory gyrus and gyrus ambiens form the prepiriform region of the cortex, passing caudally into the entorhinal area of the parahippocampal gyrus. The prepiriform and periamygdaloid regions and the entorhinal area (area 28) together make up the piriform cortex. The medial olfactory stria, covered thinly by the grey matter of the medial olfactory gyrus, passes medially along the rostral boundary of the anterior perforated substance towards the medial continuation of the diagonal band of Broca. Together, they curve up on the medial aspect of the hemisphere, anterior to the attachment of the lamina terminalis. The diagonal band enters the paraterminal gyrus. The medial stria becomes indistinct as it approaches the boundary zone, which includes the paraterminal gyrus,
parolfactory gyrus and, between them, the prehippocampal rudiment (Fig. 22.7). The olfactory cortex receives a direct input from the olfactory bulb, which arrives via the olfactory tract without relay in the thalamus. The largest cortical olfactory area is the piriform cortex. The anterior olfactory nucleus, olfactory tubercle, regions of the entorhinal and insular cortex and amygdala also receive direct projections from the olfactory bulb. The entorhinal cortex (Brodmann's area 28) is the most posterior part of the piriform cortex, and is divided into medial and lateral areas (areas 28a and 28b). The lateral parts receive fibres mainly from the olfactory bulb, and also from the piriform and periamygdaloid cortices. Projections from the piriform olfactory cortex are widespread, and include the neocortex (especially the orbitofrontal cortex), thalamus (especially the medial dorsal thalamic nucleus), hypothalamus, amygdala and hippocampal formation.
© 2008 Elsevier
VISION The anatomy of the eye is described in Chapters 41 and 42. The visual pathway is illustrated in Fig. 24.2. The first-order neurone of the visual system is a bipolar cell which is contained entirely within the retina. The second-order neurone is a ganglion cell whose axon enters the optic nerve. The optic nerves pass posteromedially into the cranial cavity and meet in the midline, forming the optic chiasma, a flat mass of decussating fibres which lies at the junction of the anterior wall and floor of the third ventricle. The tuber cinereum and infundibulum lie posterior to the chiasma, and the third ventricle is dorsal to them. The termination of the internal carotid artery and anterior perforated substance are lateral relations. The optic recess of the third ventricle passes over its superior surface to reach the lamina terminalis. Optic nerve fibres arising from the nasal half of each retina, including half of the macula, cross in the chiasma to enter the contralateral optic tract. Fibres from the temporal hemiretinae continue into the ipsilateral optic tract. Decussating fibres loop a little backwards into their ipsilateral optic nerve before crossing and then pass forwards into the contralateral optic tract after crossing. Macular fibres, and those from an adjacent central area, occupy almost two-thirds of the central chiasma, dorsal to all peripheral decussating fibres. The most ventral axons are nasal fibres concerned with monocular fringes of the binocular field. They lie beneath fibres from the extramacular parts of both nasal hemiretinae, which occupy an intermediate position in the chiasma. The optic chiasma is supplied with blood from a pial plexus which receives branches from the superior hypophyseal, internal carotid, posterior communicating, anterior cerebral and anterior communicating arteries. The venous drainage of the chiasma is into the basal and anterior cerebral venous system. Behind the optic chiasma, the optic tracts diverge dorsolaterally, each passing between the anterior perforated substance and tuber cinereum. The tract curves around the cerebral peduncle, to which it adheres. Optic tract fibres terminate primarily in the lateral geniculate nucleus of the thalamus, but also in the superior colliculus, pretectal area, suprachiasmatic nucleus of the hypothalamus and inferior pulvinar. Axons from third-order visual neurones in the lateral geniculate nucleus run in the retrolenticular part of the internal capsule and form the optic radiation, which curves dorsomedially to the occipital cortex. Fibres representing the lower half of the visual field sweep superiorly to reach the visual cortex above the calcarine sulcus. Those representing the upper half of the visual field curve inferiorly into the temporal lobe (Meyer's loop) before reaching the visual cortex below the calcarine sulcus. Some neurones in the occipital cortex send descending axons to the superior colliculus, which therefore receives cortical and retinal afferents. From here fibres travel by tectobulbar tracts to motor nuclei of the third, fourth, sixth and eleventh cranial nerves and the ventral horn of the spinal cord.
Visual field defects page 432 page 433
page 433 page 434
Figure 24.2 The visual pathway, showing the spatial arrangement of neurones and their fibres in relation to the quadrants of the retinae and visual fields. The proportions at various levels are not exactly to scale and in particular the macula is exaggerated in size in the visual fields and retinae. In each quadrant of the visual field, and the parts of the visual pathway subserving it, two shades of the respective colour are used, the paler for the peripheral fields and a darker shade for the macular part of the
quadrant. From the optic tract onwards these two shades are both made more saturated to denote intermixture of neurones from both retinae, the palest shade being reserved for parts of the visual pathway concerned with monocular vision. The pathway subserving the pupillary light reflex is also indicated.
Plotting visual field loss frequently reveals the approximate location of the causative lesion in the visual pathway and sometimes its nature. Since retinal lesions can be visualized using an ophthalmoscope, these aids might appear to be redundant, but visual field measurement is still helpful in assessing the extent of the damage and may be the key factor in confirming a diagnosis. Glaucoma serves as an example. Field defects in glaucoma, occurring as a consequence of damage to the nerve fibre bundles at the optic nerve head, may be detectable ophthalmoscopically, but confirmation of the diagnosis frequently depends on field assessment. An initial constriction of the visual field is of little clinical significance but later defects, characteristic of the disease, consist of a scotoma between 10 and 20 degrees of the fixation area, extending upwards, or less commonly downwards, from the blind spot. This later elongates circumferentially along the arcuate nerve fibres and subsequently extends further. The field defect forms a linear limit or step along the horizontal meridian nasally, the loss continuing to blindness. So far as the location of lesions central to the retina are concerned, deficits in the vision of one eye are usually attributable to optic nerve lesions. Lesions of the optic chiasma, involving crossing nerve fibres, produce a bilateral field loss as exemplified by a pituitary adenoma. The tumour expands upwards from the pituitary fossa, compressing the inferior midline of the chiasma, and eventually produces bitemporal hemianopia, starting with an early loss in the upper temporal quadrants. Field defects in the rare instances of optic tract lesions are distinctive. The tract contains contralateral nasal and ipsilateral temporal retinal projections and damage will cause a homonymous contralateral loss of field with substantial incongruity (dissimilar defects in the two fields). Incongruity probably results from a delay in achieving coincidence between retinal topographical projections of the two inputs of the visual pathway, as contiguous projections adjust their location, gradually achieving coincidence. It also likely reflects the reorganization of fibres which occurs normally in the optic tracts, as some fibres leave the tract in the superior brachium and others progress to the lateral geniculate nucleus. The two defective fields may display incongruency as a result of lesions above the level of the chiasma. It is most marked in optic tract defects, less obvious in optic radiation defects, and is usually absent in cortically induced field defects, thus providing an additional clue in assessing location of the cause. Lesions of the optic radiations are usually unilateral, and commonly vascular in origin. Field defects therefore develop abruptly, in contrast to the slow progression of defects associated with tumours. Resulting hemifield loss follows the general rule that visual field defects central to the chiasma are on the opposite side to the lesion. Little or no incongruity is seen in visual cortical lesions but they commonly display the phenomenon of macula sparing, the central 5-10° field being retained in an otherwise hemianopic defect.
Neural control of gaze Neural control systems are required to coordinate the movements of the eyes so that the image of the object of interest is simultaneously held on both foveae, despite movement of the object or the observer. A number of separate neural systems are involved: first, to shift gaze to the object of interest using rapid movements, called saccades; and second, to stabilize the image on the fovea either during movement of the object of interest (the smooth pursuit system), and/or during movement of the head or body (the vestibulo-ocular and optokinetic systems). Although the detailed anatomical substrates for these systems differ, they share common circuitry which lies mainly in the pons and midbrain, for horizontal and vertical gaze movements respectively (Fig. 24.3). The common element for all types of horizontal gaze movements is the abducens nucleus. This contains motor neurones which innervate the ipsilateral lateral rectus. It also contains interneurones which project via the medial longitudinal fasciculus (MLF) to the contralateral oculomotor nucleus controlling medial rectus. A lesion of the abducens nucleus leads to a total loss of ipsilateral horizontal conjugate gaze. A lesion of the MLF produces slowed or absent adduction of the
ipsilateral eye, usually associated with jerky movements (nystagmus) of the abducting eye, a syndrome called internuclear ophthalmoplegia. The gaze motor command involves specialized areas of the reticular formation of the brain stem which receive a variety of supranuclear inputs. The main region for horizontal gaze is the paramedian pontine reticular formation (PPRF), which is located on each side of the midline in the central paramedian part of the tegmentum, and extends from the pontomedullary junction to the pontopeduncular junction. Each PPRF contains excitatory neurones which discharge at high frequencies just prior to and during ipsilateral saccades. Pause neurones, which are located in a midline caudal pontine nucleus called the nucleus raphe interpositus, discharge tonically except just before and during saccades. They appear to exert an inhibitory influence on the burst neurones and so prevent extraneous saccades occurring during fixation.
page 434 page 435
Figure 24.3 Summary of eye movement control. The central drawing shows the supranuclear connections from the frontal eye field (FEF) and the posterior eye field (PEF) to the superior colliculus (SC), rostral interstitial nucleus of the medial longitudinal fasciculus (riMLF), and the paramedian pontine reticular formation (PPRF). The FEF and SC are involved in the production of saccades, while the PEF is thought to be important in the production of pursuit. The drawing on the left shows the brain stem pathways for horizontal gaze. Axons from the PPRF travel to the ipsilateral abducens nucleus innervating lateral rectus (LR). Abducens internuclear axons cross the midline and travel in the medial longitudinal fasciculus (MLF) to the portion of the oculomotor nucleus (III) innervating medial rectus (MR) of the contralateral eye. The drawing on the right shows the brain stem pathways for vertical gaze. Important structures include the riMLF, PPRF, the interstitial nucleus of Cajal (INC), and the posterior commissure (PC). Other abbreviations: DLPFC, dorsolateral prefrontal cortex; IV, trochlear nucleus; SEF, supplementary eye field; VN, vestibular nucleus.
The vestibular nuclei and the perihypoglossal complex (especially the nucleus prepositus hypoglossi) project directly to the abducens nuclei. These projections probably carry both smooth pursuit signals, via the cerebellum, and vestibular signals. In addition, these nuclei, via reciprocal innervation with the PPRF, contain integrator neurones which control the step change in innervation required to maintain the eccentric position of the eye against the viscoelastic forces in the orbit. These forces tend to move the eyeball back to the position of looking straight ahead, i.e. the primary position, after a saccade. The final common pathway of vertical gaze movements is formed by the oculomotor and trochlear nuclei. The rostral interstitial nucleus of the medial longitudinal fasciculus (riMLF) contains neurones that discharge in relation to upand-down vertical saccadic movements. The riMLF projects through the posterior commissure to its equivalent on the other side of the mesencephalon, as well as directly to the oculomotor nucleus. Therefore, lesions within the posterior
commissure give rise to disturbance in vertical gaze, especially up-gaze. Lesions placed more ventrally in the region of the riMLF give rise to vertical gaze disorders which may be mixed up-and-down, or mainly down-gaze. Slightly caudal to the riMLF, and directly connected to it, lies the interstitial nucleus of Cajal. It contains neurones which appear to be involved in vertical gaze in holding the vertical pursuit. The cerebral hemispheres are extremely important for the programming and coordination of both saccadic and pursuit conjugate eye movements. There appear to be four main cortical areas in the cerebral hemispheres involved in the generation of saccades (Fig. 24.3). These are: the frontal eye field (FEF), which lies laterally at the caudal end of the second frontal gyrus in the premotor cortex (Brodmann area 8); the supplementary eye field (SEF), which lies at the anterior region of the supplementary motor area in the first frontal gyrus (Brodmann area 6); the dorsolateral prefrontal cortex (DLPFC), which lies anterior to the FEF in the second frontal gyrus (Brodmann area 46); and a posterior eye field (PEF), which lies in the parietal lobe, possibly in the superior part of the angular gyrus (Brodmann area 39), and the adjacent lateral intraparietal sulcus. These areas all appear to be interconnected with each other and to send projections to the superior colliculus and the brain stem areas controlling saccades. It appears that there are two parallel pathways involved in the cortical generation of saccades. An anterior system originates in the FEF and projects both directly, and via the superior colliculus, to the brain stem saccadic generators. This pathway also passes indirectly via the basal ganglia to the superior colliculus. Projections from the frontal cortex influence cells in the pars reticulata of the substantia nigra, via a relay in the caudate nucleus. An inhibitory pathway from the pars reticulata projects directly to the superior colliculus. This appears to be a gating circuit related to volitional saccades, especially of the memory-guided type. A posterior pathway originates in the PEF and passes to the brain stem saccadic generators via the superior colliculus. To maintain foveation of a moving target, the smooth pursuit system has developed relatively independently of the saccadic oculomotor system, although there are inevitable interconnections between the two. The first task is to identify and code the velocity and direction of a moving target. This is carried out in the extrastriate visual area known as the middle temporal visual area (MT; also called visual area V5), which contains neurones sensitive to visual target motion. In man, this lies immediately posterior to the ascending limb of the inferior temporal sulcus at the occipitotemporal border. Area MT send this motion signal to the medial superior temporal visual area (MST), which is thought to lie superior and a little anterior to area MT within the inferior parietal lobe. Damage to this area results in an impairment of smooth pursuit of targets moving towards the damaged hemisphere. Both area MST and FEF send direct projections to a group of nuclei which lie in the basal part of the pons. In the monkey, the dorsolateral and lateral groups of pontine nuclei receive direct cortical inputs related to smooth pursuit. Lesions of similarly located nuclei in man result in abnormal pursuit. These nuclei transfer the pursuit signal bilaterally to the posterior vermis, contralateral flocculus and fastigial nuclei of the cerebellum. The pursuit signal ultimately passes from the cerebellum to the brain stem, specifically to the medial vestibular nucleus and nucleus propositus hypoglossi, thence to the PPRF and possibly directly to the ocular motor nuclei. This circuitry therefore involves a double decussation, firstly at the level of the midpons (pontocerebellar neurones) and secondly in the lower pons (vestibuloabducens neurones). The vestibulo-ocular reflex maintains coordination of vision during movement of the head. It results in a compensatory conjugate eye movement which is equal but opposite to the movement of the head. This is essentially a three-neurone arc. It consists of primary vestibular neurones which project to the vestibular nuclei, secondary neurones which project from these nuclei directly to the abducens nucleus, and tertiary neurones which are abducens motor neurones. The optokinetic response is another visually mediated reflex which stabilizes retinal imagery during rotational movement. As the visual scene changes, the eyes follow, holding the retinal image steady until they shift rapidly in the opposite direction to another area of the visual scene. The full field of vision, rather than
small objects within it, is the stimulus, and the alternating slow and fast phases of movement generated, describes optokinetic nystagmus. The optokinetic reflex functions in collaboration with the rotational vestibulo-ocular reflex. Because of the mechanical arrangements of the semicircular canals, in the sustained rotations of the body described above the vestibulo-ocular reflex fades. In darkness the reflex, which is initially compensatory, loses velocity, and after c.45 seconds the eyes become stationary.
Pupillary light reflex The pupillary light reflex is a dynamic system for controlling the amount of light reaching the retina (Fig. 24.4). Illumination of the retina causes reflex constriction of the pupil (miosis). There is a direct component of the light reflex which mediates the constriction of the pupil of the ipsilateral eye, and a consensual component which elicits simultaneous constriction of the contralateral pupil. A light stimulus acting upon the retinal photoreceptors gives rise to activity in retinal ganglion cells, the axons of which form the optic nerve. Activity is conducted through the optic chiasma and along the optic tract, and the majority of fibres end in the lateral geniculate nucleus of the thalamus. However, a small number of fibres leave the optic tract before it reaches the thalamus and synapse in the pretectal nucleus. The information is relayed from the pretectal nucleus by short neurones which synapse bilaterally with preganglionic parasympathetic neurones in the Edinger-Westphal nucleus of the oculomotor nerve complex in the rostral midbrain. Efferent impulses pass along parasympathetic fibres of the oculomotor nerve to the orbit where they synapse in the ciliary ganglion. Postganglionic fibres (short ciliary nerves) pass to the eyeball to supply sphincter pupillae, which reduces the size of the pupil when it contracts. There is also a connection to the spinal sympathetic centre controlling the dilator pupillae. The preganglionic fibres arise from neurones in the lateral column of the first and second thoracic segments, and pass via the sympathetic trunk to the superior cervical ganglion where they synapse on postganglionic neurones. Postganglionic fibres arising from these neurones are distributed to the cavernous plexus, whence they travel mainly through the long ciliary nerves to the anterior part of the eye where they supply the dilator pupillae. Since pupillary size results from the balanced action of these two innervations, the pupil dilates when the parasympathetic stimulus ceases. The pupil dilates also in response to painful stimulation of almost any part of the body. Presumably fibres of sensory pathways connect with the sympathetic preganglionic neurones described above.
Accommodation reflex When focussing on a nearby object, the eyes converge, the lens becomes more convex, and the pupils constrict (Fig. 24.4). Information from the retina passing to the visual cortex does not constitute the afferent limb of a simple reflex in the usual sense of the term, but permits the visual areas to assess the clarity of objects in the visual field. Cortical efferent information passes to the pretectal area and thence to the Edinger-Westphal nucleus, which contains preganglionic parasympathetic neurones whose axons travel in the oculomotor nerve. Efferent impulses pass in the oculomotor nerve to the orbit where they synapse in the ciliary ganglion. Postganglionic fibres (short ciliary nerves) pass to the eyeball and stimulate contraction of the ciliary muscle, which slackens the ligament of the lens and increases the curvature of the lens for near vision. Contraction of sphincter pupillae and relaxation of dilator pupillae constrict the pupil. Simultaneously, contraction of the medial, superior, and inferior recti (all innervated by the oculomotor nerve) converges the eyes on the near target. The pupillary changes may be secondary to the convergence. page 435 page 436
Figure 24.4 The pupillary light reflex and accommodation reflex. (From Oxford Textbook of Functional Anatomy, Vol 3 Head and Neck, MacKinnon P, Morris J (eds), 1990. By permission of Oxford University Press.)
In certain central nervous diseases (e.g. tabes dorsalis) the pupillary light reflex may be lost, but pupilloconstriction as part of the accommodation reflex is retained (the Argyll Robertson pupil). The site of a lesion producing such an effect is unclear, but may involve the periaqueductal grey matter.
© 2008 Elsevier
TASTE Afferent nerve fibres carrying taste information are the peripheral processes of neuronal cell bodies in the geniculate ganglion of the facial nerve and in the inferior ganglia of the glossopharyngeal and vagus nerves. Taste from the anterior two-thirds of the tongue, excluding the vallate papillae, and from the inferior surface of the palate, is carried in the sensory root of the facial nerve (nervus intermedius). Taste buds in the vallate papillae, posterior third of the tongue, palatoglossal arches, oropharynx and, to some extent, the palate, are innervated by the glossopharyngeal nerve. Those in the extreme pharyngeal part of the tongue and the epiglottis are innervated by fibres of the vagus nerve. On entering the brain stem, these afferent fibres constitute the tractus solitarius, and they terminate in the rostral third of the nucleus solitarius of the medulla. Second-order neurones arising from the nucleus solitarius cross the midline and many ascend through the brain stem in the dorsomedial part of the medial lemniscus. They terminate in the medial part of the ventral posteromedial (VPm) nucleus of the thalamus. From the ventral posteromedial nucleus, third-order neurones project through the internal capsule to the anteroinferior part of the sensory cortex and to the limen insulae. Other ascending projections to the hypothalamus have been described which may represent the pathway by which gustatory information reaches the limbic system.
© 2008 Elsevier
HEARING The primary afferents of the auditory pathway arise from cell bodies in the spiral ganglion of the cochlea. The axons constitute the auditory component of the vestibulocochlear nerve, which enters the brain stem at the cerebellopontine angle. Afferent fibres bifurcate, and terminate in the dorsal and ventral cochlear nuclei. Onward connections make up the ascending auditory pathway (Fig. 24.5). The dorsal cochlear nucleus projects via the dorsal acoustic stria to the contralateral inferior colliculus. The ventral cochlear nucleus projects via the trapezoid body or the intermediate acoustic stria to relay centres in either the superior olivary complex, the nuclei of the lateral lemniscus, or the inferior colliculus. The superior olivary complex is dominated by the medial superior olivary nucleus which receives direct input from the ventral cochlear nucleus on both sides, and is involved in localization of sound by measuring the time difference between afferent impulses arriving from the two ears. The inferior colliculus consists of a central nucleus and two cortical areas. The dorsal cortex lies dorsomedially, and the external cortex lies ventromedially. Secondary and tertiary fibres ascend in the lateral lemniscus. They converge in the central nucleus, which projects to the ventral division of the medial geniculate body of the thalamus. The external cortex receives both auditory and somatosensory input. It projects to the medial division of the medial geniculate body, and, together with the central nucleus, also projects to olivocochlear cells in the superior olivary complex and to cells in the cochlear nuclei. The dorsal cortex receives an input from the auditory cortex and projects to the dorsal division of the medial geniculate body. Connections also run from the nucleus of the lateral lemniscus to the deep part of the superior colliculus, to coordinate auditory and visual responses. The ascending auditory pathway crosses the midline at several points both below and at the level of the inferior colliculus. However, the input to the central nucleus of the inferior colliculus and higher centres has a clear contralateral dominance. The medial geniculate body is connected reciprocally to the primary auditory cortex, which is located in the superior temporal gyrus, buried in the lateral fissure.
© 2008 Elsevier
BALANCE page 436 page 437
Figure 24.5 Ascending auditory pathway.
The vestibular sensory pathways are concerned with perception of the position of the head in space and movement of the head. They also establish important connections for reflex movements governing the equilibrium of the body and the fixity of gaze. Functionally, the vestibular apparatus is customarily divided into two components. These are the kinetic labyrinth, which provides information about acceleration and deceleration of the head, and the static labyrinth which detects the orientation of the head in relation to the pull of gravity. In terms of structure, the kinetic labyrinth
consists of the semicircular canals and their ampullary cristae, while the static labyrinth consists of the maculae of the utricle and saccule. However, the saccular macula also responds to head movements, and both maculae can be stimulated by low frequency sound, and may, therefore, have minor auditory functions. Angular acceleration and deceleration of the head cause a counterflow of endolymph in the semicircular canals, which deflects the cupola of each crista and bends the stereocilial/kinocilial bundles. This causes a change in the membrane potential of the receptor cell, which is signalled to the brain as a change in the firing frequency of the vestibular nerve afferents (either an increase or a decrease of the basal resting discharge, depending on the direction of stimulation). When a steady velocity of head movement is reached, the endolymph rapidly adopts the same velocity as the surrounding structures because of friction with the canal walls, so that the cupula and receptor cells return to their resting state. Since the three semicircular canals are orientated at right angles to each other, all possible directions of acceleration can be detected. In addition, the labyrinths on both sides of the head provide complementary information which is integrated centrally. In the maculae, the weight of the otoconial crystals creates a gravitational pull on the otoconial membrane and thus on the stereocilial bundles of the sensory cells which are inserted into its base. Because of this, they are able to detect the static orientation of the head with respect to gravity. They also detect shifts in position according to the extent to which the stereocilia are deflected from the perpendicular. As the two maculae are set at right angles to each other, and the cells of both maculae are orientated functionally in opposite directions across their striolar boundaries, this system is very sensitive to orientation. Moreover, because the otoconia have a collective inertia/momentum, linear acceleration and deceleration along the anteroposterior axis can be detected by the lag or overshoot of the otoconial membrane with respect to the epithelial surface, and so the saccular macula is able to signal these changes of velocity. Similarly, the macular receptors can be stimulated by low frequency sound which sets up vibratory movements in the otoconial membrane, although this appears to require relatively high sound levels. Efferent synapses on the afferent endings of the type I sensory cells and on the bases of type II cells receive inputs from the brain stem which appear to be inhibitory. They serve to reduce the activity of the afferent fibres either indirectly, in the case of the type I cells, or directly, for the type II cells. The information gathered by these various receptors is carried to the CNS in the vestibular nerve, which enters the brain stem at the cerebellopontine angle, and terminates in the vestibular nuclear complex. Neurones in this complex project to motor nuclei in the brain stem and upper spinal cord, and to the cerebellum and thalamus. Thalamic efferent projections pass to a cortical vestibular area which is probably located near the intraparietal sulcus in area 2 of the primary somatosensory cortex. Another major function of the vestibular system is the control of visual reflexes,
which allow the fixation of gaze on an object in spite of movements of the head, and require the coordinated movements of the eye, neck and upper trunk. Constant adjustments of the visual axes are achieved chiefly through the medial longitudinal fasciculus, which connects the vestibular nuclear complex with neurones in the oculomotor, trochlear and abducens nuclei and with upper spinal motor neurones (Fig. 19.21), and also by the vestibulospinal tracts. page 437 page 438
Abnormal activity of the vestibular input or central connections has various effects on these reflexes, e.g. the production of nystagmus. This can be elicited by a clinical test of vestibular function by syringing the external auditory meatus with water above or below body temperature, a procedure which appears to stimulate the cristae of the lateral semi-circular canal directly. Spontaneous high activity in the afferent fibres of the vestibular nerve is seen in Ménière's disease, in which those affected experience a range of disturbances including the sensation of dizziness and nausea, the latter reflecting the vestibular input to the vagal reflex pathway. REFERENCES Cagan RH (ed) 1989 Neural Mechanisms in Taste. Boca Raton, FL: CRC Press Hubel DH 1988 Eye, brain and vision. Scientific American Library Series No. 22 Kandel ER, Schwartz JH, Jessel TM (eds) 2000 Principles of Neural Science, 4th edn. New York; McGraw-Hill Oertel D, Fay RR, Popper AN (eds) 2002 Integrative Functions in the Mammalian Auditory Pathway. Springer Handbook of Auditory Research, vol. 15. New York: Springer Zeki S 1993 A Vision of the Brain. Oxford: Blackwell Scientific
© 2008 Elsevier
SECTION 3 HEAD AND NECK Susan Standring (Lead Editor) Barry KB Berkovitz (Editor) Carole M Hackney (Editor, chapter 39) The late Gordon L Ruskell (Editor, chapters 41, 42) Patricia Collins (Embryology, Growth and Development) Caroline Wigley (Microstructure) With specialist contributions on clinical and functional anatomy by Martin E Atkinson (chapter 36), Simon A Hickey (32, 35, 36, 38, 39), John D Langdon (25, 27, 29-31, 33-35), Daniel E Lieberman (27), V Mahadev (29), BJ Moxham (27, 33), Jeff Osborn (30), and Allan Thexton (30, 35) Critical reviewers: Paul Cartwright (chapter 25), Michael Dilkes (31), Peter Morgan (29 & 33), BJ Moxham (30) page page page page
438 439 439 440
© 2008 Elsevier
25 Surface anatomy of the head and neck To avoid repetition, additional information related to clinical and surgical anatomy is provided in the appropriate chapters in this section.
© 2008 Elsevier
HEAD SKELETAL SURFACE LANDMARKS (Figs 25.1, 25.2) The skeletal surface landmarks of the head can be examined from the back, from the side and from the front. The palpable bony landmarks of the calvarium are visible in bald people but more commonly they have to be palpated through the hair. The anterior fontanelle, at the junction of the coronal and sagittal sutures, may be palpated until c.18 months after birth.
Figure 25.1 Anterior aspect of the head: bones. (Photograph by Sarah-Jane Smith. Artwork modified from Lumley JSP 2002 Surface Anatomy, 3rd edn. Edinburgh: Churchill Livingstone.)
Most of the superficial aspect of the skull is covered by skin, subcutaneous tissue and thin muscles, and so it is relatively easy to feel the bony prominences and surfaces. The pericraniocervical line demarcates the head from the neck. It runs from the midpoint of the chin anteriorly to the external occipital protuberance posteriorly. Starting anteriorly from the midline, where the mental tubercles may be felt, the lower border of the body of the mandible may be traced to the mandibular angle (the angle is often everted in the male, incurved in the female). An oblique line joins the mental tubercle to the lower end of the anterior border of
the ramus of the mandible. The teeth can be easily felt (when present) by palpating the superior border of the body of the mandible through the cheek. The mental foramen, which transmits the mental nerve and vessels, lies below the level of the premolar teeth, c.2 fingers' breadth from the median plane and c.1 finger's breadth above the lower border of the mandible. It is worth noting that the supraorbital, infraorbital and mental foramina all lie in approximately the same vertical plane.
page 441 page 442
Figure 25.2 Lateral aspect of the head: bones. (Photograph by Sarah-Jane Smith. Artwork modified from Lumley JSP 2002 Surface Anatomy, 3rd edn. Edinburgh: Churchill Livingstone.)
The posterior border of the ramus of the mandible may be palpated up to the neck of the condylar process, which lies just under the lobule of the ear. The ramus is easily palpable, although mostly covered by masseter. With the mouth loosely open, the mandibular notch between condylar and coronoid processes can be felt through masseter. As the mouth is opened and closed, articulation at the temporomandibular joint may be appreciated as the condylar process slides up and down the articular eminence of the temporal bone; anteriorly the coronoid process can be felt to move from its resting position under the zygomatic arch as the mouth is opened. A finger probing inwards just behind the condylar process
enters the small retromandibular fossa where, anteriorly, the mandibular neck, and superiorly, the inferior wall of the external acoustic meatus, may be felt. The meatus is bounded anteriorly by the tragus, a small curved flap that partly projects over the orifice of the meatus. Palpating behind the meatus, first the anterolateral aspect and tip of the mastoid process, and, on deeper palpation, a somewhat indistinct resistance offered by the styloid process and its attached structures, will be encountered. If the examining finger is then taken posteriorly over the convexity of the mastoid process, the lateral part of the superior nuchal line is felt. This curves upwards to meet the contralateral superior nuchal line at a bony midline prominence, the external occipital protuberance, which is an important surface marking for the confluence of the underlying dural venous sinuses. The lateral aspect of the face consists of the temporal region above, the cheek in the middle, and the lower jaw below. The temporal region lies in front of the external ear and above the zygomatic arch. It is demarcated superiorly by the temporal lines, which indicate the upper limit of temporalis. The temporal lines (superior and inferior) curve downwards anteriorly and posteriorly and help to delineate the temporal fossa. Below, the temporal fossa is bounded laterally by the zygomatic arch, which consists of the zygomatic process of the temporal bone behind and the temporal process of the zygomatic bone in front. The body of the zygomatic bone, which accounts for the variably prominent 'cheekbone', forms the anterior part of the arch. If the sharp posterior margin of the frontal process of the zygomatic bone is followed upwards, it fuses with the zygomatic process of the frontal bone. Continuing posteriorly, in the line of a gentle arch, the temporal lines may be palpated. The lower temporal line terminates by curving downwards and forwards to end just above the root of the mastoid process as the supramastoid crest on the squamous part of the temporal bone. The pterion is a small circular area within the temporal fossa which contains the junction of the frontal, sphenoid, parietal and temporal sutures. It usually lies 4 cm above the zygomatic arch and 3.5 cm behind the frontozygomatic suture, and marks the anterior branch of the middle meningeal artery and the Sylvian point of the brain. Its position can be estimated roughly by a shallow palpable hollow, c.3.5 cm above the centre of the zygomatic bone. The prominence of the cheek is formed by the underlying zygomatic bone. The suprameatal triangle lies above and behind the external acoustic meatus and overlies the lateral wall of the mastoid (tympanic) antrum. It is bounded above by the supramastoid crest, in front by the posterosuperior margin of the meatus, and behind by a vertical tangent to the posterior border of the meatus. The forehead extends from the hair margin of the scalp to the eyebrows. The superciliary arch is usually palpable above the orbit and is better marked in the male than the female. Above it the rounded frontal tuberosity may be felt c.3 cm above the midpoint of each supraorbital margin. Between the superciliary arches there is a small horizontal ridge called the glabella, again easily palpable. Below the glabella the nasal bones meet the frontal bone in a small depression, the nasion, at the root of the nose. With a little finger inserted into the nostril, the bony margins of the anterior nasal aperture can be felt: they are formed by the inferior border of the nasal bone, the sharp margins of the nasal notch, and the coapted nasal spines of the maxillae. The orbital opening is somewhat quadrangular. The supraorbital margin is formed entirely by the frontal bone and is easily palpable. At the junction of its sharp
lateral two-thirds and rounded medial third (c.2 fingers' breadth from the median plane) the supraorbital notch, if present, may be felt. This notch transmits the supraorbital nerve and vessels, and pressure exerted here with the fingernail can be painful. A frontal notch may also be found towards the bridge of the nose and is associated with the supratrochlear neurovascular bundle. The lateral margin of the orbit consists of the frontal process of the zygomatic bone and the zygomatic process of the frontal bone. The frontozygomatic suture between them may be felt as a palpable depression. Approximately 1 cm below this suture a tubercle (Whitnall's tubercle) may be palpated within the orbital opening: it gives attachment to the lateral palpebral and cheek ligaments. The inferior border of the orbit is formed by the zygomatic bone laterally and the maxilla medially. It blends into the less obvious medial margin formed above by the frontal bone and below by the lacrimal crest of the frontal process of the maxilla. A shallow fossa behind the lower part of the medial wall houses the nasolacrimal sac. The infraorbital foramen, which transmits the infraorbital nerve and vessels, lies c.0.5 cm below the infraorbital margin, a finger's breadth from the side of the nose and above the canine fossa. The anterior surface of the maxilla is extensive and may be palpated between the infraorbital margin and the alveolar processes that bear the upper teeth (when present). The canine eminence overlies the roots of the canine tooth and separates the incisive fossa anteriorly from the deeper canine fossa posteriorly. The palatine process of the maxilla (which forms the greater part of the roof of the mouth) is easily palpable within the mouth.
UNDERLYING SOFT TISSUES AND VISCERA The muscular, fatty and cutaneous features which so clearly differentiate individuals are readily apparent on inspection. The external features of the eyelids and eyebrows are described on page 681. The tympanic membrane may be examined under direct vision using an auroscope. The retinal vascular supply may be examined directly by ophthalmoscopy (p. 716). Two important masticatory muscles are palpable when the jaw is clenched, but are difficult to define when relaxed (Fig. 25.3). Temporalis, which lies in the temporal fossa and is covered by the temporal fascia, is palpable if the flat of the hand is placed on the side of the head and the jaw is clenched and unclenched. Masseter is similarly easily palpable with the jaw clenched, when its anterior border stands out. The palatine tonsil can be represented by an oval area over the lower part of masseter, just above and in front of the angle of the mandible.
page 442 page 443
Figure 25.3 Lateral aspect of the head and neck: soft tissues. (Photograph by SarahJane Smith. Artwork modified from Lumley JSP 2002 Surface Anatomy, 3rd edn. Edinburgh: Churchill Livingstone.)
Figure 25.4 Parotid gland. (Photograph by Sarah-Jane Smith. Artwork modified from Lumley JSP 2002 Surface Anatomy, 3rd edn. Edinburgh: Churchill Livingstone.)
The parotid gland is soft and indistinct and lies largely below the external acoustic meatus, wedged between the ramus of the mandible, the mastoid process and sternocleidomastoid (Fig. 25.4). The anterior border of the gland is represented by a line descending from the mandibular condyle to a point just above the middle of the masseter and then to a point c.2 cm below and behind the angle of the mandible. Its concave upper border corresponds to a curve traced from the mandibular condyle across the lobule of the ear to the mastoid process. The posterior border corresponds to a line drawn between the posterior ends of the anterior and upper borders. The parotid duct arises from the gland just above half-way along its anterior margin. It runs over masseter to its anterior border where it bends sharply to pierce the underlying buccinator. The orifice of the duct is visible as a small papilla within the cheek opposite the second upper molar tooth. The duct can be represented by the middle third of a line drawn from the lower border of the tragus of the auricle to a point midway between the ala of the nose and labial margin of the upper lip. It can be felt on the face (or more easily in the vestibule of the mouth) and rolled on the anterior border of masseter by pressing the finger backwards on it (preferably with the teeth clenched, to tense masseter). With the mouth open it is possible to examine all the teeth and the palatine tonsils
(when present), and to inspect and palpate the orifice of the parotid duct. The tongue may be examined for its general appearance and any abnormalities of movement, which may reflect neuronal damage. Course of vessels
Two main arteries, the facial and superficial temporal arteries, can be palpated on the face (Fig. 25.4). The pulsation of the facial artery can be felt as it crosses the lower margin of the body of the mandible immediately in front of masseter and again just over a centimetre from the angle of the mouth, between a finger placed within the mouth and a thumb placed on the skin surface. If the lateral part of the lip is gripped in a similar manner the pulsation of the labial artery can be felt beneath the mucous surface c.0.5 cm from the free margin of the lip. The facial artery subsequently runs a tortuous course towards the medial corner of the eye. The superficial temporal artery arises within the parotid gland and passes upwards across the zygomatic process of the temporal bone, immediately in front of the tragus of the auricle. Compression at this point allows palpation of a superficial temporal pulse. This is of special value during anaesthesia when access to the patient is frequently restricted to the head. The artery branches into frontal and parietal divisions c.2.5 cm above the zygomatic arch.
Figure 25.5 Cutaneous innervation of the head and neck. (Photograph by Sarah-Jane
Smith. Artwork modified from Lumley JSP 2002 Surface Anatomy, 3rd edn. Edinburgh: Churchill Livingstone.)
Course of nerves (Fig. 25.5)
The facial nerve exits the skull at the stylomastoid foramen and therefore is deep to the posterior margin of the external acoustic meatus. The surface marking of the cranial exit of the facial nerve is a point immediately in front of the intertragic notch (situated between the tragus and the antitragus of the auricle). From here the nerve crosses the styloid process to enter the substance of the parotid salivary gland, where it divides into five main branches which radiate out across the face (p. 514). The distribution of the trigeminal dermatomes is described on page 512.
© 2008 Elsevier
NECK SKELETAL SURFACE LANDMARKS (Fig. 25.6) page 443 page 444
Figure 25.6 Lateral aspect of the head and neck. (Photograph by Sarah-Jane Smith. Artwork modified from Lumley JSP 2002 Surface Anatomy, 3rd edn. Edinburgh: Churchill Livingstone.)
The superior limit of the neck is the pericraniocervical line. Inferiorly the neck blends with the thorax and upper limb at the level of the clavicle and scapula. Several skeletal features are easily palpable. At the back of the neck the cervical vertebrae may be felt in the midline. The transverse process of the first cervical vertebra (atlas) may be palpated in the uppermost part of the neck, in the hollow
between the mastoid process of the temporal bone and the external ear. Just below, the transverse process of the second cervical vertebra (axis) can be felt on deep palpation. Inferiorly the spine of the seventh cervical vertebra (vertebra prominens) is especially prominent, particularly when the neck is flexed. The remaining cervical spines are indistinct because they are covered by the ligamentum nuchae, which separates the postvertebral musculature on both sides. However, the spines of the second, fifth and sixth cervical vertebrae may be felt on deeper palpation. The hyoid bone may be felt a few centimetres below and behind the chin, especially if the neck is extended. It may be palpated between finger and thumb and moved from side to side. The hyoid bone lies approximately at the level of the third cervical vertebra. The most obvious palpable feature in the front of the neck below the hyoid bone is the thyroid cartilage, and the prominent midline subcutaneous laryngeal prominence or Adam's apple, which indicates the line of fusion of the two thyroid laminae. In the male, the prominence is usually clearly visible, whereas in the female it is not usually apparent, even when the neck is viewed from the side. The curved upper border of the thyroid cartilage and the thyroid notch are easily palpable. The middle part of the thyroid cartilage lies at the level of the fifth cervical vertebra. The anterior arch of the cricoid cartilage can be felt below the inferior border of the thyroid cartilage. The cricoid cartilage lies at the level of the sixth cervical vertebra. The groove between the thyroid and cricoid cartilages, filled by the anterior (median) cricothyroid ligament, is a useful site for emergency access to the airway if there is obstruction at or above the vocal cords (cricothyroid puncture). In the child the laryngeal structures lie at a higher level. The trachea can be palpated inferior to the cricoid cartilage. It normally lies in the midline but may be deviated by disease. The upper tracheal cartilages may be impalpable if they are covered by the thyroid isthmus, which joins the two lobes of the gland across the midline of the neck. The clavicle is a sigmoid-shaped bone which is easily visible in thin people and palpable in all except the morbidly obese. Its medial two-thirds are rounded and convex forwards and the lateral third is flat and concave forwards. The suprasternal (jugular) notch lies between the medial expanded ends of the clavicles: its inferior border is the superior edge of the manubrium sterni. For much of its length the clavicle may be almost encircled by two fingers, but medially its massive ligamentous attachments make definition more difficult. The posterior end of the first rib may sometimes be felt rather indistinctly as it lies in the supraclavicular fossa in the root of the neck, between sternocleidomastoid and trapezius and above the clavicle. It should be remembered that the head and neck are extremely mobile but, with the head held in the anatomical position, the following vertebral levels should be noted: C1 C2 C3
Dens, level of nasopharynx Level of oropharynx and dependent soft palate with the mouth open Level of body of hyoid and its greater cornu
C3-4 junction C4-5 C6
Level of upper border of thyroid cartilage and bifurcation of common carotid artery Level of thyroid cartilage level of cricoid cartilage
UNDERLYING SOFT TISSUES AND VISCERA From the front and side (Fig. 25.7) the neck is obviously divided into two major portions by sternocleidomastoid. These are the anterior and posterior cervical triangles, both of which may be subdivided. The anterior cervical triangle may be divided into a submental triangle, a muscular triangle, a carotid triangle and a digastric triangle. The posterior cervical triangle may be split into the occipital triangle and the supraclavicular triangle. The structures that form the boundaries of some of the lesser triangles are not readily palpable or visible. The base of the anterior triangle is formed by the base of the mandible and a line from its angle to the mastoid process, and the sides are formed by the midline anteriorly and the anterior edge of sternocleidomastoid laterally. The triangle is best inspected from the front, but best examined bimanually with the examiner standing behind the subject and using the fingers of both hands to examine the structures within the triangle. Inspection reveals the rounded tendinous oblique head of sternocleidomastoid, which arises from the superolateral angle of the manubrium, and the more vertical muscular portion, which arises from the upper surface of the medial third of the clavicle. Between these two heads there is usually a hollow in which the internal jugular vein lies just before it joins the subclavian vein posterior to the clavicle.
page 444 page 445
Figure 25.7 Anterior and posterior triangles of the neck. (Photograph by Sarah-Jane Smith. Artwork modified from Lumley JSP 2002 Surface Anatomy, 3rd edn. Edinburgh: Churchill Livingstone.)
Figure 25.8 Anterior aspect of the neck: bones and muscles; larynx and thyroid gland. (Photograph by Sarah-Jane Smith. Artwork modified from Lumley JSP 2002 Surface Anatomy, 3rd edn. Edinburgh: Churchill Livingstone.)
The trachea and its cartilaginous 'rings' can be palpated inferior to the cricoid cartilage. If the examining finger is moved laterally, the lobes of the thyroid gland may be felt, particularly if the subject is asked to swallow. The thyroid gland consists of two lobes lying either side of the thyroid cartilage and joined in the midline of the neck by an isthmus (Fig. 25.8). The upper border of each lobe lies alongside the lower half of the lamina of the cartilage. The lower border reaches towards the sternal end of the clavicle. The isthmus, c.2 cm wide, lies in front of the trachea just below the cricoid cartilage and usually overlies the second and third tracheal 'rings'. Taken superiorly, an examining finger will enter either the submental or submandibular triangle, in which enlarged lymph nodes or salivary glands may be felt. The submandibular gland extends c.2 cm beneath the lower border of the mandible and reaches the approximate level of the hyoid bone. Its posterior border is level with the angle of the mandible; its anterior border extends forwards c.4 cm.
It should be noted that above the hyoid bone the musculature runs in a predominantly horizontal or oblique direction, and below the hyoid it runs in a vertical direction. Lymph nodes above the hyoid bone tend to be disposed in a horizontal plane and lie mainly just below the pericraniocervical line, whereas the deep cervical nodes run vertically and are related to the internal jugular vein. Apart from a few retrovisceral nodes and some deep to sternocleidomastoid, members of all of the groups of lymph nodes in the head and neck are clinically palpable when enlarged. The boundaries of the posterior triangle are the posterior border of sternocleidomastoid, the middle third of the superior surface of the clavicle, which forms the base, and the anterior margin of trapezius. The apex is the point where sternocleidomastoid and trapezius approximate to each other at the superior nuchal line. The lower portion of the posterior triangle forms the supraclavicular fossa, a very important clinical area, which lies just above and behind the clavicle at the confluence of the thoracic inlet and the aditus to the axilla and arm. It is better inspected from in front, but palpated from behind, when a right-handed examiner stands behind and to the right of the subject. When inspecting the supraclavicular fossa, the pulsation of the great veins may be seen if the central venous pressure is raised. The external jugular vein, a prominent feature, may be distended due to kinking, raised venous pressure or obstruction. The supraclavicular fossa is a common site in which to feel pathologically enlarged lymph nodes. In particular, cancers of the upper gastrointestinal tract and of the lung frequently spread to the supraclavicular group of nodes. The posterior end of the first rib may be felt as a fullness in the posterior aspect of the fossa. The subclavian artery can be felt pulsating as it crosses the first rib, and the trunks of the brachial plexus may be felt above and behind it. A point c.2.5 cm above the middle of the medial third of the clavicle marks the level of the neck of the first rib and thus the surface marking for the apex of the dome of the cervical pleura and lung. Course of vessels
In the neck, the common carotid artery and its continuation, the internal carotid artery, may be represented by a more or less straight line from the sternoclavicular joint to a point just behind the condyle of the mandible. At the level of the upper border of the thyroid cartilage (approximately at a level between the third and fourth cervical vertebrae) the common carotid artery divides into the external and internal carotid arteries. The transverse process of the sixth cervical vertebra is prominent (Chassaignac's tubercle), and the common carotid artery may be compressed here. Above this level the artery is superficial and its pulsation can be readily felt beneath the anterior border of sternocleidomastoid. The pulse is produced as much by the roots of the main branches as it is by the internal and external carotid arteries themselves, and this is therefore one of the prime sites in the body to feel for a pulse. The subclavian artery enters the root of the neck behind the sternoclavicular joint. It arches upwards to reach a point c.2 cm above the clavicle deep to the posterior border of sternocleidomastoid, and then passes across the upper surface of the first rib behind the middle region of the clavicle. A pulse may be felt at this point
when the artery is compressed against the first rib. The anterior jugular vein runs downwards beneath the chin approximately a finger's breadth from the midline. It turns laterally c.2.5 cm from the sternal end of the clavicle and passes beneath sternocleidomastoid to drain into the external jugular vein. The veins of each side join to form a jugular arch just above the manubrium sterni. The external jugular vein lies superficial to sternocleidomastoid and can be represented by a line which starts just below and behind the angle of the mandible and runs down to the clavicle near the posterior border of sternocleidomastoid (Fig. 29.8). It drains into the subclavian vein after penetrating the investing layer of deep cervical fascia. The vein may be kinked at this point. If the proximal part of the vein is damaged it may be held open by the surrounding fascia: air can then be sucked in, resulting in an air embolus. The external jugular vein can be distended if venous pressure is raised, e.g. by performing Valsalva's manoeuvre (forced expiration against a closed mouth and blocked nostrils) or by supraclavicular digital pressure. The internal jugular vein runs in the carotid sheath, lying just lateral to the arteries. It therefore has similar surface markings to those described for the common and internal arteries and is represented in surface projection by a broad band from the lobule of the ear to the medial end of the clavicle, where it joins the subclavian vein. The inferior bulb of the internal jugular vein lies in the lesser supraclavicular fossa (the depression between the sternal and clavicular heads of sternocleidomastoid): this site is one of several at which the internal jugular vein may be accessed for central vein cannulation (see p. 948). Pulsation of the great veins may be seen in this region if the central venous pressure is raised. The subclavian vein runs just below the subclavian artery. Course of nerves page 445 page 446
The course of the spinal accessory nerve can be indicated by a line that crosses the floor of the posterior triangle passing from the tragus of the auricle to the junction of the lower and middle thirds of the anterior border of trapezius. This line will also cross the (palpable) transverse process of the atlas c.1 cm below the mastoid process and the junction of the upper and middle thirds of the posterior border of sternocleidomastoid. The sensory nerves of the cervical plexus, which supply the skin of the neck, emerge from behind the posterior border of sternocleidomastoid just below the spinal accessory nerve (Fig. 25.5). The roots and trunks of the brachial plexus can be represented by a line passing between the middle of the posterior border of sternocleidomastoid and the middle of the clavicle. With the head held to the opposite side, the upper trunk of the plexus is palpable. The trunks lie above and behind the subclavian artery. The divisions of the brachial plexus are situated behind the clavicle near the lateral border of the first rib. The cervical sympathetic chain has three ganglia and lies at the side of the neck
behind the common and internal carotid arteries. The superior cervical ganglion lies slightly anterior to the (palpable) transverse process of the second cervical vertebra, while the middle cervical ganglion lies just in front of the transverse process of the sixth cervical vertebra (which is difficult to palpate). The inferior cervical ganglion may be fused with the first thoracic cervical ganglion, forming the stellate ganglion. Stellate ganglion block is often employed to perform a sympathetic nerve block to the head and neck, or to the arm: the surface marking is approximately at the level of the transverse process of the seventh cervical vertebra (situated two finger's breadths above the sterno-clavicular joint when the neck is fully extended) (Ellis & Feldman, 1997).
© 2008 Elsevier
26 Overview of the development of the head and neck Head development is distinct from that of the trunk, utilizing region-specific genes, signalling mechanisms and morphogenetic processes. Evolution of the vertebrate head was made possible by the origin of a novel cell population, the neural crest (Gans & Northcutt 1983). Cranial neural crest cells emigrate from the neural folds before cranial neurulation is complete, and also differ from those of the trunk in having the potential to form connective and skeletal tissues; they make major contributions to the skull. Sensory placodes form the nasal pits, the lens, and the otocyst. These are areas in which the otherwise squamous epithelium develops a pseudostratified structure before undergoing morphogenesis to form a pit and then (for the lens and otic pits) a closed cyst. Epibranchial placodes are similar pseudostratified epithelial thickenings within the proximal ectoderm of each pharyngeal arch; they contribute cells by epithelial-mesenchymal transformation to underlying neural crest cell condensations to form the cranial sensory ganglia (there is no functional difference between the ganglion cells of each origin). Three separate populations of neural crest cells migrate from the cranial neural folds (Fig. 26.5). The first of these populations originates from the diencephalic region of the forebrain, the midbrain, and the first two rhombomeres of the hindbrain. They migrate to surround the telencephalon and nasal region (frontonasal mesenchyme), and the maxillary and mandibular regions of the first arch (Jiang et al., 2002). The other two populations contribute anterior neck structures. They migrate from rhombomeres 3-8, into the second and subsequent pharyngeal arches (Fig. 26.5). The segmental organization of the embryonic head caudal to rhombomere 2 is related to the expression of evolutionarily conserved HOX genes in the rhombomeres and their derived neural crest.
© 2008 Elsevier
THE SKULL The skull is composed of the neurocranium, which surrounds the brain, eyes and inner plus middle ears, and the viscerocranium, which is derived from the frontonasal and first arch mesenchyme and forms the facial skeleton. The whole of the viscerocranium is neural crest-derived; the neurocranium (skull vault and the skull base) is derived from mesenchyme of both neural crest and mesodermal origin. Bone anlagen first form as mesenchymal condensations, which may either ossify directly (intramembranous ossification) or via a cartilaginous precursor (endochondral ossification). There is no correspondence between the tissue origin and the type of ossification process. The neural crest-and mesoderm-derived components of the skull vault both form by intramembranous ossification, corresponding to the dermal bones that protect the brain of ancestral fishes. Similarly, tissues of both origins contribute to the endochondral skull base. (For details of skull development, see Chapter 28.)
© 2008 Elsevier
THE EMBRYONIC PHARYNX The embryonic pharynx forms the template for the future face (except for the nasal region, which is derived from frontonasal mesenchyme), palate, and anterior neck structures. Its development from neural crest, surface ectoderm and foregut endoderm involves spatiotemporal integration of cell movement, tissue growth and tissue interactions.
PHARYNGEAL APPARATUS(Figs 26.1, 26.2, 26.3, 26.4) After head fold formation, the stomodeum, or primitive mouth, lies between the maxillary and mandibular parts of the first pharyngeal arch; these are bounded rostrally by the projecting forebrain and caudally by the cardiac prominence (Fig. 26.1). The neck, which will subsequently intervene between the developing jaws and thorax, is absent: it is formed later by modification of the second and subsequent pairs of branchial (pharyngeal) arches anteriorly and the neural tube and somite-derived structures posteriorly (Figs 26.2, 26.3, 26.4), and by descent of the heart. There are five pairs of pharyngeal arches in mammals, numbered 1,2,3,4, and 6 by analogy with their evolutionary origin, the fifth arch having been lost. In the earliest jawless vertebrates (Agnatha), the branchial arches were a uniform series of bars behind the gill clefts. However, long before the evolution of the terrestrial vertebrates remarkable adaptations occurred. Structures commonly regarded as the first pair of arches became the upper and lower jaws of the jawbearing vertebrates (Gnathostomata). The term 'mandibular arch' is widely used to describe the first arch but is not entirely appropriate because numerous maxillofacial and palatopharyngeal structures are derived from its proximal end. (The evolutionary origin of the jaws is currently controversial, but some authorities suggest that the trabeculae cranii, probably represented by the interorbitonasal cartilage of the human embryo, are derived from a pair of ancestral premandibular skeletal structures.) The second branchial arch is the hyoid arch, whose skeletal derivatives form the varied hyoid elements present in all jawed vertebrates. The most dorsal of these elements, the hyomandibula, is already present in cartilaginous fishes as a strut between the skull and the jaw joint: it reduces the cleft between the mandibular and hyoid arches to a small opening, the spiracle. Further evolution of this region occurred in land animals in connection with the auditory apparatus. At first the arches produce rounded ridge-like prominences of both the overlying ectoderm and of the endodermal lining of the lateral walls and floor of the pharynx. Ectoderm and endoderm are in virtual contact in the depressions between these prominences, i.e. the ectodermal clefts and endodermal pouches. The thin membranes so formed break down permanently in gill-breathing vertebrates, and transiently in reptile embryos, but persist in mammals, in which open channels or 'true clefts' are not formed. The region of the embryo which contains the rostral foregut surrounded by mesenchyme and ectoderm of the branchial arches constitutes the embryonic pharynx, and the stage of development at which the arches are prominent has been termed the pharyngula stage (p. 206). The first arch, which forms most of the face, is composed largely of ectoderm, both on the outer and inner surfaces and within the arch (neural crest mesenchyme). Technically speaking therefore, the first arch is not a pharyngeal structure, unlike the subsequent arches which are composed of ectoderm externally, pharyngeal endoderm internally and neural crest mesenchyme within the arches. The arch musculature is mesodermal in origin. Since the term "branchial" means "of the gills", the term "pharyngeal arch" will be used in the following account of human embryology. page 447 page 448
Figure 26.1 Series of scanning electron micrographs of rat embryos at days 11, 12 and 13: lateral view. A, Day 11, the pharyngula stage; the otic vesicle is still open but the lens vesicle has yet to invaginate; the first, second and third pharyngeal arches are present; an upper limb bud is present dorsal to the heart. B, Day 12, the lens vesicle has invaginated but is still open; the maxillary process has developed and is beneath the eye; the upper limb is becoming paddle-shaped and the lower limb is present. C, Day 13, the eyelids are beginning to develop; the maxillary process is merging with the lateral nasal process; both limb buds are well developed. The relative size and number of somites can be seen at each age. (Photographs by P Collins; printed by S Cox, Electron Microscopy Unit, Southampton General Hospital.)
In general, each pharyngeal arch consists of a mesenchymal core covered by epithelium (Figs 26.2, 26.3). The latter may be entirely ectodermal (as in the first arch), or ectodermal externally and endodermal internally (as in the remaining
arches). The mesenchymal core is derived from neural crest, paraxial mesoderm and angiogenic mesenchyme, which give rise to region-specific structures. Consequently each arch contains a skeletal element derived from neural crest mesenchyme; associated striated muscle from paraxial mesoderm; an arch artery from angiogenic mesenchyme and motor and sensory nerves from arch-specific cranial nerves. The epithelium covering each arch is patterned by the underlying mesenchyme and endoderm. Patterning is arch-specific and produces keratinized stratified squamous epithelium (hair, sweat, sebaceous and ceruminous glands); pseudostratified columnar epithelium (teeth, salivary, mucous and lacrimal glands); glandular epithelia (thyroid, parathyroids, thymus); and lymphoid tissues of the oro-and nasopharynx. The skeletal element is formed by condensation of neural crest mesenchyme, which subsequently chondrifies either wholly or in part of its length. If chondrification is complete then the element extends dorsally until it comes into contact with the mesenchymatous cranial base lateral to the hindbrain. The arch cartilage, entirely or in part, may remain as cartilage, undergo endochondral ossification, be replaced completely by intramembranous ossification, or become ligamentous. Neural crest also gives rise to the ligaments, tendons and connective tissues in the arches and the dermis underlying the surface ectoderm, which becomes the epidermis. The striated muscle of each arch, sometimes termed branchial musculature to denote its origin, is derived from the rostral continuation of the paraxial mesenchyme (in the trunk this becomes segmented to form somites). The paraxial mesoderm of the head is unsegmented, although a segmental pattern of seven cranial somitomeres has been described (Meier 1979). Clear segmentation equivalent to the somites of the trunk is seen only in the paraxial mesoderm of the occipital region, which gives rise to the occipital part of the skull and the extrinsic musculature of the tongue. The organization of cranial muscles formed from the more rostral unsegmented paraxial mesoderm is illustrated in Fig. 26.5. Myoblasts migrate from the paraxial mesoderm to sites of future muscle differentiation and form premuscle condensations prior to the development of any skeletal elements. The pattern of primary myotube alignment for any one muscle is specified by the surrounding neural crest mesenchyme and is not related to the source of the myoblasts. The rate and pattern of muscle maturation are closely associated with the development of the skeletal elements but remain unattached until an appropriate stage. Figure 26.5 shows the relationship between the notional somitomeres and the muscle masses migrating to each arch. An arch artery develops in each arch either by vasculogenesis, where angioblastic mesenchyme migrates into a region and initiates vessel development in situ, or by angiogenesis, where vessels develop by sprouting from the endothelium of pre-existing vessels. The paired arch arteries arise from the aortic sac at the rostral end of the truncus arteriosus and pass laterally each side of the pharynx to join the dorsal aortae. The nerves associated with each arch arise from the adjacent hindbrain (Figs 14.2, 14.18). The motor nerves grow out from the basal plate of the midbrain and hindbrain to innervate the striated muscle of the arches: they are termed special branchial efferent nerves because they innervate branchial musculature. Sensory nerves form from midbrain-and hindbrain-derived neural crest, extending from cranial sensory ganglia, but also have a placodal component (see above); they convey general and special somatic afferent axons. Each arch is innervated by a mixed nerve, but each nerve also has a purely sensory branch that innervates the arch rostral to its "own" arch; this sensory nerve is called the pretrematic branch, because it lies rostral to the cleft or trema between the two arches, while the mixed branch may be referred to as the post-trematic branch (Fig. 26.2). Hence the mandibular division of the trigeminal nerve is the post-trematic nerve of the first arch; the chorda tympani and greater petrosal nerves - branches of the facial nerve - are usually regarded as the pretrematic nerves of the arch. The facial nerve supplies the second arch; the glossopharyngeal is the nerve of the third arch; the superior laryngeal branch of the vagus supplies the fourth arch; and the recurrent laryngeal branch of the vagus is the nerve of the sixth arch. The tympanic branch of the glossopharyngeal and the auricular branch of the vagus have also been described as pretrematic nerves, but this attribution is not widely accepted. The difference in the courses of the right and left recurrent laryngeal nerves can be explained by the development of the aortic arch arteries. The arch nerve for arches 1-4 enters rostral to its aortic arch artery, whereas it enters the sixth arch caudal to the aortic arch artery. The sixth arch artery retains this position on the left side, where it lies caudal to and then loops round the ligamentum arteriosum in the adult. However, on the right, the dorsal part of the sixth aortic arch arteries
disappear, and so the nerve loops round the caudal aspect of the fourth aortic arch artery, i.e. the subclavian artery. page 448 page 449
Figure 26.2 Developing pharyngeal region showing (left) the pharyngeal floor and sectioned lateral walls, viewed from the dorsal aspect, and (right) details of generalized pharyngeal constituents, including arches, endodermal pouches and ectodermal grooves. (Modified with permission from Williams PL, Wendell-Smith CP, Treadgold S 1969 Basic Human Embryology, 2nd edn. Philadelphia: Lippincott.)
Figure 26.3 Oblique section through the pharynx of a human embryo of CR length 2 mm. (Modified with permission from Norris EH 1938 The morphogenesis and histogenesis of the thymus gland in man. Contrib Embryol Carnegie Inst Washington 27: 191-207.)
DEVELOPMENT OF THE PHARYNGEAL ARCHES On each side, the human circumoral first pharyngeal arch (Figs 26.1, 26.2) consists of a ventral part or mandibular process (prominence) and a dorsal part or
maxillary process (prominence). Each mandibular process, first seen at stage 10 (22 postovulatory days), grows ventromedially in the floor of the pharynx to meet its fellow in the midline, and is situated between the primitive mouth and the cardiac (pericardial) prominence. The maxillary processes are not seen until stage 13: their enlargement coincides with the proliferation of neural crest mesenchyme between the ectoderm and prosencephalon which forms the frontonasal process. The enlargement of the first arch is rostral to the site of the buccopharyngeal membrane and so the inner and outer aspects of this arch are covered with ectoderm. The second or hyoid arches, seen from stage 11, are caudal to the maxillomandibular; they also grow ventrally to meet and fuse in the midline. The third arches are seen at stage 12 (26 days) and the fourth arches by stage 13 (28 days). The fourth arches are not especially prominent, since they are largely sunk in a depression produced by the caudal overlapping of the hyoid arch. The sixth arch cannot be recognized externally and can only be identified by the arrangement of the mesenchyme and by a slight projection on the pharyngeal aspect. page 449 page 450
Figure 26.4 Oblique section through the head of a mole embryo, 4.5 mm long. The section passes through the hindbrain, the pharynx, the second (hyoid) and a part of the third pharyngeal arches.
First pharyngeal arch
The first pharyngeal arch is sufficiently different from the subsequent arches, in both structure and development, to merit separate examination. Unlike the other arches it possesses dorsal and ventral processes, and appears C-shaped in lateral view (see Fig. 26.1). The dorsal (maxillary) processes interact with overlying ectodermal epithelium and adjacent frontonasal mesenchyme, and generally form more extensive skeletal structures than the other arches. Indeed, these skeletal elements fuse with the neurrocranium (Fig. 26.6). The first arch is completely clothed with ectoderm, unlike the caudal arches which are dependent on the proximity of pharyngeal endoderm for their development. The ectoderm originates from a territory lateral to the rostral rhombencephalic neural folds (future rhombomeres 1 and 2). It includes the first epibranchial placode, which, together with the rhombomere 1-and 2-derived neural crest cells, forms the trigeminal ganglion. The neural crest of this level streams into the mandibular and maxillary prominences. The first arch contains a dorsal and ventral cartilage on each side. The former represents the palatopterygoquadrate bar, which forms part of the upper jaw in earlier vertebrates, but is much reduced in mammals. The ventral cartilage (of Meckel, Fig. 26.7) extends from the developing middle ear into the mandibular prominence, where it meets its fellow at its ventral end. The dorsal end of Meckel's cartilage, which becomes separated, was once thought to form the rudiments of both malleus and incus. However, there is strong palaeontological and comparative anatomical evidence that at least part of the incus should be regarded as a homologue of the quadrate bone of reptiles. It is therefore probably more correct to consider the incus to be a derivative of the palatopterygoquadrate cartilage, which may also contribute to the greater wing of the sphenoid bone and
to the roots of its pterygoid plates. Other than contributing the rudiment of the malleus, the intermediate part of Meckel's cartilage disappears, while its sheath persists as the anterior malleolar and sphenomandibular ligaments. The ventral part, which is much the largest, is enveloped by the developing mandible as it undergoes intramembranous ossification. The portion which extends from the mental foramen almost to the site of the future symphysis, probably becomes ossified from, and incorporated into, invading mandibular tissue, and the remainder of the cartilage is ultimately absorbed. The cells which give rise to the muscle of the first arch arise from the paraxial mesenchyme localized to the putative somitomeres 2 and 3. The muscle mass of the mandibular part of the first arch forms tensor tympani, tensor veli palatini, mylohyoid, anterior belly of digastric, and the masticatory muscles (Fig. 26.8). Tensor tympani retains its connection with the skeletal element of the arch through its attachments to the malleus, and tensor veli palatini remains attached to the base of the medial pterygoid process - which may be derived from the dorsal cartilage of the first arch. However, the masticatory muscles transfer to the mandible, which is mainly a dermal bone. All of these muscles are supplied by the mandibular nerve, the mixed 'post-trematic' nerve of the first arch. Second pharyngeal arch
The ectoderm covering the outer aspect of the second pharyngeal arch originates from a strip of ectoderm lateral to the metencephalic neural fold, as does the otic placode. The cartilaginous element of the second arch (Reichert's cartilage) extends from the otic capsule to the midline on each side (Fig. 26.6). Its dorsal end separates and becomes enclosed in the developing tympanic cavity as the stapes. The cartilage also gives rise to the styloid process, stylohyoid ligament, and the lesser cornu and cranial rim of the body of the hyoid bone (Fig. 26.7). The remainder of the hyoid bone derives from the third arch. The fully formed hyoid bone has a relatively higher and more anterior position in the neonate; it has a small ossification centre in the body of the bone, which is mainly cartilaginous at birth. Its two constituent parts, derived from the second and third pharyngeal arch cartilages, can be identified from the horizontal groove present along the body. The length of the hyoid bone from greater cornu to greater cornu is 3 cm. The stylohyoid ligament attached to the lesser cornu of the hyoid passes to a more horizontally inclined styloid process. In infancy the hyoid bone descends with the larynx to a lower position in the neck. The muscles of the second arch derive from somitomeres 4 and 5. For the most part the muscle mass migrates widely, but retains its original nerve supply from the facial nerve: migration is facilitated by the early obliteration of some of the first groove (cleft) and pouch. Stapedius, stylohyoid and posterior belly of digastric remain attached to the hyoid skeleton, but the facial musculature, platysma, auricular muscles and epicranius all lose connection with it (Fig. 26.8). Third, fourth and sixth pharyngeal arches
The ectoderm adjacent to the myelencephalic neural fold, down to the level of somite 3, develops to cover the third and fourth pharyngeal arches, and consequently has a much smaller distribution than that of the more rostral arches. The ectoderm in this region also gives rise to placodal cells which contribute to the petrosal (distal IXth) and nodose ganglia. Chondrification does not occur in the dorsal parts of the skeletal elements of the third to sixth arches. The ventral cartilage of the third arch becomes the greater cornu of the hyoid bone and the caudal part of the body of the hyoid. The final adaptations of the cartilages of the skeletal elements in the fourth and sixth arches are a source of disagreement, but the following represents a fairly general view. The thyroid cartilage develops from the fourth arch, which may also give rise to the arytenoid, corniculate and cuneiform cartilages. The cricoid cartilage may be derived from the sixth arch cartilage, or it may be a modified tracheal cartilage. The epiglottis is developed in the substance of the hypobranchial eminence and is probably not derived from 'true' branchial cartilage (Fig. 23.8). The paraxial mesenchyme from somitomeres 6 and 7 migrates to the third arch, while somitomere 7 alone appears to invade the fourth arch. The muscle masses are adapted to form the musculature of the pharynx, larynx and soft palate. Stylopharyngeus is a third arch muscle, cricothyroid develops in the fourth arch (Fig. 23.8), and the rest of the laryngeal muscles are derived from the sixth arch. The precise origin of the remaining palatal muscles and the pharyngeal constrictors is uncertain. Sternocleidomastoid and trapezius are thought to be derived partly from paraxial mesenchyme and partly from adjacent myotomes.
PHARYNGEAL GROOVES The external contours of the arches are modified as the skeletal and muscular
elements develop. The modification of the external pharyngeal grooves or clefts produces the smooth contour of the neck. The concurrent development of the internal pharyngeal pouches also contributes to this process. page 450 page 451
page 451 page 452
Figure 26.5 The organization of the head and pharynx in an embryo at about stage 14. The individual tissue components have been separated but are aligned in register through the numbered zones. (After Noden.)
Figure 26.6 A, Cartilaginous components of the base of the neurocranium from above. B, Cartilaginous components of the skull from the lateral aspect. The basal components of the neurocranium are in register with those in A. The pharyngeal arch cartilages of the viscerocranium are also shown.
The first pharyngeal groove is obliterated ventrally; its dorsal end deepens to form the epithelium of the external acoustic meatus and the external surface of the tympanic membrane. Thickened patches of ectoderm, the epibranchial placodes, appear at the dorsal ends of the first, second and fourth pharyngeal grooves. They are closely related to the developing ganglia of the facial, glossopharyngeal and vagus nerves, to which they contribute. Together with dorsolateral and suprabranchial placodal cells, the epibranchial placodes also contribute to the trigeminal and vestibulocochlear ganglia (Fig. 14.12). At the end of the fifth week the third and fourth arches are sunk in a retrohyoid depression, the cervical sinus. Cranially the sinus is bounded by the hyoid arch, dorsally by a ridge produced by ventral extensions from the occipital myotomes and by mesenchyme developing into sternocleidomastoid and trapezius. Caudally, the smaller epipericardial ridge separates the sinus from the pericardium: it curves cranially near the midline and then with its opposite fellow reaches the lingual swelling of the mandibular prominence and the hypobranchial eminence. The cervical sinus may be obliterated by caudal growth of the hyoid arches to fuse with the cardiac elevation, so excluding the succeeding arches from any part in the formation of the skin of the neck. Alternatively, the sinus may be reduced by gradual approximation of its walls from within outwards. There is a view that the surface course of the second groove persists as the curved submandibular cervical flexure line. Whatever the mechanism, the neck becomes covered with a smooth layer of epidermis. Platysma (a second arch muscle), bounded both superficially and deep by superficial fascia, passes along the neck to the anterior thoracic wall.
Figure 26.7 The skeletal derivatives (osseous and cartilaginous) of the pharyngeal arches (viscerocranium). (Modified with permission from Williams PL, Wendell-Smith CP, Treadgold S 1969 Basic Human Embryology, 2nd edn. Philadelphia: Lippincott.)
PHARYNGEAL POUCHES The first four pharyngeal pouches appear in sequence craniocaudally during stages 10-13. The rostral pharynx - primitive rostral foregut - is wide and the endoderm of the pouches approaches the ectoderm of the overlying pharyngeal grooves to form thin closing membranes (Figs 26.2, 26.3). It is compressed dorsoventrally which means that there is limited, or virtually no, true lateral wall. The close proximity of the ectoderm and endoderm between the first cleft and pouch is maintained as the tympanic membrane: there is minimal mesenchyme between the layers. The first pouch, and possibly the dorsal part of the second pouch, expand as the tubotympanic recess which gives rise to the middle ear system. The relationship between subsequent clefts and pouches diverges as mesenchyme intervenes between ectoderm and endoderm. The blind recesses of the second, third and fourth pouches are prolonged dorsally and ventrally as angular, wing-like diverticula. The endoderm of the pouches thickens and evaginates into localized regions of neural crest and undivided lateral plate mesenchyme. The second pouch is much reduced in dimensions compared to the first, and its ventral part is the focus of lymphoid development as the palatine tonsil (p. 614). A generalized ring of lymphoid tissue develops in the primitive foregut at this region (Matsunaga & Rahman 2001). The third pouch gives rise to the thymus ventrally and the parathyroid III dorsally, whereas the fourth pouch produces the parathyroid IV and an ultimobranchial body. The dorsal and ventral portions of the fourth pouch together with the lower ultimobranchial body are collectively termed the caudal pharyngeal complex (Fig. 34.1). page 452 page 453
Figure 26.8 The muscular derivatives of the prechordal mesenchyme, unsegmented paraxial mesenchyme and rostral somites. (Modified with permission from Williams PL, Wendell-Smith CP, Treadgold S 1969 Basic Human Embryology, 2nd edn. Philadelphia: Lippincott.)
REFERENCES Gans C, Northcutt RG 1983 Neural crest and the origin of vertebrates: a new head. Science 220: 268-74. Medline Similar articles Matsunaga T, Rahman A 2001 In search of the origin of the thymus: the thymus and GALT may be evolutionarily related. Scand J Immunol 53: 1-6. Medline Similar articles Full article Meier S 1979 Development of chick mesoblast. Formation of the embryonic axis and the establishment of the metameric pattern. Dev Biol 73: 24-45. Medline Similar articles page 453 page 454
© 2008 Elsevier
27 HEAD Skull and mandible The skull is the bony skeleton of the head and is the most complex osseous structure in the body. It is protective, shielding the brain, the organs of special sense and the cranial parts of the respiratory and digestive systems, and also provides attachments for many of the muscles of the head and neck, thus allowing for movement. Of particular importance is movement of the lower jaw (mandible) which occurs at the temporomandibular joint. The marrow within the skull bones is a site of haemopoiesis, at least in the young skull. The skull is composed of 28 separate bones, of which most are paired, but some in the median plane are single. Many of the bones are flat bones, consisting of two thin plates of compact bone enclosing a narrow layer of cancellous bone containing bone marrow. In terms of shape, however, the bones are far from flat and can show pronounced curvatures. The term diploë is used to describe the cancellous bone within the flat bones of the skull. The inner table is thinner and more brittle; the outer table is generally very resilient. Many bones are so thin that the tables are fused, for example the vomer and pterygoid plates. The skull bones vary in thickness in different regions, but tend to be thinner where they are covered by muscles, for example in the temporal and posterior cranial fossae. The skull is thicker in some races, but no relationship exists between this and cranial capacity which, on average is c.1400 ml. In all races, the bone is thinner in women and children when compared with adult males. The majority of bones in the skull are held firmly together by fibrous joints termed sutures. In the developing skull sutures allow for growth. There are three main arrangements: the margins of adjacent bones of a suture may be smooth and meet end-to-end, giving a simple (butt-end) suture (e.g. median palatine suture); the margins of adjacent bones may be bevelled, so that the border of one bone overlaps the other (e.g. zygomaticomaxillary suture); or the margins of adjacent bones may present numerous projections that interlock, giving a serrated appearance (e.g. sagittal suture). The complexity of serrated sutures increases from the inner to the outer surface. Fusion across sutures (synostosis) commences at c.30 years, although its variability precludes using this information to age skulls. The process of fusion commences on the internal surface of the cranium and the sagittal suture is one of the first affected. At c.40 years of age the sphenofrontal, lambdoid and occipitomastoid sutures close. In the facial region, the posterior part of the median palatine suture starts to close at c.30 years, followed by the sutures around the nose. The squamosal, zygomaticofrontal and anterior part of the intermaxillary suture rarely exhibit synostosis. Premature fusion of sutures during the early growth phase of the skull will result in various cranial abnormalities. The bones forming the base of the skull develop endochondrally and play an important part in growth. In this region, therefore, primary cartilaginous joints are encountered during growth: one of the most important is the spheno-occipital synchondrosis that disappears at c.14-16 years of age. The skull articulates with
the first cervical vertebra at the synovial atlanto-occipital joints. These joints allow for flexion and extension of the skull. Rotation of the skull does not directly involve any joints of the skull but occurs at the atlanto-axial joint between the first and second cervical vertebrae. Many important nerves and vessels pass in and out of the skull via openings termed foramina. The skull is a prime site for fractures resulting from trauma, and these structures can be damaged as a result of head injury. Detailed clinical examination should reveal signs and symptoms that, together with radiological examination, should provide information regarding the extent and seriousness of a traumatic incident. In addition to main foramina, irregular emissary foramina allow veins situated externally on the face and scalp to communicate with those lying intra cranially: spread of infection along these routes can have serious clinical consequences. For ease of navigation, the skull can be subdivided into cranium and mandible, based upon the fact that, whereas most of the bones of the skull articulate by relatively fixed joints, the mandible is easily detached. The cranium may then itself be subdivided into a number of regions. These are: the cranial vault, which is the upper, dome-like part of the skull and includes the skullcap or calvaria; the cranial base, which consists of the inferior surface of the skull extracranially and the floor of the cranial cavity intracranially; the facial skeleton, which includes the orbital cavities and the nasal fossae; the tooth-bearing bones or jaws; the acoustic cavities which contain the middle and inner ears; and the cranial cavity which houses the brain. Alternatively, the skull can be divided into neurocranium and viscerocranium. The neurocranium is defined as that part of the skull which houses and protects the brain and the organs of special sense, while the viscerocranium is associated with the cranial parts of the respiratory and digestive tracts. In the account of the skull that follows, a generalized account of a number of standard views of the skull as a whole will be given first, followed by a more detailed account of each individual disarticulated bone.
© 2008 Elsevier
EXTERNAL APPEARANCE OF ARTICULATED SKULL (Figs 27.1, 27.2) FRONTAL (ANTERIOR) VIEW The upper part of the facial region is formed by the frontal bone. Superomedial to each orbit is a rounded superciliary arch (better marked in males) between which is a median elevation, the glabella. The glabella may show the remains of the inter-frontal (metopic) suture, which is present in c.9% of adult skulls, where it ascends to the coronal suture, indicating that the frontal bone is the result of fusion of two halves that ossify independently. Above each superciliary arch is a slightly elevated frontal tuber or tuberosity. Below, where the nasal bones meet the frontal bone, is a depression marking the root of the nose. Just below the glabella, the frontal bone meets the two nasal bones at the frontonasal sutures. The point at which the frontonasal and internasal sutures meet is the nasion. Each orbital opening is approximately quadrangular. The upper, supraorbital, margin is formed entirely by the frontal bone, interrupted at the junction of its sharp lateral two-thirds and rounded medial third by the supraorbital notch or foramen, which transmits the supraorbital vessels and nerve. The lateral margin of the orbit is formed largely by the frontal process of the zygomatic bone, completed above by the zygomatic process of the frontal bone: the suture between them is a palpable depression. The lower, infraorbital, margin is formed by the zygomatic bone laterally and maxilla medially. Both lateral and infraorbital margins are sharp and palpable. The medial margin of the orbit is formed above by the frontal bone and below by the lacrimal crest of the frontal process of the maxilla. The osteology of the orbit is considered in detail on page 688. page 455 page 456
Figure 27.1 Front view of skull. (By permission from Berkovitz and Moxham, 1994.)
The central part of the face is occupied mainly by the maxillary bones, and the anterior nasal aperture lies between them. Each maxilla therefore contributes to the upper jaw, the bridge of the nose, the floor of the orbital cavity, the nasal aperture and the bone of the cheek. The medial surface of the maxilla forms the nasal notch which is the lower and partly lateral border of the anterior nasal aperture. The prominent anterior nasal spine surmounts the intermaxillary suture at the lower margin of the anterior nasal aperture and is palpable in the nasal septum. The infraorbital foramen which transmits the infraorbital vessels and nerve lies c.1 cm below the infraorbital margin. The maxillary alveolar process contains sockets for the upper teeth. The short, thick zygomatic process of the maxilla has an oblique upper surface that articulates with the zygomatic bone at the zygomaticomaxillary suture. The frontal process of the maxilla ascends posterolateral to the nasal bone to reach the frontal bone. The anterior nasal aperture is piriform, wider below, and bounded by the nasal bones and maxillae: these bones articulate with each other, with their contralateral fellows, and with the frontal bone above. The upper boundary of the aperture is formed by the nasal bones; the remainder is formed by the maxillary bones. In life, various cartilages (septal, lateral nasal, major and minor alar) help to delineate two nasal cavities: however, the macerated skull contains a single anterior nasal aperture because these cartilages are lost during preparation. The lower part of the face is formed by the body of the mandible. In the midline the mental protuberance produces the characteristic prominence of the chin. The mental foramen, which transmits the mental nerve and accompanying vessels,
lies in the same vertical plane as the supraorbital and infraorbital foramina.
Figure 27.2 Anteroposterior radiograph of skull. (By permission from Berkovitz and Moxham, 1989.)
POSTERIOR VIEW (Fig. 27.3) When the skull is viewed from behind, the occipital bone is the most prominent bone, and consequently this is sometimes termed the occipital view. The superolateral parts are formed by the parietal bones, while the temporal bones contribute the mastoid processes to the inferolateral parts of the back of the skull. The parietal bones are separated from the occipital bone by the lambdoid suture; the latter also meets the occipitomastoid and the parietomastoid sutures above and behind the mastoid process. Sutural bones are islands of bone commonly found within the lambdoid suture: they arise from separate centres of ossification and have no clinical significance. The external occipital protuberance is a midline ridge or a distinct process on the occipital bone. Superior nuchal lines extend laterally from the protuberance to a point above the mastoid processes. Inferior nuchal lines run parallel to, and below, the superior nuchal lines, while supreme nuchal lines may sometimes be seen above the superior nuchal lines. The external occipital protuberance, nuchal lines and roughened external surface of the occipital bone between the nuchal lines all afford attachment to muscles.
SUPERIOR VIEW (Fig. 27.4) Seen from above, the contour of the cranial vault varies greatly but is usually ellipsoid, or more strictly, a modified ovoid. Its greatest width lies nearer to its occipital pole. It is formed by four bones separated by three prominent sutures. The squamous part of the frontal bone is anterior, and the squamous part of the occipital bone is posterior. The two parietal bones lie between the frontal and occipital bones. The maximal parietal convexity on each site is palpable at the parietal tuber or tuberosity. The superior and inferior temporal lines run close to the tuberosity but are best seen in a lateral view. page 456 page 457
Figure 27.3 Posterior view of skull. (By permission from Berkovitz and Moxham, 1994.)
The coronal suture marks the junction of the posterior margin of the frontal bone with the anterior margins of the two parietal bones, and it descends around and forwards across the cranial vault. The sagittal suture runs in the midline between the two parietal bones. The lambdoid suture delineates the junction between the posterior borders of the right and left parietal bones and the superior border of the occipital bone, and it descends laterally across the cranial vault. The coronal and sagittal sutures meet at the bregma: in the fetal skull, together with the temporary interfrontal suture, they form the boundaries of a diamond-shaped membranefilled anterior fontanelle that persists until c.18 months after birth. The lambda, at the junction of the sagittal and lambdoid sutures, is the site of the posterior fontanelle, which closes c.2 months after birth. A parietal foramen may pierce either or both parietal bones near the sagittal suture c.3.5 cm anterior to the lambda. It transmits a small emissary vein from the superior sagittal sinus.
LATERAL VIEW (Figs 27.5, 27.6) The skull, viewed from the side, can be subdivided into three zones: face (anterior); temporal and infratemporal fossae and zygomatic arch (intermediate); occipital region (posterior).
Figure 27.4 Superior view of skull. (By permission from Berkovitz and Moxham, 1994.)
The temporal fossa is related to the temple of the head (where greying of the hair first denotes the passage of time). It is bounded inferiorly by the zygomatic arch, superiorly and posteriorly by the temporal lines on the calvaria, and anteriorly by the frontal process of the zygomatic bone. It continues beneath the zygomatic arch into the infratemporal fossa. The temporal lines often present anteriorly as distinct ridges, but become much less prominent as they arch across the parietal bone. Indeed, the superior line usually disappears posteriorly. The inferior temporal line becomes distinct once more as it curves down the squamous part of the temporal bone, forming a supramastoid crest at the base of the mastoid process. The superior temporal line gives attachment to the temporal fascia. The inferior temporal line provides attachment for temporalis. The floor of the temporal fossa is formed by the frontal and parietal bones, the greater wing of the sphenoid, and the squamous part of the temporal bones. All four bones meet on each side at an H-shaped junction of sutures termed the pterion. This is an important landmark on the side of the skull because it overlies both the anterior branch of the middle meningeal artery and the lateral (Sylvian) cerebral fissure intracranially (it is also known as the Sylvian point). The pterion corresponds to the site of the anterolateral (sphenoidal) fontanelle on the neonatal skull, which disappears about three months after birth. UPDATE Date Added: 15 May 2006 Abstract: Variation in epipteric bones in the pterion identified as a potential surgical pitfall Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=14968406&query_hl=32&itool=pubmed_docsum Epipteric bones in the pterion may be a surgical pitfall. Ersoy M, Evliyaoglu C, Bozkurt MC et al: Minim Invasive Neurosurg 46:363-365, 2003. The suture between the temporal and parietal bones is called the squamosal suture. The sphenosquamosal suture lies between the greater wing of the sphenoid and the squamous part of the temporal bone. The lateral surface of the ramus of the mandible will be described briefly here. The ramus is a plate of bone projecting upwards from the back of the body of the mandible. It bears two prominent processes superiorly, the coronoid and condylar
processes separated by the mandibular notch. The coronoid process is the site for the insertion of temporalis. The condylar process articulates with the mandibular fossa of the temporal bone at the temporomandibular joint and is the site of attachment of lateral pterygoid. The inferior and posterior borders of the ramus meet at the angle of the mandible. page 457 page 458
Figure 27.5 Lateral view of skull. (By permission from Berkovitz and Moxham, 1994.)
Figure 27.6 Lateral radiograph of skull. (By permission from Berkovitz and Moxham, 1989.)
The zygomatic arch stands clear of the rest of the skull, and the temporal and infratemporal fossae communicate via the gap thus created. The bones of the cheek are the zygomatic bone together with the zygomatic processes of the frontal, maxillary and temporal bones; the term zygomatic arch is generally restricted to the temporal process of the zygomatic bone and the zygomatic process of the temporal bone which meet at the zygomaticotemporal suture. The suture between the zygomatic process of the frontal bone and the frontal process of the zygomatic bone is the frontozygomatic suture, that between the maxillary margin of the zygomatic bone and the zygomatic process of the maxillary bone is the zygomaticomaxillary suture, and between the sphenoid and zygomatic is the
sphenozygomatic suture. As the zygomatic process of the temporal bone passes posteriorly, it becomes associated with the mandibular fossa and the supramastoid crest. Two small foramina, the zygomaticofacial and zygomaticotemporal, lie on the outer and inner surfaces respectively, of the zygomatic bone: they transmit similarly named nerves and vessels onto the face. The temporal bone is a prominent structure on the lateral aspect of the skull. Its squamous part lies in the floor of the temporal fossa and its zygomatic process contributes to the bones of the cheek. Additional components visible in the lateral view of the skull are the mandibular fossa and its articular eminence (tubercle), the tympanic plate, the external acoustic meatus, and the mastoid and styloid processes. The mandibular (glenoid) fossa is the part of the temporomandibular joint into which the condylar process of the mandible articulates. It is bounded in front by the articular eminence and behind by the tympanic plate. The articular eminence is important functionally as it provides a surface down which the mandibular condyle glides during mandibular movements. The tympanic plate of the temporal bone contributes most of the margin of the external acoustic meatus, and the squamous part forms the posterosuperior region. The external margin is roughened to provide an attachment for the cartilaginous part of the meatus. Above and behind the meatus lies a small depression, the suprameatal triangle, which is related to the lateral wall of the mastoid antrum. The mastoid process is a small inferior projection of the temporal bone, which lies posteroinferior to the external acoustic meatus. It is in contact behind with the posteroinferior angle of the parietal bone at the parietomastoid suture and with the squamous part of the occipital bone at the occipitomastoid suture. These two sutures meet the lateral end of the lambdoid suture at the asterion. The asterion coincides with the site of the posterolateral fontanelle in the neonatal skull: this fontanelle closes during the second year. A mastoid foramen may be found near or in the occipitomastoid suture and transmits an emissary vein from the sigmoid sinus. Sutural bones may appear in the parietomastoid suture. page 458 page 459
The styloid process lies anterior and medial to the mastoid process and gives attachment to several muscles and ligaments. Its base is partly ensheathed by the tympanic plate and it descends anteromedially, its tip usually reaching a point medial to the posterior margin of the mandibular ramus. However, the styloid process is very variably developed, and ranges in length from a few millimetres to a few centimetres. Often approximately straight, it is on occasion curved, when an anteromedial concavity is common, whereas a posterior concavity is rare. The infratemporal fossa is an irregular postmaxillary space located deep to the ramus of the mandible, which communicates with the temporal fossa deep to the zygomatic arch. It is best visualized, therefore, when the mandible is removed but, for completeness, is considered here. Its roof is the infratemporal surface of the greater wing of the sphenoid. The lateral pterygoid plate lies medial to the fossa, and the ramus of the mandible and styloid process lie laterally and posteriorly respectively. The infratemporal fossa has no anatomical floor. Its anterior and medial walls are separated above by the pterygomaxillary fissure lying between the lateral pterygoid plate and the posterior wall of the maxilla. The infratemporal fossa communicates with the pterygopalatine fossa through this fissure.
INFERIOR (BASAL) SURFACE (Fig. 27.7) The inferior surface of the skull, the base of the cranium, is complex and extends from the upper incisor teeth in front to the superior nuchal lines of the occipital bone behind. The region contains many of the foramina through which structures enter and exit the cranial cavity. The inferior surface can be conveniently subdivided into anterior, middle, posterior and lateral parts. The anterior part contains the hard palate and the dentition of the upper jaw, and lies at a lower level than the rest of the cranial base. The middle and posterior parts can be arbitrarily divided by a transverse plane passing through the anterior margin of the foramen magnum. The middle part is occupied mainly by the base of the sphenoid bone, the petrous processes of the temporal bones and the basilar part of the occipital bone. The lateral part contains the zygomatic arches and the mastoid and styloid processes. Whereas the middle and posterior parts are directly related to the cranial cavity (the middle and posterior cranial fossae), the anterior part (the palate) is some distance from the anterior cranial fossa, being
separated from it by the nasal cavities. Anterior part of cranial base
The bony palate within the superior alveolar arch is formed by the palatine processes of the maxillae and the horizontal plates of the palatine bones, which meet at a cruciform system of sutures. The median palatine suture runs anteroposteriorly and divides the palate into right and left halves. This suture is continuous with the intermaxillary suture between the maxillary central incisor teeth. The transverse palatine (palatomaxillary) sutures run transversely across the palate between the maxillary and the palatine bones. The palate is arched sagittally and transversely: its depth and breadth are variable but are always greatest in the molar region, the average width between the maxillary first molars being c.50 mm. The incisive fossa lies behind the central incisor teeth, and the lateral incisive foramina, through which incisive canals pass to the nasal cavity lie in its lateral walls. Median incisive foramina, present in some skulls, open on the anterior and posterior walls of the fossa. The incisive fossa transmits the nasopalatine nerve and the termination of the greater palatine vessels. When median incisive foramina occur, the left nasopalatine nerve traverses the anterior foramen, and the right nerve traverses the posterior foramen. The greater palatine foramen lies near the lateral palatal border of the transverse palatine suture, and a vascular groove which is deep posteriorly, leads forwards from it. The lesser palatine foramina, usually two, lie behind the greater palatine foramen, and pierce the pyramidal process of the palatine bone which is wedged between the lower ends of the medial and lateral pterygoid plates. The palate is pierced by many other small foramina and is marked by pits for palatine glands. Variably prominent palatine crests extend medially from behind the greater palatine foramina. The posterior border projects back as a median posterior nasal spine. The alveolar arch has 16 sockets or alveoli for teeth, varying in size and depth, some single, some divided by septa in adaptation to tooth roots.
Figure 27.7 Inferior view of skull. (By permission from Berkovitz and Moxham, 1994.)
The nasal fossae, separated in the midline by the nasal septum, lie above the hard palate. The two posterior nasal apertures (choanae) are located where the nasal fossae end. The posterior part of the septum is formed by the vomer. The upper border of the vomer is applied to the inferior aspect of the body of the sphenoid, where it expands into an ala on each side. The lateral border of each ala reaches a thin vaginal process which projects medially from the medial pterygoid plate. The two may either touch or the vaginal process may overlap the ala of the vomer inferiorly. The inferior surface of the vaginal process bears an anteroposterior groove, which is converted into a canal anteriorly by the superior aspect of the sphenoidal process of the palatine bone. This palatovaginal canal opens anteriorly into the pterygopalatine fossa and transmits a pharyngeal branch of the pterygopalatine ganglion and a pharyngeal branch from the third part of the maxillary artery. An inconstant vomerovaginal canal may lie between the ala of the vomer and the vaginal process of the sphenoid bone, medial to the palatovaginal canal, and lead into the anterior end of the palatovaginal canal. It transmits the pharyngeal branch of the third part of the maxillary artery. Middle part of cranial base page 459 page 460
The middle part of the cranial base is made up by the occipital, sphenoid and temporal bones. The body of the sphenoid bone lies anteriorly, and the basilar part of the occipital bone lies posteriorly, just in front of the foramen magnum.
Where these meet in the growing skull, the junction between the two bones is a primary cartilaginous joint, the spheno-occipital synchondrosis. This joint is important for growth of the skull in an anteroposterior direction, and ossifies at c.14-16 years of age. The basilar part of the occipital bone bears a small midline pharyngeal tubercle, which gives attachment to the pharyngeal raphe and the highest attachment of the superior pharyngeal constrictor. The middle part of the cranial base is completed by the petrous processes of the two temporal bones, which pass from the lateral sides of the base of the skull towards the site of union of the sphenoid and occipital bones. Each petrous process meets the basilar part of the occipital bone at a petro-occipital suture, which is deficient posteriorly at the jugular foramen. The petrosphenoidal suture and the groove for the pharyngotympanic tube lie between the petrous process and the infratemporal surface of the greater wing of the sphenoid. The apex of the petrous process does not meet the spheno-occipital suture and the deficit so produced is called the foramen lacerum. Each pterygoid process of the sphenoid bone bears medial and lateral pterygoid plates separated by a pterygoid fossa. Anteriorly the plates are fused, except below, where they are separated by the pyramidal process of the palatine bone. Sutures are usually discernible at this site. Laterally the pterygoid plates are separated from the posterior maxillary surface by the pterygomaxillary fissure, which leads into the pterygopalatine fossa. The posterior border of the medial pterygoid plate is sharp, and bears a small projection near the midpoint, above which it is curved and attached to the pharyngeal end of the pharyngotympanic tube. Above, the medial pterygoid plate divides to enclose the scaphoid fossa, while below it projects as a slender pterygoid hamulus, which curves laterally and is grooved anteriorly by the tendon of tensor veli palatini. The pterygoid hamulus gives origin to the pterygomandibular raphe. The lateral pterygoid plate projects posterolaterally and its lateral surface forms the medial wall of the infratemporal fossa. Superiorly and laterally the pterygoid process is continuous with the infratemporal surface of the greater wing of the sphenoid bone that forms part of the roof of the infratemporal fossa. This surface forms the posterolateral border of the inferior orbital fissure and bears an infratemporal crest associated with the origin of the upper part of lateral pterygoid. The infraorbital and zygomatic branches of the maxillary nerve and accompanying vessels pass through the inferior orbital fissure. Laterally the greater wing of the sphenoid bone articulates with the squamous part of the temporal bone. Features associated with the pterygoid plate region may be assessed radiographically (Fig. 27.8).
Figure 27.8 Horizontal CT at level of upper part of ramus of mandible showing relationships of the pterygoid plates. (By permission from Berkovitz and Moxham, 1994.)
A thin-walled depression in the temporal bone, the mandibular fossa, may be inspected when the mandible is removed, in front of which the zygomatic arch extends laterally. A distinct ridge, the articular eminence, is anterior to the fossa, and three fissures can be distinguished behind it. The squamotympanic fissure extends from the spine of the sphenoid, between the mandibular fossa and the tympanic plate of the temporal bone, and curves up the anterior margin of the
external acoustic meatus. A thin wedge of bone forming the inferior margin of the tegmen tympani lies within the fissure and divides the squamotympanic fissure into petrotympanic and petrosquamous fissures. The petrotympanic fissure transmits the chorda tympani branch of the facial nerve from the skull into the infratemporal fossa. The foramen lacerum is bounded in front by the body and adjoining roots of the pterygoid process and greater wing of the sphenoid bone, posterolaterally by the apex of the petrous part of the temporal bone, and medially by the basilar part of the occipital bone. Although it is nearly 1 cm long, no large structure completely traverses it. A large, almost circular, foramen, the carotid canal, lies behind and posterolateral to the foramen lacerum in the petrous part of the temporal bone. The internal carotid artery enters the skull through this foramen, ascends in the carotid canal, and turns anteromedially to reach the posterior wall of the foramen lacerum. It ascends through the upper end of the foramen lacerum with its venous and sympathetic nerve plexuses. Meningeal branches of the ascending pharyngeal artery, and emissary veins from the cavernous sinus also traverse the foramen lacerum. In life, the lower part of the foramen lacerum is partially occluded by cartilaginous remnants of the developmental chondrocranium. The pterygoid canal can be seen on the base of the skull at the anterior margin of the foramen lacerum, above and between the pterygoid plates of the sphenoid bone. It leads into the pterygopalatine fossa and contains the nerve of the pterygoid canal and accompanying blood vessels. The foramen ovale and the foramen spinosum lie lateral to the foramen lacerum on the infratemporal surface of the greater wing of the sphenoid bone. The foramen ovale, near the posterior margin of the lateral pterygoid plate, transmits the mandibular nerve as well as the lesser petrosal nerve, the accessory meningeal branch of the maxillary artery, and an emissary vein which connects the cavernous venous sinus to the pterygoid venous plexus in the infratemporal fossa. Posterolaterally, the smaller and rounder foramen spinosum transmits the middle meningeal artery and a meningeal branch of the mandibular nerve. The irregular spine of the sphenoid projects posterolateral to the foramen spinosum. The medial surface of the spine is flat and forms, with the adjoining posterior border of the greater wing of the sphenoid, the anterolateral wall of a groove that is completed posteromedially by the petrous part of the temporal bone. This groove contains the cartilaginous pharyngotympanic (auditory) tube and leads posterolaterally into the bony portion of the tube lying within the petrous part of the temporal bone. Occasionally the foramen ovale and foramen spinosum are confluent. The posterior edge of the foramen spinosum may be defective. A small foramen, the sphenoidal emissary foramen (of Vesalius), is sometimes found between the foramen ovale and scaphoid fossa. When present, it contains an emissary vein linking the pterygoid venous plexus in the infratemporal fossa with the cavernous sinus in the middle cranial fossa. The zygomaticotemporal foramen passes up and backwards from the posterior surface of the zygomatic bone in the anterior wall of the infratemporal fossa. It transmits the zygomaticotemporal nerve and a small accompanying artery. Posterior part of cranial base
The posterior part of the cranial base is formed by the occipital and temporal bones. Prominent features are the foramen magnum and associated occipital condyles; jugular foramen; mastoid and styloid processes of the temporal bone; stylomastoid foramen; mastoid notch and squamous part of the occipital bone up to the external occipital protuberance and the superior nuchal lines; hypoglossal canals (anterior condylar canals) and condylar canals (posterior condylar canals). page 460 page 461
The foramen magnum lies in an anteromedian position. It is oval, wider behind, with its greatest diameter being anteroposterior. It contains the lower end of the medulla oblongata, the vertebral arteries and the spinal accessory nerve. Anteriorly, the margin of the foramen magnum is slightly overlapped by the occipital condyles which project down to articulate with the superior articular facets on the lateral masses of the atlas. Each occipital condyle is oval in outline and oriented obliquely so that its anterior end lies nearer the midline. It is markedly convex anteroposteriorly, less so transversely, and its medial aspect is roughened by ligamentous attachments. The hypoglossal canal, directed laterally and slightly forwards, traverses each condyle and transmits the hypoglossal nerve, a meningeal branch of the ascending pharyngeal artery and an emissary vein from the basilar plexus. A depression, the condylar fossa, lies immediately posterior to the condyle and sometimes contains a (posterior) condylar canal for an emissary vein from the sigmoid sinus. A jugular process joins the petrous part
of the temporal bone lateral to each condyle, and its anterior border forms the posterior boundary of the jugular foramen. Laterally, the occipital bone joins the petrous part of the temporal bone anteriorly at the petro-occipital suture, and the mastoid process of the temporal bone more posteriorly at the petromastoid suture. The jugular foramen, a large irregular hiatus, lies at the posterior end of the petro-occipital suture between the jugular process of the occipital bone and the jugular fossa of the petrous part of the temporal bone. A number of important structures pass through this foramen: inferior petrosal sinus (anterior); glossopharyngeal, vagus and accessory nerves (midway); internal jugular vein (posterior). A mastoid canaliculus runs through the lateral wall of the jugular fossa and transmits the auricular branch of the vagus nerve. The canaliculus for the tympanic nerve - a branch of the glossopharyngeal nerve to the cavity of the middle ear - lies on the ridge between the jugular fossa and the opening of the carotid canal. A small notch, related to the inferior glossopharyngeal ganglion, may be found medially, on the upper boundary of the jugular foramen (it is more easily identified internally). The orifice of the cochlear canaliculus may be found at the apex of the notch. The stylomastoid foramen lies between the mastoid and styloid processes of the temporal bone on the lateral aspect. It transmits the facial nerve and the stylomastoid artery. A groove, the mastoid notch, lies medial to the mastoid process and gives origin to the posterior belly of digastric. A groove related to the occipital artery often lies medial to the mastoid notch. A mastoid foramen may be present near or in the occipitomastoid suture and, when present, it transmits an emissary vein from the sigmoid sinus. The external acoustic meatus lies in front of the mastoid process. It is surrounded inferiorly by the tympanic plate which partly ensheathes the base of the styloid process. The squamous part of the occipital bone exhibits the external occipital protuberance, supreme, superior and inferior nuchal lines, and the external occipital crest, all of which lie in the midline, posterior to the foramen magnum. The region is roughened for the attachment of muscles whose primary function is extension of the skull. UPDATE Date Added: 20 September 2005 Publication Services, Inc. Abstract: Anatomy involved in the jugular foramen approach for jugulotympanic paraganglioma resection. Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=15329021&query_hl=8 Anatomy involved in the jugular foramen approach for jugulotympanic paraganglioma resection. Inserra MM, Pfister M, Jackler RK: Neurosurg Focus. 17(2):E6, 2004.
© 2008 Elsevier
INTERNAL APPEARANCE OF ARTICULATED SKULL The cranial cavity contains the brain, the intracranial portions of cranial and spinal nerves, blood vessels, meninges and cerebrospinal fluid. It is formed by the frontal, parietal, sphenoid, temporal and occipital bones, and a part of the ethmoid bone. All the bones are lined by fibrous endocranium, the external layer of the dura mater, a tough connective tissue which traverses various foramina to join the external periosteum, the pericranium. Both membranes fuse with sutural ligaments or cartilages in the narrow interosseous intervals.
INTERNAL SURFACE OF CRANIAL VAULT (Fig. 27.9) The gradual obliteration of the sutures that occurs with age commences on the intracranial surface. The internal surface includes most of the frontal and parietal bones and the squamous part of the occipital bone. The bones are united at the coronal, sagittal and lambdoid sutures (unless fusion has obliterated them). Parietal foramina may occur near the sagittal sulcus, c.3.5 cm anterior to the lambdoid suture, for the passage of emissary veins associated with the superior sagittal sinus. The cranial vault, deeply concave, presents numerous vascular furrows. The frontal branch of the middle meningeal vein, and sometimes the artery, groove the bone deeply just behind the coronal suture. Branches of both these vessels and their parietal branches ascend backwards, and groove the internal surface of the parietal bone. Smaller grooves may mark the frontal and occipital bones. Impressions for cerebral gyri are less distinct on the vault than on the cranial base.
Figure 27.9 Internal surface of cranial vault. (By permission from Berkovitz and Moxham, 1994.)
An anteromedian frontal crest projects backwards and gives attachment to the falx cerebri, a dural partition that passes between the two cerebral hemispheres of the brain. The crest exhibits a groove related to the origin of the sagittal sulcus that accommodates the superior sagittal sinus. This groove widens as it passes back below the sagittal suture. Irregular depressions, granular foveolae, which become larger and more numerous with age, lie on either side of the sulcus and are adapted to arachnoid granulations.
CRANIAL FOSSAE (ANTERIOR, MIDDLE, POSTERIOR) (Fig. 27.10) The base of the cranial cavity is divided into three distinct fossae, the anterior, middle and posterior cranial fossae (Fig. 27.10). The floor of the anterior cranial
fossa is at the highest level and the floor of the posterior fossa is at the lowest. Anterior cranial fossa
The anterior cranial fossa is formed at the front and sides by the frontal bone, while its floor contains the orbital plate of the frontal bone, the cribriform plate and crista galli of the ethmoid bone, and the lesser wings and anterior part of the body of the sphenoid. Unlike the other cranial fossae, it does not directly communicate with the inferior surface of the cranium but instead is related to the roofs of the orbits and the nasal fossae. page 461 page 462
Figure 27.10 Floor of cranial cavity showing the cranial fossae. (By permission from Berkovitz and Moxham, 1994.)
A perforated plate of bone, the cribriform plate of the ethmoid bone, spreads across the midline between the orbital plates of the frontal bone, and is depressed below them, forming part of the roof of the nasal cavity. Olfactory nerves pass
from the nasal mucosa to the olfactory bulb of the brain through numerous small foramina in the cribriform plate. Anteriorly a spur of bone, the crista galli, projects upwards between the cerebral hemispheres. A depression between the crista galli and the crest of the frontal bone is crossed by the fronto-ethmoidal suture and bears the foramen caecum, which is usually a small blind-ended depression, but which occasionally accommodates a vein draining from the nasal mucosa to the superior sagittal sinus. The anterior ethmoidal nerve enters the cranial cavity where the cribriform plate meets the orbital part of the frontal bone and then passes into the roof of the nose via a small foramen by the side of the crista galli: the nerve grooves the crista galli. The anterior ethmoidal vessels accompany the nerve. The posterior ethmoidal canal, which transmits the posterior ethmoidal nerve and vessels, opens at the posterolateral corner of the cribriform plate and is overhung by the sphenoid bone. The convex cranial surface of the frontal bone separates the brain from the orbit and bears impressions of cerebral gyri and small grooves for meningeal vessels. Posteriorly, it joins the anterior border of the lesser wing of the sphenoid bone which forms the posterior boundary of the anterior cranial fossa. The medial end of the lesser wing constitutes the anterior clinoid process. The lesser wing joins the body of the sphenoid body by two roots which are separated by the optic canal. The anterior root, broad and flat, is continuous with the jugum sphenoidale, while the smaller and thicker posterior root joins the body of the sphenoid bone near the posterior bank of the sulcus chiasmatis. The frontosphenoid and sphenoethmoidal sutures divide the sphenoid from the adjacent bones. The posterior border of each lesser wing fits the stem of the lateral cerebral sulcus and may be grooved by the sphenoparietal sinus. Above is the inferior surface of the frontal lobe of the cerebral hemisphere and medially is the anterior perforated substance. Inferiorly the lesser wing bounds the superior orbital fissure, and completes the orbital roof. Each anterior clinoid process gives attachment to the free margin of the tentorium cerebelli and is grooved medially by the internal carotid artery as it leaves the cavernous sinus. It may be connected to the middle clinoid process by a thin osseous bar, completing a caroticoclinoid foramen around the artery. Middle cranial fossa
The middle cranial fossa is deeper and more extensive than the anterior cranial fossa, particularly laterally. It is bounded in front by the lesser wings and part of the body of the sphenoid, behind by the superior borders of the petrous part of the temporal bone and the dorsum sellae of the sphenoid, and laterally by the squamous parts of the temporal bone, parietal bone and greater wings of the sphenoid. This region corresponds with the middle part of the cranial base. Centrally the floor is narrower and formed by the body of the sphenoid bone. The hollowed out area is the site of the hypophysial (pituitary) gland and is therefore termed the hypophysial (pituitary) fossa. The area has the shape of a Turkish saddle, and so is also known as the sella turcica. The anterior edge of the hypophysial fossa is completed laterally by a middle clinoid process, the floor forms the roof of the sphenoidal air sinuses, and the posterior boundary presents
a vertical pillar of bone, the dorsum sellae. The superolateral angles of the dorsum are expanded as the posterior clinoid processes. A fold of dura, the diaphragma sella, is attached to the anterior and posterior clinoid processes and roofs over the hypophysial fossa. The smooth upper part of the anterior wall of the fossa is the jugum sphenoidale which is bounded behind by the anterior border of the grooved sulcus chiasmatis leading laterally to the optic canals. The optic nerve and ophthalmic artery pass through the optic canal, and the optic chiasma usually lies posterosuperior to the sulcus chiasmatis. Below the sulcus chiasmatis is the tuberculum sellae. The cavernous sinus lies lateral to the hypophysial fossa, and the lateral wall of the body of the sphenoid contains a shallow carotid groove related to the internal carotid artery as it ascends from the carotid canal and runs through the cavernous sinus. Posterolaterally the groove may be deepened by a small projecting lingula. Laterally the middle cranial fossa is deep and supports the temporal lobes of the cerebral hemispheres. Anteriorly are the orbits, laterally the temporal fossae, and inferiorly the infratemporal fossae. The middle cranial fossa communicates with the orbits by the superior orbital fissures, each bounded above by a lesser wing, below by a greater wing, and medially by the body of the sphenoid bone. Each fissure is wider medially, and has a long axis sloping inferomedially and forwards. Many nerves and vessels pass through it, namely, the oculomotor, trochlear and abducens nerves, and the lacrimal, frontal and nasociliary branches of the ophthalmic division of the trigeminal nerve, together with filaments from the internal carotid plexus (sympathetic), the ophthalmic veins, the orbital branch of the middle meningeal artery, and the recurrent branch of the lacrimal artery. Three foramina can be identified in the greater wing of the sphenoid bone. The foramen rotundum is situated just below and behind the medial end of the superior orbital fissure, and leads forwards into the pterygopalatine fossa, to which it conducts the maxillary nerve. Behind the foramen rotundum is the foramen ovale which transmits the mandibular nerve. The foramen spinosum is posterolateral to the foramen ovale and transmits the middle meningeal artery. The latter, with companion veins, ascends lateral to the squamous part of the temporal bone, and turns anterolaterally across the sphenosquamosal suture to the greater wing of the sphenoid bone where it divides into frontal and parietal branches. The frontal branch ascends across the pterion to the anterior part of the parietal bone; at or near the pterion it is often in a bony canal. The parietal branch runs back and up on to the squamous part of the temporal bone, crossing the squamosal suture to gain the parietal bone. These arteries and veins groove the floor and lateral wall of the middle cranial fossa. The foramen ovale and foramen spinosum connect with the underlying infratemporal fossa. page 462 page 463
The foramen lacerum is situated at the posterior end of the carotid groove, posteromedial to the foramen ovale. Its boundaries and contents have already been considered when describing the intermediate part of the cranial base. A small foramen may occur at the root of the greater wing of the sphenoid medial to the foramen lacerum, when present; this emissary sphenoidal foramen transmits a vein from the cavernous sinus.
A shallow trigeminal impression, adapted to the trigeminal ganglion, is situated posterior to the foramen lacerum on the anterior surface of the petrous part of the temporal bone near its apex. Posterolateral to this impression is a shallow pit, limited posteriorly by a rounded arcuate eminence which is produced by the anterior semicircular canal. Lateral to the trigeminal impression a narrow groove passes posterolaterally into the hiatus for the greater petrosal nerve, and even further laterally is the hiatus for the lesser petrosal nerve. The anterior surface of the petrous part of the temporal bone is formed by the tegmen tympani, a thin osseous lamina in the roof of the tympanic cavity, which extends anteromedially above the auditory tube, anterolateral to the arcuate eminence. The posterior part of the tegmen tympani roofs the mastoid antrum, lateral to the eminence. The superior border of the petrous part of the temporal bone separates the middle and the posterior cranial fossae, and is grooved by the superior petrosal sinus. In young skulls, a petrosquamous suture may be visible at the lateral limit of the tegmen tympani but it is obliterated in adults. The tegmen tympani then turns down as the lateral wall of the osseous auditory tube and its lower border may appear in the squamotympanic fissure. Lateral to the anterior part of the tegmen tympani, the squamous part of the temporal bone is thin over a small area that coincides with the deepest part of the mandibular fossa. A smooth trigeminal notch leads into the trigeminal impression and lies on the upper border of the petrous temporal, anteromedial to the groove for the superior petrosal sinus. At this point, the trigeminal nerve separates the sinus from bone. The petrosphenoidal ligament is attached to a tiny bony spicule, directed anteromedially at the anterior end of the trigeminal notch. The abducent nerve bends sharply across the upper petrous border, passing between the ligament and the dorsum sellae anterior to the petrosphenoidal ligament. Posterior cranial fossa
The posterior cranial fossa is the largest and deepest of the cranial fossae. It is bounded in front by the dorsum sellae, posterior aspects of the sphenoidal body and basilar part of occipital bone; behind by the squamous part of the occipital bone; laterally by the petrous and mastoid parts of the temporal bone and by the lateral parts of the occipital bone, and above and behind by the mastoid angles of the parietal bones. The posterior cranial fossa contains the cerebellum, pons and medulla oblongata. The region corresponds extracranially with the posterior part of the cranial base. The most prominent feature in the floor of the posterior cranial fossa is the foramen magnum in the occipital bone. A sloping surface, the clivus, formed successively by the basilar part of the occipital bone, the posterior part of the body and then the dorsum sellae of the sphenoid bone, lies anterior to the foramen magnum. The clivus is gently concave from side to side. On each side it is separated from the petrous part of the temporal bone by a petro-occipital fissure, filled by a thin plate of cartilage and limited behind by the jugular foramen. Its margins are grooved by the inferior petrosal sinus. The spheno-occipital synchondrosis is evident on the clivus of a growing child. A large jugular foramen, sited at the posterior end of the petro-occipital fissure,
lies above and lateral to the foramen magnum. Its upper border is sharp and irregular, and contains a notch for the glossopharyngeal nerve. The cochlear canaliculus, which contains the perilymphatic 'duct' is sited in the deepest part of the notch. The lower border of the jugular foramen is smooth. Posteriorly it is grooved by the sigmoid sinus which continues into the foramen as the internal jugular vein. The accessory, vagus and glossopharyngeal nerves pass through the anterior part of the jugular foramen from behind forwards, and may groove the jugular tubercle as they enter the foramen. The hypoglossal (anterior condylar) canal lies medial to and below the lower border of the jugular foramen at the junction of the basilar and lateral parts of the occipital bone. This canal transmits the hypoglossal nerve (and its recurrent branch), the meningeal branch of the ascending pharyngeal artery and an emissary vein linking the basilar plexus intracranially with the internal jugular vein extracranially. If a posterior condylar canal is present behind the occipital condyle, its internal orifice is posterolateral to that of the hypoglossal canal and contains a sigmoid emissary vein (associated with the occipital veins) and a meningeal branch of the occipital artery. The occipital condyles lie within the anterior aspect of the foramen magnum: their medial aspects are roughened for the attachments of the alar ligaments associated with the atlanto-axial joints. The posterior surface of the petrous part of the temporal bone forms much of the anterolateral wall of the posterior cranial fossa. It contains the internal acoustic meatus, which lies anterosuperior to the jugular foramen, and transmits the facial and vestibulocochlear nerves, the nervus intermedius, and labyrinthine vessels. The mastoid part of the temporal bone lies behind the petrous part of the temporal bone in the lateral wall of the posterior cranial fossa. Anteriorly it is grooved by a wide sigmoid sulcus (groove) running forwards and downwards, then downwards and medially, and finally forwards to the jugular foramen. It contains the sigmoid sinus. Superiorly, where the groove touches the mastoid angle of the parietal bone, it is continuous with a groove transmitting the transverse sinus; it next crosses the parietomastoid suture, and then descends behind the mastoid antrum. A mastoid foramen for an emissary vein from the sigmoid sinus and a meningeal branch of the occipital artery, sometimes large enough to groove the squamous part of the occipital bone, may be sited here. The lowest part of the sigmoid sulcus crosses the occipitomastoid suture and grooves the jugular process of the occipital bone. The right sigmoid sulcus is usually larger than the left. A thin plate with an irregularly curved margin projects back behind the internal acoustic meatus and bounds a slit containing the opening of the vestibular aqueduct (which contains the saccus and ductus endolymphaticus and a small artery and vein). A small subarcuate fossa lies between the internal acoustic meatus and the aqueductal opening. It contains dura mater. Near the superior border of the petrous part of the temporal bone it is pierced by a small vein. In infants the fossa is a relatively large blind tunnel under the anterior semicircular canal. The squamous part of the occipital bone displays a median internal occipital
crest, which runs posteriorly from the foramen magnum to an internal occipital protuberance and gives attachment to the falx cerebelli. The internal occipital crest may be grooved by the occipital sinus. The internal occipital protuberance is close to the confluence of the sinuses and is grooved bilaterally by the transverse sinuses. The latter curve laterally with an upward convexity to the mastoid angles of the parietal bones. The groove for the transverse sinus is usually deeper on the right, where it is generally a continuation of the superior sagittal sinus, while on the left it is frequently a continuation of the straight sinus. On both sides the transverse sulcus is continuous with the sigmoid sulcus. Below the transverse sulcus the internal occipital crest separates two shallow fossae, adapted to the cerebellar hemispheres. The margins of the grooves for the transverse sinus and superior petrosal sinus, together with the posterior clinoid process, all provide anchorage for the attached margin of the tentorium cerebelli.
© 2008 Elsevier
DISARTICULATED INDIVIDUAL BONES OCCIPITAL BONE (Fig. 27.11) The occipital bone forms much of the back and base of the cranium. It is trapezoid, internally concave and encloses the foramen magnum. It has four parts, namely, basilar (basioccipital), which is the quadrilateral part in front of the foramen magnum; squamous, which is the expanded plate posterosuperior to the foramen; and lateral (condylar or exoccipital), on each side of the foramen magnum. The foramen magnum is situated in an anteromedian position, and is oval, being wider behind and with its greatest diameter being anteroposterior. Squamous part
The squamous part is convex externally and concave internally. On the external surface the external occipital protuberance lies midway between its summit and the foramen magnum. On each side, two curved lines extend laterally from this protuberance. The upper, faintly marked and often almost imperceptible, is the highest nuchal line, to the medial part of which the epicranial aponeurosis is attached, and the lower is the superior nuchal line. The lateral part of the highest nuchal line gives attachment to the occipital part of occipitofrontalis. The median external occipital crest, often faint, descends from the external occipital protuberance to the foramen magnum. On each side an inferior nuchal line spreads laterally from the midpoint of the crest. page 463 page 464
Figure 27.11 Occipital bone. A, External surface; B, internal surface; C, muscle attachments. (By permission from Berkovitz and Moxham, 1994.)
The internal surface of the squamous part is divided into four deep fossae by an irregular internal occipital protuberance and by ridged sagittal and horizontal extensions from it. The two superior fossae are triangular and adapted to the occipital poles of the cerebral hemispheres; the inferior fossae are quadrilateral and shaped to accommodate the cerebellar hemispheres. A wide groove with raised banks, the superior sagittal sulcus, ascends from the protuberance to the superior angle of the squamous part. The posterior part of the falx cerebri is attached to the margins of the sulcus. A prominent internal occipital crest descends from the protuberance and bifurcates near the foramen magnum, providing an attachment for the falx cerebelli. The occipital sinus, sometimes double, lies in this attachment. A small vermian fossa may exist, at the lower end of the internal occipital crest; when present, it is occupied by part of the inferior cerebellar vermis. On each side a wide sulcus for the transverse sinus extends laterally from the internal occipital protuberance. The tentorium cerebelli is attached to the margins of these sulci. The right sulcus is usually larger, passing into the sulcus for the superior sagittal sinus, the left being a continuation of the straight sinus. The position of this confluence of sinuses is indicated by a depression on one side of the protuberance. The position of the fetal posterior fontanelle coincides with the junction between the superior angle of the squamous part of the occipital bone and the occipital angles of the parietal bones. The lateral angles of the squamous part are marked internally by the ends of the transverse sulci and project between the parietal and temporal bones. The lambdoid borders extend from superior to lateral angles and are serrated for articulation with the occipital borders of the parietal bones at the lambdoid suture. The mastoid borders extend from the lateral angles to the jugular processes, articulating with the mastoid parts of the temporal bones. A variety of sutural bones (ossicles) may occur at or near the lambda, e.g. the 'interparietal' (Inca bone or ossicle of Goethe). Basilar part
The basilar part extends anterosuperiorly from the foramen magnum, fusing with the sphenoid bone in adults. In young skulls a rough and uneven surface is joined to the body of the sphenoid by a growth cartilage (spheno-occipital synchondrosis). By the twenty-fifth year, this plate has fully ossified and the occipital and sphenoid bones are fused. The inferior surface of the basilar part bears a small pharyngeal tubercle for attachment of the fibrous pharyngeal raphe c.1 cm in front of the foramen magnum. Longus capitis is attached anterolateral to the tubercle, and rectus capitis anterior is attached to a small depression immediately anterior to the occipital condyle. This depression may occasionally be replaced by a small precondylar tubercle. The anterior atlanto-occipital membrane is attached to the anterior margin of the foramen magnum. The superior surface of the basilar part is a broad groove and forms part of the clivus that ascends anteriorly from the foramen magnum, and on which rest the medulla oblongata and lower pons. Sulci of the inferior petrosal sinuses are on its lateral margins, which articulate below with the petrous part of the temporal bones. page 464 page 465
Lateral (condylar) parts
The lateral (condylar) parts of the occipital bone flank the foramen magnum. On their inferior surfaces are occipital condyles for articulation with the superior articular facets of the atlas vertebra. The condyles are oval or reniform, their long axes converging anteromedially. The articular surfaces, wholly convex, face inferolaterally. They are occasionally constricted and a condyle may be in two parts (as may be the reciprocal surfaces of the atlas vertebra). A tubercle gives attachment to an alar ligament medial to each articular facet. The hypoglossal (anterior condylar) canal, which is situated anteriorly above each condyle, starts internally a little above the anterolateral part of the foramen magnum and continues anterolaterally. It may be partly or wholly divided by a spicule of bone and transmits the hypoglossal nerve and a meningeal branch of the ascending pharyngeal artery. A condylar fossa, behind each condyle, fits the posterior margin of the superior facet of the atlas vertebra in full extension of the skull. Its floor is sometimes perforated by a posterior condylar canal for a sigmoid emissary vein. A quadrilateral plate, the jugular process, projects laterally from the posterior half of each condyle, and contributes the posterior part of the jugular foramen. The inferior surface of the jugular process is roughened by the attachment of rectus capitis lateralis. The jugular process is indented in front by a jugular notch,
which is sometimes partly divided by a small intrajugular process that projects anterolaterally. A paramastoid process sometimes projects down and may even articulate with the transverse process of the atlas vertebra. Laterally, the jugular process has a rough quadrilateral or triangular area that is joined to the jugular surface of the temporal bone by cartilage: it begins to ossify at c.25 years. An oval jugular tubercle overlies the hypoglossal canal on the superior surface of the occipital condyle. Its posterior part often bears a shallow furrow for the glossopharyngeal, vagus and accessory nerves. A deep groove containing the end of the sigmoid sinus curves anteromedially around a hook-shaped process to end at the jugular notch. The posterior condylar canal opens into the posterior cranial fossa near the medial end of the groove. UPDATE Date Added: 07 December 2005 Publication Services, Inc. Update: Infratemporal fossa approach to the hypoglossal canal. Tumors of hypoglossal nerves are uncommon. They account for only 2% of all skull base tumors. The hypoglossal nerve can be affected by glomus tumors and skull base meningiomas. For surgical treatment of these tumors, previous studies described several approaches to the hypoglossal canal including the extended posterolateral approach, the supracondylar approach, and the far-lateral transcondylar transtubercular approach. However, to reach the canal, the procedure of choice for the surgeon is the infratemporal fossa approach, since it provides a good exposure to the jugular foramen and access to the hypoglossal canal and nerves. In order to reach the hypoglossal canal, the surgeon needs guidelines based on the specific anatomy of the canal. In a recent study (published in 2004), Hadley and Shelton demonstrated that the hypoglossal canal could be reached safely and consistently by way of the temporal bone with preservation of hearing and cranial nerves (CN) IX to XI. In this prospective anatomic study, 15 cadaver temporal bones were dissected using infratemporal fossa Fisch type-A. After exposure of the hypoglossal canal, the distance from the canal to the jugular bulb, carotid artery, round window, lateral canal, and roots of CN IX to XI were recorded. The position of the hypoglossal canal was consistently located anterior, inferior, and medial to the jugular bulb. The distance from midcanal to the jugular bulb and the roots of CN IX to XI at the posterior fossa dura was 5.3 mm ± 0.82 and 7.1 mm ± 2.49, respectively. The distance from the carotid artery where it meets the jugular vein to the midcanal was 15.3 mm ± 2.09. The distance from the round window to the canal was 21.7mm ± 3.17. In their study, Hadley and Shelton demonstrated that the infratemporal fossa approach allows the surgeon to preserve the carotid artery, cochlea, and lower CN while exposing the entire canal. The distance from the jugular bulb and roots of CN IX to XI can be used as guideposts. If a tumor involves the bulb, the carotid artery and the round window are the next reliable indicators of position. The canal can be reached by identifying the anterior-inferior edge of the jugular bulb and dissecting anterior, inferor, and slightly medial for approximately 5 mm through soft medullary bone. The 5-mm measurement is approximately the same distance as the width of a standard Sheehy weapon knife (Storz, Tuttlingen, Germany), providing the surgeon with an easy way to make this measurement intraoperatively. This distance assumes that a lesion such as the glomus tumor has not altered the jugular bulb. The hypoglossal canal can be found approximately 22 mm away by using the round window as a reference point. In the case that the carotid artery has been exposed, the point where it joins the jugular vein to the hypoglossal canal is at approximately 15 mm. The condylar emissary vein can be encountered when dissecting posterior to the hypoglossal canal and, when inadvertently injured, it can be a source of significant bleeding in an area of critical adjacent structures. Hadley and Shelton provide practical guidance for surgeons approaching the hypoglossal canal during surgical procedures to treat unusual skull base tumors and related problems. Hadley KS, Shelton C: Infratemporal fossa approach to the hypoglossal canal: practical landmarks for elusive anatomy. Laryngoscope 114:1648-1651, 2004. Medline Similar articles
UPDATE Date Added: 30 November 2005 Publication Services, Inc. Update: Infratemporal fossa approach to the hypoglossal canal. Tumors of hypoglossal nerves are uncommon. They account for only 2% of all skull base tumors. The hypoglossal nerve can be affected by glomus tumors and skull base meningiomas. For surgical treatment of these tumors, previous studies described several approaches to the hypoglossal canal including the extended
posterolateral approach, the supracondylar approach, and the far-lateral transcondylar transtubercular approach. However, to reach the canal, the procedure of choice for the surgeon is the infratemporal fossa approach, since it provides a good exposure to the jugular foramen and access to the hypoglossal canal and nerves. In order to reach the hypoglossal canal, the surgeon needs guidelines based on the specific anatomy of the canal. In a recent study (published in 2004), Hadley and Shelton demonstrated that the hypoglossal canal could be reached safely and consistently by way of the temporal bone with preservation of hearing and cranial nerves (CN) IX to XI. In this prospective anatomic study, 15 cadaver temporal bones were dissected using infratemporal fossa Fisch type-A. After exposure of the hypoglossal canal, the distance from the canal to the jugular bulb, carotid artery, round window, lateral canal, and roots of CN IX to XI were recorded. The position of the hypoglossal canal was consistently located anterior, inferior, and medial to the jugular bulb. The distance from midcanal to the jugular bulb and the roots of CN IX to XI at the posterior fossa dura was 5.3 mm ± 0.82 and 7.1 mm ± 2.49, respectively. The distance from the carotid artery where it meets the jugular vein to the midcanal was 15.3 mm ± 2.09. The distance from the round window to the canal was 21.7mm ± 3.17. In their study, Hadley and Shelton demonstrated that the infratemporal fossa approach allows the surgeon to preserve the carotid artery, cochlea, and lower CN while exposing the entire canal. The distance from the jugular bulb and roots of CN IX to XI can be used as guideposts. If a tumor involves the bulb, the carotid artery and the round window are the next reliable indicators of position. The canal can be reached by identifying the anterior-inferior edge of the jugular bulb and dissecting anterior, inferor, and slightly medial for approximately 5 mm through soft medullary bone. The 5 mm measurement is approximately the same distance as the width of a standard Sheehy weapon knife (Storz, Tuttlingen, Germany), providing the surgeon with an easy way to make this measurement intraoperatively. This distance assumes that a lesion such as the glomus tumor has not altered the jugular bulb. The hypoglossal canal can be found approximately 22 mm away by using the round window as a reference point. In the case that the carotid artery has been exposed, the point were it joins the jugular vein to the hypoglossal is at approximately 15 mm. The condylar emissary vein can be encountered when dissecting posterior to the hypoglossal canal and, when inadvertently injured, it can be a source of significant bleeding in an area of critical adjacent structures. Hadley and Shelton provide practical guidance for surgeons approaching the hypoglossal canal during surgical procedures to treat unusual skull base tumors and related problems. Hadley KS, Shelton C: Infratemporal fossa approach to the hypoglossal canal: practical landmarks for elusive anatomy. Laryngoscope 114:1648-1651, 2004. Medline Similar articles
Ossification
Above the highest nuchal lines the squamous part of the occipital bone is developed in a fibrous membrane and is ossified from two centres (one on each side) from about the second fetal month. This part of the occipital bone may remain separate as the interparietal bone. The remainder of the occipital bone is preformed in cartilage. Below the highest nuchal lines, the squamous part ossifies from two centres, appearing in about the seventh week and soon uniting. The two components of the squamous part unite in the third postnatal month, but the line of union is recognizable at birth. The remainder of the cartilage of the occipital is ossified from five centres, two each for the lateral parts during the eighth week and one for the basilar part commencing around the sixth week. At birth the occipital bone consists of four separate parts (Fig. 27.12), a basilar part, two lateral parts and a squamous part, all joined by cartilage and forming a ring around the foramen magnum. The squamous and lateral parts fuse together from the second year. The lateral parts fuse with the basilar part during years 3 and 4, but fusion may be delayed until the 7th year.
Figure 27.12 The occipital bone of a newborn child: external surface. Parts of the chondrocranium which are still unossified are shown in blue.
SPHENOID BONE (Fig. 27.13) The sphenoid bone lies in the base of the skull between the frontal, temporal and occipital bones. It has a central body, paired greater and lesser wings that spread laterally from the body, and two pterygoid processes that descend from the junction of the body and greater wings. The body contains the sphenoidal air sinuses while immediately above it is a depression which contains the hypophysis cerebri (pituitary gland). Body
The body is cuboidal and contains two sphenoidal air sinuses, separated by a septum. The cerebral or superior surface articulates in front with the cribriform plate of the ethmoid bone. Anteriorly is the smooth jugum sphenoidale, related to gyri recti and olfactory tracts. The jugum is bounded behind by the anterior border of the sulcus chiasmatis that leads laterally to the optic canals. Posteriorly is the tuberculum sellae, behind which is the deeply concave sella turcica. In life the sella turcica contains the hypophysis cerebri in the hypophysial fossa. The anterior edge of the sella turcica is completed laterally by two middle clinoid processes. Posteriorly the sella turcica is bounded by a square dorsum sellae, the superior angles of which bear variable posterior clinoid processes. The clinoid process is related to the attachment of the diaphragma sella and the tentorium cerebelli. On each side, below the dorsum sellae, a small petrosal process articulates with the apex of the petrous part of the temporal bone. The body of the sphenoid slopes directly into the basilar part of the occipital bone posterior to the dorsum sellae, and together these bones form the clivus. In the growing child this is the site of the spheno-occipital synchondrosis. UPDATE Date Added: 22 May 2006 Abstract: Surgical anatomy of the anterior clinoid process Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=14975418&query_hl=20&itool=pubmed_docsum Surgical anatomy of the anterior clinoid process. Huynh-Le P, Natori Y, Sasaki T: J Clin Neurosci 11:283-287, 2004. The lateral surfaces of the body are united with the greater wings and the medial pterygoid plates. A broad carotid sulcus accommodates the internal carotid artery and a series of cranial nerves associated with the cavernous sinus above the root of each wing. The sulcus is deepest posteriorly, overhung medially by the petrosal part of the temporal and has a sharp lateral margin, the lingula. The lingula continues back over the posterior opening of the pterygoid canal. A median triangular, bilaminar sphenoidal crest on the anterior surface of the body of the sphenoid makes a small contribution to the nasal septum. The anterior border of the crest joins the perpendicular plate of the ethmoid bone, and a sphenoidal sinus opens on each side of it. The sphenoidal sinuses, which are two large, irregular cavities within the body, are usually separated by an asymmetrical septum. Each sinus varies in form and size and is partially divided by bony laminae. A lateral recess may extend into the greater wing and lingula and may even invade the basilar part of the occipital bone almost to the foramen magnum. The morphology of the sphenoidal sinus is of clinical importance in relation to the trans-sphenoidal surgical approach to the hypophysis cerebri.
Sphenoidal sinuses may be classified according to size into three main types: sellar, the commonest type, where the sinus extends for a variable distance beyond the tuberculum sellae; presellar, where the sinus occasionally extends posteriorly towards, but not beyond, the tuberculum sellae; conchal, the rarest type, where a small sinus is separated from the sella turcica by c.10 mm of trabecular bone. In the articulated state the sphenoidal sinuses are closed anteroinferiorly by the sphenoidal conchae, which are largely destroyed when disarticulating a skull. Each half of the anterior surface of the body of the sphenoid possesses a superolateral depressed area joined to the ethmoid labyrinth, which completes the posterior ethmoidal sinuses; a lateral margin which articulates with the orbital plate of the ethmoid above and the orbital process of the palatine bone below; and an inferomedial, smooth, triangular area, which forms the posterior nasal roof and near whose superior angle is the orifice of a sphenoidal sinus. The inferior surface of the body of the sphenoid bears a median triangular sphenoidal rostrum, embraced above by the diverging lower margins of the sphenoidal crest. The narrow anterior end of the rostrum fits into a fissure between the anterior parts of the alae of the vomer, and posterior ends of the sphenoidal conchae flank the rostrum, articulating with its alae. A thin vaginal process projects medially from the base of the medial pterygoid plate on each side of the posterior part of the rostrum, behind the apex of the sphenoidal concha. Greater wings page 465 page 466
Figure 27.13 Sphenoid bone. A, Anterior view; B, posterior view; C, superior view; D, inferior view. (By permission from Berkovitz and Moxham, 1994.)
The greater wings of the sphenoid bone curve broadly superolaterally from the body. Posteriorly each is triangular, fitting the angle between the petrous and squamous parts of the temporal bone at a sphenosquamosal suture. The cerebral surface contributes to the anterior part of the middle cranial fossa. Deeply concave, its undulating surface is adapted to the anterior gyri of the temporal lobe of the cerebral hemisphere. The foramen rotundum for the maxillary nerve lies anteromedially. Posterolateral to the foramen rotundum is the foramen ovale, which transmits the mandibular nerve, accessory meningeal artery and sometimes the lesser petrosal nerve, although the latter nerve may have its own canaliculus innominatus medial to the foramen spinosum. A small emissary sphenoidal foramen which transmits a small vein from the cavernous sinus is present medial to the foramen ovale (on one or both sides) in c.40% of skulls. Behind the foramen ovale is the foramen spinosum, which transmits the middle meningeal artery and meningeal branch of the mandibular nerve. The lateral surface is vertically convex and divided by a transverse infratemporal crest into temporal (upper) and infratemporal (lower) surfaces. Temporalis is attached to the temporal surface. The infratemporal surface is directed downwards and, with the infratemporal crest, is the site of attachment of the upper fibres of lateral pterygoid. It contains the foramen ovale and foramen spinosum. The small downward projecting spine of the sphenoid lies posterior to the foramen spinosum. The sphenomandibular ligament, a remnant of the first branchial arch cartilage, is attached to the tip of the spine of the sphenoid. The medial side of the spine has a faint anteroinferior groove for the chorda tympani nerve and appears in the lateral wall of the sulcus for the auditory tube. A ridge which forms a posterior boundary of the pterygomaxillary fissure descends to the front of the lateral pterygoid plate medial to the anterior end of the infratemporal crest.
The quadrilateral orbital surface of the greater wing faces anteromedially, and forms the posterior part of the lateral wall of the orbit. It has a serrated upper edge which articulates with the orbital plate of the frontal bone, and a serrated lateral margin which articulates with the zygomatic bone. Its smooth inferior border is the posterolateral edge of the inferior orbital fissure, and its sharp medial margin forms the inferolateral edge of the superior orbital fissure, on which a small tubercle gives partial attachment to the common annular ocular tendon. Below the medial end of the superior orbital fissure a grooved area forms the posterior wall of the pterygopalatine fossa, which is pierced by the foramen rotundum. page 466 page 467
The irregular margin of the greater wing, from the body of the sphenoid to the spine, is an anterior limit of the foramen lacerum, in its medial half. It also displays the posterior aperture of the pterygoid canal. Its lateral half articulates with the petrous part of the temporal bone at a sphenopetrosal synchondrosis. Inferior to this the sulcus tubae contains the cartilaginous auditory tube. Anterior to the spine of the sphenoid the concave squamosal margin is serrated - bevelled internally below, externally above - for articulation with the squamous part of the temporal bone. The tip of the greater wing, bevelled internally, articulates with the sphenoidal angle of the parietal bone at the pterion. Medial to this, a triangular rough area articulates with the frontal bone: its medial angle is continuous with the inferior boundary of the superior orbital fissure, and its anterior angle joins the zygomatic bone by a serrated articulation. Lesser wings
The lesser wings are triangular, pointed plates that protrude laterally from the anterosuperior regions of the body. The superior surface of each wing is smooth and related to the frontal lobe of the cerebral hemisphere. The inferior surface is a posterior part of the orbital roof and upper boundary of the superior orbital fissure, and overhangs the middle cranial fossa. The posterior border projects into the lateral fissure of the cerebral hemisphere. The medial end of the lesser wing forms the anterior clinoid process. The anterior and middle clinoid processes are sometimes united to form a caroticoclinoid foramen. The lesser wing is connected to the body by a thin flat anterior root and a thick triangular posterior root, between which lies the optic canal. Growth of the posterior root is closely associated with variations in the canal. The cranial opening of the canal may be duplicated, or more commonly, the division is incomplete. Superior orbital fissure
The superior orbital fissure connects the cranial cavity with the orbit. It is bounded medially by the body of the sphenoid, above by the lesser wing of the sphenoid, below by the medial margin of the orbital surface of the greater wing, and laterally, between greater and lesser wings, by the frontal bone. Pterygoid processes
The pterygoid processes descend perpendicularly from the junctions of the greater wings and body. Each consists of a medial and lateral plate, whose upper parts are fused anteriorly. The plates are separated below by the angular pterygoid fissure, whose margins articulate with the pyramidal process of the palatine bone. They diverge behind, and medial pterygoid and tensor veli palatini lie in the cuneiform pterygoid fossa between them. Above is a small, oval, shallow scaphoid fossa, formed by division of the upper posterior border of the medial plate. Part of tensor veli palatini is attached to the fossa. The anterior surface of the root of the pterygoid process is broad and triangular. It forms the posterior wall of the pterygopalatine fossa which is pierced by the anterior opening of the pterygoid canal. Lateral pterygoid plate
The lateral pterygoid plate is broad, thin and everted. Its lateral surface forms part of the medial wall of the infratemporal fossa and gives origin to the lower part of lateral pterygoid. Its medial surface is the lateral wall of the pterygoid fossa, and most of the deep head of medial pterygoid is attached to it. The upper part of its anterior border is a posterior boundary of the pterygomaxillary fissure and the lower part articulates with the palatine bone. Its posterior border is free. Medial pterygoid plate
The medial pterygoid plate is narrower and longer than the lateral. Its lower end curves to the lateral, unciform pterygoid hamulus. The tendon of tensor veli palatini winds around the hamulus, and the pterygomandibular raphe is attached
to it. The lateral surface is the medial wall of the pterygoid fossa. The medial surface is a lateral boundary of the posterior nasal aperture. The medial plate is prolonged above on the inferior aspect of the body of the sphenoid as a thin vaginal process, which articulates anteriorly with the sphenoidal process of the palatine bone and medially with the ala of the vomer. Inferiorly it bears a furrow, which is converted into the palatovaginal canal anteriorly by the sphenoidal process of the palatine bone. This canal transmits pharyngeal branches of the maxillary artery and pterygopalatine ganglion. The pharyngobasilar fascia is attached to the whole of the posterior margin of the medial plate, and the superior pharyngeal constrictor is attached to its lower end. At its upper end, just below the posterior opening of the pterygoid canal, is a small pterygoid tubercle. The processus tubarius, which supports the cartilaginous pharyngeal end of the pharyngotympanic tube, projects back near the midpoint of the margin of the medial pterygoid plate. The plate articulates with the posterior border of the perpendicular plate of the palatine bone in the lower part of its anterior margin. Sphenoidal conchae
The sphenoidal conchae are two thin, curved platelets, attached anteroinferiorly to the body of the sphenoid bone. The superior concave surface of each forms the anterior wall and part of the floor of a sphenoidal sinus. They are largely destroyed in disarticulating a skull. In situ, each has vertical quadrilateral anterior and horizontal triangular posterior parts. The anterior part consists of a superolateral depressed area, which completes the posterior ethmoidal sinuses and joins below with the orbital process of a palatine bone; and a smooth and triangular inferomedial area, which forms part of the nasal roof and is perforated above by the round opening connecting the sphenoidal sinus and sphenoethmoidal recess. Anterior parts of the two bones meet in the midline, and protrude as the sphenoidal crest. The horizontal part appears in the nasal roof and completes the sphenopalatine foramen. Its medial edge articulates with the rostrum of the sphenoid and the ala of the vomer. Its apex, directed posteriorly, is superomedial to the vaginal process of the medial pterygoid plate and joins the posterior part of the ala. A small conchal part sometimes appears in the medial wall of the orbit, lying between the orbital plate of the ethmoid in front, the orbital process of the palatine bone below and the frontal bone above. Ossification
Until the seventh or eighth month in utero the sphenoid body has a presphenoidal part, anterior to the tuberculum sellae, with which the lesser wings are continuous, and a postsphenoidal part, comprising the sella turcica and dorsum sellae, and integral with the greater wings and pterygoid processes. Much of the bone is preformed in cartilage. There are six ossification centres for the presphenoidal, and eight for postsphenoidal, parts. Presphenoidal part About the ninth week of fetal life, a centre appears in each wing, lateral to the optic canal, and a little later two bilateral centres appear in the presphenoidal body. Each sphenoidal concha has a centre, appearing superoposteriorly in the nasal capsule in the fifth month in utero. As this enlarges it partly surrounds a posterosuperior expansion of the nasal cavity, which becomes the sphenoidal sinus. The posterior conchal wall is absorbed and the sinus invades the presphenoid component. In the fourth year the concha fuses with the ethmoidal labyrinth and before puberty it fuses with the sphenoid and palatine bones. Its anterior deficiency persists as an opening for the sphenoidal sinus. Postsphenoidal part The first centres appear in the greater wings about the eighth week of fetal life, one in the basal cartilage of each wing below the foramen rotundum. These centres only contribute to the root of the greater wing (near the foramen rotundum and pterygoid canal). The remainder of the greater wing is ossified in mesenchyme, spreading also into the lateral pterygoid plate. About the fourth month of fetal life two centres appear, flanking the sella turcica, and soon fuse. The medial pterygoid plates are also ossified in 'membrane', a centre in each probably appearing about the ninth or tenth week. The hamulus is chondrified during the third fetal month and at once begins to ossify. Medial and lateral pterygoid plates join about the sixth fetal month. During the fourth month, a centre appears for each lingula, soon joining the body. The optic canal in the neonate is relatively large and has a keyhole or 'figure of eight' shape rather than the circular profile seen in the adult. Postnatal details
page 467
page 468
Presphenoidal and postsphenoidal parts fuse about the eighth month in utero, but an unciform cartilage persists after birth in lower parts of the junction. At birth the bone is tripartite (Fig. 27.13) and consists of a central part (body and lesser wings) and two lateral parts (each comprising a greater wing and pterygoid process). During the first year the greater wings and body unite around the pterygoid canals and the lesser wings extend medially above the anterior part of the body, meeting to form the smooth, elevated jugum sphenoidale. By the twenty-fifth year, sphenoid and occipital bones are completely fused. An occasional vascular foramen, often erroneously termed the craniopharyngeal canal, is occasionally seen in the anterior part of the hypophysial fossa. Although the sphenoidal sinus can be identified in the fourth month of fetal life as an evagination of the posterior part of the nasal capsule, by birth it represents an outgrowth of the sphenoethmoidal recess. Pneumatization of the body of the sphenoid bone commences in the second or third year and spreads first into the presphenoid, and later invades the postsphenoid, part. It reaches full size in adolescence, but often enlarges further by absorption of its walls as age advances. Certain sphenoidal parts are connected by ligaments that occasionally ossify, e.g. the pterygospinous, between the sphenoid spine and upper part of lateral pterygoid plate; the interclinoid, joining the anterior to the posterior clinoid process; and the caroticoclinoid, connecting the anterior to the middle clinoid process. Premature synostosis of the junction between pre-and postsphenoidal parts, or of the spheno-occipital suture, produces a characteristic appearance, obvious in profile, of an abnormal depression of the nasal bridge (hypertelorism).
TEMPORAL BONE(Fig. 27.14) UPDATE Date Added: 01 December 2004 Abstract: Temporal bone fractures: traditional classification and clinical relevance. Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=15454763 Temporal bone fractures: traditional classification and clinical relevance. Each temporal bone consists of four components: the squamous, petromastoid and tympanic parts and the styloid process. The squamous part has a shallow mandibular fossa associated with the temporomandibular joint. The petromastoid part is relatively large. Its petrous portion houses the auditory apparatus and is formed of compact bone. In contrast, the mastoid process is trabecular and variably pneumatized. The tympanic part has the form of a thin and incomplete ring whose ends are fused with the squamous part. The styloid process gives attachment to the styloid group of muscles. Two canals are associated with the temporal bone. The external acoustic meatus, visible on the lateral surface, conveys sound waves to the tympanic membrane. The internal acoustic meatus, evident on the medial surface, conveys the facial and vestibulocochlear nerves. Squamous part
The squamous part lies anterosuperiorly and is thin and partly translucent. Its external temporal surface is smooth, slightly convex, and forms part of the temporal fossa for attachment of temporalis. Above the external acoustic meatus, it is grooved vertically by the middle temporal artery. The supramastoid crest curves backwards and upwards across its posterior part and gives attachment to the temporal fascia. The junction between the squamous and mastoid parts is c.1.5 cm below this crest and traces of the squamomastoid suture may persist. The suprameatal triangle, a depression marking the position of the mastoid antrum (medial to it at a depth of c.1.25 cm), lies between the anterior end of the supramastoid crest and the posterosuperior quadrant of the external acoustic meatus. The triangle usually contains a small suprameatal spine anteriorly. The internal cerebral surface of the squamous part is concave and contains depressions corresponding to convolutions of the temporal lobe of the cerebral hemisphere. This surface is grooved by the middle meningeal vessels. Its lower border is fused to the anterior region of the petrous part, but traces of a petrosquamosal suture often appear in adult bones. The superior border is thin, bevelled internally and overlaps the inferior border of the parietal bone at the squamosal suture. Posteriorly it forms an angle with the mastoid element. The anteroinferior border, thin above and thick below, meets the greater wing of the sphenoid bone: above it is bevelled internally, below it is bevelled externally.
The squamous part has a zygomatic process and a mandibular fossa. Zygomatic process
The zygomatic process juts forwards from the lower region of the squamous part. Its triangular posterior part has a broad base directed laterally, presenting superior and inferior surfaces. The zygomatic process then twists anteromedially, so that its surfaces become medial and lateral. The superior surface of the posterior part is concave. The inferior surface is bounded by anterior and posterior roots, converging into the anterior part of the process. At the junction of the roots the tubercle of the zygomatic root gives attachment to the lateral temporomandibular ligament. The posterior root is prolonged forwards above the external acoustic meatus, its upper border continues into the supramastoid crest. Very rarely the squamous part is perforated above the posterior root by a squamosal foramen, transmitting the petrosquamous sinus. The anterior root juts almost horizontally from the squamous part. Its inferior surface, with an anteroposterior convexity, forms a short semicylindrical articular tubercle and comes in contact with the articular disc of the temporomandibular joint. The tubercle forms the anterior limit of the mandibular fossa. The anterior part of the zygomatic process is thin and flat and the temporal fascia is attached to its superior border. The inferior border is short and arched and gives origin to some fibres of masseter. The lateral surface is convex. The medial surface is concave and provides further attachment for part of masseter. The anterior end is deeply serrated and slopes obliquely posteroinferiorly to articulate with the temporal process of the zygomatic bone, forming the zygomatic arch. Anterior to the articular tubercle, a small triangular area forms part of the roof of the infratemporal fossa. It is continuous behind with the anterior root and in front with the infratemporal crest of the greater wing of the sphenoid. Mandibular fossa
The mandibular fossa is limited in front by the articular eminence of the zygomatic process. It presents an anterior articular area, formed by the squamous part, and a posterior non-articular area, formed by the tympanic element. The articular surface is smooth, oval and concave and contacts the articular disc of the temporomandibular joint. Unlike most other synovial joints, it is lined by fibrous tissue rather than hyaline cartilage, reflecting its intramembranous development. The non-articular area sometimes contains part of the parotid gland. A small, conical postglenoid tubercle separates the articular surface laterally from the tympanic plate. Posteriorly, the mandibular fossa is separated from the tympanic part by the squamotympanic fissure. Rarely, a postglenoid foramen exists anterior to the external acoustic meatus in the line of fusion of the squamous and tympanic parts. It then replaces the squamosal foramen noted above and transmits the petrosquamous sinus. Medially, a projection from the petrous part of the temporal bone (tegmen tympani) comes to lie within the squamotympanic fissure, further dividing it into petrotympanic and petrosquamous fissures. The petrotympanic fissure leads into the tympanic cavity and contains an anterior malleolar ligament and the anterior tympanic branch of the maxillary artery. At the medial end of the fissure is the anterior opening of the anterior canaliculus for the chorda tympani nerve. Petromastoid part
The petromastoid part of the temporal bone, although morphologically one element, is more conveniently described as two parts, the mastoid and petrous parts. Mastoid part
This is the posterior region of the temporal bone. It has an outer surface roughened by attachments of the occipital belly of occipitofrontalis and auricularis posterior. A mastoid foramen, of variable size and position, and traversed by a vein from the sigmoid sinus and a small dural branch of the occipital artery, frequently lies near its posterior border. The foramen may be in the occipital or occipitotemporal suture; parasutural (40-50% of crania); or may be absent. The mastoid part projects down as the conical mastoid process, and is larger in adult males. Sternocleidomastoid, splenius capitis and longissimus capitis are attached to its lateral surface. There is a deep mastoid notch on the medial aspect to which the posterior belly of digastric is attached. The occipital artery runs in a shallow occipital groove medial to this notch.
The internal surface of the mastoid process bears a deep, curved sigmoid sulcus for the sigmoid venous sinus. The sulcus is separated from the underlying innermost mastoid air cells by a thin lamina of bone. The air cells and mastoid antrum are described in Chapter 38. The superior border of the mastoid part is thick and serrated for articulation with the mastoid angle of the parietal bone. The posterior border is also serrated and articulates with the inferior border of the occipital bone between its lateral angle and jugular process. The mastoid element is fused with the descending process of the squamous part: below, it appears in the posterior wall of the tympanic cavity. Petrous part
This mass of bone is wedged between the sphenoid and occipital bones in the cranial base, and is inclined superiorly and anteromedially. It has a base, apex, three surfaces (anterior, posterior and inferior) and three borders (superior, posterior and anterior), and contains the acoustic labyrinth. page 468 page 469
Figure 27.14 Temporal bone. A, External view; B, posterior/oblique view; C, inferior view; D, internal view. (By permission from Berkovitz and Moxham, 1994.)
The base would correspond to the part that lies on the base of the skull and is separated from the squamous part by a suture. However, this suture disappears soon after birth. The subsequent development of the mastoid processes means that the precise boundaries of the base are no longer identifiable. The apex, blunt and irregular, is angled between the posterior border of the greater wing of the sphenoid and the basilar part of the occipital bone. It contains the anterior opening of the carotid canal and limits the foramen lacerum posterolaterally. page 469 page 470
The anterior surface contributes to the floor of the middle cranial fossa and is continuous with the cerebral surface of the squamous part (although the petrosquamosal suture often persists late in life). The whole surface is adapted to the inferior temporal gyri. Behind the apex is a trigeminal impression for the trigeminal ganglion. Bone anterolateral to this roofs the anterior part of the carotid canal, but is often deficient. A ridge separates the trigeminal impression from another hollow behind, which partly roofs the internal acoustic meatus and cochlea. This in turn is limited behind by the arcuate eminence raised by the anterior semicircular canal. Laterally, it roofs the vestibule and, partly, the facial canal. Between the squamous part laterally and the arcuate eminence and the hollows just described medially, the anterior surface is formed by the tegmen tympani. This thin plate of bone forms the roof of the mastoid antrum, extending forwards above the tympanic cavity and the canal for tensor tympani. Its lateral margin meets the squamous part at the petrosquamosal suture, turning down in front as the lateral wall of the canal for the tensor tympani and the osseous part of the pharyngotympanic tube: its lower edge is in the squamotympanic fissure. Anteriorly the tegmen bears a narrow groove related to the greater petrosal nerve which passes posterolaterally to enter the bone by a hiatus anterior to the arcuate eminence. The groove passes forwards to the foramen lacerum. A smaller and similar hiatus and groove may be found more laterally and are related to the lesser petrosal nerve. This nerve runs to the foramen ovale. The posterior slope of the arcuate eminence overlies the posterior and lateral semicircular canals. Lateral to the eminence the posterior part of the tegmen tympani roofs the mastoid antrum. The posterior surface contributes to the anterior part of the posterior cranial fossa and is continuous with the internal surface of the mastoid part. The opening of the internal acoustic meatus lies near its centre. A small slit leading to the vestibular aqueduct lies behind the opening of the meatus, almost hidden by a thin plate of bone. This contains the saccus and ductus endolymphaticus together with a small artery and vein. The terminal half of the saccus endolymphaticus protrudes through the slit between the periosteum and dura mater. The subarcuate fossa lies above these openings. The irregular inferior surface is part of the exterior of the cranial base. Near the apex of the petrous part, there is a quadrilateral area which is partly associated with the attachment of levator veli palatini and the cartilaginous pharyngotympanic tube, and partly connected to the basilar part of the occipital bone by dense fibrocartilage. Behind this region is the large, circular opening of the carotid canal, and behind the opening of the canal is the jugular fossa, which is of variable depth and size and contains the superior jugular bulb. The inferior ganglion of the glossopharyngeal nerve lies in a triangular depression anteromedial to the jugular fossa (below the internal acoustic meatus). At its apex is a small opening into the cochlear canaliculus, occupied by the perilymphatic duct (a tube of dura mater) and a vein draining from the cochlea to the internal jugular vein. A canaliculus for the tympanic nerve from the glossopharyngeal nerve lies on the ridge between the carotid canal and the jugular fossa. The mastoid canaliculus for the auricular branch of the vagus nerve is laterally positioned in the jugular fossa. Behind the jugular fossa, the rough quadrilateral jugular surface is covered by cartilage which joins it to the jugular process of the occipital bone. The superior border, the longest, is grooved by the superior petrosal sinus. The attached margin of the tentorium cerebelli is fixed to the edges of the groove except at its medial end, where it is crossed by the roots of the trigeminal nerve. The posterior border, intermediate in length, bears a sulcus medially which forms, together with the occipital bone, a gutter for the inferior petrosal sinus. Behind this the jugular fossa contributes (together with the occipital bone) to the jugular foramen and is notched by the glossopharyngeal nerve. Bone on either or both sides of the jugular notch may meet the occipital bone and divide the jugular foramen into two or three parts. The anterior border is joined laterally to the squamous part of the temporal bone at the petrosquamosal suture; medially it articulates with the greater wing of the sphenoid bone. At the junction of the petrous and squamous parts two canals exist, one above the other, which are separated by a thin osseous plate. Both lead to the tympanic cavity; the upper canal contains tensor tympani, the lower canal is the pharyngotympanic tube. Tympanic part
The tympanic part of the temporal bone is a curved plate below the squamous part and anterior to the mastoid process. Internally it fuses with the petrous part and appears between this and the squamous part, where it is inferolateral to the
auditory orifice. Behind, it fuses with the squamous part and mastoid process and is the anterior limit of the tympanomastoid fissure. Its concave posterior surface forms the anterior wall, floor and part of the posterior wall of the external acoustic meatus. A narrow tympanic sulcus on the medial surface serves for the attachment of the tympanic membrane. The quadrilateral concave anterior surface is the posterior wall of the mandibular fossa and may contact the parotid gland. Its rough lateral border forms most of the margin of the external acoustic meatus and is continuous with its cartilaginous part. Laterally the upper border is fused with the back of the postglenoid tubercle; medially it is the posterior edge of the petrotympanic fissure. The inferior border is sharp, splitting laterally to form, at its root, the sheath of the styloid process (vaginal process). Centrally the tympanic part is thin, and is often perforated. The stylomastoid foramen lies between the styloid and mastoid processes: it represents the external end of the facial canal and transmits the facial nerve and stylomastoid artery. Styloid process
The styloid process is slender, pointed and projects anteroinferiorly from the inferior aspect of the temporal bone. Its length is variable, ranging from a few millimetres to an average of c.2.5 cm. Often approximately straight, it can show a curvature, an anteromedial concavity being most common. Its proximal part (tympanohyal) is ensheathed by the tympanic plate, especially anterolaterally, while muscles and ligaments are attached to its distal part (stylohyal). In vivo, its relationships are important. The styloid process is covered laterally by the parotid gland; the facial nerve crosses its base; the external carotid artery crosses its tip, embedded in the parotid; and medially the process is separated from the beginning of the internal jugular vein by the attachment of stylopharyngeus. External acoustic meatus
The temporal bone contains the bony part of the external acoustic meatus. This is c.16 mm long and slopes down anteromedially while its floor is convex upwards. In sagittal section it is oval or elliptical, with a long axis directed down and slightly back. The anterior wall, floor and lower posterior wall are formed by the tympanic plate; the roof and upper posterior wall are formed by the squamous part, and the medial end is closed by the tympanic membrane. The outer wall of the meatus is bounded above by the posterior zygomatic root, below which there may be a suprameatal spine. The cartilaginous part of the external acoustic meatus is attached to the lateral surface of the bony part. Ossification
The four temporal components ossify independently (Fig. 27.15). The squamous part is ossified in a sheet of condensed mesenchyme from a single centre near the zygomatic roots, which appear in the seventh or eighth week in utero. The petromastoid part has several centres appearing in the cartilaginous otic capsule during the fifth month. As many as 14 have been described, variable in order of appearance. Several are small and inconstant, soon fusing with others. The otic capsule is almost fully ossified by the end of the sixth month. The tympanic part is also ossified in mesenchyme from a centre identifiable about the third month; at birth it is an incomplete tympanic ring, deficient above, its concavity grooved by a tympanic sulcus for the tympanic membrane. Inclined obliquely downwards and forwards across the medial aspect of the anterior part of the ring is the malleolar sulcus for the anterior malleolar process, chorda tympani and anterior tympanic artery. The styloid process develops from two centres at the cranial end of cartilage in the second visceral or hyoid arch: a proximal centre for the tympanohyal, appearing before birth and another, for the distal stylohyal, after birth. The tympanic ring unites with the squamous part shortly before birth, and the petromastoid fuses with it and the tympanohyal during the first year. The stylohyal does not unite with the rest of the process until after puberty and may never do so. Once ossified, the tympanic cavity, mastoid antrum and the posterior end of the pharyngotympanic tube become surrounded by bone. The petrous part forms the roof, floor and medial wall of the cavity, while the squamous and tympanic parts, together with the tympanic membrane, form its lateral wall. At birth the middle and inner ears are adult size, and the tympanic cavity, mastoid antrum, tympanic membrane and auditory ossicles are all almost adult size. The anterior process does not join the malleus until 6 months later. The internal acoustic meatus is c.6 mm in horizontal diameter, 4 mm in vertical diameter and 7 mm in length at birth: the adult diameters are 7.7 mm and 11 mm respectively. page 470 page 471
Figure 27.15 The left temporal bone at birth. A, The three principal parts are, from left to right, petromastoid part (lateral aspect), tympanic ring (medial aspect) and squama (medial aspect). B, Lateral aspect with the rudimentary styloid process removed (yellow: tympanic part; brown: squamous part; uncoloured: petromastoid part). C, Medial aspect.
After birth and apart from general growth, the tympanic ring extends posterolaterally to become cylindrical, growing into a fibrocartilaginous tympanic plate, which forms the adjacent part of the external acoustic meatus at this stage. This growth is not equal but is rapid in the anterior and posterior regions, which meet and blend. Thus, for a time, there is in the floor an opening (foramen of Huschke), usually closed at about the fifth year, but sometimes permanent (in 546% of adult crania from ancient and modern populations). The external acoustic meatus is relatively as long in children as it is in adults, but the canal is fibrocartilaginous, whereas its medial two-thirds are osseous in adults. Surgical access to the tympanic cavity is via the mastoid antrum, and in children it is necessary to remove only a thin scale of bone in the suprameatal triangle to reach the antrum. The tympanic plate ensheathes the styloid process by posterior extension, and extends medially over the petrous bone to the carotid canal. Initially, the mandibular fossa is shallow, facing more laterally, but it then deepens and ultimately faces downwards. Posteroinferiorly the squamous part grows down behind the tympanic ring to form the lateral wall of the mastoid antrum. The mastoid part is at first flat, so that the stylomastoid foramen and rudimentary styloid process are immediately behind the tympanic ring. The mastoid part becomes invaded by air cells, especially at puberty. The lateral mastoid region grows downwards and forwards to form the mastoid process, so that the styloid process and stylomastoid foramen become inferior. Descent of the foramen lengthens the facial canal. The mastoid process is not perceptible until late in the
second year. The subarcuate fossa gradually fills and is almost obliterated. In the neonate the petrous and squamous parts of the temporal bone are usually partially separated by the petrosquamous fissure which opens directly into the mastoid antrum of the middle ear. The fissure closes in 4% of infants during the first year, but it remains unclosed in 20-40% up to the age of 19 years. It is a route for the spread of infection from the middle ear to the meninges. In the neonate the internal acoustic meatus is about half the length of that of the adult. Its opening from the middle ear cavity is as large as it is in the adult, but the pharyngeal opening in the nasal part of the pharynx is relatively smaller. The course of the pharyngotympanic tube is horizontal in the newborn, whereas in the adult it passes from the middle ear downward, forward and medially.
PARIETAL BONE (Fig. 27.16) The two parietal bones form most of the cranial roof and sides of the skull. Each is irregularly quadrilateral and has two surfaces, four borders and four angles. The external surface is convex and smooth, with a central parietal tuber (tuberosity). Curved superior and inferior temporal lines cross it and form posterosuperior arches. The temporal fascia is attached to the superior line or arch and the inferior line or arch indicates the upper limit of attachment of temporalis. The epicranial aponeurosis (galea aponeurotica) lies above these lines, and part of the temporal fossa lies below. Posteriorly, close to the sagittal (superior) border, an inconstant parietal foramen transmits a vein from the superior sagittal sinus and sometimes a branch of the occipital artery. The internal surface is concave and marked by cerebral gyri and grooves for the middle meningeal vessels. The vessels ascend, inclining backwards, from the sphenoidal (anteroinferior) angle and posterior half (or more) of its inferior border. A groove for the superior sagittal sinus lies along the sagittal border, and is completed by the groove on the opposite parietal bone. The falx cerebri is attached to the edges of the groove. Granular foveolae for arachnoid granulations flank the sagittal sulcus, being most pronounced in old age. The dentated sagittal border, longest and thickest, articulates with the opposite parietal bone at the sagittal suture. In the squamosal (inferior) border the anterior part is short, thin and truncated, bevelled externally and overlapped by the greater wing of the sphenoid. The middle part of the inferior border is arched, bevelled externally and overlapped by the squamous part of the temporal bone. The posterior part of the inferior border is short, thick and serrated for articulation with the mastoid part. The frontal border is deeply serrated, bevelled externally above, internally below, and articulates with the frontal bone to form one half of the coronal suture. The occipital border, deeply dentated, articulates with the occipital, forming one half of the lambdoid suture. The frontal (anterosuperior) angle, c.90°, is at the bregma, the meeting of the sagittal and coronal sutures. In the neonatal skull, this is the site of the anterior fontanelle. The sphenoidal (anteroinferior) angle is between the frontal bone and greater wing of the sphenoid. Its internal surface is marked by a deep groove or sometimes even a canal related to the frontal branches of the middle meningeal vessels. The frontal bone sometimes meets the squamous part of the temporal bone which means that the parietal bone then fails to reach the greater wing of the sphenoid bone. These four bones meet at the pterion. The rounded occipital (posterosuperior) angle is at the lambda, the meeting of the sagittal and lambdoid sutures. In the neonatal skull this marks the site of the posterior fontanelle. The blunt mastoid (posteroinferior) angle articulates with the occipital bone and the mastoid portion of the temporal bones. This is the site of the asterion. Internally it bears a broad, shallow groove for the junction of the transverse and sigmoid sinuses. Ossification
Each parietal bone is ossified from two centres which appear in dense mesenchyme near the tuberosity, one above the other, at about the seventh week in utero. These two centres unite early and the ossification radiates towards the margins. This means that the angles are the last to be ossified and so fontanelles occur at these sites. At birth the temporal lines are low down, and they only reach their final position after the eruption of the molar teeth. Occasionally the parietal bone is divided by an anteroposterior suture. page 471 page 472
Figure 27.16 Left parietal bone. A, External view; B, internal view. (By permission from Berkovitz and Moxham, 1994.)
FRONTAL BONE (Fig. 27.17) The frontal bone is like half a shallow, irregular cap forming the forehead or frons. It has three parts, and contains two cavities, which are the frontal sinuses. Squamous part of the frontal bone
This forms the major portion of the bone. The external surface has a rounded frontal tuber (tuberosity) c.3 cm above the midpoint of each supraorbital margin. These tubera vary, but are especially prominent in young skulls and more so in adult females than males. Below them and separated by a shallow groove, are two curved superciliary arches, medially prominent and joined by a smooth median elevated glabella. The arches are more prominent in males, depending partly on the size of the frontal sinuses; but prominence is occasionally associated with small sinuses. The curved supraorbital margins of the orbital openings lie inferior to the superciliary arches. The lateral two-thirds of each margin are sharp, the medial third rounded, and a supraorbital notch (or foramen), which transmits the supraorbital vessels and nerve, lies at the junction between them. A small frontal notch or foramen occurs medial to it in 50% of skulls. A supraorbital notch or foramen occurs equally in some populations, whereas the appearance of a frontal foramen is more variable (15-87% in various ethnic groups). Both features show sexual dimorphism. The supraorbital margin ends laterally in a strong, prominent zygomatic process that meets the zygomatic bone. From here a line curves posterosuperiorly, and divides into superior and inferior temporal lines, which are continued on the squamous part of the temporal bone. The area of the
frontal bone below and behind the temporal lines is known as the temporal surface and forms the anterior part of the temporal fossa; it gives an attachment to temporalis. The parietal (posterior) margin is thick, deeply serrated, bevelled internally above and externally below. Inferiorly it becomes a rough, triangular surface for the greater wing of the sphenoid. UPDATE Date Added: 08 July 2005 Abstract: The sequence in appearance and disappearance of impressiones gyrorum cerebri and cerebelli. Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=15666620&query_hl=7 The sequence in appearance and disappearance of impressiones gyrorum cerebri and cerebelli. The internal surface is concave. Its upper, median part has a vertical sulcus for the sagittal sinus, the edges of which unite below as the frontal crest. The sulcus contains the anterior part of the superior sagittal sinus, and part of the falx cerebri is attached to its margins and frontal crest. The crest ends in a small notch, which is completed by the ethmoid bone to form a foramen caecum. The internal surface shows impressions of cerebral gyri and small furrows for meningeal vessels. Several granular foveolae for arachnoid granulations usually exist near the sagittal sulcus. Nasal part of the frontal bone page 472 page 473
page 473 page 474
Figure 27.17 Frontal bone. A, External view; B, inferior view; C, internal view. (By permission from Berkovitz and Moxham, 1994.)
The region between the supraorbital margins is the nasal part. A serrated nasal notch articulates with the nasal bones inferiorly and with the frontal processes of the maxillae and the lacrimal bones laterally. From the centre of the notch posteriorly the bone projects anteroinferiorly behind the nasal bones and the frontal processes of the maxillae, and supports the nasal bridge. The region ends in a sharp nasal spine, on each side of which a small grooved surface partly roofs the ipsilateral nasal cavity. The nasal spine makes a very small contribution to the nasal septum. In front it articulates with the crest of the nasal bones and behind with the perpendicular plate of the ethmoid bone. Orbital parts of the frontal bone
The orbital parts of the frontal bone are two thin, curved, triangular laminae which form the largest part of the orbital roofs and are separated by a wide ethmoidal notch. Most of the frontal bone is thick, and displays trabecular tissue lying between two compact laminae. In contrast, the orbital plates consist entirely of compact bone and are thin and often translucent posteriorly, indeed they may be partly absorbed in old age. The orbital surface of each plate is smooth and concave, and bears a shallow anterolateral fossa for the lacrimal gland. Below and behind the medial end of the supraorbital margin, midway between the supraorbital notch and frontolacrimal suture, is the trochlear fovea (or spine) for attachment of a fibrocartilaginous trochlea for superior oblique. The convex cerebral surface is marked by frontal gyri and faint grooves for meningeal vessels. The quadrilateral ethmoidal notch is occupied by the cribriform plate of the ethmoid bone. Inferior to its lateral margins, the bone articulates with the labyrinths of the ethmoid bone and impressions of the ethmoidal air cells can be seen on this surface. Two transverse grooves across each margin are converted into anterior and posterior ethmoidal canals by articulation with the ethmoid bone. These canals open on the medial orbital wall, transmitting anterior and posterior ethmoidal nerves and vessels. Openings of the frontal sinuses are anterior to the ethmoidal notch and lateral to the nasal spine. These two irregular cavities ascend posterolaterally for a variable distance between the frontal laminae, separated by a thin septum and usually deflected from the median plane. The sinuses are consequently rarely symmetrical, are variable in size, and larger in males. Each communicates with the middle meatus in the ipsilateral nasal cavity by a frontonasal canal. The degree of development is linked to the prominence of the superciliary arches, which is thought to be a response to masticatory stresses. The posterior borders of the orbital plates are thin and serrated to articulate with the lesser wings of the sphenoid; their lateral parts usually appear in the middle cranial fossa between greater and lesser wings. Ossification
The frontal bone is ossified in fibrous mesenchyme from two primary centres that appear in the eighth week in utero, one near each frontal tuber. Ossification extends superiorly to form half of the main part of the bone; posteriorly to form the orbital part, and inferiorly to form nasal parts (Fig. 27.18). Two secondary centres occur for the nasal spine, and appear about the tenth year. At birth the bone consists of two halves, and the median suture usually disappears by about 8 years. However, it may persist as the metopic suture. Such metopism has been assessed at 0-7.4% of individuals in various ethnic groups. The sinuses are rudimentary at birth and can barely be distinguished. Growth is slow in the early years but it can be detected radiographically by 6 years. The sinuses show a primary expansion with eruption of the first deciduous molars at c.1! years, and again when the permanent molars begin to appear in the sixth year. They reach full size after puberty, although with advancing age osseous absorption may lead to further enlargement.
ETHMOID BONE (Fig. 27.19) The ethmoid bone is cuboidal and fragile, and lies anteriorly in the cranial base. It contributes to the medial orbital walls, nasal septum, and the roof and lateral walls of the nasal cavity. It has a horizontal, perforated cribriform plate, a median perpendicular plate, and two lateral labyrinths which contain the ethmoidal air cells. Cribriform plate
Figure 27.18 The frontal bone at birth: anterior aspect. Note that at this stage the bone consists of right and left halves connected by the frontal suture.
The cribriform plate fills the ethmoidal notch of the frontal bone and forms a large part of the nasal roof. It derives its name from the fact that it is penetrated by numerous foramina containing branches of the olfactory nerves and their associated meninges. A thick, smooth, triangular, median crista galli projects up from the horizontally aligned plate of bone. The falx cerebri is attached to its thin and curved posterior border. Its shorter, thick, anterior border joins the frontal bone by two small alae, to complete the foramen caecum. Its sides are generally smooth, but may show slight bulges related to underlying ethmoidal air cells. On both sides of the crista galli, the cribriform plate is narrow and depressed: it is related to the gyrus rectus and the olfactory bulb which lie above it. On each side of the crista anteriorly there is a small slit occupied by dura mater, and just anterolateral to the slit, a foramen which transmits the anterior ethmoidal nerve and vessels to the nasal cavity: a groove runs forwards to the foramen caecum from the anterior ethmoidal canal. Perpendicular plate
The perpendicular plate is thin, flat, quadrilateral and median, and it descends from the cribriform plate to form the upper part of the nasal septum, usually deviating slightly from the midline. Its anterior border meets the nasal spine of the frontal bone and the crests of the nasal bones. Its posterior border joins the crest
of the body of the sphenoid bone above and vomer below. The thick inferior border is attached to the nasal septal cartilage. Its surfaces are smooth, except above, where numerous grooves and canals lead to medial foramina in the cribriform plate for filaments of the olfactory nerves. Ethmoidal labyrinths
The ethmoidal labyrinths consist of thin-walled ethmoidal air cells between two vertical plates and are arranged in anterior, middle and posterior groups. On average there are c.11 anterior ethmoidal air cells, 3 middle, and 6 posterior. The lateral surface (orbital plate) of the labyrinth is part of the medial orbital wall. In the disarticulated bone, many air cells are open, but in life these are closed when articulated with adjoining bones, except where they open into the nasal cavity. The superior surface shows open air cells that will be covered by the edges of the ethmoidal notch of the frontal bone. It is crossed by two grooves that will be converted into anterior and posterior ethmoidal canals by the frontal bone. On the posterior surface open air cells are present that will be covered by the sphenoidal conchae and the orbital process of the palatine bone. The thin, smooth, oblong orbital plate covers the middle and posterior ethmoidal air cells. It articulates superiorly with the orbital plate of the frontal bone, inferiorly with the maxilla and orbital process of the palatine bone, anteriorly with the lacrimal bone and posteriorly with the sphenoid bone. The walls of the air cells lying anterior to the orbital plate are completed by the lacrimal bone and frontal process of the maxilla. page 474 page 475
Figure 27.19 Ethmoid bone. A, Superior view; B, inferior view; C, posterior view; D, lateral view. (By permission from Berkovitz and Moxham, 1994.)
A thin, curved uncinate process, variable in size, projects posteroinferiorly from the labyrinth. The upper edge of this process is a medial boundary of the hiatus semilunaris in the middle meatus. The uncinate process appears in the medial wall of the maxillary sinus as it crosses the ostium of the maxillary sinus to join the ethmoidal process of the inferior nasal concha. The medial surface of the labyrinth forms part of the lateral nasal wall as a thin lamella descending from the inferior surface of the cribriform plate to end as the convoluted middle nasal concha. Above this the surface contains many vertical grooves for olfactory nerves. Posteriorly it is divided by the narrow, oblique superior meatus, bounded above by the thin, curved superior nasal concha. Posterior ethmoidal air cells open into this meatus. Anteroinferior to the superior meatus, the convex surface of the middle nasal concha extends along the entire medial surface of the labyrinth, its lower edge being thick. Its lateral surface is concave and forms part of the middle meatus. Middle ethmoidal air cells produce a swelling (bulla ethmoidalis) on the lateral wall of the middle meatus. These air cells open into the meatus, on the bulla or above it. A curved infundibulum extends up and forwards from the middle meatus and communicates with the anterior ethmoidal sinuses. In more than 50% of crania it continues up as the frontonasal duct to include the drainage point for the frontal sinus. Ossification
The ethmoid bone ossifies in the cartilaginous nasal capsule from three centres, one in the perpendicular plate, and one in each labyrinth. The latter two appear in the orbital plates between the fourth and fifth months in utero, and extend into the ethmoid conchae. At birth, the labyrinths, although ill-developed, are partially ossified, and the remainder are cartilaginous. During the first year the perpendicular plate begins to ossify from the median centre, and fuses with the labyrinths early in the second year. The cribriform plate is ossified partly from the perpendicular plate, and partly from the labyrinths. The crista galli ossifies during the second year. The parts of the ethmoid bone unite to form a single bone at c.3 years of age. Ethmoidal air cells begin to develop c.3 months in utero. Although present at birth, they are difficult to visualize radiographically until the end of the first year. They grow slowly and have almost reached adult size by the age of 12 years.
INFERIOR NASAL CONCHA (Fig. 27.20)
page 475 page 476
Figure 27.20 Inferior nasal concha. A, Medial view; B, lateral view; C, posterior view. (By permission from Berkovitz and Moxham, 1994.)
The inferior nasal conchae are curved horizontal laminae in the lateral nasal walls. Each has two surfaces (medial and lateral), two borders (superior and inferior) and two ends (anterior and posterior). The medial surface is convex, much perforated, and longitudinally grooved by vessels. The lateral surface is concave
and part of the inferior meatus. The superior border, thin and irregular, may be divided into three regions, namely, an anterior region articulating with the conchal crest of the maxilla, a posterior region articulating with the conchal crest of the palatine bone, and a middle region with three processes, which are variable in size and form. The lacrimal process is small and pointed and lies towards the front, and articulates apically with a descending process from the lacrimal bone. It also articulates at its margins with the edges of the nasolacrimal groove on the medial surface of the maxilla, and so helps complete the nasolacrimal canal. Most posteriorly, a thin ethmoidal process ascends to meet the uncinate process of the ethmoid bone. An intermediate thin maxillary process curves inferolaterally to articulate with the medial surface of the maxilla at the opening of the maxillary sinus. The inferior border is thick and spongiose, especially in its midpart. Both anterior and posterior ends of the inferior nasal concha are more or less tapered, the posterior more than the anterior. Ossification
Ossification is from one centre which appears at about the fifth month in utero in the incurved lower border of the cartilaginous lateral wall of the nasal capsule. It loses continuity with the capsule during ossification.
LACRIMAL BONE (Fig. 27.21) The lacrimal bones, the smallest and most fragile of the cranial bones, lie anteriorly in the medial walls of the orbits. Each has two surfaces (medial and lateral) and four borders (anterior, posterior, superior and inferior). The lateral (orbital) surface is divided by a vertical posterior lacrimal crest. There is a vertical groove anterior to the crest, and the anterior edge of this groove meets the posterior border of the frontal process of the maxilla to complete the fossa for the lacrimal sac. The medial wall of the groove is prolonged by a descending process that contributes to the formation of the nasolacrimal canal by joining the lips of the nasolacrimal groove of the maxilla and the lacrimal process of the inferior nasal concha. A smooth part of the medial orbital wall lies behind the posterior lacrimal crest: the lacrimal part of orbicularis oculi is attached to this surface and crest. The surface ends below in the lacrimal hamulus which, together with the maxilla, completes the upper opening of the nasolacrimal canal. The hamulus may appear as a separate lesser lacrimal bone. The anteroinferior region of the medial (nasal) surface is part of the middle meatus. Its posterosuperior part meets the ethmoid to complete some anterior ethmoidal air cells. The anterior lacrimal border articulates with the frontal process of the maxilla, the posterior border with the orbital plate of the ethmoid bone, the superior border with the frontal bone, and the inferior border with the orbital surface of the maxilla. Ossification
Ossification is from a centre that appears at about the twelfth week in utero in mesenchyme around the nasal capsule. In later life the lacrimal bone is subject to patchy erosion.
NASAL BONE (Fig. 27.22)
Figure 27.21 Lacrimal bone. A, External view; B, internal view. (By permission from Berkovitz and Moxham, 1994.)
Figure 27.22 Left nasal bone. A, External view; B, internal view. (By permission from Berkovitz and Moxham, 1994.)
The nasal bones are small, oblong, variable in size and form, and placed side by side between the frontal processes of the maxillae. They jointly form the nasal bridge. Each nasal bone has two surfaces (external and internal) and four borders (superior, inferior, lateral and mesial). The external surface has a descending concavo-convex profile and is transversely convex. It is covered by procerus and nasalis and perforated centrally by a small foramen traversed by a vein. The internal surface, transversely concave, is traversed by a longitudinal groove for the anterior ethmoidal nerve. The superior border, thick and serrated, articulates with the nasal part of the frontal bone. The inferior border, thin and notched, is continuous with the lateral nasal cartilage. The lateral border joins the frontal process of the maxilla. The medial border, thicker above, meets its fellow and projects behind as a vertical crest, so forming a small part of the nasal septum. It articulates from above with the nasal spine of the frontal bone, the perpendicular plate of the ethmoid bone, and the nasal septal cartilage. Ossification
Ossification is from a centre that appears early in the third month in utero in mesenchyme overlying the cartilaginous anterior part of the nasal capsule.
VOMER (Fig. 27.23) The vomer is thin, flat, and almost trapezoid. It forms the posteroinferior part of the nasal septum and presents two surfaces and four borders. Both surfaces are marked by grooves for nerves and vessels, and a prominent groove for the nasopalatine nerve and vessels lies obliquely in an anteroinferior plane. The superior border is thickest, and possesses a deep furrow between projecting alae which fits the rostrum of the body of the sphenoid bone. The alae articulate with the sphenoidal conchae, the vaginal processes of the medial pterygoid plates of the sphenoid bone, and the sphenoidal processes of the palatine bones. Where each ala lies between the body of the sphenoid and the vaginal process, its inferior surface helps to form the vomerovaginal canal. The inferior border articulates with the median nasal crests of the maxilla and palatine bones. The anterior border is the longest, and articulates in its upper half with the perpendicular plate of the ethmoid bone. Its lower half is cleft to receive the inferior margin of the nasal septal cartilage. The posterior border is concave, separating the posterior nasal apertures. It is thick and bifid above, thin below. The anterior extremity of the vomer articulates with the posterior margin of the maxillary incisor crest and descends between the incisive canals. Ossification
The nasal septum is at first a plate of cartilage, part of which is ossified above to form the perpendicular plate of the ethmoid. Its anteroinferior region persists as septal cartilage. The vomer is ossified in a layer of connective tissue which covers cartilage on each aspect in its posteroinferior part. About the eighth week in utero, two centres appear flanking the midline, and in the twelfth week these unite below the cartilage, to form a deep groove for the nasal septal cartilage. Union of the bony lamellae progresses anterosuperiorly while intervening cartilage is absorbed. By puberty they are almost united, but the bilaminar origin remains in the everted alae and anterior marginal groove.
ZYGOMATIC BONE (Fig. 27.24) Each zygomatic bone forms the prominence of a cheek, contributes to the floor
and lateral wall of the orbit and the walls of the temporal and infratemporal fossae, and completes the zygomatic arch. It is roughly quadrangular and is described as having three surfaces, five borders and two processes. page 476 page 477
Figure 27.23 Vomer. A, Lateral view; B, posterior view. (By permission from Berkovitz and Moxham, 1994.)
The lateral (facial) surface is convex and is pierced near its orbital border by the zygomaticofacial foramen, which is often double and occasionally absent, for the zygomaticofacial nerve and vessels. This surface gives attachment to zygomaticus major posteriorly and zygomaticus minor anteriorly. The posteromedial (temporal) surface has a rough anterior area for articulation with the zygomatic process of the maxilla, and a smooth, concave posterior area that extends up posteriorly on its frontal process as the anterior aspect of the temporal fossa. It also extends back on the medial aspect of the temporal process as an incomplete lateral wall for the infratemporal fossa. The zygomaticotemporal foramen pierces this surface near the base of the frontal process. The smooth and concave orbital surface forms the anterolateral part of the orbital floor and adjoining lateral wall, and extends up on the medial aspect of its frontal process. It usually bears zygomatico-orbital foramina which represent the openings of canals leading to zygomaticofacial and zygomaticotemporal foramina. The smoothly concave anterosuperior (orbital) border forms the inferolateral circumference of the orbital opening, and separates the orbital and lateral surfaces of the bone. The anteroinferior (maxillary) border articulates with the maxilla. Its medial end tapers to a point above the infraorbital foramen. A part of levator labii superioris is attached at this surface. The posterosuperior (temporal) border is sinuous, convex above and concave below, and is continuous with the posterior border of the frontal process and upper border of the zygomatic arch. The temporal fascia is attached to this border. There is often a small, easily palpable, marginal tubercle below the frontozygomatic suture. The posteroinferior border is roughened for the attachment of masseter. The serrated posteromedial border articulates with the greater wing of the sphenoid bone above, and the orbital surface of the maxilla below. Between these serrated regions a short, concave, non-articular part usually forms the lateral edge of the inferior orbital fissure. Occasionally absent, the fissure is then completed by junction of the maxilla and sphenoid bones (or with a small sutural bone between them).
Figure 27.24 Left zygomatic bone. A, External view; B, internal view. (By permission from Berkovitz and Moxham, 1994.)
The frontal process, thick and serrated, articulates above with the zygomatic process of the frontal bone and behind with the greater wing of the sphenoid bone. A tubercle of varying size and form, Whitnall's tubercle, is usually present on its orbital aspect, within the orbital opening and c.1 cm below the frontozygomatic suture. This tubercle provides attachment for the lateral palpebral ligament, the suspensory ligament of the eye, and part of the aponeurosis of levator palpebrae superioris. The temporal process, directed backwards, has an oblique, serrated end that articulates with the zygomatic process of the temporal bone to complete the zygomatic arch. Ossification
Ossification is from one centre, which appears in fibrous tissue about the eighth week in utero. The bone is sometimes divided by a horizontal suture into a larger upper and smaller lower part.
MAXILLA (Fig. 27.25) The maxillae are the largest of the facial bones, other than the mandible, and jointly form the whole of the upper jaw. Each bone forms the greater part of the floor and lateral wall of the nasal cavity, and of the floor of the orbit. It contributes to the infratemporal and pterygopalatine fossae and bounds the inferior orbital and pterygomaxillary fissures. The maxilla has a body and four processes, namely the zygomatic, frontal, alveolar and palatine processes. Body
The body of the maxilla is roughly pyramidal, and has anterior, infratemporal (posterior), orbital and nasal surfaces that enclose the maxillary sinus. page 477 page 478
Figure 27.25 Left maxilla. A, Lateral view; B, superior view; C, medial view. (By permission from Berkovitz and Moxham, 1994.)
Anterior surface
This surface faces anterolaterally and displays inferior elevations overlying the roots of teeth. There is a shallow incisive fossa above the incisors to which depressor septi is attached. A slip of orbicularis oris is attached to the alveolar border below this fossa, and nasalis is attached superolateral to it. Lateral to the incisive fossa is a larger, deeper canine fossa and levator anguli oris is attached to the bone of this fossa. The incisive and canine fossae are separated by the canine eminence, which overlies the socket of the canine tooth. The infraorbital foramen lies above the fossa and transmits the infraorbital vessels and nerve. Above the foramen a sharp border separates the anterior and orbital surfaces of the bone and contributes to the infraorbital margin. Levator labii superioris is attached here above the infraorbital foramen and levator anguli oris below it. Medially the anterior surface ends at a deeply concave nasal notch, and terminates in a pointed process which, with its fellow of the opposite side, forms the anterior nasal spine. Nasalis and depressor septi are attached to the anterior surface near the notch. Infratemporal surface
This surface is concave and faces posterolaterally, forming the anterior wall of the infratemporal fossa. It is separated from the anterior surface by the zygomatic process and a ridge (jugal crest) that ascends to it from the first molar socket. Near its centre are the openings of two or three alveolar canals, which transmit posterior superior alveolar vessels and nerves. Posteroinferior is the maxillary tuberosity which is roughened superomedially where it meets the pyramidal process of the palatine bone. A few fibres of medial pterygoid are attached here. Above the tuberosity the smooth anterior boundary of the pterygopalatine fossa is grooved by the maxillary nerve as it passes laterally and slightly upwards into the infraorbital groove on the orbital surface. Orbital surface page 478 page 479
This surface is smooth and triangular, and forms most of the floor of the orbit. Anteriorly its medial border bears a lacrimal notch, behind which it articulates with the lacrimal bone, the orbital plate of the ethmoid and, posteriorly, with the orbital process of the palatine bone. Its posterior border is smoothly rounded, and forms most of the anterior edge of the inferior orbital fissure. The infraorbital groove lies centrally. The anterior border is part of the orbital margin, and is continuous medially with the lacrimal crest of the frontal process of the maxilla. The infraorbital groove transmits similarly named vessels and a nerve, and begins midway on the posterior border, where it is continuous with a groove on the
posterior surface. It passes forwards into the infraorbital canal which opens on the anterior surface below the infraorbital margin. Near its midpoint, the infraorbital canal has a small lateral branch, the canalis sinuosus, for the anterior superior alveolar nerve and vessels. The canalis sinuosus descends in the orbital floor lateral to the infraorbital canal and curves medially in the anterior wall of the maxillary sinus. It then passes below the infraorbital foramen to the margin of the anterior nasal aperture in front of the anterior end of the inferior concha. Here it follows the lower margin of the aperture to open near the nasal septum in front of the incisive canal. The site of the attachment of inferior oblique may be indicated by a small depression in the bone at the anteromedial corner of the orbital surface, lateral to the lacrimal groove. Nasal surface
This surface displays posterosuperiorly the large, irregular maxillary hiatus leading into the maxillary sinus. Parts of air sinuses which are completed by articulation with the ethmoid and lacrimal bones lie at the upper border of the hiatus. The smooth concave surface below the hiatus is part of the inferior meatus. Posteriorly, the surface is roughened where it meets the perpendicular plate of the palatine bone. This surface is traversed by a groove which descends forwards from the midposterior border, and is converted into a greater palatine canal by the perpendicular plate. Anterior to the hiatus, a deep groove, the nasolacrimal groove, which is continuous above with the lacrimal groove, makes up about two-thirds of the circumference of the nasolacrimal canal. The rest is contributed by the descending part of the lacrimal bone and the lacrimal process of the inferior nasal concha. This canal leads the nasolacrimal duct to the inferior meatus. More anterior is an oblique conchal crest which articulates with the inferior nasal concha. The concavity below it is part of the inferior meatus, while the surface above it is part of the atrium of the middle meatus. Zygomatic process
Anterior, infratemporal and orbital surfaces converge at a pyramidal projection, the zygomatic process. Anteriorly, the process merges into the facial surface of the body of the maxilla. Posteriorly, it is concave and continuous with the infratemporal surface. Superiorly, it is roughly serrated for articulation with the zygomatic bone. Inferiorly, a bony arched ridge, the zygomaticoalveolar ridge or jugal crest, separates the facial (anterior) and infratemporal surfaces. Frontal process
The frontal process projects posterosuperiorly between the nasal and lacrimal bones. Its lateral surface is divided by a vertical anterior lacrimal crest which gives attachment to the medial palpebral ligament and is continuous below with the infraorbital margin. A small palpable tubercle at the junction of the crest and orbital surface is a guide to the lacrimal sac. The smooth area anterior to the lacrimal crest merges below with the anterior surface of the body of the maxilla. Parts of orbicularis oculi and levator labii superioris alaeque nasi are attached here. Behind the crest, a vertical groove combines with one on the lacrimal bone to complete the lacrimal fossa. The medial surface is part of the lateral nasal wall. A rough subapical area articulates with the ethmoid, and closes anterior ethmoidal air cells. Below this an oblique ethmoidal crest articulates posteriorly with the middle nasal concha, and anteriorly underlies the agger nasi, which is a ridge anterior to the concha on the lateral nasal wall. The ethmoidal crest forms the upper limit of the atrium of the middle meatus. The frontal process articulates above with the nasal part of the frontal bone. Its anterior border articulates with the nasal bone and its posterior border articulates with the lacrimal bone. Alveolar process
The alveolar process is thick and arched, wide behind, and socketed for the roots of the upper teeth. The eight sockets on each side vary according to the tooth type. The socket for the canine is deepest, the sockets for molars are widest and subdivided into three by septa, those for incisors and second premolar are single, and that for the first premolar usually double. Buccinator is attached to the external alveolar aspect as far forwards as the first molar. In articulated maxillae the processes form the alveolar arch. Occasionally a variably prominent longitudinal maxillary torus is present on the palatal aspect of the process near the molar sockets. Palatine process
The palatine process, thick and horizontal, projects medially from the lowest part of the medial aspect of the maxilla. It forms a large part of the nasal floor and hard palate and is much thicker in front. Its inferior surface is concave and uneven, and it forms with its contralateral fellow the anterior three-fourths of the
osseous (hard) palate. The palatine process displays numerous vascular foramina and depressions for palatine glands and, posterolaterally, two grooves that transmit the greater palatine vessels and nerves. The infundibular incisive fossa is placed between the two maxillae, behind the incisor teeth. The median intermaxillary palatal suture runs posterior to the fossa, and although a little uneven, is usually relatively flat on its oral aspect. Its bony margins are sometimes raised into a prominent longitudinal palatine torus. Two lateral incisive canals, each ascending into its half of the nasal cavity, open in the incisive fossa: they transmit the terminations of the greater palatine artery and nasopalatine nerve. Two additional median openings, anterior and posterior incisive foramina, are occasionally present: they transmit the nasopalatine nerves, the left passing through the anterior and the right through the posterior foramen. On the inferior palatine surface a fine groove, sometimes termed the incisive suture, and prominent in young skulls, may be observed in adults. It extends anterolaterally from the incisive fossa to the interval between the lateral incisor and canine teeth. The superior surface of the palatine process is smooth, concave transversely, and forms most of the nasal floor. The incisive canal lies anteriorly, near its median margin. The lateral border is continuous with the body of the maxilla. The medial border, thicker in front, is raised into a nasal crest that, with its contralateral fellow, forms a groove for the vomer. The front of this ridge rises higher as an incisor crest, prolonged forwards into a sharp process which, with its fellow, forms an anterior nasal spine. The posterior border is serrated for articulation with the horizontal plate of the palatine bone. Maxillary sinus
The maxillary sinus is the largest of the paranasal sinuses and is situated in the body of the maxilla. Pyramidal in shape, the base (medial wall) forms part of the lateral wall of the nose, while the apex extends into the zygomatic process of the maxilla. The floor is formed by the alveolar process and part of the palatine process of the maxilla, and the roof contributes the major part of the floor of the orbit. The facial and infratemporal surfaces of the maxilla form its anterior and posterior walls respectively. The sinus may be partially divided by bony septa. The medial wall of the maxilla sinus bears the opening (ostium) of the sinus. The infraorbital nerve and vessels lie within the infraorbital canal in the roof. The floor of the sinus lies below the level of the floor of the nose, and is related to the roots of the cheek teeth, although this relationship is variable according to the size of the sinus. Usually, at least the upper second premolar and first molar are related to the floor of the sinus, however the sinus may extend anteriorly to the first premolar tooth - and sometimes even to the canine - and posteriorly to the third molar tooth. The anterior superior alveolar nerve and vessels that arise within the infraorbital canal pass downwards in a fine canal (canalis sinuosus) in the anterior wall of the maxillary sinus, to be distributed to the anterior upper teeth. The posterior superior alveolar nerve and vessels pass through canals in the posterior surface of the sinus. In an isolated maxillary, the ostium of the maxillary sinus appears large. However, in life, the size of the ostium is considerably reduced by portions of the perpendicular plate of the palatine bone, the uncinate process of the ethmoid bone, the inferior nasal concha and the lacrimal bone, and by the overlying nasal mucosa. The ostium lies high up at the back of the medial wall of the maxillary sinus, where it is unfavourably situated for drainage. It usually opens into the posterior part of the ethmoidal infundibulum, and thence into the hiatus semilunaris of the middle meatus of the lateral wall of the nose. An accessory ostium is sometimes present behind the major ostium. Ossification
The maxilla ossifies from a single centre in a sheet of mesenchyme that appears above the canine fossa at about the sixth week in utero and spreads into the rest of the maxilla and its processes. The pattern of spread of ossification may initially leave an unmineralized zone in a region roughly corresponding to a site where a premaxillary suture may occur. However, this deficiency is soon ossified and there is no evidence of a separate centre of ossification for the incisor-bearing portion of the maxilla (i.e. premaxilla). The maxillary sinus appears as a shallow groove on the nasal aspect at about the fourth month in utero. Though small at birth, the sinus is identifiable radiologically. After birth the maxillary sinus enlarges with the growing maxilla, though it is only fully developed following the eruption of the permanent dentition. The infraorbital vessels and nerve are for a time in an open groove in the orbital floor, and the anterior part is subsequently converted into a canal by a lamina that grows in from the lateral side.
page 479 page 480
Figure 27.26 Maxilla at birth.
Age changes in the maxilla
At birth the transverse and sagittal maxillary dimensions are greater than the vertical. The frontal process is prominent, but the body is little more than an alveolar process, since the alveoli reach almost to the orbital floor (Fig. 27.26). In adults the vertical dimension is greatest, reflecting the development of the alveolar process and enlargement of the sinus. When teeth are lost, the bone reverts towards its infantile shape. Thus, its height diminishes, the alveolar process is absorbed, and the lower parts of the bone contract and become reduced in thickness at the expense of the labial wall.
PALATINE BONE (Fig. 27.27) The palatine bones are posteriorly placed in the nasal cavity between the maxillae and the pterygoid processes of the sphenoid bones. They contribute to the floor and lateral wall of the nose, to the floors of the palate and orbit, to the pterygopalatine and pterygoid fossae, and to the inferior orbital fissures. Each has two plates (horizontal and perpendicular plates) arranged as an L-shape, and three processes (pyramidal, orbital and sphenoidal). Horizontal plate
The horizontal plate is quadrilateral, with two surfaces (nasal and palatine) and four borders (anterior, posterior, lateral and medial). The nasal surface, transversely concave, forms the posterior nasal floor. The palatine surface forms, with its fellow, the posterior quarter of the bony palate. There is often a curved palatine crest near its posterior margin. The posterior border is thin and concave: the expanded tendon of tensor veli palatini is attached to it and its adjacent surface behind the palatine crest. Medially, with its fellow from the opposite side, the posterior border forms a median posterior nasal spine to which the uvular muscle is attached. The serrated anterior border articulates with the palatine process of the maxilla. The lateral border is continuous with the perpendicular plate of the palatine bone and is marked by a greater palatine groove. The medial border, thick and serrated, articulates with its fellow in the midline, and forms the posterior part of the nasal crest which articulates with the posterior part of the lower edge of the vomer. Perpendicular plate
The perpendicular plate is thin and oblong, and has two surfaces (nasal and maxillary) and four borders (anterior, posterior, superior and inferior). The nasal surface bears two crests (conchal and ethmoidal) and shows areas which contribute to the inferior middle and superior meatuses. Inferiorly, the nasal surface is concave where it contributes to part of the inferior meatus. Above this is a horizontal conchal crest that articulates with the inferior concha. Above the conchal crest the surface presents a shallow depression which forms part of the middle meatus. This depression is limited above by an ethmoidal crest for the middle nasal concha, above which a narrow, horizontal groove forms part of the
superior meatus. The maxillary surface is largely rough and irregular and articulates with the nasal surface of the maxilla. Posterosuperiorly it forms a smooth medial wall to the pterygopalatine fossa. Its anterior area, also smooth, overlaps the maxillary hiatus from behind to form a posterior part of the medial wall of the maxillary sinus. A deep, obliquely descending greater palatine groove - converted into a canal by the maxilla - lies posteriorly on this maxillary surface and transmits the greater palatine vessels and nerve.
Figure 27.27 Left palatine bone. A, Anterior view; B, medial view; C, Inferior view; D, Posterior view and muscle attachment. (By permission from Berkovitz and Moxham, 1994.)
The anterior border is thin and irregular. Level with the conchal crest, a pointed lamina projects below and behind the maxillary process of the inferior concha: it articulates with it and so appears in the medial wall of the maxillary sinus. The posterior border has a serrated suture with the medial pterygoid plate. It is continuous above with the sphenoidal process of the palatine bone and expands below into its pyramidal process. Orbital and sphenoidal processes project from the superior border, and are separated by the sphenopalatine notch (converted into a foramen by articulation with the body of the sphenoid). This foramen connects the pterygopalatine fossa to the posterior part of the superior meatus, and transmits sphenopalatine vessels and the posterior superior nasal nerves. The inferior border is continuous with the lateral border of the horizontal plate and
bears the lower end of the greater palatine groove in front of the pyramidal process. Pyramidal process page 480 page 481
The pyramidal process slopes down posterolaterally from the junction of the horizontal and perpendicular palatine plates into the angle between the pterygoid plates of the sphenoid bone. On its posterior surface a smooth, grooved triangular area, limited on each side by rough articular furrows which articulate with the pterygoid plates, completes the lower part of the pterygoid fossa. Anteriorly the lateral surface articulates with the maxillary tuberosity. This area gives attachment to fibres of the superficial head of medial pterygoid. Posteriorly a smooth triangular area appears low in the infratemporal fossa between the tuberosity and lateral pterygoid plate. The inferior surface, near its union with the horizontal plate, bears the lesser palatine foramina for the corresponding nerves and arteries. Orbital process
The orbital process is directed superolaterally from in front of the perpendicular plate, and has a constricted 'neck'. It encloses an air sinus and presents three articular and two non-articular surfaces. Of the articular surfaces, the oblong anterior, or maxillary, surface faces down and anterolaterally to articulate with the maxilla; the posterior, or sphenoidal, surface, directed up and posteromedially, bears the opening of an air sinus. It usually communicates with the sphenoidal sinus, and is completed by a sphenoidal concha. The medial, or ethmoidal, surface, faces anteromedially and articulates with the labyrinth of the ethmoid bone. The sinus of the orbital process sometimes opens on the surface, and communicates with the posterior ethmoidal air cells. More rarely it opens on both the ethmoidal and sphenoidal surfaces, and communicates with both posterior ethmoidal air cells and the sphenoidal sinus. Of the non-articular surfaces, the triangular superior or orbital surface is directed superolaterally to the posterior part of the orbital floor. The lateral surface is oblong, faces the pterygopalatine fossa and is separated from the orbital surface by a rounded border that forms a medial part of the lower margin of the inferior orbital fissure. This surface may present a groove, directed superolaterally, for the maxillary nerve and is continuous with the groove on the upper posterior surface of the maxilla. The border between the lateral and posterior surfaces descends anterior to the sphenopalatine notch. Sphenoidal process
The sphenoidal process is a thin plate that is smaller and lower than the orbital process, and is directed superomedially. Its superior surface articulates with the sphenoidal concha and, above it, the root of the medial pterygoid plate. It carries a groove that contributes to the formation of the palatovaginal canal. The concave inferomedial surface forms part of the roof and lateral wall of the nose. Posteriorly the lateral surface articulates with the medial pterygoid plate, while its smooth anterior region forms part of the medial wall of the pterygopalatine fossa. The posterior border articulates with the vaginal process of the medial pterygoid plate. The anterior border is the posterior edge of the sphenopalatine notch. The medial border articulates with the ala of the vomer. The sphenopalatine notch, between the two processes, becomes a foramen by articulation with the body of the sphenoid bone. Ossification
Ossification is in mesenchyme from one centre in the perpendicular plate that appears during the eighth week in utero. From this, ossification spreads into all parts. At birth the height of the perpendicular plate equals the width of the horizontal plate, but in adults it is almost twice as great, a change in proportions that accords with those that occur in the maxilla.
MANDIBLE (Fig. 27.28) The mandible is the largest, strongest and lowest bone in the face. It has a horizontally curved body that is convex forwards, and two broad rami, that ascend posteriorly. The body of the mandible supports the mandibular teeth within the alveolar process. The rami bear the coronoid and condylar processes, and the latter articulate with the temporal bones at the temporomandibular joints. Body
The body, is somewhat U-shaped, and has external and internal surfaces separated by upper and lower borders. Anteriorly, the upper external surface
shows an inconstant faint median ridge, which indicates fusion of the halves of the fetal bone at the symphysis menti. Inferiorly this ridge divides to enclose a triangular mental protuberance; its base is centrally depressed but raised on each side as a mental tubercle. The mental protuberance and mental tubercles constitute the chin. The mental foramen, from which the mental nerve and vessels emerge, lies below either the interval between the premolar teeth, or the second premolar tooth. The posterior border of the foramen is smooth, and accommodates the nerve which emerges posterolaterally. A faint external oblique line ascends backwards from each mental tubercle, and sweeps below the foramen: it becomes more marked as it continues into the anterior border of the ramus. UPDATE Date Added: 13 December 2005 Publication Services, Inc. Abstract: Foramina on the internal aspect of the alveolar part of the mandible. Click on the following link to view the abstract: Foramina on the internal aspect of the alveolar part of the mandible. http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=16121325&query_hl=8 Przystanska A, Bruska M: Folia Morphol (Warsz) 64(2):89-91, 2005. The lower border of the body, the base, extends posterolaterally from the mandibular symphysis into the lower border of the ramus behind the third molar tooth. Near the midline, on each side, is a rough digastric fossa that gives attachment to the anterior belly of digastric. Behind the fossa the base is thick and rounded, and has a slight anteroposterior convexity. This changes to a gentle concavity as the ramus is approached, which gives the base a sinuous profile. The upper border, the alveolar part, contains 16 alveoli for the roots of the lower teeth. It consists of buccal and lingual plates of bone joined by interdental and inter-radicular septa. Near the second and third molar teeth the external oblique line is superimposed upon the buccal plate. Like the maxilla, the form and depth of the tooth sockets is related to the morphology of the roots of the mandibular teeth. Usually, the sockets of the incisors, canines and premolar teeth contain a single root, while those for the three molar teeth each contain two roots. The external surface of the alveolus adjacent to the molar teeth gives attachment to buccinator. A number of muscles of facial expression are attached to the lateral surface of the mandible.
page 481 page 482
Figure 27.28 Mandible. A, Lateral view; B, medial view. (By permission from Berkovitz and Moxham, 1994.)
The internal surface of the mandible is divided by an oblique mylohyoid line to which mylohyoid is attached (as are, above its posterior end, the superior pharyngeal constrictor, some retromolar fascicles of buccinator, and the pterygomandibular raphe behind the third molar). This line, which extends from a point a centimetre from the upper border behind the third molar to the mental symphysis, is sharp and distinct near the molars, but faint in front. A slightly concave submandibular fossa related to the submandibular gland lies below the mylohyoid line. The area above the line widens anteriorly into a triangular sublingual fossa related to the sublingual gland. The bone is covered by oral mucosa above the sublingual fossa as far back as the third molar. Above the anterior ends of the mylohyoid lines, the posterior symphyseal aspect bears a small elevation, often divided into upper and lower parts, the mental spines (genial tubercles). The upper part gives attachment to genioglossus, the lower part to geniohyoid. Posteriorly the mylohyoid groove extends downwards and forwards from the ramus below the posterior part the mylohyoid line and contains the mylohyoid nerve and vessels. Superior to the mental spines, most mandibles display a lingual (genial) foramen that opens into a canal which traverses the bone to c.50% of the buccomandibular dimension of the mandible: it contains a branch of the lingual artery. As yet its development is uncertain, although it is a useful radiological landmark (see also accessory mandibular foramina). A rounded torus mandibularis sometimes occurs above the mylohyoid line, medial to the molar roots. Ramus
The mandibular ramus is quadrilateral, and has two surfaces (lateral and medial), four borders (superior, inferior, anterior and posterior) and two processes (coronoid and condylar). The lateral surface is relatively featureless and bears the (external) oblique ridge in its lower part. The medial surface presents, a little above centre, an irregular mandibular foramen, that leads into the mandibular canal. This canal curves downwards and forwards into the body to the mental foramen. Anteromedially the mental foramen is overlapped by a thin, triangular lingula. The mylohyoid groove descends forwards from behind the lingula. The inferior border, continuous with the mandibular base, meets the posterior border at the angle. This is typically everted in males, but in females is frequently inverted. The thin superior border bounds the mandibular incisure, surmounted in front by the somewhat triangular, flat, coronoid process and behind by the condylar process. The posterior border, thick and rounded, extends from the condyle to the angle, being gently convex backwards above, and concave below. The anterior border is thin above where it is continuous with that of the coronoid process, and thicker below where it is continuous with the external oblique line. The temporal crest is a ridge that runs down from the tip of the coronoid process on its medial side to the bone just behind the third molar tooth. The triangular depression between the temporal crest and the anterior border of the ramus is called the retromolar fossa. Coronoid process
The coronoid process projects upwards and slightly forwards as a triangular plate of bone. Its posterior border bounds the mandibular incisure, and its anterior border continues into that of the ramus. The temporal crest is a ridge that runs down from the tip of the coronoid process on its medial side. Condylar process
The mandibular condyle varies considerably both in size and shape. When viewed from above, the condyle is roughly ovoid in outline, the anteroposterior dimension of the condyle (c.1 cm) being approximately half the mediolateral dimension. The medial aspect of the condyle is wider than the lateral. However, the long axis of the condyle is not at right angles to the ramus, but diverges posteriorly from a strictly coronal plane. Thus, the lateral pole of the condyle lies slightly anterior to the medial, and if the long axes of the two condyles are extended, they meet at an obtuse angle (c.145°) at the anterior border of the foramen magnum. The articular head of the condyle joins the ramus through a thin bony projection, the neck of the condyle. A small depression situated on the anterior surface of the neck, below the articular surface, termed the pterygoid fovea, receives part of the attachment of lateral pterygoid. The condyle is composed of a core of cancellous bone covered by a thin layer of compact bone. During the period of growth a layer of hyaline cartilage forms a secondary condylar cartilage and lies immediately beneath the fibrous articulating surface of the condyle. The ramus and its processes provide attachment for the four primary muscles of mastication. Masseter is attached to the lateral surface, medial pterygoid is attached to the medial surface, temporalis is inserted into the coronoid process and lateral pterygoid is attached to the condyle. The sphenomandibular ligament is attached to the lingula. Accessory foramina of the mandible
These are usually unnamed and infrequently described, yet they are numerous. Accessory foramina of the mandible are common. They may transmit auxiliary nerves to the teeth (from facial, mylohyoid, buccal, transverse cervical cutaneous and other nerves), and their occurrence is significant in dental anaesthetic blocking techniques. Ossification
The mandible forms in dense fibromembranous tissue lateral to the inferior alveolar nerve and its incisive branch, and also in the lower parts of Meckel's cartilage. Each half is ossified from a centre that appears near the mental foramen about the sixth week in utero. From this site, ossification spreads medially and posterocranially to form the body and ramus, first below, and then around, the inferior alveolar nerve and its incisive branch. It then spreads upwards, initially forming a trough, and later crypts, for developing teeth. By the tenth week, Meckel's cartilage below the incisor rudiments is surrounded and invaded by bone. Secondary cartilages appear later (Fig. 27.29): a conical mass, the condylar cartilage, extends from the mandibular head downwards and forwards in the ramus, and contributes to its growth in height. Although it is largely replaced by bone by midfetal life, its proximal end persists as proliferating cartilage under the fibrous articular lining until about the third decade. Another secondary cartilage, which soon ossifies, appears along the anterior coronoid border, and disappears before birth. One or two cartilaginous nodules also occur at the symphysis menti. At about the seventh month in utero these may ossify as variable mental ossicles in symphyseal fibrous tissue, and unite to adjacent bone before the end of the first postnatal year. Age changes in the mandible (Fig. 27.30)
page 482 page 483
Figure 27.29 The left half of the mandible of a human embryo, 95 mm long. A, Lateral aspect; B, medial aspect. Blue: cartilage; yellow: bone. (By kind permission from A Low.)
Figure 27.30 The mandible at different ages. (Photographs by Andrew Dyer.)
At birth the two halves of the mandible are united by a fibrous symphysis menti. Anterior ends of both rudiments are covered by cartilage, separated only by a symphysis. Until fusion occurs new cells are added to each cartilage from symphyseal fibrous tissue, and ossification on its mandibular side proceeds towards the midline. When the latter process overtakes the former, and ossification extends into median fibrous tissue, the symphysis fuses. At this stage the body is a mere shell which encloses the imperfectly separated sockets of deciduous teeth. The mandibular canal is near the lower border, and the mental foramen opens below the first deciduous molar and is directed forwards. The coronoid process projects above the condyle. In the first to third postnatal years the two halves join at their symphysis from below upwards, although separation near the alveolar margin may persist into the second year. The body elongates, especially behind the mental foramen, providing space for three additional teeth. During the first and second years, as a chin develops, the mental foramen alters direction from facing forwards to facing backwards, as in adult mandibles, to accommodate a changing direction of the emerging mental nerve. page 483 page 484
In general terms, increase in height of the body of the mandible occurs primarily by formation of alveolar bone associated with the developing and erupting teeth, although some bone is deposited on the lower border. Increase in length of the mandible is accomplished by deposition of bone on the posterior surface of the ramus with compensatory resorption on its anterior surface (accompanied by deposition of bone on the posterior surface of the coronoid process and resorption on the anterior surface of the condylar process). Increase in width of the mandible is produced by deposition of bone on the outer surface of the mandible and resorption on the inner surface. An increase in the comparative size of the ramus compared with the body of the mandible occurs during postnatal growth and tooth eruption. There is some controversy concerning the role of the condylar cartilages in mandibular growth. One view states that continued proliferation of this cartilage is primarily responsible for the increase in both the mandibular length and the height of the ramus. Alternatively, there is experimental evidence which supports the view that proliferation of the condylar cartilage is a response to growth and not its cause. In adults, alveolar and subalveolar regions are about equal in depth, the mental foramen appears midway between the upper and lower borders, and the mandibular canal nearly parallels the mylohyoid line. If the teeth are lost, alveolar bone is resorbed and the mandibular canal and mental foramen come to lie nearer the superior border. Indeed, both may disappear, so that the nerves lie just beneath the oral mucosa.
© 2008 Elsevier
JOINTS The general characteristics of cranial sutures and the detailed anatomy of the temporomandibular joint are described in Chapters 6 and 30, respectively. Sutural bones are described on page 486.
© 2008 Elsevier
NEONATAL, PAEDIATRIC AND SENESCENT ANATOMY THE SKULL AT BIRTH At birth the skull is large in proportion to other skeletal parts, but the facial region is relatively small, and constitutes only about one-eighth of the neonatal cranium, compared with half in adult life. Smallness of the face at birth is due to the rudimentary stage of the mandible and maxillae, non-eruption of the teeth, and the small size of the maxillary sinuses and the nasal cavity. The latter is almost entirely between the orbits, the lower border of the piriform nasal aperture being only slightly below the orbital floors. The large size of the calvaria, especially the cranial vault, is related to precocious cerebral growth. The cranial base is relatively short and narrow and, although the middle and internal auditory parts are almost adult in size, the petrous temporal bones are generally far from adult dimensions. Bones of the cranial vault are unilaminar and without diploë. Frontal and parietal tuberosities are prominent and in norma verticalis the greatest width is between the parietal tuberosities. The glabella, superciliary arches and mastoid processes are not developed. Ossification is incomplete, and many bones are still in several elements united by fibrous tissue or cartilage. The 'os incisivum' is continuous with the maxilla; preand postsphenoids have just united, but the halves of frontal bone and mandible, and the squamous, lateral and basilar parts of the occipital bone are all separate. A second styloid centre (stylohyal) has not appeared and parts of the temporal bones are separate except for the commencing fusion of the tympanic with the petrous and squamous parts. The fibrous membrane, forming the cranial vault before ossification, is unossified at the angles of the parietal bones, leaving six fonticuli (fontanelles), two median (anterior and posterior) and two lateral pairs (sphenoidal and mastoid). The anterior fontanelle (Fig. 27.31), the largest, at the junction of the sagittal, coronal and frontal sutures, and hence rhomboid in shape, is c.4 cm in anteroposterior and 2.5 cm in transverse dimensions. The posterior fontanelle, at the junction of the sagittal and lambdoid sutures, is hence triangular. The sphenoidal (anterolateral) and mastoid (posterolateral) fontanelles (Fig. 27.31) are small, irregular and at the sphenoidal and mastoid angles of the parietal bones. At birth the orbits are large and the germs of the developing teeth are near their orbital floors. Temporal bones differ greatly from their adult form. The internal ear, tympanic cavity, auditory ossicles and mastoid antrum are almost adult in size, the tympanic plate is an incomplete ring, and the mastoid process is absent. Hence the external acoustic meatus is short, straight, unossified and wholly fibrocartilaginous. The external aspect of the tympanic membrane faces down rather than laterally, in accord with the basal cranial contour. The stylomastoid foramen is exposed on the lateral surface of the skull; the styloid process has not fused with the temporal bone; the mandibular fossa is flat and more lateral, and its articular tubercle undeveloped. Paranasal sinuses are rudimentary or absent; only the maxillary sinuses are usually identifiable.
During birth the skull is moulded by slow compression. That part of the scalp which is more central in the birth canal is often temporarily oedematous as a result of interference with venous return, and is called the caput succedaneum. Fontanelles and the width of the sutures allow bones of the cranial vault some overlap. The skull is compressed in one plane with compensatory elongation orthogonal to this. These effects disappear within the first week.
POSTNATAL GROWTH Coordinated postnatal growth of the calvarial and facial skeleton proceeds at different rates and periods, that of the cranial cavity being related to cerebral growth, the facial skeleton to the development of the teeth, muscles of mastication and tongue. Growth of the cranial base is not at the same rate as that of the vault. Therefore the three regions must be considered separately. The anterior part of the cranial base is a zone of interaction between facial and cerebral growth. Growth of the vault
Growth of the vault is rapid during the first year and then slower to the seventh, by which time it has reached almost adult dimensions. For most of this period expansion is largely concentric; form is determined early in the first year, remaining thereafter largely unaltered. That shape of the vault is not directly related to cerebral growth but to genetic factors is supported by the great range of cranial indices and shapes in racial groups. During the first and early second years, growth of the vault is mainly by ossification at apposed margins of bones which possess an osteogenic layer - accompanied by some accretion and absorption of bone at surfaces to adapt to continually altering curvatures. Growth in breadth occurs at the sagittal, sphenofrontal, sphenotemporal and occipitomastoid sutures and petro-occipital cartilaginous joints. Growth in height occurs at the frontozygomatic and squamosal sutures, pterion and asterion. During this period fontanelles are closed by ossification of the bones around them, but separate centres may convert them into sutural bones. The sphenoidal and posterior fontanelles 'fill in' within 2 or 3 months of birth; mastoid fontanelles usually near the end of the first year; and the anterior fontanelle at about the middle of the second, by which time calvarial bones have interlocked at sutures, a process which commences early in the first year. Further expansion is chiefly by accretion and absorption on external and internal surfaces respectively. Meanwhile the bones thicken, although not uniformly. At birth the vault is unilaminar. Tables and intervening diploë appear about the fourth year, with maximal differentiation at about 35 years, when diploic veins are prominent in radiograms. Thickening of the vault and development of external muscular markings reflect the development of the masticatory and neck muscles. The mastoid process is a visible bulge in the second year, and is invaded by air cells in the sixth year. Growth of the base
This is responsible for much of the cranial lengthening, mostly at cartilaginous joints between the sphenoid and ethmoid, and especially between the sphenoid and occipital bones. Largely independent of cerebral growth, it continues at the occipitosphenoid synchondrosis until the eighteenth to twenty-fifth year, a period
prolonged by continued expansion of the jaws to accommodate erupting teeth, and by growth in the muscles of mastication and those of the nasopharynx. However, there is some evidence that growth may cease at about 15 years. A pubertal growth spurt has been ascribed to both sexes, about 2 years earlier in females; considerable postpubertal growth, up to 17.5 years in males, has been described. Multivariate analysis of Cartesian coordinates of cranial landmarks in Polynesian crania suggests that there is considerable independence in the growth and positioning of the segments noted above. Growth of the face page 484 page 485
page 485 page 486
Figure 27.31 Skull of a newborn infant. A, Anterior aspect; B, lateral aspect; C,
superior aspect; D, basal aspect; E, posterior occipital aspect. (Photographs by Kevin Fitzpatrick on behalf of GKT School of Medicine, London.)
Growth of the face occupies a longer period than does that of the calvaria. Much information has been derived from serial radiography. The ethmoid and the orbital and upper nasal cavities have almost completed growth by the seventh year. Orbital and upper nasal growth is achieved by sutural accretion, with deposition of bone on the facial aspects of the margins. The maxilla is carried downwards and forwards by expansion of the orbits and nasal septum and by sutural growth, especially at the fontanelles and zygomaticomaxillary and pterygomaxillary sutures. In the first year, growth in width occurs at the symphysis menti and midpalatal, internasal and frontal sutures; but such growth diminishes or even ceases when the symphysis menti and frontal suture close during the first few years, even though the midpalatal suture persists until mature years. Facial growth in this period continues to puberty and later, linked with the eruption of the permanent teeth. After sutural growth, near the end of the second year, expansion of the facial skeleton is by surface accretion on the face, alveolar processes and palate, and there is resorption in the walls of the maxillary sinuses, the upper surface of the hard palate and the labial aspect of the alveolar process. Co-ordinated growth and divergence of the pterygoid processes is due to deposition and resorption of bone on appropriate surfaces. Mandibular growth is described on page 482. Obliteration of the calvarial sutures progresses with age, commencing between 30 and 40 years internally, and 10 years later on the exterior. Closure times vary greatly. Obliteration usually begins at the bregma, extending into the sagittal, coronal and lambdoid sutures, in that order. In old age the skull becomes thinner and lighter, but occasionally the reverse may occur. The most striking senile feature is diminution in size of the mandible and maxillae following the loss of teeth and absorption of alveolar bone. This reduces the vertical depth of the face and increases the mandibular angles. Sutural bones
Additional ossificatory centres may occur in or near sutures, giving rise to isolated sutural bones (Fig. 27.32). Usually irregular in size and shape, and most frequent in the lambdoid suture, they sometimes occur at fontanelles, especially the posterior. They may represent a pre-interparietal element, a true interparietal, or some composite. An isolated bone at the lambda is sometimes dubbed the Inca bone or Goethe's ossicle. One or more pterion ossicles or epipteric bones may appear between the sphenoidal angle of the parietal and the greater wing of the sphenoid, varying much in size, but more or less symmetrical. Sutural bones usually have little morphological significance. However, they appear in great numbers in hydrocephalic skulls (Figs 27.32, 16.12), and they have therefore been linked with rapid cranial expansion: this is unproven. For a detailed analysis of these and other epigenetic variations in adult crania, consult Berry and Berry (1967) and (Berry 1975). UPDATE Publication Services, Inc.
Date Added: 07 December 2005
Anatomical observations on os inca and associated cranial deformities. Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=16121331&query_hl=12 Anatomical observations on os inca and associated cranial deformities. Das S, Suri R, Kapur V. Folia Morphol (Warsz). 2005 May;64(2):118-21. Craniosynostosis
Sutural growth makes an important contribution to growth of the skull, especially during the first few years of life. The reasons for premature fusion of sutures (synostosis) are varied, but many occur as a result of small brain size or failure of the development of dural bands between the sutures. Premature fusion may occur in one or more of the cranial sutures. When the sutures around the skull base are involved in premature fusion, severe limitation of facial bone growth will occur. Metabolic disorders such as rickets and familial hypophosphatasia can also result in synostosis. Raised intracranial pressure with or without hydrocephalus, visual deterioration and mental retardation may result. Scaphocephaly (sagittal craniosynostosis) is the commonest and leads to lengthening of the vault in an anteroposterior direction. Sagittal craniosynostosis occurs in conjunction with other sutures, e.g. Crouzon's syndrome. Coronal synostosis, either unilateral (plagiocephaly) or bilateral (brachycephaly/oxycephaly) is the next most frequently seen. Premature fusion of the coronal suture results in reduced anteroposterior development with marked supraorbital recession. When it is unilateral the face develops asymmetrically and is rotated away from the side with premature fusion. Metopic craniosynostosis (trigonocephaly) and pansynostosis (turricephaly, where both the coronal and sphenofrontal sutures are involved) are much less common.
Figure 27.32 Lateral (A) and posterior (B) views of the hydrocephalic skull of a 25year old male showing numerous sutural bones. Courtesy of the Museum of the Royal College of Surgeon of England. (Photographer Mr J Carr).
Treatment of these premature sutural fusions is critical to prevent abnormalities of skull growth. It is usually carried out between 3-6 months of age. Early release of the synostosis is also indicated to relieve increased intracranial pressure. Treatment consists of release of the sutures involved and the prevention of refusion. This is usually achieved by covering the bone edges with silastic sheeting after the radical removal of bone from either side of the suture line. Simultaneous expansion and reshaping of the skull is often required, particularly if diagnosis has been delayed. It is usually necessary to advance the supraorbital ridge and to straighten it if there is asymmetrical growth (as seen in plagiocephaly). The craniofacial dysostosis syndromes such as Crouzon's, Apert's, SaethreChotzen, Pfeiffer's and Carpenter's, show a varying degree of calvarial synostoses, which are usually accompanied by a significant lack of growth in the mid-face. Early release of the calvarial synostoses does not result in normal facial growth, and a midfacial osteotomy at the Le Fort III level is usually required later in life. When significant orbital hypertelorism develops, a transcranial bipartitioning procedure is needed in order to bring the two orbits together. Skull deformities can also be derived deliberately by affecting sutural growth using binding and other pressure, as has been practised in certain tribal regions of the world (Fig. 27.33).
CONGENITAL ANOMALIES AFFECTING THE SKULL A large number of malformations and anomalies affect the bones and associated soft tissue structures of the skull. They result from a localized error of morphogenesis during embryological development. Many of these are recognized patterns of malformation which are presumed to have the same aetiology. They do not arise as the result of just one isolated error in morphogenesis, and are described as syndromes. Hemifacial microsomia (Goldenhar syndrome) page 486 page 487
Figure 27.33 Skull binding. Egyptian Copts. (Provided by John Langdon.)
The term hemifacial microsomia is used to describe patients with unilateral microtia, macrostomia and varying degrees of failure of formation of the mandibular ramus and condyle. Vertebral anomalies and epibulbar dermoids are common. Most cases are sporadic, but familial instances have been described. The facies are strikingly asymmetric as a result of hypoplasia or displacement of the ear. The maxilla, temporal and zygomatic bones show varying degrees of hypoplasia on the affected side. The mandible similarly shows varying degrees of hypoplasia ranging from mild asymmetry to major failure of development of the ramus and condyle. The mastoid process also shows degrees of hypoplasia. Often there is frontal bossing. Ten per cent of cases are bilateral, but invariably one side is more severely affected. There is concomitant hypoplasia of the main masticatory muscles on the affected side, and occasionally the muscles of facial expression are involved. In 10% of
cases there is a lower motor neurone weakness of the facial nerve. Mental retardation is unusual. The ear deformities vary from complete aplasia to minor distortions of the pinna, which is displaced anteriorly and inferiorly. Absence of the external auditory meatus is common, as are middle ear deficiencies, which result in conduction deafness. Supernumerary eartags are present and occur anywhere along a line from the tragus to the angle of the mouth. Epibulbar dermoids occur and are usually located at the limbus or lower outer quadrant. A coloboma of the upper lid is present in most cases. Unilateral microphthalmia or anophthalmia can occur and is associated with mental retardation. Vertebral anomalies are common and include occipitalization of the atlas, cuneiform vertebrae, and fusion of several adjacent vertebrae. A variety of cardiac anomalies have been described ranging from ventricular septal defects to Fallot's tetralogy. Pulmonary hypoplasia and renal anomalies have been recorded. Mandibulofacial dysostosis (Treacher Collins syndrome)
Mandibulofacial dysostosis mainly involves structures derived from the first branchial arch, groove and pouch. It is inherited as an autosomal dominant trait with variable penetration. The facial appearance is characteristic. There are downward sloping palpebral fissures, depressed cheek prominences, deformed ears, mandibular hypoplasia and a large fishlike mouth. The hairline often shows a tongue-shaped extension toward the cheek. Clinically the skull vault appears normal, but on imaging it is seen that the supraorbital ridges are poorly developed and despite normal sutural development there may be increased digital markings (copper beating) on the inner table. The zygomatic bones may be totally absent, or more frequently are grossly deficient in a symmetrical manner with failure of fusion of the zygomatic arches. The mastoid processes are not pneumatized and may be sclerotic. The paranasal sinuses are frequently abnormally small or even absent. The infraorbital rims are also poorly developed. There may be various eye anomalies. Colobomas affecting the lateral third of the lower eyelid are present in 75% of cases. Microphthalmia may also be present. The ears are usually severely deformed, have a crumpled appearance and are often wrongly positioned. In a third of patients the external auditory meatus is absent and there may be ossicular defects resulting in conduction deafness. Additional ear tags and blind fistulae may be present anywhere between the tragus and the angle of the mouth. The nasofrontal angle is usually obliterated and the bridge of the nose elevated. The alar cartilages are hypoplastic and choanal atresia may be present. The mandible is almost always hypoplastic, the angle is obtuse and the ramus deficient. The coronoid and condylar processes may also be hypoplastic. A cleft palate is present in 30% of cases and there is a high arched palate. Mental retardation is common. Distraction osteogenesis
The pioneering Russian orthopaedic surgeon Ilizaroff demonstrated that long bones could be lengthened by performing osteotomies in the axial plane and then
slowly separating the two parts of the bone. If the distraction is performed slowly, bone morphogenetic proteins are released and new bone is formed between the sectioned bone ends. When the desired length has been achieved, the long bone is immobilized in its desired position. After a period of a few weeks the initial woven bone is replaced by normal mature bone and the resulting lengthened bone is stable and functional. These techniques are now applied to the bones of the skull. Mandibular distraction is commonly used, particularly when there is asymmetry of the mandible as seen in hemifacial microsomia. By performing the osteotomies at the angles of the mandible, and using carefully adjusted distraction devices, the mandible can be lengthened in the vertical and anteroposterior planes. Distraction is usually performed at one millimetre per day, and the jaw can be lengthened indefinitely until the distraction is stopped and the callus matures and unites. Using complicated frames bolted onto the skull vault, distraction techniques have now been applied to the middle third of the facial skeleton. The technique is particularly useful for the management of the craniofacial syndromes such as Crouzon's, in which the facial bones which articulate with the sphenoid fail to develop normally. If the facial bones can be released at an early age and the mobilized bones distracted, the facial profile can develop normally.
© 2008 Elsevier
SEXUAL AND GEOGRAPHIC VARIATION IN THE SKULL MEASUREMENT OF SHAPE AND SIZE (Fig. 27.34) A. Maximal cranial Summit of glabella to furthest occipital point length B. Maximal cranial Greatest breadth, at right angles to median plane breadth C. Cranial height From basion (median point on anterior rim of foramen magnum) to bregma
Metrical studies are used to compare shapes and sizes of skulls. Frequently, these analyses are accomplished using internationally standard techniques of craniometry in which linear (chord) or surface (arc) distances were measured between a variety of defined cranial and mandibular landmarks (Fig. 27.34). For example, the calvarial part of the skull is measured (usually to the nearest millimetre) as follows: From these three dimensions, three indices are calculated: B/A, C/A and C/B and expressed as percentages. The most frequently used, the breadth/length ratio, is called the cranial index (cephalic index in the living), and has often been used to classify skull types because of its high degree of variability both within and between populations. By convention, skulls with a cranial or cephalic index (CI) below 75.0 were classified as dolichocranic or dolichocephalic; skulls with a CI between 75.0 and 79.9 were classified as mesocranic or mesocephalic; and skulls with a CI greater than 80.0 were classified as brachycranic or brachycephalic. Variations in the CI have little utility in distinguishing skulls from different geographic regions, and mostly reflect interactions between the width of the cranial base and the volume of the brain (Lieberman et al 2000). page 487 page 488
Figure 27.34 The cranial points used, by international agreement, in making linear and certain angular measurements in anthropometry. In all four views (A-D) the skull is in the standard orientation, that is, with the Frankfurt plane as a horizontal.
(a) Total facial
= (nasion-gnathion height/bizygomatic breadth) !100 The nasion is the point where the internasal suture meets the frontal bone. The gnathion is the midpoint of the lower
index (b) Upper facial index (c) Nasal index
mandibular border = (nasion-prosthion length/bizygomatic breadth) !100 The prosthion is the midpoint of the maxillary alveolar rim, between the central incisors. The bizygomatic breadth is the greatest distance measured by trial between zygomatic arches on external aspects = (nasal breadth/nasal height) !100 Breadth is the horizontal maximum across the nasal aperture, and height is from nasion to the mean between the two lowest points on the lower border of the aperture = (maximal orbital height/maximal orbital breadth) !100
(d) Orbital index (e) = (maximal palatal breadth/maximal palatal length) !100 Palatal index (f) = (basion-prosthion/basion-nasion) !100 Gnathic index
page 488 page 489
Other indices in common use are: Similar measurements are frequently taken on mandibles between landmarks including the superoinferior height and anteroposterior thickness of the symphysis; the anteroposterior length of the mandibular body; the anteroposterior length and superoinferior height of the ramus; bigonial width, the bicondylar width; and so on. In addition, a variety of radiographic measurements using landmarks visible in radiographs make it possible to extend classic craniometry to measure certain angles directly, e.g. the gnathic angle (between the basion-nasion and basion-prosthion lines). The cranial base angle is of special interest and represents the orientation of the anterior cranial base relative to the posterior cranial base. The cranial base angle can be measured in many ways. The most common measurement, the angle between the chord from the sella to the foramen caecum relative to the chord between the sella and basion, has an average value in humans of about 135°, which it attains by flexion c.2 years after birth (Lieberman & McCarthy 1999). The cranial base angle is an important determinant of craniofacial form because it influences the position of the face relative to the neurocranium, the protrusion of the mandible relative to the maxilla and the shape of the pharynx. Males Females
: 0.000337 (L-11) (B-11) (H-11) + 406.01 cc : 0.000400 (L-11) (B-11) (H-11) + 206.60 cc
Another important variable, cranial capacity which is correlated to brain volume can be assessed directly by filling the cavity with lead shot, millet seed or other
particulate materials suitable for volumetric measurement when poured out again. Cranial capacity can also be measured using CT scans. In addition, several formulae have also been calculated to estimate cranial capacity from linear measurements of the length, breadth and height of the cranium (in millimetres). Examples are: In these formulae, L and B are length and breadth, and H is the auricular height, measured to the vertex from the external acoustic meatus. Such methods involve some inaccuracy. Variations in skull shape and form have most commonly been analysed using multivariate statistical methods using many measurements taken on large sets of individuals. Radiographic and three-dimensional imaging technology, especially computer tomography (CT) and magnetic resonance imaging (MRI), in combination with new methods of analysing shape from three-dimensional landmarks have revolutionized the ability to study the shapes of skulls by including information about the relative position between multiple landmarks in multivariate space. Several analytical techniques are most commonly used, including Euclidean distance matrix analysis (EDMA), which examines variations in size and shape (hence form, which is shape corrected by size) by comparing matrices of all the interlandmark distances between individuals or sets of individuals (Lele & Richtsmeier 1993). Another set of techniques, termed geometric morphometrics, analyses the relative spatial arrangements of landmarks between individuals or sets of individuals, allowing the full threedimensionality of form differences to be modelled, compared and visualized.
SEXUAL VARIATION Until puberty there is little sexual difference in skulls. Adult males tend to be larger than females in a number of features due to a combination of faster rates of growth during puberty and longer period of growth. However, ranges of variation between sexes overlap considerably. In general, the adult male cranium has c.11% larger cranial capacity than females, mostly reflecting larger male body mass. In terms of shape differences, the male cranium tends to have thicker bones in the neurocranial vault; and more marked muscle origins and insertions, e.g. the temporal and nuchal lines; the frontal sinuses are larger, as are the glabella and the superciliary arches; the external occipital protuberance and the mastoid processes are more prominent; the superior margin of the orbits tends to be squarer; and the mandibular and maxillary arches are larger, in part due to larger tooth size. In addition, the male mandible tends to have larger coronoid processes; larger, more flared, gonial regions; longer rami; and a more pronounced mental eminence. Because variation is greater within than between sexes, diagnosis of sex is difficult or impossible for many crania, and is most accurately assessed using multivariate statistical techniques such as discriminant function analysis.
GEOGRAPHIC VARIATION Several major studies have assessed variation in cranial shape among and between populations (Howells 1973; Lahr 1996). As with the genotype, variation in human cranial shape is far greater within than between populations (Relethford 1994). Nonetheless some tendencies are evident when comparing average
cranial shapes from populations of different geographic origin. Summarizing these studies, there is more variation in Africa than elsewhere, with marked differences between Bushmen, Bantu, and other groups. On average, African crania are broader, with taller upper faces, more inferiorly positioned nasal regions, and more prognathic mandibular and maxillary arches than crania from other parts of the globe. European skulls tend to be narrow, with concomitantly narrow faces, retracted zygomatic arches, tall nasal regions, and prominent midfaces. Europeans and American Indians share many cranial similarities. Asian skulls are typically wide (brachycephalic), with wide faces, a high degree of facial flatness, and flat supranasal regions. Australian aborigines are often characterized by narrow skulls (dolichocephaly), and large, low projecting faces with prominent subnasal regions. Attempts to differentiate crania by region of geographic origin using multivariate methods such as discriminant function analysis can have accuracies of over 90%.
© 2008 Elsevier
FRACTURES OF THE FACIAL SKELETON Fractures affecting the jaw bones are common. They result from road traffic accidents, sports injuries, accidents at work and increasingly as a result of interpersonal violence. Many of these injuries are sustained by those intoxicated with alcohol or are as a result of an assailant who is drunk. Given that the majority of people are right-handed, most jaw fractures resulting from assault affect either the left cheek bone (zygomaticomaxillary complex) or the left angle of the mandible. Because of the shape and structure of the jaw bones, skull fractures tend to adopt well-recognized patterns, which are determined by the concentration of stresses and lines of weakness in the bones affected. Although in severe injuries the fractures are often complex, it is convenient to describe them as arising in the upper, middle and lower thirds of the face. Often a subject will have sustained fractures involving more than one of these areas.
UPPER THIRD OF FACE (NASOETHMOIDAL COMPLEX) Fractures in the upper third of the face are almost invariably comminuted and involve many bones. The skeletal foundation of the nasoethmoidal complex consists of a strong triangular-shaped frame. However, all these structures are fragile and any force sufficient to fracture the frame results in severe comminution and displacement. A severe impact delivered to the midface, particularly over the bridge of the nose, may result in these structures being driven backwards between the orbits. This may result in traumatic hypertelorism, producing an increase in distance between the pupils. Associated displacement of the medial canthal ligaments results in traumatic telecanthus. Comminution of the cribriform plates of the ethmoid results in dural tears and cerebrospinal rhinorrhoea. Often these fractures are combined with more extensive fractures of the frontal bone. Such fractures involve the orbital roofs and, if displaced significantly, will in turn displace the globe of the eye. Fractures that involve both the inner and outer walls of the frontal sinus carry a risk of both early and delayed intracranial infection, and often it is necessary to obliterate the frontal sinuses in order to prevent this complication.
MIDDLE THIRD OF THE FACE page 489 page 490
The middle third of the face is defined as that area bounded above by a transverse line connecting the two zygomaticofrontal sutures, passing through the frontomaxillary and frontonasal sutures, and limited below by the occlusal plane of the maxillary teeth. Posteriorly the region is limited by the sphenoethmoidal junction, but it includes the free margins of the pterygoid plates inferiorly. Fractures of the middle third of the facial skeleton may involve the two maxillae, the two palatine bones, the two zygomatic bones, the zygomatic processes of the temporal bones, the two nasal bones, the vomer, the ethmoid bone together with its nasal conchae and the body and greater and lesser wings of the sphenoid bone. They are divided into those involving the central block and those involving the lateral middle thirds. Central middle third of the face
The majority of the skeleton of the central middle third is composed of wafer thin
sheets of cortical bone with stronger reinforcements, i.e. the palate and alveolar process; the lateral rim of the piriform aperture extending upwards (via the canine fossa) to the medial orbital rim, and finally to the glabella; the zygomatic buttress and its connections to the inferior and lateral orbital margins and the zygomatic arch; the orbital rims and the pterygoid plates. The strength lies in the facial surface of the skeleton which, although thin in most areas, is cross-braced. The design is ideally suited to transmit occlusal forces vertically to the skull base. The fracture lines of the central middle third are customarily described with respect to the Le Fort lines. Although comminuted fractures are common in this region, the various fracture lines do follow the Le Fort lines. Le Fort I fractures (Guerin's fracture) Le Fort I fractures consist of a horizontal fracture line above the level of the floor of the nose involving the lower third of the nasal septum. The mobile segment consists of the palate, the alveolar process and the lower thirds of the pterygoid plates. UPDATE Date Added: 23 November 2005 Publication Services, Inc. Update: Location of the descending palatine artery in relation to the Le Fort I osteotomy. One of the risks of the Le Fort I osteotomy (a maxillary procedure for the correction of dentofacial deformity) is that a major hemorrhage will form as a result of injury to the descending palatine artery. This artery lies within the greater palatine canal, located in the perpendicular portion of the palatine bone. The estimated blood loss in a routine Le Fort I osteotomy ranges from 100 ml to 1,500 ml, which contributes to postoperative fatigue syndrome and the complications associated with autologous blood transfusion. In order to minimize injury to the descending palatine artery and blood loss, Li and colleagues (1996) evaluated the position of descending palatine artery as it relates to Le Fort I osteotomy in 3 separate studies. In the first study, they measured the greater palatine canal and the foramen in relation to maxillary landmarks pertaining to Le Fort I osteotomy in 30 human skulls. In the second study, the computed tomography (CT) of 40 selected patients with normal or minimal sinus mucosal thickening was analyzed and the distance from the greater palatine canal to the piriform rim was measured. In the third study, 8 fresh cadavers were used to measure the distance from the internal maxillary artery to the nasal floor. These measurements showed that the internal maxillary artery enters the pterygopalatine fossa approximately 16.6 mm above the nasal flow and gives off the descending palatine artery. The descending palatine artery travels a short distance within the pterygopalatine fossa and then enters the greater palatine canal. It travels approximately 10 mm within the canal in an inferior, anterior, and slightly medial direction to exit the greater palatine foramen in the region of the second and third molars. Li et al. suggest that the following guidelines will help during Le Fort I osteotomy to minimize injury to the descending palatine artery: 1. Osteotomy of the lateral wall of the maxillary sinus should be extended just beyond the second molar. 2. Osteotomy of the medial wall of the maxillary sinus should usually extend 30 mm posterior to the piriform rim in females. In males, it can be carried back to 35 mm. 3. Pterygomaxillary separation should be made along the pterygomaxillary fissure with either a curved osteotome or a right-angled oscillating saw. Because the descending palatine artery travels in an anterior-inferior direction as it enters the
descending palatine artery travels in an anterior-inferior direction as it enters the greater palatine canal, injury can be prevented by closely adapting the cutting edge of the curved osteotome or right-angled saw to the pterygomaxillary fissure, while avoiding excessive anterior angulation. Furthermore, the superior cutting edge of the osteotome or saw blade should be less than 10 mm above the nasal floor. If damage occurs to the descending palatine artery after following these guidelines, the posterior wall of the maxillary sinus can be removed at least 10 mm to 15 mm above the nasal floor to achieve proximal control of the vessel without significant risk of injury to the internal maxillary artery. Li KK, Meara JG, Alexander A: Location of the descending palatine artery in relation to the Le Fort I osteotomy. J Oral Maxillofac Surg 54:822-825, 1996. Medline Similar articles
Le Fort II fractures (pyramidal fracture) Le Fort II fractures are pyramidal fractures involving the maxillary bones. From the nasal bridge, the fracture enters the medial wall of the orbit to involve the lacrimal bone and then crosses the inferior orbital rim, usually at the junction of the medial and lateral two-thirds, and often involves the infraorbital foramen. The fracture line then runs beneath the zygomaticomaxillary suture, traversing the lateral wall of the maxillary sinus to extend posteriorly horizontally across the pterygoid plates. The zygomatic bones and arches remain attached to the skull base. Le Fort III fractures Le Fort III fractures run parallel with the base of the skull, separating the entire mid-facial skeleton from the cranial base. The fracture extends through the nasal base and continues posteriorly across the ethmoid bone. The fracture then crosses the lesser wing of the sphenoid and, on occasion, involves the optic foramen. Usually, however, it slopes down medially passing below the optic foramen to reach the pterygomaxillary fissure and pterygopalatine fossa. From the base of the inferior orbital fissure the fracture runs laterally and upwards, separating the greater wing of the sphenoid from the zygomatic bone, to reach the frontozygomatic suture. It also extends downwards and backwards across the pterygopalatine fossa to involve the root of the pterygoid plates. The zygomatic arch is usually fractured at the zygomaticotemporal suture. Frequently these fractures do not occur as bilateral symmetrical fractures but occur in various combinations, e.g. both together, on the same side, and involving both sides. Typically these fractures arise from force applied anteriorly over a wide area. Such injuries are seen in road traffic accidents where, e.g. a driver or passenger is thrown forwards on to the steering wheel or dashboard. The direction of the applied force determines the displacement of these fractures. With the possible exception of the relatively weak lateral pterygoids, muscle pull plays a relatively small part. As the fractures are generally displaced backwards, because of the angulation of the strong skull base, there is also a downward component, which results clinically in a lengthening of the face and a dished-in appearance. There may be airway obstruction if this downwards and backwards displacement is severe. Lateral middle third of the face
Fractures of the lateral middle third involve the zygomaticomaxillary complex. The zygomatic bone forms the prominence of the cheek and so is subject to direct trauma to the side of the face. As the most common cause of a zygomatic fracture is a blow from a fist, depressed fractures of the zygomaticomaxillary complex are a common injury. There is separation at both the zygomaticofrontal and zygomaticotemporal sutures. The major damage is at the lateral wall of the
maxilla, which is usually comminuted. Displacement of the zygomatic bone into the lateral wall of the maxilla leads to damage to the infraorbital nerve. Isolated fractures of the zygomatic arch are relatively unusual. Orbital fractures The orbit is invariably involved in depressed fractures of the zygomatic bone and in Le Fort II and III fractures. The orbit is also involved in fractures of the frontal bone and in extensive nasal complex injuries. A fracture of the orbital floor without associated rim involvement is known as a 'blow-out' fracture. Fortunately the optic foramen which is situated within the lesser wing of the sphenoid bone is surrounded by dense bone and is only rarely involved in fractures. Direct injury to the optic nerve is therefore unusual.
LOWER THIRD OF FACE (MANDIBLE) The bone of the lower third of the face is the mandible, which is essentially a tubular bone bent into a blunt V-shape. This basic configuration is modified by sites of muscle attachments, notably masseter and medial pterygoid around the angle, and temporalis around the coronoid process. The presence of teeth, particularly those with long roots such as the canines, or unerupted teeth, produces lines of weakness in the mandible. When the teeth are lost, or fail to develop, the subsequent progressive resorption of the alveolar bone means that the mandible reverts to its underlying tubular structure. Like all tubular bones, the strength of the mandible resides in a dense cortical plate, thickened anteriorly and at the lower border of the mandible. It follows that the mandible is strongest anteriorly in the midline and is progressively weaker posteriorly towards the condylar processes. Again, like all tubular bones, the mandible has great resistance to compressive forces, but fractures at sites of tensile strain. The mandible is liable to particular patterns of distribution of tensile strain when forces are applied to it. Anterior forces applied to the mental symphysis, or over the body of the mandible, lead to strain at the condylar necks and also along the lingual cortical plates on the contralateral side in the molar region. In order of frequency, fractures occur most commonly at the neck of the condyle, the angle, the parasymphyseal region and the body of the mandible. Most often the mandible fractures occur at two of these sites: isolated fractures are relatively unusual. Condylar process The condyle is protected from direct injury by the zygomatic arches. Fractures occur usually by the transmission of force following a blow to the front of the mandible or to the contralateral body. Except in children most condylar fractures are not intracapsular, and occur in the neck. They typically run obliquely downwards and backwards from the mandibular notch. The condyle is usually displaced anteromedially (because of the attachment of lateral pterygoid to the temporomandibular joint disc, capsule and anterior border of the neck of the condyle). Angle of the mandible The majority of fractures at the angle of the mandible run vertically downwards and backwards from the alveolar bone to the angle. When a third molar tooth is present the fracture line will pass through its socket. The presence of the tooth results in a line of weakness. A fracture at the angle prevents the powerful elevator muscles (masseter, medial pterygoid and temporalis) from having any direct effect on the tooth-bearing part of the jaw. Thus, the posterior fragment is typically displaced upwards, forwards and inwards as a result of the unopposed pull of these powerful muscles.
Ramus and coronoid process Fractures at the ramus exhibit very little displacement due to the splinting activity of medial pterygoid medially and masseter laterally. These two bulky muscles are widely attached to the ramus and their attachments extend across the fracture lines. Similarly the coronoid process is rarely significantly displaced because it is splinted by the tendinous insertion of temporalis. Body of the mandible Most fractures of the body of the mandible occur as the result of direct trauma and tend to be concentrated in the first molar or canine region. The further forward the site of the fracture, the more the upward displacement of the elevators is counteracted by the downward pull of geniohyoid and the anterior belly of digastric. When teeth are present displacement is limited by the dental occlusion, since further displacement is resisted by the lower and upper teeth. Displacement may be considerable in the edentulous patient.
CRANIAL BASE page 490 page 491
The cranial base - composed of the frontal, ethmoid, sphenoid and occipital bones - is a relatively solid platform inclined at an angle of 45° to the maxillary occlusal plane. Fractures of the cranial base are not readily visible on normal radiographs. They result in bleeding in the floor of the middle cranial fossa. This often presents as bruising over the mastoid process and is known as Battle's sign. Such fractures also result in escape of cerebrospinal fluid which may be seen leaking from a ruptured tympanic membrane. Alternatively, if this membrane remains intact it will be seen as blue and bulging, possibly with a transmitted pulsation.
© 2008 Elsevier
SKELETAL ACCESS SURGERY The craniofacial skeleton has an excellent blood supply, and so can be dismantled as a series of osteoplastic flaps. The surgical disarticulation of the craniofacial skeleton has been used to gain access to otherwise inaccessible sites in order to allow the surgeon to attend to pathology in the skull base, cervical spine and anterior and posterior cranial fossae. The aim is to provide increased and more direct exposure of both the pathology and the adjacent vital structures without the need to resect uninvolved structures. The craniofacial skeleton can be divided into a series of modular osteotomies, which permit both independent and conjoined mobilization. The zygomatic and nasal bones and the maxilla may be exposed and mobilized, pedicled on the overlying soft tissues either unilaterally or bilaterally. These approaches improve access to the nasal cavity, maxillary, ethmoid and sphenoid sinuses, the soft palate and nasopharynx, and the infratemporal fossa and pharyngeal space. The exposures may be extended to gain access to the anterior and middle cranial fossae, cavernous sinus, clivus, craniocervical junction and upper cervical vertebrae. A variety of different access osteotomies have been described and found to be useful in specific clinical situations. Most of the osteotomies described follow the conventional patterns of facial fractures described above. The entire hemimaxilla and zygoma can be mobilized, and pedicled on the soft tissues of the face by making bone cuts that follow the lines of a Le Fort II fracture on one side. The osteotomy is completed by dividing the upper alveolus and palate just to the side of the nasal septum and perpendicular plate of the vomer. The maxilla may be mobilized at the Le Fort I level and downfractured, pedicled on the palatoglossal muscles and soft tissue attachments. This gives good access to the nasopharynx, clivus and upper cervical spine, particularly if the palate is divided in the midline. Lateral zygomatic osteotomies may be performed to gain access to the orbital apex and infratemporal fossa. The surgical approach is from behind using a hemior bicoronal flap. The zygomatic complex is mobilized inferiorly pedicled on masseter. When combined with a mandibular ramus osteotomy, access is gained to the retromaxillary area and pterygoid space as well as to the infratemporal fossa. In combination with a frontotemporal craniotomy, the zygomatic osteotomy has been used for access to the middle cranial fossa, cavernous sinus, apex of the petrous temporal bone and the interpeduncular cistern. Dividing the lower lip in the midline, and dividing the mandible either in the midline or just in front of the mental foramen, allows the hemimandible to be swung laterally. The technique is used to give improved access to the floor of the mouth, the base of the tongue, tonsillar fossa, soft palate, oropharynx, posterior pharyngeal wall, supraglottic larynx and pterygomandibular region. By extending the dissection laterally access is gained to the pterygoid space, infratemporal fossa and parapharyngeal space. By dissecting more medially access is gained to the nasopharynx, lower part of the clivus and all seven of the cervical
vertebrae. A modification of the mandibular swing procedure increases access up to the skull base, by combining the classic mandibular swing with a horizontal osteotomy of the mandibular ramus above the level of the lingula. REFERENCES Berkovitz BKB, Moxham BJ 1989 Colour atlas of the skull. London: Mosby-Wolfe Berkovitz BKB, Moxham BJ 1994 Color atlas of the skull. London: Mosby-Wolfe Berry AC 1975 Factors affecting the incidence of non-metrical skeletal variants. J Anat 120: 519-35. Medline Similar articles Berry AC, Berry RJ 1967 Epigenetic variation in the human cranium. J Anat 101: 361-80. Medline Similar articles Howells WW 1973 Cranial Variation in Man. Cambridge MA: Papers of the Peabody Museum of Archaeology and Ethnography, vol 67. Lahr MM 1996 The Evolution of Modern Human Diversity: A Study of Cranial Variation. Cambridge: Cambridge University Press. Lele S, Richtsmeier JT 1991 Euclidean distance matrix analysis: a coordinate-free approach for comparing biological shapes using landmark data. Am J Phys Anthropol 86: 415-27. Medline Similar articles Full article Lieberman DE, McCarthy RC 1999 The ontogeny of cranial base angulation in humans and chimpanzees and its implications for reconstructing pharyngeal dimensions. J Hum Evol 36:487-517. Medline Similar articles Full article Lieberman DE, Pearson OM, Mowbray KM 2000 Basicranial influence on overall cranial shape. J Hum Evol 38(2): 291-315. Medline Similar articles Full article Moos KF, Baker AW 1998 Craniofacial surgery. Assessment and techniques. In: Langdon JD, Patel MF (eds) Operative Maxillofacial Surgery. London: Chapman & Hall, 407-436. A detailed description of the various craniosynostoses. Relethford JH. 1994 Craniometric variation among modern human populations. Am J Phys Anthropol 95(1): 53-62. Medline Similar articles Full article Sperber GH 2001 Craniofacial Development, 4th edn. Edinburgh: Churchill Livingstone. Vidarsdottir US, O'Higgins P, Stringer C 2002 A geometric morphometric study of regional differences in the ontogeny of the modern human facial skeleton. J Anat 201: 211-29. Medline Similar articles Full article Williams JL 1994 Rowe and Williams' Maxillofacial Injuries. Edinburgh: Churchill Livingstone. page 491 page 492
© 2008 Elsevier
28 HEAD Development of the skull BASAL REGIONS OF THE SKULL PATTERNING OF THE BASAL REGION OF THE SKULL The skull base forms entirely by endochondral ossification. The pattern of the initial chondrogenic anlagen is closely related to that of the brain and olfactory epithelium and may involve an epithelial-mesenchymal interaction between these structures, which transiently express type II collagen, and the adjacentskeletogenic mesenchyme (Thorogood, 1988). This intimate relationship between brain and skull during development means that evolutionary changes in the brain and skull are co-ordinated. After neural expression of type II collagen has ceased, chondrogenesis is initiated in the mesenchyme. The skeletogenic mesenchyme of the skull base is of neural crest origin rostral to the tip of the notochord and of mesodermal origin more caudally, i.e. in the notochordal region.
CARTILAGINOUS NEUROCRANIUM Differentiation of the endochondral component of the skull is initiated in several centres: in the prechordal region, rostral to the notochord; in the parachordal region, from the caudal end of the rhombencephalon to the tip of the notochord (caudal to the hypophysis); and in sense capsules, around the olfactory pits and otocyst. In mammals, a complete optic cartilaginous capsule does not differentiate, but a partial capsule is represented by the orbital wing of the sphenoid. The initial sites of chondrogenesis fuse to form a continuous cartilaginous framework prior to endochondral ossification Cranial chondrification begins in the second month. Cartilaginous foci appear in the occipital plate, one on each side of the notochord (parachordal cartilages), and fuse at the end of the seventh week to surround the notochord. The otic capsules, presphenoid, bases of the greater and lesser wings of the sphenoid, and the nasal capsules, become chondrified in sequence (Figs 26.6, 28.1). Prechordal region
The rostral neural crest mesenchyme on a level with, and in front of, the hypophysis develops two pairs of centres of chondrogenic differentiation. The front pair form the trabeculae cranii which fuse to form the trabecular cartilage. The other pair form the polar hypophyseal cartilages on each side of the craniopharyngeal duct (remnant of Rathke's pouch) which leads to the hypophysis. They unite at first behind, then in front, to enclose the craniopharyngeal canal which contains the hypophyseal diverticulum. The canal is usually obliterated by the third month. The hypophyseal cartilage derives from both paraxial mesenchyme and neural crest. The neural crest forms the more rostral portion of the sella turcica, whereas the paraxial mesenchyme contributes to the caudal part, and forms each side of the rostral end of the notochord (Couly et al 1993).
Two chondrogenic centres for the sphenoid appear lateral to the trabecular cartilage. The orbitosphenoid forms part of the back of the orbit and lesser wing of the sphenoid. Medial processes extend around the optic nerve and fuse with the trabecular cartilage to form the optic canal. The alisphenoid is separated from the orbitosphenoid by the oculomotor, trochlear and abducens cranial nerves and by the first and second divisions of the trigeminal nerve. Posteriorly the alisphenoid is separated from the otic capsule by the mandibular division of the trigeminal nerve and the internal carotid artery. The mandibular nerve becomes surrounded by cartilage to form the foramen ovale. A large portion of the alisphenoid forms the greater wing of the sphenoid by membranous ossification. Parachordal region
Immediately caudal to the hypophysis, the unsegmented paraxial mesenchyme gives rise to a sclerotomal component. This condenses to form an unpaired, plate-like parachordal cartilage which lies at first between the notochord and brain stem. The region of fusion of the hypophyseal and parachordal cartilages corresponds to the sphenooccipital synchondrosis, which is a site of growth until up to 20 years of age. More caudally, the segmented portion of paraxial mesenchyme gives rise to four occipital somites. Although the most rostral is rudimentary and does not form a clear sclerotome, the next three occipital somites produce sclerotomes which fuse rostrally with the parachordal cartilage and caudally with each other to form the clivus. The hypoglossal canal forms between the lower occipital somites. In the region of the foramen magnum the occipital sclerotomes extend dorsally to enclose the neural tube like the neural arch of a vertebra. Sense capsules
Each otic vesicle is surrounded by skeletogenic mesenchyme which becomes chondrified to form the otic capsule (Chapter 40). Each capsule lies lateral to the parachordal cartilage and fuses with its lateral margin, except caudally, where fusion is incomplete at the site of the jugular foramen. The capsule is pierced medially by the internal auditory meatus. A cartilaginous nasal capsule develops within the frontonasal mesenchyme around each olfactory pit (Chapter 34). These capsules unite with each other and with the trabecular cartilage. The whole nasal capsule is well-developed by the end of the third month, and consists of a common median septal part - sometimes initially termed the interorbitonasal septum - and two lateral regions. The free caudal borders of the lateral regions incurve to form the interior nasal conchae: these ossify during the fifth month and become separate elements. Posteriorly, each lateral part of the nasal capsule becomes ossified as the ethmoidal labyrinth. This bears ridges on its medial surface which will become the middle and superior conchae. Part of the rest of the capsule remains cartilaginous as the septal and alar cartilages of the nose, and part is replaced by the mesenchymatous vomer and nasal bones.
OSSIFICATION OF THE BASE OF THE SKULL Ossification starts before the chondrocranium has fully developed, and bone
replaces cartilage until little of the chondrocranium remains. However, at birth unossified chondrocranium persists in the alae, lateral nasal cartilage and septum of the nose; the spheno-ethmoidal junction; the spheno-occipital and sphenopetrous junctions; the apex of the petrous bone (foramen lacerum), and also between ossifying elements of the sphenoid bone and between elements of the occipital bone. Most of these regions function as growth cartilages and are termed synchondroses. Small areas of unossified cartilage remain in the adult skull.
© 2008 Elsevier
BASAL REGIONS OF THE SKULL PATTERNING OF THE BASAL REGION OF THE SKULL The skull base forms entirely by endochondral ossification. The pattern of the initial chondrogenic anlagen is closely related to that of the brain and olfactory epithelium and may involve an epithelial-mesenchymal interaction between these structures, which transiently express type II collagen, and the adjacentskeletogenic mesenchyme (Thorogood, 1988). This intimate relationship between brain and skull during development means that evolutionary changes in the brain and skull are co-ordinated. After neural expression of type II collagen has ceased, chondrogenesis is initiated in the mesenchyme. The skeletogenic mesenchyme of the skull base is of neural crest origin rostral to the tip of the notochord and of mesodermal origin more caudally, i.e. in the notochordal region.
CARTILAGINOUS NEUROCRANIUM Differentiation of the endochondral component of the skull is initiated in several centres: in the prechordal region, rostral to the notochord; in the parachordal region, from the caudal end of the rhombencephalon to the tip of the notochord (caudal to the hypophysis); and in sense capsules, around the olfactory pits and otocyst. In mammals, a complete optic cartilaginous capsule does not differentiate, but a partial capsule is represented by the orbital wing of the sphenoid. The initial sites of chondrogenesis fuse to form a continuous cartilaginous framework prior to endochondral ossification Cranial chondrification begins in the second month. Cartilaginous foci appear in the occipital plate, one on each side of the notochord (parachordal cartilages), and fuse at the end of the seventh week to surround the notochord. The otic capsules, presphenoid, bases of the greater and lesser wings of the sphenoid, and the nasal capsules, become chondrified in sequence (Figs 26.6, 28.1). Prechordal region
The rostral neural crest mesenchyme on a level with, and in front of, the hypophysis develops two pairs of centres of chondrogenic differentiation. The front pair form the trabeculae cranii which fuse to form the trabecular cartilage. The other pair form the polar hypophyseal cartilages on each side of the craniopharyngeal duct (remnant of Rathke's pouch) which leads to the hypophysis. They unite at first behind, then in front, to enclose the craniopharyngeal canal which contains the hypophyseal diverticulum. The canal is usually obliterated by the third month. The hypophyseal cartilage derives from both paraxial mesenchyme and neural crest. The neural crest forms the more rostral portion of the sella turcica, whereas the paraxial mesenchyme contributes to the caudal part, and forms each side of the rostral end of the notochord (Couly et al 1993). Two chondrogenic centres for the sphenoid appear lateral to the trabecular cartilage. The orbitosphenoid forms part of the back of the orbit and lesser wing of the sphenoid. Medial processes extend around the optic nerve and fuse with the trabecular cartilage to form the optic canal. The alisphenoid is separated from the
orbitosphenoid by the oculomotor, trochlear and abducens cranial nerves and by the first and second divisions of the trigeminal nerve. Posteriorly the alisphenoid is separated from the otic capsule by the mandibular division of the trigeminal nerve and the internal carotid artery. The mandibular nerve becomes surrounded by cartilage to form the foramen ovale. A large portion of the alisphenoid forms the greater wing of the sphenoid by membranous ossification. Parachordal region
Immediately caudal to the hypophysis, the unsegmented paraxial mesenchyme gives rise to a sclerotomal component. This condenses to form an unpaired, plate-like parachordal cartilage which lies at first between the notochord and brain stem. The region of fusion of the hypophyseal and parachordal cartilages corresponds to the sphenooccipital synchondrosis, which is a site of growth until up to 20 years of age. More caudally, the segmented portion of paraxial mesenchyme gives rise to four occipital somites. Although the most rostral is rudimentary and does not form a clear sclerotome, the next three occipital somites produce sclerotomes which fuse rostrally with the parachordal cartilage and caudally with each other to form the clivus. The hypoglossal canal forms between the lower occipital somites. In the region of the foramen magnum the occipital sclerotomes extend dorsally to enclose the neural tube like the neural arch of a vertebra. Sense capsules
Each otic vesicle is surrounded by skeletogenic mesenchyme which becomes chondrified to form the otic capsule (Chapter 40). Each capsule lies lateral to the parachordal cartilage and fuses with its lateral margin, except caudally, where fusion is incomplete at the site of the jugular foramen. The capsule is pierced medially by the internal auditory meatus. A cartilaginous nasal capsule develops within the frontonasal mesenchyme around each olfactory pit (Chapter 34). These capsules unite with each other and with the trabecular cartilage. The whole nasal capsule is well-developed by the end of the third month, and consists of a common median septal part - sometimes initially termed the interorbitonasal septum - and two lateral regions. The free caudal borders of the lateral regions incurve to form the interior nasal conchae: these ossify during the fifth month and become separate elements. Posteriorly, each lateral part of the nasal capsule becomes ossified as the ethmoidal labyrinth. This bears ridges on its medial surface which will become the middle and superior conchae. Part of the rest of the capsule remains cartilaginous as the septal and alar cartilages of the nose, and part is replaced by the mesenchymatous vomer and nasal bones.
OSSIFICATION OF THE BASE OF THE SKULL Ossification starts before the chondrocranium has fully developed, and bone replaces cartilage until little of the chondrocranium remains. However, at birth unossified chondrocranium persists in the alae, lateral nasal cartilage and septum of the nose; the spheno-ethmoidal junction; the spheno-occipital and sphenopetrous junctions; the apex of the petrous bone (foramen lacerum), and
also between ossifying elements of the sphenoid bone and between elements of the occipital bone. Most of these regions function as growth cartilages and are termed synchondroses. Small areas of unossified cartilage remain in the adult skull.
© 2008 Elsevier
VAULT OR UPPER REGIONS OF THE SKULL page 493 page 494
Figure 28.1 Representative stages in the development of the cranium. In all the diagrams the chondrocranium and cartilaginous stages of vertebrae are shown in blue, except where ossification is occurring and here the colour is green. The desmocranium, consisting of elements ossifying directly in mesenchyme, is shown in yellow. Cranial nerves are indicated by the appropriate Roman numeral. A, Sagittal section through the cranial end of the developing axial skeleton in an early human embryo of c.10 mm, showing the extent of the notochord. B, Superior aspect of cranium of human embryo at 40 mm. C, Lateral aspect of B. D, Lateral aspect of cranium of human embryo at 80 mm.
The vault of the neurocranium is formed entirely by intramembranous ossification and its elements are frequently described as dermal bones. They are the frontal and parietal bones, the squamous part of the temporal bones and the upper part (interparietal) of the occipital bone. The frontal and squamous temporal bones are of neural crest origin and the parietals are of mesodermal origin; the interparietal is mixed (Jiang et al., 2002). The coronal suture thus forms at the neural crestmesoderm interface, as does the sagittal suture, due to a small tongue of neural crest tissue lying between the two developing bones. These tissue interfaces may be significant for initiating the signaling system that governs growth of the skull vault. The bones of the vault of the skull first appear at about day 30. They consist of curved plates of mesenchyme at the sides of the skull which gradually extend towards each other, and towards the cartilaginous base of the skull (Fig. 28.1). The dermal bones are formed by the initiation of a wave of osteodifferentiation which extends radially from ossification centres within the desmocranial (skeletogenic) mesenchyme. When the paired bones meet in the midline, metopic and sagittal suture formation is induced. In contrast, the coronal suture, between the frontal and parietal bones, is present from the onset of ossification. Once sutures have been established and the fibrous desmocranium has been replaced by mineralized bone, growth continues within the sutural growth centres until growth of the brain is complete. page 494 page 495
There is a close association between the developing meninges, particularly the dura mater, and the calvarial bones. Transplants of sutures in which the fetal dura mater is left intact result in a continuous fibrous suture between developing vault bones, whereas in transplants in which the fetal dura mater is removed, bony fusion occurs. This interaction of underlying dura mater with the developing calvarial bones has been demonstrated experimentally in the rabbit, showing that
the dura not only promotes the position and maintenance of sutures, but also that dura can re-pattern both the appearance and position of the bones and sutures of the cranial vault after removal of the calvaria in the neonate (Opperman et al 1993). At the site of a developing suture the osteogenic fronts of two adjacent bones meet and overlap. Initially there is a highly cellular suture blastema between the bones which later becomes more dense and acellular. In the mature suture there is a narrow overlap of compact bone which contains a dense, narrow band of cells continuous with the periosteum.
© 2008 Elsevier
POSTNATAL GROWTH OF THE SKULL The brain, skull, eyes and ears all develop earlier than other parts of the body. After birth, the skull thickens with age and continues ossification towards the sutures. The face is relatively underdeveloped at birth and undergoes profound changes throughout childhood and at the adolescent spurt, in response to the eruption of the deciduous and permanent teeth, the formation of the sinuses, and the elongation of the maxilla and mandible (Fig. 34.9). REFERENCES Couly GF, Coltey PM, Le Douarin NM 1993 The triple origin of the skull in higher vertebrates: a study in quail-chick chimeras. Development 117: 409-29. Medline Similar articles Jiang,X, Iseki S, Maxson RE, Sucov HM, Morriss-Kay GM 2002. Tissue origins and interactions in the mammalian skull vault. Developmental Biology 241, 106-116. Opperman LA, Sweeney TM, Redmon J, Persing JA, Ogle RC 1993 Tissue interactions with underlying dura mater inhibit osseous obliteration of developing cranial sutures. Dev Dyn 198: 312-22. Medline Similar articles Full article Thorogood P 1988 The developmental specification of the vertebrate skull. Development 103 (suppl): 14153. page 495 page 496
© 2008 Elsevier
29 HEAD Face and scalp SKIN The scalp and buccolabial tissues are described here. The structure of the eyelids is described on page 681.
SCALP The scalp extends from the top of the forehead in front to the superior nuchal line behind. Laterally it projects down to the zygomatic arch and external acoustic meatus. It consists of five layers: skin, subcutaneous tissue, occipitofrontalis (epicranius) and its aponeurosis, subaponeurotic areolar tissue and pericranium (Fig. 29.1). The skin of the scalp contains the hair and associated glands. There are many sebaceous glands, and the scalp is the commonest site for sebaceous cysts. The dense subcutaneous connective tissue has the richest cutaneous blood supply in the body. The third layer contains occipitofrontalis whose anterior and posterior muscular components are connected by a tough, fibrous, epicranial aponeurosis, and consequently this layer is called the aponeurotic layer (galea aponeurotica). Beneath the aponeurotic layer is a layer of loose connective tissue over which the upper three layers of the scalp can easily slide. The deepest layer is the periosteum of the skull. It is very easy to raise a scalp flap within the plane between the galea and the pericranium without compromising the blood or nerve supply of the scalp, because all of these structures lie in the superficial fascia. Scalp flaps are used in craniofacial surgery - e.g. for the correction of congenital deformity, for the release of craniosynostoses and for the treatment of craniofacial fractures - and also for repairing scalp defects following the excision of skin tumours. An anteriorly based scalp flap gives excellent access to the frontal bone and upper facial skeleton including the orbits and the infratemporal fossa and temporomandibular joint. Similar flaps are seen in traumatic scalp avulsions, which occur when the hair is trapped in moving machinery, and are also used electively in surgery. The arterial blood supply to the scalp is particularly rich, and there are free anastomoses between branches of the occipital and superficial temporal vessels. Scalp lacerations continue to bleed profusely because the elastic fibres of the underlying galea aponeurotica prevent initial vessel retraction: these wounds may be associated with significant blood loss which can result in clinical shock. When suturing scalp lacerations it is essential to control all the bleeding points before repairing the scalp itself. Usually it is necessary to tie off any larger arterioles and use bipolar diathermy to control smaller arterioles and veins. Failure to control the bleeding as a separate step can result in significant haematomas, often subgaleal, leading to breakdown of the original wound and sometimes necessitating surgical drainage. Repair of scalp lacerations usually requires full thickness tension sutures because the galea aponeurotica will otherwise gape as the occipital and frontal muscle bellies contract. However, a wound that does not involve epicranius or its aponeurosis does not gape. UPDATE Date Added: 08 May 2006 Abstract: Anatomy and blood supply of the subgaleal fascia flap Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=15176036&query_hl=22&itool=pubmed_docsum Anatomy and blood supply of the subgaleal fascia flap. Casoli V, Dauphin N, Taki C et al: Clin Anat 17:392-399, 2004. Eyebrows
The eyebrows are two arched eminences of skin which surmount the orbits. Numerous short, thick hairs are set obliquely in them. Fibres of orbicularis oculi, corrugator and the frontal part of occipitofrontalis are inserted into the dermis of the eyebrows.
BUCCOLABIAL TISSUE Cheeks
The cheeks are continuous in front with the lips. The external junction is indicated by the nasolabial groove (sulcus) (see Fig. 29.6), and further laterally by the nasolabial fold, which descends from the side of the nose to the angle of the mouth. The cheek is covered on the outer surface by skin and on the inner surface by mucosa. Each cheek contains the buccinator muscle, and a variable, but usually considerable, amount of adipose tissue often encapsulated to form a biconcave mass, the buccal fat pad (of Bichat), which is particularly evident in infants. Indeed, this fat was originally named the suctorial pad, although its association with suckling is far from obvious. The walls of the cheek also contain fibrous connective tissue, vessels, nerves and numerous small buccal mucous (salivary) glands. UPDATE Date Added: 22 May 2006 Abstract: Anatomical interrelation among the buccal fat pad, buccal branches of the facial nerve, and parotid duct Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=16077311&query_hl=21&itool=pubmed_docsum Interrelated buccal fat pad with facial buccal branches and parotid duct. Hwang K, Cho HJ, Battuvshin D et al: J Craniofac Surg 16:658-660, 2005. UPDATE Date Added: 28 February 2006 Publication Services, Inc. Abstract: Bilateral protrusion of the buccal fat pad into the mouth of an infant: Report of a case Click on the following link to view the abstract:http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=15719927&query_hl=3&itool=pubmed_docsum Bilateral protrusion of the buccal fat pad into the mouth of an infant: Report of a case. Santiago BM, Damasceno LM, Primo LG: J Clin Pediatr Dent 29(2):181-184, 2005. UPDATE Date Added: 21 February 2006 Publication Services, Inc. Abstract: Interrelated buccal fat pad with facial buccal branches and parotid duct Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=16077311&query_hl=8&itool=pubmed_docsum Interrelated buccal fat pad with facial buccal branches and parotid duct. Hwang K, Cho HJ, Battuvshin D et al: J Craniofac Surg 16(4):658-660, 2005. UPDATE Date Added: 09 November 2005 Publication Services, Inc. Update: The interdomal fat pad of the nose: a new anatomical structure. The removal of the interdomal fat pad may contribute to nasal tip reduction in varying degrees. The surgical approach to interdomal fat pad removal can be accomplished only by the open rhinoplasty technique. Although rhinoplasty is the most commonly performed operation in plastic surgery, some unexpected results and complications may occur related to unknown or un-clarified anatomical structures in the nose. Copcu et al. (2004) evaluated the interdomal region in 4 fresh cadavers and 24 patients who underwent open rhinoplasty. Using histochemistry, they demonstrated the existence of the interdomal fat pad as a separate anatomical structure in necroscopy specimens from cadavers. Preoperative nasal tip ultrasonography showed the existence of fat pads in the interdomal space of all
patients in varying size, with larger fat pads in bulbous and fatty noses. The size of the interdomal fat pads varied from 1.2 mm x 2.4 mm to 3.6 mm x 5.2 mm. The interdomal fat pads started at the antero-superior surfaces of the altar cartilages and ended at the supratip region. Excision of fat pads during rhinoplasty did not result in damage to the dermal plexus, skin, or any postoperative damage in the altar cartilages. In patients with bulbous noses and/or divergent intermediate crura of the altar cartilage (larger fat pad), excision of fat pads was easy, since the fat pad is a separate anatomical structure that is not attached to subcutaneous tissue. Histochemical analysis of all biopsies revealed that the interdomal fat pad consisted of pure and mature adipocytes, while fibroadipose tissues were mainly observed in biopsies of nasal subcutaneous tissues. Copcu et al. suggest that the interdomal fat pad is an important anatomical structure that may contribute to unexpected postoperative results in rhinoplasty. Therefore, identification of the interdomal fat pad and detection of its size before the operation may help the surgeon decide on the type of operation and whether different or additional techniques (such as a cartilage-splitting procedure and tip scoring) should be used. In addition, they showed that the interdomal fat pad could be accurately assessed by ultrasonography, which is a safe and inexpensive technique. Copcu E, Metin K, Özsunar Y et al: The interdomal fat pad of the nose: a new anatomical structure. Surg Radiol Anat 26:14-18, 2004. Medline Similar articles
UPDATE Date Added: 25 October 2005 Publication Services, Inc. Abstract: Interrelated buccal fat pad with facial buccal branches and parotid duct. Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve=pubmed=Abstract_uids=16077311_hl=7 Interrelated buccal fat pad with facial buccal branches and parotid duct. Hwang K, Cho HJ, Battuvshin D et al: J Craniofac Surg 16(4):658-660, 2005. Lips
page 497 page 498
Figure 29.1 Coronal section through the scalp, skull and brain. Note: loculated fat between fibrous septa blending with dermis and epicranial aponeurosis (galea aponeurotica); loose subaponeurotic areolar tissue; emissary, diploic, dural and neuropial veins. The superior sagittal sinus and lateral lacunae are more complex than depicted here.
The lips are two fleshy folds surrounding the oral orifice (see Fig. 29.6). The centre of each lip contains a thick fibrous strand, consisting of parallel bundles of skeletal muscle fibres (orbicularis oris, together with incisivus superior and inferior, and the direct labial tractors (p. 508), and their attachments to skin, mucosa or other muscle fibres. The free external surface of each lip is covered by a thin keratinized epidermis, and is continuous with the mucosa at the vermilion (red) zone of the lip. The dermis is well vascularized and accommodates numerous hair follicles (many of them large in the male), sebaceous glands and sweat glands. Subcutaneous adipose tissue is scanty. The internal mucous surfaces are lined with a thick non-keratinizing stratified squamous epithelium, and the submucosa is well vascularized and accommodates numerous labial mucous glands, which may be several millimetres in diameter, the largest being palpable with the tip of the tongue. Because of the thickness of its semi-opaque epithelium, the mucosa of the everted lip appears moist, glistening and pink. Between the skin and mucosa, the vermilion zone is covered with a specialized keratinized stratified squamous epithelium which is thin near the skin, increases in thickness slightly as the mucosa is approached, and then thickens abruptly when true mucosa is reached. The epithelium is covered with transparent, dead squames and its deep surface is highly convoluted, interdigitating with abundant long dermal papillae. The latter carry a rich capillary plexus which imparts a dusky red colour. These surfaces are hairless, their dermis carries no sebaceous, sweat or mucous glands, and they are moistened with saliva by the tip of the tongue. The dense innervation of the lips is consistent with their acute sensitivity to light touch sensation. This is due mainly to the increased density of Meissner's corpuscles (p. 61) in the dermal papillae. The size and curvature of the exposed red lip surfaces is subject to considerable individual, gender, and ethnic variation. The line of contact between the lips, the oral fissure, lies just above the cutting edges of the maxillary incisor teeth. On each side a labial commissure forms the angle (corner) of the mouth, usually near the first premolar tooth. The labial epithelia and internal tissues radiate over the boundaries of the commissure to become continuous with those of the cheek. With age, buccolabial (labiomarginal) grooves appear at the corners of the mouth. On each side the upper lip is separated from the cheek laterally by the nasolabial groove and is continuous above the nasal ala with the circumalar groove (sulcus). The lower lip is separated from the chin by the mentolabial groove (sulcus). Externally, the central region in the upper lip presents a shallow vertical groove, the philtrum, which ends below in a slight tubercle limited by lateral ridges. The lower lip shows a small depression in the midline that corresponds to the tubercle. The junction between the external, hair-bearing skin and the red, hairless surface of the upper lip almost invariably takes the form of a double-curved Cupid's bow. From the centre it rises rapidly on each side to an apex that corresponds to the lower end of each ridge of the philtrum. It then slopes gently downwards and usually ends horizontally but sometimes curves slightly upwards (infrequently downwards). The line of contact between the red lip surfaces is often almost horizontal but quite frequently takes the form of a much less wavy Cupid's bow. In the lower lip the junction between the skin and the red lip varies greatly between individuals in its vertical depth at the centre, whereas the lateral extremities descend medially for a few millimetres in all individuals. In the upper lip, a narrow band of smooth tissue related to the subnasal maxillae marks the point at which labial mucosa becomes continuous with gingival mucosa. The corresponding reflexion in the lower lip coincides approximately with the mentolabial sulcus, and here the lip is continuous with mental tissues. The upper and lower lips differ in cross-sectional profile in that neither is a simple fold
of uniform thickness. The upper lip has a bulbous asymmetrical profile, the skin and red-lip having a slight external convexity, and the adjoining red-lip and mucosa a pronounced internal convexity, creating a mucosal ridge or shelf that can be wrapped around the incisal edges of the parted teeth. The lower lip is on a more posterior plane than the upper lip. In the position of neutral lip contact, the external surface of the lower lip is concave, and there is minimal or no elevation of the internal mucosal surface. The profile of the lips can be modified by muscular activity (p. 508).
RELAXED SKIN TENSION LINES AND SKIN FLAPS ON THE FACE The direction in which facial skin tension is greatest varies regionally. Skin tension lines which follow the furrows formed when the skin is relaxed are known as 'relaxed skin tension lines' (p. 173) (Borges & Alexander 1962). In the living face, these lines frequently (but not always) coincide with wrinkle lines (Fig. 29.2) and can therefore act as a guide in planning elective incisions. When lesions on the face such as scars, pigmented lesions and skin cancers are excised, the dimensions of these lesions often permit excision as an ellipse, so that the resulting defect can be closed as a straight line. If the resulting scar is to be aesthetically acceptable it is important to make the long axis of the ellipse parallel to the natural relaxed skin tension lines, so that the scar will look like a natural skin crease. If the excision line runs contrary to the skin tension lines, the scar may be more conspicuous and will tend to stretch transversely as a result of natural expressive facial movements. When larger lesions are excised it may be necessary to advance or rotate other adjacent soft tissue to fill the defect. The ability to raise these skin flaps is entirely dependent on the regional blood supply and both random pattern and axial pattern skin flaps (p. 169) are used surgically. Because of the richness of the subdermal plexus in the face, random pattern flaps can be raised with a greater length:breadth ratio than in any other area of the body.
page 498 page 499
Figure 29.2 A, Distribution of relaxed skin tension lines (Kraissl's lines) lateral view. B, Anterior view.
The following are examples of axial pattern flaps that can be used to reconstruct defects on the face and scalp. Supratrochlear/supraorbital arteries support forehead flaps that are useful for nasal reconstruction. There is usually enough skin laxity to allow the majority of the donor site to be closed directly. The frontal branch of the superficial temporal artery anastomoses in the midline with its opposite number, and consequently the entire forehead skin can be raised on a narrow pedicle based on just one of the superficial temporal arteries. These flaps can be used to repair many facial defects and also intraoral defects, but the donor site defect cannot be closed directly and must be covered by a skin graft. The parietal branch of the superficial temporal artery and the occipital artery can support hair-bearing flaps from the scalp which are useful for reconstructing defects involving the scalp. The nasolabial flap utilizes the lax skin just lateral to the nasolabial groove. It is not supplied by a named axial artery but rather its blood supply is provided by many small branches from the underlying facial artery, which run perpendicular to the skin surface, allowing these flaps to be raised with their base either superior or inferior.
© 2008 Elsevier
SKIN The scalp and buccolabial tissues are described here. The structure of the eyelids is described on page 681.
SCALP The scalp extends from the top of the forehead in front to the superior nuchal line behind. Laterally it projects down to the zygomatic arch and external acoustic meatus. It consists of five layers: skin, subcutaneous tissue, occipitofrontalis (epicranius) and its aponeurosis, subaponeurotic areolar tissue and pericranium (Fig. 29.1). The skin of the scalp contains the hair and associated glands. There are many sebaceous glands, and the scalp is the commonest site for sebaceous cysts. The dense subcutaneous connective tissue has the richest cutaneous blood supply in the body. The third layer contains occipitofrontalis whose anterior and posterior muscular components are connected by a tough, fibrous, epicranial aponeurosis, and consequently this layer is called the aponeurotic layer (galea aponeurotica). Beneath the aponeurotic layer is a layer of loose connective tissue over which the upper three layers of the scalp can easily slide. The deepest layer is the periosteum of the skull. It is very easy to raise a scalp flap within the plane between the galea and the pericranium without compromising the blood or nerve supply of the scalp, because all of these structures lie in the superficial fascia. Scalp flaps are used in craniofacial surgery - e.g. for the correction of congenital deformity, for the release of craniosynostoses and for the treatment of craniofacial fractures - and also for repairing scalp defects following the excision of skin tumours. An anteriorly based scalp flap gives excellent access to the frontal bone and upper facial skeleton including the orbits and the infratemporal fossa and temporomandibular joint. Similar flaps are seen in traumatic scalp avulsions, which occur when the hair is trapped in moving machinery, and are also used electively in surgery. The arterial blood supply to the scalp is particularly rich, and there are free anastomoses between branches of the occipital and superficial temporal vessels. Scalp lacerations continue to bleed profusely because the elastic fibres of the underlying galea aponeurotica prevent initial vessel retraction: these wounds may be associated with significant blood loss which can result in clinical shock. When suturing scalp lacerations it is essential to control all the bleeding points before repairing the scalp itself. Usually it is necessary to tie off any larger arterioles and use bipolar diathermy to control smaller arterioles and veins. Failure to control the bleeding as a separate step can result in significant haematomas, often subgaleal, leading to breakdown of the original wound and sometimes necessitating surgical drainage. Repair of scalp lacerations usually requires full thickness tension sutures because the galea aponeurotica will otherwise gape as the occipital and frontal muscle bellies contract. However, a wound that does not involve epicranius or its aponeurosis does not gape. UPDATE Date Added: 08 May 2006 Abstract: Anatomy and blood supply of the subgaleal fascia flap Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=15176036&query_hl=22&itool=pubmed_docsum Anatomy and blood supply of the subgaleal fascia flap. Casoli V, Dauphin N, Taki C et al: Clin Anat 17:392-399, 2004. Eyebrows
The eyebrows are two arched eminences of skin which surmount the orbits. Numerous short, thick hairs are set obliquely in them. Fibres of orbicularis oculi, corrugator and the frontal part of occipitofrontalis are inserted into the dermis of the eyebrows.
BUCCOLABIAL TISSUE Cheeks
The cheeks are continuous in front with the lips. The external junction is indicated by the nasolabial groove (sulcus) (see Fig. 29.6), and further laterally by the nasolabial fold, which descends from the side of the nose to the angle of the mouth. The cheek is covered on the outer surface by skin and on the inner surface by mucosa. Each cheek contains the buccinator muscle, and a variable, but usually considerable, amount of adipose tissue often encapsulated to form a biconcave mass, the buccal fat pad (of Bichat), which is particularly evident in infants. Indeed, this fat was originally named the suctorial pad, although its association with suckling is far from obvious. The walls of the cheek also contain fibrous connective tissue, vessels, nerves and numerous small buccal mucous (salivary) glands. UPDATE Date Added: 22 May 2006 Abstract: Anatomical interrelation among the buccal fat pad, buccal branches of the facial nerve, and parotid duct Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=16077311&query_hl=21&itool=pubmed_docsum Interrelated buccal fat pad with facial buccal branches and parotid duct. Hwang K, Cho HJ, Battuvshin D et al: J Craniofac Surg 16:658-660, 2005. UPDATE Date Added: 28 February 2006 Publication Services, Inc. Abstract: Bilateral protrusion of the buccal fat pad into the mouth of an infant: Report of a case Click on the following link to view the abstract:http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=15719927&query_hl=3&itool=pubmed_docsum Bilateral protrusion of the buccal fat pad into the mouth of an infant: Report of a case. Santiago BM, Damasceno LM, Primo LG: J Clin Pediatr Dent 29(2):181-184, 2005. UPDATE Date Added: 21 February 2006 Publication Services, Inc. Abstract: Interrelated buccal fat pad with facial buccal branches and parotid duct Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=16077311&query_hl=8&itool=pubmed_docsum Interrelated buccal fat pad with facial buccal branches and parotid duct. Hwang K, Cho HJ, Battuvshin D et al: J Craniofac Surg 16(4):658-660, 2005. UPDATE Date Added: 09 November 2005 Publication Services, Inc. Update: The interdomal fat pad of the nose: a new anatomical structure. The removal of the interdomal fat pad may contribute to nasal tip reduction in varying degrees. The surgical approach to interdomal fat pad removal can be accomplished only by the open rhinoplasty technique. Although rhinoplasty is the most commonly performed operation in plastic surgery, some unexpected results and complications may occur related to unknown or un-clarified anatomical structures in the nose. Copcu et al. (2004) evaluated the interdomal region in 4 fresh cadavers and 24 patients who underwent open rhinoplasty. Using histochemistry, they demonstrated the existence of the interdomal fat pad as a separate anatomical structure in necroscopy specimens from cadavers. Preoperative nasal tip ultrasonography showed the existence of fat pads in the interdomal space of all patients in varying size, with larger fat pads in bulbous and fatty noses. The size of the interdomal fat pads varied from 1.2 mm x 2.4 mm to 3.6 mm x 5.2 mm. The interdomal fat pads started at the antero-superior surfaces of the altar cartilages and ended at the supratip region.
Excision of fat pads during rhinoplasty did not result in damage to the dermal plexus, skin, or any postoperative damage in the altar cartilages. In patients with bulbous noses and/or divergent intermediate crura of the altar cartilage (larger fat pad), excision of fat pads was easy, since the fat pad is a separate anatomical structure that is not attached to subcutaneous tissue. Histochemical analysis of all biopsies revealed that the interdomal fat pad consisted of pure and mature adipocytes, while fibroadipose tissues were mainly observed in biopsies of nasal subcutaneous tissues. Copcu et al. suggest that the interdomal fat pad is an important anatomical structure that may contribute to unexpected postoperative results in rhinoplasty. Therefore, identification of the interdomal fat pad and detection of its size before the operation may help the surgeon decide on the type of operation and whether different or additional techniques (such as a cartilage-splitting procedure and tip scoring) should be used. In addition, they showed that the interdomal fat pad could be accurately assessed by ultrasonography, which is a safe and inexpensive technique. Copcu E, Metin K, Özsunar Y et al: The interdomal fat pad of the nose: a new anatomical structure. Surg Radiol Anat 26:14-18, 2004. Medline Similar articles
UPDATE Date Added: 25 October 2005 Publication Services, Inc. Abstract: Interrelated buccal fat pad with facial buccal branches and parotid duct. Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve=pubmed=Abstract_uids=16077311_hl=7 Interrelated buccal fat pad with facial buccal branches and parotid duct. Hwang K, Cho HJ, Battuvshin D et al: J Craniofac Surg 16(4):658-660, 2005. Lips
page 497 page 498
Figure 29.1 Coronal section through the scalp, skull and brain. Note: loculated fat between fibrous septa blending with dermis and epicranial aponeurosis (galea aponeurotica); loose subaponeurotic areolar tissue; emissary, diploic, dural and neuropial veins. The superior sagittal sinus and lateral lacunae are more complex than depicted here.
The lips are two fleshy folds surrounding the oral orifice (see Fig. 29.6). The centre of each lip contains a thick fibrous strand, consisting of parallel bundles of
skeletal muscle fibres (orbicularis oris, together with incisivus superior and inferior, and the direct labial tractors (p. 508), and their attachments to skin, mucosa or other muscle fibres. The free external surface of each lip is covered by a thin keratinized epidermis, and is continuous with the mucosa at the vermilion (red) zone of the lip. The dermis is well vascularized and accommodates numerous hair follicles (many of them large in the male), sebaceous glands and sweat glands. Subcutaneous adipose tissue is scanty. The internal mucous surfaces are lined with a thick non-keratinizing stratified squamous epithelium, and the submucosa is well vascularized and accommodates numerous labial mucous glands, which may be several millimetres in diameter, the largest being palpable with the tip of the tongue. Because of the thickness of its semi-opaque epithelium, the mucosa of the everted lip appears moist, glistening and pink. Between the skin and mucosa, the vermilion zone is covered with a specialized keratinized stratified squamous epithelium which is thin near the skin, increases in thickness slightly as the mucosa is approached, and then thickens abruptly when true mucosa is reached. The epithelium is covered with transparent, dead squames and its deep surface is highly convoluted, interdigitating with abundant long dermal papillae. The latter carry a rich capillary plexus which imparts a dusky red colour. These surfaces are hairless, their dermis carries no sebaceous, sweat or mucous glands, and they are moistened with saliva by the tip of the tongue. The dense innervation of the lips is consistent with their acute sensitivity to light touch sensation. This is due mainly to the increased density of Meissner's corpuscles (p. 61) in the dermal papillae. The size and curvature of the exposed red lip surfaces is subject to considerable individual, gender, and ethnic variation. The line of contact between the lips, the oral fissure, lies just above the cutting edges of the maxillary incisor teeth. On each side a labial commissure forms the angle (corner) of the mouth, usually near the first premolar tooth. The labial epithelia and internal tissues radiate over the boundaries of the commissure to become continuous with those of the cheek. With age, buccolabial (labiomarginal) grooves appear at the corners of the mouth. On each side the upper lip is separated from the cheek laterally by the nasolabial groove and is continuous above the nasal ala with the circumalar groove (sulcus). The lower lip is separated from the chin by the mentolabial groove (sulcus). Externally, the central region in the upper lip presents a shallow vertical groove, the philtrum, which ends below in a slight tubercle limited by lateral ridges. The lower lip shows a small depression in the midline that corresponds to the tubercle. The junction between the external, hair-bearing skin and the red, hairless surface of the upper lip almost invariably takes the form of a double-curved Cupid's bow. From the centre it rises rapidly on each side to an apex that corresponds to the lower end of each ridge of the philtrum. It then slopes gently downwards and usually ends horizontally but sometimes curves slightly upwards (infrequently downwards). The line of contact between the red lip surfaces is often almost horizontal but quite frequently takes the form of a much less wavy Cupid's bow. In the lower lip the junction between the skin and the red lip varies greatly between individuals in its vertical depth at the centre, whereas the lateral extremities descend medially for a few millimetres in all individuals. In the upper lip, a narrow band of smooth tissue related to the subnasal maxillae marks the point at which labial mucosa becomes continuous with gingival mucosa. The corresponding reflexion in the lower lip coincides approximately with the mentolabial sulcus, and here the lip is continuous with mental tissues. The upper and lower lips differ in cross-sectional profile in that neither is a simple fold of uniform thickness. The upper lip has a bulbous asymmetrical profile, the skin and red-lip having a slight external convexity, and the adjoining red-lip and mucosa a pronounced internal convexity, creating a mucosal ridge or shelf that can be wrapped around the incisal edges of the parted teeth. The lower lip is on a
more posterior plane than the upper lip. In the position of neutral lip contact, the external surface of the lower lip is concave, and there is minimal or no elevation of the internal mucosal surface. The profile of the lips can be modified by muscular activity (p. 508).
RELAXED SKIN TENSION LINES AND SKIN FLAPS ON THE FACE The direction in which facial skin tension is greatest varies regionally. Skin tension lines which follow the furrows formed when the skin is relaxed are known as 'relaxed skin tension lines' (p. 173) (Borges & Alexander 1962). In the living face, these lines frequently (but not always) coincide with wrinkle lines (Fig. 29.2) and can therefore act as a guide in planning elective incisions. When lesions on the face such as scars, pigmented lesions and skin cancers are excised, the dimensions of these lesions often permit excision as an ellipse, so that the resulting defect can be closed as a straight line. If the resulting scar is to be aesthetically acceptable it is important to make the long axis of the ellipse parallel to the natural relaxed skin tension lines, so that the scar will look like a natural skin crease. If the excision line runs contrary to the skin tension lines, the scar may be more conspicuous and will tend to stretch transversely as a result of natural expressive facial movements. When larger lesions are excised it may be necessary to advance or rotate other adjacent soft tissue to fill the defect. The ability to raise these skin flaps is entirely dependent on the regional blood supply and both random pattern and axial pattern skin flaps (p. 169) are used surgically. Because of the richness of the subdermal plexus in the face, random pattern flaps can be raised with a greater length:breadth ratio than in any other area of the body.
page 498 page 499
Figure 29.2 A, Distribution of relaxed skin tension lines (Kraissl's lines) lateral view. B, Anterior view.
The following are examples of axial pattern flaps that can be used to reconstruct defects on the face and scalp. Supratrochlear/supraorbital arteries support forehead flaps that are useful for nasal reconstruction. There is usually enough skin laxity to allow the majority of the donor site to be closed directly. The frontal branch of the superficial temporal artery anastomoses in the midline with its opposite number, and consequently the entire forehead skin can be raised on a narrow pedicle based on just one of the superficial temporal arteries. These flaps can be used to repair many facial defects and also intraoral defects, but the donor site defect cannot be closed directly and must be covered by a skin graft. The parietal branch of the superficial temporal artery and the occipital artery can support hair-bearing flaps from the scalp which are useful for reconstructing defects involving the scalp. The nasolabial flap utilizes the lax skin just lateral to the nasolabial groove. It is not supplied by a named axial artery but rather its blood supply is provided by many small branches from the underlying facial artery, which run perpendicular to the skin surface, allowing these flaps to be raised with their base either superior or inferior.
© 2008 Elsevier
SOFT TISSUE FASCIAL LAYERS Fascia of scalp
The superficial fascia of the scalp is firm, dense and fibroadipose, and adheres closely to both skin and the underlying epicranius including its epicranial aponeurosis, the galea aponeurotica. Posteriorly it is continuous with the superficial fascia of the back of the neck, and laterally it is prolonged into the temporal region, where it is looser in texture.
FASCIAL LAYERS AND TISSUE PLANES IN THE FACE UPDATE Date Added: 07 December 2004 Update: The safe face lift with bony anatomic landmarks to elevate the SMAS Face lifts that require resuspension of the superficial musculoaponeurotic system (SMAS) are accompanied by a risk for facial nerve injury because the branches of the facial nerve that emerge from the anterior end of the parotid gland are exposed when the SMAS is elevated anterior to the parotid gland. Identification of bony anatomic landmarks to predict the location of the anterior edge of the parotid gland would decrease this risk. A recent study using 20 cadaver face halves provides accurate measurements to assist in the location of the anterior edge of the parotid gland and the exiting facial nerve branches. The most anterior portion of the parotid gland was 2.7 ± 1.0 mm anterior to the vector from the inferior lateral orbital rim to the masseteric tuberosity. The most posterior part of the anterior edge of the parotid gland was 1.0 ± 1.5 mm posterior to the aforementioned vector. The mean width of the parotid gland, from the tragus to the anterior parotid edge was 38.8 ± 3.5 mm. These results indicate that the anterior edge of the parotid gland where the facial nerve branches exit is 38.8 mm anterior to the tragus along the transverse axis of the zygomatic arch and near the oblique vector from the inferior lateral orbital wall to the masseteric tuberosity. Knowing this location facilitates resuspension of the SMAS during face lift surgery without damage to the facial nerve. Wilhelmi BJ, Mowlavi A, Neumeister MW. The safe face lift with bony anatomic landmarks to elevate the SMAS. Plast Reconstr Surg. 2003; 111: 1723. Medline Similar articles
On the basis of gross dissection and complementary histological studies, four distinct tissue planes are recognized on the face superficial to the plane of the facial nerve and its branches. From superficial to deep, these layers are the skin; a subcutaneous layer of fibro-adipose tissue; the superficial musculo-aponeurotic system (SMAS); and the parotid-masseteric fascia. Subcutaneous fibroadipose tissue
This homogeneous layer is present throughout the face, although the degree of adiposity varies in different parts of the face. Anteriorly, it crosses the nasolabial fold onto the lip, and superiorly it crosses the zygomatic arch. In both locations the layer is more fascial than fatty. The fat content of the subcutaneous tissue in the cheek accounts for the cheek mass: part of the subcutaneous adipose tissue is the malar fat pad, a more or less discrete aggregation of fatty tissue inferolateral to the orbital margin. UPDATE Abstract: The anatomy of temporal hollowing
Date Added: 30 May 2006
Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=16077309&query_hl=13&itool=pubmed_docsum The anatomy of temporal hollowing: The superficial temporal fat pad. Kim S, Matic DB: J Craniofac Surg 16:651-654, 2005. Superficial musculo-aponeurotic system (SMAS)
This is described as a single tissue plane in the face. In some areas it is composed of muscle fibres, and elsewhere it is composed of fibrous or fibroaponeurotic tissue: it is not directly attached to bone. When traced below the level of the lower border of the mandible it becomes continuous with platysma in the neck. Microdissection has revealed that the SMAS becomes indistinct on the lateral aspect of the face c.1 cm below the level of the zygomatic arch. Anteromedially, the SMAS layer becomes continuous with some of the mimetic muscles including zygomaticus major, frontalis and the peri-orbital fibres of orbicularis oculi. UPDATE
Date Added: 05 June 2006
Abstract: Identification of the lateral zygomaticus major muscle border using bony anatomic landmarks Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=15084877&query_hl=3&itool=pubmed_docsum The extended SMAS facelift: Identifying the lateral zygomaticus major muscle border using bony anatomic landmarks. Mowlavi A, Wilhelmi BJ: Ann Plast Surg 52:353-357, 2004. In most areas of the face, a distinct sub-SMAS plane can be defined deep to SMAS. It is continuous with the plane between platysma and the underlying investing layer of deep cervical fascia in the neck. However, where it overlies the parotid gland, the SMAS is firmly blended with the superficial layer of the parotid fascia, which means that a clear sub-SMAS plane is difficult, if not impossible, to define in the region of the parotid. Parotid-masseteric fascia
This is a filmy areolar layer that overlies the filamentous branches of the facial nerve and the parotid duct as these structures lie on the surface of masseter. Further anteriorly the parotid-masseteric fascia overlies the buccal fat pad which lies superficial to buccinator. Having crossed the surface of the buccal fat pad, the fascia blends with the epimysium on the surface of buccinator. Below the lower border of the mandible, it is continuous with the investing layer of deep cervical fascia. Parotid fascia (capsule)
The parotid gland is surrounded by a fibrous capsule called the parotid fascia or capsule. Traditionally this has been described as an upward continuation of the investing layer of deep cervical fascia in the neck which splits to enclose the gland within a superficial and a deep layer. The superficial layer is attached above to the zygomatic process of the temporal bone, the cartilaginous part of the external acoustic meatus, and the mastoid process. The deep layer is attached to the mandible, and to the tympanic plate, styloid and mastoid processes of the temporal bone. The prevailing view is that the deep layer of the parotid gland is derived from the deep cervical fascia. However, the superficial layer of the parotid capsule appears to be continuous with the fascia associated with platysma, and is now regarded as a component of the SMAS (Mitz & Peyronie 1976, Wassef 1987, Gosain et al 1993). It varies in thickness from a thick fibrous layer anteriorly to a thin translucent membrane posteriorly. It may be traced forwards as a separate layer which passes over the masseteric fascia (itself derived from the deep cervical fascia), separated from it by a cellular layer which contains branches of the facial nerve and the parotid duct. Histologically, the parotid fascia is atypical in that it contains muscle fibres which parallel those of platysma, especially in the lower part of the parotid capsule. Although thin fibrous septa may be seen in the subcutaneous layer at the histological level, macroscopically there is little evidence of a distinct layer of superficial fascia. The deep fascia covering the muscles forming the parotid bed (digastric and styloid group of muscles) contains the stylomandibular and mandibulostylohyoid ligaments. The stylomandibular ligament passes from the styloid process to the angle of the mandible. The more extensive mandibulostylohyoid ligament (angular tract) passes between the angle of the mandible and the stylohyoid ligament for varying distances, generally reaching the hyoid bone (Fig. 29.3). It is thick posteriorly but thins anteriorly in the region of the angle of the mandible (Ziarah & Atkinson 1981, Shimada & Gasser 1988). There is some dispute as to whether the mandibulostylohyoid ligament is part of the deep cervical fascia (Ziarah and Atkinson 1981), or lies deep to it (Shimada & Gasser 1988). The stylomandibular and mandibulostylohyoid ligaments separate the parotid gland region from the superficial part of the submandibular gland, and so are landmarks of surgical interest. UPDATE Date Added: 28 February 2006 Publication Services, Inc. Abstract: The surgical anatomy of the parotid fascia Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=16270161&query_hl=2&itool=pubmed_docsum The surgical anatomy of the parotid fascia. Ramsaroop L, Singh B, Allopi L et al: Surg Radiol Anat 15, 2005. Temporo-parietal and temporal fasciae
Above the level of the zygomatic arch, on the lateral side of the head, the temporo-parietal fascia (superficial temporal fascia) constitutes a fascial layer
which lies in the same plane as, but is not continuous with, the SMAS. It is quite separate from, and superficial to, the temporal fascia (deep temporal fascia). More superiorly, it blends with the galea aponeurotica. The plane between the temporo-parietal fascia and the underlying deep temporal fascia contains loose areolar tissue and a small amount of fat. This tissue plane, the temporo-parietal fat pad, is continuous superiorly with the subgaleal plane of loose areolar tissue in the scalp. Running superiorly in the temporo-parietal fascia or just deep to it are the superficial temporal vessels, the auriculotemporal nerve and its branches, and the temporal branch of the facial nerve. The temporal fascia (deep temporal fascia) is a dense aponeurotic layer which lies deep to the temporo-parietal fat pad and covers temporalis: the deep surface of the fascia affords attachment to the superficial fibres of temporalis. Above, it is a single layer attached along the length of the superior temporal line where it blends with the periosteum. Below, at approximately the level of the superior orbital rim, it splits into superficial and deep laminae which run downwards to attach to the lateral and medial margins of the upper surface of the zygomatic arch respectively. The fat enclosed between these two layers is termed the superficial temporal fat pad, and contains the zygomatico-orbital branch of the superficial temporal artery and a cutaneous nerve, the zygomatico-temporal branch of the maxillary nerve. The temporal fascia is overlapped by auriculares anterior and superior, the epicranial aponeurosis and part of orbicularis oculi; the superficial temporal vessels and the auriculotemporal nerve ascend over it. Buccopharyngeal fascia
Buccinator is covered by a thin layer of fascia, the buccopharyngeal fascia, which also covers the superior constrictor of the pharynx. page 499 page 500
Figure 29.3 The mandibulostylohyoid ligament.
Retaining ligaments of the face
These ligaments are fascial bands at specific sites which serve to anchor the skin to the underlying bone. The general cutaneous laxity that attends the ageing process renders facial skin subject to gravitational pull. However, at sites where retaining ligaments are present, the effect of gravitational pull is resisted. When performing facelift procedures, these ligaments must be surgically divided in order to facilitate redraping of facial skin. Examples of retaining ligaments in the face are the zygomatic ligament (also known as McGregor's patch) and the mandibular
ligament. UPDATE Date Added: 02 November 2005 Publication Services, Inc. Update: Surgical anatomy of the midface as applied to facial rejuvenation. Mobilization of soft tissues, which allows re-elevation of the malar fat pad, is an effective method to rejuvenate the midface and eliminate the effects of aging. A clear understanding of the malar fat pad and its ligamentous structures is necessary to permit mobilization and vertical elevation of the midface. To evaluate the anatomy of structures pertinent to midface rejuvenation procedures, Gamboa and colleagues (2004) performed 16 hemifacial dissections on eight fresh adult cadaver heads. They dissected and evaluated four softtissues structures that affect aging including the orbicularis retaining ligaments (ORL), the lateral orbital thickening (LOT), the prezygomatic space, and the zygomatic cutaneous ligaments. The zygomatic ligaments affix the malar fat pad and the skin of the cheek to the underlying zygomatic eminence. Attenuation of support from these ligaments is responsible for many of the stigmata seen in the aging face. The orbicularis retaining ligaments (ORL) attach the orbicularis oculi muscle to the periosteum at the junction of the pars palpebrarum and pars orbitalis. Centrally the ORL reach their maximum length, varying from 1 to 1.6 cm. Medially, the ORL attaches to the orbital rim with maximum distance at the central area diminishing laterally to merge to the lateral orbital thickening (LOT). This anatomic distribution of the ORL leads to laxity within the central third of the face and sagging deformity of the aged face. The lateral orbital thickening (LOT) is a triangular condensation of the superficial musculoaponeurotic system in continuity with the temporoparietal fascia and orbicularis muscle fascia. The LOT attaches to the orbital rim by a fibrous condensation of the ligament and to the lateral orbital tubercle by fibrous connections from the orbicularis muscle fascia to the lateral canthal tendon. Releasing both the ORL and the LOT in combination with suture suspension of these ligaments to the orbital rim or to the temporalis fascia will allow effective improvement of the midface cheek junction. This technique also improves the position of both the lower eyelid and the lateral commissure through redraping the orbicularis oculi muscle, which tightens the lower lid and supports the cheek. The prezygomatic space is a virtual space where the orbicularis oculi muscle overlies the body of the zygoma and the origins of the levator muscle of the upper lip. The anatomic boundaries of this space are formed superiorly by the orbicularis retaining ligament, inferiorly by the zygomatic retaining ligaments, medially by the medial orbital rim and the origin of the levator labii suprioris, and laterally by the LOT. The overlying structure of the space is formed by skin, subcutaneous fat, orbicularis oculi muscle, orbicularis oculi fascia, and suborbicularis oculi fat. The facial nerve branches around and through the suborbicularis fat at the roof of the prezygomatic space. The retaining zygomatic ligament at the lower boundary of the space attaches to the malar fat pad. Ptosis or descend of the malar fat pad can occur secondarily to laxity of the upper boundary (specifically the ORL), and is accentuated by the resistance of the lower boundary of the zygomatic cutaneous retaining ligaments. To achieve sufficient mobilization of the malar fat pad and to correct the infrazygomatic region of the cheek, the inferior and lateral zygomatic retaining ligament should be released. Gamboa GM, de la Torre JI, Vasconez LO: Surgical anatomy of the midface as applied to facial rejuvenation. Ann Plast Surg 52(3):240-245, 2004.
FASCIAL SPACES Two tissue spaces on the face may be involved in spread of odontogenic infection. They are the buccal tissue space, lying between the skin and surface of buccinator, and the infraorbital tissue space, lying between the bony attachments of levator labii superioris and levator anguli oris.
© 2008 Elsevier
CRANIOFACIAL MUSCLES (Fig. 29.4) Muscles of the head can be divided into craniofacial and masticatory groups. Craniofacial muscles are often referred to, not very accurately, as 'muscles of facial expression', and are related mainly to the orbital margins and eyelids, the external nose and nostrils, the lips, cheeks and mouth, the pinna, scalp and cervical skin. Masticatory muscles are concerned primarily with movements of the temporomandibular joint. This division of head musculature reflects differences in embryonic origin and innervation. In functional terms, activities such as mastication, deglutition, vocalization, communication, emotional expression, respiration, ocular, aural and nasal action reflect the close cooperation and interdependence between muscles in the two groups. The organization of the muscles of facial expression differs from that of muscles in most other regions of the body because there is no deep membranous fascia beneath the skin. Instead, many small slips of muscle attached to the facial skeleton insert directly into the skin. Although the muscles can cause movement of the facial skin that reflects emotions, because they are grouped mainly around the orifices of the face, it is often argued that their primary function is to act as sphincters and dilators of the facial orifices and that the function of facial expression has developed secondarily. Embryologically, they are derived from the mesenchyme of the second branchial arch and so are innervated by the facial nerve. Topographically and functionally the muscles of facial expression may be subdivided into epicranial, circumorbital and palpebral, nasal, and buccolabial groups (Fig. 29.5).
EPICRANIAL MUSCLE GROUP Epicranius
Epicranius consists of occipitofrontalis and temporoparietalis. Occipitofrontalis (Fig. 29.5)
Occipitofrontalis covers the dome of the skull from the highest nuchal lines to the eyebrows. It is a broad, musculofibrous layer consisting of four thin, muscular quadrilateral parts, two occipital and two frontal, connected by the epicranial aponeurosis. Each occipital part (occipitalis) arises by tendinous fibres from the lateral two-thirds of the highest nuchal line of the occipital bone and the adjacent region of the mastoid part of the temporal bone, and extends forwards to join the aponeurosis. The gap between the two occipital parts is occupied by an extension of the epicranial aponeurosis. Each frontal part (frontalis) is adherent to the superficial fascia, particularly of the eyebrows. Although frontalis has no bony attachments of its own, its fibres blend with those of adjacent muscles - procerus, corrugator supercilii and orbicularis oculi - and ascend to join the epicranial aponeurosis in front of the coronal suture. Vascular supply Occipitofrontalis is supplied by branches of the superficial temporal, ophthalmic, posterior auricular and occipital arteries. Innervation The occipital part of occipitofrontalis is supplied by the posterior auricular branch of the facial nerve and the frontal part is supplied by the temporal branches of the facial nerve. page 500 page 501
Figure 29.4 A, Anterior view of the skull, showing muscle attachments. B, Basal view of the skull, showing muscle attachments. C, Lateral view of the skull, showing muscle attachments.
Actions Acting from above, the frontal parts raise the eyebrows and the skin over the root of the nose (e.g. as in expressions of surprise or horror). Acting from below, the frontal parts draw the scalp forwards, throwing the forehead into transverse wrinkles. The occipital parts draw the scalp backwards. Acting alternately, the occipital and frontal parts can move the entire scalp backwards and forwards. Variations A thin muscular slip, transversus nuchae, is present in c.25% of people. It arises from the external occipital protuberance or from the superior nuchal line, either superficial or deep to trapezius. It is frequently inserted with auricularis posterior, but may blend with the posterior edge of sternocleidomastoid. The epicranial aponeurosis (Fig. 29.5) The epicranial aponeurosis covers the upper part of the cranium and, with the epicranial muscle, forms a continuous fibromuscular sheet extending from the occiput to the eyebrows. Posteriorly, between the occipital parts of occipitofrontalis, it is attached to the external protuberance and highest nuchal line of the occipital bone. Anteriorly it splits to enclose the frontal parts and sends a short narrow prolongation between them. Laterally, the anterior and superior auricular muscles are attached to it, and the aponeurosis is thinner, and continues over the temporal fascia to the zygomatic arch. It is united to the skin lying over the cranial vault by fibrous superficial fascia, but it is connected more loosely to the underlying pericranium by areolar tissue, and this arrangement allows it to move freely, carrying with it the skin of the scalp. Temporoparietalis
Temporoparietalis is a variably developed sheet of muscle which lies between the frontal parts of occipitofrontalis and the anterior and superior auricular muscles.
CIRCUMORBITAL AND PALPEBRAL MUSCLE GROUP
The circumorbital and palpebral group of muscles are orbicularis oculi, corrugator supercilii and levator palpebrae superioris. The first two are described here and levator palpebrae superioris is described in the context of the eye (p. 691). Orbicularis oculi (Fig. 29.5)
Orbicularis oculi is a broad, flat, elliptical muscle which surrounds the circumference of the orbit and spreads into the adjacent regions of the eyelids, anterior temporal region, infraorbital cheek and superciliary region. It has orbital, palpebral and lacrimal parts. page 501 page 502
The orbital part arises from the nasal component of the frontal bone, the frontal process of the maxilla and from the medial palpebral ligament. The fibres form complete ellipses, without interruption on the lateral side, where there is no bony attachment. The upper orbital fibres blend with the frontal part of occipitofrontalis and the corrugator supercilii. Many of them are inserted into the skin and subcutaneous tissue of the eyebrow, constituting depressor supercilii. Inferiorly and medially, the ellipses overlap or blend to some extent with adjacent muscles (levator labii superioris alaeque nasi, levator labii superioris and zygomaticus minor). At the extreme periphery, sectors of complete, and sometimes incomplete, ellipses have a loose areolar connection with the temporal extension of the epicranial aponeurosis. The palpebral part arises from the medial palpebral ligament, mainly from its superficial surface, and from the bone immediately above and below the ligament. The fibres sweep across the eyelids anterior to the orbital septum, interlacing at the lateral commissure to form the lateral palpebral raphe. A small group of fine fibres, close to the margin of each eyelid behind the eyelashes, constitutes the ciliary bundle. The lacrimal part arises from the upper part of the lacrimal crest, and the adjacent lateral surface, of the lacrimal bone. It passes laterally behind the nasolacrimal sac (where some fibres are inserted into the associated fascia), and divides into upper and lower slips. Some fibres are inserted into the tarsi of the eyelids close to the lacrimal canaliculi, but most continue across in front of the tarsi and interlace in the lateral palpebral raphe. Vascular supply Orbicularis oculi is supplied by branches of the facial, superficial temporal, maxillary and ophthalmic arteries. Innervation Orbicularis oculi is supplied by temporal and zygomatic branches of the facial nerve. UPDATE Date Added: 26 April 2006 Abstract: The temporal branch of the facial nerve in the upper orbicularis oculi muscle Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=15111793&query_hl=20&itool=pubmed_docsum Pattern of the temporal branch of the facial nerve in the upper orbicularis oculi muscle. Hwang K, Cho HJ, Chung IH: J Craniofac Surg 15:373-376, 2004. Actions Orbicularis oculi is the sphincter muscle of the eyelids and plays an important role in facial expression and various ocular reflexes. The orbital portion is usually activated under voluntary control. Contraction of the upper orbital fibres produces vertical furrowing above the bridge of the nose, narrowing of the palpebral fissure, and bunching and protrusion of the eyebrows, which reduces the amount of light entering the eyes. Eye closure is largely affected by lowering of the upper eyelid, but there is also considerable elevation of the lower eyelid. The palpebral portion can be contracted voluntarily, to close the lids gently as in sleep, or reflexly, to close the lids protectively in blinking. The palpebral part has upper depressor and lower elevator fascicles. The lacrimal part of the muscle draws the eyelids and the lacrimal papillae medially, exerting traction on the lacrimal fascia and may aid drainage of tears by dilating the lacrimal sac. It may also influence pressure gradients within the lacrimal gland and ducts. This activity may assist in the sinuous flow of tears across the cornea, direct the lacrimal punctum into the lacus lacrimalis, and express secretions of the ciliary and tarsal glands. When the entire orbicularis oculi muscle contracts, the skin is thrown into folds which radiate from the lateral angle of the eyelids. Such folds, when permanent, cause wrinkles in middle age (the so-called 'crow's feet').
UPDATE Date Added: 08 February 2005 Shanida Helena Nataraja, PhD (Dianthus Medical Limited) Update: Use of the orbicularis oculi myocutaneous flap in the repair of the medial canthal region This paper reviews the surgical anatomy of the orbicularis oculi myocutaneous flap (OOMF) in the repair of the periorbital region. The authors demonstrate its blood and nerve supply, its relationship with the surrounding tissue, and its arc of rotation, in fresh cadaveric heads. Blood supply was from the marginal and peripheral arcades of the medial palpebral artery, a branch of the ophthalmic artery. Venous drainage followed the arterial network. Nerves to orbicularis oculi were derived from the frontal and zygomatic branches of the facial nerve. The authors also report their experience with the flap in patients with epitheliomas of the inner canthus. Four patients (mean age 60 years; two men and two women) with medium size (no more than 2.5 cm diameter) epitheliomas of the inner canthus were treated with surgical excision and subsequent reconstruction with either a medially based island (2 patients) or a non-island (2 patients) OOMF. Mean follow up was 36 months, and healing was successful with no cases of persistent edema, deformities of the flap, or abnormal scarring. Scars were reported to be well concealed, soft, and flat, with good color matching. These findings suggest that the OOMF can effectively correct defects of the inner canthus. The flap is most suitable for reconstruction in older patients, who have an excess of skin in the eyelid, but is recommended for use in all patients with a small to medium size defect and eyelid skin excess. Stagno d'Alcontres F, D'Amico E, Colonna MR, Quatra F, Lupo F. The orbicularis oculi myocutaneus flap in the repair of the medial canthal region. A new strategy for canthal resurfacing. Br J Plast Surg. 2004;57(6):540-2. Medline Similar articles
Corrugator supercilii (Fig. 29.5)
Corrugator supercilii is a small pyramidal muscle located at the medial end of each eyebrow. It lies deep to the frontal part of occipitofrontalis and orbicularis oculi, with which it is partially blended. Its fibres arise from bone at the medial end of the superciliary arch and pass laterally and slightly upwards to exert traction on the skin above the middle of the supraorbital margin. Vascular supply Corrugator supercilii is supplied by branches from adjacent arteries, mainly from the superficial temporal and ophthalmic arteries. Innervation Corrugator supercilii is innervated by temporal branches of the facial nerve. page 502 page 503
Actions Corrugator supercilii cooperates with orbicularis oculi, drawing the eyebrows medially and downwards to shield the eyes in bright sunlight. It is also involved in frowning. The combined action of the two muscles produces mainly vertical wrinkles on the supranasal strip of the forehead.
BUCCOLABIAL MUSCLE GROUP The nasal muscle group comprises procerus, nasalis and depressor septi. Procerus (Fig. 29.5)
Procerus is a small pyramidal slip close to, and often partially blended with, the medial side of the frontal part of occipitofrontalis. It arises from a fascial aponeurosis covering the lower part of the nasal bone and the upper part of the lateral nasal cartilage. It is inserted into the skin over the lower part of the forehead between the eyebrows. Normally its lower aponeurosis blends with that of the transverse part of nasalis. A few muscle fascicles of procerus occasionally continue to the nasal ala, some even reaching the upper lip. Vascular supply Procerus is supplied mainly by branches from the facial artery. Innervation Procerus is supplied by temporal and lower zygomatic branches from the facial nerve (although a supply from the buccal branch has been described). Actions Procerus draws down the medial angle of the eyebrow and produces transverse wrinkles over the bridge of the nose. It is active in frowning and 'concentration', and helps to reduce the glare of bright sunlight. Nasalis (Fig. 29.5)
Nasalis consists of transverse and alar parts that may be continuous at their origins. The transverse part (compressor naris) arises from the maxilla just lateral to the nasal notch. Its fibres pass upwards and medially and expand into a thin aponeurosis. At the bridge of the nose, the aponeuroses of the paired muscles merge with each other and with the aponeurosis of procerus. The alar part (dilatator naris) arises from the maxilla below and medial to the transverse part, with which it partly merges, and is attached to the cartilaginous ala nasi. Vascular supply Nasalis is supplied by branches from the facial artery and the infraorbital branch of the maxillary artery. Innervation Nasalis is supplied by the buccal branch of the facial nerve, although there may also be a contribution from the zygomatic branch. Actions The transverse part of nasalis compresses the nasal aperture at the junction of the vestibule of the nose with the nasal cavity. The alar part draws the ala downwards and laterally and so assists in widening the anterior nasal aperture. These actions accompany deep inspiration, and are thus associated with exertion, and also with some emotional states. Depressor septi page 503 page 504
Figure 29.5 The superficial muscles of the head and neck.
Depressor septi is often regarded as part of dilatator naris. It arises from the maxilla above the central incisor tooth and ascends to attach to the mobile part of the nasal septum. It is immediately deep to the mucous membrane of the upper lip. Vascular supply Depressor septi is supplied by the superior labial branch of the facial artery.
Innervation Depressor septi is innervated by the buccal branch, and sometimes by the zygomatic branch, of the facial nerve. Actions Depressor septi pulls the nasal septum downwards and, with the alar part of nasalis, widens the nasal aperture.
BUCCOLABIAL GROUP OF MUSCLES The shape of the mouth and the posture of the lips are controlled by a complex three-dimensional assembly of muscular slips. These include elevators, retractors and evertors of the upper lip (levator labii superioris alaeque nasi, levator labii superioris, zygomaticus major and minor, levator anguli oris and risorius); depressors, retractors and evertors of the lower lip (depressor labii inferioris, depressor anguli oris, and mentalis); a compound sphincter (orbicularis oris, incisivus superior and inferior); buccinator. Levator labii superioris alaequae nasi (Fig. 29.5)
Levator labii superioris alaequae nasi arises from the upper part of the frontal process of the maxilla and, passing obliquely downwards and laterally, divides into medial and lateral slips. The medial slip is inserted into the greater alar cartilage of the nose and the skin over it. The lateral slip is prolonged into the lateral part of the upper lip, where it blends with levator labii superioris and orbicularis oris. Superficial fibres of the lateral slip curve laterally across the front of levator labii superioris and attach along the floor of the dermis at the upper part of the nasolabial furrow and ridge. Vascular supply Levator labii superioris alaequae nasi is supplied by the facial artery and the infraorbital branch of the maxillary artery. Innervation Levator labii superioris alaequae nasi is innervated by zygomatic and buccal branches of the facial nerve. Actions The lateral slip raises and everts the upper lip and raises, deepens and increases the curvature of the top of the nasolabial furrow. The medial slip dilates the nostril, displaces the circumalar furrow laterally, and modifies its curvature. Levator labii superioris (Fig. 29.5)
Levator labii superioris starts from the infraorbital margin, where it arises from the maxilla and zygomatic bone above the infraorbital foramen. Its fibres converge into the muscular substance of the upper lip between the lateral slip of levator labii superioris alaequae nasi and zygomaticus minor. Vascular supply Levator labii superioris is supplied by the facial artery and the infraorbital branch of the maxillary artery. Innervation Levator labii superioris is innervated by the zygomatic and buccal branches of the facial nerve. page 504 page 505
Actions Levator labii superioris elevates and everts the upper lip. Acting with other muscles, it modifies the nasolabial furrow. In some faces, this furrow is a highly characteristic feature and it is often deepened in expressions of sadness or seriousness. Zygomaticus major (Fig. 29.5)
Zygomaticus major arises from the zygomatic bone, just in front of the zygomaticotemporal suture, and passes to the angle of the mouth where it blends with the fibres of levator anguli oris, orbicularis oris and more deeply placed muscular bands. Vascular supply Zygomaticus major is supplied by the superior labial branch of the facial artery. Innervation Zygomaticus major is innervated by the zygomatic and buccal branches of the facial nerve.
Actions Zygomaticus major draws the angle of the mouth upwards and laterally as in laughing. UPDATE Date Added: 25 October 2005 Publication Services, Inc. Update: The extended SMAS facelift: Identifying the lateral zygomaticus major muscle border using bony anatomic landmarks. The extended superficial musculoaponeurotic system (SMAS) rhytidectomy was introduced, following delineation of the medial SMAS anatomy as well as the nasolabial fold and crease, to improve nasolabial fold prominence. This technique requires release of the SMAS from the upper lateral border of the zygmaticus major muscle and continued dissection medial to this muscle. As a result, more dramatic effacement of the nasolabial crease is achieved due to the release of the zygomatic retaining ligaments and effective mobilization and elevation of the ptotic malar soft tissues. Despite these beneficial effects, several studies reported greater risks of complications, such as transient paresis of the lower eyelid muscles and injury to the upper lid nerves, when compared to other limited subSMAS dissection techniques. To minimize nerve injury during extended SMAS rhytidectomy, Mowlavi and colleagues (2004) attempted to define bony anatomical landmarks that reliably identify the upper extent of the zygomaticus major lateral border along its entire cephalad length. They dissected 13 cadaver heads using standard rhytidectomy skin incision. The authors identified palpable bony anatomic landmarks, which include the mental protuberance and the most anterior inferior temporal fossa notch at the junction of the frontal process and temporal process of the zygomatic bone. The lateral border of the zygomaticus major muscle was observed 4.4 ± 2.2 mm lateral and parallel to the oblique line spanning the above landmarks. The authors suggested that predicting the location of the lateral border using the described palpable anatomical landmarks preoperatively would potentially minimize nerve damage during SMAS dissections in extended SMAS rhytidectomy. In addition, because the SMAS fascia tends to become thinner as it spans medially over the zygomaticus major muscle, predicting the lateral border also could help limit the subcutaneous dissection to the lateral border of the zygomaticus major muscle, thus preventing the skin-SMAS overlap especially over this region of SMAS thinning. Mowlavi A, Wilhelmi BJ: The extended SMAS facelift: Identifying the lateral zygomaticus major muscle border using bony anatomic landmarks. Ann Plast Surg 52(4):353-357, 2004.
Zygomaticus minor (Fig. 29.5)
Zygomaticus minor arises from the lateral surface of the zygomatic bone immediately behind the zygomaticomaxillary suture, and passes downwards and medially into the muscular substance of the upper lip. Superiorly it is separated from levator labii superioris by a narrow triangular interval, and inferiorly it blends with this muscle. Vascular supply Zygomaticus minor is supplied by the superior labial branch of the facial artery. Innervation Zygomaticus minor is innervated by the zygomatic and buccal branches of the facial nerve. Actions Zygomaticus minor elevates the upper lip, exposing the maxillary teeth. It also assists in deepening and elevating the nasolabial furrow. Acting together, the main elevators of the lip - levator labii superioris alaequae nasi, levator labii superioris and zygomaticus minor - curl the upper lip in smiling, and in expressing smugness, contempt or disdain. Levator anguli oris (Fig. 29.5)
Levator anguli oris arises from the canine fossa of the maxilla, just below the infraorbital foramen and inserts into and below the angle of the mouth. Its fibres mingle there with other muscle fibres (zygomaticus major, depressor anguli oris, orbicularis oris). Some superficial fibres curve anteriorly and attach to the dermal floor of the lower part of the nasolabial furrow. The infraorbital nerve and accompanying vessels enter the face via the infraorbital foramen between the origins of levator anguli oris and levator labii superioris. Vascular supply Levator anguli oris is supplied by the superior labial branch of the facial artery
and the infraorbital branch of the maxillary artery. Innervation Levator anguli oris is innervated by the zygomatic and buccal branches of the facial nerve. Actions Levator anguli oris raises the angle of the mouth in smiling, and contributes to the depth and contour of the nasolabial furrow. Malaris
Malaris is a thin sheet of muscle that is sometimes found covering and blending with zygomaticus major and minor and the levator labii superioris muscles. It is subject to considerable variation. When present it is continuous with the inferior limit of orbicularis oculi, from which it is possibly derived. Its fibres incline medially and downwards. Some of its superficial fascicles have a dermal attachment to the nasolabial ridge and sulcus, and others pass directly to the angle of the mouth and to the outer third of the upper lip to intersect with bundles of orbicularis oris. Mentalis
Mentalis is a conical fasciculus lying at the side of the frenulum of the lower lip. The fibres arise from the incisive fossa of the mandible and descend to attach to the skin of the chin. Vascular supply Mentalis is supplied by the inferior labial branch of the facial artery and the mental branch of the maxillary artery. Innervation Mentalis is innervated by the mandibular branch of the facial nerve. Actions Mentalis raises the lower lip, wrinkling the skin of the chin. Since it raises the base of the lower lip, it helps in protruding and everting the lower lip in drinking and also in expressing doubt or disdain. Depressor labii inferioris (Fig. 29.5)
Depressor labii inferioris is a quadrilateral muscle that arises from the oblique line of the mandible, between the symphysis menti and the mental foramen. It passes upwards and medially into the skin and mucosa of the lower lip, blending with the paired muscle from the opposite side and with orbicularis oris. Below and laterally it is continuous with platysma. Vascular supply Depressor labii inferioris is supplied by the inferior labial branch of the facial artery and the mental branch of the maxillary artery. Innervation Depressor labii inferioris is innervated by the mandibular branch of the facial nerve. Actions Depressor labii inferioris draws the lower lip downwards and a little laterally in masticatory activity, and may assist in eversion of the lower lip. It contributes to the expressions of irony, sorrow, melancholy and doubt. Depressor anguli oris (Fig. 29.5)
Depressor anguli oris has a long, linear origin from the mental tubercle of the mandible and its continuation, the oblique line, below and lateral to depressor labii inferioris. It converges into a narrow fasciculus that blends at the angle of the mouth with orbicularis oris and risorius. Some fibres continue into the levator anguli oris muscle. Depressor anguli oris is continuous below with platysma and cervical fasciae. Some of its fibres may pass below the mental tubercle and cross the midline to interlace with their contralateral fellows; these constitute the transversus menti (the 'mental sling'). Vascular supply Depressor anguli oris is supplied by the inferior labial branch of the facial artery and the mental branch of the maxillary artery. Innervation Depressor anguli oris is innervated by the buccal and mandibular branches of the facial nerve.
Actions Depressor anguli oris draws the angle of the mouth downwards and laterally in opening the mouth and in expressing sadness. During opening of the mouth the mentolabial sulcus becomes more horizontal and its central part deeper. Buccinator (Fig. 29.5) page 505 page 506
The muscle of the cheek, buccinator, is a thin quadrilateral muscle which occupies the interval between the maxilla and the mandible. Its upper and lower boundaries are attached respectively to the outer surfaces of the alveolar processes of the maxilla and mandible opposite the molar teeth. Its posterior border is attached to the anterior margin of the pterygomandibular raphe. In addition, a few fibres spring from a fine tendinous band that bridges the interval between the maxilla and the pterygoid hamulus, between the tuberosity of the maxilla and the upper end of the pterygomandibular raphe. On its way to the soft palate the tendon of tensor veli palatini pierces the pharyngeal wall in the small gap that lies behind this tendinous band. The posterior part of buccinator is deeply placed, internal to the mandibular ramus and in the plane of the medial pterygoid plate. Its anterior component curves out behind the third molar tooth to lie in the submucosa of the cheek and lips. The fibres of buccinator converge towards the modiolus near the angle of the mouth. Here the central (pterygomandibular) fibres intersect, those from below crossing to the upper part of orbicularis oris, and those from above crossing to the lower part. The highest (maxillary) and lowest (mandibular) fibres of buccinator continue forward to enter their corresponding lips without decussation. As buccinator courses through the cheek and modiolus substantial numbers of its fibres are diverted internally to attach to submucosa. Relations Posteriorly, buccinator lies in the same plane as the superior pharyngeal constrictor, which arises from the posterior margin of the pterygomandibular raphe, and is covered there by the buccopharyngeal fascia. Superficially, the buccal pad of fat separates the posterior part of buccinator from the ramus of the mandible, masseter and part of temporalis. Anteriorly, the superficial surface of buccinator is related to zygomaticus major, risorius, levator and depressor anguli oris, and the parotid duct. It is crossed by the facial artery, facial vein and branches of the facial and buccal nerves. The deep surface of buccinator is related to the buccal glands and mucous membrane of the mouth. The parotid duct pierces buccinator opposite the third upper molar tooth, and lies on the deep surface of the muscle before opening into the mouth opposite the maxillary second molar tooth. Vascular supply Buccinator is supplied by branches from the facial artery and the buccal branch of the maxillary artery. Innervation Buccinator is supplied by the buccal branch of the facial nerve. Actions Buccinator compresses the cheek against the teeth and gums during mastication, and assists the tongue in directing food between the teeth. As the mouth closes, the teeth glide over the buccolabial mucosa, which must be retracted progressively from their occlusal surfaces by buccinator and other submucosally attached muscles. When the cheeks have been distended with air, the buccinators expel it between the lips, an activity important when playing wind instruments, accounting for the name of the muscle (Latin buccinator = trumpeter). Pterygomandibular raphe The pterygomandibular raphe is a thin band of tendinous fibres that stretches from the hamulus of the medial pterygoid plate down to the posterior end of the mylohyoid line of the mandible. It is easily palpated medially, where it is covered by the mucous membrane of the mouth, and laterally it is separated from the ramus of the mandible by a quantity of adipose tissue. It gives attachment posteriorly to the superior constrictor of the pharynx, and anteriorly to the central part of buccinator. Orbicularis oris (Figs 29.5, 29.6)
Orbicularis oris is so named because it was once assumed that the oral fissure was surrounded by a series of complete ellipses of striated muscle which acted
together in the manner of a sphincter. However, it is now recognized that the muscle actually consists of four substantially independent quadrants (upper, lower, left and right), each of which contains a larger pars peripheralis and a smaller pars marginalis. Marginal and peripheral parts are apposed along lines that correspond externally to the lines of junction between the vermilion zone of the lip and the skin. Thus, orbicularis oris is composed of eight segments, each of which is named systematically according to its location. Each segment resembles a fan that has its stem at the modiolus and is open in peripheral segments and almost closed in marginal segments. Pars peripheralis Pars peripheralis has, in each quadrant, a lateral stem attached to the labial side of the modiolus over its full thickness, from apex to base, including the corresponding upper or lower cornu. Most of these stem fibres are thought to originate within the modiolus (although it is possible that some are direct continuations from the other modiolar muscles). The consensus view is that stem fibres are reinforced directly by fibres from buccinator (upper fibres and decussating lower central fibres), levator anguli oris and the superficial part of zygomaticus major in the upper lip, and from buccinator (lower fibres and decussating upper central fibres), and depressor anguli oris in the lower lip. The fibres of orbicularis oris enter their respective superior and inferior labial areas and diverge to form triangular muscular sheets. These are thickest at the junctions between skin and the vermilion zone and become progressively thinner as they reach the limits of the labial region (as defined above). The greater part of each sheet enters the free lip, where its fibres aggregate into cylindrical bundles orientated parallel to the vermilion zone. Fibres of the direct labial tractors pass to their submucosal attachments between these cylindrical bundles and between pars peripheralis and pars marginalis. In the upper lip, the highest fibres run near the nasolabial sulcus, a few fibres attach to the sulcus, and a few to the nasal ala and septum. In the lower lip, the lowest fibres reach and attach to the mentolabial sulcus. A small proportion of the main body of fibres is thought to end in the labial connective tissue, dermis or submucosa as it traverses its quadrant of free lip. Most fibres continue towards the median plane and cross some 5 mm into the opposite half-lip. At this point the fibres from the two sides interlace on their way to their dermal insertions, creating the ridges of the philtrum of the upper lip and the less marked corresponding depression in the lower lip. Pars marginalis Pars marginalis of orbicularis oris is developed to a unique extent in human lips and is closely associated with speech and the production of some kinds of musical tone. In each quadrant the pars marginalis consists of a single (occasionally double) band of narrow diameter muscle fibres lodged within the tissues of each vermilion zone. At their medial end, the marginal fibres meet and interlace with their contralateral fellows and then attach to the dermis of the vermilion zone a few millimetres beyond the median plane in a manner similar to pars peripheralis. At their lateral ends, the fibres converge and attach to the deepest part of the modiolar base along a horizontal strip level with the buccal angle. The relations between pars marginalis and pars peripheralis are complex. In a full thickness section of an upper lip at right angles to the vermilion zone, the cylindrical bundles of peripheralis fibres form an S-shape, with an external convexity above, and an internal convexity below: the classic analogy is to the shank and initial curved part of a hook. Beyond peripheralis, the hook-shape is completed by the blunted triangular profile of marginalis, which occupies the core of the vermilion zone with its base adjacent to peripheralis and its apex reaching upwards and anteriorly towards the junction between vermilion zone and skin. In a similar section through the lower lip, peripheralis bundles form a continuous curve that is concave towards the external surface. This is surmounted by the flattened triangular profile of marginalis, which curves anteriorly, its apex again nearing the vermilion/cutaneous junction. Thus, throughout the vermilion zones of both lips, marginalis lies substantially anterior to the adjacent bundles of peripheralis. However, as the muscles are traced laterally beyond the vermilion zone and across the buccal angle, this relationship alters and marginalis becomes inverted as it wraps progressively around the adjacent edge of peripheralis to reach its deep (submucosal) surface, and maintains this position up to its attachment at the modiolar base. Vascular supply Orbicularis oris is supplied mainly by the superior and inferior labial branches of the facial artery, the mental and infraorbital branches of the maxillary artery and
the transverse facial branch of the superficial temporal artery. Nerve supply Orbicularis oris is supplied by the buccal and mandibular branches of the facial nerve. Actions The actions of orbicularis oris are considered in detail on page 508. Incisivus labii superioris
Incisivus labii superioris has a bony origin from the floor of the incisive fossa of the maxilla above the eminence of the lateral incisor tooth. Initially it lies deep to orbicularis oris pars peripheralis superior. Arching laterally, its fibre bundles become intercalated between, and parallel to, the orbicular bundles. Approaching the modiolus, it segregates into superficial and deep parts: the former blends partially with levator anguli oris and attaches to the body and apex of the modiolus and the latter is attached to the superior cornu and base of the modiolus. Incisivus labii inferioris
Incisivus labii inferioris, an accessory muscle of the orbicularis oris muscle complex, has many features in common with incisivus labii superioris. Its osseous attachment is to the floor of the incisive fossa of the mandible, lateral to mentalis and below the eminence of the lateral incisor tooth. Curving laterally and upwards, it blends to some extent with orbicularis oris pars peripheralis inferior before reaching the modiolus, where superficial bundles attach to the apex and body, and deep bundles attach to the base and inferior cornu. page 506 page 507
Figure 29.6 A, The principal sulci, creases and ridges of the face referred to at various points in the text. Note particularly those defining the 'labial hexagon' (see text). B, The disposition of the modiolus and orbicularis oris pars peripheralis and pars marginalis (on the left); the successively transected laminae of the direct labial tractors of both upper and lower lips (on the right). C, Parasagittal section of the upper lip in repose. On the left is thin skin with oblique hair follicles; on the right is thick mucosa with mucous glands and mucosal shelf; between them is the vermilion zone. D, as C but slightly contracted, forming a narrowed profile (labial cord). E, Superimposed outlines of C (magenta) and D (blue).
Platysma (Fig. 29.5)
Platysma is described as a muscle of the neck (p. 535) but it is considered here as a contributor to the orbicularis oris muscle complex. The pars mandibularis attaches to the lower border of the body of the mandible. Posterior to this attachment, a substantial flattened bundle separates and passes superomedially to the lateral border of depressor anguli oris, where a few fibres join this muscle. The remainder continue deep to depressor anguli oris and reappear at its medial border. Here they continue within the tissue of the lateral half of the lower lip, as a direct labial tractor, platysma pars labialis. Pars labialis occupies the interval between depressor anguli oris and depressor labii inferioris and is in the same plane as these muscles. The adjacent margins of all three muscles blend and they have similar labial attachments. Platysma pars modiolaris constitutes all the remaining bundles posterior to pars labialis, other than a few fine fascicles that end directly in buccal dermis or submucosa. Pars modiolaris is posterolateral to depressor anguli oris and passes superomedially, deep to risorius, to apical and subapical modiolar attachments. Risorius
Risorius is a highly variable muscle that ranges from one or more slender fascicles to a wide, thin superficial fan. Its peripheral attachments may include some or all of the following: the zygomatic arch, parotid fascia, fascia over the masseter anterior to the parotid, fascia enclosing pars modiolaris of platysma, and fascia over the mastoid process. Its fibres converge to apical and subapical attachments at the modiolus. page 507 page 508
Vascular supply Risorius is supplied mainly by the superior labial branch of the facial artery. Nerve supply Risorius is supplied by buccal branches of the facial nerve. Actions Risorius pulls the corner of the mouth laterally in numerous facial activities, including grinning and laughing.
MOVEMENTS OF THE FACE AND LIPS Direct labial tractors (Fig. 29.6)
Direct labial tractors, as their name suggests, pass directly into the tissues of the lips and not via the modioli. In broad terms, the force exerted by tractors is directed vertically at an approximate right angle to the oral fissure. Their action will therefore elevate and/or evert the whole, or part, of the upper lip and depress and/or evert the whole, or part, of the lower lip. The tractors are, from medial to lateral: the labial part of levator labii superioris alaequae nasi, levator labii
superioris and zygomaticus minor in the upper lip, and depressor labii inferioris and platysma pars labialis in the lower lip. In both upper and lower lips the tractors blend into a continuous sheet that divides into a series of superimposed coronal sheets which are anterior to the muscle bundles of pars peripheralis orbicularis oris as they enter the free lip. The sheets may be divided into three groups at increasing depths from the skin surface, each with a distinct zone of attachment. The superficial group comprises a succession of fine fibre bundles which curve anteriorly a short distance before attaching in a series of horizontal rows to the dermis between the hair follicles, sebaceous glands and sweat glands. The intermediate group attaches to the dermis of the vermilion zone, which they reach by two routes: the more superficial bundles continue past the skin/vermilion junction, then curve posteriorly over pars marginalis orbicularis oris to punctate attachments on the ventral half of the dermis of the vermilion zone; the deeper bundles first pass posteriorly between pars peripheralis and pars marginalis, then curve anteriorly to punctate attachments on the dorsal half of the dermis of the vermilion zone. The deep group is closely applied to the anterior surface of pars peripheralis orbicularis oris, and sends fine tractor fibres between its parallel bundles to attach posteriorly into the submucosa and periglandular connective tissue. Movements of the lips
The various groups of direct labial tractors may act together or individually, and their effects may involve a complete labial quadrant, or be restricted to a short segment. For example, partial contraction of the superior labial tractors can result in localized elevation of a segment of the upper lip, in a postural expression reminiscent of the 'canine snarl'. Normally, however, the activity of the tractors is modified by the superimposed activity of orbicularis oris and the modiolar muscles. The resultant actions range from delicate adjustments of the tension and profile of the lip margins to large increases of the oral fissure with eversion of the lips. Lip protrusion is passive in its initial stages. It may be suppressed by powerful contraction of the whole of orbicularis oris or enhanced by selective activation of parts of the direct labial tractors. However, lip movements must accommodate separation of the teeth brought about by mandibular depression at the temporomandibular joints. Beyond a certain range of mouth opening, labial movements are almost completely dominated by mandibular movements. Thus over the last 2.5-3 cm interincisal distance of wide jaw separation, strong contraction of orbicularis oris cannot effect lip contact, and instead it causes fullthickness inflection of upper and lower lips, including the vermilion zone, towards the oral cavity, wrapping them around the incisal edges, canine cusps and premolar occlusal surfaces. The involvement of the lips in speech is described elsewhere, but some aspects relevant to the actions of orbicularis oris pars marginalis will be described here. Contraction of marginalis is considered to alter the cross-sectional profile of the free margin of the vermilion zone such that both the gentle bulbous profile of the upper lip and the smooth posterosuperior convexity of the lower lip change to a narrow, symmetrical triangular profile. The transformed rims, whose length and tension can be delicately controlled, have been named labial cords. They are known to be involved in the production of some consonantal (labial) sounds. A labial cord may also function as a 'vibrating reed' in whistling or playing a wind instrument such as the trumpet. The modiolus and its role in facial movements Modiolus
On each side of the face a number of muscles converge towards a focus just lateral to the buccal angle, where they interlace to form a dense, compact, mobile, fibromuscular mass called the modiolus (Fig. 29.6B). This can be palpated most effectively by using the opposed thumb and index finger to compress the mucosa and skin simultaneously. At least nine muscles, depending on the classification employed, are attached to each modiolus. Moreover, the muscles lie in different planes, their modiolar stems are often spiralized, and most divide into two bundles, some into three or four, each of them interlacing and attaching in a distinctive way. Not surprisingly, therefore, the three-dimensional organization of the modiolus has proved difficult to analyse. The shape and dimensions of the modiolus are given approximately because they are subject to individual, age, sexual and ethnic variation. Furthermore the modiolus has no precise histological boundaries, and is an irregular zone where dense, compact interlacing tissue grades into the stems of individually recognizable muscles. The modiolus has the rough form of a blunt cone. The
base of the cone (basis moduli) is adjacent and adherent to the mucosa. It is roughly elliptical in outline and extends vertically c.20 mm above and 20 mm below a horizontal line through the buccal angle. It also extends laterally a similar distance from the angle. The blunt apex of the cone (apex moduli) is c.4 mm across, and is centred c.12 mm lateral to the buccal angle. From mucosa to dermis the thickness of the mass is c.10 mm, divided approximately equally into basal, central and apical parts. The central body has an oblique fibrous cleft or channel that transmits the facial artery, an arrangement that may limit the extent to which it is compressed by contraction of the buccolabial musculature. The cone shape is modified by two round-edged flanges (or cornua) that extend into the lateral free lip tissues above and below the corner of the mouth. The tip of the superior cornu extends 5-5 mm medial to the buccal angle, the tip of the inferior cornu only 3-5 mm. With these additions, the modiolar base becomes kidneyshaped, with the buccal angle projecting towards the hilum. The apex of the modiolus is deep and adherent to the panniculus carnosus, which extends posteromedially as a thin sloping sheet down to the buccal angle. There, its free border forms a crescentic, narrow, flexible, subcutaneous fibroelastic cord that accommodates the varying postures of the modioli, lips, mouth and jaws. Modiolar movements
Controlled three-dimensional mobility of the modioli enables them to integrate the activities of the cheeks, lips and oral fissure, the oral vestibule and the jaws. Such activities include biting, chewing, drinking, sucking, swallowing, changes in vestibular contents and pressure, the innumerable subtle variations involved in speech, the modulation (and occasional generation) of musical tones, production of harsher sounds in shouting and screaming, crying, and all the permutations of facial expression, ranging from mere hints to gross distortion, symmetrical or asymmetrical. Major modiolar movements appear to involve many, if not all, of its associated muscles, and there is little value in considering the actions of the individual muscles in isolation. While the most obvious determinant of modiolar position and mobility is the balance between the forces exerted by muscles that are directly attached to it, another influential factor is the degree of separation or 'gape' between the upper and lower teeth. Starting from the occlusal position, and with the lips maintained in contact, the teeth can be separated by c.1.25 cm near the midline, and the mentolabial sulcus descends by a similar distance. With further separation the lips part, and as gape increases to its maximum, interlabial and interdental distances approach 4 cm, at which point the mentolabial sulcus has descended a further 2 cm. In this posture the modiolus has descended c.1 cm to lie over the interdental space, into which its basal and surrounding buccal mucosa projects a few millimetres, and its cornua diverge into their respective lips at an obtuse angle to each other, the dispositions of the modiolar muscles being correspondingly modified. The general hexagonal shape of the labial area changes as the mouth and jaws open progressively. In maximal opening, the distance between the superior and inferior boundaries has increased by 3-3.5 cm at the centre; the transverse distance between its lateral angles has decreased by c.1 cm and the angles are obtuse; the nasolabial sulci are longer, straighter and more vertical; and the inferior buccolabial sulci are less deep and curved. These soft tissue changes radiate from the bilateral modioli. page 508 page 509
With the lips in contact and the teeth in tight occlusion, the modiolus can move a few millimetres in all directions. However, mobility is maximal when there is 2-3 mm clearance between the teeth: the apex of the modiolus may then move vertically upwards c.10 mm, downwards 5 mm, posterolaterally 10 mm, and anteromedially 10 mm, these movements occurring in the curved planes of the cheek and lips. Specific movements of the modiolus may occur to any point, and along any path, within the boundaries of the envelope of movement thus defined. When the mouth is opened wide, the modiolus becomes immobile. From the neutral position the modiolus may be displaced superficially along its apicobasal axis for up to 5-10 mm by liquids or solids in the vestibule, or by an increase in air pressure that 'balloons' the cheeks and lips. Many activities take place in three phases. Initially, a particular modiolar muscle group becomes dominant over its antagonists and the modiolus is rapidly relocated. Next, the modiolus is transiently fixed in this new site by simultaneous contraction of modiolar muscles, principally zygomaticus major, levator anguli oris, depressor anguli oris, platysma pars modiolaris, and this provides a fixed base from which the main physiological effectors, buccinator and orbicularis oris, carry out their specific actions. These actions are usually integrated with partial separation or closure of the jaws, and with varying degrees of activity in the direct
labial tractors. All these factors combine to determine the positions of the lips and oral fissure from moment to moment. Modiolar movements may be bilaterally symmetrical, unilateral or asymmetrical.
MUSCLES OF MASTICATION Two muscles of mastication are to be seen on the face; masseter, which covers the ramus of the mandible and on which the parotid gland lies, and temporalis, which lies over the temporal fossa. These muscles are described in detail with the other main muscles of mastication on pages 519 and 520.
© 2008 Elsevier
VASCULAR SUPPLY AND LYMPHATIC DRAINAGE ARTERIAL SUPPLY TO THE FACE The main arterial supply to the face is derived from the facial and superficial temporal arteries. Blood is also supplied by branches of the maxillary and ophthalmic arteries. The back of the scalp is supplied by the posterior auricular and occipital arteries. There are numerous anastomoses between the branches. UPDATE Date Added: 15 May 2006 Abstract: The anatomy of the angiosomes of the head and neck Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=10845282&query_hl=31&itool=pubmed_docsum The angiosomes of the head and neck: Anatomic study and clinical applications. Houseman ND, Taylor GI, Pan WR: Plast Reconstr Surg 105:2287-2313, 2000.
FACIAL ARTERY (Fig. 29.7) The facial artery arises in the neck from the external carotid artery. It passes onto the face at the anteroinferior border of masseter (where its pulse can be felt as it crosses the mandible). It is superficial and at first lies beneath platysma. It is covered by skin, the fat of the cheek and, near the angle of the mouth, by zygomaticus major and risorius. It pursues a tortuous course - that presumably allows it to stretch when the face is distorted during jaw opening - by the side of the nose towards the medial corner of the eye. Buccinator and levator anguli oris lie deep to the facial artery, and it may pass over or through levator labii superioris. At its termination it is embedded in levator labii superioris alaequae nasi. Occasionally, the facial artery barely extends beyond the angle of the mouth, in which case its normal territory beyond this region is taken over by an enlarged transverse facial branch from the superficial temporal artery, and by branches from the contralateral facial artery. The facial vein is posterior, running a more direct course across the face. At the anterior border of masseter, the two vessels are in contact, but in the neck the vein is superficial. The facial artery supplies branches to the adjacent muscles and skin of the face. Its named branches on the face are the premasseteric artery, the superior and inferior labial arteries and the lateral nasal artery. The part of the artery distal to its terminal branch is called the angular artery. Premasseteric artery This small inconstant artery passes upwards along the anterior border of masseter and supplies the adjacent tissue. Inferior labial artery The inferior labial artery arises near the angle of the mouth, passes upwards and forwards under depressor anguli oris, penetrates orbicularis oris and runs sinuously near the margin of the lower lip, between the muscle and the mucous membrane. It supplies the inferior labial glands, mucous membrane and muscles, and anastomoses with its fellow of the opposite side and with the mental branch of the inferior alveolar artery. Superior labial artery
page 509 page 510
Figure 29.7 The arteries of the left side of the face and their main branches.
Larger and more tortuous than the inferior labial artery, the superior labial artery has a similar course along the superior labial margin, between the mucous membrane and orbicularis oris. It anastomoses with its fellow of the opposite side, and supplies the upper lip. It gives off a septal branch, which ramifies anteroinferiorly in the nasal septum, and an alar branch. Lateral nasal artery This is given off by the side of the nose and supplies the dorsum and alae of nose, and anastomoses with its fellow of the opposite side. It may be replaced by a branch from the superior labial artery.
SUPERFICIAL TEMPORAL ARTERY (Fig. 29.7) This is the smaller terminal branch of the external carotid artery. It arises in the parotid gland behind the neck of the mandible, where it is crossed by temporal and zygomatic branches of the facial nerve. Initially deep, it becomes superficial as it passes over the posterior root of the zygomatic process of the temporal bone, where its pulse can be felt. It then runs up the scalp for c.4 cm and divides into frontal (anterior) and parietal (posterior) branches. It is accompanied by corresponding veins, and the auriculotemporal nerve lies just posterior to it. The superficial temporal artery supplies skin and muscles at the side of the face and in the scalp, the parotid gland and the temporomandibular joint. Its named branches are the transverse facial, auricular, zygomatico-orbital, middle temporal, frontal and parietal arteries. There are variations in the relative sizes of the frontal, parietal and transverse facial branches, in that the frontal and parietal branches may be absent, and the transverse facial may replace a shortened transverse facial artery. Transverse facial artery
The transverse facial artery arises before the superficial temporal artery emerges from the parotid gland. It traverses the gland, crosses masseter between the parotid duct and the zygomatic arch (accompanied by one or two facial nerve branches) and divides into numerous branches that supply the parotid gland and duct, masseter and adjacent skin. The branches anastomose with the facial, masseteric, buccal, lacrimal and infraorbital arteries, and may have a direct origin from the external carotid artery. Auricular artery The branches of the auricular artery are distributed to the lobule and lateral surface of the auricle and the external acoustic meatus. Zygomatico-orbital artery The zygomatico-orbital artery may arise independently from the superficial temporal artery or from its middle temporal or parietal branches. It runs close to the upper border of the zygomatic arch, between the two layers of temporal fascia, to the lateral orbital angle. It supplies orbicularis oculi and anastomoses with the lacrimal and palpebral branches of the ophthalmic artery. A welldeveloped zygomatico-orbital artery is associated with a delayed division into frontal and parietal branches. Middle temporal artery The middle temporal artery arises just above the zygomatic arch and perforates the temporal fascia to supply temporalis. It anastomoses with the deep temporal branches of the maxillary artery. Frontal (anterior) branch The artery passes upwards towards the frontal tuberosity and supplies adjacent muscles, skin and pericranium in this region. It anastomoses with its fellow of the opposite side and with the supraorbital and supratrochlear branches of the ophthalmic artery. Parietal (posterior) branch The parietal branch is larger than the frontal branch of the superficial temporal artery, and curves upwards and backwards. It remains superficial to the temporal fascia, and anastomoses with its fellow of the opposite side and with the posterior auricular and occipital arteries. Facial branches of the maxillary artery
The maxillary artery is the larger of the two terminal branches of the external carotid artery, and has three branches that supply the face, namely the mental, buccal and infraorbital arteries. Mental artery The mental artery arises from the first part of the maxillary artery as a terminal branch of the inferior alveolar artery. It emerges onto the face from the mandibular canal at the mental foramen, and supplies muscles and skin in the chin region. The mental artery anastomoses with the inferior labial and submental arteries. Buccal artery The buccal artery is a branch of the second part of the maxillary artery. It emerges onto the face from the infratemporal fossa and crosses buccinator to supply the cheek. The buccal artery anastomoses with the infraorbital artery and with branches of the facial artery. Infraorbital artery The infraorbital artery arises from the third part of the maxillary artery. It runs through the infraorbital foramen and onto the face, supplying the lower eyelid, the lateral aspect of the nose and the upper lip. The infraorbital artery has extensive anastomoses with the transverse facial and buccal arteries and branches of the ophthalmic and facial arteries.
Facial branches of the ophthalmic artery
The ophthalmic artery is a branch of the internal carotid artery. Its supratrochlear, supraorbital, lacrimal, medial palpebral, dorsal nasal and external nasal branches supply the face. Supratrochlear artery The supratrochlear artery emerges from the orbit onto the face at the frontal notch. It supplies the medial parts of the upper eyelid, forehead and scalp. The supratrochlear artery anastomoses with the supraorbital artery and with its contralateral fellow. Supraorbital artery The supraorbital artery leaves the orbit through the supraorbital notch (or foramen). It divides into superficial and deep branches, supplying skin and muscle of the upper eyelid, forehead and scalp. It anastomoses with the supratrochlear artery, frontal branch of the superficial temporal and its contralateral fellow. At the supraorbital margin it often sends a branch to the diploë of the frontal bone and may also supply the mucoperiosteum in the frontal sinus. Lacrimal artery The lacrimal artery appears on the face at the upper lateral corner of the orbit and supplies the lateral part of the eyelids. Within the orbit, it gives off a zygomatic artery which subdivides into zygomaticofacial and zygomaticotemporal arteries. The zygomaticofacial artery passes through the lateral wall of the orbit to emerge onto the face at the zygomaticofacial foramen, and supplies the region overlying the prominence of the cheek. The zygomaticotemporal artery also passes through the lateral wall of the orbit, via the zygomaticotemporal foramen, to supply the skin over the non-beard part of the temple. The lacrimal artery anastomoses with the deep temporal branch of the maxillary artery and the transverse facial branch of the superficial temporal artery. Medial palpebral arteries Superior and inferior medial palpebral arteries arise from the ophthalmic artery below the trochlea. They descend behind the nasolacrimal sac to enter the eyelids where each divides into two branches that course laterally along the edges of the tarsal plates to form the superior and inferior arches, and supply the eyelids. They anastomose with branches of the supraorbital, zygomatico-orbital and lacrimal arteries. The inferior arch also anastomoses with the facial artery. External (dorsal) nasal artery The external nasal artery is a terminal branch of the anterior ethmoidal artery from the ophthalmic artery. It supplies skin on the external nose, and emerges at the junction of the nasal bone and the lateral nasal cartilage. Occipital artery
The occipital artery runs in a groove on the temporal bone, medial to the mastoid process. It arises in the neck from the external carotid artery, and enters the back of the scalp by piercing the investing layer of deep cervical fascia connecting the cranial attachments of trapezius and sternocleidomastoid, accompanied by the greater occipital nerve. Tortuous branches run between the skin and the occipital belly of occipitofrontalis, anastomosing with the opposite occipital, posterior auricular and superficial temporal arteries as well as the transverse cervical branch of the subclavian artery. They supply the occipital belly of occipitofrontalis and skin and pericranium associated with the scalp as far forward as the vertex. There may be a meningeal lateral branch, traversing the parietal foramen. page 510 page 511
Posterior auricular artery
The posterior auricular artery arises in the neck from the external carotid artery,
ascends between the auricle and mastoid process and gives off an auricular branch supplying the cranial surface of the auricle and an occipital branch to supply the occipital belly of occipitofrontalis and the scalp behind and above the auricle. The posterior auricular artery anastomoses with the occipital artery.
VEINS OF THE FACE (Fig. 29.8) The veins of the face are subject to considerable variations, and therefore the following description concerns those which are relatively constant. Supratrochlear vein
The supratrochlear vein starts on the forehead from a venous network connected to the frontal tributaries of the superficial temporal vein. Veins from this network form a single trunk, descending near the midline parallel with its fellow to the bridge of the nose. Each vein is joined by a nasal arch across the nose. The veins then diverge, each joining a supraorbital vein to form the facial vein near the medial canthus of the eye. Supraorbital vein
The supraorbital vein begins near the zygomatic process of the frontal bone, connecting with branches of the superficial and middle temporal veins. It passes medially above the orbital opening, pierces the orbicularis oculi and unites with the supratrochlear vein near the medial canthus of the eye to form the facial vein. A branch passes through the supraorbital notch to connect with the superior ophthalmic vein. In the notch it receives veins from the frontal sinus and frontal diploë. Facial vein
Figure 29.8 The veins of the left side of the head and neck. Parts of the left sternocleidomastoid and platysma have been excised to expose the trunk of the internal jugular vein. The external jugular vein is visible through the lower part of
platysma.
The facial vein is the main vein of the face. After receiving the supratrochlear and supraorbital veins, it travels obliquely downwards by the side of the nose, passes under zygomaticus major, risorius and platysma, descends to the anterior border and then passes over the surface of masseter. It crosses the body of the mandible, and runs in the neck to drain into the internal jugular vein. The uppermost segment of the facial vein - above its junction with the superior labial vein - is also termed the angular vein. The facial vein initially lies behind the more tortuous facial artery, but crosses it at the lower border of the mandible. The fact that the vein lacks valves and that it is connected with the cavernous sinus is of considerable clinical significance in terms of the spread of infection. Tributaries Near its origin, the facial vein connects with the superior ophthalmic vein, both directly and via the supraorbital vein, and so is linked to the cavernous sinus. The facial vein receives tributaries from the side of the nose and, below this, an important deep facial vein from the pterygoid venous plexus. It also receives the inferior palpebral, superior and inferior labial, buccinator, parotid and masseteric veins, and other tributaries which join it below the mandible. Cavernous sinus thrombosis (See also Section 2 and intracranial venous sinuses) Any spreading infection involving the cheek, upper lip, anterior nares or even an upper incisor or canine tooth can result in cavernous sinus thrombosis (p. 279). Superficial temporal vein
The superficial temporal vein begins in a widespread network joined across the scalp to the contralateral vein and to the ipsilateral supratrochlear, supraorbital, posterior auricular and occipital veins that all drain the same network. Anterior and posterior tributaries unite above the zygomatic arch to form the superficial temporal vein. Accompanying its artery (behind in c.70% of cases), the vein crosses the posterior root of the zygoma and enters the parotid gland. Here, the superficial temporal vein joins the maxillary vein, to form the retromandibular vein. Tributaries
The tributaries are the parotid veins, rami draining the temporomandibular joint, anterior auricular veins and the transverse facial vein. The middle temporal vein receives the orbital vein (formed by the lateral palpebral veins), and passes back between layers of temporal fascia, which it pierces to join the superficial temporal vein just above the level of the zygomatic arch. Buccal, mental and infraorbital veins
The buccal, mental and infraorbital veins drain the cheek and chin regions and pass into the pterygoid venous plexus page 511 page 512
Posterior auricular and occipital veins
The posterior auricular vein arises in a parieto-occipital network that also drains into tributaries of the occipital and superficial temporal veins. It descends behind the auricle to join the posterior division of the retromandibular vein in, or just below, the parotid gland, to form the external jugular vein. It receives a stylomastoid vein and tributaries from the cranial surface of the auricle, drains the region of the scalp behind the ear and drains into the external jugular vein. The occipital vein begins in a posterior network in the scalp, pierces the cranial attachment of trapezius, turns into the suboccipital triangle and joins the deep cervical and vertebral veins. It may be joined by a vein draining the diploë in the occipital bone and then passes to either the internal jugular, posterior auricular, deep cervical or vertebral veins. Emissary veins connect the occipital vein to the
intracranial venous sinuses via the mastoid and parietal foramina and through the posterior condylar canal and occipital protuberances.
LYMPHATIC DRAINAGE OF THE FACE AND SCALP (Fig. 31.1) Lymph vessels from the frontal region above the root of the nose drain to the submandibular nodes. Vessels from the rest of the forehead, temporal region, upper half of the lateral auricular aspect and anterior wall of the external acoustic meatus drain to the superficial parotid nodes, just anterior to the tragus, on or deep to the parotid fascia. These nodes also drain lateral vessels from the eyelids and skin of the zygomatic region. Their efferent vessels pass to the upper deep cervical nodes. A strip of scalp above the auricle, the upper half of the cranial aspect and margin of the auricle, and the posterior wall of the external acoustic meatus all drain to the upper deep cervical and retroauricular nodes. The retroauricular nodes, superficial to the mastoid attachment of sternocleidomastoid and deep to auricularis posterior, drain to the upper deep cervical nodes. The auricular lobule, floor of the external acoustic meatus and skin over the mandibular angle and lower parotid region are drained to the superficial cervical or upper deep cervical nodes. Superficial cervical nodes spread along the external jugular vein superficial to sternocleidomastoid. Some efferents pass round the anterior border of sternocleidomastoid to the upper deep cervical nodes; others follow the external jugular vein to the lower deep cervical nodes in the subclavian triangle. The occipital region of the scalp is drained partly to the occipital nodes, and partly by a vessel that runs along the posterior border of sternocleidomastoid to the lower deep cervical nodes. Occipital nodes are commonly superficial to the upper attachment of trapezius, but occasionally lie in the superior angle of the posterior triangle. There are usually three submandibular nodes, internal to the deep cervical fascia in the submandibular triangle. There is one at the anterior pole of the submandibular gland, and two flanking the facial artery as it reaches the mandible. Other nodes are often embedded in the gland or deep to it. Submandibular nodes drain a wide area, including vessels from the submental, buccal and lingual groups of nodes and their efferents pass to the upper and lower deep cervical nodes. The external nose, cheek, upper lip and lateral parts of the lower lip drain directly to the submandibular nodes; the afferent vessels may have a few buccal nodes along their course and near the facial vein. The mucous membrane of the lips and cheek drains to the submandibular nodes and the lateral part of the cheek drains to the parotid nodes. The central part of the lower lip, buccal floor and tip of the tongue drain to the submental nodes on mylohyoid between the anterior bellies of the digastric muscles. They receive afferents from both sides, some decussate across the chin, and their efferents pass to the submandibular and jugulo-omohyoid nodes.
© 2008 Elsevier
INNERVATION OF THE FACE AND SCALP The numerous muscles of facial expression are supplied by the facial nerve, while the two muscles of mastication that relate to the face are innervated by the mandibular division of the trigeminal nerve. The sensory innervation is primarily from the three divisions of the mandibular nerve, with smaller contributions from the cervical spinal nerves. The detailed innervation of the auricle is considered on page 651. UPDATE Date Added: 17 December 2004 Abstract: New one-stage nerve pedicle grafting technique using the great auricular nerve for reconstruction of facial nerve defects. Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=15237353 New one-stage nerve pedicle grafting technique using the great auricular nerve for reconstruction of facial nerve defects. Koshima I, Nanba Y, Tsutsui T, Takahashi Y, Itoh S. New one-stage nerve pedicle grafting technique using the great auricular nerve for reconstruction of facial nerve defects. J Reconst Microsurg. 2004; Jul;20(5): 357-61.
TRIGEMINAL NERVE Three large areas of the face can be mapped out to indicate the peripheral nerve fields associated with the three divisions of the trigeminal nerve. The fields are not horizontal but curve upwards (Fig. 29.9A), apparently because the facial skin moves upwards with growth of the brain and skull. Embryologically, each division of the trigeminal nerve is associated with a developing facial process which gives rise to a specific area of the face in the adult. Thus the ophthalmic nerve is associated with the frontonasal process, the maxillary nerve with the maxillary process and the mandibular nerve with the mandibular process.
page 512 page 513
Figure 29.9 A, The sensory nerves of the left side of the scalp, face and neck, and the branches of the facial nerve, which are distributed to the muscles of 'facial expression'. The pinna has been reflected forwards. B, Cutaneous innervation of the face and neck, showing dermatomes.
Ophthalmic nerve (Fig. 29.9B)
The cutaneous branches of the ophthalmic nerve supply the conjunctiva, skin over the forehead, upper eyelid and much of the external surface of the nose. Supratrochlear nerve
The supratrochlear nerve is the smaller terminal branch of the frontal nerve. It runs anteromedially in the roof of the orbit, passes above the trochlea, and supplies a descending filament to the infratrochlear branch of the nasociliary nerve. The nerve emerges between the trochlea and the supraorbital foramen at the frontal notch, curves up on the forehead close to the bone with the supratrochlear artery and supplies the conjunctiva and the skin of the upper eyelid. It then ascends beneath the corrugator and the frontal belly of occipitofrontalis before dividing into branches which pierce these muscles to supply the skin of the lower forehead near the midline. Supraorbital nerve
The supraorbital nerve is the larger terminal branch of the frontal nerve. It traverses the supraorbital notch or foramen and supplies palpebral filaments to the upper eyelid and conjunctiva. It ascends on the forehead with the supraorbital artery, and divides into medial and lateral branches, which supply the skin of the scalp nearly as far back as the lambdoid suture. These branches are at first deep to the frontal belly of the occipitofrontalis. The medial branch perforates the muscle to reach the skin, while the lateral pierces the epicranial aponeurosis. Lacrimal nerve
The lacrimal nerve is the smallest of the main ophthalmic branches and pierces the orbital septum to end in the lateral region of the upper eyelid, which it supplies. It joins filaments of the facial nerve. Occasionally it is absent, in which case it is replaced by the zygomaticotemporal nerve: the relationship is reciprocal, and when the zygomaticotemporal nerve is absent it is replaced by a branch of the lacrimal nerve. Infratrochlear nerve
The infratrochlear nerve branches from the nasociliary nerve. It leaves the orbit below the trochlea and supplies the skin of the eyelids, the conjunctiva, lacrimal sac, lacrimal caruncle and the side of the nose above the medial canthus. External nasal nerve (Fig. 29.9A,B)
The external nasal nerve is the terminal branch of the anterior ethmoidal nerve. It descends through the lateral wall of the nose, and supplies the skin of the nose below the nasal bones, excluding the alar portion around the external nares. Maxillary nerve
The maxillary nerve passes through the orbit to supply the skin of the lower eyelid, the prominence of the cheek, the alar part of the nose, part of the temple, and the upper lip. It has three cutaneous branches, namely the zygomaticotemporal, zygomaticofacial and infraorbital nerves. Zygomaticotemporal nerve
The zygomaticotemporal nerve is a terminal branch of the zygomatic nerve. It traverses a canal in the zygomatic bone to emerge into the anterior part of the temporal fossa, ascends between the bone and temporalis and finally pierces the temporal fascia c.2 cm above the zygomatic arch to supply the skin of the temple. It communicates with the facial and auriculotemporal nerves. As it pierces the deep layer of the temporal fascia it sends a slender twig between the two layers of the fascia towards the lateral angle of the eye. This lacrimal ramus conveys parasympathetic postganglionic fibres from the pterygopalatine ganglion to the lacrimal gland. UPDATE Date Added: 22 May 2006 Abstract: Surgical anatomy of the zygomaticotemporal nerve in the orbit and temporal area Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=15167230&query_hl=24&itool=pubmed_docsum Zygomaticotemporal nerve passage in the orbit and temporal area. Hwang K, Suh MS, Lee SI, Chung IH: J Craniofac Surg 15:209-214, 2004. UPDATE Date Added: 25 October 2005 Publication Services, Inc. Abstract: Zygomaticotemporal nerve passage in the orbit and temporal area Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve=pubmed=Abstract_uids=15167230_hl=9 Zygomaticotemporal nerve passage in the orbit and temporal area. Hwang K, Suh MS, Lee SI et al: J Craniofac Surg 15(2):209-214, 2004.
Zygomaticofacial nerve
The zygomaticofacial nerve is a terminal branch of the zygomatic nerve. It traverses the inferolateral angle of the orbit, and emerges on the face through a foramen in the zygomatic bone. It next perforates orbicularis oculi to supply the skin on the prominence of the cheek. It forms a plexus with zygomatic branches of the facial nerve and palpebral branches of the maxillary nerve. Occasionally the nerve is absent. Infraorbital nerve
The infraorbital nerve emerges onto the face at the infraorbital foramen, where it lies between levator labii superioris and levator anguli oris. It divides into three further groups of branches. The palpebral branches ascend deep to orbicularis oculi, and pierce the muscle to supply the skin in the lower eyelid and join with the facial and zygomaticofacial nerves near the lateral canthus. Nasal branches supply the skin of the side of the nose and of the movable part of the nasal septum, and join the external nasal branch of the anterior ethmoidal nerve. Superior labial branches, large and numerous, descend behind levator labii superioris, to supply the skin of the anterior part of the cheek and upper lip. They are joined by branches from the facial nerve to form the infraorbital plexus. Mandibular nerve
The mandibular nerve supplies skin over the mandible, the lower lip, the fleshy part of the cheek, part of the auricle of the ear and part of the temple via the buccal, mental and auriculotemporal nerves. Buccal nerve
The buccal nerve emerges onto the face from behind the ramus of the mandible and passes laterally in front of the masseter to unite with the buccal branches of the facial nerve. It supplies the skin over the anterior part of buccinator. Mental nerve
The mental nerve is the terminal branch of the inferior alveolar nerve. It enters the face through the mental foramen, where it is directed backwards. It supplies the skin of the lower lip. Auriculotemporal nerve
The auriculotemporal nerve emerges onto the face behind the temporomandibular joint within the superior surface of the parotid gland. It ascends posterior to the superficial temporal vessels, over the posterior root of the zygoma, and divides into superficial temporal branches. The cutaneous branches of the auriculotemporal nerve supply the tragus and part of the adjoining auricle of the ear and the posterior part of the temple. It communicates with the facial nerve, usually by two rami that pass anterolaterally behind the neck of the mandible. The communications with the temporofacial division of the facial nerve anchor the facial nerve close to the lateral surface of the condylar process of the mandible, limiting its mobility during surgery. Communications with the temporal and zygomatic branches loop around the transverse facial and superficial temporal vessels.
FACIAL NERVE The facial nerve emerges from the base of the skull at the stylomastoid foramen. At this point the facial nerve lies c.9 mm from the posterior belly of the digastric muscle and 11 mm from the bony external acoustic meatus. It gains access to the face by passing through the substance of the parotid gland. Although mainly motor, there are some cutaneous fibres from the facial nerve which accompany the auricular branch of the vagus and which probably innervate the skin on both auricular aspects, in the conchal depression and over its eminence. Close to the stylomastoid foramen the facial nerve gives off the posterior auricular nerve, which supplies the occipital belly of occipitofrontalis, and some of the auricular muscles, and the nerves to the posterior belly of digastric and stylohyoid. The nerve then enters the parotid gland high up on the posteromedial surface and passes forwards and downwards behind the mandibular ramus. Within the substance of the gland the facial nerve branches into the temporofacial and cervicofacial trunks, just behind (within c.5 mm) the retromandibular vein. In c.90% of cases, the two trunks lie superficial to the vein, in intimate contact with it. Occasionally the trunks pass beneath the retromandibular vein (temporofacial trunk c.9%; cervicofacial trunk c.2%). The trunks branch further to form a parotid plexus (pes anserinus), which exhibits variations in branching pattern. Five main terminal branches arise from the plexus and diverge within the gland. They leave
the parotid gland by its anteromedial surface, medial to its anterior margin and supply the muscles of facial expression. page 513 page 514
The temporal branches are generally multiple and pass across the zygomatic arch to the temple to supply intrinsic muscles on the lateral surface of the auricle, and the anterior and superior auricular muscles. They join with the zygomaticotemporal branch of the maxillary nerve and the auriculotemporal branch of the mandibular nerve. The more anterior branches supply the frontal belly of occipitofrontalis, orbicularis oculi and corrugator and join the supraorbital and lacrimal branches of the ophthalmic nerve. Zygomatic branches are generally multiple and cross the zygomatic bone to the lateral canthus of the eye, to supply the orbicularis oculi and join filaments of the lacrimal nerve and zygomaticofacial branch of the maxillary nerve. The branches may also help supply muscles associated with the buccal branch of the facial nerve. The buccal branch has a variable origin and passes horizontally to a distribution below the orbit and around the mouth. It is usually single, but two branches occur in 15% of cases. The buccal branch has a close relationship to the parotid duct, and usually lies below it. Superficial branches run deep to subcutaneous fat and the superficial musculo-aponeurotic system (SMAS) (p. 499). Some branches pass deep to procerus and join the infratrochlear and external nasal nerves. Upper deep branches pass under zygomaticus major and levator labii superioris, supply them and form an infraorbital plexus with the superior labial branches of the infraorbital nerve. They also supply levator anguli oris, zygomaticus minor, levator labii superioris alaequae nasi and the small nasal muscles. These branches are sometimes described as lower zygomatic branches. Lower deep branches supply the buccinator and orbicularis oris, and join filaments of the buccal branch of the mandibular nerve. The marginal mandibular branches, of which there are usually two, run forwards towards the angle of the mandible under platysma, at first superficial to the upper part of the digastric triangle, then turning up and forwards across the body of the mandible to pass under depressor anguli oris. The branches supply risorius and the muscles of the lower lip and chin, and join the mental nerve. The marginal mandibular branch has an important surgical relationship with the lower border of the mandible, and may pass below the lower border with a reported incidence varying between 20% and 50%, the furthest distance being 1.2 cm. The cervical branch issues from the lower part of the parotid gland and runs anteroinferiorly under platysma to the front of the neck, to supply platysma and communicate with the transverse cutaneous cervical nerve. In 20% of cases, there are two branches. The peripheral branches of the facial nerve described above are joined by anastomotic arcades between adjacent branches to form the parotid plexus of nerves which shows considerable variation. In surgical terms these anastomoses are important, and presumably explain why accidental or essential division of a small branch often fails to result in the expected facial nerve weakness. Six distinctive anastomotic patterns were originally classified by Davis et al (1956) and these are illustrated in Fig. 29.10. These observations have been confirmed by others, although some variation in the frequency has been reported. Surgery of the facial nerve
When operating on the face, a detailed understanding of the anatomy of the facial nerve is essential if iatrogenic trauma is to be avoided. Three surgical manoeuvres are used to identify the facial nerve trunk as it exits the stylomastoid foramen. The blood-free plane immediately in front of the cartilaginous external acoustic meatus can be opened up by blunt dissection, and this leads the surgeon to the skull bases just superficial to the styloid process and the stylomastoid foramen. This plane can then be gently opened up in an inferior direction by further blunt dissection until the trunk of the facial nerve is encountered. Second, the trunk of the facial nerve can be identified by exposing the anterior border of sternocleidomastoid just below its insertion into the mastoid process, and retracting the muscle posteriorly to expose the posterior belly of digastric, which is then traced upwards and backwards to the mastoid process. This point lies immediately below the stylomastoid foramen and the facial nerve trunk. A third option is to identify a terminal branch of the facial nerve peripherally - commonly the marginal mandibular branch - and to trace it back centripetally until the facial nerve trunk is identified.
Complications of facial nerve dissection
UPDATE Date Added: 01 December 2004 Update: Preoperative determination of the location of parotid gland tumors by analysis of the position of the facial nerve Knowledge of the course of the facial nerve, which separates the deep and superficial lobes of the parotid gland, is helpful in determining the location of parotid neoplasms and the resultant need for either superficial lobectomy or total parotidectomy, respectively. To date, there is no radiologic technique to visualize the facial nerve,1 but there are several predictive tools for determination of the course of the facial nerve.2-5 In this study, 2 new hypothetical lines for the determination of the facial nerve location were compared with the hypothetical facial nerve line (FN-line; connecting the lateral surface of the posterior belly of the digastric muscle to the lateral surface of the cortical bone of the ascending ramus), which has been shown to successfully diagnose the location of parotid gland tumors. 5 The two new lines, connecting the most dorsal (Line 1) and the most lateral (Line 2) points visible on the ipsilateral half of the vertebra to the dorsal side of the retromandibular vein, and the FN-line, were drawn on transverse sections of 5 cadaver heads. The shortest and longest distances to the facial nerve from these lines were determined, scored, and compared (chi-squared test for significance). Line 1 had the shortest distance to the facial nerve (average shortest distance=0.8 mm), and was closest to the facial nerve in 44 cases compared with 16 cases for the FN-line (p60 years), a flocculated appearance of the muscle mass and "verticalization" of the aponeurotic layer were consistent features. When group 4 was compared with group 3, it was apparent that more of the muscle volume had been replaced by fat and the contractile portion was shorter in older subjects. The clinical study confirmed that masseter was organized into alternating
muscular and aponeurotic layers. Muscular volume was diminished with age as fatty tracts replaced the muscle (this may have reflected the fact that many of the older subjects lacked teeth). The progressive verticalization of the aponeurotic sheets with age consistently corresponded to a diminution in muscle mass. Brunel G et al: General organization of the human intra-masseteric aponeuroses: changes with ageing. Surg Radio Anat 25:270-283, 2003.
Relations Skin, platysma, risorius, zygomaticus major, the parotid gland and duct, branches of the facial nerve and the transverse facial branches of the superficial temporal vessels are all superficial relations. Temporalis and the ramus of the mandible lie deep to masseter. The anterior margin of masseter is separated from buccinator and the buccal branch of the mandibular nerve by a buccal pad of fat and crossed by the facial vein. The posterior margin of the muscle is overlapped by the parotid gland. The masseteric nerve and artery reach the deep surface of masseter by passing over the mandibular incisure (mandibular notch). Vascular supply Masseter is supplied by the masseteric branch of the maxillary artery, the facial artery and the transverse facial branch of the superficial temporal artery. Innervation Masseter is supplied by the masseteric branch of the anterior trunk of the mandibular nerve. Actions Masseter elevates the mandible to occlude the teeth in mastication and has a small effect in side-to-side movements, protraction and retraction. Its electrical activity in the resting position of the mandible is minimal. Submasseteric space infections Sometimes infection around a mandibular third molar tooth tracks backwards, lateral to the mandibular ramus and pus localizes deep to the attachment of masseter in the submasseteric tissue space. Such an abscess, lying deep to this thick muscle produces little visible swelling, but is accompanied by profound muscle spasm and limitation of jaw opening. Temporalis
Figure 30.3 Left temporalis: the zygomatic arch and masseter have been removed. Note the changing orientations of the muscle fibres, from vertical anteriorly to horizontal posteriorly.
Temporalis (Fig. 30.3) arises from the whole of the temporal fossa up to the inferior temporal line - except the part formed by the zygomatic bone - and from the deep surface of the temporal fascia. Its fibres converge and descend into a tendon which passes through the gap between the zygomatic arch and the side of the skull. The muscle is attached to the medial surface, apex, anterior and posterior borders of the coronoid process and to the anterior border of the mandibular ramus almost up to the third molar tooth. The anterior fibres of temporalis are orientated vertically, the most posterior fibres almost horizontally,
and the intervening fibres with intermediate degrees of obliquity, in the manner of a fan. Fibres of temporalis may occasionally gain attachment to the articular disc. UPDATE Date Added: 02 November 2005 Publication Services, Inc. Abstract: The anatomical basis for surgical preservation of temporal muscle. Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=15035289&query_hl=1 The anatomical basis for surgical preservation of temporal muscle. Kadri PA, Al-Mefty O: J Neurosurg 100(3):517-522, 2004. Relations Skin, auriculares anterior and superior, temporal fascia, superficial temporal vessels, the auriculotemporal nerve, temporal branches of the facial nerve, the zygomaticotemporal nerve, the epicranial aponeurosis, the zygomatic arch and the masseter muscle are all superficial relations. Posterior relations of temporalis are the temporal fossa above and the major components of the infratemporal fossa below. Behind the tendon of the muscle, the masseteric nerve and vessels traverse the mandibular notch. The anterior border is separated from the zygomatic bone by a mass of fat. UPDATE Date Added: 25 October 2005 Publication Services, Inc. Abstract: The anatomy of temporal hollowing: the superficial temporal fat pad. Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve=pubmed=Abstract_uids=16077309_hl=8 The anatomy of temporal hollowing: the superficial temporal fat pad. Kim S, Matic DB: The anatomy of temporal hollowing: J Craniofac Surg 16(4):651-654, 2005. Vascular supply Temporalis is supplied by the deep temporal branches from the second part of the maxillary artery. The anterior deep temporal artery supplies c.20% of the muscle anteriorly, the posterior deep temporal supplies c.40% of the muscle in the posterior region and the middle temporal artery supplies c.40% of the muscle in its mid-region. Innervation Temporalis is supplied by the deep temporal branches of the anterior trunk of the mandibular nerve. Actions Temporalis elevates the mandible and so closes the mouth and approximates the teeth. This movement requires both the upward pull of the anterior fibres and the backward pull of the posterior fibres, because the head of the mandibular condyle rests on the articular eminence when the mouth is open. The muscle also contributes to side-to-side grinding movements. The posterior fibres retract the mandible after it has been protruded. UPDATE Abstract: Surgical preservation of temporal muscle
Date Added: 22 May 2006
Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=15035289&query_hl=27&itool=pubmed_ExternalLink The anatomical basis for surgical preservation of temporal muscle. Kadri PA, Al-Mefty O: J Neurosurg 15:209-214, 2004. Lateral pterygoid
Lateral pterygoid (Figs 30.2, 30.4, 30.5) is a short, thick muscle consisting of two parts. The upper head arises from the infratemporal surface and infratemporal crest of the greater wing of the sphenoid bone. The lower head arises from the lateral surface of the lateral pterygoid plate. From the two origins, the fibres converge, and pass backwards and laterally, to be inserted into a depression on the front of the neck of the mandible (the pterygoid fovea). A part of the upper head may be attached to the capsule of the temporomandibular joint and to the anterior and medial borders of its articular disc. Unlike the other muscles of mastication, lateral pterygoid is not pennate, nor does it have a significant number of Golgi tendon organs associated with its attachments. Relations The mandibular ramus and masseter, the maxillary artery - which crosses either deep or superficial to the muscle - and the superficial head of medial pterygoid and the tendon of temporalis, are all superficial relations. Deep to the muscle are
the deep head of medial pterygoid, the sphenomandibular ligament, the middle meningeal artery, and the mandibular nerve. The upper border is related to the temporal and masseteric branches of the mandibular nerve and the lower border is related to the lingual and inferior alveolar nerves. The buccal nerve and the maxillary artery pass between the two heads of the muscles (Fig. 30.4). Vascular supply Lateral pterygoid is supplied by pterygoid branches from the maxillary artery which are given off as the artery crosses the muscle and from the ascending palatine branch of the facial artery. Innervation The nerves to lateral pterygoid (one for each head) arise from the anterior trunk of the mandibular nerve, deep to the muscle. The upper head and the lateral part of the lower head receive their innervation from a branch given off from the buccal nerve. However, the medial part of the lower head has a branch arising directly from the anterior trunk of the mandibular nerve. page 520 page 521
Figure 30.4 A dissection of the left pterygoid region, showing some of the branches of the mandibular nerve and maxillary artery. Temporalis and the coronoid process of the mandible have been reflected upwards. Masseter has been removed (with the exception of a small inferior portion). The zygomatic arch has been removed.
Figure 30.5 A transverse section through the anterior part of the head at a level just inferior to the apex of the odontoid process: inferior aspect.
Actions When left and right muscles contract together the condyle is pulled forward and slightly downward. This protrusive movement alone has little or no function except to assist opening the jaw. Digastric and geniohyoid are the main jaw opening muscles: unlike lateral pterygoid, when acting alone they rotate the jaw open, provided other muscles attached to the hyoid prevent if from being pulled forward. If only one lateral pterygoid contracts, the jaw rotates about a vertical axis passing roughly through the opposite condyle and is pulled medially toward the opposite side. This contraction together with that of the adjacent medial pterygoid (both attached to the lateral pterygoid plate) provides most of the strong medially directed component of the force used when grinding food between teeth of the same side. It is arguably the most important function of the inferior head of lateral pterygoid. It is often stated that the upper head is used to pull the articular disc forward when the jaw is opened. But electromyography studies have proved that the upper head is inactive during jaw opening and most active when the jaws are clenched. An explanation for this surprising activity is as follows (Osborn 1995a). Most of the power of a clenching force is due to contractions of masseter and temporalis. The associated backward pull of temporalis is greater than the associated forward pull of (superficial) masseter, and so their combined jaw closing action potentially pulls the condyle backward. This is prevented by the simultaneous contraction of the upper head of lateral pterygoid. Medial pterygoid
Medial pterygoid (Figs 30.2, 30.4, 30.5) is a thick, quadrilateral muscle with two heads of origin. The major component is the deep head which arises from the medial surface of the lateral pterygoid plate of the sphenoid bone and is therefore deep to the lower head of lateral pterygoid. The small, superficial head arises from the maxillary tuberosity and the pyramidal process of the palatine bone, and therefore lies on the lower head of lateral pterygoid. The fibres of medial pterygoid descend posterolaterally and are attached by a strong tendinous lamina to the posteroinferior part of the medial surface of the ramus and angle of the mandible, as high as the mandibular foramen and almost as far forwards as the mylohyoid groove. This area of attachment is often ridged. Relations The lateral surface of medial pterygoid is related to the mandibular ramus, from which it is separated above its insertion by lateral pterygoid, the sphenomandibular ligament, the maxillary artery, the inferior alveolar vessels and nerve, the lingual nerve and a process of the parotid gland. The medial surface is related to tensor veli palatini and is separated from the superior pharyngeal constrictor by styloglossus and stylopharyngeus and by some areolar tissue. Vascular supply Medial pterygoid derives its main arterial supply from the pterygoid branches of the maxillary artery.
Innervation Medial pterygoid is innervated by the medial pterygoid branch of the mandibular nerve. Actions The medial pterygoid muscles assist in elevating the mandible. Acting with the lateral pterygoids they protrude it. When the medial and lateral pterygoids of one side act together, the corresponding side of the mandible is rotated forwards and to the opposite side, with the opposite mandibular head as a vertical axis. Alternating activity in the left and right sets of muscles produces side-to-side movements, which are used to triturate food. Pterygospinous ligament The pterygospinous ligament, which is occasionally replaced by muscle fibres, stretches between the spine of the sphenoid bone and the posterior border of the lateral pterygoid plate near its upper end. It is sometimes ossified, and then completes a foramen which transmits the branches of the mandibular nerve to temporalis, masseter and lateral pterygoid.
VASCULAR SUPPLY AND LYMPHATIC DRAINAGE Maxillary artery (Fig. 30.6) page 521 page 522
Figure 30.6 The left ophthalmic, maxillary and mandibular nerves and the submandibular and pterygopalatine ganglia (semi-diagrammatic).
The maxillary artery, the larger terminal branch of the external carotid artery, arises behind the neck of the mandible, and is at first embedded in the parotid gland. It then crosses the infratemporal fossa to enter the pterygopalatine fossa through the pterygomaxillary fissure. The artery is widely distributed to the mandible, maxilla, teeth, muscles of mastication, palate, nose and cranial dura mater. It will be described in three parts, mandibular, pterygoid and pterygopalatine. The mandibular part runs horizontally by the medial surface of the ramus. It passes between the neck of the mandible and the sphenomandibular ligament, parallel with and slightly below the auriculotemporal nerve. It next crosses the inferior alveolar nerve and skirts the lower border of lateral pterygoid. The pterygoid part ascends obliquely forwards medial to temporalis and in c.60% cases is superficial to the lower head of lateral pterygoid. When it runs deep to lateral pterygoid it lies between the muscle and branches of the mandibular nerve, and may project as a lateral loop between the two parts of lateral pterygoid. Asymmetry in this pattern of distribution may occur between the right and left infratemporal fossae and ethnic differences have been reported. Where the maxillary artery runs superficial to the lower head of lateral pterygoid, the commonest pattern is that the artery passes lateral to the inferior alveolar, lingual and buccal nerves. Less frequently, only the buccal nerve crosses the artery laterally, and rarely the artery passes deep to all the branches of the mandibular nerve. The pterygopalatine part passes between the two heads of lateral pterygoid to reach the pterygomaxillary fissure before it passes into the pterygopalatine fossa. page 522 page 523
Branches
The mandibular part of the maxillary artery has five branches, namely, deep auricular, anterior tympanic, middle meningeal, accessory meningeal and inferior alveolar: they all enter bone. The pterygoid part of the maxillary artery also has five branches. They do not enter bone, but supply muscle, and include deep temporal, pterygoid, masseteric and buccal arteries. The distribution of the branches from the pterygopalatine segment is described on page 579. Deep auricular artery The deep auricular artery pierces the osseous or cartilaginous wall of the external acoustic meatus and supplies the skin of the external acoustic meatus and part of the tympanic membrane. A small branch contributes to the arterial supply of the temporomandibular joint. Anterior tympanic artery The anterior tympanic artery passes through the petrotympanic fissure to supply part of the lining of the middle ear and accompanies the chorda tympani nerve. Middle meningeal artery The middle meningeal artery is the main source of blood to the bones of the vault of the skull. It may arise either directly from the first part of the maxillary artery or from a common trunk with the inferior alveolar artery. When the maxillary artery lies superficial to lateral pterygoid, the middle meningeal artery is usually the first branch of the maxillary artery. However, when the maxillary artery takes a deep
course in relation to the muscle this is not usually the case. The middle meningeal artery ascends between the sphenomandibular ligament and lateral pterygoid, passes between the two roots of the auriculotemporal nerve and leaves the infratemporal fossa through the foramen spinosum to enter the cranial cavity medial to the midpoint of the zygomatic bone. Its further course is described on page 281. Accessory meningeal artery The accessory meningeal artery runs through the foramen ovale into the middle cranial fossa and may arise directly from the maxillary artery or as a branch of the middle meningeal artery itself. In its course in the infratemporal fossa, the accessory meningeal artery is closely related to tensor and levator veli palatini and usually runs deep to the mandibular nerve. Although it runs intracranially, its main distribution is extracranial, principally to medial pterygoid, lateral pterygoid (upper head), tensor veli palatini, the greater wing and pterygoid processes of the sphenoid, branches of the mandibular nerve and the otic ganglion. The accessory meningeal artery is sometimes replaced by separate small arteries. Deep temporal arteries The anterior, middle and posterior branches of the deep temporal arteries pass between temporalis and the pericranium, producing shallow grooves in the bone. They anastomose with the middle temporal branch of the superficial temporal artery. The anterior deep temporal artery connects with the lacrimal artery by small branches which perforate the zygomatic bone and greater wing of the sphenoid. Masseteric artery The masseteric artery, which is small, accompanies the masseteric nerve as it passes behind the tendon of temporalis through the mandibular incisure (notch) to enter the deep surface of masseter. Its branches can also supply the temporomandibular joint. The masseteric artery anastomoses with the masseteric branches of the facial artery and with the transverse facial branch of the superficial temporal artery. Pterygoid arteries The pterygoid arteries are irregular in number and origin, and are distributed to lateral and medial pterygoid. Buccal artery The buccal artery runs obliquely forwards between medial pterygoid and the attachment of temporalis and supplies the skin and mucosa over buccinator, accompanying the lower part of the buccal branch of the mandibular nerve. It anastomoses with branches of the facial and infraorbital arteries. A small lingual branch may be given off to accompany the lingual nerve and supply structures in the floor of the mouth. Maxillary veins and the pterygoid venous plexus Maxillary vein
The maxillary vein is a short trunk which accompanies the first part of the maxillary artery. It is formed from the confluence of veins from the pterygoid plexus and passes back between the sphenomandibular ligament and the neck of the mandible, to enter the parotid gland. It unites within the substance of the gland with the superficial temporal vein to form the retromandibular vein (Fig. 29.12). Pterygoid venous plexus
The pterygoid plexus of veins is found partly between temporalis and lateral pterygoid and partly between the two pterygoid muscles. Sphenopalatine, deep temporal, pterygoid, masseteric, buccal, alveolar (dental), greater palatine and middle meningeal veins and a branch or branches from the inferior ophthalmic vein are all tributaries. The plexus connects with the facial vein via the deep facial vein and with the cavernous sinus through veins that pass through the sphenoidal emissary foramen, foramen ovale and foramen lacerum. Its deep temporal tributaries often connect with tributaries of the anterior diploic veins and thus with the middle meningeal veins.
INNERVATION The infratemporal fossa contains the major subdivisions of the mandibular branch of the trigeminal nerve, together with the chorda tympani, which enters the fossa and joins the lingual nerve, and the otic ganglion, which is functionally related to the parotid gland. The main sensory branches of the mandibular nerve extend beyond the infratemporal fossa and their distribution to the face is described on page 512.
Mandibular nerve (Figs 29.9, 30.6, 30.7)
The mandibular nerve is the largest trigeminal division and is a mixed nerve. Its sensory branches supply the teeth and gums of the mandible, the skin in the temporal region, part of the auricle - including the external meatus and tympanic membrane - and the lower lip, the lower part of the face (Fig. 29.9), and the mucosa of the anterior two-thirds (presulcal part) of the tongue and the floor of the oral cavity. The motor branches innervate the muscles of mastication. The large sensory root emerges from the lateral part of the trigeminal ganglion and exits the cranial cavity through the foramen ovale. The small motor root passes under the ganglion and through the foramen ovale to unite with the sensory root just outside the skull. As it descends from the foramen ovale, the nerve is c.4 cm from the surface and a little anterior to the neck of the mandible. The mandibular nerve immediately passes between tensor veli palatini, which is medial, and lateral pterygoid, which is lateral, and gives off a meningeal branch and the nerve to medial pterygoid from its medial side. The nerve then divides into a small anterior and large posterior trunk. The anterior division gives off branches to the four main muscles of mastication and a buccal branch which is sensory to the cheek. The posterior division gives off three main sensory branches, the auriculotemporal, lingual and inferior alveolar nerves, and motor fibres to supply mylohyoid and the anterior belly of digastric (Figs 30.6, 30.7). Meningeal branch (nervus spinosus) The meningeal branch re-enters the cranium through the foramen spinosum with the middle meningeal artery. It divides into anterior and posterior branches which accompany the main divisions of the middle meningeal artery and supply the dura mater in the middle cranial fossa and, to a lesser extent, in the anterior fossa and calvarium. Nerve to medial pterygoid The nerve to medial pterygoid is a slender ramus which enters the deep aspect of the muscle. It supplies one or two filaments that pass through the otic ganglion without interruption to supply tensor tympani and tensor veli palatini (see Fig. 30.7). Anterior trunk of mandibular nerve
The anterior trunk of the mandibular nerve gives rise to the buccal nerve, which is sensory, and the masseteric, deep temporal and lateral pterygoid nerves, which are all motor. Buccal nerve
page 523 page 524
Figure 30.7 The right otic and pterygopalatine ganglia and their branches displayed from the medial side (semi-diagrammatic).
The buccal nerve (Fig. 30.4) passes between the two heads of lateral pterygoid. It descends deep to the tendon of temporalis, passes laterally in front of masseter, and anastomoses with the buccal branches of the facial nerve. It carries the motor fibres to lateral pterygoid, and these are given off as the buccal nerve passes through the muscle. It may also give off the anterior deep temporal nerve. The buccal nerve supplies sensation to the skin over the anterior part of buccinator and the buccal mucous membrane, together with the posterior part of the buccal gingivae adjacent to the second and third molar teeth. Nerve to masseter The nerve to masseter (Fig. 30.4) passes laterally above lateral pterygoid, on to the skull base, anterior to the temporomandibular joint and posterior to the tendon of temporalis. It crosses the posterior part of the mandibular notch with the masseteric artery and ramifies on and enters the deep surface of masseter. It also provides articular branches which supply the temporomandibular joint. Deep temporal nerves The deep temporal nerves usually consist of two branches, anterior and posterior, although there may be a middle branch. They pass above lateral pterygoid to enter the deep surface of temporalis. The anterior nerve frequently arises as a branch of the buccal nerve. The small posterior nerve sometimes arises in common with the nerve to masseter. Nerve to lateral pterygoid The nerve to lateral pterygoid enters the deep surface of the muscle. It may arise separately from the anterior division of the mandibular nerve or from the buccal nerve. Posterior trunk of mandibular nerve
The posterior trunk of the mandibular nerve is larger than the anterior and is mainly sensory, although it receives fibres from the motor root for the nerve to mylohyoid. It divides into auriculotemporal, lingual and inferior alveolar (dental) nerves. Auriculotemporal nerve The auriculotemporal nerve usually has two roots which encircle the middle meningeal artery (Figs 30.6, 30.7). It runs back under lateral pterygoid on the
surface of tensor veli palatini, passes between the sphenomandibular ligament and the neck of the mandible, and then runs laterally behind the temporomandibular joint related to the upper part of the parotid gland. Emerging from behind the joint, it ascends over the posterior root of the zygoma, posterior to the superficial temporal vessels, and divides into superficial temporal branches. It communicates with the facial nerve and otic ganglion. The rami to the facial nerve, usually two, pass anterolaterally behind the neck of the mandible to join the facial nerve at the posterior border of masseter. Filaments from the otic ganglion join the roots of the auriculotemporal nerve close to their origin (Figs 30.7, 30.8). The sensory distribution of the auriculotemporal nerve on the face is described on page 513. Lingual nerve
page 524 page 525
Figure 30.8 The parasympathetic connections of the pterygopalatine, otic and submandibular ganglia. The parasympathetic fibres, both pre-and postganglionic, are shown as blue lines. The parasympathetic fibres in the palatine nerves are secretomotor to the nasal, palatine and pharyngeal glands.
The lingual nerve (Figs 30.4, 30.6, 30.7) is sensory to the mucosa of the anterior two-thirds of the tongue, the floor of the mouth and the mandibular lingual gingivae. It arises from the posterior trunk of the mandibular nerve and at first runs beneath lateral pterygoid and superficial to tensor veli palatini, where it is joined by the chorda tympani branch of the facial nerve, and often by a branch of the inferior alveolar nerve. Emerging from under cover of lateral pterygoid, the lingual nerve then runs downwards and forwards on the surface of medial pterygoid, and is thus carried progressively closer to the medial surface of the mandibular ramus. It becomes intimately related to the bone a few millimetres below and behind the junction of the vertical ramus and horizontal body of the mandible. Here it lies anterior to, and slightly deeper than, the inferior alveolar (dental) nerve. It next passes below the mandibular attachment of the superior pharyngeal constrictor and pterygomandibular raphe, closely applied to the periosteum of the medial surface of the mandible, until it lies opposite the posterior root of the third molar tooth, where it is covered only by the gingival mucoperiosteum. At this point it usually lies 2-3 mm below the alveolar crest and 0.6 mm from the bone, however in c.5% of cases it lies above the alveolar crest. It next passes medial to the mandibular origin of mylohyoid, and this carries it progressively away from the mandible, and separates it from the alveolar bone covering the mesial root of the third molar tooth. The rest of the nerve is described with the mouth and oral cavity (p. 588). Inferior alveolar (dental) nerve
The inferior alveolar nerve descends behind lateral pterygoid. At the lower border of the muscle the nerve passes between the sphenomandibular ligament and the mandibular ramus and enters the mandibular canal via the mandibular foramen. Below lateral pterygoid it is accompanied by the inferior alveolar artery, a branch of the first part of the maxillary artery, which also enters the canal with associated veins. The subsequent course of the inferior alveolar nerve is described on page 601. Otic ganglion
This is a small, oval, flat reddish-grey ganglion (see Figs 30.7, 30.8) situated just below the foramen ovale. It is a peripheral parasympathetic ganglion related topographically to the mandibular nerve, but connected functionally with the glossopharyngeal nerve. Near its junction with the trigeminal motor root, the mandibular nerve lies lateral to the ganglion; tensor veli palatini lies medially, separating the ganglion from the cartilaginous part of the pharyngotympanic tube, and the middle meningeal artery is posterior to the ganglion. The otic ganglion usually surrounds the origin of the nerve to medial pterygoid. Like all parasympathetic ganglia, there are three roots, motor, sympathetic and sensory. Only the parasympathetic fibres relay in the ganglion. The motor, parasympathetic, root of the otic ganglion is the lesser petrosal nerve, conveying preganglionic fibres from the glossopharyngeal nerve which originate from neurones in the inferior salivatory nucleus. The lesser petrosal nerve runs intracranially in the middle cranial fossa on the anterior surface of the petrous bone before passing through the foramen ovale to join the otic ganglion. The nerve synapses in the otic ganglion, and postganglionic fibres pass by a communicating branch to the auriculotemporal nerve and so to the parotid gland. The sympathetic root is from a plexus on the middle meningeal artery. It contains postganglionic fibres from the superior cervical sympathetic ganglion which traverse the otic ganglion without relay and emerge with parasympathetic fibres in the connection with the auriculotemporal nerve to supply blood vessels in the parotid gland. The sensory fibres from the gland are derived from the auriculotemporal nerve. Clinical observations suggest that in humans the gland also receives secretomotor fibres through the chorda tympani. Branches A branch connects the otic ganglion to the chorda tympani nerve, while another ramus ascends to join the nerve of the pterygoid canal. These branches may form an additional pathway by which gustatory fibres from the anterior two-thirds of the tongue may reach the facial ganglion without traversing the middle ear, and they do not synapse in the otic ganglion. Motor branches to tensor veli palatini and tensor tympani, derived from the nerve to medial pterygoid, also pass through the ganglion without synapsing. Chorda tympani nerve
The chorda tympani nerve enters the infratemporal fossa region by passing through the medial end of the petrotympanic fissure behind the capsule of the temporomandibular joint. The nerve descends medial to the spine of the sphenoid bone - which it sometimes grooves - lying posterolateral to tensor veli palatini. It is crossed medially by the middle meningeal artery, the roots of the auriculotemporal nerve and by the inferior alveolar nerve (Fig. 30.6). The chorda tympani joins the posterior aspect of the lingual nerve at an acute angle. It carries taste fibres for the anterior two-thirds of the tongue and efferent preganglionic parasympathetic (secretomotor) fibres destined for the submandibular ganglion in the floor of the mouth.
LE FORT AND ZYGOMATIC FRACTURES (p. 489) Le Fort I, II or III fractures inevitably involve the infratemporal fossa. The bones of the midface transmit the forces of impact directly to the cranium. The most important strut related to the infratemporal and pterygopalatine fossae is the pterygomaxillary strut. Fractures involving this strut may extend elsewhere to involve the cranial base and orbit. The associated soft tissue damage which accompanies these fractures may damage nerves, blood vessels and muscles. Injuries to the second or third divisions of the trigeminal nerve or the chorda tympani nerve result in altered sensation to the oral cavity, face and jaws, including impaired taste. Fractures extending into the orbit may result in decreased visual acuity and ophthalmoplegia. Neural damage to motor nerves or direct damage to muscles may result in problems with chewing, swallowing, speech, middle ear function and eye movements. Injuries that involve the pterygopalatine or otic ganglia interfere with lacrimation, nasal secretions and salivation. Damage to adjacent blood vessels may result in haemorrhage, thrombosis emboli and the formation of false aneurisms or arteriovenous fistulae.
Classic zygomatic complex fractures involve the lateral wall of the orbit and cross laterally into the infratemporal fossa at the frontozygomatic suture. From this point the fracture line extends inferiorly to join the most lateral aspect of the inferior orbital fissure, continues inferiorly along the posterior surface of the zygomatic buttress - where it communicates with the lateral bulge of the maxillary antrum and runs around the zygomatic buttress, high in the buccal sulcus in the upper molar region, and then extends upwards towards the infraorbital foramen. It finally runs laterally along the floor of the orbit to reach the lateral extension of the inferior orbital fissure. These fractures involve the maxillary sinus and the infratemporal fossa and orbit, which means that any blood which collects in the antrum, if it becomes infected, will allow infection to spread into the infratemporal fossa. Infection in this area can have grave consequences and can rapidly spread through the foramina in the skull base into the middle cranial fossa. For this reason patients presenting with zygomatic complex fractures are placed on prophylactic antibiotic therapy to prevent infection.
SPREAD OF INFECTION FROM THE INFRATEMPORAL FOSSA (p. 607) page 525 page 526
The majority of infected teeth in the upper jaw and those in the front part of the lower jaw will generally drain harmlessly into the oral cavity - either via the vestibule buccally, or via the palate or mouth lingually - and they are of little clinical significance. In contrast, a pericoronitis affecting a partially impacted mandibular third molar tooth, or less commonly either a dental abscess of this tooth, or an infection following tooth extraction, spreads into the infratemporal fossa. Infection may also result from an infected needle used during an inferior alveolar nerve block, or as a result of spread from an adjacent infected tissue space. The main symptom caused by infection of the pterygomandibular space is trismus - painful reflex muscle spasm - which usually affects medial pterygoid. Infection may potentially spread some distance from the infratemporal fossa because the latter lies between the tissue spaces of the face above and the tissue spaces of the neck below. Thus infection may spread to involve the buccal tissue space, or directly around the back of the maxillary tuberosity and into the orbit via the inferior orbital fissure, which may result in a cavernous sinus thrombosis. Once in the orbit, infection may spread directly through the superior orbital fissure into the cranial cavity. Infection may also spread from the infratemporal fossa via the pterygomaxillary fissure to involve the pterygopalatine fossa and its contents, and may spread further via a number of small canals which lead from the fossa into the nose, pharynx and palate.
© 2008 Elsevier
INFRATEMPORAL FOSSA The infratemporal fossa is the space located deep to the ramus of the mandible. It communicates with the temporal fossa deep to the zygomatic arch and the pterygopalatine fossa through the pterygomaxillary fissure (p. 578). The major structures which occupy the infratemporal fossa are the lateral and medial pterygoid muscles, the mandibular division of the trigeminal nerve, the chorda tympani branch of the facial nerve, the otic parasympathetic ganglion, the maxillary artery and the pterygoid venous plexus. UPDATE Date Added: 16 August 2005 Shanida Helena Nataraja, PhD (Dianthus Medical Limited) Update: Surgical anatomy of infratemporal fossa: transmaxillary approach. The transmaxillary approach to the infratemporal fossa (ITF) has been evaluated in an anatomic study of five adult cadaver heads (10 specimens). In three of these specimens, the vascular structures were injected with colored latex to improve their visibility. All specimens were dissected using both macro- and micro-surgical procedures and instrumentation. Two compartments were identified whilst exposing the ITF using the transmaxillary approach: the retroantral space and the lateropterygoid space. The retroantral space was 5 to 10 mm in anteroposterior depth and was filled with fat and a loose venous network surrounding the maxillary artery (MA). In all specimens, the anterior loop of the MA and the sphenopalatine artery were located in the proximal retroantral space and could, therefore, be ligated without optic magnification. The transmaxillary approach allowed for a mean angle of vision of the foramen rotundum of 31°. The lateropterygoid space was located posterior to the retroantral space and was filled with the lateral pterygoid (LP). Within the lateropterygoid space, the MA was observed to be intimately related with the pterygoid head of the LP. Access to the foramen ovale was deep (mean depth of 56mm) and narrow (20°) and required extensive desinsertion and partial removal of the superior bundle of the LP. The results of this study indicate that the transmaxillary approach is a convenient, minimally invasive route to expose the anterior portion of the ITF. Not only does this procedure expose the entire extracranial portion of the maxillary nerve, but it also allows for the complete dissection of any lesions of the orbital roof and the inferior orbital fissure whilst preserving the maxillary nerve. Roche PH, Fournier HD, Laccourreye L et al: Surgical anatomy of the infratemporal fossa using the transmaxillary approach. Surg Radiol Anat 23(4):209-213, 2001. Medline Similar articles
The temporomandibular joint is a synovial joint between the condyle of the mandible below and the mandibular fossa of the temporal bone above. The joint is unusual in that its articular surfaces are lined by fibrous tissue (rather than hyaline cartilage) and its joint cavity is divided into two by an articular disc. The roof of the infratemporal fossa is the infratemporal surface of the greater wing of the sphenoid, its medial wall is formed by the lateral pterygoid plate of the sphenoid bone, separated from the maxilla by the pterygomaxillary fissure (Fig. 30.1) and the lateral wall is formed by the ramus of the mandible. The posterior wall shows the presence of the styloid process of the temporal bone. The infratemporal fossa has no anatomical floor.
Figure 30.1 The left infratemporal fossa: seen from the side after detachment of the mandible and removal of the zygomatic arch. Blue: frontal bone; yellow: sphenoid and lacrimal bones; brown: temporal and nasal bones; green: maxilla. The parts shown of the parietal, zygomatic, ethmoid and palatine bones are uncoloured.
Lateral pterygoid provides a key to understanding the relationships of structures within the infratemporal fossa. This muscle lies in the roof of the fossa and runs anteroposteriorly in a more or less horizontal plane from the region of the pterygoid plates to the mandibular condyle (Figs 30.1, 30.2). Branches of the mandibular nerve and the main origin of medial pterygoid are deep relations and the maxillary artery is superficial. The buccal branch of the mandibular nerve passes between the two heads of lateral pterygoid. Medial pterygoid and the lingual and inferior alveolar nerves emerge below its inferior border and the deep temporal nerves and vessels emerge from its upper border. A venous network, the pterygoid venous plexus, lies around and within lateral pterygoid.
MUSCLES The infratemporal fossa contains two of the four principal muscles of mastication, medial and lateral pterygoid. The tendon of a third muscle, temporalis, is also found in this region. The remaining muscle, masseter, lies on the face on the lateral surface of the ramus but is considered here with the muscles of mastication. The masticatory muscles are most immediately concerned with movements of the mandible at the temporomandibular joints. Masseter
page 519 page 520
Figure 30.2 Left pterygoid muscles: the zygomatic arch and part of the ramus of the mandible have been removed, and temporalis has been reflected back.
Masseter (Fig. 29.12) consists of three layers which blend anteriorly. The superficial layer is the largest. It arises by a thick aponeurosis from the maxillary process of the zygomatic bone and from the anterior two-thirds of the inferior border of the zygomatic arch. Its fibres pass downwards and backwards, to insert into the angle and lower posterior half of the lateral surface of the mandibular ramus. Intramuscular tendinous septa in this layer are responsible for the ridges on the surface of the ramus. The middle layer of masseter arises from the medial aspect of the anterior two-thirds of the zygomatic arch and from the lower border of the posterior third of this arch. It inserts into the central part of the ramus of the mandible. The deep layer arises from the deep surface of the zygomatic arch and inserts into the upper part of the mandibular ramus and into its coronoid process. There is still debate as to whether fibres of masseter are attached to the anterolateral part of the articular disc of the temporomandibular joint. UPDATE Date Added: 07 December 2005 Publication Services, Inc. Update: General organization of the human intra-masseteric aponeuroses: changes with ageing. Brunel et al. compared the appearance of the living masseter muscle using magnetic resonance imaging (18 patients, age range 6 to 79 years) with its cadaveric appearance (n = 169). The dissected masseter consisted of 6 alternating musculo-aponeurotic layers, which could be divided into 3 planes (superficial, intermediate, and deep masseter). The superficial masseter consisted of 2 alternating musculoaponeurotic layers that were defined as the superficial or prior lamina and the deep or altera lamina orientated at 60° with respect to the Frankfurt plane. Intermediate masseter was generally oriented at 90° to the reference plane and demonstrated structural changes in edentulous subjects. Deep masseter consisted of a muscular tendinous fan separated into 2 portions by the masseteric pedicle. The anterior fan was almost vertical with respect to the reference plane and consisted of a single layer, while the posterior fan exhibited 3 alternating musculo-aponeurotic layers. In the clinical study, subjects were divided into 4 groups. In group 1 (subjects aged 6 to 16 years), only 2 aponeurotic structures were consistently visible. In group 2 (17 to 40 years), muscular volume was increased compared with that in groups 1 and 4 and aponeurotic structures were consistently visible. Compared with group 2, group 3 (41-60 years) demonstrated less muscular homogeneity; fatty infiltration of the muscular zones was found in some patients. In group 4 (>60 years), a flocculated appearance of the muscle mass and "verticalization" of the aponeurotic layer were consistent features. When group 4 was compared with group 3, it was apparent that more of the muscle volume had been replaced by fat and the contractile portion was shorter in older subjects. The clinical study confirmed that masseter was organized into alternating
muscular and aponeurotic layers. Muscular volume was diminished with age as fatty tracts replaced the muscle (this may have reflected the fact that many of the older subjects lacked teeth). The progressive verticalization of the aponeurotic sheets with age consistently corresponded to a diminution in muscle mass. Brunel G et al: General organization of the human intra-masseteric aponeuroses: changes with ageing. Surg Radio Anat 25:270-283, 2003.
Relations Skin, platysma, risorius, zygomaticus major, the parotid gland and duct, branches of the facial nerve and the transverse facial branches of the superficial temporal vessels are all superficial relations. Temporalis and the ramus of the mandible lie deep to masseter. The anterior margin of masseter is separated from buccinator and the buccal branch of the mandibular nerve by a buccal pad of fat and crossed by the facial vein. The posterior margin of the muscle is overlapped by the parotid gland. The masseteric nerve and artery reach the deep surface of masseter by passing over the mandibular incisure (mandibular notch). Vascular supply Masseter is supplied by the masseteric branch of the maxillary artery, the facial artery and the transverse facial branch of the superficial temporal artery. Innervation Masseter is supplied by the masseteric branch of the anterior trunk of the mandibular nerve. Actions Masseter elevates the mandible to occlude the teeth in mastication and has a small effect in side-to-side movements, protraction and retraction. Its electrical activity in the resting position of the mandible is minimal. Submasseteric space infections Sometimes infection around a mandibular third molar tooth tracks backwards, lateral to the mandibular ramus and pus localizes deep to the attachment of masseter in the submasseteric tissue space. Such an abscess, lying deep to this thick muscle produces little visible swelling, but is accompanied by profound muscle spasm and limitation of jaw opening. Temporalis
Figure 30.3 Left temporalis: the zygomatic arch and masseter have been removed. Note the changing orientations of the muscle fibres, from vertical anteriorly to horizontal posteriorly.
Temporalis (Fig. 30.3) arises from the whole of the temporal fossa up to the inferior temporal line - except the part formed by the zygomatic bone - and from the deep surface of the temporal fascia. Its fibres converge and descend into a tendon which passes through the gap between the zygomatic arch and the side of the skull. The muscle is attached to the medial surface, apex, anterior and posterior borders of the coronoid process and to the anterior border of the mandibular ramus almost up to the third molar tooth. The anterior fibres of temporalis are orientated vertically, the most posterior fibres almost horizontally,
and the intervening fibres with intermediate degrees of obliquity, in the manner of a fan. Fibres of temporalis may occasionally gain attachment to the articular disc. UPDATE Date Added: 02 November 2005 Publication Services, Inc. Abstract: The anatomical basis for surgical preservation of temporal muscle. Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=15035289&query_hl=1 The anatomical basis for surgical preservation of temporal muscle. Kadri PA, Al-Mefty O: J Neurosurg 100(3):517-522, 2004. Relations Skin, auriculares anterior and superior, temporal fascia, superficial temporal vessels, the auriculotemporal nerve, temporal branches of the facial nerve, the zygomaticotemporal nerve, the epicranial aponeurosis, the zygomatic arch and the masseter muscle are all superficial relations. Posterior relations of temporalis are the temporal fossa above and the major components of the infratemporal fossa below. Behind the tendon of the muscle, the masseteric nerve and vessels traverse the mandibular notch. The anterior border is separated from the zygomatic bone by a mass of fat. UPDATE Date Added: 25 October 2005 Publication Services, Inc. Abstract: The anatomy of temporal hollowing: the superficial temporal fat pad. Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve=pubmed=Abstract_uids=16077309_hl=8 The anatomy of temporal hollowing: the superficial temporal fat pad. Kim S, Matic DB: The anatomy of temporal hollowing: J Craniofac Surg 16(4):651-654, 2005. Vascular supply Temporalis is supplied by the deep temporal branches from the second part of the maxillary artery. The anterior deep temporal artery supplies c.20% of the muscle anteriorly, the posterior deep temporal supplies c.40% of the muscle in the posterior region and the middle temporal artery supplies c.40% of the muscle in its mid-region. Innervation Temporalis is supplied by the deep temporal branches of the anterior trunk of the mandibular nerve. Actions Temporalis elevates the mandible and so closes the mouth and approximates the teeth. This movement requires both the upward pull of the anterior fibres and the backward pull of the posterior fibres, because the head of the mandibular condyle rests on the articular eminence when the mouth is open. The muscle also contributes to side-to-side grinding movements. The posterior fibres retract the mandible after it has been protruded. UPDATE Abstract: Surgical preservation of temporal muscle
Date Added: 22 May 2006
Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=15035289&query_hl=27&itool=pubmed_ExternalLink The anatomical basis for surgical preservation of temporal muscle. Kadri PA, Al-Mefty O: J Neurosurg 15:209-214, 2004. Lateral pterygoid
Lateral pterygoid (Figs 30.2, 30.4, 30.5) is a short, thick muscle consisting of two parts. The upper head arises from the infratemporal surface and infratemporal crest of the greater wing of the sphenoid bone. The lower head arises from the lateral surface of the lateral pterygoid plate. From the two origins, the fibres converge, and pass backwards and laterally, to be inserted into a depression on the front of the neck of the mandible (the pterygoid fovea). A part of the upper head may be attached to the capsule of the temporomandibular joint and to the anterior and medial borders of its articular disc. Unlike the other muscles of mastication, lateral pterygoid is not pennate, nor does it have a significant number of Golgi tendon organs associated with its attachments. Relations The mandibular ramus and masseter, the maxillary artery - which crosses either deep or superficial to the muscle - and the superficial head of medial pterygoid and the tendon of temporalis, are all superficial relations. Deep to the muscle are
the deep head of medial pterygoid, the sphenomandibular ligament, the middle meningeal artery, and the mandibular nerve. The upper border is related to the temporal and masseteric branches of the mandibular nerve and the lower border is related to the lingual and inferior alveolar nerves. The buccal nerve and the maxillary artery pass between the two heads of the muscles (Fig. 30.4). Vascular supply Lateral pterygoid is supplied by pterygoid branches from the maxillary artery which are given off as the artery crosses the muscle and from the ascending palatine branch of the facial artery. Innervation The nerves to lateral pterygoid (one for each head) arise from the anterior trunk of the mandibular nerve, deep to the muscle. The upper head and the lateral part of the lower head receive their innervation from a branch given off from the buccal nerve. However, the medial part of the lower head has a branch arising directly from the anterior trunk of the mandibular nerve. page 520 page 521
Figure 30.4 A dissection of the left pterygoid region, showing some of the branches of the mandibular nerve and maxillary artery. Temporalis and the coronoid process of the mandible have been reflected upwards. Masseter has been removed (with the exception of a small inferior portion). The zygomatic arch has been removed.
Figure 30.5 A transverse section through the anterior part of the head at a level just inferior to the apex of the odontoid process: inferior aspect.
Actions When left and right muscles contract together the condyle is pulled forward and slightly downward. This protrusive movement alone has little or no function except to assist opening the jaw. Digastric and geniohyoid are the main jaw opening muscles: unlike lateral pterygoid, when acting alone they rotate the jaw open, provided other muscles attached to the hyoid prevent if from being pulled forward. If only one lateral pterygoid contracts, the jaw rotates about a vertical axis passing roughly through the opposite condyle and is pulled medially toward the opposite side. This contraction together with that of the adjacent medial pterygoid (both attached to the lateral pterygoid plate) provides most of the strong medially directed component of the force used when grinding food between teeth of the same side. It is arguably the most important function of the inferior head of lateral pterygoid. It is often stated that the upper head is used to pull the articular disc forward when the jaw is opened. But electromyography studies have proved that the upper head is inactive during jaw opening and most active when the jaws are clenched. An explanation for this surprising activity is as follows (Osborn 1995a). Most of the power of a clenching force is due to contractions of masseter and temporalis. The associated backward pull of temporalis is greater than the associated forward pull of (superficial) masseter, and so their combined jaw closing action potentially pulls the condyle backward. This is prevented by the simultaneous contraction of the upper head of lateral pterygoid. Medial pterygoid
Medial pterygoid (Figs 30.2, 30.4, 30.5) is a thick, quadrilateral muscle with two heads of origin. The major component is the deep head which arises from the medial surface of the lateral pterygoid plate of the sphenoid bone and is therefore deep to the lower head of lateral pterygoid. The small, superficial head arises from the maxillary tuberosity and the pyramidal process of the palatine bone, and therefore lies on the lower head of lateral pterygoid. The fibres of medial pterygoid descend posterolaterally and are attached by a strong tendinous lamina to the posteroinferior part of the medial surface of the ramus and angle of the mandible, as high as the mandibular foramen and almost as far forwards as the mylohyoid groove. This area of attachment is often ridged. Relations The lateral surface of medial pterygoid is related to the mandibular ramus, from which it is separated above its insertion by lateral pterygoid, the sphenomandibular ligament, the maxillary artery, the inferior alveolar vessels and nerve, the lingual nerve and a process of the parotid gland. The medial surface is related to tensor veli palatini and is separated from the superior pharyngeal constrictor by styloglossus and stylopharyngeus and by some areolar tissue. Vascular supply Medial pterygoid derives its main arterial supply from the pterygoid branches of the maxillary artery.
Innervation Medial pterygoid is innervated by the medial pterygoid branch of the mandibular nerve. Actions The medial pterygoid muscles assist in elevating the mandible. Acting with the lateral pterygoids they protrude it. When the medial and lateral pterygoids of one side act together, the corresponding side of the mandible is rotated forwards and to the opposite side, with the opposite mandibular head as a vertical axis. Alternating activity in the left and right sets of muscles produces side-to-side movements, which are used to triturate food. Pterygospinous ligament The pterygospinous ligament, which is occasionally replaced by muscle fibres, stretches between the spine of the sphenoid bone and the posterior border of the lateral pterygoid plate near its upper end. It is sometimes ossified, and then completes a foramen which transmits the branches of the mandibular nerve to temporalis, masseter and lateral pterygoid.
VASCULAR SUPPLY AND LYMPHATIC DRAINAGE Maxillary artery (Fig. 30.6) page 521 page 522
Figure 30.6 The left ophthalmic, maxillary and mandibular nerves and the submandibular and pterygopalatine ganglia (semi-diagrammatic).
The maxillary artery, the larger terminal branch of the external carotid artery, arises behind the neck of the mandible, and is at first embedded in the parotid gland. It then crosses the infratemporal fossa to enter the pterygopalatine fossa through the pterygomaxillary fissure. The artery is widely distributed to the mandible, maxilla, teeth, muscles of mastication, palate, nose and cranial dura mater. It will be described in three parts, mandibular, pterygoid and pterygopalatine. The mandibular part runs horizontally by the medial surface of the ramus. It passes between the neck of the mandible and the sphenomandibular ligament, parallel with and slightly below the auriculotemporal nerve. It next crosses the inferior alveolar nerve and skirts the lower border of lateral pterygoid. The pterygoid part ascends obliquely forwards medial to temporalis and in c.60% cases is superficial to the lower head of lateral pterygoid. When it runs deep to lateral pterygoid it lies between the muscle and branches of the mandibular nerve, and may project as a lateral loop between the two parts of lateral pterygoid. Asymmetry in this pattern of distribution may occur between the right and left infratemporal fossae and ethnic differences have been reported. Where the maxillary artery runs superficial to the lower head of lateral pterygoid, the commonest pattern is that the artery passes lateral to the inferior alveolar, lingual and buccal nerves. Less frequently, only the buccal nerve crosses the artery laterally, and rarely the artery passes deep to all the branches of the mandibular nerve. The pterygopalatine part passes between the two heads of lateral pterygoid to reach the pterygomaxillary fissure before it passes into the pterygopalatine fossa. page 522 page 523
Branches
The mandibular part of the maxillary artery has five branches, namely, deep auricular, anterior tympanic, middle meningeal, accessory meningeal and inferior alveolar: they all enter bone. The pterygoid part of the maxillary artery also has five branches. They do not enter bone, but supply muscle, and include deep temporal, pterygoid, masseteric and buccal arteries. The distribution of the branches from the pterygopalatine segment is described on page 579. Deep auricular artery The deep auricular artery pierces the osseous or cartilaginous wall of the external acoustic meatus and supplies the skin of the external acoustic meatus and part of the tympanic membrane. A small branch contributes to the arterial supply of the temporomandibular joint. Anterior tympanic artery The anterior tympanic artery passes through the petrotympanic fissure to supply part of the lining of the middle ear and accompanies the chorda tympani nerve. Middle meningeal artery The middle meningeal artery is the main source of blood to the bones of the vault of the skull. It may arise either directly from the first part of the maxillary artery or from a common trunk with the inferior alveolar artery. When the maxillary artery lies superficial to lateral pterygoid, the middle meningeal artery is usually the first branch of the maxillary artery. However, when the maxillary artery takes a deep
course in relation to the muscle this is not usually the case. The middle meningeal artery ascends between the sphenomandibular ligament and lateral pterygoid, passes between the two roots of the auriculotemporal nerve and leaves the infratemporal fossa through the foramen spinosum to enter the cranial cavity medial to the midpoint of the zygomatic bone. Its further course is described on page 281. Accessory meningeal artery The accessory meningeal artery runs through the foramen ovale into the middle cranial fossa and may arise directly from the maxillary artery or as a branch of the middle meningeal artery itself. In its course in the infratemporal fossa, the accessory meningeal artery is closely related to tensor and levator veli palatini and usually runs deep to the mandibular nerve. Although it runs intracranially, its main distribution is extracranial, principally to medial pterygoid, lateral pterygoid (upper head), tensor veli palatini, the greater wing and pterygoid processes of the sphenoid, branches of the mandibular nerve and the otic ganglion. The accessory meningeal artery is sometimes replaced by separate small arteries. Deep temporal arteries The anterior, middle and posterior branches of the deep temporal arteries pass between temporalis and the pericranium, producing shallow grooves in the bone. They anastomose with the middle temporal branch of the superficial temporal artery. The anterior deep temporal artery connects with the lacrimal artery by small branches which perforate the zygomatic bone and greater wing of the sphenoid. Masseteric artery The masseteric artery, which is small, accompanies the masseteric nerve as it passes behind the tendon of temporalis through the mandibular incisure (notch) to enter the deep surface of masseter. Its branches can also supply the temporomandibular joint. The masseteric artery anastomoses with the masseteric branches of the facial artery and with the transverse facial branch of the superficial temporal artery. Pterygoid arteries The pterygoid arteries are irregular in number and origin, and are distributed to lateral and medial pterygoid. Buccal artery The buccal artery runs obliquely forwards between medial pterygoid and the attachment of temporalis and supplies the skin and mucosa over buccinator, accompanying the lower part of the buccal branch of the mandibular nerve. It anastomoses with branches of the facial and infraorbital arteries. A small lingual branch may be given off to accompany the lingual nerve and supply structures in the floor of the mouth. Maxillary veins and the pterygoid venous plexus Maxillary vein
The maxillary vein is a short trunk which accompanies the first part of the maxillary artery. It is formed from the confluence of veins from the pterygoid plexus and passes back between the sphenomandibular ligament and the neck of the mandible, to enter the parotid gland. It unites within the substance of the gland with the superficial temporal vein to form the retromandibular vein (Fig. 29.12). Pterygoid venous plexus
The pterygoid plexus of veins is found partly between temporalis and lateral pterygoid and partly between the two pterygoid muscles. Sphenopalatine, deep temporal, pterygoid, masseteric, buccal, alveolar (dental), greater palatine and middle meningeal veins and a branch or branches from the inferior ophthalmic vein are all tributaries. The plexus connects with the facial vein via the deep facial vein and with the cavernous sinus through veins that pass through the sphenoidal emissary foramen, foramen ovale and foramen lacerum. Its deep temporal tributaries often connect with tributaries of the anterior diploic veins and thus with the middle meningeal veins.
INNERVATION The infratemporal fossa contains the major subdivisions of the mandibular branch of the trigeminal nerve, together with the chorda tympani, which enters the fossa and joins the lingual nerve, and the otic ganglion, which is functionally related to the parotid gland. The main sensory branches of the mandibular nerve extend beyond the infratemporal fossa and their distribution to the face is described on page 512.
Mandibular nerve (Figs 29.9, 30.6, 30.7)
The mandibular nerve is the largest trigeminal division and is a mixed nerve. Its sensory branches supply the teeth and gums of the mandible, the skin in the temporal region, part of the auricle - including the external meatus and tympanic membrane - and the lower lip, the lower part of the face (Fig. 29.9), and the mucosa of the anterior two-thirds (presulcal part) of the tongue and the floor of the oral cavity. The motor branches innervate the muscles of mastication. The large sensory root emerges from the lateral part of the trigeminal ganglion and exits the cranial cavity through the foramen ovale. The small motor root passes under the ganglion and through the foramen ovale to unite with the sensory root just outside the skull. As it descends from the foramen ovale, the nerve is c.4 cm from the surface and a little anterior to the neck of the mandible. The mandibular nerve immediately passes between tensor veli palatini, which is medial, and lateral pterygoid, which is lateral, and gives off a meningeal branch and the nerve to medial pterygoid from its medial side. The nerve then divides into a small anterior and large posterior trunk. The anterior division gives off branches to the four main muscles of mastication and a buccal branch which is sensory to the cheek. The posterior division gives off three main sensory branches, the auriculotemporal, lingual and inferior alveolar nerves, and motor fibres to supply mylohyoid and the anterior belly of digastric (Figs 30.6, 30.7). Meningeal branch (nervus spinosus) The meningeal branch re-enters the cranium through the foramen spinosum with the middle meningeal artery. It divides into anterior and posterior branches which accompany the main divisions of the middle meningeal artery and supply the dura mater in the middle cranial fossa and, to a lesser extent, in the anterior fossa and calvarium. Nerve to medial pterygoid The nerve to medial pterygoid is a slender ramus which enters the deep aspect of the muscle. It supplies one or two filaments that pass through the otic ganglion without interruption to supply tensor tympani and tensor veli palatini (see Fig. 30.7). Anterior trunk of mandibular nerve
The anterior trunk of the mandibular nerve gives rise to the buccal nerve, which is sensory, and the masseteric, deep temporal and lateral pterygoid nerves, which are all motor. Buccal nerve
page 523 page 524
Figure 30.7 The right otic and pterygopalatine ganglia and their branches displayed from the medial side (semi-diagrammatic).
The buccal nerve (Fig. 30.4) passes between the two heads of lateral pterygoid. It descends deep to the tendon of temporalis, passes laterally in front of masseter, and anastomoses with the buccal branches of the facial nerve. It carries the motor fibres to lateral pterygoid, and these are given off as the buccal nerve passes through the muscle. It may also give off the anterior deep temporal nerve. The buccal nerve supplies sensation to the skin over the anterior part of buccinator and the buccal mucous membrane, together with the posterior part of the buccal gingivae adjacent to the second and third molar teeth. Nerve to masseter The nerve to masseter (Fig. 30.4) passes laterally above lateral pterygoid, on to the skull base, anterior to the temporomandibular joint and posterior to the tendon of temporalis. It crosses the posterior part of the mandibular notch with the masseteric artery and ramifies on and enters the deep surface of masseter. It also provides articular branches which supply the temporomandibular joint. Deep temporal nerves The deep temporal nerves usually consist of two branches, anterior and posterior, although there may be a middle branch. They pass above lateral pterygoid to enter the deep surface of temporalis. The anterior nerve frequently arises as a branch of the buccal nerve. The small posterior nerve sometimes arises in common with the nerve to masseter. Nerve to lateral pterygoid The nerve to lateral pterygoid enters the deep surface of the muscle. It may arise separately from the anterior division of the mandibular nerve or from the buccal nerve. Posterior trunk of mandibular nerve
The posterior trunk of the mandibular nerve is larger than the anterior and is mainly sensory, although it receives fibres from the motor root for the nerve to mylohyoid. It divides into auriculotemporal, lingual and inferior alveolar (dental) nerves. Auriculotemporal nerve The auriculotemporal nerve usually has two roots which encircle the middle meningeal artery (Figs 30.6, 30.7). It runs back under lateral pterygoid on the
surface of tensor veli palatini, passes between the sphenomandibular ligament and the neck of the mandible, and then runs laterally behind the temporomandibular joint related to the upper part of the parotid gland. Emerging from behind the joint, it ascends over the posterior root of the zygoma, posterior to the superficial temporal vessels, and divides into superficial temporal branches. It communicates with the facial nerve and otic ganglion. The rami to the facial nerve, usually two, pass anterolaterally behind the neck of the mandible to join the facial nerve at the posterior border of masseter. Filaments from the otic ganglion join the roots of the auriculotemporal nerve close to their origin (Figs 30.7, 30.8). The sensory distribution of the auriculotemporal nerve on the face is described on page 513. Lingual nerve
page 524 page 525
Figure 30.8 The parasympathetic connections of the pterygopalatine, otic and submandibular ganglia. The parasympathetic fibres, both pre-and postganglionic, are shown as blue lines. The parasympathetic fibres in the palatine nerves are secretomotor to the nasal, palatine and pharyngeal glands.
The lingual nerve (Figs 30.4, 30.6, 30.7) is sensory to the mucosa of the anterior two-thirds of the tongue, the floor of the mouth and the mandibular lingual gingivae. It arises from the posterior trunk of the mandibular nerve and at first runs beneath lateral pterygoid and superficial to tensor veli palatini, where it is joined by the chorda tympani branch of the facial nerve, and often by a branch of the inferior alveolar nerve. Emerging from under cover of lateral pterygoid, the lingual nerve then runs downwards and forwards on the surface of medial pterygoid, and is thus carried progressively closer to the medial surface of the mandibular ramus. It becomes intimately related to the bone a few millimetres below and behind the junction of the vertical ramus and horizontal body of the mandible. Here it lies anterior to, and slightly deeper than, the inferior alveolar (dental) nerve. It next passes below the mandibular attachment of the superior pharyngeal constrictor and pterygomandibular raphe, closely applied to the periosteum of the medial surface of the mandible, until it lies opposite the posterior root of the third molar tooth, where it is covered only by the gingival mucoperiosteum. At this point it usually lies 2-3 mm below the alveolar crest and 0.6 mm from the bone, however in c.5% of cases it lies above the alveolar crest. It next passes medial to the mandibular origin of mylohyoid, and this carries it progressively away from the mandible, and separates it from the alveolar bone covering the mesial root of the third molar tooth. The rest of the nerve is described with the mouth and oral cavity (p. 588). Inferior alveolar (dental) nerve
The inferior alveolar nerve descends behind lateral pterygoid. At the lower border of the muscle the nerve passes between the sphenomandibular ligament and the mandibular ramus and enters the mandibular canal via the mandibular foramen. Below lateral pterygoid it is accompanied by the inferior alveolar artery, a branch of the first part of the maxillary artery, which also enters the canal with associated veins. The subsequent course of the inferior alveolar nerve is described on page 601. Otic ganglion
This is a small, oval, flat reddish-grey ganglion (see Figs 30.7, 30.8) situated just below the foramen ovale. It is a peripheral parasympathetic ganglion related topographically to the mandibular nerve, but connected functionally with the glossopharyngeal nerve. Near its junction with the trigeminal motor root, the mandibular nerve lies lateral to the ganglion; tensor veli palatini lies medially, separating the ganglion from the cartilaginous part of the pharyngotympanic tube, and the middle meningeal artery is posterior to the ganglion. The otic ganglion usually surrounds the origin of the nerve to medial pterygoid. Like all parasympathetic ganglia, there are three roots, motor, sympathetic and sensory. Only the parasympathetic fibres relay in the ganglion. The motor, parasympathetic, root of the otic ganglion is the lesser petrosal nerve, conveying preganglionic fibres from the glossopharyngeal nerve which originate from neurones in the inferior salivatory nucleus. The lesser petrosal nerve runs intracranially in the middle cranial fossa on the anterior surface of the petrous bone before passing through the foramen ovale to join the otic ganglion. The nerve synapses in the otic ganglion, and postganglionic fibres pass by a communicating branch to the auriculotemporal nerve and so to the parotid gland. The sympathetic root is from a plexus on the middle meningeal artery. It contains postganglionic fibres from the superior cervical sympathetic ganglion which traverse the otic ganglion without relay and emerge with parasympathetic fibres in the connection with the auriculotemporal nerve to supply blood vessels in the parotid gland. The sensory fibres from the gland are derived from the auriculotemporal nerve. Clinical observations suggest that in humans the gland also receives secretomotor fibres through the chorda tympani. Branches A branch connects the otic ganglion to the chorda tympani nerve, while another ramus ascends to join the nerve of the pterygoid canal. These branches may form an additional pathway by which gustatory fibres from the anterior two-thirds of the tongue may reach the facial ganglion without traversing the middle ear, and they do not synapse in the otic ganglion. Motor branches to tensor veli palatini and tensor tympani, derived from the nerve to medial pterygoid, also pass through the ganglion without synapsing. Chorda tympani nerve
The chorda tympani nerve enters the infratemporal fossa region by passing through the medial end of the petrotympanic fissure behind the capsule of the temporomandibular joint. The nerve descends medial to the spine of the sphenoid bone - which it sometimes grooves - lying posterolateral to tensor veli palatini. It is crossed medially by the middle meningeal artery, the roots of the auriculotemporal nerve and by the inferior alveolar nerve (Fig. 30.6). The chorda tympani joins the posterior aspect of the lingual nerve at an acute angle. It carries taste fibres for the anterior two-thirds of the tongue and efferent preganglionic parasympathetic (secretomotor) fibres destined for the submandibular ganglion in the floor of the mouth.
LE FORT AND ZYGOMATIC FRACTURES (p. 489) Le Fort I, II or III fractures inevitably involve the infratemporal fossa. The bones of the midface transmit the forces of impact directly to the cranium. The most important strut related to the infratemporal and pterygopalatine fossae is the pterygomaxillary strut. Fractures involving this strut may extend elsewhere to involve the cranial base and orbit. The associated soft tissue damage which accompanies these fractures may damage nerves, blood vessels and muscles. Injuries to the second or third divisions of the trigeminal nerve or the chorda tympani nerve result in altered sensation to the oral cavity, face and jaws, including impaired taste. Fractures extending into the orbit may result in decreased visual acuity and ophthalmoplegia. Neural damage to motor nerves or direct damage to muscles may result in problems with chewing, swallowing, speech, middle ear function and eye movements. Injuries that involve the pterygopalatine or otic ganglia interfere with lacrimation, nasal secretions and salivation. Damage to adjacent blood vessels may result in haemorrhage, thrombosis emboli and the formation of false aneurisms or arteriovenous fistulae.
Classic zygomatic complex fractures involve the lateral wall of the orbit and cross laterally into the infratemporal fossa at the frontozygomatic suture. From this point the fracture line extends inferiorly to join the most lateral aspect of the inferior orbital fissure, continues inferiorly along the posterior surface of the zygomatic buttress - where it communicates with the lateral bulge of the maxillary antrum and runs around the zygomatic buttress, high in the buccal sulcus in the upper molar region, and then extends upwards towards the infraorbital foramen. It finally runs laterally along the floor of the orbit to reach the lateral extension of the inferior orbital fissure. These fractures involve the maxillary sinus and the infratemporal fossa and orbit, which means that any blood which collects in the antrum, if it becomes infected, will allow infection to spread into the infratemporal fossa. Infection in this area can have grave consequences and can rapidly spread through the foramina in the skull base into the middle cranial fossa. For this reason patients presenting with zygomatic complex fractures are placed on prophylactic antibiotic therapy to prevent infection.
SPREAD OF INFECTION FROM THE INFRATEMPORAL FOSSA (p. 607) page 525 page 526
The majority of infected teeth in the upper jaw and those in the front part of the lower jaw will generally drain harmlessly into the oral cavity - either via the vestibule buccally, or via the palate or mouth lingually - and they are of little clinical significance. In contrast, a pericoronitis affecting a partially impacted mandibular third molar tooth, or less commonly either a dental abscess of this tooth, or an infection following tooth extraction, spreads into the infratemporal fossa. Infection may also result from an infected needle used during an inferior alveolar nerve block, or as a result of spread from an adjacent infected tissue space. The main symptom caused by infection of the pterygomandibular space is trismus - painful reflex muscle spasm - which usually affects medial pterygoid. Infection may potentially spread some distance from the infratemporal fossa because the latter lies between the tissue spaces of the face above and the tissue spaces of the neck below. Thus infection may spread to involve the buccal tissue space, or directly around the back of the maxillary tuberosity and into the orbit via the inferior orbital fissure, which may result in a cavernous sinus thrombosis. Once in the orbit, infection may spread directly through the superior orbital fissure into the cranial cavity. Infection may also spread from the infratemporal fossa via the pterygomaxillary fissure to involve the pterygopalatine fossa and its contents, and may spread further via a number of small canals which lead from the fossa into the nose, pharynx and palate.
© 2008 Elsevier
TEMPOROMANDIBULAR JOINT UPDATE Date Added: 01 December 2004 Abstract: The deep subfacial approach to the temporomandibular joint. Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=15346360 The deep subfacial approach to the temporomandibular joint. Each joint involves the articular fossa (also known as the mandibular fossa or glenoid fossa) above and the mandibular condyle below. These have been described on page 482, and are considered here in more detail. The articular eminence, a transversely elliptical region sinuously curved in the sagittal plane and tilted forward at c.25° to the occlusal plane, forms most of the articular surface of the articular fossa. Its steepness is variable, and it becomes flatter in the edentulous. Its anterior limit is the summit of the articular eminence, a transverse ridge that extends laterally out to the zygomatic arch as far as the articular tubercle. Articular tissue extends anteriorly beyond the articular summit and on to the preglenoid plane. Posteriorly it extends behind the depth of the fossa as far as the squamotympanic fissure. A postglenoid tubercle (at the root of the zygomatic arch, just anterior to the fissure) is usually poorly developed in human skulls. The articular surface of the mandibular condyle is slightly curved and tilted forward at c.25° to the occlusal plane. Like the articular eminence, its slope is variable. In the coronal plane its shape varies (Osborn & Baranger 1992) from that of a gable (particularly marked in those whose diet is hard), to roughly horizontal in the edentulous. It is probably impossible to measure the pressure developed on the articular surfaces of the human jaw joint when biting. There is, however, irrefutable theoretical evidence based on Newtonian mechanics that the jaw joint is a weightbearing joint. With a vertical bite force of 500N on the left first molar the right condyle must support a load of well over 300N (Osborn 1995a). The non-working condyle is more loaded than the condyle on the working side, which may help explain why patients with a fractured condyle choose to bite on the side of the fracture. UPDATE Date Added: 19 July 2005 Abstract: Radiographic examination of temporomandibular joint. Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=15371321&query_hl=9 Radiographic examination of temporomandibular joint. UPDATE Date Added: 08 July 2005 Abstract: Improved Interaction models of temporomandibular joint Click on the following link to view the abstract:
http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=15508998&query_hl=11 Improved interaction models of temporomandibular joint anatomic relationships.
FIBROUS CAPSULE The lower part of the joint is surrounded by tight fibres which attach the condyle of the mandible to the disc. The upper part of the joint is surrounded by loose fibres which attach the disc to the temporal bone (Fig. 30.9). Thus the articular disc is attached separately to the temporal bone and to the mandibular condyle forming what could be considered two joint capsules. Longer fibres joining the condyle directly to the temporal bone may be regarded as reinforcing. The capsule is attached above to the anterior edge of the preglenoid plane, posteriorly to the lips of the squamotympanic fissure, between these to the edges of the articular fossa, and below to the periphery of the neck of the mandible.
LIGAMENTS Sphenomandibular ligament
The sphenomandibular ligament (Fig. 30.9) is medial to, and normally separate from, the capsule. It is a flat, thin band that descends from the spine of the sphenoid and widens as it reaches the lingula of the mandibular foramen. Some fibres traverse the medial end of the petrotympanic fissure and attach to the anterior malleolar process. This part is a vestige of the dorsal end of Meckel's cartilage.
page 526 page 527
Figure 30.9 The left temporomandibular joint. A, lateral aspect; B, medial aspect; C, sagittal section.
With the jaw closed, there is c.5 mm slack within the ligament, but it becomes taut when the jaw is about half open. Lateral pterygoid and the auriculotemporal nerve are lateral relations, the chorda tympani nerve lies medial near its upper end and medial pterygoid is an inferomedial relation. The sphenomandibular ligament is
separated from the neck of the mandible below lateral pterygoid by the maxillary artery and from the ramus of the mandible by the inferior alveolar vessels and nerve and a parotid lobule. At this point the vessels and nerve to mylohyoid pierce the ligament. It is separated from the pharynx by fat and a pharyngeal vein. Stylomandibular ligament
The stylomandibular ligament (Fig. 30.9) is a thickened band of deep cervical fascia that stretches from the apex and adjacent anterior aspect of the styloid process to the angle and posterior border of the mandible. Its position and orientation indicate that it cannot mechanically constrain any normal movements of the mandible and does not seem to warrant the status of a ligament of the joint. Temporomandibular (lateral) ligament
This broad ligament (Fig. 30.9) is attached above to the articular tubercle on the root of the zygomatic process of the temporal bone. It extends downwards and backwards at an angle of c.45° to the horizontal, to attach to the lateral surface and posterior border of the neck of the condyle, deep to the parotid gland. It appears to be poorly developed in the edentulous. A short, almost horizontal, band of collagen connects the articular tubercle in front to the lateral pole of the condyle behind. It may function to prevent posterior displacement of the resting condyle.
SYNOVIAL MEMBRANE The synovial membrane lines the inside of the capsule of the joint but does not extend to cover the disc or the articular surfaces.
ARTICULAR DISC
Figure 30.10 A sagittal section of the temporomandibular joint. The upper and lower joint spaces are normally compressed. They have been widened to illustrate the anteroposterior extent of each. The bilaminar posterior region contains a venous plexus.
The transversely oval articular disc is composed predominantly of dense fibrous connective tissue (Fig. 30.10). It has a thick margin which forms a peripheral annulus and a central depression in its lower surface that accommodates the articular surface of the mandibular condyle. The depression probably develops as a mechanical response to pressure from the condyle as it rotates inside the annulus. The disc is stabilized on the condyle in three ways. Its edges are fused with the part of the capsular ligament that tightly surrounds the lower joint compartment and is attached around the neck of the condyle. Well-defined bands in the capsular ligament attach the disc to the medial and lateral poles of the condyle. The thick annulus prevents the disc sliding off the condyle, provided that the condyle and disc are firmly lodged against the articular fossa (as is normally the case). In sagittal section, the disc appears to possess a thin intermediate zone and thickened anterior and posterior bands, and its upper surface appears concavoconvex where it fits against the convex articular eminence and the concavity of the articular fossa. Posteriorly the disc is attached to a region of loose vascular and nervous tissue which splits into two laminae, the bilaminar region: unlike the rest of the disc, its normal function is to provide attachment rather than intra-
articular support. The upper lamina, composed of fibroelastic tissue, is attached to the squamotympanic fissure, and the lower lamina, composed of fibrous nonelastic tissue, is attached to the back of the condyle. The bilaminar region contains a venous plexus, but the central part of the disc is avascular. The collagen is crimped, and this probably serves to absorb energy when a sudden tensile force is applied, and so briefly protects the disc from potential rupture. Cells in the disc also secrete chondroitin sulphate - a glycosaminoglycan found in cartilage - which is most heavily concentrated in the centre of the disc and which probably gives the disc some of the resilience and compressive strength of cartilage. The amount increases in response to load and to age, and by the fifth decade the disc shows signs of ageing including fraying, thinning and perforation. Functions of the articular disc
The functions of the articular disc remain controversial. It is generally believed that the disc helps to stabilize the temporomandibular joint. The articulating surfaces of the mandibular condyle and the articular fossa fit together poorly (Fig. 30.10) and are therefore separated by an irregular space. Muscle forces control the position of the mandible, and therefore of the condyle, in relation to the articular eminence, and these in turn set the shape and thickness of the irregular space. The position of the disc is controlled by neuromuscular forces: the upper head of lateral pterygoid anteriorly, and the elastic tissue in the bilaminar region posteriorly, together pull the disc backward or forward to keep the joint space filled and thereby stabilize the condyle. The articular disc may reduce wear, because the frictional wear on the condyle and the articular eminence is halved by separating slide and rotation into different joint compartments. It may also aid lubrication of the joint because it stores fluid that is squeezed out to create a weeping lubricant from the loaded part of the disc. A final view is based on the fact that the addition of a slippery disc doubles the number of virtually friction free sliding surfaces suggesting that its function is to destabilize the condyle (certainly not stabilize it) in the same way that stepping on a banana skin destabilizes the foot. All other joints are most heavily loaded when their articular surfaces are closely fitted together, creating a large area of contact, and braced to prevent further movement. However the condyle of the mandible is most heavily loaded when it is required to move, sliding backward during the buccal phase of the power stroke of a masticatory cycle on the opposite side of the jaw. If the articular tissues were composed of hyaline cartilage, the small area of contact between poorly fitting surfaces would promote free sliding (Fig. 30.11A) but simultaneously create a potentially damaging pressure (force per unit area). Making them of compressible fibrous tissue would reduce the pressure by increasing the surface area of contact and thereby spread the load, but the compressed tissues would interfere with free sliding movements (Fig. 30.11B). The problem is overcome by fitting a disc between them (Fig. 30.11C), which therefore destabilizes the condyle. In this context it is perhaps telling that the pathological absence of a disc, and not its presence, stabilizes the condyle during grinding movements and thereby renders the articular tissues vulnerable to damage. If these tissues respond by increasing their cartilaginous properties, the
increased resistance to compression reduces the area of contact and results in damagingly large articular pressures. If they respond by becoming more fibrous the sliding condyle destructively gouges through the compressed fibrous tissue on the articular eminence. Temporomandibular joint syndrome page 527 page 528
Figure 30.11 The advantages and disadvantages of covering articular surfaces with cartilage (A) and fibrous tissue (B). The addition of a fibrous disc (C) decreases the intra-articular pressure while simultaneously facilitating loaded sliding movements, unique requirements of the temporomandibular joint. (By kind permission from Dr JW Osborn.)
Symptoms arising from the temporomandibular joints and their associated masticatory muscles are very common (temporomandibular joint syndrome/internal derangement). Diffuse facial pain due to masseteric muscle spasm, headache due to temporalis muscle spasm and jaw ache due to lateral pterygoid spasm are typical presenting symptoms. These may be associated with clicking, which is often audible whilst the patient is chewing, and sometimes locking, when the patient is unable to open fully. Changes in the normal structure of the articular disc occur and the disc does not smoothly follow the movements of the condyle. Clicking and locking occur when the attachment of the articular disc posteriorly to the squamotympanic fissure becomes stretched or detached, allowing the disc to become temporarily or permanently trapped anteriorly. These symptoms affect predominantly adolescents and young adults and affect females more frequently than males. The symptoms occur particularly when the subject is under stress. Although predisposing factors have been implicated, such as the nature of the dental occlusion, the morphology of the head of the condyle, and variations in the attachments of lateral pterygoid, the precise aetiology of temporomandibular joint syndrome awaits clarification.
VASCULAR SUPPLY AND INNERVATION The articular tissues and the dense part of the articular disc have no nerve supply. Branches of the auriculotemporal and masseteric nerves and postganglionic sympathetic nerves supply the tissues associated with the capsular ligament and the looser posterior bilaminar extension of the disc. The temporomandibular joint
capsule, lateral ligament and retroarticular tissue contain mechanoreceptors and nociceptors. The input from mechanoreceptors provides a source of proprioceptive sensation that helps control mandibular posture and movement. The joint derives its arterial supply from the superficial temporal artery laterally and the maxillary artery medially. Penetrating vessels that supply lateral pterygoid may also supply the condyle of the mandible. Veins drain the anterior aspect of the joint and associated tissues into the plexus surrounding lateral pterygoid, and posteriorly they drain into the vascular region that separates the two laminae of the bilaminar region of the disc. Pressure produced by forward and backward movement of the condyle shunts blood between these regions. Lymphatics drain deeply to the upper cervical lymph nodes surrounding the internal jugular vein.
JAW MOVEMENTS Muscle synergism
During opening of the mouth, the incisors of adults may be separated by 50-60 mm, and this involves c.35° rotation of the mandible. The mandible may be protruded or laterally displaced c.10 mm, although this is very variable. The adult range of movements is reached at c.10 years in females and 15 years in males. When a jaw muscle contracts in the absence of a restraint, the mandible is pulled in the direction of the shortening muscle. The muscles have therefore been classified as protruders (lateral and medial pterygoids); retractors (posterior fibres of temporalis assisted by digastric and geniohyoid); elevators (anterior and middle fibres of temporalis, superficial and deep fibres of masseter and medial pterygoid); depressors (lateral pterygoids aided by digastric, geniohyoid and mylohyoid); lateral movers (medial and lateral pterygoids of each side). While these descriptions are correct, they fail to explain why the jaw muscles are so powerful, and ignore the synergism of these muscles in the generation of bite force. For example, the lateral pterygoids are each capable of exerting about 160N force, a total of 320N, and yet less than 5% of this is used to protrude or open the jaw. The jaw muscles normally exert large forces only during the power stroke of mastication. In general, a jaw muscle is most active when a line joining its attachments is more parallel to the bite force. Each jaw muscle has a component of force in parasagittal, coronal and horizontal planes. For example, masseter pulls the mandible upward, forward and outward. Moreover, active jaw muscles often have a component of force in one plane that is unwanted, e.g. the outward pull of the left masseter opposes a medially directed power stroke on the left molars. This wasteful lateral component must be counteracted by another muscle, even one that is poorly placed to help produce the required output force. If the activity of all the jaw muscles is analysed when they create a given bite force, it is possible to subdivide them into 'power' muscles that largely create the force, and 'balancing' muscles that largely counteract unwanted components and help to increase the output force by more than their contribution to that force. For example, the superior head of lateral pterygoid acts as a balancing muscle during clenching, even though it is very poorly placed to increase the bite force, because it allows temporalis to be much more active by balancing its unwanted backward
force component. It becomes a power muscle during protrusion because its activity increases the output force by an amount equal to its contribution to a protrusive force (see Osborne 1995a). Movements of the condyle in the temporomandibular joint
The major function of the mandible is to exert, via the teeth, the force necessary to break down food into smaller particles and so facilitate digestion. Pure vertical movements of the lower teeth create a crushing force that is ineffective in breaking up tough fibrous food. Man uses a lateral movement of the lower jaw to create a shear component of force that enhances the effectiveness of the power stroke of mastication. Bodily lateral movement of the whole jaw, the Bennet shift, is insignificant. Extensive lateral movement is only possible when the jaw is rotated horizontally about one condyle while the other condyle slides backward and forward. UPDATE Date Added: 07 December 2005 Publication Services, Inc. Condylar bone change and sagittal incisal and condylar paths during mandibular protrusive excursion. Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=16128352&query_hl=14 Condylar bone change and sagittal incisal and condylar paths during mandibular protrusive excursion. Yamada K, Tsuruta A, Hosogai A, et al. Cranio. 2005 Jul;23(3):179-87. The temporomandibular joint is structurally adapted to accommodate both sliding and rotation in a parasagittal plane. Sliding occurs because the capsular ligament which surrounds the upper joint compartment is loose, whereas the capsular ligament which encloses the lower joint compartment is tight, and only allows the condyle to rotate over the depression inside the annulus of the articular disc. Symmetrical opening
Symmetrical jaw opening is associated with preparation for incising. At the start, each mandibular condyle rotates in the lower joint compartment inside the annulus of its disc. After a few degrees of opening, the condyle continues rotating inside its disc, and, in addition, both slide forward down the articular eminence of the upper joint compartment. Without this forward slide, it rapidly becomes impossible to continue opening the jaw. page 528 page 529
There are conflicting views about the reason why forward slide occurs, probably because direct experimental testing is not possible. No other animal has an articular eminence constraint and ligaments comparable to man, which means that most supporting evidence for any theory is based on analyses of human joint dysfunction. It has been argued that a sensory input from the rotary movement, possibly from either the joint capsule or the jaw muscles, initiates a response that reflexly activates muscles that cause the slide. The fact that the condyle in a cadaver still slides forward when the jaw is rotated open suggests that it is not a
neuromuscular response but is the mechanical result of physical constraints. When the jaw is rotated open the temporomandibular ligament rapidly becomes taut (Osborn 1995b). The taut ligament acts as a constraint that allows the mandible only two rotary movements: it can swing about the upper attachment of the ligament and rotate about the lower attachment. The lower end of the taut ligament acts as a moving fulcrum that converts the downward and backward pull of the opening rotary force (created at the front by digastric and geniohyoid) into one that drives the condyle upward and forward into the concavity of the overlying articular disc. This now pushes the disc forward. Swing about the upper attachment creates space above for the disc to slide further forward which is possible because the upper part of the capsular ligament is loose. The two movements, rotation and swing, are inextricably linked by the taut ligament and, via the condyle, combine to keep the disc in firm contact with the articular eminence while the jaw is opened. The disc is stabilized by its tight attachment to the condyle and by the thickened margins of its anulus that prevent it sliding through the thinner compressed region between the centre of the condyle and the articular eminence. As forward slide of the condyle continues, the controlling influence exerted by the temporomandibular ligament diminishes. The lingula of the mandible moves away from the spine of the sphenoid, tautening the originally slack sphenomandibular ligament, which now acts in the same way as the temporomandibular ligament, to maintain the condyle against the articular eminence. Symmetrical opening thus appears to consist of at least three separate phases: an early phase controlled by the temporomandibular ligament and articular eminence; a short middle phase in which either both, or neither, temporomandibular and sphenomandibular ligaments act to constrain movements; and a late phase controlled by the sphenomandibular ligament and articular eminence. Eccentric jaw opening
Eccentric jaw opening is associated with preparing for the power stroke of mastication. The mandibular condyle on the non-working side slides back and forth during lateral movements associated with the power stroke on the working side. Although the jaw muscles now have the major control over mandibular movements, the temporomandibular and sphenomandibular ligaments keep the condyle firmly against its articular eminence during opening. Eccentric and symmetrical jaw closing
The resultant of the jaw closing muscles of the mandible has a component that forces the joint surfaces together. This compresses the joint tissues and potentially shortens the ligaments so that they no longer constrain jaw movements. Under these conditions, jaw movements and the positions of the condyles are controlled by neuromuscular processes (within the limits of constraints imposed by the articular eminence, the occluding surfaces of the teeth and the presence of food between them). Note that the non-working condyle moves the furthest, and is the most heavily loaded, during the power stroke of mastication. The loads on each joint, balancing and working, drive each condyle more forcefully into its articular eminence.
Temporomandibular joint dislocation
The mandible is dislocated only forwards. With the mouth open, the mandibular condyles are on the articular eminences and sudden violence, even muscular spasm - a convulsive yawn - may displace one or both condyles onto the preglenoid plane. To reduce the dislocation the condyle must be lowered and pushed back behind the summit of the articular eminence into the articular fossa. This is achieved by first rotating the jaw closed: the chin is elevated and the lower molar region depressed by pushing down with the thumbs in the buccal sulci. Once the condyle has been lowered below the articular summit the jaw can easily be pushed back.
ANATOMY OF MASTICATION Intake of food is carried out primarily by a series of intraoral, pharyngeal and oesophageal transport mechanisms. Mastication is an interruption in the intraoral transport process which occurs when the ingested material is not of a consistency suitable for further onward transport. It is the process, characteristic of adult mammals, in which ingested food is cut or crushed into small pieces, mixed with saliva and formed into a bolus in preparation for swallowing. The intraoral transport of ingested food relies primarily upon tongue movements, whereas food breakage can be considered primarily a function of the teeth, jaws and jaw closing muscles. Nevertheless, efficient food breakage is heavily dependent upon the ability of the tongue to select and place appropriate sized food items between the occluding teeth. Initially puncturing forces are applied by the tips of the cusps of premolar/molar teeth during a simple vertical closure of the jaw (essentially the same breakage mechanism is used by the incisors when hard food is fractured prior to ingestion during the initial stages of intraoral transport). In subsequent cycles shearing forces are generated as the inclined planes of cusps of the premolar/molar teeth of one side move past each other: this occurs as the jaw closes, initially with a slight lateral movement and finally, after tooth-food-tooth contact, with the medial movement necessary for the shear production. Once the food has been mechanically processed to the point where it is suitable for swallowing, the jaw movements revert to cycles with a simple vertical path. During these cycles, tongue movements transport the processed food into the swallow. Jaw closing movements are produced by contraction of masseter, temporalis and medial pterygoid. In each cycle closing movement is initially carried out largely against gravity until tooth-food-tooth contact is made, after which closure can only occur if the food is deformed or fractured. Electromyographic activity in the jaw elevator muscles increases only moderately - from the negligible level seen just prior to jaw closure - as the jaw accelerates in closure against gravity, but the activity then increases significantly as tooth-food-tooth contact is made. The recruitment of the additional muscle activity is thought to be due to sensory feedback, probably from subsets of periodontal and muscle afferents. However, a large part of the sensory feedback from periodontal afferents is associated with tooth overload protection, and inhibits the activity of the jaw-closing muscles. Jaw opening is produced by relaxation of the jaw-closing muscles associated with
activation of the lateral pterygoid muscles to bring the condyles forward, and activation of submandibular muscles (including the digastric muscles). Profiles of cycles of jaw movement (plotted as gape versus time) for hard food and for soft food are different because the former is dominated by the proportion of the cycle devoted to the breakage of food, whereas the latter is devoted to the proportion of the cycle dedicated to tongue movement (much of which occurs in opening). When a piece of hard food is being mechanically processed, the form of the jaw cycle changes as the consistency of the food item changes. When jaw movements in eating are plotted as X-Y coordinates in the coronal plane, the mandible usually moves laterally (towards the side containing the food bolus) during closure until a time which broadly corresponds to when food contact is made. After this the teeth move medially while still closing (although the extent of this movement varies, depending upon whether puncture crushing or shearing is being performed). When the food is soft, the jaw movement tends to be a simple vertical one with little or no lateral excursion. It should be noted that the profiles of naturally occurring movements in the coronal plane are not limited anatomically except by tooth contact; the anatomically limited envelope of motion is normally substantially larger. The envelope of motion
The envelope of motion is the volume of space within which all movements of a point on the mandible have to occur because the limits are set by anatomical features i.e. by the shape or size of the upper and lower jaws, by tooth contacts, by the attachment and insertions of muscle and ligaments. In consciously controlled movement of the jaw from the rest position to the fully opened position, the trajectory of the mandibular incisal edge is two-phased. The first phase is a hinge-like movement during which the condyles are retruded within the mandibular fossae. When the teeth are opened by c.25 mm, the second phase of opening occurs by anterior movement or protrusion of the condyles down the articular eminences with further rotation. If conscious effort is used, a closure path can then be followed in which the jaw is closed to an extreme protruded tooth contact position after which it has to be retruded to the starting position. The free, habitual, unconscious movement during both jaw opening and closing has a significantly more limited trajectory. Similar considerations apply to lateral movements: mandibular rotation around a retruding condyle and the protraction of the opposite condyle are anatomically limiting factors that are again rarely encountered in normal function. page 529 page 530
The majority of normal movements of the jaw are not consciously controlled in the sense that they are totally deliberate throughout their course. They are largely automatic, indeed, some rhythmic movements can even be performed in the absence of functioning cerebral hemispheres. The current view is that the rhythmic movements of the jaw are produced by a central pattern generator (CPG) in the brainstem, which is activated by sensory input from nerves of the orofacial region (e.g. by material in the mouth) and/or by intact descending
influences from cerebrocortical sources. The nature of the rhythmic output from the CPG may also be subject to subsequent modification by oral sensory input and by descending influences. This helps to explain the change in cycle type as different foods are processed. In all such cyclical movements anatomical constraints can act only during tooth contact in closing and during the transition from hinge axis opening to condylar sliding. The main limitations set on rhythmic movements, so that they conform to particular profiles, are those set by neural controls. REFERENCES Barker BCW, Davies PL 1972 The applied anatomy of the pterygomandibular space. Br J Surg 10: 43-55. Describes the relationships of the structures within the pterygomandibular space, with particular reference to anaesthesia associated with an inferior alveolar nerve block. Bertilsson O, Strom D 1995 A literature survey of a hundred years of anatomic and functional lateral pterygoid muscle research. J Orofac Pain 9: 17-23. Medline Similar articles Pogrel MA, Renaut A, Schmidt B, Ammar A 1995 The relationship of the lingual nerve to the mandibular third molar region. J Oral Maxillofac Surg 53: 1178-81. Describes the relationships between the lingual nerve and the third molar tooth and the clinical relevance of this knowledge to the extraction of such teeth. Medline Similar articles Full article Lang J 1995 Clinical Anatomy of the Masticatory Apparatus and Peripharyngeal Spaces. New York: Thième Medical Publishers. Provides detailed anatomical information (including quantitation) of the infratemporal fossa, relating such information to the clinic. Langdon JD, Patel MF (eds) 1998 Operative Maxillofacial Surgery. London: Chapman and Hall. Langdon JD, Berkovitz BKB, Moxham BJ 2002 The Surgical Anatomy of the Infratemporal Fossa. London: Dunitz. The two previous references are textbooks which contain detailed information about surgical approaches associated with the infratemporal fossa, including access to the brain. Orchardson R, Cadden W 1998 Mastication. In: Linden RWA (ed) The Scientific Basis of Eating. Taste and Smell, Salivation, Mastication and Swallowing and their Dysfunctions. Frontiers of Oral Biology Series Vol 9. Basel: Karge: 76-121. Turker KS 2002 Reflex control of human jaw muscles. Crit Rev Oral Biol Med 13: 85-104. These two last papers contain detailed descriptions of the process of mastication. Medline Similar articles Osborn JW 1985 The disc of the human temporomandibular joint: design, function and failure. J Oral Rehab 12: 279-93. Analyses of the design, function and physical properties of the articular disc and the capsular ligament including explanations for the origin of disc displacement and clicking joints. Osborn JW 1995a Biomechanical implications of lateral pterygoid contributions to biting and jaw opening in humans. Arch Oral Biol 40: 1099-108. Medline Similar articles Full article Osborn JW 1995b Internal derangement and the accessory ligaments around the temporomandibular joint. J Oral Rehab 22: 731-40. Analyses of the anatomy of the ligaments associated with the jaw joint and how they control the movements of the condyle during border movements of the jaw associated with jaw opening and the aetiology of jaw dislocation and disc displacement. Osborn JW, Baranger FA 1992 Observed shapes of human condyles. J Biomechan 25: 967. Describes the shapes of human jaw joints in three dimensions, and analyses how the shapes are related to the different loads that need to be supported at the joint surfaces during biting. Includes predictions about changes in joint loads and jaw muscle activity in response to theoretical changes in joint and muscle receptors.
© 2008 Elsevier
31 NECK AND UPPER AERODIGESTIVE TRACT Neck The neck extends from the base of the cranium and the inferior border of the mandible to the thoracic inlet. For gross descriptive purposes it can be divided into musculoskeletal, visceral and two laterally positioned neurovascular compartments, that are each delineated by fascia. The investing layer of deep cervical fascia encloses trapezius and sternocleidomastoid. Anteriorly and on each side, platysma lies between the fascia and the skin, and the infrahyoid muscles lie posterior to the fascia. The musculoskeletal compartment is posterior and contains the vertebral column and the prevertebral and postvertebral muscles, enclosed by the prevertebral fascia. The visceral compartment occupies the midline of the neck and, depending on the level, may contain the pharynx or the oesophagus; the larynx or the trachea; the thyroid and parathyroid glands. The compartment is enclosed by pretracheal fascia. The neurovascular compartments lie on each side of the neck beneath sternocleidomastoid and contain the common carotid or internal carotid artery, the internal jugular vein, and the vagus nerve, all enclosed within the carotid sheath.
© 2008 Elsevier
SKIN The skin in the neck is normally under tension, and the direction in which this is greatest varies regionally. In the living face, these lines often coincide with wrinkle lines. Lines of greatest tension have been termed 'relaxed skin tension lines': surgical incisions made along these lines are said to heal with minimal postoperative scarring.
Cutaneous vascular supply and lymphatic drainage The blood vessels supplying the skin of the neck are derived chiefly from the facial, occipital, posterior auricular and subclavian arteries. They form a rich network within platysma and in the subdermal plexus, and account for the viability of the various skin flaps raised during block dissection of the neck, irrespective of whether they include platysma. The vessels supplying the anterior skin of the neck are derived mainly from the superior thyroid artery and the transverse cervical branch of the subclavian artery. The posterior skin is supplied by branches from the occipital artery and the deep cervical and transverse cervical branches of the subclavian artery. Superiorly, the skin is supplied from the occipital artery and its upper sternocleidomastoid branch, and the submandibular and submental branches of the facial artery. Inferiorly, the skin of the neck is supplied from the transverse cervical and/or suprascapular branches of the subclavian artery. The pattern of venous drainage of the skin of the neck mirrors the arterial supply, and drains into the jugular and facial veins. UPDATE Date Added: 25 January 2005 Shanida Helena Nataraja, PhD (Dianthus Medical Limited) Update: Population-based atlas aids with determination of clinical target volume of the head and neck lymph nodes Target volume definition is essential for successful lymph node radiotherapy; however, many small lymph nodes are below the resolution of conventional computerized tomography (CT) or magnetic resonance imaging (MRI). Accordingly, the delineation of nodal target volume therefore relies on the correlation of lymph nodes with other anatomical structures visible on CT images. Poon et al. have recently assembled a population-based, three-dimensional, lymph node target volume atlas of the head and neck lymph nodes. T2 weighted axial MRI images, taken from patients with suspected head and neck cancers, were retrospectively reviewed at the University of California San Francisco Head and Neck oncology service. Each lymph node visible on these images was marked by an experienced head and neck radiologist. Furthermore, to elucidate the relationship between identifiable anatomical landmarks and specific lymph node groups, each image was assigned to one of 12 distinct baseline levels on the basis of the presence of specific external contours and anatomical structures: four levels (Jaw 1, 2, 3, and 4) for the jaw and 8 levels (A-H: nasopharynx; maxillary alveolus/hard palate; upper oropharynx/ascending ramus of mandible/parotid gland; thyroid cartilage and glottic structures; thyroid gland; apices of lungs and clavicles) for all other areas of the neck. During a threemonth period, 503 MRI images were taken from 35 patients (22 men; 13 women). Within levels A, B, C, D, E, F, G, and H, there were 22, 44, 206, 199, 196, 175, 63, and 35 lymph nodes, respectively. A further 121 marked lymph nodes were distributed among the four jaw levels. Significant variations in the lymph node location were seen for all lymph node groups: 89 retropharyngeal (RP) lymph nodes were found, predominantly in levels A, B, and C; 576 jugular lymph nodes were found at all levels, but mostly in levels C, D, and E; and 204 spinal accessory lymph nodes were found, mainly in levels E and F, although smaller number of lymph nodes were also seen in levels D, G, and H. The findings of this study have produced a population-based lymph node map that can be used to define the clinical target volume before radiotherapy. Poon I, Fischbein N, Lee N, Akazawa P, Xia P, Quivey J, Phillips T. A population-based atlas and clinical target volume for the head-and-neck lymph nodes. Int J Radiat Oncol Biol Phys. 2004;59(5):1301-11. Medline Similar articles
page 531 page 532
Figure 31.1 Lymph nodes of the head and neck. Adapted from Montgomery WW 2002 Surgery of the Larynx, Trachea, Esophagus and Neck. Philadelphia: Saunders. Inset courtesy of Professor John D Langdon, GKT Schools of Medicine, Dentistry and Biomedical Sciences.
Many lymphatic vessels draining the superficial cervical tissues skirt the borders of sternocleidomastoid to reach the superior or inferior deep cervical nodes. Some pass over sternocleidomastoid and the posterior triangle to the superficial cervical and occipital nodes (Fig. 31.1). Lymph from the superior region of the anterior triangle drains to the submandibular and submental nodes. Vessels from the anterior cervical skin inferior to the hyoid bone pass to the anterior cervical lymph nodes near the anterior jugular veins, and their efferents go to the deep cervical nodes of both sides, including the infrahyoid, prelaryngeal and pretracheal groups. An anterior cervical node often occupies the suprasternal space. UPDATE Date Added: 19 April 2005 Abstract: A new imaging-based classification for describing location of lymph nodes in neck Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=10859519 A new imagingbased classification for describing location of lymph nodes in neck.
Cutaneous innervation (Figs 29.9, 31.2, 46.14, 45.58) The skin of the neck is innervated by branches of cervical spinal nerves, via both dorsal and ventral rami. The dorsal rami supply skin over the back of the neck and scalp, and the ventral rami supply skin covering the lateral and anterior portions of the neck, and extend onto the face over the angle of the mandible. The dorsal rami of the first, sixth, seventh and eighth cervical nerves have no cutaneous distribution in the neck. The greater occipital nerve mainly supplies the scalp and comes from the medial branch of the dorsal ramus of the second cervical nerve. The medial branches of the dorsal rami of the third, fourth and fifth cervical nerves pierce trapezius to supply skin over the back of the neck sequentially. The ventral rami of the second, third and fourth cervical nerves supply named cutaneous branches (the lesser occipital, great auricular, transverse cutaneous and supraclavicular nerves), via the cervical plexus located deep to sternocleidomastoid. The cervical plexus is described in detail on page 555.
LESSER OCCIPITAL NERVE
Figure 31.2 A plan of the cervical plexus.
The lesser occipital nerve is derived mainly from the second cervical nerve (although fibres from the third cervical nerve may sometimes contribute). It curves around the accessory nerve and ascends along the posterior margin of sternocleidomastoid. Near the cranium it perforates the deep fascia and passes up onto the scalp behind the auricle. It supplies the skin and connects with the great auricular and greater occipital nerves and the auricular branch of the facial nerve. Its auricular branch supplies the skin on the upper third of the medial aspect of the auricle and connects with the posterior branch of the great auricular nerve. The auricular branch is occasionally derived from the greater occipital nerve. It has been suggested that compression or stretching of the lesser occipital nerve contributes to cervicogenic headache.
GREAT AURICULAR NERVE This is the largest ascending branch of the cervical plexus. It arises from the second and third cervical rami, encircles the posterior border of sternocleidomastoid, perforates the deep fascia and ascends on the muscle beneath platysma with the external jugular vein. It passes to the parotid gland, dividing into anterior and posterior branches. The anterior branch is distributed to the facial skin over the parotid gland, connecting in the gland with the facial nerve. The posterior branch supplies the skin over the mastoid process and on the back of the auricle (except its upper part); a filament pierces the auricle to reach the lateral surface where it is distributed to the lobule and concha. The posterior branch communicates with the lesser occipital, the auricular branch of the vagus and the posterior auricular branch of the facial nerve.
TRANSVERSE CUTANEOUS (CERVICAL) NERVE OF THE NECK The transverse cutaneous nerve arises from the second and third cervical rami, curves round the posterior border of sternocleidomastoid near its midpoint and runs obliquely forwards, deep to the external jugular vein, to the anterior border of the muscle. It perforates the deep cervical fascia, and divides under platysma into
ascending and descending branches that are distributed to the anterolateral areas of the neck. The ascending branches ascend to the submandibular region, forming a plexus with the cervical branch of the facial nerve beneath platysma. Some branches pierce platysma and are distributed to the skin of the upper anterior areas of the neck. The descending branches pierce platysma and are distributed anterolaterally to the skin of the neck, as low as the sternum.
SUPRACLAVICULAR NERVES The supraclavicular nerves arise from a common trunk formed from rami from the third and fourth cervical nerves and emerge at the posterior border of sternocleidomastoid. Descending under platysma and the deep cervical fascia, the trunk divides into medial, intermediate and lateral (posterior) branches, which diverge to pierce the deep fascia a little above the clavicle. The medial supraclavicular nerves run inferomedially across the external jugular vein and the clavicular and sternal heads of sternocleidomastoid to supply the skin as far as the midline and as low as the second rib. They supply the sternoclavicular joint. The intermediate supraclavicular nerves cross the clavicle to supply the skin over pectoralis major and deltoid down to the level of the second rib, next to the area of supply of the second thoracic nerve. Overlap between these nerves is minimal. The lateral supraclavicular nerves descend superficially across trapezius and the acromion, supplying the skin of the upper and posterior parts of the shoulder.
© 2008 Elsevier
TRIANGLES OF THE NECK (Figs 31.3, 31.4) page 532 page 533
Figure 31.3 Muscles of the front of the neck. Sternocleidomastoid has been removed on the right side. In this subject, the origin of scalenus medius is extended up to the transverse process of the atlas.
Anterolaterally the neck appears as a somewhat quadrilateral area, limited superiorly by the base of the mandible and a line continued from the angle of the mandible to the mastoid process, inferiorly by the upper border of the clavicle, anteriorly by the anterior median line, and posteriorly by the anterior margin of trapezius. This quadrilateral area can be further divided into anterior and posterior triangles by sternocleidomastoid, which passes obliquely from the sternum and clavicle to the mastoid process and occipital bone. It is true that these triangles and their subdivisions are somewhat arbitrary, because many major structuresarteries, veins, lymphatics, nerves, and some viscera-transgress their boundaries without interruption, nevertheless they have a topographical value in description. Moreover, some of their subdivisions are easily identified by inspection and palpation and provide invaluable assistance in surface anatomical and clinical examination.
Anterior triangle of the neck
Figure 31.4 The triangles of the left side of the neck. This is a highly schematic two dimensional representation of what in reality are non-planar trigones distributed over a waisted column.
The anterior triangle of the neck is bounded anteriorly by the median line of the neck and posteriorly by the anterior margin of sterno-cleidomastoid. Its base is the inferior border of the mandible and its projection to the mastoid process, and its apex is at the manubrium sterni. It can be subdivided into suprahyoid and infrahyoid areas above and below the hyoid bone, and into digastric, submental, muscular and carotid triangles by the passage of digastric and omohyoid across the anterior triangle.
DIGASTRIC TRIANGLE The digastric triangle is bordered above by the base of the mandible and its projection to the mastoid process, posteroinferiorly by the posterior belly of digastric and by stylohyoid, and anteroinferiorly by the anterior belly of digastric (Fig. 31.5). It is covered by the skin, superficial fascia, platysma and deep fascia, which contain branches of the facial and transverse cutaneous cervical nerves. Its floor is formed by mylohyoid and hyoglossus. The anterior region of the digastric triangle contains the submandibular gland, which has the facial vein superficial to it and the facial artery deep to it. The submental and mylohyoid arteries and nerves lie on mylohyoid. The submandibular lymph nodes are variably related to the submandibular gland. The posterior region of the digastric triangle contains the lower part of the parotid gland. The external carotid artery, passing deep to stylohyoid, curves above the muscle, and overlaps its superficial surface as it ascends deep to the parotid gland before entering it. The internal carotid artery, internal jugular vein and vagus nerve lie deeper and are separated from the external carotid artery by styloglossus, stylopharyngeus and the glossopharyngeal nerve.
SUBMENTAL TRIANGLE
The single submental triangle is demarcated by the anterior bellies of both digastric muscles. Its apex is at the chin, its base is the body of the hyoid bone and its floor is formed by both mylohyoid muscles. It contains lymph nodes and small veins that unite to form the anterior jugular vein. The structures within the digastric and submental triangles are described in more detail with the floor of the mouth (p. 583).
MUSCULAR TRIANGLE The muscular triangle is bounded anteriorly by the median line of the neck from the hyoid bone to the sternum, inferoposteriorly by the anterior margin of sternocleidomastoid and posterosuperiorly by the superior belly of omohyoid. The triangle contains omohyoid, sternohyoid, sternothyroid and thyrohyoid. page 533 page 534
Figure 31.5 Dissection of the left anterior triangle. Platysma has been divided transversely: its upper part has been turned upwards on to the face, and its lower part turned backwards, exposing the lower part of sternocleidomastoid.
CAROTID TRIANGLE The carotid triangle is limited posteriorly by sternocleidomastoid, anteroinferiorly by the superior belly of omohyoid and superiorly by stylohyoid and the posterior belly of digastric (Fig. 31.5). In the living (except the obese) the triangle is usually a small visible triangular depression, sometimes best seen with the head and cervical vertebral column slightly extended and the head contralaterally rotated. The carotid triangle is covered by the skin, superficial fascia, platysma and deep fascia containing branches of the facial and cutaneous cervical nerves. The hyoid bone forms its anterior angle and adjacent floor and can be located on simple
inspection, verified by palpation. Parts of thyrohyoid, hyoglossus and inferior and middle pharyngeal constrictor muscles form its floor. The carotid triangle contains the upper part of the common carotid artery and its division into external and internal carotid arteries. Overlapped by the anterior margin of sternocleidomastoid, the external carotid artery is first anteromedial, then anterior to the internal carotid artery. Branches of the external carotid artery are encountered in the carotid triangle. Thus the superior thyroid artery runs anteroinferiorly, the lingual artery anteriorly with a characteristic upward loop, the facial artery anterosuperiorly, the occipital artery posterosuperiorly and the ascending pharyngeal artery medial to the internal carotid artery. Arterial pulsation greets the examining finger. The superior thyroid, lingual, facial, ascending pharyngeal and sometimes the occipital, veins, correspond to the branches of the external carotid artery, and all drain into the internal jugular vein. The hypoglossal nerve crosses the external and internal carotid arteries. It curves round the origin of the lower sternocleidomastoid branch of the occipital artery, and at this point the superior root of the ansa cervicalis leaves it to descend anteriorly in the carotid sheath. The internal laryngeal nerve and, below it, the external laryngeal nerve, lie medial to the external carotid artery below the hyoid bone. Many structures in this region, such as all or part of the internal jugular vein, associated deep cervical lymph nodes, and the vagus nerve, may be variably obscured by sternocleidomastoid, and, pedantically, are thus 'outside the triangle'.
Posterior triangle of the neck The posterior triangle is delimited anteriorly by sternocleidomastoid, posteriorly by the anterior edge of trapezius, and inferiorly by the middle third of the clavicle. Its apex is between the attachments of sternocleidomastoid and trapezius to the occiput and is often blunted, so that the 'triangle' becomes quadrilateral. The roof of the posterior triangle is formed by the investing layer of the deep cervical fascia. The floor of the triangle is formed by the prevertebral fascia overlying splenius capitis, levator scapulae and the scalene muscles. It is crossed, c.2.5 cm above the clavicle, by the inferior belly of omohyoid, which subdivides it into occipital and supraclavicular triangles. Collectively these contain the cervical and brachial plexuses, the subclavian artery and the spinal accessory nerve. The muscles forming the floor of the posterior triangle constitute the anterior and lateral groups of the prevertebral musculature (Fig. 31.6). UPDATE Date Added: 31 October 2006 Helen E Wiggett, PhD (Dianthus Medical Limited) Update: Arteries in the posterior cervical triangle in humans It is difficult to understand and compare anatomic and surgical studies of arteries in the posterior cervical triangle (lateral cervical region) because the anatomic nomenclature is constantly changing. This also makes musculocutaneous flap planning in plastic and reconstructive surgery difficult. A recent study attempted to standardize the nomenclature of these vessels. The study included 498 cadaver neck halves at three investigation sites (Graz, Innsbruck, and Munich). The arteries in the Graz neck halves were injected with Thiel's DGM 85 substance to facilitate identification of the arteries. In all samples, the arteries in the right and left posterior cervical triangles were dissected to determine their origin and terminal distribution. Within the lateral cervical regions, three arteries and four trunks were identified. The four trunks were named according to the branches that arose from them. In 20% of cases, the three arteries (superficial cervical, dorsal scapular, and suprascapular) arose directly from a vessel, and in the remaining 20% of cases they arose from one of the four trunks. The four trunks (cervicodorsal, cervicoscapular, dorsoscapular, and cervicodorsoscapular) originated from the
thyrocervical trunk, the subclavian artery, or the internal thoracic artery. The superficial cervical artery arose from the cervicodorsal trunk in 30% of cases, the cervicoscapular trunk in 22% of cases, and the cervicodorsoscapular trunk in 24% of cases. In 24% of cases, it originated independently from the thyrocervical trunk (22%) or the subclavian artery (2%). The dorsal scapular artery arose from the cervicodorsal trunk (30%), the dorsoscapular trunk (4%), or the cervicodorsoscapular trunk (24%) and originated directly in 42% of cases, most commonly from the subclavian artery (37%). The suprascapular artery arose from the cervicoscapular trunk (22%), the dorsoscapular trunk (4%), or the cervicodorsoscapular trunk (24%) and arose directly in 50% of cases, most commonly from the thyrocervical trunk (27%). The authors propose replacing the term transverse cervical artery with the term cervicodorsal trunk. In this study, the rationale for identifying trunks was consistent with the convention used elsewhere in the body. This study gives a more precise and specific identification of branches arising from trunks with the aim of improving and standardizing the nomenclature of arteries in this region. Weiglein AH, Moriggl B, Schalk C, et al: Arteries in the posterior cervical triangle in man. Clin Anat 18(8):553-557, 2005.
OCCIPITAL TRIANGLE The occipital triangle constitutes the upper and larger part of the posterior triangle, with which it shares the same borders, except that inferiorly it is limited by the inferior belly of omohyoid. Its floor is constituted, from above down, by splenius capitis, levator scapulae, and scaleni medius and posterior, and semispinalis capitis occasionally appears at the apex (Fig. 31.4). It is covered by the skin, superficial and deep fasciae and below by platysma. The spinal accessory nerve pierces sternocleidomastoid and crosses levator scapulae obliquely downwards and backwards to reach the deep surface of trapezius. Cutaneous and muscular branches of the cervical plexus emerge at the posterior border of sternocleidomastoid. Inferiorly, supraclavicular nerves, transverse cervical vessels and the uppermost part of the brachial plexus cross the triangle. Lymph nodes lie along the posterior border of sternocleidomastoid from the mastoid process to the root of the neck.
SUPRACLAVICULAR TRIANGLE The supraclavicular triangle is the lower and smaller division of the posterior triangle, with which it shares the same boundaries, except that superiorly it is limited by omohyoid (Fig. 31.4). It corresponds in the living neck with the lower part of a deep, prominent hollow, namely, the greater supraclavicular fossa. Its floor contains the first rib, scalenus medius and the first slip of serra tus anterior. Its size varies with the extent of the clavicular attachments of sternocleidomastoid and trapezius and also the level of the inferior belly of omohyoid. The triangle is covered by the skin, superficial and deep fasciae and platysma and crossed by the supraclavicular nerves. Just above the clavicle, the third part of the subclavian artery curves inferolaterally from the lateral margin of scalenus anterior across the first rib to the axilla. The subclavian vein is behind the clavicle and is not usually in the triangle; but it may rise as high as the artery and even accompany it behind scalenus anterior. The brachial plexus is partly superior, and partly posterior to the artery and is always closely related to it. The trunks of the brachial plexus may easily be palpated here if the neck is contralaterally flexed and the examining finger is drawn across the trunks at right angles to their length. With the musculature relaxed, pulsation of the subclavian artery may be felt and the arterial flow can be controlled by retroclavicular compression against the first rib. The suprascapular vessels pass transversely behind the clavicle, below the transverse cervical artery and vein. The external jugular vein descends behind the posterior border of sternocleidomastoid to end in the subclavian vein. It
receives the transverse cervical and suprascapular veins, which form a plexus in front of the third part of the subclavian artery; occasionally it is joined by a small vein crossing the clavicle anteriorly from the cephalic vein. The nerve to subclavius crosses the triangle. The triangle contains some lymph nodes.
© 2008 Elsevier
ROOT OF THE NECK (Figs 31.7, 31.8, 31.9, 31.24, 31.25, 63.5) page 534 page 535
Figure 31.6 A dissection to show the prevertebral region and the superior mediastinum. On the right the costal elements of the upper six cervical vertebrae have been removed to expose the cervical part of the vertebral artery. On the left most of the deep relations of the common carotid artery and the internal jugular vein are exposed.
A number of important structures and tissue spaces pass between the neck and thorax or upper limb in the root of the neck. They include the subclavian vessels;
common carotid artery; trunks of the brachial plexus; sympathetic trunk; phrenic, vagus and recurrent laryngeal nerves (on both sides); the terminal portion of the thoracic duct (on the left side only); the terminal portion of the right lymphatic duct (on the right side only); and the oesophagus and trachea (in the midline). The brachiocephalic veins are formed by the union of the internal jugular and subclavian veins at the junction of the neck and thorax. On each side, the apical (cervical) pleura and the apex of the lung bulge up into the root of the neck. The height to which the apical pleura rises-with reference to the first pair of ribs and costal cartilages-varies in different individuals according to the obliquity of the thoracic inlet. Posteriorly the apical pleura typically reaches the level of the neck of the first rib, and forms a domed roof over each side of the thoracic cavity, strengthened by the suprapleural membrane. Scalenus anterior covers the anterolateral part of the dome of the pleura, and separates it from the subclavian vein. The subclavian artery crosses the dome below its summit and immediately above the vein. The internal thoracic artery descends from the first part of the subclavian artery, passes behind the brachiocephalic vein and, on the right side, is crossed by the phrenic nerve. The costocervical trunk arches backwards from the subclavian artery and crosses the summit of the dome of the lung: its superior intercostal branch descends behind the dome, between the first intercostal nerve laterally and the first thoracic sympathetic ganglion medially. The vagus descends anterior to the first part of the subclavian artery, and on the right side its recurrent laryngeal branch turns around the lower border of the artery.
CERVICAL RIB A small extra rib (cervical rib) may develop in the root of the neck in association with the seventh cervical vertebra. Its presence may result in symptoms associated with compression of adjacent structures, particularly the brachial plexus and subclavian artery. Anatomical variations in the relationship between scalenus anterior and the brachial plexus can also give rise to compression syndromes of the brachial plexus.
© 2008 Elsevier
MUSCLES Superficial and lateral cervical muscles PLATYSMA Platysma is a broad sheet of muscle of varying prominence which arises from the fascia covering the upper parts of pectoralis major and deltoid. Its fibres cross the clavicle and ascend medially in the side of the neck. Anterior fibres interlace across the midline with the fibres of the contralateral muscle, below and behind the symphysis menti. Other fibres attach to the lower border of the mandible or to the lower lip or cross the mandible to attach to skin and subcutaneous tissue of the lower face. page 535 page 536
Figure 31.7 Dissection to show the course of the right vertebral and internal carotid arteries and some of their branches.
Vascular supply Platysma receives its blood supply from the submental branch of the facial artery and the suprascapular artery from the thyrocervical trunk of the subclavian artery. Innervation Platysma is innervated by the cervical branch of the facial nerve which descends on the deep surface of the muscle close to the angle of the mandible. Actions Contraction diminishes the concavity between the jaw and the side of the neck and produces tense oblique ridges in the skin of the neck. Platysma may assist in depressing the mandible, and via its labial and modiolar attachments it can draw down the lower lip and corners of the mouth in expressions of horror or surprise.
STERNOCLEIDOMASTOID Sternocleidomastoid (Fig. 31.10) descends obliquely across the side of the neck and forms a prominent surface landmark, especially when contracted. It is thick and narrow centrally, and broader and thinner at each end. The muscle is attached inferiorly by two heads. The medial or sternal head is rounded and tendinous, arises from the upper part of the anterior surface of the manubrium and sterni and ascends posterolaterally. The lateral or clavicular head, which is
variable in width and contains muscular and fibrous elements, ascends almost vertically from the superior surface of the medial third of the clavicle. The two heads are separated near their attachments by a triangular interval which corresponds to a surface depression, the lesser supraclavicular fossa. As they ascend, the clavicular head spirals behind the sternal head and blends with its deep surface below the middle of the neck, forming a thick, rounded belly. Sternocleidomastoid inserts superiorly by a strong tendon into the lateral surface of the mastoid process from its apex to its superior border, and by a thin aponeurosis into the lateral half of the superior nuchal line. The clavicular fibres are directed mainly to the mastoid process; the sternal fibres are more oblique and superficial, and extend to the occiput. The direction of pull of the two heads is therefore different, and the muscle may be classed as 'cruciate' and slightly 'spiralized'. Relations The superficial surface of sternocleidomastoid is covered by skin and platysma, between which lie the external jugular vein, the great auricular and transverse cervical nerves and the superficial lamina of the deep cervical fascia. Near its insertion the muscle is overlapped by a small part of the parotid gland. The deep surface of the muscle near its origin is related to the sternoclavicular joint and sternohyoid, sternothyroid and omohyoid. The anterior jugular vein crosses deep to it, but superficial to the infrahyoid muscles, immediately above the clavicle. The carotid sheath and the subclavian artery are deep to these muscles. Between omohyoid and the posterior belly of digastric, the anterior part of sternocleidomastoid lies superficial to the common, internal and external carotid arteries, the internal jugular, facial and lingual veins, the deep cervical lymph nodes, the vagus nerve and the rami of the ansa cervicalis. The sternocleidomastoid branch of the superior thyroid artery crosses deep to the muscle at the upper border of omohyoid. The posterior part of sternocleidomastoid is related on its internal surface to splenius capitis, levator scapulae and the scalene muscles, the cervical plexus, the upper part of the brachial plexus, the phrenic nerve and the transverse cervical and suprascapular arteries. The occipital artery crosses deep to the muscle at, or under cover of, the lower border of the posterior belly of digastric. At this point the accessory nerve passes deep to sternocleidomastoid, then pierces and supplies the muscle, before it reappears just above the middle of the posterior border. At its insertion the muscle lies superficial to the mastoid process, splenius capitis, longissimus capitis and the posterior belly of digastric. UPDATE Date Added: 11 October 2005 Publication Services, Inc. Abstract: How cranial could the sternocleidomastoid muscle be split? Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=15750415&query_hl=6 How cranial could the sternocleidomastoid muscle be split? Alagoz MS, Cagri Uysal A, Tuccar E et al: J Craniofac Surg 16(2):201-204, 2005. Vascular supply Sternocleidomastoid receives its blood supply from branches of the occipital and posterior auricular arteries (upper part of muscle), the superior thyroid artery (middle part of muscle), and the suprascapular artery (lower part of muscle). A superiorly based flap can be raised on sternocleidomastoid to include a paddle of skin supplied by perforator vessels. This flap has been used to reconstruct the lips, floor of mouth and inner aspect of the cheeks, however its use has been superseded by microvascular free transfer flaps or by conventional myocutaneous flaps such as the pectoralis major flap. Innervation Sternocleidomastoid is supplied by the spinal part of the accessory nerve. Branches from the ventral rami of the second, third, and sometimes fourth, cervical spinal nerves also enter the muscle. Although these cervical rami were believed to be solely proprioceptive, clinical evidence suggests that some of their fibres are motor. Actions Acting alone, each sternocleidomastoid will tilt the head towards the ipsilateral shoulder, simultaneously rotating the head so as to turn the face towards the opposite side. This movement occurs in an upward, sideways glance. A more common visual movement is a level rotation from side to side, and this probably represents the most frequent use of the sternocleidomastoids. Acting together
from below, the muscles draw the head forwards and so help longi colli to flex the cervical part of the vertebral column, which is a common movement in feeding. The two sternocleidomastoids are also used to raise the head when the body is supine, and when the head is fixed, they help to elevate the thorax in forced inspiration. Branchial cysts and fistulae Branchial cysts usually present in the upper neck in early adulthood as fluctuant swellings at the junction of the upper and middle thirds of the anterior border of sternocleidomastoid. The cyst typically passes backwards and upwards through the carotid bifurcation and ends at the pharyngeal constrictor muscles, a course which brings it into intimate association with the hypoglossal, glossopharyngeal and spinal accessory nerves. Great care must be taken to avoid damage to these nerves during surgical removal of a branchial cyst. Branchial fistulae represent a persistent connection between the second branchial pouch and the cervical sinus. The fistula typically presents as a small pit adjacent to the anterior border of the lower third of sternocleidomastoid, which may weep saliva and become intermittently infected. Excision involves following the tract of the fistula up the neck-often through the carotid bifurcation-and into the distal tonsillar fossa where it opens into the pharynx. page 536 page 537
Figure 31.8 The lower part of the posterior triangle to show the relations of the third part of the right subclavian artery. The clavicle has been removed, but its outline is indicated by a dashed line. In this dissection, the middle trunk of the brachial plexus gives an unusual contribution to the medial cord.
Branchial cysts, sinuses and fistulae are thought to arise from inclusions of salivary gland tissue in lymph nodes: they may also occur around the parotid gland.
Muscles of the anterior triangle of the neck Apart from the superficial neck muscles already described, the anterior triangle contains two of the suprahyoid muscles, namely digastric and stylohyoid, and the four infrahyoid strap muscles. The other suprahyoid muscles, namely mylohyoid and geniohyoid, are described with the floor of the mouth (p. 583).
DIGASTRIC Digastric has two bellies and lies below the mandible, extending from the mastoid process to the chin (Figs 31.3, 31.5). The posterior belly, which is longer than the anterior, is attached in the mastoid notch of the temporal bone, and passes downwards and forwards. The anterior belly is attached to the digastric fossa on the base of the mandible near the midline, and slopes downwards and backwards. The two bellies meet in an intermediate tendon which runs in a fibrous sling attached to the body and greater cornu of the hyoid bone and is sometimes lined by a synovial sheath. The tendon perforates stylohyoid. Variations Digastric may lack the intermediate tendon and is then attached midway along the body of the mandible. The posterior belly may be augmented by a slip from the styloid process or arise wholly from it. The anterior belly may cross the midline and it is not uncommon for it to fuse with mylohyoid. Relations Superficial to digastric are platysma, sternocleidomastoid, splenius capitis, longissimus capitis and stylohyoid, the mastoid process, the retromandibular vein and the parotid and submandibular salivary glands. Mylohyoid is medial to the anterior belly, and hyoglossus, superior oblique and rectus capitis lateralis, the transverse process of the atlas vertebra, the accessory nerve, internal jugular vein, occipital artery, hypoglossal nerve, internal and external carotid, facial and lingual arteries are all medial to the posterior belly. Vascular supply The posterior belly is supplied by the posterior auricular and occipital arteries. The anterior belly of digastric receives its blood supply chiefly from the submental branch of the facial artery. UPDATE Abstract: Vascular anatomy of the digastric muscle
Date Added: 10 April 2006
Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=14704576&query_hl=16&itool=pubmed_docsum The vascular anatomy of the digastric muscle. Alagoz MS, Uysal AC, Tuccar E, Sensoz O: J Craniofac Surg 15:114-117, 2004. Innervation The anterior belly of digastric is supplied by the mylohyoid branch of the inferior alveolar nerve, and the posterior belly is supplied by the facial nerve. The different innervation of the two parts reflects their separate derivations from the mesenchyme of the first and second branchial arches. Actions Digastric depresses the mandible and can elevate the hyoid bone. The posterior bellies are especially active during swallowing and chewing.
STYLOHYOID Stylohyoid arises by a small tendon from the posterior surface of the styloid process, near its base. Passing downwards and forwards, it inserts into the body of the hyoid bone at its junction with the greater cornu (and just above the attachment of the superior belly of omohyoid) (Fig. 31.3). It is perforated near its insertion by the intermediate tendon of digastric. The muscle may be absent or double. It may lie medial to the external carotid artery and may end in the suprahyoid or infrahyoid muscles. Vascular supply Stylohyoid receives its blood supply from branches of the facial, posterior auricular and occipital arteries. page 537 page 538
Figure 31.9 Root of the neck. (From Brash JC 1958 Cunningham's Manual of Practical Anatomy, Vol 3. Head and Neck: Brain. London: Oxford University Press. By permission of Oxford University Press.)
Innervation Stylohyoid is innervated by the stylohyoid branch of the facial nerve, which frequently arises with the digastric branch, and enters the middle part of the muscle. Actions Stylohyoid elevates the hyoid bone and draws it backwards, elongating the floor of the mouth.
STYLOHYOID LIGAMENT The stylohyoid ligament is a fibrous cord extending from the tip of the styloid process to the lesser cornu of the hyoid bone. It gives attachment to the highest fibres of the middle pharyngeal constrictor and is intimately related to the lateral wall of the oropharynx. Below, it is overlapped by hyoglossus. The ligament is derived from the cartilage of the second branchial arch, and may be partially calcified.
Infrahyoid muscles The infrahyoid muscles are organized so that sternohyoid and omohyoid lie superficially and sternothyroid and thyrohyoid lie more deeply. The muscles are involved in movements of the hyoid bone and thyroid cartilage during vocalization, swallowing and mastication and are mainly innervated from the ansa cervicalis. UPDATE Date Added: 11 October 2005 Publication Services, Inc. Abstract: Modification of the infra hyoid musculo-cutaneous flap. Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=15780566&query_hl=5 Modification of the infra hyoid musculo-cutaneous flap. Dolivet G, Gangloff P, Sarini J et al: Eur J Surg Oncol 31(3):294-298, 2005.
STERNOHYOID Sternohyoid (Fig. 31.3) is a thin, narrow strap muscle that arises from the posterior surface of the medial end of the clavicle, the posterior sternoclavicular ligament and the upper posterior aspect of the manubrium sterni. It ascends medially and is attached to the inferior border of the body of the hyoid bone. Inferiorly, there is a considerable gap between the muscle and its contralateral fellow, but the two usually come together in the middle of their course, and are contiguous above this. Sternohyoid may be absent or double, augmented by a clavicular slip (cleidohyoid), or interrupted by a tendinous intersection. Vascular supply Sternohyoid is supplied by branches from the superior thyroid artery. Innervation Sternohyoid is innervated by branches from the ansa cervicalis (C1, 2, 3). Action Sternohyoid depresses the hyoid bone after it has been elevated.
OMOHYOID page 538 page 539
Figure 31.10 Anterior view of the veins of the neck. The head has been extended.
Omohyoid consists of two bellies united at an angle by an intermediate tendon (Fig. 31.3). The inferior belly is a flat, narrow band, which inclines forwards and slightly upwards across the lower part of the neck. It arises from the upper border of the scapula, near the scapular notch, and occasionally from the superior transverse scapular ligament. It then passes behind sternocleidomastoid and ends there in the intermediate tendon. The superior belly begins at the intermediate tendon, passes almost vertically upwards near the lateral border of sternohyoid and is attached to the lower border of the body of the hyoid bone lateral to the insertion of sternohyoid. The length and form of the intermediate tendon varies, although it usually lies adjacent to the internal jugular vein at the level of the arch of the cricoid cartilage. The angulated course of the muscle is maintained by a band of deep cervical fascia, attached below to the clavicle and the first rib, which ensheathes the tendon. A variable amount of skeletal muscle may be present in the fascial band; either belly may be absent or double; the
inferior belly may be attached directly to the clavicle and the superior is sometimes fused with sternohyoid. Vascular supply Omohyoid is supplied by branches from the superior thyroid and lingual arteries. Innervation The superior belly of omohyoid is innervated by branches from the superior ramus of the ansa cervicalis (C1). The inferior belly is innervated from the ansa cervicalis itself (C1, 2 and 3). Actions Omohyoid depresses the hyoid bone after it has been elevated. It has been speculated that the muscle tenses the lower part of the deep cervical fascia in prolonged inspiratory efforts, reducing the tendency for soft parts to be sucked inward.
STERNOTHYROID Sternothyroid (Fig. 31.3) is shorter and wider than sternohyoid, and lies deep and partly medial to it. It arises from the posterior surface of the manubrium sterni inferior to the origin of sternohyoid and from the posterior edge of the cartilage of the first rib. It is attached above to the oblique line on the lamina of the thyroid cartilage, where it delineates the upward extent of the thyroid gland. In the lower part of the neck the muscle is in contact with its contralateral fellow, but the two diverge as they ascend. Vascular supply Sternothyroid is supplied by branches from the superior thyroid and lingual arteries. Innervation Sternothyroid is innervated by branches from the ansa cervicalis (C1, 2 and 3). Action Sternothyroid draws the larynx downwards after it has been elevated by swallowing or vocal movements. In the singing of low notes, this downward traction would be exerted with the hyoid bone relatively fixed.
THYROHYOID Thyrohyoid is a small, quadrilateral muscle that may be regarded as an upward continuation of sternothyroid (Fig. 31.5). It arises from the oblique line on the lamina of the thyroid cartilage, and passes upwards to attach to the lower border of the greater cornu and adjacent part of the body of the hyoid bone. Vascular supply Thyrohyoid is supplied by branches from the superior thyroid and lingual arteries. Innervation Unlike the other infrahyoid muscles, thyrohyoid is not innervated by the ansa cervicalis. In common with geniohyoid, it is supplied by fibres from the first cervical spinal nerve which branch off from the hypoglossal nerve beyond the descendens hypoglossi. Actions Thyrohyoid depresses the hyoid bone. With the hyoid bone stabilized, it pulls the larynx upwards, e.g. when high notes are sung.
Anterior vertebral muscles (Fig. 31.11) The anterior vertebral group of muscles consists of longi colli and capitis, and recti capitis anterior and lateralis, all of which are flexors of the head and neck to varying degrees. Together with the lateral vertebral muscles they form the prevertebral muscle group.
RECTUS CAPITIS ANTERIOR Rectus capitis anterior is a short, flat muscle situated behind the upper part of longus capitis. It arises from the anterior surface of the lateral mass of the atlas and the root of its transverse process, and ascends almost vertically to the inferior surface of the basilar part of the occipital bone immediately anterior to the occipital condyle. Vascular supply Rectus capitis anterior is supplied by branches from the vertebral and ascending
pharyngeal arteries. page 539 page 540
Figure 31.11 The anterior and lateral vertebral muscles. Scalenus anterior and longus capitis have been removed on the right side.
Innervation Rectus capitis anterior is innervated by branches from the loop between the ventral rami of the first and second cervical spinal nerves. Actions Rectus capitis anterior flexes the head at the atlanto-occipital joints.
RECTUS CAPITIS LATERALIS Rectus capitis lateralis is a short, flat muscle that arises from the upper surface of the transverse process of the atlas and inserts into the inferior surface of the jugular process of the occipital bone. In view of its attachments and its relation to the ventral ramus of the first spinal nerve, rectus capitis lateralis is regarded as homologous with the posterior intertransverse muscles. Vascular supply Rectus capitis lateralis is supplied by branches from the vertebral, occipital and ascending pharyngeal arteries. Innervation Rectus capitis lateralis is innervated by branches from the loop between the ventral rami of the first and second cervical spinal nerves. Actions Rectus capitis lateralis flexes the head laterally to the same side.
LONGUS CAPITIS Longus capitis (Fig. 31.6) has a narrow origin from tendinous slips from the anterior tubercles of the transverse processes of the third, fourth, fifth and sixth cervical vertebrae and becomes broad and thick above, where it is inserted into the inferior surface of the basilar part of the occipital bone. Vascular supply Longus capitis is supplied by the ascending pharyngeal, ascending cervical branch of the inferior thyroid and the vertebral arteries. Innervation Longus capitis is innervated by branches from the ventral rami of the first, second and third cervical spinal nerves.
Actions Longus capitis flexes the head.
LONGUS COLLI Longus colli (Fig. 31.6) is applied to the anterior surface of the vertebral column, between the atlas and the third thoracic vertebra. It can be divided into three parts which all arise by tendinous slips. The inferior oblique part is the smallest, running upwards and laterally from the fronts of the bodies of the first two or three thoracic vertebrae to the anterior tubercles of the transverse processes of the fifth and sixth cervical vertebrae. The superior oblique part passes upwards and medially from the anterior tubercles of the transverse processes of the third, fourth and fifth cervical vertebrae to be attached by a narrow tendon to the anterolateral surface of the tubercle on the anterior arch of the atlas. The vertical intermediate part ascends from the fronts of the bodies of the upper three thoracic and lower three cervical vertebrae to the fronts of the bodies of the second, third and fourth cervical vertebrae. Vascular supply Longus colli is supplied by branches from the vertebral, inferior thyroid and ascending pharyngeal arteries. Innervation Longus colli is innervated by branches from the ventral rami of the second, third, fourth, fifth and sixth cervical spinal nerves. Actions Longus colli flexes the neck forwards. In addition, the oblique parts may flex it laterally, and the inferior oblique part rotates it to the opposite side. Its main antagonist is longissimus cervicis.
Lateral vertebral muscles (Figs 31.6, 31.11) The scaleni, anterior, medius and posterior, extend obliquely between the upper two ribs and the cervical transverse processes.
SCALENUS ANTERIOR Scalenus anterior lies at the side of the neck deep (posteromedial) to sternocleidomastoid. Above, it is attached by musculotendinous fascicles to the anterior tubercles of the transverse processes of the third, fourth, fifth and sixth cervical vertebrae. These converge, blend and descend almost vertically, to be attached by a narrow, flat tendon to the scalene tubercle on the inner border of the first rib, and to a ridge on the upper surface of the rib anterior to the groove for the subclavian artery. Relations Scalenus anterior forms an important landmark in the root of the neck, because the phrenic nerve passes above it, the subclavian artery below it, and the brachial plexus lies at its lateral border. The clavicle, subclavius, sternocleidomastoid and omohyoid, lateral part of the carotid sheath, transverse cervical, suprascapular and ascending cervical arteries, subclavian vein, prevertebral fascia and phrenic nerve are all anterior to scalenus anterior. Posteriorly are the suprapleural membrane, pleura, roots of the brachial plexus and the subclavian artery: the latter two separate scalenus anterior from scalenus medius. The proximity of the muscle to the brachial plexus, subclavian artery and vein can give rise to compression syndromes. Below its attachment to the sixth cervical vertebra, the medial border of the muscle is separated from longus colli by an angular interval in which the vertebral artery and vein pass to and from the foramen transversarium of the sixth cervical vertebra. The inferior thyroid artery crosses this interval from the lateral to the medial side near its apex. The sympathetic trunk and its cervicothoracic ganglion are closely related to the posteromedial side of this part of the vertebral artery. On the left side the thoracic duct crosses the triangular interval at the level of the seventh cervical vertebra and usually comes into contact with the medial edge of scalenus anterior. The musculotendinous attachments of scalenus anterior to anterior tubercles are separated from those of longus capitis by the ascending cervical branch of the inferior thyroid artery. Innervation Scalenus anterior is innervated by branches from the ventral rami of the fourth, fifth and sixth cervical spinal nerves. Actions Acting from below, scalenus anterior bends the cervical portion of the vertebral
column forwards and laterally and rotates it towards the opposite side. Acting from above, the muscle helps to elevate the first rib.
SCALENUS MEDIUS page 540 page 541
Scalenus medius, the largest and longest of the scaleni, is attached above to the transverse process of the axis and the front of the posterior tubercles of the transverse processes of the lower five cervical vertebrae. It frequently extends upwards to the transverse process of the atlas. Below it is attached to the upper surface of the first rib between the tubercle of the rib and the groove for the subclavian artery. Relations The anterolateral surface of the muscle is related to sternocleidomastoid. It is crossed anteriorly by the clavicle and omohyoid, and it is separated from scalenus anterior by the subclavian artery and ventral rami of the cervical spinal nerves. Levator scapulae and scalenus posterior lie posterolateral to it. The upper two roots of the nerve to serratus anterior and the dorsal scapular nerve (to the rhomboids) pierce the muscle and appear on its lateral surface. Innervation Scalenus medius is supplied by branches from the ventral rami of the third to eighth cervical spinal nerves. Actions Acting from below, scalenus medius bends the cervical part of the vertebral column to the same side. Acting from above, it helps to raise the first rib. The scalene muscles, particularly scalenus medius, are active during inspiration, even during quiet breathing in the erect attitude.
SCALENUS POSTERIOR Scalenus posterior is the smallest and most deeply situated of the scalene muscles. It passes from the posterior tubercles of the transverse processes of the fourth, fifth, and sixth cervical vertebrae to the outer surface of the second rib, behind the tubercle for serratus anterior, where it is attached by a thin tendon. Scalenus posterior is occasionally blended with scalenus medius. The scalene muscles vary a little in the number of vertebrae to which they are attached, in their degree of separation, and their segmental innervation. Vascular supply All the scalene muscles are chiefly supplied by the ascending cervical branch of the inferior thyroid artery. Scalenus posterior receives an additional supply from the superficial cervical artery. Innervation Scalenus posterior is innervated by branches from the ventral rami of the lower three cervical spinal nerves. Actions When the second rib is fixed, scalenus posterior bends the lower end of the cervical part of the vertebral column to the same side. When its upper attachment is fixed, it helps to elevate the second rib.
SCALENUS MINIMUS Scalenus minimus (pleuralis) is associated with the suprapleural membrane and cervical pleura, and is described in that context (pp. 1065, 742).
SPLENIUS CAPITIS AND CERVICIS Splenius capitis and splenius cervicis are described in detail in Chapter 45.
© 2008 Elsevier
HYOID BONE The U-shaped hyoid bone (Fig. 31.12) is suspended from the tips of the styloid processes by the stylohyoid ligaments. It has a body, two greater and two lesser horns, or cornua. Body The body is irregular, elongated and quadrilateral. Its anterior surface is convex, faces anterosuperiorly, and is crossed by a transverse ridge with a slight downward convexity. A vertical median ridge often bisects the upper part of the body; its presence on the lower part is rare. The posterior surface is smooth, concave, faces posteroinferiorly, and is separated from the epiglottis by the thyrohyoid membrane and loose areolar tissue. There is a bursa between the hyoid bone and the membrane.
Figure 31.12 A, B, The hyoid bone: anterosuperior aspect. B shows positions of muscular attachments.
Geniohyoid is attached to most of the anterior surface of the body, above and below the transverse ridge, although the medial part of hyoglossus invades the lateral geniohyoid area. The lower anterior surface gives attachment to mylohyoid; the line of attachment lies above the sternohyoid medially and omohyoid laterally.
The lowest fibres of the genioglossus, the hyoepiglottic ligament and (most posteriorly) the thyrohyoid membrane are attached to the rounded superior border. Sternohyoid is attached to the inferior border medially and omohyoid laterally. Occasionally the medial fibres of thyrohyoid and of levator glandulae thyroideae, when present, are attached along the inferior border. Greater cornua In early life, the greater cornua are connected to the body by cartilage, but after middle age they are usually united by bone. They project backwards (curving posterolaterally) from the lateral ends of the body. They are horizontally flattened, taper posteriorly, and each ends in a tubercle. When the throat is gripped between finger and thumb above the thyroid cartilage, the greater cornua can be identified and the bone can be moved from side to side. The middle pharyngeal constrictor and, more laterally (i.e. superficially), hyoglossus, are attached along the whole length of the upper surface of each greater cornu. Stylohyoid is attached near the junction of the cornu with the body. The fibrous loop for the digastric tendon is attached lateral and a little posterior to hyoglossus. The thyrohyoid membrane is attached to the medial border and thyrohyoid is attached to the lateral border. The oblique inferior surface is separated from the thyrohyoid membrane by fibroareolar tissue. Lesser cornua The lesser cornua are two small conical projections at the junctions of the body and greater cornua. At its base, each is connected to the body by fibrous tissue and occasionally to the greater cornu by a synovial joint which occasionally becomes ankylosed. The middle pharyngeal constrictors are attached to the posterior and lateral aspects of the lesser cornua The stylohyoid ligaments are attached to their apices and are often partly calcified, and the chondroglossi are attached to the medial aspects of their bases. Ossification The hyoid bone develops from cartilages of the second and third pharyngeal arches, the lesser cornua from the second, the greater cornua from the third and the body from the fused ventral ends of both. Chondrification begins in the fifth fetal week in these elements, and is completed in the third and fourth months. Ossification proceeds from six centres, i.e. a pair for the body and one for each cornu. Ossification begins in the greater cornua towards the end of intrauterine life, in the body shortly before or after birth, and in the lesser cornua around puberty. The greater cornual apices remain cartilaginous until the third decade and epiphyses may occur here. They fuse with the body. Synovial joints between the greater and lesser cornua may be obliterated by ossification in later decades.
© 2008 Elsevier
CERVICAL FASCIA Superficial fascia The superficial cervical fascia is usually a thin lamina covering platysma and is hardly demonstrable as a separate layer. It may, however, contain considerable amounts of adipose tissue, especially in females. Like all superficial fascia it is not a separate stratum, but merely a zone of loose connective tissue between dermis and deep fascia, and is joined to both. page 541 page 542
Deep cervical fascia Deep fascia in the neck is organized into an investing layer, prevertebral fascia, and pretracheal fascia (Fig. 31.13). The carotid sheath is a non-membranous layer of fascia that is conventionally described as part of the deep cervical fascia.
INVESTING LAYER OF DEEP FASCIA The investing layer of deep cervical fascia is continuous behind with the ligamentum nuchae and the periosteum covering the spine of the seventh cervical vertebra. It forms a thin covering for trapezius and continues forwards from the anterior border of this muscle as a loose areolar layer roofing over the posterior triangle of the neck to the posterior border of sternocleidomastoid, where it becomes denser. It divides around sternocleidomastoid, enclosing it, and reunites at the anterior margin as a single sheet, which covers the anterior triangle of the neck and reaches forwards to the midline. Here it meets the corresponding sheet from the opposite side and adheres to the symphysis menti and the body of the hyoid bone. Superiorly, the deep fascia fuses with periosteum along the superior nuchal line of the occipital bone, over the mastoid process and along the entire base of the mandible. Between the angle of the mandible and the anterior edge of sternocleidomastoid it is particularly strong. Between the mandible and the mastoid process it is related to the parotid gland, extending beneath it to become attached to the zygomatic arch. From this region the strong stylomandibular ligament ascends to the styloid process. Inferiorly, the investing layer of deep fascia is attached to the acromion, clavicle and manubrium sterni, fusing with their periostea. A short distance above the manubrium it splits into superficial and deep layers. The superficial layer is attached to the anterior border of the manubrium, the deep layer to its posterior border and to the interclavicular ligament. Between these two layers is a slit-like interval, the suprasternal space. This contains a small amount of areolar tissue, the lower parts of the anterior jugular veins and the jugular venous arch, the sternal heads of the sternocleidomastoid muscles and sometimes a lymph node. Over the lower part of the posterior triangle, between trapezius and sternocleidomastoid, the deep fascia again divides into superficial and deep layers. The superficial layer is attached below to the superior border of the clavicle. The deep layer surrounds the inferior belly of omohyoid and, deep to sternocleidomastoid, the intermediate tendon of omohyoid. The deep layer blends inferiorly with the fascia around subclavius and the periosteum on the posterior surface of both the clavicle and anterior end of the first rib. UPDATE Date Added: 28 February 2006 Publication Services, Inc. Abstract: Does the investing layer of the deep cervical fascia exist? Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=16249670&query_hl=1&itool=pubmed_docsum Does the investing layer of the deep cervical fascia exist? Nash L, Nicholson HD, Zhang M: Anesthesiology 103(5):962-968, 2005.
PRETRACHEAL FASCIA
Figure 31.13 Transverse section through the lower part of the neck at the level of the seventh cervical vertebra, showing the arrangement of the deep cervical fascia, much of which has been coloured blue. (Specimen provided by REM Bowden.)
The pretracheal layer of the deep cervical fascia is very thin. It provides a fascial sheath for the thyroid gland and lies deep to the strap muscles. Superiorly, it attaches to the arch of the cricoid cartilage, and inferiorly it continues into the superior mediastinum along the great vessels to merge with the fibrous pericardium. The pretracheal fascia merges laterally with the investing layer of deep cervical fascia and with the carotid sheath.
PREVERTEBRAL FASCIA The prevertebral fascia covers the anterior vertebral muscles and extends laterally on scalenus anterior, scalenus medius and levator scapulae, forming a fascial floor for the posterior triangle of the neck. As the subclavian artery and the brachial plexus emerge from behind scalenus anterior they carry the prevertebral fascia downwards and laterally behind the clavicle as the axillary sheath. The prevertebral fascia is particularly prominent in front of the vertebral column, where there may be two distinct layers of fascia. Traced laterally, it becomes thin and areolar and is lost as a definite fibrous layer under cover of trapezius. Superiorly the prevertebral fascia is attached to the base of the skull. Inferiorly it descends in front of longus colli into the superior mediastinum, where it blends with the anterior longitudinal ligament. Anteriorly the prevertebral fascia is separated from the pharynx and its covering buccopharyngeal fascia by a loose areolar zone, the retropharyngeal space. Laterally this loose tissue connects the prevertebral fascia to the carotid sheath and the fascia on the deep surface of sternocleidomastoid. All the ventral rami of the cervical nerves are initially behind the prevertebral fascia. The nerves to the rhomboids and serratus anterior and the phrenic nerve retain this position throughout their course in the neck, but the accessory nerve lies superficial to the prevertebral fascia.
CAROTID SHEATH The carotid sheath is a condensation of deep cervical fascia around the common and internal carotid arteries, the internal jugular vein, the vagus nerve and the constituents of the ansa cervicalis. It is thicker around the arteries than the vein, an arrangement that allows the vein to expand. Peripherally the carotid sheath is connected to adjacent fascial layers by loose areolar tissue.
TISSUE SPACES AND THE SPREAD OF INFECTION The fascial layers of the neck define a number of potential tissue 'spaces' above
and below the hyoid bone. In health, the tissues within these spaces are closely applied to each other or are filled with relatively loose connective tissue. However, infection, which often arises superiorly in the region of the infratemporal fossa as a result of dental or tonsillar infections, can alter these relationships. The organisms responsible are often beta-haemolytic streptococci or a variety of anaerobes. Streptococci produce proteolytic enzymes which digest the loose connective tissue and so open up the tissue spaces. Since there are no tissue barriers running horizontally or vertically in the neck, infections which are not treated promptly can spread from the infratemporal fossa down to the mediastinum below, and can even cross the midline through the sublingual and submental spaces. The tissue spaces above the hyoid bone are the submandibular and submental spaces beneath the inferior border of the mandible; the pharyngeal spaces; and the prevertebral space near the base of the skull. These spaces are described on pages 607 and 626. Tissue spaces around the larynx are described on page 640. Tissue spaces below the hyoid bone are the pretracheal and retrovisceral tissue spaces in the visceral compartment of the neck; the prevertebral space in front of the vertebral column; and a space associated with the carotid sheath. Pretracheal space The pretracheal tissue space lies behind the pretracheal fascia and the infrahyoid strap muscles, and in front of the anterior wall of the oesophagus, and therefore immediately surrounds the trachea. It is bounded superiorly by the attachments of the infrahyoid muscles to the thyroid cartilage of the larynx. Inferiorly, it extends down into the anterior portion of the superior mediastinum. Infection usually spreads into the pretracheal space either by perforating the anterior wall of the oesophagus or from the retrovisceral space. Retrovisceral space
page 542 page 543
The retrovisceral space is continuous superiorly with the retropharyngeal space. It is situated between the posterior wall of the oesophagus and the prevertebral fascia. Inferiorly, the retrovisceral space extends into the superior mediastinum. Should the prevertebral fascia merge with the connective tissue on the posterior surface of the oesophagus-usually at the level of the fourth thoracic vertebra-the retrovisceral space then has a distinct inferior boundary. Prevertebral space The prevertebral tissue space has been variously described as the potential space lying between the prevertebral fascia and the vertebral column, and as the space between the two layers of the prevertebral fascia. Infection usually spreads into the space via its fascial walls from the retrovisceral area because it is closed superiorly and laterally. Inferiorly, it extends into the posterior mediastinum. Carotid space The carotid sheath is a layer of loose connective tissue demarcated by adjacent portions of the investing layer of deep cervical fascia, the pretracheal fascia, and the prevertebral fascia. Nevertheless, it delineates a potential space into which infections from the visceral spaces may track. Infections around the carotid sheath may be restricted because superiorly (near the hyoid bone) and inferiorly (near the root of the neck) the connective tissues adhere to the vessels. Cellulitis in the neck The main cause of cellulitis of the neck is infection arising from the region of the mandibular molar teeth. Several fascial spaces are accessible from this area, and several factors contribute to the spread of infection. Thus, the apices of the second and, more especially, the third, mandibular molar teeth are often close to the lingual surface of the mandible. The apices of the roots of the third mandibular
molars are usually-and the second molars are often-below the attachment of mylohyoid on the inner aspect of the mandible and so drain directly into the submandibular tissue space. The posterior free border of mylohyoid is close to the sockets of the third mandibular molars, and at this point, the floor of the mouth consists only of mucous membrane covering part of the submandibular salivary gland. Any virulent periapical infection of the mandibular third molar teeth may therefore penetrate the lingual plate of the mandible and is then at the entrance to several fascial spaces, namely, the submandibular and sublingual spaces anteriorly, and the parapharyngeal and pterygoid spaces posteriorly. Infection in this area may also spread from an acute pericoronitis, particularly when the deeper tissues are opened to infection by extraction of the tooth during the acute phase. In general, cellulitis around the jaw is only likely to develop when the tissues are infected by virulent and invasive organisms at a point where there is access to the fascial spaces. As the predisposing causes do not often coincide, cellulitis is uncommon. Cellulitis in the region of the maxilla is even more uncommon, but fascial space infections may develop in various sites as the result of infected local anaesthetic needles. It is evident that there are no barriers running vertically with respect to the tissue spaces in the neck. Thus, infection entering in this site can rapidly spread more or less unhindered down the neck and may enter the mediastinum.
© 2008 Elsevier
VASCULAR SUPPLY AND LYMPHATIC DRAINAGE Arteries of the neck The common carotid, internal carotid, and external carotid arteries provide the major source of blood to the head and neck. Additional arteries arise from branches of the subclavian artery, particularly the vertebral artery. The common, internal and external carotid arteries and accompanying veins and nerves, all lie in a cleft that is bound posteriorly by the transverse processes of cervical vertebrae and attached muscles, medially by the trachea, oesophagus, thyroid gland, larynx and pharyngeal constrictors, and anterolaterally by sternocleidomastoid and, at different levels, omohyoid, sternohyoid, sternothyroid, digastric and stylohyoid muscles. The common and internal carotid arteries lie within the carotid sheath, accompanied by the internal jugular vein and the vagus nerve.
COMMON CAROTID ARTERY (Figs 31.14, 31.15, 31.16) The common carotid arteries differ on the right and left sides with respect to their origins. On the right, the common carotid arises from the brachiocephalic artery as it passes behind the sternoclavicular joint. On the left, the common carotid artery comes directly from the arch of the aorta in the superior mediastinum. The right common carotid has, therefore, only a cervical part whereas the left common carotid has cervical and thoracic parts. Following a similar course on both sides, the common carotid artery ascends, diverging laterally from behind the sternoclavicular joint to the level of the upper border of the thyroid cartilage of the larynx (C3-4 junction), where it divides into external and internal carotid arteries. This bifurcation can sometimes be at a higher level. The artery may be compressed against the prominent transverse process of the sixth cervical vertebra (Chassaignac's tubercle), and above this level it is superficial and its pulsation can be easily felt. Relations In the lower part of the neck the common carotid arteries are separated by a narrow gap which contains the trachea. Above this, the arteries are separated by the thyroid gland, larynx and pharynx. Each artery is contained within the carotid sheath of deep cervical fascia, which also encloses the internal jugular vein and vagus nerve. The vein lies lateral to the artery, and the nerve lies between them and posterior to both. The artery is crossed anterolaterally, at the level of the cricoid cartilage, by the intermediate tendon-sometimes the superior belly-of omohyoid. Below omohyoid it is sited deeply, covered by skin, superficial fascia, platysma, deep cervical fascia, and sternocleidomastoid, sternohyoid and sternothyroid. Above omohyoid it is more superficial, covered merely by skin, superficial fascia, platysma, deep cervical fascia and the medial margin of sternocleidomastoid, and it is crossed obliquely from its medial to lateral side by the sternocleidomastoid branch of the superior thyroid artery. The superior root of the ansa cervicalis, joined by its inferior root from the second and third cervical spinal nerves, lies anterior to, or embedded within, the carotid sheath as it crosses it obliquely. The superior thyroid vein usually crosses near the upper border of the thyroid cartilage, and the middle thyroid vein crosses a little below the level of the cricoid cartilage. The anterior jugular vein crosses the common carotid artery above the clavicle, separated from it by sternohyoid and sternothyroid. Posterior to the carotid sheath are the transverse processes of the fourth to sixth cervical vertebrae, to which are attached longus colli, longus capitis and tendinous slips of scalenus anterior. The sympathetic trunk and ascending cervical branch of the inferior thyroid artery lie between the common carotid artery and the muscles. Below the level of the sixth cervical vertebra the artery is in an angle between scalenus anterior and longus colli, anterior to the vertebral vessels, inferior thyroid and subclavian arteries, sympathetic trunk and, on the left, the thoracic duct. The oesophagus, trachea, inferior thyroid artery and recurrent laryngeal nerve and, at a higher level, the larynx and pharynx are medial to the sheath and its contents. The thyroid gland overlaps it anteromedially. The internal jugular vein lies lateral, and, in the lower neck also anterior, to the artery, while the vagus nerve lies posterolaterally in the angle between artery and vein. On the right side, low in the neck, the recurrent laryngeal nerve crosses obliquely behind the artery. The right internal jugular vein diverges from it below but the left vein approaches and often overlaps its artery. In c.12% of cases the right common carotid artery arises above the level of the sternoclavicular joint, or it may be a separate branch from the aorta. The left common carotid artery varies in origin more than the right and may arise with the brachiocephalic artery. Division of the common carotid may occur higher, near the level of the hyoid bone, or, more rarely, at a lower level alongside the larynx. Very rarely it ascends without division, so that either the external or internal carotid is
absent, or it may be replaced by separate external and internal carotid arteries which arise directly from the aorta, on one side, or bilaterally. Although the common carotid artery usually has no branches, it may occasionally give rise to the vertebral, superior thyroid, superior laryngeal, ascending pharyngeal, inferior thyroid or occipital arteries.
EXTERNAL CAROTID ARTERY page 543 page 544
Figure 31.14 Dissection of the right side of the neck, showing the carotid and subclavian arteries and their branches. The parotid and submandibular glands have been removed together with the lower part of the internal jugular vein, most of the sternocleidomastoid and the upper parts of stylohyoid and the posterior belly of digastric.
The external carotid artery (Figs 31.7, 31.14, 31.15, 31.16) begins lateral to the upper border of the thyroid cartilage, level with the intervertebral disc between the third and fourth cervical vertebrae. A little curved and with a gentle spiral, it first ascends slightly forwards and then inclines backwards and a little laterally, to pass midway between the tip of the mastoid process and the angle of the mandible. Here, in the substance of the parotid gland behind the neck of the mandible, it divides into its terminal branches, the superficial temporal and maxillary arteries. As it ascends, it gives off several large branches, and diminishes rapidly in calibre. In children the external carotid is smaller than the internal carotid, but in adults the two are of almost equal size. At its origin, it is in the carotid triangle and lies anteromedial to the internal carotid artery. It later becomes anterior, then lateral, to the internal carotid as it ascends. At mandibular levels the styloid process and its attached structures intervene between the vessels: the internal carotid is deep, and the external carotid superficial, to the styloid process. A fingertip placed in the carotid triangle perceives a powerful arterial pulsation, which represents the termination of the common carotid, the origins of external and internal carotids and the stems of the initial branches of the external carotid. Relations The skin and superficial fascia, the loop between the cervical branch of the facial nerve and the transverse cutaneous nerve of the neck, the deep cervical fascia and the anterior margin of sternocleidomastoid all lie superficial to the external carotid artery in the carotid triangle. The artery is crossed by the hypoglossal nerve and its vena comitans and by the lingual, facial and, sometimes, the superior thyroid veins. Leaving the carotid triangle, the external carotid artery is
crossed by the posterior belly of digastric and by stylohyoid, and ascends between these muscles and the posteromedial surface of the parotid gland, which it next enters. Within the parotid, the artery lies medial to the facial nerve and the junction of the superficial temporal and maxillary veins. The pharyngeal wall, superior laryngeal nerve and ascending pharyngeal artery are the initial medial relations of the artery. At a higher level, it is separated from the internal carotid artery by the styloid process, styloglossus and stylopharyngeus, glossopharyngeal nerve, pharyngeal branch of vagus nerve and part of the parotid gland. The artery is equally likely to lie medial to the parotid gland, or within it. The external carotid artery has eight named branches distributed to the head and neck. The superior thyroid, lingual and facial arteries arise from its anterior surface, the occipital and posterior auricular arteries arise from its posterior surface and the ascending pharyngeal artery arises from its medial surface. The maxillary and superficial temporal arteries are its terminal branches within the parotid gland.
SUPERIOR THYROID ARTERY (Figs 31.7, 31.14) The superior thyroid artery is the first branch of the external carotid artery, and arises from the anterior surface of the external carotid just below the level of the greater cornu of the hyoid bone. It descends along the lateral border of thyrohyoid to reach the apex of the lobe of the thyroid gland. Lying medially are the inferior constrictor muscle and the external laryngeal nerve: the nerve is often posteromedial, and therefore at risk when the artery is being ligatured. Occasionally it may issue directly from the common carotid. Branches
The superior thyroid artery supplies the thyroid gland and some adjacent skin. Glandular branches are: anterior, which runs along the medial side of the upper pole of the lateral lobe to supply mainly the anterior surface; a branch which crosses above the isthmus to anastomose with its fellow of the opposite side; and posterior, which descends on the posterior border to supply the medial and lateral surfaces and anastomoses with the inferior thyroid artery. Sometimes a lateral branch supplies the lateral surface. The artery also has the following named branches: infrahyoid, superior laryngeal, sternocleidomastoid and cricothyroid. Infrahyoid artery
The infrahyoid artery is a small branch which runs along the lower border of the hyoid bone deep to thyrohyoid and anastomoses with its fellow of the opposite side to supply the infrahyoid strap muscles. page 544 page 545
Figure 31.15 Dissection of the lower part of the front of the neck and the superior mediastinum. The manubrium sterni and the sternal ends of the clavicles and the first costal cartilages have been removed and the pleural sac and lung have been retracted on each side. In this specimen, each superior thyroid artery arose from the
common carotid artery.
Figure 31.16 The structures crossing the internal jugular vein and carotid arteries and those intervening between the external and internal carotid arteries.
Superior laryngeal artery (Fig. 31.10)
The superior laryngeal artery accompanies the internal laryngeal nerve. Deep to thyrohyoid it pierces the lower part of the thyrohyoid membrane to supply the tissues of the upper part of the larynx. It anastomoses with its fellow of the opposite side and with the inferior laryngeal branch of the inferior thyroid artery. Sternocleidomastoid artery
The sternocleidomastoid artery descends laterally across the carotid sheath and supplies the middle region of sternocleidomastoid. Like the parent artery itself, it may arise directly from the external carotid artery. Cricothyroid artery
The cricothyroid artery crosses high on the anterior cricothyroid ligament, anastomoses with its fellow of the opposite side and supplies cricothyroid.
ASCENDING PHARYNGEAL ARTERY The ascending pharyngeal artery is the smallest branch of the external carotid. It is a long, slender vessel which arises from the medial (deep) surface of the external carotid artery near the origin of that artery. It ascends between the internal carotid artery and the pharynx to the base of the cranium. The ascending pharyngeal artery is crossed by styloglossus and stylopharyngeus, and longus capitis lies posterior to it. It gives off numerous small branches to supply longus capitis and longus colli, the sympathetic trunk, the hypoglossal, glossopharyngeal and vagus nerves and some of the cervical lymph nodes. It anastomoses with the ascending palatine branch of the facial artery and the ascending cervical branch of the vertebral artery. Its named branches are the pharyngeal, inferior tympanic and meningeal arteries. Pharyngeal artery page 545 page 546
The pharyngeal artery gives off three or four branches to supply the constrictor muscles of the pharynx and stylopharyngeus. A variable ramus supplies the palate, and may replace the ascending palatine branch of the facial artery. The artery descends forwards between the superior border of the superior constrictor and levator veli palatini to the soft palate, and also supplies a branch to the palatine tonsil and the pharyngotympanic tube. Inferior tympanic artery
The inferior tympanic artery is a small branch which traverses the temporal canaliculus with the tympanic branch of the glossopharyngeal nerve and supplies the medial wall of the tympanic cavity.
Meningeal branches
The meningeal branches are small vessels which supply the nerves that traverse the foramen lacerum, jugular foramen and hypoglossal canal, and the associated dura mater and adjoining bone. One branch, the posterior meningeal artery, reaches the cerebellar fossa via the jugular foramen, and is usually regarded as the terminal branch of the ascending pharyngeal artery.
LINGUAL ARTERY (Figs 31.14, 31.16) The lingual artery provides the chief blood supply to the tongue and the floor of the mouth. It arises anteromedially from the external carotid artery opposite the tip of the greater cornu of the hyoid bone, between the superior thyroid and facial arteries. It often arises with the facial or, less often, with the superior thyroid artery. It may be replaced by a ramus of the maxillary artery. Ascending medially at first, it loops down and forwards, passes medial to the posterior border of hyoglossus and then runs horizontally forwards deep to it. The lingual artery next ascends again almost vertically, and courses sinuously forwards on the inferior surface of the tongue as far as its tip. The further course of the lingual artery is described on page 587. Relations Its relationship to hyoglossus naturally divides the lingual artery into descriptive 'thirds'. In its first part the lingual artery is in the carotid triangle. Skin, fascia and platysma are superficial to it, while the middle pharyngeal constrictor muscle is medial. The artery ascends a little medially, then descends to the level of the hyoid bone, and the loop so formed is crossed externally by the hypoglossal nerve. The second part passes along the upper border of the hyoid bone, deep to hyoglossus, the tendons of digastric and stylohyoid, the lower part of the submandibular gland and the posterior part of mylohyoid. Hyoglossus separates it from the hypoglossal nerve and its vena comitans. Here its medial aspect adjoins the middle constrictor muscle and it crosses the stylohyoid ligament accompanied by lingual veins. The third part is the arteria profunda linguae which turns upward near the anterior border of hyoglossus and then passes forwards close to the inferior lingual surface near the frenulum, accompanied by the lingual nerve. Genioglossus is a medial relation, and the inferior longitudinal muscle of the tongue lies lateral to it below the lingual mucous membrane. Near the tip of the tongue the lingual artery anastomoses with its fellow of the opposite side. Its named branches are the suprahyoid, dorsal lingual and sublingual arteries.
SUPRAHYOID ARTERY The suprahyoid artery is a small branch which runs along the upper border of the hyoid bone to anastomose with the contralateral artery. It supplies adjacent structures.
DORSAL LINGUAL ARTERIES The dorsal lingual arteries are described on page 587.
SUBLINGUAL ARTERY The sublingual artery is described on page 587.
FACIAL ARTERY The facial artery (Figs 31.7, 31.14, 29.7, 29.12) arises anteriorly from the external carotid in the carotid triangle, above the lingual artery and immediately above the greater cornu of the hyoid bone. In the neck, at its origin, it is covered only by the skin, platysma, fasciae and often by the hypoglossal nerve. It runs up and forwards, deep to digastric and stylohyoid. At first on the middle pharyngeal constrictor, it may reach the lateral surface of styloglossus, separated there from the palatine tonsil only by this muscle and the lingual fibres of the superior constrictor. Medial to the mandibular ramus it arches upwards and grooves the posterior aspect of the submandibular gland. It then turns down and descends to the lower border of the mandible in a lateral groove on the submandibular gland, between the gland and medial pterygoid. Reaching the surface of the mandible, the facial artery curves round its inferior border, anterior to masseter, to enter the face: its further course is described on page 509. The artery is very sinuous throughout its extent. In the neck this may be so that the artery is able to adapt to the movements of the pharynx during deglutition, and similarly on the face, so that the artery can adapt to movements of the mandible, lips and cheeks. Facial artery pulsation is most palpable where the artery crosses the mandibular base, and again near the corner of the mouth. Its branches in the neck are the ascending palatine, tonsillar, submental and glandular arteries. Ascending palatine artery (Fig. 31.7)
The ascending palatine artery arises close to the origin of the facial artery. It passes up between styloglossus and stylopharyngeus to reach the side of the pharynx, along which it ascends between the superior constrictor of the pharynx and medial pterygoid towards the cranial base. It bifurcates near levator veli palatini. One branch follows this muscle, winding over the upper border of the
superior constrictor of the pharynx to supply the soft palate and to anastomose with its fellow of the opposite side and the greater palatine branch of the maxillary artery. The other branch pierces the superior constrictor muscle to supply the tonsil and pharyngotympanic tube and to anastomose with the tonsillar and ascending pharyngeal arteries. Tonsillar artery
The tonsillar artery provides the main blood supply to the palatine tonsil, and may sometimes arise from the ascending palatine artery. It ascends between medial pterygoid and styloglossus, and penetrates the superior constrictor of the pharynx at the upper border of styloglossus to ramify in the tonsil and the musculature of the posterior part of the tongue. Submental artery (Figs 31.14, 29.7)
The submental artery is the largest cervical branch of the facial artery. It arises as the facial artery separates from the submandibular gland and turns forwards on mylohyoid below the mandible. It supplies the overlying skin and muscles, and anastomoses with a sublingual branch of the lingual and mylohyoid branch of the inferior alveolar artery. At the chin it ascends over the mandible, and divides into superficial and deep branches, which anastomose with the inferior labial and mental arteries to supply the chin and lower lip. Glandular branches
Three or four large vessels supply the submandibular salivary gland and associated lymph nodes, adjacent muscles and skin.
OCCIPITAL ARTERY (Figs 31.14, 31.16, 31.17) The occipital artery arises posteriorly from the external carotid artery, c.2 cm from its origin. At its origin, the artery is crossed superficially by the hypoglossal nerve, which winds round it from behind. The artery next passes backwards, up and deep to the posterior belly of digastric, and crosses the internal carotid artery, internal jugular vein, hypoglossal, vagus and accessory nerves. Between the transverse process of the atlas and the mastoid process, the occipital artery reaches the lateral border of rectus capitis lateralis. It then runs in the occipital groove of the temporal bone, medial to the mastoid process and attachments of sternocleidomastoid, splenius capitis, longissimus capitis and digastric, and lies successively on rectus capitis lateralis, obliquus superior and semispinalis capitis. Finally, accompanied by the greater occipital nerve, it turns upwards to pierce the investing layer of the deep cervical fascia connecting the cranial attachments of trapezius and sternocleidomastoid, and ascends tortuously in the dense superficial fascia of the scalp where it divides into many branches. The occipital artery has two main branches (upper and lower) to the upper part of sternocleidomastoid in the neck. The lower branch arises near the origin of the occipital artery, and may sometimes arise directly from the external carotid artery. It descends backwards over the hypoglossal nerve and internal jugular vein, enters sternocleidomastoid and anastomoses with the sternocleidomastoid branch of the superior thyroid artery. The upper branch arises as the occipital artery crosses the accessory nerve, and runs down and backwards superficial to the internal jugular vein. It enters the deep surface of sternocleidomastoid with the accessory nerve. UPDATE Abstract: The microanatomy of the occipital artery
Date Added: 07 August 2006
Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? db=pubmed&cmd=Retrieve&dopt=AbstractPlus&list_uids=16543868&query_hl=3&itool=pubmed_docsum The microanatomy of the occipital artery. Alvernia JE, Fraser K, Lanzino G: Neurosurgery 58(1 Suppl):ONS114-122, discussion ONS114-122, 2006. page 546 page 547
Figure 31.17 Dissection to show the course of the right occipital artery. The upper and lower sternocleidomastoid branches of the artery have been transected and are not labelled.
POSTERIOR AURICULAR ARTERY (Figs 31.14, 31.16) The posterior auricular artery is a small vessel which branches posteriorly from the external carotid just above digastric and stylohyoid. It ascends between the parotid gland and the styloid process to the groove between the auricular cartilage and mastoid process, and divides into auricular and occipital branches which are described with the face on page 510, 511. In the neck, it provides branches to supply digastric, stylohyoid, sternocleidomastoid and the parotid gland. It also gives origin to the stylomastoid artery-described as an indirect branch of the posterior auricular artery in about a third of subjects-which enters the stylomastoid foramen to supply the facial nerve, tympanic cavity, mastoid antrum air cells and semicircular canals. In the young, its posterior tympanic ramus forms a circular anastomosis with the anterior tympanic branch of the maxillary artery.
INTERNAL CAROTID ARTERY The internal carotid artery supplies most of the ipsilateral cerebral hemisphere, eye and accessory organs, forehead and, in part, the nose (Figs 31.14, 31.16). From its origin at the carotid bifurcation (where it usually has a carotid sinus), it ascends in front of the transverse processes of the upper three cervical vertebrae to the inferior aperture of the carotid canal in the petrous part of the temporal bone. Here it enters the cranial cavity and turns anteriorly through the cavernous sinus in the carotid groove on the side of the body of the sphenoid bone. It terminates below the anterior perforated substance by division into the anterior and middle cerebral arteries. It may be divided conveniently into cervical, petrous, cavernous and cerebral parts. Relations The internal carotid artery is initially superficial in the carotid triangle, then passes deeper, medial to the posterior belly of digastric. Except near the skull, the internal jugular vein and vagus nerve are lateral to it within the carotid sheath. The external carotid artery is first anteromedial, but then curves back to lie superficial. Posteriorly the internal carotid adjoins longus capitis, and the superior cervical sympathetic ganglion lies between them. The superior laryngeal nerve crosses obliquely behind it. The pharyngeal wall lies medial to the artery, which is separated by fat and pharyngeal veins from the ascending pharyngeal artery and superior laryngeal nerve. Anterolaterally the internal carotid artery is covered by sternocleidomastoid. Below the posterior belly of digastric, the hypoglossal nerve and superior root of the ansa cervicalis and the lingual and facial veins are superficial to the artery. At the level of the digastric, the internal carotid is crossed by stylohyoid and the occipital and posterior auricular arteries. Above the digastric it is separated from the external carotid artery by the styloid process, styloglossus and stylopharyngeus, the glossopharyngeal nerve and the pharyngeal branch of the vagus, and the deeper part of the parotid gland (Fig. 31.16). At the base of the skull the glossopharyngeal, vagus, accessory and hypoglossal nerves lies between the internal carotid artery and the internal jugular vein, which here has become posterior. The length of the artery varies with the length of the neck and the point of the carotid bifurcation. It may arise from the aortic arch and then lies medial to the external carotid as far as the larynx, where it crosses behind it. The cervical portion is normally straight but may be very tortuous, in which case it lies closer to the pharynx than usual and very near the tonsil. Its absence has also been recorded. The cervical portion of the internal carotid artery has no branches.
CAROTID SINUS AND CAROTID BODY
The common carotid artery shows two specialized organs near its bifurcation, the carotid sinus and the carotid body. They relay information concerning the pressure and chemical composition of the arterial blood respectively, and are innervated principally by carotid branch(es) of the glossopharyngeal nerve, with small contributions from the cervical sympathetic trunk and the vagus nerve. The carotid sinus usually appears as a dilation of the lower end of the internal carotid, and functions as a baroreceptor. The carotid body is a reddish-brown, oval structure, 5-7 mm in height and 2.5-4 mm in width. It lies either posterior to the carotid bifurcation or between its branches, and is attached to, or sometimes partly embedded in, their adventitia. Occasionally it takes the form of a group of separate nodules. Aberrant miniature carotid bodies, microstructurally similar but with diameters of 600 µm or less, may appear in the adventitia and adipose tissue near the carotid sinus. The carotid body is surrounded by a fibrous capsule from which septa divide the enclosed tissue into lobules. Each lobule contains glomus (Type I) cells which are separated from an extensive network of fenestrated sinusoids by sustentacular (type II) cells (Fig. 31.18). Glomus cells store a number of peptides, particularly enkephalins, bombesin and neurotensin, and amines including dopamine, serotonin, adrenaline (epinephrine ) and noradrenaline (norepinephrine), and are therefore regarded as paraneurones. Unmyelinated axons lie in a collagenous matrix between the sustentacular cells and the sinusoidal endothelium, and many synapse on the glomus cells. They are visceral afferents which travel in the carotid sinus nerve to join the glossopharyngeal nerve. Preganglionic sympathetic axons and fibres from the carotid sinus synapse on parasympathetic and sympathetic ganglion cells, which lie either in isolation or in small groups near the surface of each carotid body. Postganglionic axons travel to local blood vessels: the parasympathetic efferent fibres are probably vasodilatory and the sympathetic ones are vasoconstrictor. The carotid body receives a rich blood supply from branches of the adjacent external carotid artery, which is consistent with its role as an arterial chemoreceptor. When stimulated by hypoxia, hypercapnia or increased hydrogen ion concentration (low pH) in the blood flowing through it, it elicits reflex increases in the rate and volume of ventilation via connections with brain stem respiratory centres. The bodies are most prominent in children and normally involute in older age, when they are infiltrated by lymphocytes and fibrous tissue. Individuals with chronic hypoxia, or who live at high altitude or suffer from lung disease, may have enlarged carotid bodies as a result of hyperplasia. Other small bodies, resembling carotid bodies, and also considered to be chemoreceptors, occur near the arteries of the fourth and sixth pharyngeal arches and hence are found near the aortic arch, ligamentum arteriosum and right subclavian artery, and are supplied by the vagus nerve.
SUBCLAVIAN ARTERY (Fig. 31.19) The right subclavian artery arises from the brachiocephalic trunk, the left from the aortic arch. For description, each is divided into a first part, from its origin to the medial border of scalenus anterior, a second part behind this muscle and a third part from the lateral margin of scalenus anterior to the outer border of the first rib, where the artery becomes the axillary artery. Each subclavian artery arches over the cervical pleura and pulmonary apex. Their first parts differ, whereas the second and third parts are almost identical. First part of right subclavian artery page 547 page 548
Figure 31.18 The cellular, neural and vascular architecture of the carotid body. Functional pathways are indicated.
The right subclavian artery branches from the brachiocephalic trunk behind the upper border of the right sternoclavicular joint, and passes superolaterally to the medial margin of scalenus anterior (Figs 31.7, 60.27). It ascends c.2 cm above the clavicle but this varies. Relations
page 548 page 549
Figure 31.19 The subclavian arteries and their branches. (From Brash JC 1979 Cunningham's Manual of Practical Anatomy, Vol 3. Head, Neck and Brain, 14th edn. London: Oxford University Press. By permission of Oxford University Press.)
The artery is deep to the skin, superficial fascia, platysma, supraclavicular nerves, deep fascia, clavicular attachment of sternocleidomastoid, sternohyoid and sternothyroid. It is at first behind the origin of the right common carotid artery; more laterally it is crossed by the vagus nerve, the cardiac branches of the vagus and the sympathetic chain and by internal jugular and vertebral veins; the subclavian sympathetic loop encircles it. The anterior jugular vein diverges laterally in front of it, separated by sternohyoid and sternothyroid. Below and behind the artery are the pleura and pulmonary apex: they are separated from the artery by the suprapleural membrane, the ansa subclavia, a small accessory vertebral vein and the right recurrent laryngeal nerve which winds round the lower and posterior part of the vessel. First part of left subclavian artery
The first part of the left subclavian artery springs from the aortic arch, behind the left common carotid, level with the disc between the third and fourth thoracic vertebrae. It ascends into the neck, then arches laterally to the medial border of scalenus anterior. Relations In the neck, near the medial border of scalenus anterior, the artery is crossed anteriorly by the left phrenic nerve and the termination of the thoracic duct. Otherwise anterior relations are the same as those of the first part of the right subclavian artery. Posteriorly and inferiorly, the relations of both vessels are identical but the left recurrent laryngeal nerve, medial to the left subclavian artery in the thorax, is not directly related to its cervical part. Second part of subclavian artery
The second part of the subclavian artery lies behind scalenus anterior; it is short and the highest part of the vessel (Figs 31.8, 31.14). Relations The skin, superficial fascia, platysma, deep cervical fascia, sternocleidomastoid and scalenus anterior are anterior. The right phrenic nerve is often described as being separated from the second part of the subclavian artery by scalenus anterior, whereas it crosses the first part of the left subclavian artery. However, both nerves may sometimes lie anterior to the muscle. The suprapleural membrane, pleura and lung and the lower trunk of the brachial plexus are posteroinferior; the upper and middle trunks of the plexus are superior; the subclavian vein is anteroinferior, separated by scalenus anterior (Fig. 31.8). Third part of subclavian artery
The third part of the subclavian artery descends laterally from the lateral margin of scalenus anterior to the outer border of the first rib, where it becomes the axillary artery. It is the most superficial part of the artery and lies partly in the supraclavicular triangle, where its pulsations may be felt and it may be compressed. The third part of the subclavian artery is the most accessible segment of the artery. Since the line of the posterior border of sternocleidomastoid approximates to the (deeper) lateral border of scalenus
anterior, the artery can be felt in the anteroinferior angle of the posterior triangle. It can only be effectively compressed against the first rib: with the shoulder depressed, pressure is exerted down, back and medially in the angle between sternocleidomastoid and the clavicle. The palpable trunks of the brachial plexus may be injected with local anaesthetic allowing major surgical procedures to the arm. Relations The skin, superficial fascia, platysma, supraclavicular nerves and deep cervical fascia are anterior (Figs 31.8, 60.4). The external jugular vein crosses its medial end and here receives the suprascapular, transverse cervical and anterior jugular veins, which collectively often form a venous plexus. The nerve to subclavius descends between the veins and the artery; the latter is terminally behind the clavicle and subclavius, where it is crossed by the suprascapular vessels. The subclavian vein is anteroinferior and the lower trunk of the brachial plexus is posteroinferior, between the subclavian artery and the scalenus medius (and on the first rib). The upper and middle trunks of the brachial plexus (which are palpable here) and the inferior belly of omohyoid are superolateral. The first rib is inferior. The right subclavian artery may arise above or below sternoclavicular level; it may be a separate aortic branch and be the first or last branch of the arch. When it is the first branch, it is in the position of a brachiocephalic trunk. When it is the last branch, it arises from the left end of the arch, and ascends obliquely to the right behind the trachea, oesophagus and right common carotid to the first rib. When this occurs, the right recurrent laryngeal nerve hooks round the common carotid artery. Sometimes, when the right subclavian artery is the last aortic branch, it passes between the trachea and oesophagus. It may perforate scalenus anterior, and very rarely may pass anterior to it. Sometimes the subclavian vein accompanies the artery behind scalenus anterior. The artery may ascend as high as 4 cm above the clavicle or it may reach only its upper border. The left subclavian artery is occasionally combined at its origin with the left common carotid artery
VERTEBRAL ARTERY (Figs 31.20, 31.24, 31.25) The vertebral artery arises from the superoposterior aspect of the first part of the subclavian artery. It passes through the foramina in the transverse processes of all of the cervical vertebrae except the seventh, curves medially behind the lateral mass of the atlas and enters the cranium via the foramen magnum. At the lower pontine border it joins its fellow to form the basilar artery. Occasionally it may enter the cervical vertebral column via the fourth, fifth or seventh cervical vertebra (Figs 31.7, 63.5) Relations The first part passes back and upwards between longus colli and scalenus anterior, behind the common carotid artery and the vertebral vein. It is crossed by the inferior thyroid artery, and by the thoracic duct on the left side and the right lymphatic duct on the right side. The seventh cervical transverse process, the inferior cervical ganglion and ventral rami of the seventh and eighth cervical spinal nerves lie posterior to the artery. The second part ascends through the transverse foramina of the remaining cervical vertebrae, accompanied by a large branch from the inferior cervical ganglion and a plexus of veins which form the vertebral vein low in the neck. It lies anterior to the ventral rami of the cervical spinal nerves (C.2-C.6), and ascends almost vertically to pass through the transverse process of the axis, where it turns laterally to gain access to the transverse foramen of the atlas. The third part issues medial to rectus capitis lateralis, and curves backwards and medially behind the lateral mass of the atlas, with the first cervical ventral spinal ramus lying on its medial side. In this position it lies in a groove on the upper surface of the posterior arch of the atlas, and it enters the vertebral canal below the inferior border of the posterior atlantooccipital membrane. This part of the artery, covered by semispinalis capitis, lies in the suboccipital triangle. The first cervical dorsal spinal ramus separates the artery from the posterior arch. The fourth part pierces the dura and arachnoid mater, and ascends anterior to the hypoglossal roots. It inclines anterior to the medulla oblongata and unites with its contralateral fellow to form the midline basilar artery at the lower border of the pons. page 549 page 550
Figure 31.20 A dissection of the brain stem and the upper part of the spinal cord after removal of large portions of the occipital and parietal bones, the cerebellum and the roof of the fourth ventricle. On the left side, the foramina transversaria of the atlas and the third, fourth and fifth cervical vertebrae have been opened to expose the vertebral artery. On the right side the posterior arch of the atlas and the laminae of the succeeding cervical vertebrae have been divided and have been removed together with the vertebral spines and the contralateral laminae. The tentorium cerebelli and the transverse sinuses have been divided and their posterior portions removed.
CERVICAL BRANCHES OF THE VERTEBRAL ARTERY Spinal branches
The spinal branches enter the vertebral canal via the intervertebral foramina, and supply the spinal cord and its membranes. They fork into ascending and descending rami, which unite with those above and below, to form two lateral anastomotic chains on the posterior surfaces of the vertebral bodies near the attachment of their pedicles. Branches from these chains supply the periosteum and vertebral bodies, and others communicate with similar branches across the midline; from these connections small rami join similar ones above and below, to form a median anastomotic chain on the posterior surfaces of the vertebral bodies. Muscular branches
Muscular branches arise from the vertebral artery as it curves round the lateral mass of the atlas. They supply the deep muscles of the suboccipital region and anastomose with the occipital, ascending and deep cervical arteries.
INTERNAL THORACIC ARTERY (Fig. 63.5) The internal thoracic artery arises inferiorly from the first part of the subclavian artery, c.2 cm above the sternal end of the clavicle, opposite the root of the thyrocervical trunk.
THYROCERVICAL TRUNK (Fig. 31.6) The thyrocervical trunk is a short wide artery which arises from the front of the first part of the subclavian artery near the medial border of scalenus anterior, and
divides almost at once into the inferior thyroid, suprascapular and superficial cervical arteries. Inferior thyroid artery (Figs 31.6, 31.25) page 550 page 551
The inferior thyroid artery loops upwards anterior to the medial border of the scalenus anterior, turns medially just below the sixth cervical transverse process, then descends on longus colli to the lower border of the thyroid gland. It passes anterior to the vertebral vessels and posterior to the carotid sheath and its contents (and usually the sympathetic trunk, whose middle cervical ganglion frequently adjoins the vessel). On the left, near its origin, the artery is crossed anteriorly by the thoracic duct as the latter curves inferolaterally to its termination. Relations between the terminal branches of the artery and recurrent laryngeal nerve are very variable and of considerable surgical importance. The artery usually passes behind the nerve as it nears the gland. However, very close to the gland, the right nerve is equally likely to be anterior, posterior or amongst, the branches of the artery, and the left nerve is usually posterior. The artery is not accompanied by the inferior thyroid vein. Muscular branches
These supply the infrahyoid muscles, longus colli, scalenus anterior and the inferior pharyngeal constrictor. Ascending cervical artery
The ascending cervical artery is a small branch which arises as the inferior thyroid turns medially behind the carotid sheath and ascends on the anterior tubercles of the cervical transverse processes between scalenus anterior and longus capitis. It supplies the adjacent muscles and gives off one or two spinal branches which enter the vertebral canal through the intervertebral foramina to supply the spinal cord and membranes and vertebral bodies, and thereby supplement the spinal branches of the vertebral artery. The ascending cervical artery anastomoses with the vertebral, ascending pharyngeal, occipital and deep cervical arteries. Inferior laryngeal artery
The inferior laryngeal artery ascends on the trachea with the recurrent laryngeal nerve, enters the larynx at the lower border of the inferior constrictor and supplies the laryngeal muscles and mucosa. It anastomoses with its contralateral fellow, and with the superior laryngeal branch of the superior thyroid artery. Pharyngeal branches
These supply the lower part of the pharynx. Tracheal branches supply the trachea and anastomose with the bronchial arteries; oesophageal branches supply the oesophagus and anastomose with the oesophageal branches of the thoracic aorta; inferior and ascending glandular branches supply the posterior and inferior regions of the thyroid gland, and anastomose with the contralateral inferior and ipsilateral superior thyroid arteries. The ascending branch also supplies the parathyroid glands. Suprascapular artery (Fig. 31.8)
The suprascapular artery descends laterally across scalenus anterior and the phrenic nerve, posterior to the internal jugular vein and sternocleidomastoid. It then crosses anterior to the subclavian artery and brachial plexus, posterior to, and parallel with, the clavicle, subclavius and the inferior belly of omohyoid, to reach the superior scapular border. Superficial cervical artery (Fig. 31.8)
The superficial cervical artery is given off at a higher level than the suprascapular artery. It crosses anterior to the phrenic nerve, scalenus anterior and the brachial plexus and is covered by the internal jugular vein, sternocleidomastoid and platysma. It crosses the floor of the posterior triangle to reach the anterior margin of levator scapulae, and ascends deep to the anterior part of the trapezius, which it supplies, together with the adjoining muscles and the cervical lymph nodes. It anastomoses with the superficial ramus of the descending branch of the occipital artery. About a third of the superficial cervical and dorsal scapular arteries arise in common from the thyrocervical trunk, with a superficial (superficial cervical artery) and a deep (dorsal scapular artery) branch. The latter passes laterally anterior to the brachial plexus and then posterior to levator scapulae.
COSTOCERVICAL TRUNK (Fig. 63.5) On the right, this short vessel arises posteriorly from the second part of the subclavian artery, and, on the left, from its first part (Fig. 31.8). It arches back above the cervical pleura to the neck of the first rib, where it divides into superior intercostal and deep cervical branches. Deep cervical artery
The deep cervical artery usually arises from the costocervical trunk. It is analogous in its first segment to a posterior branch of a posterior intercostal
artery, and occasionally is a separate branch of the subclavian artery. It passes back above the eighth cervical spinal nerve between the transverse process of the seventh cervical vertebra and the neck of the first rib (sometimes between the transverse processes of the sixth and seventh cervical vertebrae). It then ascends between semispinales capitis and cervicis to the level of the second cervical vertebra. It supplies adjacent muscles and anastomoses with the deep branch of the descending branch of the occipital artery and branches of the vertebral artery. A spinal branch enters the vertebral canal between the seventh cervical and first thoracic vertebrae. Dorsal scapular artery
The dorsal scapular artery arises from the third, or less often the second, part of the subclavian artery. It gives off a small branch (which sometimes arises directly from the subclavian artery), to scalenus anterior. It passes laterally through the brachial plexus in front of scalenus medius and then deep to levator scapulae to the superior scapular angle. About a third of the superficial cervical and dorsal scapular arteries arise in common from the thyrocervical trunk as a transverse cervical artery, with a superficial (superficial cervical artery) and a deep (dorsal scapular artery) branch; the latter passes laterally anterior to the brachial plexus and then posterior to levator scapulae.
VEINS OF THE NECK Veins in the neck show considerable variation. They are superficial or deep to the deep fascia but are not entirely separate systems. Superficial veins are tributaries, some with specific names, given below, of the anterior, external and posterior jugular veins. They drain a much smaller volume of tissue than the deep veins. The latter drain all but the subcutaneous structures, mostly into the internal jugular vein and also into the subclavian vein.
EXTERNAL JUGULAR VEIN The external jugular vein mainly drains the scalp and face, although it also drains some deeper parts. The vein is formed by the union of the posterior division of the retromandibular vein with the posterior auricular vein and begins near the mandibular angle just below or in the parotid gland. It descends from the angle to the midclavicle, running obliquely, superficial to sternocleidomastoid, to the root of the neck. Here it crosses the deep fascia and ends in the subclavian vein, lateral or anterior to scalenus anterior. There are valves at its entrance into the subclavian, but they do not prevent regurgitation. Its wall is adherent to the rim of the fascial opening. It is covered by platysma, superficial fascia and skin, and is separated from sternocleidomastoid by deep cervical fascia. The vein crosses the transverse cutaneous nerve and lies parallel with the great auricular nerve, posterior to its upper half. In size the external jugular vein is inversely proportional to the other veins in the neck, and may be double. Between the entrance into the subclavian vein and a point c.4 cm above the clavicle, the vein is often dilated, producing a so-called sinus. Tributaries In addition to formative tributaries, the external jugular receives the posterior external jugular and, near its end, transverse cervical, suprascapular and anterior jugular veins. In the parotid gland it is often joined by a branch from the internal jugular. The occipital vein occasionally joins it.
POSTERIOR EXTERNAL JUGULAR VEIN The posterior external jugular vein begins in the occipital scalp, and drains the skin and the superficial muscles which lie posterosuperior in the neck. It usually joins the middle part of the external jugular vein.
ANTERIOR JUGULAR VEIN page 551 page 552
The anterior jugular vein arises near the hyoid bone from the confluence of the superficial submandibular veins. It descends between the midline and the anterior border of sternocleidomastoid. Turning laterally, low in the neck, deep to sternocleidomastoid but superficial to the infrahyoid strap muscles, it joins either the end of the external jugular vein or may enter the subclavian vein directly. In size it is usually inverse to the external jugular vein. It communicates with the internal jugular vein, and receives the laryngeal veins and sometimes a small thyroid vein. There are usually two anterior jugular veins, united just above the manubrium by a large transverse jugular arch, receiving the inferior thyroid tributaries. They have no valves and may be replaced by a midline trunk.
INTERNAL JUGULAR VEIN (Figs 29.8, 31.8, 31.10, 31.15, 31.16, 31.21) The internal jugular vein collects blood from the skull, brain, superficial parts of face and much of the neck. It begins at the cranial base in the posterior compartment of the jugular foramen, where it is continuous with the sigmoid
sinus. At its origin it is dilated as the superior bulb, which lies below the posterior part of the tympanic floor. The internal jugular vein descends in the carotid sheath, and unites with the subclavian vein, posterior to the sternal end of the clavicle, to form the brachiocephalic vein. Near its termination the vein dilates into the inferior bulb, above which is a pair of valves. Relations
Figure 31.21 A dissection to show the general distribution of the left hypoglossal and lingual nerves and the position and constitution of some parts of the cervical plexus of the left side.
From above, rectus capitis lateralis, the transverse process of the atlas, levator scapulae, scalenus medius, scalenus anterior, the cervical plexus, the phrenic nerve, thyrocervical trunk, vertebral vein and the first part of the subclavian artery all lie posterior to the vein. On the left, the internal jugular crosses anterior to the thoracic duct (Fig. 31.6). The internal and common carotid arteries and the vagus nerve are medial to the vein: the nerve lies between vein and arteries but posterior to them. Superficially the internal jugular vein is overlapped above, then covered below, by sternocleidomastoid and is crossed by the posterior belly of digastric and the superior belly of omohyoid. Superior to digastric the parotid gland, styloid process, accessory nerve, and posterior auricular and occipital arteries cross the vein. Between digastric and omohyoid, the sternocleidomastoid arteries and inferior root of the ansa cervicalis cross it, although the nerve often passes between the vein and the common carotid artery. Below omohyoid, the vein is covered by the infrahyoid muscles and sternocleidomastoid and is crossed by the anterior jugular vein. Deep cervical lymph nodes lie along the internal jugular, mainly on its superficial aspect. At the root of the neck the right internal jugular vein is separated from the common carotid artery, but the left usually overlaps its artery. At the base of the skull the internal carotid artery is anterior to the vein, separated from it by the ninth to twelfth cranial nerves. Tributaries The inferior petrosal sinus, facial, lingual, pharyngeal, superior and middle thyroid veins, and occasionally the occipital vein, are all tributaries of the internal jugular vein. The internal jugular vein may communicate with the external jugular vein. The thoracic duct opens near the union of the left subclavian and internal jugular veins, and the right lymphatic duct opens at the same site on the right.
INFERIOR PETROSAL SINUS The inferior petrosal sinus leaves the skull through the anterior part of the jugular foramen, crosses lateral or medial to the ninth to eleventh cranial nerves and joins the superior jugular bulb.
FACIAL VEIN (Figs 29.8, 31.10) The initial part of the facial vein as it lies on the face is described on page 511. From the face it passes over the surface of masseter, crosses the body of the mandible and enters the neck where it runs obliquely back under platysma. Here it lies superficial to the submandibular gland, digastric and stylohyoid. Just
anteroinferior to the mandibular angle it is joined by the anterior division of the retromandibular vein, and then descends superficial to the loop of the lingual artery, the hypoglossal nerve and external and internal carotid arteries, to enter the internal jugular near the greater cornu of the hyoid bone, i.e. in the upper angle of the carotid triangle. Near its end a large branch often descends along the anterior border of sternocleidomastoid to the anterior jugular vein. Its uppermost segment, above its junction with the superior labial vein, is often termed the angular vein. Tributaries Submental, tonsillar, external palatine (paratonsillar), submandibular, vena comitans of the hypoglossal nerve (sometimes), pharyngeal and superior thyroid veins are all tributaries of the portion of the facial vein that lies below the mandible.
LINGUAL VEIN (Fig. 29.8) The lingual veins follow two routes. The dorsal lingual veins drain the dorsum and sides of the tongue, join the lingual veins accompanying the lingual artery between hyoglossus and genioglossus, and enter the internal jugular near the greater cornu of the hyoid bone. The deep lingual vein begins near the tip of the tongue and runs back, lying near the mucous membrane on the inferior surface of the tongue. Near the anterior border of hyoglossus it joins a sublingual vein, from the sublingual salivary gland, to form the vena comitans nervi hypoglossi which runs back between mylohyoid and hyoglossus with the hypoglossal nerve to join the facial, internal jugular or lingual vein.
PHARYNGEAL VEINS (See Chapter 35) The pharyngeal veins begin in a pharyngeal plexus external to the pharynx. They receive meningeal veins and a vein from the pterygoid canal, and usually end in the internal jugular vein, but may sometimes end in the facial, lingual or superior thyroid vein.
SUPERIOR THYROID VEIN (Figs 29.8, 31.15) The superior thyroid vein is formed by deep and superficial tributaries corresponding to the arterial branches in the upper part of the thyroid gland (Fig. 31.15). It accompanies the superior thyroid artery, receives the superior laryngeal and cricothyroid veins, and ends in the internal jugular or facial vein.
MIDDLE THYROID VEIN The middle thyroid vein drains the lower part of the gland and also receives veins from the larynx and trachea (Fig. 31.15).It crosses anterior to the common carotid artery to join the internal jugular vein behind the superior belly of omohyoid.
TYMPANIC BODY page 552 page 553
The tympanic body (glomus jugulare) is ovoid, c.0.5 mm long and 0.25 mm broad, and lies in the adventitia of the upper part of the superior bulb of the internal jugular vein. It is similar in structure to the carotid body (p. 547) and is presumed to have a similar function. The predominant cell type has morphological similarities to adrenal chromaffin cells, and is derived from the neural crest. Cells obtained from glomus jugulare paragangliomas show spontaneous neurite outgrowth in culture, and have vasoactive intestinal peptide (VIP)-like activity. The tympanic body may be present as two or more parts near the tympanic branch of the glossopharyngeal nerve or the auricular branch of the vagus as they lie within their canals in the petrous part of the temporal bone. Tumours of tympanic bodies may involve the adjacent cranial nerves and the middle ear.
SUBCLAVIAN VEIN (Fig. 31.8) The subclavian vein is a continuation of the axillary vein and extends from the outer border of the first rib to the medial border of scalenus anterior, where it joins the internal jugular vein to form the brachiocephalic vein. The clavicle and subclavius lie anterior to it, the subclavian artery is posterosuperior, separated by scalenus anterior and the phrenic nerve, and the first rib and pleura are inferior. The vein usually has a pair of valves, c.2 cm from its end. Its tributaries are the external jugular, dorsal scapular and sometimes the anterior jugular vein. At its junction with the internal jugular, the left subclavian vein receives the thoracic duct, and the right subclavian vein receives the right lymphatic duct.
VERTEBRAL VEIN Numerous small tributaries from internal vertebral plexuses leave the vertebral canal above the posterior arch of the atlas and join small veins from local deep muscles in the suboccipital triangle. Their union produces a vessel which enters the foramen in the transverse process of the atlas and forms a plexus around the vertebral artery. It descends through successive transverse foramina and ends as the vertebral vein. The vein emerges from the sixth cervical transverse foramen, whence it descends, at first anterior, then anterolateral, to the vertebral artery, to
open superoposteriorly into the brachiocephalic vein: the opening has a paired valve. As it descends it passes behind the internal jugular vein and in front of the first part of the subclavian artery. A small accessory vertebral vein usually descends from the vertebral plexus, traverses the seventh cervical transverse foramen and turns forwards between the subclavian artery and the cervical pleura to join the brachiocephalic vein. Tributaries The vertebral vein connects with the sigmoid sinus by a vessel in the posterior condylar canal, when this exists. It also receives branches from the occipital vein, prevertebral muscles, internal and external vertebral plexuses. It is joined by anterior vertebral and deep cervical veins (see below) and, sometimes near its end, by the first intercostal vein.
ANTERIOR VERTEBRAL VEIN The anterior vertebral vein starts in a plexus around the upper cervical transverse processes, descends near the ascending cervical artery between attachments of scalenus anterior and longus capitis, and opens into the end of the vertebral vein.
DEEP CERVICAL VEIN The deep cervical vein accompanies its artery between semispinales capitis and cervicis. It is formed in the suboccipital region by the union of communicating branches of the occipital vein; veins from suboccipital muscles; and veins from plexuses around the cervical spines. It passes forwards between the seventh cervical transverse process and the neck of the first rib to end in the lower part of the vertebral vein.
Cervical groups of lymph nodes (Fig. 31.1) Lymph nodes in the head and neck are distributed in terminal and outlying groups. The terminal group is related to the carotid sheath and the nodes it contains are the deep cervical lymph nodes. All lymph vessels of the head and neck drain into this group, either directly from tissues or indirectly through nodes in the outlying groups. Efferents of the deep cervical nodes form the jugular trunk. The right jugular trunk collects lymph from the right arm and right half of the thorax and the right head and neck and may end in the jugulosubclavian junction or the right lymphatic duct. The left jugular trunk usually enters the thoracic duct, but it may join the internal jugular or subclavian vein. UPDATE Date Added: 26 April 2005 Abstract: Current concepts in cervical lymph node imaging. Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=15347718 Current concepts in cervical lymph node imaging.
LYMPHATIC DRAINAGE OF THE NECK (Fig. 31.1) Many vessels draining the superficial cervical tissues skirt the borders of sternocleidomastoid to reach the superior or inferior deep cervical nodes. Others pass to the superficial cervical and occipital nodes. Lymph from the superior region of the anterior triangle drains to the submandibular and submental nodes. Vessels from the anterior cervical skin inferior to the hyoid bone pass to the anterior cervical lymph nodes near the anterior jugular veins. Their efferents go to the deep cervical nodes of both sides, including the infrahyoid, prelaryngeal and pretracheal groups. An anterior cervical node often occupies the suprasternal space.Lymph from tissues of the head and neck internal to the deep fascia drains to the deep cervical nodes directly or through outlying groups that include the retropharyngeal, paratracheal, lingual, infrahyoid, prelaryngeal and pretracheal groups. The lymphatic drainage associated with the nasal region, larynx and oral cavity is described in the appropriate regions. The deep cervical lymphatic nodes lie alongside the carotid sheath, and form superior and inferior groups. Superior deep cervical nodes The superior deep cervical nodes adjoin the upper part of the internal jugular vein. Most are deep to sternocleidomastoid, but a few extend beyond it. One subgroup, consisting of one large and several small nodes, is in a triangular region bounded by the posterior belly of digastric and the facial and internal jugular veins, and is known as the jugulodigastric group. It is concerned specially with drainage of the tongue. Efferents from the upper deep cervical nodes drain either to the lower group or direct to the jugular trunk. Inferior deep cervical nodes The inferior deep cervical nodes are partly deep to the sternocleidomastoid, and are particularly related to the lower part of the internal jugular vein. Some are closely related to the brachial plexus and subclavian vessels. The juguloomohyoid node lies on, or just above, the intermediate tendon of omohyoid, and is concerned especially with lymphatic drainage from the tongue. Efferents from this lower group join the jugular lymph trunk.
Retropharyngeal nodes Retropharyngeal nodes lie between the pharyngeal and prevertebral fasciae and form a median and two lateral groups, the latter anterior to the lateral masses of the atlas along the lateral borders of longus capitis. The nodes receive afferents from the nasopharynx, pharyngotympanic tube and atlanto-occipital and atlantoaxial joints and drain to the upper deep cervical nodes. Paratracheal nodes The paratracheal nodes flank both trachea and oesophagus along the recurrent laryngeal nerves. Efferents pass to the corresponding deep cervical nodes. Infrahyoid, prelaryngeal and pretracheal nodes The infrahyoid, prelaryngeal and pretracheal nodes lie beneath the deep cervical fascia. They drain afferents from the anterior cervical nodes, and their efferents join the deep cervical nodes. Infrahyoid nodes are anterior to the thyrohyoid membrane, prelaryngeal nodes lie on the cricovocal membrane, and pretracheal nodes lie anterior to the trachea near the inferior thyroid veins. Lingual nodes Lingual nodes are small and inconstant, and are situated on the external surface of hyoglossus and also between the genioglossi. They drain to the upper deep cervical nodes.
SPREAD OF MALIGNANT DISEASE IN THE NECK page 553 page 554
Cancers arising in the head and neck from regions such as the thyroid gland, larynx, oral cavity and oropharynx, nasopharynx and paranasal sinuses have predictable patterns of spread through the chains of lymph nodes in the neck. When operating on malignant disease in this region it is vitally important to understand these patterns of spread so that for any individual cancer the appropriate operation is undertaken. Clinical experience has shown that the lymph nodes in the neck fall into five distinct groups (Fig. 31.1). Level I nodes lie in the submandibular triangle bounded by the anterior and posterior bellies of digastric and the lower border of the mandible above. Level II (upper jugular) nodes lie around the upper portion of the internal jugular vein and the upper part of the spinal accessory nerve. They extend from the skull base to the bifurcation of the common carotid artery or the hyoid bone. Level III (middle jugular) nodes lie around the middle third of the internal jugular vein from the inferior border of level II to the superior belly of omohyoid or cricothyroid membrane. Level IV (lower jugular) nodes lie around the lower third of the internal jugular vein from the inferior border of level III to the clavicle. The anterior and posterior borders for levels II, III and IV are the lateral border of sternohyoid and the posterior border of sternocleidomastoid respectively. Level V (posterior triangle) nodes lie around the lower part of the spinal accessory nerve and the transverse cervical vessels. Knowing which levels of nodes are likely to be involved in the metastatic spread of a particular cancer arising in the head and neck means that appropriate nodal clearance can be undertaken. The classic radical neck dissection first described by Crile in 1906 involves a thorough clearance of levels I to V including the sacrifice of sternocleidomastoid, the internal jugular vein and the spinal accessory nerve. Modified radical neck dissections (so-called functional neck dissections) still remove level I to V nodes, but spare either or all of sternocleidomastoid, the internal jugular vein and the spinal accessory. Selective neck dissections remove selected groups of nodes, e.g. the supraomohyoid neck dissection removes level I to III nodes, the lateral neck dissection removes level II to IV nodes, and the posterolateral neck dissection removes level II to V nodes.
CERVICAL LYMPHOVENOUS PORTALS Lymph is returned to the systemic venous circulation via right and left lymphovenous portals sited at, or near, the junctions of the internal jugular and subclavian veins. The arrangement of these terminations is variable. Usually, three small lymph trunks converge towards their venous junctions on either side of the body, and they are joined, on the left side only, by the larger thoracic duct. On the right side, the three trunks are the right jugular, right subclavian and right bronchomediastinal. The right jugular trunk extends from the terminal lower deep cervical nodes along the ventrolateral aspect of the internal jugular vein, and conveys all the lymph from the right half of the head and neck. The right subclavian trunk drains from the terminal apical axillary group. It extends along the axillary and subclavian veins, and conveys lymph from the right upper limb and superficial tissues of the right half of the thoracoabdominal wall, down to the umbilicus anteriorly and iliac crest posteriorly (and includes much of the breast). The right bronchomediastinal trunk, ascends over the trachea towards the lymphovenous portal and conveys lymph from the thoracic walls, the right cupola of the diaphragm and subjacent liver, the right lung, bronchi and trachea, the greater part of the 'right heart'-of clinical parlance, not the geometric right half-and a proportionately small drainage from the thoracic oesophagus.
The three right lymphatic trunks usually open independently (Fig. 31.22). Their orifices are clustered either on the ventral aspect of the jugulo/subclavian junction, or in the nearby wall of either of the great veins. Sometimes one or more of the trunks may bifurcate (or even trifurcate) preterminally and then terminate via multiple orifices. Rarely, the three trunks fuse to form a short, single, right lymphatic duct (c.1 cm long) that inclines across the medial border of scalenus anterior at the root of the neck to reach the ventral aspect of the venous junction, where its orifice is guarded by a bicuspid semilunar valve. An incomplete right lymphatic duct may be present if the subclavian and jugular trunks, or any combination of their terminals, are fused. When this occurs, the bronchomediastinal trunk almost invariably opens separately. On the left, the four trunks that converge on the left lymphovenous portal are the left jugular and left subclavian trunks, which have a disposition corresponding to that of their counterparts on the right; the left bronchomediastinal trunk, which has a drainage similar to the right trunk, but which drains more of the heart-the 'left' and part of the 'right' hearts of clinical parlance-and more of the oesophagus; and the thoracic duct, which drains all of the rest of the body.
© 2008 Elsevier
INNERVATION The skin, joints, viscera and muscles of the neck are innervated by branches of the glossopharyngeal, vagus and spinal accessory nerves, the cervical spinal nerves and the cervical sympathetic trunk.
Figure 31.22 Variation in the terminal lymphatic trunks nodes of the right side. (By permission from Poirier P, Charpy A 1901 Traite d'Anatomie Humaine. Paris: Masson et Cie.)
The first and second cervical dorsal root ganglia lie on the vertebral arches of the atlas and axis respectively. The first cervical ganglion may be absent. Smaller aberrant ganglia sometimes occur on the upper cervical dorsal roots between the ganglia and cord. The upper four cervical roots are small, the lower four are large. In general, cervical dorsal roots have a thickness ratio to the ventral roots of three to one, which is greater than is seen in other regions. The first dorsal root is an exception, being smaller than the ventral, and in c.8% cases it is absent. The first and second cervical roots are short, and run almost horizontally to their exit from the vertebral canal. From the third to the eighth cervical they slope obliquely down. Obliquity and length increase successively, however the distance between spinal attachment and vertebral exit never exceeds the height of one vertebra.
Cervical ventral rami
Cervical ventral rami, except the first, appear between the anterior and posterior intertransverse muscles. The upper four form the cervical plexus, and the lower four, together with most of the first thoracic ventral ramus, form the brachial plexus. Each receives at least one grey ramus communicans, the upper four from the superior cervical sympathetic ganglion, the fifth and sixth from the middle ganglion and the seventh and eighth from the cervicothoracic ganglion. The first cervical ventral ramus, the suboccipital nerve, emerges above the posterior arch of the atlas, passes forwards lateral to its lateral mass and medial to the vertebral artery. It supplies rectus capitis lateralis, emerges medial to it, descends anterior to the transverse process of the atlas and posterior to the internal jugular vein and joins the ascending branch of the second cervical ventral ramus. page 554 page 555
The second cervical ventral ramus issues between the vertebral arches of the atlas and axis. It ascends between their transverse processes, passes anterior to the first posterior intertransverse muscle and emerges lateral to the vertebral artery generally between longus capitis and levator scapulae. The ramus divides into an ascending branch which joins the first cervical nerve and a descending branch which joins the ascending branch of the third cervical ventral ramus. The third cervical ventral ramus appears between longus capitis and scalenus medius. The remaining ventral rami emerge between scalenus anterior and the scalenus medius.
CERVICAL PLEXUS (Figs 31.2, 31.6, 31.21, 29.9) The cervical plexus is formed by the ventral rami of the upper four cervical nerves, and supplies some neck muscles and the diaphragm, and areas of skin on the head, neck and chest. It is situated in the neck opposite a line drawn down the side of the neck from the root of the auricle to the level of the upper border of the thyroid cartilage. It is deep to the internal jugular vein, the deep fascia and sternocleidomastoid, and anterior to scalenus medius and levator scapulae. Each ramus, except the first, divides into ascending and descending parts, which unite in communicating loops. From the first loop (C2 and 3) superficial branches supply the head and neck; cutaneous nerves of the shoulder and chest arise from the second loop (C3 and 4). Muscular and communicating branches arise from the same nerves. The branches are superficial or deep. The superficial branches perforate the cervical fascia to supply the skin while the deep branches in general supply muscles. The superficial branches either ascend (the lesser occipital, great auricular and the transverse cutaneous nerves) or descend (supraclavicular nerves). These nerves are described in detail on page 532. The deep branches form medial and lateral series.
DEEP BRANCHES-MEDIAL SERIES Communicating branches Communicating branches pass from the loop between the first and second cervical rami to the vagus and hypoglossal nerves and to the sympathetic trunk. The hypoglossal branch later leaves the hypoglossal nerve as a series of branches, viz. the meningeal, superior root of ansa cervicalis, nerves to thyrohyoid and to geniohyoid. A branch also connects the fourth and fifth cervical rami. The first four cervical ventral rami each receive a grey ramus communicans from the superior cervical sympathetic ganglion. The superior root of the ansa cervicalis (descendens hypoglossi) (Fig. 31.2) leaves the hypoglossal nerve where it curves round the occipital artery and then descends anterior to or in the carotid sheath. It contains only fibres from the first cervical spinal nerve. After giving a branch to the superior belly of omohyoid, it is joined by the inferior root of the ansa from the second and third cervical spinal
nerves. The two roots form the ansa cervicalis (ansa hypoglossi), from which branches supply sternohyoid, sternothyroid and the inferior belly of omohyoid. Another branch is said to descend anterior to the vessels into the thorax to join the cardiac and phrenic nerves. Muscular branches Muscular branches supply rectus capitis lateralis (C1), rectus capitis anterior (C1, 2), longus capitis (C1-3) and longus colli (C2-4). The inferior root of the ansa cervicalis and the phrenic nerve are additional muscular branches. Inferior root of the ansa cervicalis (nervus descendens cervicalis) (Fig. 31.2)
The inferior root of the ansa cervicalis is formed by the union of a branch from the second with another from the third cervical ramus. It descends on the lateral side of the internal jugular vein, crosses it a little below the middle of the neck, and continues forwards to join the superior root anterior to the common carotid artery, forming the ansa cervicalis (ansa hypoglossi), from which all infrahyoid muscles except thyrohyoid are supplied. The inferior root comes from the second and third cervical ventral rami in c.75% cases, from the second to fourth in c.15%, from the third alone in c.5%. Occasionally it may be derived from either the second alone or from the first to third. Phrenic nerve
The phrenic nerve arises chiefly from the fourth cervical ventral ramus, but also has contributions from the third and fifth. It is formed at the upper part of the lateral border of scalenus anterior and descends almost vertically across its anterior surface behind the prevertebral fascia. It descends posterior to sternocleidomastoid, the inferior belly of omohyoid (near its intermediate tendon), the internal jugular vein, transverse cervical and suprascapular arteries and, on the left, the thoracic duct. At the root of the neck, it runs anterior to the second part of the subclavian artery, from which it is separated by the scalenus anterior (some accounts state that on the left side the nerve passes anterior to the first part of the subclavian artery), and posterior to the subclavian vein. The phrenic nerve enters the thorax by crossing medially in front of the internal thoracic artery. In the neck, each nerve receives variable filaments from the cervical sympathetic ganglia or their branches and may also connect with internal thoracic sympathetic plexuses. Accessory phrenic nerve
The accessory phrenic nerve is composed of fibres from the fifth cervical ventral ramus which run in a branch of the nerve to subclavius. This lies lateral to the phrenic nerve and descends posterior (occasionally anterior) to the subclavian vein. The accessory phrenic nerve usually joins the phrenic nerve near the first rib, but may not do so until near the pulmonary hilum or beyond. The accessory phrenic nerve may be derived from the fourth or sixth cervical ventral rami or from the ansa cervicalis.
DEEP BRANCHES-LATERAL SERIES Communicating branches Lateral deep branches of the cervical plexus (C2, 3, 4) may connect with the spinal accessory nerve within sternocleidomastoid, the posterior triangle or beneath trapezius. Muscular branches (Fig. 31.15) Muscular branches are distributed to sternocleidomastoid (C2, 3, 4), trapezius (C2 and possibly C3), levator scapulae (C3, 4) and scalenus medius (C3, 4). Branches to trapezius cross the posterior triangle obliquely below the spinal accessory nerve.
BRACHIAL PLEXUS (Fig. 31.8)
The brachial plexus is formed by the union of the ventral rami of the lower four cervical nerves and the greater part of the ventral ramus of the first thoracic ventral ramus. It may also receive contributions from the fourth cervical and second thoracic spinal nerves. As its name suggests, its branches supply the muscles, joints and skin of the upper limb. The relations and distribution of the brachial plexus are described in detail in Section 5. However, it is also mentioned here because, at its origin, the brachial plexus lies in the posterior triangle of the neck, in the angle between the clavicle and the lower posterior border of sternocleidomastoid. It emerges between the scaleni anterior and medius, superior to the third part of the subclavian artery, and is covered by platysma, deep fascia and skin, through which it is palpable. It is crossed by the supraclavicular nerves, the nerve to subclavius, the inferior belly of omohyoid, the external jugular vein and the superficial ramus of the transverse cervical artery. The plexus passes posterior to the medial two-thirds of the clavicle, subclavius and the suprascapular vessels, and lies on the first digitation of serratus anterior and on subscapularis. UPDATE Date Added: 07 March 2006 Publication Services, Inc. Abstract: Donor, recipient, and nerve grafts in brachial plexus reconstruction: anatomical and technical features for facilitating the exposure Click on the following line to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=16132194&query_hl=8&itool=pubmed_docsum Donor, recipient, and nerve grafts in brachial plexus reconstruction: anatomical and technical features for facilitating the exposure. Norkus T, Norkus M, Ramanauskas T: Surg Radiol Anat 27:524-530, 2005.
Cervical dorsal rami (Figs 46.14, 29.9) Each cervical spinal dorsal ramus except the first divides into medial and lateral branches, and all innervate muscles. In general, only medial branches of the second to fourth, and usually the fifth, supply skin. Except for the first (sometimes called the suboccipital nerve) and second, each dorsal ramus passes back medial to a posterior intertransverse muscle, curving round the articular process into the interval between semispinalis capitis and semispinalis cervicis. The cervical dorsal rami are described in detail on pages 783-784.
Cranial nerves UPDATE Date Added: 01 December 2004 Abstract: Incidence of cranial nerve injuries after carotid eversion endartectomy with a transverse skin incision under regional anaesthesia. Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=15350567 Incidence of cranial nerve injuries after carotid eversion endartectomy with a transverse skin incision under regional anaesthesia.
GLOSSOPHARYNGEAL NERVE (Fig. 31.23) The glossopharyngeal nerve (Figs 31.16, 31.20) supplies motor fibres to stylopharyngeus, parasympathetic secretomotor fibres to the parotid gland (derived from the inferior salivatory nucleus), sensory fibres to the tympanic cavity, pharyngotympanic tube, fauces, tonsils, nasopharynx, uvula and posterior (postsulcal) third of the tongue, and gustatory fibres for the postsulcal part of the tongue. page 555 page 556
Figure 31.23 The communications between the last four cranial nerves of the left side viewed from the dorsolateral aspect. The hypoglossal canal has been split in its long axis and the transverse process of the atlas has been divided close to the lateral mass. The descending branch of the hypoglossal nerve is not shown.
The nerve leaves the skull through the anteromedial part of the jugular foramen, anterior to the vagus and accessory nerves, and in a separate dural sheath. In the foramen it is lodged in a deep groove leading from the cochlear aqueductal depression, and is separated by the inferior petrosal sinus from the vagus and accessory nerves. The groove is bridged by fibrous tissue, which is calcified in c.25% of skulls. After leaving the foramen, the nerve passes forwards between the internal jugular vein and internal carotid artery and then descends anterior to the latter, deep to the styloid process and its attached muscles, to reach the posterior border of stylopharyngeus. It curves forwards on stylopharyngeus and either pierces the lower fibres of the superior pharyngeal constrictor or passes between it and the middle constrictor to be distributed to the tonsil, the mucosae of the pharynx and postsulcal part of the tongue, the vallate papillae, and oral mucous glands. Two ganglia, superior and inferior, are situated on the glossopharyngeal nerve as it traverses the jugular foramen (Fig. 31.23). The superior ganglion is in the upper part of the groove occupied by the nerve in the jugular foramen. It is small, has no branches and is usually regarded as a detached part of the inferior ganglion. The inferior ganglion is larger and lies in a notch in the lower border of the petrous part of the temporal bone. Its cells are typical unipolar neurones, whose peripheral branches convey gustatory and tactile signals from the mucosa of the tongue (posterior third including the sulcus terminalis and vallate papillae) and general sensation from the oropharynx, where it is responsible for initiating the gag reflex. UPDATE Date Added: 23 August 2005 Shanida Helena Nataraja, PhD (Dianthus Medical Limited) Update: Microsurgical anatomic study of glossopharyngeal nerve landmarks. A number of landmarks can be used to guide the exposure of the glossopharyngeal nerve. Ten cadaveric heads (20 specimens) were fixed with glycerol and the arteries and veins injected with colored latex to enhance visibility.
The entire length of the glossopharyngeal nerve was exposed both intra- and extra-durally and its course was examined using a surgical microscope. The nerve can be divided into three main sections: cisternal, jugular foramen, and extracranial. In the cisternal portion (average length 15mm), it emerges from the medulla oblongata as three to five filaments, which usually then form a single root that travels forwards and laterally on the anterior aspect of the flocculus and choroid plexus. It next rests on the jugular tubercle of the occipital bone. In the jugular foramen portion (average length, 11mm), the nerve travels upwards from its entrance porus forming a genu inferiorly before passing through the jugular foramen where it is separated from the vagus and accessory nerves by a thick, fibrous band (80% of cases) and by a bony canal (20% of cases). Inside the jugular foramen, the nerve has two expansions in its course, namely the superior and inferior ganglia. The tympanic (Jacobson's) nerve originates from the inferior ganglion. The superior ganglion is located slightly below the opening of the cochlear aqueducts. Arnold's nerve, which originates from the superior ganglion of the vagus, also contains branches of the glossopharyngeal. The nerve then exits the jugular foramen posteromedial to the styloid process and associated muscles. In the extracranial portion (average length, 75mm), it travels posterior to, and innervates, stylopharyngeus. It then courses to the medial aspect of stylopharyngeaus and penetrates the pharyngeal wall slightly above the level of the middle constrictor. In the jugular foramen, the opening of the cochlear aqueduct, the mastoid canaliculus, and the inferior tympanic canaliculus are valuable landmarks, and in the extracranial portion, the transverse process of the atlas, the base of the styloid process, and the base of the styloid pyramid can be used to guide nerve exposure. Ozveren MF, Ture U, Ozek MM et al: Anatomic landmarks of the glossopharyngeal nerve: a microsurgical anatomic study. Neurosurgery 52(6):1400-1410, 2003. Medline Similar articles
COMMUNICATING BRANCHES The glossopharyngeal nerve communicates with the sympathetic trunk, vagus and facial nerves. The inferior ganglion is connected with the superior cervical sympathetic ganglion. Two filaments from the inferior ganglion pass to the vagus, one to its auricular branch and the other to its superior ganglion. A branch to the facial nerve arises from the glossopharyngeal nerve below the inferior ganglion, and perforates the posterior belly of digastric to join the facial nerve near the stylomastoid foramen.
BRANCHES OF DISTRIBUTION These are tympanic, carotid, pharyngeal, muscular, tonsillar and lingual. Tympanic nerve The tympanic nerve leaves the inferior ganglion, ascends to the tympanic cavity through the inferior tympanic canaliculus and divides into branches that contribute to the tympanic plexus. The lesser petrosal nerve is derived from the tympanic plexus. Carotid branch The carotid branch is often double. It arises just below the jugular foramen and descends on the internal carotid artery to the wall of the carotid sinus and to the carotid body. The nerve contains primary afferent fibres from chemoreceptors in the carotid body and from the baroreceptors lying in the carotid sinus wall. It may communicate with the inferior ganglion of the vagus, or with one of its branches, and with a sympathetic branch from the superior cervical ganglion. Pharyngeal branches The pharyngeal branches are three or four filaments which unite with the pharyngeal branch of the vagus and the layngopharyngeal branches of the sympathetic trunk to form the pharyngeal plexus near the middle pharyngeal constrictor. They constitute the route by which the glossopharyngeal nerve supplies sensory fibres to the mucosa of the pharynx.
Muscular branch The muscular branch supplies stylopharyngeus. Tonsillar, lingual and inferior petrosal branches The tonsillar, lingual and inferior petrosal branches are described on pages 623, 588, and 237 respectively.
LESIONS OF THE GLOSSOPHARYNGEAL NERVE Damage to the glossopharyngeal nerve rarely occurs without involvement of other lower cranial nerves. Transient or sustained hypertension may follow surgical section of the nerve, reflecting involvement of the carotid branch. Isolated lesions of the glossopharyngeal nerve lead to loss of sensation over the ipsilateral soft palate, fauces, pharynx and posterior third of the tongue, although this is difficult to assess clinically and requires galvanic stimulation. The palatal and pharyngeal (gag) reflexes are reduced or absent and salivary secretion from the parotid gland may also be reduced. Weakness of stylopharyngeus cannot be tested individually. Glossopharyngeal neuralgia consists of episodic brief but severe pain, often precipitated by swallowing, and experienced in the throat, behind the angle of the jaw and within the ear. Superior jugular bulb thromboses (e.g. in otitis media) and jugular foramen syndrome (associated with nasopharyngeal carcinoma and a glomus tumour) may cause lesions of the adjacent glossopharyngeal, vagus and accessory nerves, with associated weakness in the muscles supplied (in the pharynx and larynx).
VAGUS NERVE page 556 page 557
The vagus is a large mixed nerve. It has a more extensive course and distribution than any other cranial nerve, and traverses the neck, thorax and abdomen (Figs 31.14, 31.16, 19.3, 19.7, 31.23). Its central connections are described in Chapter 19). The vagus exits the skull through the jugular foramen accompanied by the accessory nerve, with which it shares an arachnoid and a dural sheath. Both nerves lie anterior to a fibrous septum that separates them from the glossopharyngeal nerve. The vagus descends vertically in the neck in the carotid sheath, between the internal jugular vein and the internal carotid artery, to the upper border of the thyroid cartilage, and then passes between the vein and the common carotid artery to the root of the neck. Its relationships in this part of its course are therefore similar to those described for these structures. Its further course differs on the two sides. The right vagus descends posterior to the internal jugular vein to cross the first part of the subclavian artery and enter the thorax. The left vagus enters the thorax between the left common carotid and subclavian arteries and behind the left brachiocephalic vein. After emerging from the jugular foramen, the vagus bears two marked enlargements, a small, round, superior ganglion and a larger inferior ganglion (Fig. 31.23).
SUPERIOR (JUGULAR) GANGLION The superior ganglion is greyish, spherical and c.4 mm in diameter. It is connected to the cranial root of the accessory nerve, the inferior glossopharyngeal ganglion, and the sympathetic trunk, the latter by a filament from the superior cervical ganglion. The significance of these connections is not entirely clear, but the first probably contains aberrant motor fibres from the nucleus ambiguus which issue in the accessory nerve, to be distributed to the palatal, pharyngeal, laryngeal and upper oesophageal musculature via the vagus.
INFERIOR (NODOSE) GANGLION
The inferior or nodose ganglion is larger than the superior ganglion, and is elongated and cylindrical in shape with a length of c.25 mm and a maximum breadth of 5 mm. It is connected with the hypoglossal nerve, the loop between the first and second cervical spinal nerves, and with the superior cervical sympathetic ganglion. Just above the ganglion the cranial accessory blends with the vagus nerve, its fibres being distributed mainly in pharyngeal and recurrent laryngeal vagal branches. Most visceral afferent fibres have their cell bodies in the nodose ganglion. Both vagal ganglia are exclusively sensory, and contain somatic, special visceral and general visceral afferent neurones. The superior ganglion is chiefly somatic, and most of its neurones enter the auricular nerve, whilst neurones in the inferior ganglion are concerned with visceral sensation from the heart, larynx, lungs and the alimentary tract from the pharynx to the transverse colon. Some fibres transmit impulses from taste endings in the vallecula and epiglottis. Large afferent fibres are derived from muscle spindles in the laryngeal muscles. Vagal sensory neurones in the nodose ganglion may show some somatotopic organization. Both ganglia are traversed by parasympathetic, and perhaps some sympathetic fibres, but there is no evidence that vagal parasympathetic components relay in the inferior ganglion. Preganglionic motor fibres from the dorsal vagal nucleus and the special visceral efferents from the nucleus ambiguus, which descend to the inferior vagal ganglion, commonly form a visible band, skirting the ganglion in some mammals. These larger fibres probably provide motor innervation to the larynx in the recurrent laryngeal nerve, together with some contribution to the superior laryngeal nerve supplying cricothyroid.
BRANCHES IN THE NECK The branches of the vagus in the neck are the meningeal, auricular, pharyngeal, carotid body, superior and recurrent laryngeal nerves and cardiac branches. Meningeal branch(es) Meningeal branches appear to start from the superior vagal ganglion and pass through the jugular foramen to be distributed to the dura mater in the posterior cranial fossa. Auricular branch The auricular branch arises from the superior vagal ganglion and is joined by a branch from the inferior ganglion of the glossopharyngeal nerve. It passes behind the internal jugular vein and enters the mastoid canaliculus on the lateral wall of the jugular fossa. Traversing the temporal bone, it crosses the facial canal c.4 mm above the stylomastoid foramen and here supplies an ascending branch to the facial nerve. Fibres of the nervus intermedius may pass to the auricular branch at this point, which may explain the cutaneous vesiculation in the auricle that sometimes accompanies geniculate herpes. The auricular branch then traverses the tympanomastoid fissure, and divides into two rami. One ramus joins the posterior auricular nerve and the other is distributed to the skin of part of the ear and to the external acoustic meatus. Pharyngeal branch The pharyngeal branch of the vagus is the main motor nerve of the pharynx. It emerges from the upper part of the inferior vagal ganglion and consists chiefly of filaments from the cranial accessory nerve. It passes between the external and internal carotid arteries to the upper border of the middle pharyngeal constrictor, and divides into numerous filaments which join rami of the sympathetic trunk and glossopharyngeal nerve to form a pharyngeal plexus. A minute filament, the ramus lingualis vagi, joins the hypoglossal nerve as it curves round the occipital artery. Branches to the carotid body Branches to the carotid body are variable in number. They may arise from the
inferior ganglion or travel in the pharyngeal branch, and sometimes in the superior laryngeal nerve. They form a plexus with the glossopharyngeal rami and branches of the cervical sympathetic trunk. Superior laryngeal nerve The superior laryngeal nerve is larger than the pharyngeal branch, and issues from the middle of the inferior vagal ganglion. It receives a branch from the superior cervical sympathetic ganglion and descends alongside the pharynx, at first posterior, then medial, to the internal carotid artery, and divides into the internal and external laryngeal nerves. The internal laryngeal nerve is sensory to the laryngeal mucosa down to the level of the vocal folds. It also carries afferent fibres from the laryngeal neuromuscular spindles and other stretch receptors. It descends to the thyrohyoid membrane, pierces it above the superior laryngeal artery and divides into an upper and lower branch. The upper branch is horizontal and supplies the mucosa of the pharynx, the epiglottis, vallecula and laryngeal vestibule. The lower branch descends in the medial wall of the piriform recess, supplies the aryepiglottic fold, the mucosa on the back of the arytenoid cartilage and one or two branches to transverse arytenoid (the latter unite with twigs from the recurrent laryngeal to supply the same muscle). The internal laryngeal nerve ends by piercing the inferior pharyngeal constrictor to unite with an ascending branch from the recurrent laryngeal nerve. As it ascends in the neck it supplies branches, more numerous on the left, to the mucosa and tunica muscularis of the oesophagus and trachea and to the inferior constrictor. The external laryngeal nerve, smaller than the internal, descends behind sternohyoid with the superior thyroid artery, but on a deeper plane. It lies first on the inferior pharyngeal constrictor, then pierces it to curve round the inferior thyroid tubercle and reach cricothyroid, which it supplies. The nerve also gives branches to the pharyngeal plexus and the inferior constrictor. Behind the common carotid artery, the external laryngeal nerve communicates with the superior cardiac nerve and superior cervical sympathetic ganglion. Recurrent laryngeal nerve
page 557 page 558
The recurrent laryngeal nerve differs, in origin and course, on the two sides. On the right it arises from the vagus anterior to the first part of the subclavian artery, and curves backwards below and then behind it to ascend obliquely to the side of the trachea behind the common carotid artery. Near the lower pole of the lateral lobe of the thyroid gland it is closely related to the inferior thyroid artery, and crosses either in front of, behind, or between, its branches. On the left, the nerve arises from the vagus on the left of the aortic arch, curves below it immediately behind the attachment of the ligamentum arteriosum to the concavity of the aortic arch (Fig. 30.31) and ascends to the side of the trachea. As the recurrent laryngeal nerve curves round the subclavian artery, or the aortic arch, it gives cardiac filaments to the deep cardiac plexus. On both sides the recurrent laryngeal nerve ascends in or near a groove between the trachea and oesophagus. It is closely related to the medial surface of the thyroid gland before it passes under the lower border of the inferior constrictor, and it enters the larynx behind the articulation of the inferior thyroid cornu with the cricoid cartilage. The recurrent laryngeal nerve supplies all laryngeal muscles, except the cricothyroid, and it communicates with the internal laryngeal nerve, supplying sensory filaments to the laryngeal mucosa below the vocal folds. It also carries afferent fibres from laryngeal stretch receptors. The recurrent laryngeal nerve is described further with the larynx (p. 644).
ACCESSORY NERVE (Figs 31.20, 31.21, 31.23) The accessory nerve is conventionally described as a single entity, even though
its two components, which join for a relatively short part of its course, are of quite separate origin.
CRANIAL ROOT The cranial root of the accessory nerve is smaller than the spinal root. It exits the skull through the jugular foramen, and unites for a short distance with the spinal root. It is also connected to the superior vagal ganglion. After traversing the foramen, the cranial root separates from the spinal part and immediately joins the vagus nerve superior to the inferior vagal ganglion. Those of its fibres that are distributed in the pharyngeal branches of the vagus are derived from the nucleus ambiguus, and probably innervate the pharyngeal and palatal muscles except tensor veli palatini. Other fibres enter the recurrent laryngeal nerve to supply the adductor muscles of the vocal cords, thyroarytenoid and lateral cricoarytenoid.
SPINAL ROOT The spinal root arises from an elongated nucleus of motor cells situated in the lateral aspect of the ventral horn which extends from the junction of the spinal cord and medulla to the sixth cervical segment (Fig. 31.20). Some rootlets emerge directly, others turn cranially before exiting. Their line of exit is irregular rather than linear, and the spinal root usually passes through the first cervical dorsal root ganglion. The rootlets form a trunk which ascends between the ligamentum denticulatum and the dorsal roots of the spinal nerves and enters the skull via the foramen magnum, behind the vertebral artery. It then turns upwards and passes laterally to reach the jugular foramen, which it traverses in a common dural sheath with the vagus, but separated from that nerve by a fold of arachnoid mater. As the spinal root exits the jugular foramen it runs posterolaterally and passes either medial or lateral to the internal jugular vein. Occasionally it passes through the vein. The nerve then crosses the transverse process of the atlas and is itself crossed by the occipital artery. It descends obliquely, medial to the styloid process, stylohyoid and the posterior belly of the digastric. Running with the superior sternocleidomastoid branch of the occipital artery, it reaches the upper part of sternocleidomastoid and enters its deep surface, to form an anastomosis with fibres from C2 alone, C3 alone, or C2 and C3, the ansa of Maubrac. The nerve occasionally terminates in the muscle. More commonly it emerges a little above the midpoint of the posterior border of sternocleidomastoid, generally above the emergence of the great auricular nerve (usually within 2 cm of it) and between 4-6 cm from the tip of the mastoid process. However, the point of emergence is very variable. It crosses the posterior triangle on levator scapulae (Fig. 31.21), separated from it by the prevertebral layer of deep cervical fascia and adipose tissue. Here the nerve is relatively superficial and related to the superficial cervical lymph nodes. About 3-5 cm above the clavicle it passes behind the anterior border of trapezius, often dividing to form a plexus on its deep surface which receives contributions from C3 and C4, or C4 alone. It then enters the deep surface of the muscle. The cervical course of the nerve follows a line from the lower anterior part of the tragus to the tip of the transverse process of the atlas and then across the sternocleidomastoid and the posterior triangle to a point on the anterior border of the trapezius 3-5 cm above the clavicle. Conventionally, the spinal root is thought to provide the sole motor supply to sternocleidomastoid, and the second and third cervical nerves are believed to carry proprioceptive fibres from it. The supranuclear pathway of fibres destined for sternocleidomastoid is not simple: fibres may undergo a double decussation in the brainstem and/or there may be a bilateral projection to the muscle from each hemisphere. The motor supply to the upper and middle portions of trapezius is primarily from the spinal accessory nerve. However, the lower two-thirds of the muscle, in c.75% of subjects, receives an innervation from the cervical plexus. On the basis of the
incomplete denervation of the muscle that sometimes occurs following sacrifice of both the accessory nerve and the cervical plexus, it has been suggested that the trapezius receives a partial motor supply from other sources, possibly via thoracic roots. In addition to their motor contribution, C3 and 4 carry proprioceptive fibres from trapezius. In c.25% of subjects the spinal accessory nerve receives no fibres from the cervical plexus. Sensory ganglia have been described along the course of the spinal root. UPDATE Date Added: 21 February 2006 Publication Services, Inc. Abstract: Surgical anatomy of the spinal accessory nerve in the posterior triangle of the neck Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=16024309&query_hl=10&itool=pubmed_docsum Surgical anatomy of the spinal accessory nerve in the posterior triangle of the neck. Aramrattana A, Sittitrai P, Harnsiriwattanagit K: Asian J Surg 28(3):171-173, 2005.
LESIONS AFFECTING THE ACCESSORY NERVE Lesions of the accessory nerve may occur centrally; at its exit from the skull; in the neck. The supranuclear fibres which influence motor neurones innervating sternocleidomastoid decussate twice, therefore a lesion of the pyramidal system above the pons produces weakness of the ipsilateral sternocleidomastoid and contralateral trapezius. Episodic contraction of sternocleidomastoid and trapezius, often accompanied by contraction of other muscle groups, e.g. splenius capitis, occurs in spasmodic torticollis, a focal dystonia. In jugular foramen syndrome, caused by pathologies including nasopharyngeal carcinoma or a glomus tumour, lesions of the glossopharyngeal, vagus and accessory nerves coexist. The accessory nerve can be injured more distally in the neck by trauma or by surgical exploration in the posterior triangle. If the accessory nerve is sacrificed as part of a radical neck dissection, and the innervation of trapezius is lost, the patient develops intractable neuralgia due to traction on the brachial plexus caused by the unsupported weight of the shoulder and arm. UPDATE Date Added: 21 March 2006 Nancy Milligan, MSc (Dianthus Medical Limited) Update: Mapping the surface anatomy of the spinal accessory nerve in the posterior triangle Iatrogenic injury of the spinal accessory nerve (usually via accidental damage during minor neck surgery) is the most common cause of accessory nerve palsy. Although rare, this kind of damage can cause extensive morbidity, including numbness, paralysis, pain, and winging of the scapula. It can also have medicolegal implications. In one study attempting to create a map to define the surface anatomy of the spinal accessory nerve in the area of the posterior triangle, bilateral neck dissections to expose the spinal accessory nerve were done on 25 cadavers (age range 31-93 years). Differences in the course and distribution of the spinal accessory nerve were recorded, including its relationship to the borders of the sternocleidomastoid and the trapezius. The researchers found considerable variation in the surface and regional anatomy of the nerve. The mean nerve length from the tip of the mastoid process to the point of emergence from the posterior border of sternocleidomastoid was 6.13 ± 1.43 cm standard deviation (SD), with a range of 3.1-9.8 cm. The mean ratio of the nerve length to the length of the posterior border of the sternocleidomastoid was 0.36 ± 0.08 SD, with a range of 0.22-0.55 (95% confidence interval [95% CI] 0.34-0.38). The mean distance from the clavicular insertion of the anterior border of the trapezius to the point of disappearance of the nerve behind the same border was 4.08 ± 1.47 cm SD, with a range of 1.3-8.1 cm. The mean ratio of this distance to the length of the anterior border of the trapezius from the clavicle to its insertion at the superior nuchal line was 0.26 ± 0.09 SD, with a range of 0.08-0.49 (95% CI 0.24-0.28).
Variation in the way the nerve crossed the posterior triangle was also found. In 74% of dissections the nerve crossed in a straight line, whereas in 26%, the nerve looped and twisted within the boundaries of the posterior triangle. In addition, the nerve gave varying numbers of branches to the trapezius: in 16% of dissections there were no branches, in 18% one branch, in 28% two branches, in 22% three branches, in 14% four branches, and in 2% one branch. Considerable variation was also found in the contribution the spinal accessory nerve received from the cervical plexus; in 36% of dissections it received no branches from the cervical plexus, whereas in 64% it received one or more nerve branches. This study revealed considerable intra- and inter-individual variations in the anatomy and course of the spinal accessory nerve, and many of the values were different from those reported in previous studies. The authors conclude that mapping the surface anatomy of the nerve in order to produce a tool for minor surgical procedures is unrealistic, and therefore caution is still needed during operations in the area to avoid damage to the nerve. Symes A, Ellis H: Variations in the surface anatomy of the spinal accessory nerve in the posterior triangle. Surg Radiol Anat 27:404-408, 2005. Medline Similar articles
HYPOGLOSSAL NERVE (Figs 31.2, 31.14, 31.16, 31.21) The hypoglossal nerve is motor to all the muscles of the tongue, except palatoglossus. The hypoglossal rootlets run laterally behind the vertebral artery, collected into two bundles which perforate the dura mater separately opposite the hypoglossal canal in the occipital bone, and unite after traversing it. The canal is sometimes divided by a spicule of bone. The nerve emerges from the canal in a plane medial to the internal jugular vein, internal carotid artery, ninth, tenth and eleventh cranial nerves, and passes inferolaterally behind the internal carotid artery and glossopharyngeal and vagus nerves to the interval between the artery and the internal jugular vein. Here it makes a half-spiral turn round the inferior vagal ganglion, and is united with it by connective tissue. It then descends almost vertically between the vessels and anterior to the vagus to a point level with the angle of the mandible, becoming superficial below the posterior belly of digastric and emerging between the internal jugular vein and internal carotid artery. It loops round the inferior sternocleidomastoid branch of the occipital artery, crosses lateral to both internal and external carotid arteries and the loop of the lingual artery a little above the tip of the greater cornu of the hyoid, and is itself crossed by the facial vein. Its course is described further on page 588.
COMMUNICATIONS The hypoglossal nerve communicates with the sympathetic trunk, vagus, first and second cervical nerves and lingual nerve. Near the atlas it is joined by branches from the superior cervical sympathetic ganglion and by a filament from the loop between the first and second cervical nerves which leaves the hypoglossal as the upper root of the ansa cervicalis (Fig. 31.2). The vagal connections occur close to the skull, and numerous filaments pass between the hypoglossal nerve and the inferior vagal ganglion in the connective tissue uniting them. As the hypoglossal nerve curves round the occipital artery it receives the ramus lingualis vagi from the pharyngeal plexus. Near the anterior border of hyoglossus it is connected with the lingual nerve by many filaments which ascend on the muscle.
BRANCHES OF DISTRIBUTION The branches of distribution of the hypoglossal nerve are meningeal, descending, thyrohyoid and muscular nerves. Meningeal branches Meningeal branches leave the nerve in the hypoglossal canal and return through it to supply the diploë of the occipital bone, the dural walls of the occipital and inferior petrosal sinuses and much of the floor of the anterior wall of the posterior cranial fossa, probably through pathways other than that of the hypoglossal
nerve, e.g. upper cervical spinal nerves. Descending branch
page 558 page 559
The descending branch (descendens hypoglossi) contains fibres from the first cervical spinal nerve. It leaves the hypoglossal nerve when it curves around the occipital artery, and runs down on the carotid sheath. It provides a branch to the superior belly of omohyoid before joining with the descendens cervicalis to form the ansa cervicalis (Fig. 31.2). Nerves to thyrohyoid and geniohyoid The nerves to thyrohyoid and geniohyoid arise near the posterior border of hyoglossus. They represent the remaining fibres from the first cervical spinal nerves.
LESIONS OF THE HYPOGLOSSAL NERVE The hypoglossal nerve may be damaged during neck dissection. Complete hypoglossal division causes unilateral lingual paralysis and eventual hemiatrophy; the protruded tongue deviates to the paralysed side, and, on retraction, the wasted and paralysed side rises higher than the unaffected side. The larynx may deviate towards the active side in swallowing, due to unilateral paralysis of the hyoid depressors associated with loss of the first cervical spinal nerve which runs with the hypoglossal nerve. If paralysis is bilateral, the tongue is motionless. Taste and tactile sensibility are unaffected but articulation is slow and swallowing very difficult.
Cervical sympathetic trunk (Figs 31.6, 31.7) The cervical sympathetic trunk lies on the prevertebral fascia behind the carotid sheath and contains three interconnected ganglia, the superior, middle and inferior (stellate or cervicothoracic). However there may occasionally be two or four ganglia. The cervical sympathetic ganglia send grey rami communicantes to all the cervical spinal nerves but receive no white rami communicantes from them. Their spinal preganglionic fibres emerge in the white rami communicantes of the upper five thoracic spinal nerves (mainly the upper three), and ascend in the sympathetic trunk to synapse in the cervical ganglia. In their course, the grey rami communicantes may pierce longus capitis or scalenus anterior.
SUPERIOR CERVICAL GANGLION The superior cervical ganglion is the largest of the three ganglia. It lies on the transverse processes of the second and third cervical vertebrae and is probably formed from four fused ganglia judging by its grey rami to C1-4. The internal carotid artery within the carotid sheath is anterior, and longus capitis is posterior (Fig. 31.6). The lower end of the ganglion is united by a connecting trunk to the middle cervical ganglion. Postganglionic branches are distributed in the internal carotid nerve, which ascends with the internal carotid artery into the carotid canal to enter the cranial cavity, and in lateral, medial and anterior branches. They supply vasoconstrictor and sudomotor nerves to the face and neck, dilator pupillae and smooth muscle in the eyelids and orbitalis. Lateral branches The lateral branches are grey rami communicantes to the upper four cervical spinal nerves and to some of the cranial nerves. Branches pass to the inferior vagal ganglion, the hypoglossal nerve, the superior jugular bulb and associated jugular glomus or glomera, and to the meninges in the posterior cranial fossa. Another branch, the jugular nerve, ascends to the cranial base and divides into two; one part joins the inferior glossopharyngeal ganglion and the other joins the superior vagal ganglion. Medial branches
The medial branches of the superior cervical ganglion are the laryngopharyngeal and cardiac. The laryngopharyngeal branches supply the carotid body and pass to the side of the pharynx, joining glossopharyngeal and vagal rami to form the pharyngeal plexus. A cardiac branch arises by two or more filaments from the lower part of the superior cervical ganglion and occasionally receives a twig from the trunk between the superior and middle cervical ganglia. It is thought to contain only efferent fibres, the preganglionic outflow being from the upper thoracic segments of the spinal cord, and to be devoid of pain fibres from the heart. It descends behind the common carotid artery, in front of longus colli, and crosses anterior to the inferior thyroid artery and recurrent laryngeal nerve. The courses on the two sides then differ. The right cardiac branch usually passes behind, but sometimes in front of, the subclavian artery and runs posterolateral to the brachiocephalic trunk to join the deep (dorsal) part of the cardiac plexus behind the aortic arch. It has other sympathetic connections. About midneck it receives filaments from the external laryngeal nerve. Inferiorly, one or two vagal cardiac branches join it. As it enters the thorax it is joined by a filament from the recurrent laryngeal nerve. Filaments from the nerve also communicate with the thyroid branches of the middle cervical ganglion. The left cardiac branch, in the thorax, is anterior to the left common carotid artery and crosses in front of the left side of the aortic arch to reach the superficial (ventral) part of the cardiac plexus. Sometimes it descends on the right of the aorta to end in the deep (dorsal) part of the cardiac plexus. It communicates with the cardiac branches of the middle and inferior cervical sympathetic ganglia and sometimes with the inferior cervical cardiac branches of the left vagus, and branches from these mixed nerves form a plexus on the ascending aorta. Anterior branches The anterior branches of the superior cervical ganglion ramify on the common and external carotid arteries and the branches of the external carotid, and form a delicate plexus around each in which small ganglia are occasionally found. The plexus around the facial artery supplies a filament to the submandibular ganglion; the plexus on the middle meningeal artery sends one ramus to the otic ganglion and another, the external petrosal nerve, to the facial ganglion. Many of the fibres coursing along the external carotid and its branches ultimately leave them to travel to facial sweat glands via branches of the trigeminal nerve.
MIDDLE CERVICAL GANGLION (Figs 31.24, 31.25) The middle cervical ganglion is the smallest of the three, and is occasionally absent, in which case it may be replaced by minute ganglia in the sympathetic trunk or may be fused with the superior ganglion. It is usually found at the level of the sixth cervical vertebra, anterior or just superior to the inferior thyroid artery, or it may adjoin the inferior cervical ganglion. It probably represents a coalescence of the ganglia of the fifth and sixth cervical segments, judging by its postganglionic rami, which join the fifth and sixth cervical spinal nerves (but sometimes also the fourth and seventh). It is connected to the inferior cervical ganglion by two or more very variable cords. The posterior cord usually splits to enclose the vertebral artery, while the anterior cord loops down anterior to, and then below, the first part of the subclavian artery, medial to the origin of its internal thoracic branch, and supplies rami to it. This loop is the ansa subclavia and is frequently multiple, lies closely in contact with the cervical pleura and typically connects with the phrenic nerve, and sometimes with the vagus.
page 559 page 560
Figure 31.24 The middle and inferior cervical ganglia of the right side, viewed from the right. Note the proximity of the inferior cervical and first thoracic ganglia, which often fuse to form a cervicothoracic (stellate) ganglion.
Figure 31.25 The middle and inferior cervical ganglia of the right side, anterior view. Part of the vertebral artery has been excised to show the inferior cervical ganglion.
The middle cervical ganglion gives off thyroid and cardiac branches. The thyroid branches accompany the inferior thyroid artery to the thyroid gland. They communicate with the superior cardiac, external laryngeal and recurrent laryngeal nerves, and send branches to the parathyroid glands. Fibres to both glands are largely vasomotor but some reach the secretory cells. The cardiac branch, the largest sympathetic cardiac nerve, either arises from the ganglion itself or more often from the sympathetic trunk cranial or caudal to it. On the right side it
descends behind the common carotid artery, in front of or behind the subclavian artery, to the trachea where it receives a few filaments from the recurrent laryngeal nerve before joining the right half of the deep (dorsal) part of the cardiac plexus. In the neck, it connects with the superior cardiac and recurrent laryngeal nerves. On the left side the cardiac nerve enters the thorax between the left common carotid and subclavian arteries to join the left half of the deep (dorsal) part of the cardiac plexus. Fine branches from the middle cervical ganglion also pass to the trachea and oesophagus.
INFERIOR (OR CERVICOTHORACIC/STELLATE) GANGLION (Figs 31.24, 31.25) The inferior cervical ganglion (cervicothoracic/stellate) is irregular in shape and much larger than the middle cervical ganglion. It is probably formed by a fusion of the lower two cervical and first thoracic segmental ganglia, sometimes including the second and even third and fourth thoracic ganglia. The first thoracic ganglion may be separate, leaving an inferior cervical ganglion above it. The sympathetic trunk turns backwards at the junction of the neck and thorax and so the long axis of the cervicothoracic ganglion becomes almost anteroposterior. The ganglion lies on or just lateral to the lateral border of longus colli between the base of the seventh cervical transverse process and the neck of the first rib (which are both posterior to it). The vertebral vessels are anterior, and the ganglion is separated from the posterior aspect of the cervical pleura inferiorly by the suprapleural membrane. The costocervical trunk of the subclavian artery branches near the lower pole of the ganglion, and the superior intercostal artery is lateral. A small vertebral ganglion may be present on the sympathetic trunk anterior or anteromedial to the origin of the vertebral artery and directly above the subclavian artery. When present, it may provide the ansa subclavia and is also joined to the inferior cervical ganglion by fibres enclosing the vertebral artery. It is usually regarded as a detached part of the middle cervical or inferior cervical ganglion. Like the middle cervical ganglion it may supply grey rami communicantes to the fourth and fifth cervical spinal nerves. The inferior cervical ganglion sends grey rami communicantes to the seventh and eighth cervical and first thoracic spinal nerves, and gives off a cardiac branch, branches to nearby vessels and sometimes a branch to the vagus nerve. The grey rami communicantes to the seventh cervical spinal nerve vary from one to five (two being the usual number). A third often ascends medial to the vertebral artery in front of the seventh cervical transverse process. It connects with the seventh cervical nerve, and sends a filament upwards through the sixth cervical transverse foramen in company with the vertebral vessels to join the sixth cervical spinal nerve as it emerges from the intervertebral foramen. An inconstant ramus may traverse the seventh cervical transverse foramen. Grey rami to the eighth cervical spinal nerve vary from three to six in number. The cardiac branch descends behind the subclavian artery and along the front of the trachea to the deep cardiac plexus. Behind the artery it connects with the recurrent laryngeal nerve and the cardiac branch of the middle cervical ganglion (the latter is often replaced by fine branches of the inferior cervical ganglion and ansa subclavia). The branches to blood vessels form plexuses on the subclavian artery and its branches. The subclavian supply is derived from the inferior cervical ganglion and ansa subclavia, and typically extends to the first part of the axillary artery, although a few fibres may extend further. An extension of the subclavian plexus to the internal thoracic artery may be joined by a branch of the phrenic nerve. The vertebral plexus is derived mainly from a large branch of the inferior cervical ganglion that ascends behind the vertebral artery to the sixth transverse foramen. Here it is reinforced by branches of the vertebral ganglion or the cervical
sympathetic trunk that pass cranially on the ventral aspect of the artery. Deep rami communicantes from this plexus join the ventral rami of the upper five or six cervical spinal nerves. The plexus, which contains some neuronal cell bodies, continues into the skull along the vertebral and basilar arteries and their branches as far as the posterior cerebral artery, where it meets a plexus from the internal carotid artery. The plexus on the inferior thyroid artery reaches the thyroid gland, and connects with the recurrent and external laryngeal nerves, the cardiac branch of the superior cervical ganglion, and the common carotid plexus.
HORNER'S SYNDROME (Fig. 31.26) Any condition or injury that destroys the sympathetic trunk ascending from the thorax through the neck into the face results in Horner's syndrome, characterized by a drooping eyelid (ptosis), sunken globe (enophthalmos), narrow palpebral fissure, contracted pupil (meiosis), vasodilatation and lack of thermal sweating (anhydrosis) on the affected side. This occurs classically when a bronchial carcinoma invades the sympathetic trunk. It also occurs as a complication of cervical sympathectomy or a radical neck dissection.
© 2008 Elsevier
VISCERA The main viscera to be found in the neck are the submandibular salivary glands, the thyroid and parathyroid glands and the upper components of the respiratory and alimentary systems.
Submandibular salivary gland Each submandibular salivary gland is situated behind and below the ramus of the mandible, in the region of the submandibular triangle, between the anterior and posterior bellies of digastric. The gland is described in detail on page 602.
Thyroid gland (Fig. 31.27) page 560 page 561
Figure 31.26 Horner's syndrome.
The thyroid gland, brownish-red and highly vascular, is placed anteriorly in the lower neck, level with the fifth cervical to the first thoracic vertebrae. It is ensheathed by the pretracheal layer of deep cervical fascia and consists of right and left lobes connected by a narrow, median isthmus. Its weight is usually c.25g, but this varies. The gland is slightly heavier in females, and enlarges during menstruation and pregnancy. Estimation of the size of the thyroid gland is clinically important in the evaluation and management of thyroid disorders and can be achieved non-invasively by means of diagnostic ultrasound. No significant difference in thyroid gland volume has been observed between males and females from 8 months to 15 years. The lobes of the thyroid gland are approximately conical. Their ascending apices diverge laterally to the level of the oblique lines on the laminae of the thyroid
cartilage, and their bases are level with the fourth or fifth tracheal cartilages. Each lobe is c.5 cm long, its greatest transverse and anteroposterior extents being c.3 cm and 2 cm respectively. The posteromedial aspects of the lobes are attached to the side of the cricoid cartilage by a lateral thyroid ligament. The isthmus connects the lower parts of the two lobes, although occasionally it may be absent. It measures c.1.25 cm transversely and vertically, and is usually anterior to the second and third tracheal cartilages, though often higher or sometimes lower because its site and size vary greatly. A conical pyramidal lobe often ascends towards the hyoid bone from the isthmus or the adjacent part of either lobe (more often the left). It is occasionally detached or in two or more parts. A fibrous or fibromuscular band, the levator of the thyroid gland, musculus levator glandulae thyroideae, sometimes descends from the body of the hyoid to the isthmus or pyramidal lobe. Small detached masses of thyroid tissue may occur above the lobes or isthmus as accessory thyroid glands. Vestiges of the thyroglossal duct may persist between the isthmus and the foramen caecum of the tongue, sometimes as accessory nodules or cysts of thyroid tissue near the midline or even in the tongue.
SURFACES AND RELATIONS (Fig. 31.15) The convex lateral (superficial) surface is covered by sternothyroid, whose attachment to the oblique thyroid line prevents the upper pole of the gland from extending on to thyrohyoid. More anteriorly lie sternohyoid and the superior belly of omohyoid, overlapped inferiorly by the anterior border of sternocleidomastoid. The medial surface of the gland is adapted to the larynx and trachea, contacting at its superior pole the inferior pharyngeal constrictor and the posterior part of cricothyroid, which separate it from the posterior part of the thyroid lamina and the side of the cricoid cartilage. The external laryngeal nerve is medial to this part of the gland as it passes to supply cricothyroid. Inferiorly, the trachea and, more posteriorly, the recurrent laryngeal nerve and oesophagus (which is closer on the left) are medial relations. The posterolateral surface of the thyroid gland is close to the carotid sheath, and overlaps the common carotid artery. The thin anterior border of the gland, near the anterior branch of the superior thyroid artery, slants down medially. The rounded posterior border is related below to the inferior thyroid artery and its anastomosis with the posterior branch of the superior thyroid artery. The parathyroid glands are usually related to this border. The lower end of the posterior border on the left side lies near the thoracic duct. The isthmus is covered by sternothyoid, from which it is separated by pretracheal fascia. More superficially it is covered by sternohyoid, the anterior jugular veins, fascia and skin. The superior thyroid arteries anastomose along its upper border and the inferior thyroid veins leave the gland at its lower border. page 561 page 562
page 562 page 563
Figure 31.27 The thyroid and parathyroid glands and their roles in the control of calcium metabolism.
VASCULAR SUPPLY AND LYMPHATIC DRAINAGE (Figs 31.15, 31.25, 31.27) The thyroid gland is supplied by the superior and inferior thyroid arteries and sometimes by an arteria thyroidea ima from the brachiocephalic trunk or aortic arch. The arteries are large and their branches anastomose frequently on and in the gland, both ipsilaterally and contralaterally. The superior thyroid artery pierces the thyroid fascia and then divides into anterior and posterior branches. The anterior branch supplies the anterior surface of the gland, the posterior branch supplies the lateral and medial surfaces. The superior thyroid artery is closely related to the external laryngeal nerve. The inferior thyroid artery approaches the base of the thyroid gland and divides into superior (ascending) and inferior thyroid branches which supply the inferior and posterior surfaces of the gland. The superior branch also supplies the parathyroid glands. The relationship between the inferior thyroid artery and the recurrent laryngeal nerve has clinical importance. UPDATE Date Added: 23 December 2005 Publication Services, Inc. Update: A meta-analysis of inferior thyroid artery variations in different human ethnic groups and their clinical implications. This meta-analysis was performed on library and Medline-selected publications. A total of 6,285 Caucasoid and 847 East Asian items were analyzed. The authors evaluated: (a) whether presence, numerical variations, and site of origin of the inferior thyroid artery (ITA) were influenced by ethnic group and gender; (b) whether and which neck side had the largest arterial caliber; (c) whether there were differences between the presence of the ITA and the superior thyroid artery (STA); (d) to what extent non-selective thyroid angiography was effective in visualizing the ITA; and (e) the clinical implications of these findings. The analysis revealed that the ITA was absent more frequently in East Asian subjects than in Caucasoids; it was more common in East Asian males than females. Non-selective thyroid angiography showed higher numbers for the ITA than the STA in Caucasoids. In Caucasoids, the ITA was more likely to originate from the subclavian artery, whereas in East Asian subjects, it arose more often from the thyrocervical trunk. In both Caucasoids and East Asians, the ITA was more frequently present on the right side than on the left side. The structural differences should be noted during surgical localization of inferior parthyroid adenomas, especially mediastinal tumors. The presence of the ITA should be ascertained by imaging techniques before raising an infrahyoid muscular flap to reconstruct laryngeal defects. Specific attention should be paid to the arrangement of the thyroid arteries during surgery involving the cervical sympathetic chain. Toni R, Della Casa C, Castorina S et al: A meta-analysis of inferior thyroid artery variations in different human ethnic groups and their clinical implications. Ann Anat 187(2):371-385, 2005.
The venous drainage of the thyroid gland is usually via superior, middle, and inferior thyroid veins. The superior thyroid vein emerges from the upper part of the gland and runs with the superior thyroid artery towards the carotid sheath. It drains into the internal jugular vein. The middle thyroid vein collects blood from the lower part of the gland. It emerges from the lateral surface of the gland and drains into the internal jugular vein. The inferior thyroid vein forms a plexus with the vein on the opposite side. This plexus is located below the thyroid gland and in front of the trachea. From the plexus, the left vein descends into the thorax to terminate at the left brachiocephalic vein. The right inferior thyroid vein drains into the right brachiocephalic vein. Alternatively, there may be a common trunk draining into the left brachiocephalic vein. UPDATE Date Added: 03 January 2006 Publication Services, Inc. Update: Superior thyroid artery variations in different human groups and their clinical implications. A meta-analysis of whether: (a) the presence, numerical variations, and site of origin of the superior thyroid artery (STA) are influenced by ethnic group and gender; (b) the origin and caliber of the STA are symmetrical; (c) the extent to which a non-selective thyroid angiography is effective in visualizing the STA; or (d) the clinical implications of these findings. The investigation covered 24 library and MEDLINE-selected publications,
analyzing a total of 3,453 Caucasoid and 931 East Asian items. Effectiveness of non-selective thyroid angiography was determined by using sensitivity, specificity, and positive and negative predictive values. It was found that in Caucasoids, the STA was more likely to originate from the external carotid artery, whereas in East Asians, it was more likely to arise from the common carotid artery. No gender differences were observed. There was an equal probability of either asymmetrical or symmetrical origin on the two sides of the neck for the STA in East Asians. The caliber of the STA was symmetrical in Caucasoids. Non-selective thyroid angiography, either conventional or by digital subtraction, was only moderately effective in visualizing the STA in Caucasoids. Toni R, Della Casa C, Castorina S et al: A meta-analysis of superior thyroid artery variations in different human groups and their clinical implications. Ann Anat 186:255-262, 2004. Medline Similar articles
Thyroid lymphatic vessels communicate with the tracheal plexus, and pass to the prelaryngeal nodes just above the thyroid isthmus and to the pretracheal and paratracheal nodes; some may also drain into the brachiocephalic nodes related to the thymus in the superior mediastinum. Laterally the gland is drained by vessels lying along the superior thyroid veins to the deep cervical nodes. Thyroid lymphatics may drain directly, with no intervening node, to the thoracic duct.
INNERVATION The thyroid gland receives its innervation from the superior, middle and inferior cervical sympathetic ganglia.
IMAGING The follicular nature of the thyroid gland is not resolved by current imaging techniques and thus presents a homogeneous texture on cross-sectional imaging (US, CT, MRI). Its superficial location makes the thyroid an ideal organ for sonographic examination (Fig. 31.28). The thyroid gland is highly vascular and demonstrates intense contrast enhancement and increased signal on T 2weighted MRI (Fig. 31.29). Radionuclide imaging of the thyroid may be performed with technetium (Tc99m ) pertechnetate. This readily available radionuclide is trapped by the thyroid in the same way as iodine, but is not organified. It yields morphological information and will reveal the presence of ectopic thyroid tissue. Functional data can be obtained with the use of 131 iodine which is trapped and organified.
MICROSTRUCTURE The thyroid gland has a thin capsule of connective tissue, which extends into the glandular parenchyma and divides each lobe into irregularly shaped and sized lobules. The functional units of the thyroid are follicles, which are spherical and cyst-like, between 0.02 and 0.9 mm in diameter, and consist of a central colloid core surrounded by a single-layered epithelium resting on a basal lamina (Fig. 31.30). Colloid consists almost entirely of an iodinated glycoprotein, iodothyroglobulin. This is the inactive, stored form of the active thyroid hormones, tri-iodothyronine (T 3) and tetraiodothyronine or thyroxine (T 4), and is produced by the follicular epithelial cells. Sufficient iodothyroglobulin is stored extracellularly within follicles to regulate the metabolic activity of the body for up to three months. Follicles are surrounded by a delicate connective tissue stroma, containing dense plexuses of fenestrated capillaries, extensive lymphatic networks and sympathetic nerve fibres which supply the arterioles and capillaries. Some nerve fibres end close to the follicular epithelial cells.
Figure 31.28 Thyroid sonogram.
Figure 31.29 T2 -weighted MRI demonstrating high vascularity of the thyroid gland.
Follicular cells
page 563 page 564
Figure 31.30 A section through parts of two adjacent thyroid follicles, showing the follicular epithelium enclosing a colloid-filled lumen. Pale cavities in the otherwise homogeneous, eosinophilic colloid are sites of thyroglobulin resorption. Calcitoninsecreting C cells are present but not readily identifiable in routine preparations. A capillary network surrounds the follicles within connective tissue septa. (By permission from Stevens A, Lowe JS 1996 Human Histology, 2nd edn. London: Mosby.)
Follicular cells vary from squamous or low cuboidal to columnar, depending on their level of activity, which is controlled mainly by circulating hypophyseal thyroid-stimulating hormone (TSH, thyrotropin). Resting follicles are large, lined by squamous or low cuboidal epithelium with abundant luminal colloid. Active follicular cells are highly polarized functionally. Synthesis and exocytosis of thyroglobulin occurs at the apices of the cells, and thyroglobulin endocytosis (from stored colloid), lysosomal degradation and liberation of thyroid hormones (T 3 and T 4) occur basally. Follicles showing differing levels of activity may coexist. The secretion of TSH leads to endocytosis of colloidal droplets at the luminal epithelium (Fig. 31.30). Prolonged high levels of circulating TSH induce follicular cell hypertrophy, with progressive resorption of colloid and increased stromal vascularity. Apical microvilli are short in resting cells but elongate and often branch on stimulation by TSH, which also provokes extension of cytoplasmic processes into the luminal colloid. The processes fuse around portions of the colloid and take it into the cell. After colloid endocytosis, lysosomes migrate towards the lumen to fuse with the intracellular droplets of colloid, forming secondary lysosomes. During this period the cytoplasmic colloid gradually disappears as lysosomal acid proteases degrade the iodinated thyroglobulin, releasing the thyroid hormones T 3 and T 4, which pass basally for release, leaving the gland mainly via the blood capillaries and lymphatics. C cells Thyroid parenchyma also contains C (clear) cells, so-called from their palestaining cytoplasm, which is more pronounced in some species than in the human thyroid. They belong to the amine precursor uptake and decarboxylation (APUD) system of dispersed neuroendocrine cells (p. 180), and produce the peptide hormone calcitonin (thyrocalcitonin) which lowers blood calcium by inhibiting bone resorption and calcium recovery from renal tubule ultrafiltrate. C cells populate the middle third of each lateral lobe of the thyroid and are typically found scattered within thyroid follicles, inside the basal lamina but not reaching the follicle lumen: they are occasionally seen in clusters in the interfollicular stroma (which is why they are also called parafollicular cells).
THYROIDECTOMY Apart from variable enlargement during menstruation and pregnancy, any thyroid swelling is a goitre, which may press on related structures. Symptoms are most commonly due to pressure on the trachea or on the recurrent laryngeal nerves, and there may also be venous engorgement. If thyroidectomy is undertaken, care must be taken when tying off the superior and inferior thyroid arteries to avoid damage to adjacent nerves. The external laryngeal nerve runs close to the superior thyroid artery and the recurrent laryngeal nerve runs close to the inferior
thyroid artery. Partial thyroidectomy is often necessary in hyperthyroidism and thyroid enlargement: the posterior parts of both lobes are left intact to preserve the parathyroid glands.
Parathyroid glands (Fig. 31.27) The parathyroid glands are small, yellowish-brown, ovoid or lentiform structures, usually lying between the posterior lobar borders of the thyroid gland and its capsule. They are commonly c.6 mm long, 3-4 mm across, and 1-2 mm from back to front, each weighing about 50 mg. Usually there are two on each side, superior and inferior. However, there may be only three or many minute parathyroid islands scattered in connective tissue near the usual sites. Normally the inferior parathyroids migrate only to the inferior thyroid poles, but they may descend with the thymus into the thorax or not descend at all, remaining above their normal level near the carotid bifurcation. To help identification, the anastomotic connection between the superior and inferior thyroid arteries along the posterior border of the thyroid gland usually passes very close to the parathyroids. The superior parathyroid glands are more constant in location than the inferior and are usually to be found midway along the posterior borders of the thyroid gland, although they may be higher. The inferior pair are more variably situated (related to their embryological development-see p. 617) and may be within the fascial thyroid sheath, below the inferior thyroid arteries and near the inferior lobar poles; or outside the sheath, immediately above an inferior thyroid artery; or in the thyroid gland near its inferior pole. These variations are surgically important. A tumour of the inferior parathyroid situated within the fascial thyroid sheath may descend along the inferior thyroid veins anterior to the trachea into the superior mediastinum, whereas if it is outside the sheath it may extend posteroinferiorly behind the oesophagus into the posterior mediastinum. The superior parathyroids are usually dorsal, the inferior parathyroids ventral, to the recurrent laryngeal nerves. The parathyroid glands are very flattened in cross-section and are not normally visible by current imaging methods, including scintigraphy.
VASCULAR SUPPLY AND LYMPHATIC DRAINAGE The parathyroid glands have a rich blood supply from the inferior thyroid arteries or from anastomoses between the superior and inferior vessels. Approximately one-third of human parathyroid glands have two or more parathyroid arteries. Lymph vessels are numerous and associated with those of the thyroid and thymus glands.
INNERVATION The nerve supply is sympathetic, either direct from the superior or middle cervical ganglia or via a plexus in the fascia on the posterior lobar aspects. Parathyroid activity is controlled by variations in blood calcium level: it is inhibited by a rise and stimulated by a fall. The nerves are believed to be vasomotor but not secretomotor. Microstructure (Figs 31.27, 31.31)
page 564 page 565
Figure 31.31 A section through a parathyroid gland, showing the small, tightly packed chief cells and occasional larger, eosinophilic oxyphil cells. A thin septum separates the parathyroid gland from the thyroid gland, seen in the top left corner. (Photograph by Sarah-Jane Smith.)
Each parathyroid gland has a thin connective tissue capsule with intraglandular septa but lacks distinct lobules. The parathyroids synthesize and secrete parathyroid hormone (PTH, parathormone), a single-chain polypeptide of 84 amino-acid residues concerned with the control of the level and distribution of calcium and phosphorus. In childhood, the gland consists of wide, irregular, interconnecting columns of chief or principal cells separated by a dense plexus of fenestrated sinusoidal capillaries. After puberty, adipose tissue accumulates in the stroma and typically accounts for about one third of the adult tissue mass, increasing further with age. Chief cells differ ultrastructurally according to their level of activity: active chief cells have large Golgi complexes with numerous vesicles and small membranebound granules. Glycogen granules are most abundant in inactive cells, which have few of the cytoplasmic features of synthetic or secretory activity, and appear histologically as 'clear' cells. In normal human parathyroid glands, inactive chief cells outnumber active cells in a ratio of 3-5:1. In contrast to the thyroid, where the activities of adjacent follicular cells are coordinated, individual chief cells of the parathyroid glands go through cycles of secretory activity and rest independently, according to serum calcium levels. A second cell type, the oxyphil (eosinophil) cell, appears just before puberty and increases in number with age. Oxyphil cells are larger than chief cells and contain more cytoplasm, which stains deeply with eosin. Their nuclei are smaller and more darkly staining than those of chief cells, and their cytoplasm is unusually rich in mitochondria. The functional significance of oxyphil cells and their relationship to chief cells are uncertain.
Oesophagus-cervical portion (Fig. 31.13) The oesophagus is a muscular tube c.25 cm long, connecting the pharynx to the stomach. It begins in the neck, level with the lower border of the cricoid cartilage and the sixth cervical vertebra. It descends largely anterior to the vertebral column into the superior mediastinum in the thorax. Generally vertical and median, it inclines to the left as far as the root of the neck, and also bends in an anteroposterior plane to follow the cervical curvature of the vertebral column. Relations The trachea lies anterior to the oesophagus, attached to it by loose connective
tissue. The vertebral column, longus colli and prevertebral layer of deep cervical fascia are posterior, and the common carotid artery and posterior part of the thyroid gland are lateral on each side. In the lower neck, where the oesophagus deviates to the left, it becomes closer to the left carotid sheath and thyroid gland than it is on the right. The thoracic duct ascends for a short distance along its left side. The recurrent laryngeal nerves ascend on each side in or near the groove between the trachea and the oesophagus. Vascular supply and lymphatic drainage The cervical part of the oesophagus is mainly supplied by branches from the inferior thyroid arteries. The oesophageal veins drain into the brachiocephalic veins, and lymphatic vessels pass to retropharyngeal, paratracheal, or deep cervical lymph nodes. Innervation The cervical part of the oesophagus is innervated by the recurrent laryngeal nerves and by the sympathetic plexus around the inferior thyroid artery.
Trachea-cervical portion (Fig. 31.15) The trachea is a tube c.10-11 cm long, formed of cartilage and fibromuscular membrane. It descends from the larynx, and extends from the level of the sixth cervical vertebra to the upper border of the fifth thoracic vertebra. It lies approximately in the sagittal plane but its point of bifurcation is usually a little to the right. The trachea is flexible and can rapidly alter in length. It is flattened posteriorly so that in transverse section it is shaped, with some individual variation, like a letter D. Its external transverse diameter is c.2 cm in adult males, and 1.5 cm in adult females. The lumen in live adults is c.1.2 cm in transverse diameter. In children the trachea is smaller, more deeply placed and more mobile. Tracheal diameter does not exceed 3 mm in the first postnatal year: during later childhood its diameter in millimetres is about equal to age in years. Relations The relationships of the trachea to other cervical structures is of clinical significance: tracheostomy is not an uncommon clinical procedure. Anteriorly the cervical part of the trachea is crossed by skin and by the superficial and deep fasciae. It is also crossed by the jugular arch and overlapped by sternohyoid and sternothyroid. The second to fourth tracheal cartilages are crossed by the isthmus of the thyroid gland, above which an anastomotic artery connects the bilateral superior thyroid arteries. Below and in front are the pretracheal fascia, inferior thyroid veins, thymic remnants and the arteria thyroidea ima (when it exists). In children the brachiocephalic artery crosses obliquely in front of the trachea at or a little above the upper border of the manubrium. The left brachiocephalic vein may also rise a little above this level. The oesophagus lies posterior to the trachea, and separates it from the vertebral column. The paired lobes of the thyroid gland, which descend to the fifth or sixth tracheal cartilage, and the common carotid and inferior thyroid arteries, all lie lateral to the trachea. The recurrent laryngeal nerves ascend on each side, in or near the grooves between the sides of the trachea and oesophagus. Vascular supply and lymphatic drainage The cervical part of the trachea is mainly supplied by branches from the inferior thyroid arteries. The tracheal veins drain into the bracheocephalic veins via the inferior thyroid plexus, and lymphatic vessels drain into the pretracheal and paratracheal nodes. Innervation The trachea is innervated by branches from the vagi, recurrent laryngeal nerves and sympathetic trunks. REFERENCES Berkovitz BKB, Kirsch C, Moxham BJ, Alusi G, Cheeseman T 2002 Interactive Head and Neck. London: Primal Pictures. Cady B, Rossi RL (eds) 1991 Surgery of the Thyroid and Parathyroid Glands. Philadelphia: Saunders. Crile G 1906 Excision of cancer of the head and neck with special reference to the plan of dissection based on one hundred and thirty two operations. J Am Med Assoc 47: 1780-6. Seminal paper which considers the surgical anatomy of radical neck dissection. Froes LB, De Tolosa EMC, Camargo RDC, Pompeu E, Liberti EA 1999 Blood supply to the human sternocleidomastoid muscle by the sternocleidomastoid branch of the occipital artery. Clin Anat 12: 412-6. Medline Similar articles Full article Ger R, Evans JT 1993 Tracheostomy: an anatomico-clinical review. Clin Anat 6: 337-41. Kapandji IA 1975 The Physiology of Joints. Ediburgh: Churchill Livingstone. Lingeman RE 1998 Surgical anatomy. In: Cummings CW et al (eds) Otolaryngology, Head and Neck
Surgery, vol.2. 3rd edition. St Louis: Mosby: 1673-85. Lucas GDA, Laudanna A, Chopard RP, Raffaelli E Jr 1994 Anatomy of the lesser occipital nerve in relation to cervicogenic headache. Clin Anat 7: 90-6. Matthers LH Jr, Smith DW, Frankel L 1992 Anatomical considerations in placement of central venous catheters. Clin Anat 5: 89-106. Robbins KT 1998 Neck dissection. In: Cummings CW et al (eds) Otolaryngology, Head and Neck Surgery, vol. 2. 3rd edition. St Louis: Mosby: 1787-1819. Considers the relationships of anatomical structures of the neck with reference to radical neck dissection. Shah JP, Patel SJ 2003 Cervical lymph nodes. In: Head and Neck Surgery and Oncology, 3rd edn. Edinburgh: Mosby: 353-94. Wilson-Pauwels L, Akesson EJ, Stewart PA 1998 Cranial Nerves: Anatomy and Clinical Comments. Toronto: Decker. page 565 page 566
© 2008 Elsevier
32 NECK AND UPPER AERODIGESTIVE TRACT Nose, nasal cavity, paranasal sinuses and pterygopalatine fossa The first part of the upper respiratory tract consists of paired nasal cavities divided from each other sagittally by the nasal septum and housed in a bony and cartilaginous framework that extends anteriorly as the external nose. The two halves of the nasal cavity open onto the face through the nares, and are continuous posteriorly with the nasopharynx through the posterior nasal apertures or choanae. The cavity is divisible into three regions, the nasal vestibule anteriorly, the chemosensory olfactory area posterosuperiorly and the respiratory region between them which constitutes the majority of the nasal cavity. The anterior nasal vestibule narrows posteriorly to form the nasal valve (the narrowest portion of the nasal airway). A series of air-filled expansions, the paranasal sinuses, lie within either the lateral walls of the nasal cavities, or in communication with them in adjacent bones. The nasal apparatus serves to warm, humidify, and to some extent filter, particles from the inhaled air, and the olfactory epithelium senses and discriminates between airborne chemicals and mediates the sense of olfaction. UPDATE Abstract: Virtual reality model of human nose.
Date Added: 19 July 2005
Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=15381580&query_hl=10 Virtual reality model of human nose.
© 2008 Elsevier
SKIN OF THE EXTERNAL NOSE The skin covering the nose is thin and loosely connected to the underlying structures. Over the apex and alae it is thicker and more adherent and bears numerous large sebaceous glands: their orifices are usually very distinct. Vascular supply, lymphatic drainage and innervation of the external nose The skin of the nose receives its blood supply from branches of the facial, ophthalmic and infraorbital arteries. The alae and lower part of the nasal septum are supplied by lateral nasal and septal branches of the facial artery The dorsal nasal branch of the ophthalmic artery and the infraorbital branch of the maxillary artery supply the lateral aspects and the dorsum of the nose. The venous networks draining the external nose do not run parallel to the arteries. Instead, they correspond to arteriovenous territories of the face: thus, the frontomedian region of the face, including the nose, drains to the facial vein, and the orbitopalpebral area of the face, including the root of the nose, drains to the ophthalmic veins. The connections of the veins of the nose, upper lip and cheek with the drainage area of the ophthalmic veins are clinically significant. Lymph drainage is primarily to the submandibular group of nodes. Lymph draining from the root of the nose drains to superficial parotid nodes. The nasal muscles are innervated mainly from the buccal branches of the facial nerve. The nasal skin is innervated by the infratrochlear and external nasal branches of the nasociliary nerve, and from the nasal branch of the infraorbital nerve.
© 2008 Elsevier
SKELETON OF THE NOSE BONY SKELETON OF THE EXTERNAL NOSE The supporting framework is composed of bone and fibro- or elastic cartilages. The bony framework supporting the upper part of the nose consists of the nasal bones, the frontal processes of the maxillae and the nasal process of the frontal bones (Figs 32.1, 32.2). The cartilaginous framework consists of the median quadrilateral septal cartilage and the paired upper lateral and major and minor alar nasal cartilages (Fig. 32.3), which are connected to each other and to nearby bones by the continuity between the perichondrium and periosteum. The strut formed by the medial crura of the alar cartilages and the overlying skin which lies between the tip of the nose and the philtrum of the upper lip is termed the columella. It is connected to the nasal septum posteriorly by the membranous septum which lacks the central cartilaginous component seen more posteriorly. Congenital nasal deformities can occur, for example a complete absence of the external nose, with only one aperture existing, or else suppression or malformation on one side e.g. atresia or failed perforation of the choanal plate (the embryonic barrier between the nasopharynx and the posterior nasal cavity). UPDATE Date Added: 30 November 2005 Publication Services, Inc. Abstract: Neural net applied to anthropological material: a methodical study on the human nasal skeleton. Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=16130825&query_hl=6 Neural net applied to anthropological material: a methodical study on the human nasal skeleton. Prescher A, Meyers A, Gerf von Keyserlingk D. Ann Anat. 2005 Jul;187(3):261-9.
page 567 page 568
Figure 32.1 Osteology of lateral wall of the nose. (By permission from Berkovitz BKB, Moxham BJ 1994 Color Atlas of the Skull. London: Mosby.)
Figure 32.2 Osteology of medial wall (septum) of the nose. (By permission from Berkovitz BKB, Moxham BJ 1994 Color Atlas of the Skull. London: Mosby.)
Nasal fractures
Trauma to the midface may cause fractures of the bony skeleton ranging from simple displaced fractures of the nasal bones to complex fractures of the midfacial skeleton. The upper dentition may be mobilized as a result of a fracture through the maxilla just above the tooth roots. The fracture line may also run through the ethmoidal air cells resulting in mobility of the whole midface below the orbits. The zygomatic bone may also be depressed by blunt trauma causing damage to the adjacent maxillary nerve and loss of contour of the cheek as the support of the zygomatic process and the lateral margin of the orbit is lost. Direct compressive injury to the contents of the orbit may result in a so called 'blow out' fracture in which the orbital contents prolapse through the disrupted inferior or medial walls of the orbits, often resulting in entrapment of the extra ocular muscles and restriction of eye movements.
CARTILAGINOUS SKELETON OF THE EXTERNAL NOSE (Fig. 32.3) Septal cartilage
Almost quadrilateral in side view, the septal cartilage is sandwiched between two layers of mucoperichondrium and lies often eccentrically between the anterior parts of the nasal cavity. The anterosuperior margin is connected above to the posterior border of the internasal suture. The middle part is continuous with the
upper lateral cartilages, and the lowest part is attached to these cartilages by perichondrial extensions. The anteroinferior border is connected on each side with the medial crurae of the major alar cartilage. The posterosuperior border joins the perpendicular plate of the ethmoid, while the posteroinferior border is attached to the vomer and to the nasal crest and anterior nasal spine of the maxilla. The septal cartilage may extend back (especially in children) as a narrow sphenoidal process for some distance between the vomer and the perpendicular plate of the ethmoid. The anteroinferior part of the nasal septum between the nares is devoid of cartilage and is called the membranous septum. It is continuous with the columnella anteriorly. Lateral (superior) nasal cartilage
The lateral nasal cartilage is triangular, its anterior margin being thicker than the posterior. The upper part is continuous with the septal cartilage, but anteroinferiorly it may be separated from it by a narrow fissure. The superior margin of the lateral nasal cartilage is attached to the nasal bone and frontal process of the maxilla, while the inferior margin is connected by fibrous tissue to the lateral crus of the major alar cartilage. Major alar cartilage
The major alar cartilage is a thin flexible plate lying below the upper lateral cartilage, and curved acutely around the anterior part of its naris. The medial part, the narrow medial crus (septal process), is loosely connected by fibrous tissue to its contralateral fellow and to the anteroinferior part of the septal cartilage, thus forming part of the septum mobile nasi. The lateral crus lies lateral to the naris and runs superolaterally away from the margin of the nasal ala. The upper border of the lateral crus of the major alar cartilage is attached by fibrous tissue to the lower border of the lateral nasal cartilage. Its lateral border is connected to the frontal process of the maxilla by a tough fibrous membrane containing three or four minor cartilages of the ala. The junction between the lateral crura of the major alar and the lateral cartilages is variable. The two edges may abut or overlap; the lateral crus is then the more lateral at the junction. The lateral crus of the major alar cartilage is shorter than the lateral margin of the naris and runs away from the margin of the ala nasi. The lateral part of the margin of the ala nasi is fibroadipose tissue covered by skin. In front, the angulations or 'domes' between the medial and lateral crurae of the major alar cartilages are separated by a notch palpable at the tip of the nose.
NASAL CAVITY The nasal cavity is an irregular space between the roof of the mouth and the cranial base, divided by a vertical osseocartilaginous septum that is approximately median in position. The bony septum reaches the posterior limit of the cavity, which leads into the nasopharynx through a pair of posterior nasal apertures, or choanae, lying above the posterior hard palatal border (Fig. 27.1) The medial border of the choanae is formed by the posterior edge of the vomer, and its posteroinferior boundary by the horizontal plate of the palatine bone with the nasal crest of the palatine bone. Lateral to the ala of the vomer the choanae are bounded superiorly by the vaginal processes of the pterygoid processes
above, and by the perpendicular plates of the palatine bones laterally. The size of each choana is not usually affected by deviations of the nasal septum. The cavity is wider below than above, and widest and vertically deepest in its central region. It communicates with the frontal, ethmoidal, maxillary and sphenoidal paranasal sinuses. The posterior nasal apertures are oval openings separated by the posterior border of the vomer, each being limited below by the horizontal plate of the palatine bone, above by the sphenoid and laterally by the medial pterygoid plate. In the adult each is c.2.5 cm in vertical height and 1.3 cm transversely. The vomerovaginal and palatovaginal canals are found in the roof of this region. Each half of the nasal cavity has a roof, floor, lateral and medial (septal) walls. Roof
The roof (Figs 32.1, 32.2) is horizontal in its central part but slopes downwards in front and behind. The anterior slope is formed by the nasal spine of the frontal bones and by the nasal bones, which contribute to the external nose. The central horizontal region is formed by the cribriform plate of the ethmoid bone which separates the nasal cavity from the floor of the anterior cranial fossa. The cribriform plate contains a separate anterior foramen for the anterior ethmoidal nerve and vessels, and numerous small perforations which transmit the olfactory nerves. The posterior slope is formed by the anterior aspect of the body of the sphenoid - interrupted on each side by an opening of a sphenoidal sinus - and the sphenoidal conchae or superior conchae. The alae of the vomer and the sphenoidal processes of the palatine bones lie below. Floor (Fig. 32.1)
The floor of the nasal cavity is smooth, concave transversely, and slopes up from anterior to posterior apertures. It constitutes the upper surface of the hard palate. Anteriorly, the palatine processes of the maxillae and, behind them, the horizontal plates of the palatine bones, articulate in the midline and with each other. The nasal floor is therefore crossed at the junction of its middle and posterior thirds by the palatomaxillary suture. Anteriorly, near the septum, a small infundibular opening in the nasal floor leads into the incisive canals that descend to the incisive fossa: the opening is marked by a slight depression in the mucosa. The floor of the nose may be deficient as a result of congenital clefting of the hard and/or soft palate. page 568 page 569
Figure 32.3 The bone and cartilages of the nose. A, lateral view; B, nasal septum; C, frontal view; D, inferior view.
Medial wall (Fig. 32.2)
The medial wall of the nasal cavity is the nasal septum. Relatively featureless, it lies between the roof and floor and is a thin sheet of bone with a wide anterior deficiency occupied by the septal cartilage. Ridges or spurs of bone sometimes project from the septum on either side. The bony part is formed primarily by the vomer and the perpendicular plate of the ethmoid. The vomer extends from the body of the sphenoid to the hard palate, forming the posteroinferior septum (including the posterior border). The surface contains grooves related to the
nasopalatine nerves and accompanying vessels. The perpendicular plate of the ethmoid forms the anterosuperior part of the bony nasal septum and is continuous above with its cribriform plate. The nasal septum is often deviated - more usually to the left - and particularly affects the perpendicular plate of the ethmoid. Other bones make minor contributions to the septum at the upper and lower limits of the medial wall. The nasal bones and the nasal spine of the frontal bones are anterosuperior, the rostrum and crest of the sphenoid bone are posterosuperior, and the nasal crests of the maxillary and palatine bones are inferior. Above the incisive canals, at the lower edge of the septal cartilage, there is sometimes a depression pointing downwards and forwards: it occupies the position of the nasopalatine canal which connected the nasal and buccal cavities in early fetal life. A minute orifice may be seen leading back into a blind tubule, 2-6 mm long, on each side of the septum near this recess. The tubules house the vomeronasal organ, a paired accessory olfactory organ similar to the olfactory epithelium in amphibians and reptiles, but believed to be vestigial in man. Posteriorly the mucoperiosteum of the septum may be thickened by a cushion of very vascular tissue. The nasal septum may be displaced by injury or by some congenital defect, and sometimes the deviation may be so great as to bring the septum and one lateral wall into contact, causing complete unilateral nasal obstruction. Lateral wall (Fig. 32.1)
The lateral wall of the nasal cavity contains three projections of variable size called the inferior, middle and superior nasal conchae or turbinates. It is formed largely by the maxilla and its anterior and posterior fontanelles (bony deficiencies in the medial wall of the maxilla obliterated to varying degrees by fibrous tissue) anteroinferiorly; by the perpendicular plate of the palatine bone posteriorly; and superiorly by the labyrinth of the ethmoid bone which separates the nasal cavity from the orbit. The nasal conchae curve generally inferomedially, each roofing a groove, or meatus, open to the nasal cavity. The middle conchae may also curve inferolaterally or be expanded by an enclosed air cell to form a so called 'concha bullosa'. The opening associated with the maxillary sinus, the maxillary hiatus, appears as a wide defect in the nasal surface of the isolated maxilla. However, in the articulated state in life, the hiatus is greatly reduced in size by neighbouring bones. Thus it is covered by the maxillary process of the inferior concha below, by the uncinate process of the ethmoid bone above, by the perpendicular plate of the palatine bone behind, and by small parts of ethmoidal labyrinth and lacrimal bone anterosuperiorly (Fig. 32.1). Inferior concha or turbinate page 569 page 570
The inferior concha is a thin, curved, independent bone which articulates with the nasal surface of the maxilla and the perpendicular plate of the palatine bone. The free lower border is gently curved and the subjacent inferior meatus reaches the nasal floor. The inferior meatus is the largest meatus, and it extends along almost all of the lateral nasal wall. It is deepest at the junction of its anterior and middle thirds, where the inferior opening of the nasolacrimal canal appears. The
nasolacrimal canal is formed by the articulations between the lacrimal groove of the maxilla and the descending process of the lacrimal bone and the lacrimal process of the inferior nasal concha. During postnatal development, the ostium of the nasolacrimal duct moves upwards and is increasingly arched over by the inferior concha. Middle concha or turbinate
The middle concha is a medial process of the ethmoidal labyrinth, and extends back to articulate with the perpendicular plate of the palatine bone. The middle concha itself may be pneumatized (conchal sinus). The region beneath it is the middle meatus, which is deeper in front than behind, lies below and lateral to the middle concha and continues anteriorly into a shallow fossa above the vestibule, termed the atrium of the middle meatus. Lateral to the atrium an ill-defined curved ridge, the agger nasi, slopes downwards and forwards from the upper end of the anterior free border of the middle concha. The agger nasi is better developed in the newborn than in adults. The middle concha must be displaced to display the lateral wall of the middle meatus fully. The main features of this wall are a rounded elevation, the bulla ethmoidalis and, below it and extending up in front of it, a curved cleft, the hiatus semilunaris (Figs 32.1, 32.4). The bulla ethmoidalis is formed by the expansion of the middle ethmoidal sinuses, which open on or just above it. Its size varies according to that of the contained sinuses. The hiatus semilunaris opens laterally into a curved channel, the ethmoidal infundibulum, into which the anterior ethmoidal sinuses open. In at least 50% of subjects the openings of the ethmoidal sinuses are continuous with the opening of the frontonasal duct which drains the frontal sinus. Alternately, the infundibulum may end blindly in front in one or more anterior ethmoidal sinuses (infundibular sinuses) and the frontonasal duct then opens more medially directly into the anterior end of the middle meatus. The opening of the maxillary sinus lies below the bulla, usually hidden by the flange-like lower edge of the uncinate process. This opening is near the roof of the sinus and is therefore unfavourable for drainage. The coordinated beating of the cilia of the mucociliary clearance system of the maxillary sinus is directed towards it. An accessory opening of the maxillary sinus through the posterior fontanelle of the medial wall of the maxillary sinus frequently exists inferoposterior to the hiatus. Superior concha or turbinate (Fig. 32.1)
Figure 32.4 Lateral wall of the left nasal cavity; the conchae have been partially removed. (By permission from Ellis H, Feldman S 1997 Anatomy for Anaesthetists, 7th edn. Oxford: Blackwell.)
The superior concha is a medial process of the ethmoidal labyrinth and presents as a small curved lamina, posterosuperior to the middle concha. It roofs the superior meatus and is the shortest and shallowest of the three conchae. Above the superior concha, the sphenoidal sinus opens into a triangular sphenoethmoidal recess which separates the superior concha and anterior aspect of the body of the sphenoid. Occasionally a fourth concha, the highest or supreme nasal concha, appears on the lateral wall of this recess: the passage immediately below it is called the supreme nasal meatus, and it sometimes displays an opening of the posterior ethmoidal sinus. The superior meatus is a short oblique passage extending about halfway along the upper border of the middle concha. The posterior ethmoidal sinuses open, via a variable number of apertures, into its anterior part. The sphenopalatine foramen (Fig. 32.1) can be approached through the middle meatus. Clinically it is posterior to the middle meatus and transmits the sphenopalatine artery and nasopalatine and superior nasal nerves from the pterygopalatine fossa. It is bounded above by the body and concha of the sphenoid, below by the superior border of the perpendicular plate of the palatine bone, and in front and behind by the orbital and sphenoidal processes of the palatine bone. Nasal obstruction
Variations in the anatomy of the structures of the lateral nasal wall, e.g. oversized bulla ethmoidalis air cells, paradoxically curved middle conchae, concha bullosae of the middle concha, or so called 'compensatory hypertrophy' of the inferior concha into a congenital concavity of the nasal septum, may all cause nasal obstruction or impaired sinus ventilation and drainage. Conchae are often excised
to open the airway. They may also be 'out-fractured' to lateralize them and improve the airway. The degree of congestion with blood and hence swelling of the vascular mucoperiosteum of the conchae varies cyclically every few hours and may be interfered with by allergy or infection, both of which will cause swelling and inflammation of the conchal mucous membranes. These mucous membranes may be shrunk with vasoconstrictor medication, excised surgically or burned to cause scarring so as to restrict swelling.
© 2008 Elsevier
SOFT TISSUE SOFT TISSUE FEATURES OF EXTERNAL NOSE Externally the nose is pyramidal in shape, its upper angle or root being continuous with the forehead, and its free tip forming the apex. Two ellipsoidal apertures, the external nares or nostrils, are inferior and separated by the nasal septum and columnella. The external nares are narrower in front, and usually measure 1.5-2 cm anteroposteriorly and 0.5-1 cm transversely. By their union in the median plane, the lateral surfaces of the nose form the dorsum nasi, the shape of which varies greatly between individuals. The upper part of the external nose is kept patent by the nasal bones and the frontal processes of the maxillae. Below this the nasal cartilages form the walls of the external nose. The lateral surfaces end below in the rounded alae nasi. UPDATE Date Added: 15 May 2006 Abstract: Anatomy of the interdomal fat pad of the nose Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=14574464&query_hl=22&itool=pubmed_docsum The interdomal fat pad of the nose: A new anatomical structure. Copcu E, Metin K, Ozsunar Y et al: Surg Radiol Anat 26:14-18, 2004.
SOFT TISSUE FEATURES OF THE INTERNAL NASAL CAVITY AND SINUS MUCOSA Nasal vestibule
The nasal vestibule is a slightly expanded anterior part of the air passage just inside the naris. It is bounded laterally by the ala and lateral part of the major alar cartilage, and medially by the medial crus or septal process of this cartilage. The vestibule extends as a small recess towards the apex of the nose. Its lumen is lined with skin, the inferior region bearing sebaceous and sweat glands, and coarse hairs (vibrissae) curving towards the naris and helping to arrest the passage of particles in inspired air. In males, after middle age, these hairs increase considerably in size. The vestibule is limited above and behind by a curved ridge, the limen nasi or nasal valve, corresponding to the lower margin of the upper lateral cartilage anteriorly and the pyriform nasal aperture posteroinferiorly. At this demarcation, the skin of the vestibule is continuous with the nasal mucosa. Nasal mucosa and respiratory epithelium page 570 page 571
The lining of the anterior part of the nasal cavity and vestibule is continuous with the skin, and consists of keratinized stratified squamous epithelium overlying a connective tissue lamina propria. Further posteriorly, at the limen nasi, this grades into a mucosa lined at first by non-keratinizing stratified squamous epithelium, then by pseudostratified ciliated (respiratory) epithelium with numerous goblet cells (p. 31, Fig. 3.6). Respiratory epithelium forms most of the surface of the nasal cavity, and so covers the conchae, meatuses, floor and roof, except where the olfactory epithelium is present. In some areas, cells of the respiratory epithelium may be low columnar or cuboidal, and the proportion of ciliated to nonciliated cells is variable. The nasal mucosa has numerous underlying seromucous glands within its lamina propria, which makes the surface sticky so that particles in the inspired air are deposited on the surface. It is adherent to the periosteum or perichondrium of the neighbouring skeletal structures. The mucosa is continuous with the nasopharyngeal mucosa through the posterior nasal apertures, the conjunctiva through the nasolacrimal duct and lacrimal canaliculi, and the mucosae of the sphenoidal, ethmoidal, frontal and maxillary sinuses through their openings into
the meatuses. The epithelium in the sinuses is thinner and has fewer goblet cells than elsewhere. Subepithelial glands are sparse: their combined secretions are directed towards the nasal cavity by ciliary action. The mucosa is thickest and most vascular over the conchae, especially at their extremities, and also on the nasal septum, but is very thin in the meatuses, on the floor and in the sinuses. Its thickness reduces the volume of the nasal cavity and its apertures significantly. The lamina propria contains cavernous vascular tissue with large venous sinusoids. Local vascular changes, controlled by the vasomotor autonomic innervation and possibly by endocrine stimuli, alter the thickness and contours of the mucosal surfaces, and this is visible as a swelling or shrinkage of the nasal lining. These changes produce periodic alterations in the rate of airflow through the nasal passages, alternating between nares, which may serve to protect their mucosae from desiccation. The conchae add greatly to the surface area of the nasal cavity which increases the turbulence of inhaled air, and may improve olfaction by slowing the passage of air past the olfactory area. Humidification and warming of the inhaled air are also augmented by the increased mucosal area and turbulence. The mucous film is continually moved by ciliary action backwards into the nasopharynx at a rate of c.6 mm per minute. Palatal movements transfer the mucus and its entrapped particles to the oropharynx for swallowing, but some also enters the nasal vestibule anteriorly. The secretions of the nasal mucosa contain the bacteriocides lysozyme and lactoferrin, and also secretory immunoglobulins (IgA). Respiratory epithelium extends through the apertures of the paranasal sinuses to line them, and is closely bound to the underlying periosteum in the walls of the sinuses to form a combined mucoperiosteum. Olfactory epithelium
The peripheral receptors for olfactory sensation are located bilaterally in areas of sensory epithelium lining the posterodorsal parts of the nasal cavities. The sensory epithelium occupies an area of c.5 cm2 covering the posterior upper parts of the lateral nasal wall, including the back of the superior concha, the sphenoethmoidal recess, the upper part of the perpendicular plate of the ethmoid and the roof of the nose arching between the septum and lateral wall, including the underside of the cribriform plate. This area is pigmented yellowish brown in contrast to the pinker colour of the respiratory mucosa. Microstructure of the olfactory mucosa
The olfactory mucosa consists of a pseudostratified olfactory epithelium, derived from the embryonic olfactory placodes (p. 245). It contains sensory receptor neurones, and its underlying lamina propria contains their axons, which lie within numerous olfactory nerve fascicles, and subepithelial olfactory glands (of Bowman). The glands secrete a predominantly serous fluid through ducts which open onto the epithelial surface. Their secretions form a thin fluid layer in which sensory cilia and the microvilli of sustentacular cells are embedded. Olfactory epithelium The olfactory epithelium (Figs 32.5, 32.6) is considerably thicker (up to 100 µm) than the respiratory epithelium. It contains olfactory receptor neurones, sustentacular cells and two classes of basal cell, horizontal basal cells (closest to the basal lamina) and globose basal cells (Fig. 32.7). The nuclei of these various cells occupy specific zones within the epithelial thickness. Most superficially is a layer of sustentacular cell nuclei; beneath this, and occupying much of the epithelial thickness, are several tiers of receptor cell bodies and nuclei. Basal cells lie between this zone and the basal lamina underlying the epithelium.
Figure 32.5 Low power micrograph of the olfactory mucosa covering the superior concha. The olfactory epithelium is shown overlying a lamina propria containing cavernous vascular tissue and olfactory glands (of Bowman). Bundles of olfactory axons (fila olfactoria) pass through the mucosa towards the cribriform plate. A thinner respiratory epithelium covers the lower surface, shown here beneath the bone of the concha. (By permission from Kierszenbaum AL 2002 Histology and Cell Biology. St Louis: Mosby.)
Figure 32.6 High power micrograph of the olfactory mucosa from an 18-week fetus, showing the nuclei of basal cells, olfactory neurones and sustentacular cells. The olfactory neuronal endings (knobs or vesicles) can be seen projecting from the free surface. (By permission from Stevens A, Lowe JS 1996 Human Histology, 2nd edn. London: Mosby.)
Olfactory receptor neurons
page 571
page 572
Figure 32.7 The chief cytological features of the olfactory epithelium. Receptor cells (yellow) are situated among columnar sustentacular cells. The axons of the receptor cells emerge from the epithelium in bundles enclosed by ensheathing glial cells. Rounded globose basal cells (brown) and flattened horizontal basal cells (not shown) lie on the basal lamina and the subepithelial glands (of Bowman) open on to the surface via their intraepithelial ducts (green). At the surface are cilia of the receptor cells and microvilli of the supporting cells.
Olfactory receptor neurones are bipolar cells. They have a cell body and nucleus located in the middle zone of the epithelium, a single unbranched apical dendrite c.2 µm across which extends to the epithelial surface, and a basally directed unmyelinated axon c.0.2 µm in diameter, which passes out of the epithelium. Each dendrite projects into the overlying secretory fluid as an expanded cylindrical olfactory ending (rod, knob or vesicle). Groups of up to 20 cilia radiate from the circumference of each ending and extend for long distances parallel to the epithelial surface. Internally, the short proximal part of each cilium has the '9 + 2' pattern of microtubules typical of motile cilia (p. 19), while the longer distal trailing end contains only the central pair of microtubules. The cilia lack dynein arms on the peripheral microtubule doublets and are thought to be non-motile, serving to project a large area of sensory surface for the efficient detection of odorants. Individual receptor neurones express receptors for a single (or very few) odorant molecules. Although neurones with the same receptor specificity are randomly distributed within anatomical zones of the epithelium, all project to the same target dendritic field (glomeruli) of the olfactory bulb, and there is a considerable degree of convergence. Specific odours activate a unique spectrum of receptors which in turn activate restricted groups of glomeruli and their second order neurones. The axons form small intraepithelial fascicles among the processes of sustentacular and basal cells. The fascicles penetrate the basal lamina, and are immediately ensheathed by olfactory ensheathing cells. Groups of up to 50 fascicles join to form larger olfactory nerve rootlets which pass through the cribriform plate to enter the olfactory bulb, there synapsing in glomeruli with secondary sensory neurones, principally mitral cells and, to a lesser extent, smaller tufted cells.
Sustentacular cells Sustentacular, or supporting, cells (Fig. 32.7) are columnar cells that separate and partially ensheathe the olfactory receptors. Their large nuclei form a layer superficial to the receptor nuclei. They send numerous long, irregular microvilli into the secretory fluid layer covering the surface of the epithelium, where they lie among the long trailing ends of olfactory receptor cilia. At their bases, facing the basal lamina, they have expanded end-feet containing numerous lamellated dense bodies resembling neuronal lipofuscin granules. These are remains of secondary lysosomes formed as a result of phagocytic activity, and are largely responsible for the pigmentation of the olfactory area. The granules gradually accumulate with age, and because these cells are long-lived, pigmentation also increases in intensity with age. Cells are linked by desmosomes near the epithelial surface, which gives mechanical coherence to the epithelium, and by tight junctions between the sustentacular cells and olfactory receptors at the level of the epithelial surface, which provides a protective barrier. Basal cells There are two types of basal cell; horizontal basal cells and globose basal cells. Horizontal basal cells are flattened against the basal lamina, and have condensed nuclei and darkly staining cytoplasm containing numerous intermediate filaments of the cytokeratin family, inserted into desmosomes contacting surrounding sustentacular cells. In contrast, globose cells are rounded or elliptical in shape, with a pale, euchromatic nucleus, and a pale cytoplasm. They form a distinct zone spaced slightly from the basal surface of the epithelium. Mitoses are found within this zone because globose basal cells are the immediate source of new olfactory receptor neurones. Olfactory ensheathing (glial) cells Olfactory ensheathing cells share properties with astrocytes and non-myelinating Schwann cells, but they possess distinctive features that indicate they are a separate class of glia. Developmentally they are derived from the olfactory placode rather than neural crest. They ensheath olfactory axons in a unique manner throughout their entire course, and accompany them into the central nervous tissue of the olfactory bulb, where they contribute to the glia limitans. Olfactory glands The olfactory (Bowman's) glands (Fig. 32.7) are branched tubuloalveolar structures that lie beneath the olfactory epithelium and secrete onto the epithelial surface through narrow, vertical ducts. Their secretions, which include defensive substances, lysozyme, lactoferrin, IgA and sulphated proteoglycans, bathe the dendritic endings and cilia of the olfactory receptors. The fluid acts as a solvent for odorant molecules, allowing their diffusion to the sensory receptors. The glands also secrete odorant-binding proteins into the fluid, which increase the efficiency of odour detection. Turnover of olfactory receptor neurons Receptor neurones are lost and replaced throughout life. Individual receptor cells have a variable lifespan, averaging 1-3 months. Stem cells situated near the base of the epithelium undergo periodic mitotic division throughout life, giving rise to new olfactory receptor cells which then grow a dendrite to the olfactory surface and an axon to the olfactory bulb. The cell bodies of these new receptor neurones gradually move apically until they reach the region just below the supporting cell nuclei. When they degenerate, dead cells are either shed from the epithelium or are phagocytosed by sustentacular cells. The rate of receptor cell loss and replacement increases after exposure to damaging stimuli, but their capacity to turnover declines slowly with age, and this contributes to diminishing olfactory sensory function in old age. Vomeronasal organ
Vomeronasal organs are important in sexual behaviour in several species, where
they are especially concerned with detecting pheromones, however the organ is considered to be non-functional in adult humans. Putative sensory cells are replaced by non-sensory epithelium if the organ persists into postnatal life.
© 2008 Elsevier
VASCULAR SUPPLY AND LYMPHATIC DRAINAGE OF THE NASAL CAVITY Many of the vessels and nerves supplying the nasal cavities arise within the pterygopalatine fossa and these origins are described on page 578.
ARTERIES page 572 page 573
Figure 32.8 Arteries of the lateral wall of the nose.
These arise as branches of the ophthalmic, maxillary and facial arteries which run to supply different territories within the walls, floor and roof of the nose (Fig. 32.8). They ramify to form anastomotic plexuses within and deep to the nasal mucosa. Anastomoses also occur between some larger arterial branches. The anterior and posterior ethmoidal branches of the ophthalmic artery supply the ethmoidal and frontal sinuses and the roof of the nose (including the septum). The sphenopalatine branch of the maxillary artery supplies the mucosa of the conchae, meatuses and posteroinferior part of the nasal septum, i.e. it is the principal vessel supplying the nasal mucosa. The greater palatine branch of the maxillary artery supplies the region of the inferior meatus. Its terminal part ascends through the incisive canal to anastomose on the septum with branches of the sphenopalatine and anterior ethmoidal arteries and with the septal branch of the superior labial artery. This septal region (Little's area) is a common site of bleeding from the nose. The infraorbital artery and the superior, anterior, and posterior alveolar branches of the maxillary artery supply the mucosa of the maxillary sinus. The pharyngeal branch of the maxillary artery supplies the sphenoidal sinus. UPDATE Date Added: 02 August 2005 Shanida Helena Nataraja, PhD (Dianthus Medical Limited) Update: Anatomic variations of the arteries of the nasal fossa. Knowledge of the blood supply of the nasal fossa is important for the development of an appropriate therapeutic strategy for the management of epistaxis, particularly intractable epistaxis. Ten fresh, non-embalmed, cadaveric heads were injected with red colored latex to enhance visibility of the vascular supply of the nasal fossa and were subsequently cut sagittally parallel to the septal partition. Septal vascularization was investigated in 10 hemi-heads and external septum vascularization in 20 hemi-heads. In 18 cases, the division of the sphenopalatine artery occurred in the infratemporal fossa, where it was seen to branch into two
branches (10 cases), three branches (six cases) or four and five branches (one case each) at the exit point of the sphenopalatine foramen. In the remaining two cases, the sphenopalatine artery divided in the nasal fossa. The nasopalatine artery is the internal posterior branch of the sphenopalatine artery and is responsible for the supply of the septal circulation. It gives rise to three arterial branches: a superior branch that supplies Kisselbach's plexus and two inferior branches that supply the inferior septal area. The external branch of the sphenopalatine artery supplies the external septum of the nasal fossa via the medium and low turbinate arteries (2mm and 1.5mm diameter, respectively). The anterior and posterior ethmoidal arteries and the sphenopalatine artery jointly supply the superior aspect of the outer wall of the nasal fossa. This study confirms that the sphenopalatine artery divides into multiple arterial trunks, most frequently in the infratemporal fossa. Babin E, Moreau S, de Rugy MG et al: Anatomic variations of the arteries of the nasal fossa. Otolaryngol Head Neck Surg 128(2):236-239, 2003. Medline Similar articles
VEINS These form a rich submucosal cavernous plexus that is especially dense in the posterior part of the septum and in the middle and inferior conchae. Numerous arteriovenous anastomoses are present in the deep layer of the mucosa and around the mucosal glands. The cavernous conchal plexuses resemble those in erectile tissue: the nasal cavity is susceptible to blockage should they become engorged. Veins from the posterior part of the nose generally pass to the sphenopalatine vein that runs back through the sphenopalatine foramen to drain into the pterygoid venous plexus. The anterior part of the nose is drained mainly through veins accompanying the anterior ethmoidal arteries, and these veins subsequently pass into the ophthalmic or facial veins. A few veins pass through the cribriform plate to connect with those on the orbital surface of the frontal lobes of the brain. When the foramen caecum is patent, it transmits a vein from the nasal cavity to the superior sagittal sinus.
LYMPHATIC DRAINAGE Lymph vessels from the anterior region of the nasal cavity pass superficially to join those of the external nasal skin, which end in the submandibular nodes. The rest of the nasal cavity, paranasal sinuses, nasopharynx and pharyngeal end of the pharyngotympanic tube all drain to the upper deep cervical nodes, directly or through the retropharyngeal nodes. The posterior nasal floor probably drains to the parotid nodes.
© 2008 Elsevier
INNERVATION OF THE NASAL CAVITY Special sensation related to olfaction is associated with the olfactory nerves. General sensation to the nasal mucosa is related to branches from the ophthalmic and maxillary divisions of the trigeminal nerves. The general sensations mediated are touch, pain and temperature. The trigeminal fibres close to, and within, the epithelial layer are also sensitive to noxious chemicals, e.g. ammonia and sulphur dioxide. These latter stimuli may therefore be perceived by the trigeminal nerve even when the olfactory nerves have been damaged. In addition, autonomic nerves from the pterygopalatine ganglion innervate mucous glands and control cyclical and reactive vasomotor activity.
NERVES OF ORDINARY SENSATION (Figs 32.9, 32.10) These are all derived from the maxillary nerve, with an additional contribution from the nasociliary branch of the ophthalmic nerve. The anterior ethmoidal branch of the nasociliary nerve leaves the cranial cavity through a small slit near the crista galli and enters the roof of the nasal cavity. Here it runs in a groove on the inner surface of the nasal bone, and supplies the roof of the nasal cavity. It gives off a lateral internal branch to supply the anterior part of the lateral wall and a medial internal branch to the anterior and upper parts of the septum, before emerging at the inferior margin of the nasal bone as the external nasal nerve to supply the skin of the external nose to the nasal tip. The infraorbital nerve supplies the nasal vestibule; the anterior superior alveolar nerve supplies part of the septum, the floor near the anterior nasal spine and the anterior part of the lateral wall as high as the opening of the maxillary sinus; the lateral posterior superior nasal and the posterior inferior nasal branches of the greater palatine nerve together supply the posterior three-quarters of the lateral wall, roof and floor; the medial posterior superior nasal nerves and the nasopalatine nerve supply the inferior part of the nasal septum; branches from the nerve of the pterygoid canal supply the upper and posterior part of the roof and septum. Autonomic nerves accompany the sensory innervation. Sympathetic postganglionic vasomotor fibres are distributed to the nasal blood vessels, and postganglionic parasympathetic fibres from the pterygopalatine ganglion provide the secretomotor supply to the nasal glands.
OLFACTORY NERVES (Figs 32.9, 32.10) The olfactory nerves serving the sense of smell have their cells of origin in the olfactory mucosa covering the superior nasal concha, the upper part of the vertical portion of the middle concha and the opposite part of the nasal septum. The axons, which are unmyelinated, originate as the central or deep processes of the olfactory neurones, and collect into bundles that cross in various directions, forming a plexiform network in the mucosa. These bundles finally form c.20 branches that traverse the cribriform plate in lateral and medial groups and end in the glomeruli of the olfactory bulb. Each branch has a sheath consisting of dura mater and pia-arachnoid, the former continuing into the nasal periosteum, the latter into the connective tissue sheaths surrounding the nerve bundles: this arrangement may favour the spread of infection into the cranial cavity from the nasal cavity. page 573 page 574
Figure 32.9 The sensory innervation of the lateral wall of the nasal cavity, hard and soft palates, and nasopharynx. Secretomotor fibres to mucous glands are distributed in branches from the pterygopalatine ganglion.
Figure 32.10 Bundles of olfactory nerve fibres and nerves associated with the septum (left side).
In severe injuries involving the anterior cranial fossa, the olfactory bulb may be separated from the olfactory nerves or the nerves may be torn, producing anosmia, loss of olfaction. Fractures may involve the meninges, admitting cerebrospinal fluid (CSF) into the nose resulting in cerebrospinal rhinorrhoea. Such injuries also open up avenues for infection from the nasal cavity.
© 2008 Elsevier
VISCERA: PARANASAL SINUSES The paranasal sinuses are the frontal, ethmoidal, sphenoidal and maxillary sinuses, housed within the bones of the same name. The ethmoidal sinuses differ from the others in being formed of small multiple cavities, divisible into anterior, middle and posterior groups. All sinuses open into the lateral wall of the nasal cavity by small apertures that allow the equilibration of air and movement of mucus. The detailed position of these apertures, and the precise form and sizes of the sinuses, vary enormously between individuals. Their mucosa is continuous with that of the nasal cavity - a feature unfortunately favouring the spread of infections - and is similar histologically, although thinner, less vascular and less adherent to bone. Mucus is secreted by glands within their mucosa and is swept through their apertures into the nose by cilia. Cilia are not uniformly distributed but are always present near the apertures of the sinuses. The mucociliary escalator is the normal mechanism for clearing the sinuses and maintaining aeration and forms the theoretical basis of functional endoscopic sinus surgery (FESS). Most sinuses are rudimentary or absent at birth, but enlarge appreciably during the eruption of the permanent teeth and after puberty, markedly altering the size and shape of the face at these times. The functions of the sinuses remain speculative. They clearly add some resonance to the voice, and also allow the enlargement of local areas of the skull whilst minimizing a corresponding increase in bony mass. It is likely that such growth-related changes serve to strengthen particular regions, e.g. the alveolar process of the maxilla when the secondary dentition erupts, but they may also function in contouring the head to provide visual signals indicating the individual's status in a social context (e.g. gender, sexual maturity and group identity).
IMAGING OF THE PARANASAL SINUSES On standard radiological images, normal sinuses are radiolucent, whereas when they are diseased they show varying degrees of opacity. In lateral views, the extent of the frontal sinus both upwards into the frontal bone and back into the orbital roof can be assessed. The ethmoidal sinuses are seen to extend back from the frontal process of the maxilla as far as the sphenoidal sinus, which is clearly visible below and in front of the hypophyseal fossa, although the two sphenoidal sinuses appear superimposed, and the individual sphenoidal sinuses are seen better from above. The maxillary sinus is clearly seen in a lateral view, lying below the orbit, and its relationship to the roots of the teeth is obvious. In posteroanterior views of the skull, most of the sinuses are visible. The frontal sinuses appear above the nasal cavity and the medial part of the orbits. Their asymmetry, vertical extent and the position of their septa can be assessed. The ethmoidal sinuses are superimposed on each other and on the sphenoidal sinuses in this view, lying between the orbits below the cribriform plate. The sphenoidal sinuses are obscured in this view. Each maxillary sinus is a pyramidal radiolucent area below the orbit and lateral to the lower part of the nasal cavity, extending inferiorly into the alveolar process of the maxilla. The frontal, maxillary and ethmoidal sinuses are particularly well demonstrated in occipitomental views. The introduction of imaging techniques such as CT has provided infinitely clearer images of the air sinuses and this has significantly aided diagnosis (Figs 32.11, 32.12, 32.13). page 574 page 575
Figure 32.11 Coronal CT scan showing frontal sinus. (By kind permission from Dr N Drage.)
Figure 32.12 Coronal CT scan showing sphenoidal air sinus. (By kind permission from Dr N Drage.)
Figure 32.13 Coronal CT scan showing ethmoidal and maxillary sinuses. (By kind permission from Dr N Drage.)
FRONTAL SINUS (Figs 32.8, 32.10) The paired frontal sinuses, situated posterior to the superciliary arches, lie between the outer and inner tables of the frontal bone. Each usually underlies a triangular area on the surface, its angles formed by the nasion, a point 3 cm above the nasion and the junction of the medial third and lateral two-thirds of the supraorbital margin. The two sinuses are rarely symmetrical, the septum between them usually deviating from the median plane. Their average dimensions are: height 3.2 cm; breadth 2.6 cm; depth 1.8 cm. Each extends upwards above the medial part of the eyebrow and back into the medial part of the roof of the orbit. The frontal sinus is sometimes divided into a number of communicating recesses by incomplete bony septa. Rarely, one or both sinuses may be absent, and racial differences have been reported. The prominence of the superciliary arches is no indication of the presence or size of the frontal sinuses. The part of the sinus extending upwards in the frontal bone may be small and the orbital part large, or vice versa. Sometimes one sinus may overlap in front of the other. A sinus may extend posteriorly as far as the lesser wing of the sphenoid bone but does not invade it. The morphology of the frontal sinus may be specific enough to allow identification of individuals from radiological evidence for forensic purposes. page 575 page 576
The aperture of each frontal sinus usually opens into the anterior part of the corresponding middle meatus by the ethmoidal infundibulum as a hiatus or, more often, as a more elongated frontonasal duct. It may also open medial to the hiatus semilunaris. Rudimentary or absent at birth, the frontal sinuses are generally well
developed between the seventh and eighth years, but reach full size only after puberty. They are more prominent in males, giving the forehead an obliquity contrasting with the vertical or convex profile typical of children and females. However, the shape and size of the frontal sinus is highly variable and may be hypoplastic or even absent. In the presence of a persistent metopic suture, the frontal sinuses develop separately on either side of the suture, which can be helpful in excluding frontal fractures. Vascular supply and innervation The arterial supply of the frontal sinuses is from the supraorbital and anterior ethmoidal arteries. The veins drain into the anastomotic vein in the supraorbital notch connecting the supraorbital and superior ophthalmic veins. Lymphatic drainage is to the submandibular nodes. The sinuses are innervated by branches from the supraorbital branch of the ophthalmic nerve.
SPHENOIDAL SINUS (Figs 27.13A, 32.8, 32.12, 32.14) The paired sphenoidal sinuses lie posterior to the upper part of the nasal cavity, within the body of the sphenoid bone. As the sphenoidal septum often deviates from the midline, the sinuses are often unequal in size. They also vary in size and may be further subdivided by accessory septa, especially in the region of former synchondroses. Occasionally one overlaps the other above and, rarely, they intercommunicate. One or both may approach closely to the optic canal or even partly encircle it. The bone overlying the optic nerves and the internal carotid arteries which lie in the lateral wall of the sinus may be dehiscent. The sinuses have average dimensions of: vertical height 2 cm; transverse breadth 1.8 cm; anteroposterior depth 2.1 cm. The degree of pneumatization is highly variable. In nearly 50% they extend into the greater and lesser wing of the sphenoid or the pterygoid process, and may also invade the basilar part of the occipital bone. Gaps in their osseous walls may occasionally leave their mucosa in contact with the overlying dura mater. Bony ridges, produced by the internal carotid artery or pterygoid canal, may project into the sinuses from their lateral walls and floor respectively. A posterior ethmoidal sinus may extend posterosuperior to the relatively smaller sphenoidal sinuses. UPDATE Date Added: 09 August 2005 Shanida Helena Nataraja, PhD (Dianthus Medical Limited) Update: Surgical anatomy of the natural ostium of the sphenoid sinus. The location of the natural ostium of the sphenoid sinus is important in a range of different surgical procedures, including the assessment of inflammatory lesions and tissue biopsy of sphenoid sinus tumors. In this study of 100 sagittally sectioned cadaveric heads (with an intact nasal septum), the distance and angle between the anterior portion of the nasal cavity and the natural ostium of the sphenoid sinus were measured, as was the location of the natural ostium on the anterior wall of the sphenoid sinus. The distance between the limen nasi to the inferior aspect of the natural ostium was 56.5 ± 3.2mm, and the angle between a line passing through the limen nasi to the inferior aspect of the natural ostium and a horizontal line containing the limen nasi was 35.9° ± 3.8°. The distance between the inferior aspect of the natural ostium and the sill was 62.7 ± 9.0mm. The angle between the line connecting the sill and the inferior aspect of the natural ostium and the horizontal line containing the sill was 34.3° ± 3.8°. The natural ostium was located in the middle of the anterior wall of the sphenoid sinus: the distance between the inferior aspect of the natural ostium and the skull was 10.6 ± 3.0mm, whereas that between the natural ostium and the posteroinferior end of superior turbinate was 10.3 ± 4.3mm. In the majority of cases (83%), the natural ostium drained medially into the nasal septum, whereas in the remaining cases it drained laterally. The results of this study suggest that the posteroinferior end of the superior turbinate is the most appropriate anatomic landmark for the identification of the natural ostium of the sphenoid sinus. Kim HU, Kim SS, Kang SS et al: Surgical anatomy of the natural ostium of the sphenoid sinus.
Laryngoscope 111(9):1599-1602, 2001. Medline
Similar articles
The sphenoidal sinuses are related above to the optic chiasma and hypophysis cerebri and on each side to the internal carotid artery and cavernous sinus (Fig 32.14). If the sinuses are small, they lie anterior to the hypophysis cerebri. The anterior midline septum often becomes deviated to one side posteriorly and identification of this septation is important prior to trans-sphenoidal surgery.
Figure 32.14 Horizontal section of head showing ethmoidal and sphenoidal sinuses. (By permission from Berkovitz BKB, Moxham BJ 2002 Head and Neck Anatomy. London: Martin Dunitz.)
The aperture of each sphenoidal sinus opens into the corresponding sphenoethmoidal recess high on the anterior wall of the sinus. At birth the sinuses are minute cavities, and their main development occurs after puberty. Vascular supply and innervation The arterial supply of the sphenoidal sinus is via the posterior ethmoidal branch of the ophthalmic artery and nasal branch of the sphenopalatine artery. Venous drainage is through the posterior ethmoidal vein draining into the superior ophthalmic vein. Lymph drainage is to the retropharyngeal nodes. The sensory nerve supply arises from the posterior ethmoidal nerves, while parasympathetic secretomotor fibres are derived from orbital branches of the pterygopalatine ganglion. UPDATE Date Added: 08 May 2006 Abstract: Optic nerve position and paranasal sinus pneumatization patterns studied by computed tomography Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=15490566&query_hl=26&itool=pubmed_docsum Software-enabled computed tomography analysis of the carotid artery and sphenoid sinus
pneumatization patterns. Batra PS, Citardi MJ, Gallivan RP et al: Am J Rhinol 18:203-208, 2004. UPDATE Date Added: 08 May 2006 Abstract: Anatomy of sphenoid sinus studied by quantitative computer-aided computed tomography Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=15283492&query_hl=25&itool=pubmed_docsum Quantitative computer-aided computed tomography analysis of sphenoid sinus anatomical relationships. Citardi MJ, Gallivan RP, Batra PS et al: Am J Rhinol 18:173-178, 2004.
ETHMOIDAL SINUSES (Figs 27.19C, 32.13, 32.14) Ethmoidal sinuses are small, thin-walled cavities in the ethmoidal labyrinth, completed by the frontal, maxillary, lacrimal, sphenoid and palatine bones. They range from 3 large to 18 small sinuses on each side, and their openings into the nasal cavity are also very variable in position. They lie between the upper part of the nasal cavity and the orbit, separated from the latter by the paper-thin lamina papyracea or orbital plate of the ethmoid (a poor barrier to infection that may therefore spread into the orbit). Pneumatization may extend into the middle concha in 4-12% of individuals and into the body and wings of the sphenoid bone lateral to the sphenoid sinus. Although traditionally divided into anterior, middle and posterior ethmoidal air cells, the ethmoidal sinuses are now commonly considered by clinicians as consisting of anterior and posterior groups on each side, the middle ethmoidal air cells being incorporated into the anterior group. The groups are separated from each other by the basal lamella of the middle concha which may be indented by cells from either group so that it forms a rather tortuous barrier between the two groups. They are, however, distinguished by their sites of communication with the nasal cavity. In each group the sinuses are only partially separated by incomplete bony septa. The ethmoidal sinuses, though small, are of clinical importance at birth because they are susceptible to inflammation. They grow rapidly between 6 and 8 years and after puberty. Anterior group
Peri-infundibular sinuses (anterior ethmoidal air cells) There are up to 11 anterior ethmoidal air cells and they open into the ethmoidal infundibulum or the frontonasal duct by one or more orifices. The most anterior group, the agger nasi cells, invaginate beneath the ridge of the same name on the lateral wall of the nasal cavity anteriorly, and are medial relations of the lacrimal sac and duct. Larger anterior and middle cells may develop medially beneath the orbital floor and are known as Haller's cells, and the most anterior supraorbital ethmoidal sinus cells may encroach on the frontal sinus. Bullar sinuses (middle ethmoidal air cells) There are usually less than three middle ethmoidal air cells. They open into the middle meatus by one or more orifices on or above the ethmoidal bulla. Posterior group
The posterior group of ethmoidal air cells vary in number from 1 to 7, and usually open by a single orifice into the superior meatus, although one may open into the supreme meatus when present, and one or more into the sphenoidal sinus. The posterior group lies very close to the optic canal and optic nerve. Vascular supply and innervation
The ethmoidal sinuses receive their arterial supply from nasal branches of the sphenopalatine artery and the anterior and posterior ethmoidal branches of the ophthalmic artery. Venous drainage is by the corresponding veins. The lymphatics of the anterior group drain to the submandibular nodes, and those of the posterior group to the retropharyngeal nodes. The sensory innervation is from the anterior
and posterior ethmoidal branches of the ophthalmic nerve, and the orbital branch of the pterygopalatine ganglion supplies parasympathetic secretomotor fibres. UPDATE Date Added: 02 May 2006 Abstract: Description of anatomic topographic landmarks for transethmoidal approach to optic canal Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=15336879&query_hl=23&itool=pubmed_docsum Transethmoidal approach to the optic canal: Surgical and radiological microanatomy. Akdemir G, Tekdemir I, Altin L: Surg Neurol 62:268-274, 2004.
MAXILLARY SINUS (Figs 27.25C, 32.13, 32.15) page 576 page 577
Figure 32.15 A coronal section through the nasal cavity, viewed from the posterior aspect. On the right side the plane of the section is more anterior. The normal orifice of the maxillary sinus is shown on the right side and a not uncommon accessory orifice on the left side.
The maxillary sinus is the largest of the paranasal sinuses and is situated in the body of the maxilla. It is pyramidal in shape and its thin walls correspond to the orbital (roof), alveolar (floor), facial (anterior) and infratemporal (posterior) aspects of the maxilla. Its lateral, truncated apex extends into the zygomatic process, and sometimes even extends into the zygomatic bone forming the zygomatic recess which produces the V-shaped shadows over the antra on a lateral radiograph. Its base is medial and provides much of the lateral wall of the nasal cavity. Its posterior wall contains alveolar canals that may produce ridges in the sinus and that also conduct posterior superior alveolar vessels and nerves to molar teeth. The floor of the sinus is formed by the alveolar process and it often lies below the nasal floor. It is related to the roots of the teeth, especially the second premolar and first molar. However, as the size of the sinus varies, it may extend anteriorly to incorporate the first premolar, and sometimes even the canine, and posteriorly to the third molar tooth. Defects in the bone overlying the roots are not uncommon. The infraorbital canal forms a ridge in the roof of the sinus, and
exhibits dehiscences in c.14% of cases. It gives off a fine canal (canalis sinuosus) containing the anterior superior alveolar nerve and vessels that groove the anterior wall. The maxillary sinus may be incompletely divided by septa, complete septa being present only very rarely. The bony medial wall of the sinus is deficient posterosuperiorly at the maxillary hiatus. This large opening is partially closed by the inferior concha and the uncinate process of the ethmoid bone, forming an ostium and anterior and posterior fontanelles. The ostium opens into the middle meatus, usually via the middle part of the hiatus semilunaris: it lies in the angle between the roof and the medial wall and forms the focus of the directional beating of the cilia of the sinus mucoperiosteum. The fontanelles are covered only by periosteum and mucosa and may contain accessory ostia which may be visible on axial CT. All of the openings are nearer the roof than the floor of the sinus which means that the natural drainage of the maxillary sinus is reliant on an intact mucociliary clearance system. Gravitational drainage may be affected by puncture in the lateral wall of the inferior nasal meatus, which is nearer the level of the floor of the sinus. This is useful in disorders of mucociliary clearance such as cystic fibrosis. Because of the extreme thinness of the sinus walls, a tumour may push up the orbital floor and displace the eyeball; project into the nasal cavity, causing nasal obstruction and bleeding; protrude onto the cheek, causing swelling and numbness when the infraorbital nerve is damaged; spread back into the infratemporal fossa, causing restriction of mouth opening due to pterygoid muscle damage and pain; or spread down into the mouth, loosening teeth and causing malocclusion of the teeth. Extraction of molar teeth may damage the floor, and impact may fracture its walls. Hypoplasia of one maxillary antrum is present in up to 10% of the population, and results in increased density on the plain films. Osteomeatal complex The term osteomeatal complex, or osteomeatal unit, refers to the maxillary sinus ostium, ethmoid infundibulum, hiatus semilunaris and frontal recess. It is the final common pathway for drainage of secretions from the maxillary, frontal, anterior and middle ethmoid sinuses into the middle meatus, and obstruction plays a pivotal role in the development and persistence of sinusitis. Coronal high resolution CT (HRCT) reveals these structures in exquisite detail, and is the imaging modality of choice. Vascular supply and innervation The arterial supply of the maxillary sinus is derived mainly from the maxillary artery via anterior, middle and superior posterior alveolar branches, and infraorbital and greater palatine arteries. Veins corresponding to the arteries drain into the facial vein or pterygoid venous plexus. Lymph drainage is to the submandibular nodes. The nerve supply is derived from the maxillary nerve via the infraorbital and anterior, middle and posterior superior alveolar nerves. For further details concerning the dimensions of the various air sinuses and the size and positions of their openings the reader is referred to Lang (1989).
SPREAD OF INFECTION TO CRANIAL CAVITY page 577 page 578
Suppuration in the paranasal sinuses occurs frequently. The state of health of the paranasal sinuses is dependent on the state of health of the 'prechambers' through which the sinuses exchange air with the nasal airway and drain mucus. The middle meatus and hiatus semilunaris act as the prechamber for the anterior ethmoidal, frontal and maxillary sinuses, the osteomeatal complex, and can be readily examined with a fibreoptic endoscope. The superior meatus and sphenoethmoidal recess acts as the prechamber for the posterior ethmoidal and sphenoidal sinuses. Obstruction of these prechambers by swollen mucous membranes, polyps or tumours interferes with the ventilation of the sinuses and arrests mucus clearance, which in turn causes overgrowth of bacteria or viruses
and consequent sinusitis. Endoscopic examination will show infected mucus draining from the anterior sinus group from the osteomeatal complex onto the superomedial surface of the inferior concha into the pharynx anterior to the pharyngotympanic tubal orifice whilst mucus from the posterior group will drain into the pharynx above and behind the tubal orifice. Maxillary sinus infection may also spread from the infected teeth.
© 2008 Elsevier
PTERYGOPALATINE FOSSA OSTEOLOGY OF THE PTERYGOPALATINE FOSSA The pterygopalatine fossa is a small pyramidal space below the apex of the orbit on the lateral side of the skull. The posterior boundary is the root of the pterygoid process and adjoining anterior surface of the greater wing of the sphenoid, and the anterior boundary is the superomedial part of the infratemporal surface of the maxilla. The perpendicular plate of the palatine bone, with its orbital and sphenoidal processes forms the medial boundary, and the pterygomaxillary fissure is the lateral boundary. The fossa communicates with the nasal cavity via the sphenopalatine foramen, the orbit via the medial end of the inferior orbital fissure, and the infratemporal fossa via the pterygomaxillary fissure. The latter lies between the back of the maxilla and the pterygoid process of the sphenoid and transmits the maxillary artery. There are two openings in the posterior wall of the pterygopalatine fossa, namely the foramen rotundum, which transmits the maxillary nerve, and the pterygoid canal, which transmits the nerve of the pterygoid canal. When the anterior aspect of the pterygoid plate is examined in a disarticulated sphenoid, it will be seen that the foramen rotundum lies above and lateral to the pterygoid canal (Fig. 27.13).
CONTENTS OF THE PTERYGOPALATINE FOSSA The main contents of the pterygopalatine fossa are the maxillary nerve and many of its branches, the pterygopalatine ganglion and the terminal (third) part of the maxillary artery. Maxillary nerve (Figs 41.25, 30.6)
The maxillary division of the trigeminal nerve is wholly sensory. It leaves the skull via the foramen rotundum, which leads directly into the posterior wall of the pterygopalatine fossa. Crossing the upper part of the pterygopalatine fossa, the nerve gives off two large ganglionic branches which contain fibres destined for the nose, palate and pharynx, and these pass through the pterygopalatine ganglion without synapsing. It then inclines sharply laterally on the posterior surface of the orbital process of the palatine bone and on the upper part of the posterior surface of the maxilla in the inferior orbital fissure (which is continuous posteriorly with the pterygopalatine fossa): it lies outside the orbital periosteum, and gives off its zygomatic, and then posterior superior alveolar branches. About halfway between the orbital apex and the orbital rim the maxillary nerve turns medially to enter the infraorbital canal as the infraorbital nerve. The subsequent course of the maxillary nerve is described on page 513. The maxillary nerve gives off many of its branches in the pterygopalatine fossa. They can be subdivided into those that come directly from the nerve, and those that are associated with the pterygopalatine parasympathetic ganglion. Named branches from the main trunk are meningeal, ganglionic, zygomatic, posterior, middle and anterior superior alveolar and infraorbital nerves. Named branches from the pterygopalatine ganglion are orbital, nasopalatine, posterior superior nasal, greater (anterior) palatine, lesser (posterior) palatine and pharyngeal.
Meningeal nerve
The meningeal branch of the maxillary nerve arises within the middle cranial fossa and runs with the middle meningeal vessels. It contributes to the innervation of the dura mater. Ganglionic branches
There are usually two ganglionic branches that connect the maxillary nerve to the pterygopalatine ganglion. Zygomatic nerve
The zygomatic branch of the maxillary nerve leaves the pterygopalatine fossa through the inferior orbital fissure together with the maxillary nerve. Its subsequent course is described on page 513. Posterior superior alveolar nerve
The posterior superior alveolar nerve leaves the maxillary nerve in the pterygopalatine fossa. Its subsequent course and distribution is described in detail on page 601. Infraorbital nerve
The infraorbital nerve can be regarded as the terminal branch of the maxillary nerve. It leaves the pterygopalatine fossa to enter the orbit at the inferior orbital fissure, and its subsequent course and distribution are described on page 513. Pterygopalatine ganglion
The pterygopalatine ganglion is the largest of the peripheral parasympathetic ganglia. It is placed deeply in the pterygopalatine fossa, near the sphenopalatine foramen, and anterior to the pterygoid canal and foramen rotundum. It is flattened, reddish-grey in colour, and lies just below the maxillary nerve as it crosses the pterygopalatine fossa. The majority of the 'branches' of the ganglion are connected with it morphologically, but not functionally, because they are primarily sensory branches of the maxillary nerve. Thus they pass through the ganglion without synapsing, and enter the maxillary nerve through its ganglionic branches, but they convey some parasympathetic fibres to the palatine, pharyngeal and nasal mucous glands. Preganglionic parasympathetic fibres destined for the pterygopalatine ganglion run initially in the greater petrosal branch of the facial nerve, and then in the nerve of the pterygoid canal (Vidian nerve), after the greater petrosal unites with the deep petrosal nerve. The nerve of the pterygoid canal enters the ganglion posteriorly. Postganglionic parasympathetic fibres leave the ganglion and join the maxillary nerve via a ganglionic branch, then travel via the zygomatic and zygomaticotemporal branches of the maxillary nerve to the lacrimal gland (see Fig. 30.6A). Preganglionic secretomotor fibres of uncertain origin also travel in the nerve of the pterygoid canal. They synapse in the pterygopalatine ganglion, and postganglionic fibres are distributed to palatine, pharyngeal and nasal mucous glands via palatine and nasal branches of the maxillary nerve (Fig. 32.9). Postganglionic sympathetic fibres pass through the ganglion without synapsing
and supply blood vessels and orbitalis. They arise in the superior cervical ganglion and travel via the internal carotid plexus and deep petrosal nerve to enter the pterygopalatine ganglion within the nerve of the pterygoid canal. General sensory fibres destined for distribution via orbital, nasopalatine, superior alveolar, palatine and pharyngeal branches of the maxillary division of the trigeminal nerve run through the ganglion without synapsing. Orbital branches
Fine orbital branches enter the orbit through the inferior orbital fissure and supply orbital periosteum. Some fibres also pass through the posterior ethmoidal foramen to supply the sphenoidal and ethmoidal sinuses. The orbital branches probably join branches of the internal carotid nerve to form a 'retro-orbital' plexus from which orbital structures such as the lacrimal gland and orbitalis receive an autonomic innervation. Nasopalatine nerve (Fig. 32.9)
The nasopalatine nerve leaves the pterygopalatine fossa through the sphenopalatine foramen and enters the nasal cavity. It passes across the cavity to the back of the nasal septum, runs downwards and forwards on the septum in a groove in the vomer, and then turns down through the incisive fossa in the anterior part of the hard palate to enter the roof of the mouth. When an anterior and a posterior incisive foramen exist in this fossa, the left nasopalatine nerve passes through the anterior foramen, and the right nerve passes through the posterior foramen. The nasopalatine nerve supplies the lower part of the nasal septum and the anterior part of the hard palate, where it communicates with the greater palatine nerve. Posterior superior nasal nerves (lateral and medial) page 578 page 579
The posterior superior alveolar nerves enter the back of the nasal cavity through the sphenopalatine foramen. Lateral posterior superior nasal nerves (c.6) innervate the mucosa lining the posterior part of the superior and middle nasal conchae and the posterior ethmoidal sinuses. Two or three medial posterior superior nasal nerves cross the nasal roof below the opening of the sphenoidal sinus to supply the mucosa of the posterior part of the roof and of the nasal septum. Palatine nerves (greater and lesser)
The greater and lesser palatine nerves pass downwards from the pterygopalatine ganglion through the greater palatine canal. The greater palatine nerve descends through the greater palatine canal, emerges on the hard palate from the greater palatine foramen and runs forwards in a groove on the inferior surface of the bony palate almost to the incisor teeth. It supplies the gingivae, mucosa and glands of the hard palate and also communicates with the terminal filaments of the nasopalatine nerve. In the greater palatine canal it gives off posterior inferior nasal branches that emerge through the perpendicular plate of the palatine bone and ramify over the inferior nasal concha and walls of the middle and inferior
meatuses. As it leaves the greater palatine canal, it gives off branches which are distributed to both surfaces of the adjacent part of the soft palate. UPDATE Date Added: 30 August 2005 Shanida Helena Nataraja, PhD (Dianthus Medical Limited) Update: Endoscopic study of the greater palatine nerve. The pathways of the greater palatine nerve (GPN) have been examined in an endoscopic study of seven cadaveric heads. Five heads were dissected bilaterally, one was dissected unilaterally, and one was injected with latex and sectioned coronally and dissected bilaterally. The sphenopalatine and posterior nasal arteries were separated by a thin layer of fibrous tissue and crossed almost perpendicular and just anterior to the GPN. The GPN was located 2 to 7mm from the inferior medial bony margin of the sphenopalatine foramen (SPF) and traveled inferiorly and anteriorly in the greater palatine canal to the greater palatine foramen. The greater palatine artery joined the GPN in the greater palatine canal inferior to the SPF. Several anatomic landmarks can be used to guide the safe surgical dissection of the pterygomaxillary fossa. The surgeon should attempt to remain superficial or equal to the depth of the internal maxillary artery, because the vascular structures lie anteriorly to the nervous structures. The proximity of the GPN to the lateral wall of the nasal cavity inferior to the SPF warrants avoiding aggressive instrumentation in this region. Resection of the thickened orbital process of the palatine bone should be avoided given the proximity of the underlying foramen rotundum, maxillary nerve and GPN. The SPF and lateral nasal wall should be left intact when dissecting out vascular structures in order to avoid damaging the nerve. Adherence to these guidelines should decrease the risk of damage to the GPN during surgical interventions. Mellema JW, Tami TA: An endoscopic study of the greater palatine nerve. Am J Rhinol 18(2):99-103, 2004.
The lesser (middle and posterior) palatine nerves are much smaller than the greater palatine nerve. They descend through the greater palatine canal, from which they diverge low down to emerge through the lesser palatine foramina in the tubercle or pyramidal process of the palatine bone. They innervate the uvula, tonsil and soft palate. Fibres conveying taste impulses from the palate probably pass via the palatine nerves to the pterygopalatine ganglion. They pass through the ganglion without synapsing, and leave via the greater petrosal nerve. Their cell bodies are located in the facial ganglion and their central processes pass via the sensory root of the facial nerve (nervus intermedius) to the gustatory nucleus in the nucleus of the tractus solitarius. Pharyngeal nerve (Fig. 32.9)
The pharyngeal branch of the maxillary nerve leaves the pterygopalatine ganglion posteriorly. It passes through the palatovaginal canal with the pharyngeal branch of the maxillary artery and supplies the mucosa of the nasopharynx behind the pharyngotympanic tube. Vascular supply Maxillary artery
The maxillary artery passes through the pterygomaxillary fissure from the infratemporal fossa into the pterygopalatine fossa, where it terminates as the third part of the maxillary artery. This part of the artery gives branches which accompany branches of the maxillary nerve (including those associated with the pterygopalatine ganglion). UPDATE Date Added: 13 September 2005 Shanida Helena Nataraja, PhD (Dianthus Medical Limited) Update: Anatomic study of maxillary artery at the pterygomaxillary fissure in Thai population. An evaluation was made of the relationship between the maxillary artery (MA) at the pterygomaxillary fissure (PMF) and the pterygomaxillary junction (PMJ) in an anatomical study of 100 Thai adult cadavers. The mean ± SD age at death of the 50 female and 50 male subjects was 64.5 ± 12.8 years (range: 33 to 87 years); 42.5 percent of cadavers were noted to have upper molar teeth. The MA was found to enter the pterygopalatine fossa (PPF) 23.5 ± 2.5mm (mean ± SD) above the inferior margin of the PMJ. Its mean external diameter as it entered the PPF was 2.8 ± 0.6 mm. The vertical heights of the PMJ, maxillary tuberosity, and posterior maxilla were 19.5 ± 2.3mm, 6.1 ± 2.7mm, and 25.6 ± 3.3mm, respectively. None of the study measurements differed significantly between the left and right side of each cadaver, nor between female and male cadavers, with the exception of the mean distance from the most inferior margin of the PMJ to the inferior border of the MA, which was longer in males than females. Furthermore, the mean heights of the maxillary tuberosity and posterior maxilla were longer in cadavers with upper molar teeth than in those without. The MA is particularly vulnerable to damage during osteotomy of the medial and lateral maxillary sinus walls, pterygomaxillary dysjunction, or the down fracturing of the maxilla. The findings of this study indicate that pterygomaxillary dysjunction can be done both safely and without damage to the MA or hemorrhage. Damage to the MA can be avoided by the inferior placement of the osteotome in the PMJ and its direction medially and anteriorly but not superiorly. When placed properly, the margin of safety from the superior cutting edge of an osteotome (15mm cutting edge to the MA) is approximately 8mm in Thai adults. These results were confirmed in another anatomic study of 15 Korean adult cadavers (30 sides). In 85.7 percent of cadavers, the following branching pattern was observed: posterior superior alveolar artery, infraorbital artery, artery of the pterygoid canal, descending palatine artery, and sphenopalatine artery. The posterior superior alveolar artery (PSAA) and the infraorbital artery (IOA) exhibited two patterns of branching from the MA. Either the PSAA and the IOA arose jointly from the short branch of the MA (57.1% of cases) or they branched separately from the MA (42.9% of cases). The mean (± SD) vertical distance between the inferior margin of the PMJ and the PSAA was 15.2 (± 2.4) mm, whereas that between the PMJ and the IOA was 32.3 (± 3.7) mm. The descending palatine artery (DPA) divided from the MA at a mean (± SD) vertical distance of 24.8 (± 2.8) mm from the inferior margin of the PMJ. In 95.2 percent of cases, the greater and lesser palatine arteries were divided from the short DPA. However, in a single case (4.8 percent), these arteries were divided directly from the MA rather than the DPA. The patterns of the MA in the posterior maxilla were
classified into one of five types according to the patterns of the sphenopalatine (SPA) and DPA: Y type (180° pattern; 19 percent of cases), intermediate type (90° pattern; 33.3 percent of cases), T type (>90° pattern; 23.8 percent of cases), M type (0° pattern; 14.3 percent), and other type (i.e. none of the other types; 9.6 percent of cases). These data indicate that pterygomaxillary dysjunction can be performed safely and without damage to the MA. In order to prevent damage to the MA, the osteotome should be placed inferiorly in the PMJ and then directed medially and anteriorly. Apinhasmit W, Methathrathip D, Ploytubtim S et al: Anatomical study of the maxillary artery at the pterygomaxillary fissure in a Thai population: its relationship to maxillary osteotomy. J Med Assoc Thai 87(10):1212-1217, 2004. Choi J, Park HS: The clinical anatomy of the maxillary artery in the pterygopalatine fossa. J Oral Maxillofac Surg 61(1):72-78, 2003.
Posterior superior alveolar artery The posterior superior alveolar artery arises from the maxillary artery within the pterygopalatine fossa and runs through the pterygomaxillary fissure onto the maxillary tuberosity. It gives off branches which penetrate the bone here to supply the maxillary molar and premolar teeth and the maxillary air sinus, and other branches that supply the buccal mucosa. Occasionally the posterior superior alveolar artery arises from the infraorbital artery. Infraorbital artery The infraorbital artery enters the orbit through the inferior orbital fissure. It runs on the floor of the orbit in the infraorbital groove and infraorbital canal and emerges onto the face at the infraorbital foramen to supply the lower eyelid, part of the cheek, the side of the external nose, and the upper lip. While within the infraorbital canal it gives off the anterior superior alveolar artery which runs downwards to supply the anterior teeth and the anterior part of the maxillary sinus. Artery of the pterygoid canal The artery of the pterygoid canal passes through the pterygoid canal and supplies part of the pharyngotympanic tube, tympanic cavity, and the upper part of the pharynx. Pharyngeal artery The pharyngeal branch of the maxillary artery passes through the palatovaginal canal, accompanying the nerve of the same name, and is distributed to the mucosa of the nasal roof, nasopharynx, sphenoidal air sinus and pharyngotympanic tube. Greater (descending) palatine artery The greater palatine artery leaves the pterygopalatine fossa through the greater (anterior) palatine canal, within which it gives off two or three lesser palatine arteries. The greater palatine artery supplies the inferior meatus of the nose, then passes onto the roof of the hard palate at the greater palatine foramen and runs forwards to supply the hard palate and the palatal gingivae of the maxillary teeth. It gives off a branch that runs up into the incisive canal to anastomose with the sphenopalatine artery, and so contribute to the arterial supply of the nasal septum. The lesser palatine arteries emerge onto the palate through the lesser
(posterior) palatine foramen, or foramina, and supply the soft palate. Sphenopalatine artery The sphenopalatine branch of the maxillary artery passes through the sphenopalatine foramen and enters the nasal cavity posterior to the superior meatus. From here its posterior lateral nasal branches ramify over the conchae and meatuses, anastomosing with the ethmoidal arteries and nasal branches of the greater palatine artery to supply the frontal, maxillary, ethmoidal and sphenoidal air sinuses. The sphenopalatine artery next crosses anteriorly on the inferior surface of the sphenoid and ends on the nasal septum in a series of posterior septal branches which anastomose with the ethmoidal arteries. A branch descends on the vomer to the incisive canal to anastomose with the greater palatine artery and the septal branch of the superior labial artery. UPDATE Date Added: 09 August 2005 Shanida Helena Nataraja, PhD (Dianthus Medical Limited) Update: Surgical anatomy of the sphenopalatine artery. Arterial damage, and thus bleeding, may occur during ethmoidectomy, middle meatal antrostomy, conchotomy and endoscopic ligation of the sphenopalatine artery. This study describes the surgical anatomy of the sphenopalatine artery in midsagittal sections through a series of 50 adult Korean cadaveric heads. All heads were injected with red latex via the external carotid artery, and the mucosa of the sphenopalatine foramen was removed to expose the underlying sphenopalatine artery. The foramen was usually (90%) located between the middle concha and posterior horizontal border of the lamella of the superior concha within the superior meatus. Less often (10%) the foramen extended beyond the posterior horizontal portion of the lamella of the superior concha. The sphenopalatine artery divided into two (76%), three (22%) or four (2%) branches before exiting the sphenopalatine foramen. The nasal septal branch and the posterior lateral nasal artery were consistently observed in all specimens. In the majority of heads (72%), the superior concha was supplied by the septal artery. Less often (18%) it was supplied by the posterior lateral nasal artery. The feeding vessels to the middle conchal branch arose from the proximal (88%) or distal (12%) portions of the posterior lateral nasal artery. The latter traveled downwards across the perpendicular plate of the palatine bone to course either just posterior (42%), close to (20%), posterior (20%) or anterior (18%) to the posterior wall of the maxillary sinus. The major feeding arteries to the fontanelle were derived from the inferior (50%) and middle (20%) conchal branches. The posterior lateral nasal artery gave origin to an inferior conchal branch. Lee HY, Kim HU, Kim SS et al: Surgical anatomy of the sphenopalatine artery in lateral nasal wall. Laryngoscope 112(10):1813-1818, 2002.
Nose bleeds
The vast majority of nose bleeds, particularly in children, occur as a result of digital trauma to the anastomosis of arterioles and veins in Little's area (Kiesselbach's plexus), on the nasal septum just inside the nasal vestibule. In older patients brisker bleeding may occur as a result of the spontaneous rupture of arteries further back in the nose. These may be controlled by applying pressure with a nasal pack but where this fails knowledge of the pattern of arterial blood
supply to the nasal cavity permits interruption of the appropriate blood supply by ligation or embolization of the feeding vessel. The sphenopalatine artery may be ligated as it enters the nose under endoscopic visualization. The ethmoidal arteries may be exposed within the orbit and ligated to arrest bleeding high up in the nasal cavity. The maxillary artery may be exposed surgically behind the posterior wall of the maxillary sinus and ligated. Alternatively it may be identified radiologically, by instilling a radio-opaque dye, allowing it to be blocked by releasing objects into the artery to embolize and block the bleeding vessel. UPDATE Date Added: 02 August 2005 Shanida Helena Nataraja, PhD (Dianthus Medical Limited) Update: Endoscopic anatomy of sphenopalatine and posterior nasal arteries: implications for the endoscopic management of epistaxis. An understanding of the anatomic relationship between the sphenopalatine and posterior nasal arteries is essential for the appropriate surgical management of refractory posterior epistaxis. Accordingly, this relationship has recently been defined in an endoscopic study of the sphenopalatine foramen. Eighteen dissections were undertaken on nine fresh cadaver heads. A tenth cadaver head was injected with red and blue latex to reveal the arteries and veins, respectively. This specimen was then bisected sagittally, the sphenopalatine foramen dissected, and the arteries identified. In 42% of cases, the sphenopalatine artery emerged from a slightly elongated foramen and traveled posteriorly to the posterior nasal artery. This configuration would, therefore, require significantly more posterior dissection in order to reveal the sphenopalatine artery. In a further 42% of cases, the sphenopalatine artery emerged from a separate foramen located posteriorly to the sphenopalatine foramen and, in 16% of cases, the two arteries branched within the foramen and, therefore, emerged from the foramen adjacent to each other. The findings of this study suggest that dissection should continue posteriorly following the identification and ligation of the posterior nasal artery to include the location and ligation of the sphenopalatine artery. Ligation of the posterior nasal artery is necessary to control bleeding from the lateral nasal wall, whereas sphenopalatine artery is necessary to control bleeding from the posterior septum. Once both of these vascular elements have been ligated, the procedure should be successful and epistaxis appropriately controlled. Schwartzbauer HR, Shete M, Tami TA: Endoscopic anatomy of the sphenopalatine and posterior nasal arteries: implications for the endoscopic management of epistaxis. Am J Rhinol 17(1):63-66, 2003. Medline Similar articles
Veins of the pterygopalatine fossa
The veins of the pterygopalatine fossa are small and variable. The most consistent is the sphenopalatine vein which drains the posterior aspect of the nose, then passes into the pterygopalatine fossa through the sphenopalatine foramen and ultimately drains into the pterygoid venous plexus via the pterygomaxillary fissure. The inferior ophthalmic vein passes to the pterygoid venous plexus through the inferior orbital fissure. REFERENCES Doig TN, McDonald SW, McGregor OA 1998 Possible routes of spread of carcinoma of the maxillary sinus to the oral cavity. Clin Anat 11: 149-56. Medline Similar articles Full article Emanuel JM 1998 Epistaxis. In: Cummings CW et al (eds) Otolaryngology Head and Neck Surgery, vol. 2.
3rd edition. St Louis: Mosby: 852-65. Describes the blood supply to the nose and the surgical approaches to control epistaxis. Lang J 1989 Clinical Anatomy of the Nose, Nasal Cavity and Paranasal Sinuses. Stuttgart: Thième. McGowan DA, Baxter PW, James J 1993 The Maxillary Sinus. Oxford: Wright. Navarro JAC 1997 The Nasal Cavity and Paranasal Sinuses: Surgical Anatomy. Berlin: Springer. Stammberger H, Kennedy DW (eds) 1995 Paranasal sinuses: anatomic terminology and nomenclature. Ann Otol Rhinol Laryngol (suppl 167) 104(10) Part 2: 7-16. Traxler H, Windisch A, Geyerhofer U, Surd R, Solar P, Firbas W 1999 Arterial blood supply of the maxillary sinus. Clin Anat 12: 417-21. Medline Similar articles Full article page 579 page 580
© 2008 Elsevier
33 NECK AND UPPER AERODIGESTIVE TRACT Oral cavity The mouth or oral cavity extends from the lips and cheeks externally to the anterior pillars of the fauces internally, where it continues into the oropharynx. The mouth can be subdivided into the vestibule external to the teeth and the oral cavity proper internal to the teeth. The palate forms the roof of the mouth and separates the oral and nasal cavities. The floor of the mouth is formed by the mylohyoid muscles and is occupied mainly by the tongue. The lateral walls of the mouth are defined by the cheeks and retromolar regions. Three pairs of major salivary glands (parotid, submandibular and sublingual) and numerous minor salivary glands (labial, buccal, palatal, lingual) open into the mouth. The muscles in the oral cavity are associated with the lips, cheeks, floor of the mouth and tongue. The muscles of the lips and cheeks are described with the face on page 504. The muscles of the soft palate are described with the pharynx on page 627. UPDATE Date Added: 19 April 2005 Shanida Helena Nataraja, PhD (Dianthus Medical Limited) Update: Lymph IIb nodes can be preserved in elective supraomohyoid neck dissection for oral cavity squamous cell carcinoma This study was a prospective analysis of a consecutive series of patients presenting between January 01 1997 and December 31 2001 at Yonsei University, Severence Hospital, Seoul, Korea. The aim was to determine whether level IIb lymph nodes can be saved in elective supraomohyoid neck dissection (SOHND) as treatment for squamous cell carcinoma (SCC) of the oral cavity. Seventy-four previously untreated consecutive patients (64 male; 10 female) with SCC of the oral cavity underwent SOHND. The median age of patients was 56.2 years (range: 9 to 74 years). The mean number of lymph nodes collected in the SOHND specimens was 22.2 (range: 11 to 39). The mean number of lymph nodes collected according to level was 3.8, 11.3, and 7.1 at level I, II, and III, respectively. Of the 74 patients included in this analysis, 24 (32%) had been staged as clinically N0, but were found to have pathologic positive nodes; 18 of this group had a single positive node; and 6 had positive nodes at multiple neck levels. The prevalence of metastases at level IIb was 5% (4/ 74). All of these patients had a positive neck node at level IIa. Six (8%) developed a neck recurrence within 5 years, of which five occurred in the dissected neck. Only two recurrences occurred at level IIb. The findings of this study therefore reveal that recurrence in neck level II is extremely rare. This suggests that the level IIb region can be preserved during elective SOHND in patients with SCC of the oral cavity. Lim YC, Song MH, Kim SC, Kim KM, Choi EC. Preserving level IIb lymph nodes in elective supraomohyoid neck dissection for oral cavity squamous cell carcinoma., Arch Otolaryngol Head Neck Surg., 2004;130(9):1088-91.
The mouth is concerned primarily with the ingestion and mastication of food, which is mainly the function of the teeth. The mouth is also associated with phonation and ventilation, but these are secondary functions.
© 2008 Elsevier
CHEEKS The external features of the cheeks are described on page 497. Internally, the mucosa of the cheek is tightly adherent to buccinator and is thus stretched when the mouth is opened and wrinkled when closed. Ectopic sebaceous glands may be evident as yellow patches (Fordyce's spots). Their numbers increase in puberty and in later life. Few structural landmarks are visible. The parotid duct drains into the cheek opposite the maxillary second molar tooth at a small parotid papilla. A hyperkeratinized line (the linea alba) may be seen at a position related to the occlusal plane of the teeth. In the retromolar region, a fold of mucosa which contains the pterygomandibular raphe extends from the upper to the lower alveolus. The entrance to the pterygomandibular space (which contains the lingual and inferior alveolar nerves) lies lateral to this fold and medial to the ridge produced by the anterior border of the ramus of the mandible. This is the site for injection for an inferior alveolar nerve block. Vascular supply and innervation The cheek receives its arterial blood supply principally from the buccal branch of the maxillary artery, and is innervated by cutaneous branches of the maxillary division of the trigeminal nerve, via the zygomaticofacial and infraorbital nerves, and by the buccal branch of the mandibular division of the trigeminal nerve.
© 2008 Elsevier
LIPS The external features of the lips are described on page 497. The central part of the lips contain orbicularis oris. Internally, the labial mucosa is smooth and shiny and shows small elevations caused by underlying mucous glands. The position and activity of the lips are important in controlling the degree of protrusion of the incisors. With normal (competent) lips, the tips of the maxillary incisors lie below the upper border of the lower lip, and this arrangement helps to maintain the 'normal' inclination of the incisors. When the lips are incompetent, the maxillary incisors may not be so controlled and the lower lip may even lie behind them, thus producing an exaggerated proclination of these teeth. A tight, or overactive, lip musculature may be associated with retroclined maxillary incisors. Vascular supply and innervation The lips are mainly supplied by the superior and inferior labial branches of the facial artery. The upper lip is innervated by superior labial branches of the infraorbital nerve and the lower lip is innervated by the mental branch of the mandibular division of the trigeminal.
© 2008 Elsevier
ORAL VESTIBULE The oral vestibule is a slit-like space between the lips or cheeks on one side and the teeth on the other. When the teeth occlude, the vestibule is a closed space that only communicates with the oral cavity proper in the retromolar regions behind the last molar tooth on each side. Where the mucosa that covers the alveolus of the jaw is reflected onto the lips and cheeks, a trough or sulcus is formed which is called the fornix vestibuli. A variable number of sickle-shaped folds containing loose connective tissue run across the fornix vestibuli. In the midline these are the upper and lower labial frena (or frenula). Other folds may traverse the fornix near the canines or premolars. The folds in the lower fornix are said to be more pronounced than those in the upper fornix (Fig. 33.1). The upper labial frenum is normally attached well below the alveolar crest. A large frenum with an attachment near the crest may be associated with a midline gap (diastema) between the maxillary first incisors. This can be corrected by simple surgical removal of the frenum, as it contains no structures of clinical importance. Prominent frena may also influence the stability of dentures.
© 2008 Elsevier
ORAL MUCOSA The oral mucosa is continuous with the skin at the labial margins and with the pharyngeal mucosa at the oropharyngeal isthmus. It varies in structure, function and appearance in different regions of the oral cavity and is traditionally divided into lining, masticatory and specialized mucosae.
page 581 page 582
Figure 33.1 Anterior view of the dentition in centric occlusion, with the lips retracted. Note the pale pink, stippled gingivae and the red, shiny, smooth alveolar mucosa. The degree of overbite is rather pronounced and the gingiva and its epithelial attachment have receded onto the root of the upper left canine. Note frena (arrows).
LINING MUCOSA The lining mucosa is red in colour, and covers the soft palate, ventral surface of the tongue, floor of the mouth, alveolar processes excluding the gingivae and the internal surfaces of the lips and cheeks. It has a non-keratinized stratified squamous epithelium which overlies a loosely fibrous lamina propria, and the submucosa contains some fat deposits and collections of minor mucous salivary glands. The oral mucosa covering the alveolar bone - which supports the roots of the teeth - and the necks (cervical region) of the teeth is divided into two main components. That portion lining the lower part of the alveolus is loosely attached to the periosteum via a diffuse submucosa and is termed the alveolar mucosa. It is delineated from the masticatory gingival mucosa, which covers the upper part of the alveolar bone and the necks of the teeth, by a well-defined junction, the mucogingival junction. The alveolar mucosa appears dark red, the gingival appears pale pink. These colour differences relate to differences in the type of keratinization and the proximity to the surface of underlying small blood vessels which may sometimes be seen coursing beneath the alveolar mucosa.
MASTICATORY MUCOSA AND THE GINGIVAE
Masticatory mucosa, i.e. mucosa that is subjected to masticatory stress, is bound firmly to underlying bone or to the necks of the teeth, and forms a mucoperiosteum in the gingivae and palatine raphe. Gingival, palatal and dorsal lingual mucosae are keratinized or parakeratinized. The gingivae may be further subdivided into the attached gingivae and the free gingivae. Attached gingivae are firmly bound to the periosteum of the alveolus and to the teeth, whereas free gingivae, which constitute c.1 mm margin of the gingivae, lie unattached around the cervical region of each tooth. The free gingival groove between the free and attached gingivae corresponds roughly to the floor of the gingival sulcus which separates the inner surface of the attached gingivae from the enamel. The interdental papilla is that part of the gingivae which fills the space between adjacent teeth. The surface of the attached gingivae is characteristically stippled, although there is considerable inter-individual variation in the degree of stippling, and variation according to age, sex and the health of the gingivae. The free gingivae are not stippled. A mucogingival line delineates the attached gingivae on the lingual surface of the lower jaw from the alveolar mucosa towards the floor of the mouth. There is no corresponding obvious division between the attached gingivae and the remainder of the palatal mucosa because this whole surface is orthokeratinized masticatory mucosa, which is pink. A submucosa is absent from the gingivae and the midline palatine raphe, but is present over the rest of the hard palate. Posterolaterally it is thick where it contains mucous salivary glands and the greater palatine nerves and vessels, and it is anchored to the periosteum of the maxillae and palatine bones by collagenous septa.
VASCULAR SUPPLY AND LYMPHATIC DRAINAGE The gingival tissues derive their blood supply from the maxillary and lingual arteries. The buccal gingivae around the maxillary cheek teeth are supplied by gingival and perforating branches from the posterior superior alveolar artery and by the buccal branch of the maxillary artery. The labial gingivae of anterior teeth are supplied by labial branches of the infraorbital artery and by perforating branches of the anterior superior alveolar artery. The palatal gingivae are supplied primarily by branches of the greater palatine artery. The buccal gingivae associated with the mandibular cheek teeth are supplied by the buccal branch of the maxillary artery and by perforating branches from the inferior alveolar artery. The labial gingivae around the anterior teeth are supplied by the mental artery and by perforating branches of the incisive artery. The lingual gingivae are supplied by perforating branches from the inferior alveolar artery and by its lingual branch, and by the main lingual artery, a branch of the external carotid artery. No accurate description is available concerning the venous drainage of the gingivae, although it may be assumed that buccal, lingual, greater palatine and nasopalatine veins are involved. These veins run into the pterygoid plexuses (apart from the lingual veins, which pass directly into the internal jugular veins). The lymph vessels of the labial and buccal gingivae of the maxillary and
mandibular teeth unite to drain into the submandibular nodes, though in the labial region of the mandibular incisors they may drain into the submental lymph nodes. The lingual and palatal gingivae drain into the jugulodigastric group of nodes, either directly or indirectly through the submandibular nodes. Table 33-1. Nerve supply to the teeth and gingivae Nasopalatine nerve
Palatal gingivae
Greater palatine nerve
Maxilla
Posterior Anterior superior Middle superior superior alveolar alveolar nerve alveolar nerve nerve Infraorbital Posterior superior alveolar nerve Buccal gingivae nerve and buccal nerve 12345678 Tooth position Buccal nerve and perforating Mental nerve branches of inferior alveolar nerve Mandible Incisive nerve Inferior alveolar nerve Lingual nerve and perforating branches of Lingual gingivae inferior alveolar nerve
Teeth
Buccal gingivae Teeth
(By permission from Berkovitz BKB, Holland GR, Moxham BJ 2002 Oral Anatomy, Embryology and Histology, 3rd edn. Edinburgh: Mosby.)
INNERVATION (Table 33.1) The nerves supplying the gingivae in the upper jaw come from the maxillary nerve via its greater palatine, nasopalatine and anterior, middle and posterior superior alveolar branches. Surgical division of the nasopalatine nerve causes no obvious sensory deficit in the anterior part of the palate, which suggests that the territory of the greater palatine nerve reaches as far forwards as the gingivae lingual to the incisor teeth. The mandibular nerve innervates the gingivae in the lower jaw by its inferior alveolar, lingual and buccal branches.
SPECIALIZED ORAL MUCOSA The specialized mucosa which covers the anterior two-thirds of the dorsum of the tongue is described on page 584.
© 2008 Elsevier
OROPHARYNGEAL ISTHMUS (Fig. 33.2) The oropharyngeal isthmus lies between the soft palate and the dorsum of the tongue, and is bounded on both sides by the palatoglossal arches. Each palatoglossal arch runs downwards, laterally and forwards, from the soft palate to the side of the tongue and consists of palatoglossus and its covering mucous membrane. The approximation of the arches shuts off the mouth from the oropharynx, and is essential to deglutition (p. 630).
page 582 page 583
Figure 33.2 Back of the mouth showing the soft palate and oropharyngeal isthmus. A, palatoglossal fold; B, palatopharyngeal fold; C, palatine tonsil; D, uvula. (By permission from Berkovitz BKB, Holland GR, Moxham BJ 2002 Oral Anatomy, Embryology and Histology, 3rd edn. Edinburgh: Mosby.)
Figure 33.3 Cavity of the mouth. The tip of the tongue is turned upwards.
© 2008 Elsevier
FLOOR OF THE MOUTH (Fig. 33.3) The floor of the mouth is a small horseshoe-shaped region situated beneath the movable part of the tongue and above the muscular diaphragm formed by the mylohyoid muscles. A fold of tissue, the lingual frenum, extends onto the inferior surface of the tongue from near the base of the tongue. It occasionally extends across the floor of the mouth to be attached onto the mandibular alveolus. The submandibular salivary ducts open into the mouth at the sublingual papilla, which is a large centrally positioned protuberance at the base of the tongue. The sublingual folds lie on either side of the sublingual papilla and cover the underlying submandibular ducts and sublingual salivary glands. The blood supply of the floor of the mouth is described with the blood supply of the tongue (p. 587). The main muscle forming the floor of the mouth is mylohyoid. Immediately above it is geniohyoid.
MYLOHYOID (Fig. 31.10) Mylohyoid lies superior to the anterior belly of digastric and, with its contralateral fellow, forms a muscular floor for the oral cavity. It is a flat, triangular sheet attached to the whole length of the mylohyoid line of the mandible. The posterior fibres pass medially and slightly downwards to the front of the body of the hyoid bone near its lower border. The middle and anterior fibres from each side decussate in a median fibrous raphe that stretches from the symphysis menti to the hyoid bone. The median raphe is sometimes absent, in which case the two muscles form a continuous sheet, or it may be fused with the anterior belly of digastric. In about one-third of subjects there is a hiatus in the muscle through which a process of the sublingual gland protrudes. Relations The inferior (external) surface is related to platysma, anterior belly of digastric, the superficial part of the submandibular gland, the facial and submental vessels, and the mylohyoid vessels and nerve. The superior (internal) surface is related to geniohyoid, part of hyoglossus and styloglossus, the hypoglossal and lingual nerves, the submandibular ganglion, the sublingual gland, the deep part of the submandibular gland and its duct, the lingual and sublingual vessels and, posteriorly, the mucous membrane of the mouth. Vascular supply Mylohyoid receives its arterial supply from the sublingual branch of the lingual artery, the maxillary artery, via the mylohyoid branch of the inferior alveolar artery, and the submental branch of the facial artery. Innervation Mylohyoid is supplied by the mylohyoid branch of the inferior alveolar nerve. Actions Mylohyoid elevates the floor of the mouth in the first stage of deglutition. It may also elevate the hyoid bone or depress the mandible.
GENIOHYOID (Fig. 33.6) Geniohyoid is a narrow muscle which lies above the medial part of mylohyoid. It arises from the inferior mental spine (genial tubercle) on the back of the symphysis menti, and runs backwards and slightly downwards to attach to the anterior surface of the body of the hyoid bone. The paired muscles are contiguous and may occasionally fuse with each other or with genioglossus. Vascular supply The blood supply to geniohyoid is derived from the lingual artery (sublingual branch). Innervation Geniohyoid is supplied by the first cervical spinal nerve, through the hypoglossal nerve. Actions Geniohyoid elevates the hyoid bone and draws it forwards, and therefore acts partly as an antagonist to stylohyoid. When the hyoid bone is fixed, geniohyoid depresses the mandible.
© 2008 Elsevier
PALATE The palate forms the roof of the mouth and is divisible into two regions, namely, the hard palate in front and soft palate behind.
SOFT PALATE The soft palate is described with the pharynx on page 623.
HARD PALATE (Fig. 33.13A) The hard palate is formed by the palatine processes of the maxillae and the horizontal plates of the palatine bones. The hard palate is bounded in front and at the sides by the tooth-bearing alveolus of the upper jaw and is continuous posteriorly with the soft palate. It is covered by a thick mucosa bound tightly to the underlying periosteum. In its more lateral regions it also possesses a submucosa containing the main neurovascular bundle. The mucosa is covered by keratinized stratified squamous epithelium which shows regional variations and may be ortho- or parakeratinized. The periphery of the hard palate consists of gingivae. A narrow ridge, the palatine raphe, devoid of submucosa, runs anteroposteriorly in the midline. An oval prominence, the incisive papilla, lies at the anterior extremity of the raphe and covers the incisive fossa at the oral opening of the incisive canal. It also marks the position of the fetal nasopalatine canal. Irregular transverse ridges or rugae, each containing a core of dense connective tissue, radiate outwards from the palatine raphe in the anterior half of the hard palate: their pattern is unique. The submucosa in the posterior half of the hard palate contains minor salivary glands of the mucous type. These secrete through numerous small ducts, although bilaterally a larger duct collecting from many of these glands often opens at the paired palatine foveae. These depressions, sometimes a few millimetres deep, flank the midline raphe at the posterior border of the hard palate. They provide a useful landmark for the extent of an upper denture. The upper surface of the hard palate is the floor of the nasal cavity and is covered by ciliated respiratory epithelium. Vascular supply and lymphatic drainage of the hard palate The palate derives its blood supply principally from the greater palatine artery, a branch of the third part of the maxillary artery. The greater palatine artery descends with its accompanying nerve in the palatine canal, where it gives off two or three lesser palatine arteries which are transmitted through lesser palatine canals to supply the soft palate and tonsil, and anastomose with the ascending palatine branch of the facial artery. The greater palatine artery emerges on to the oral surface of the palate at the greater palatine foramen and runs in a curved groove near the alveolar border of the hard palate to the incisive canal. It ascends this canal and anastomoses with septal branches of the nasopalatine artery to supply the gingivae, palatine glands and mucous membrane. The veins of the hard palate accompany the arteries and drain largely to the
pterygoid plexus. Innervation of the hard palate (Fig. 32.9)
page 583 page 584
The sensory nerves of the hard palate are the greater palatine and nasopalatine branches of the maxillary nerve, which all pass through the pterygopalatine ganglion (p. 579) The greater palatine nerve descends through the greater palatine canal, emerges on the hard palate from the greater palatine foramen, runs forwards in a groove on the inferior surface of the bony palate almost to the incisor teeth and supplies the gums and the mucosa and glands of the hard palate. It also communicates with the terminal filaments of the nasopalatine nerve. As it leaves the greater palatine canal, it supplies palatine branches to both surfaces of the soft palate. The lesser (middle and posterior) palatine nerves, which are much smaller, descend through the greater palatine canal and emerge through the lesser palatine foramina in the tubercle of the palatine bone to supply the uvula, tonsil and soft palate. The nasopalatine nerves enter the palate at the incisive foramen and are branches of the maxillary nerve which pass through the pterygopalatine ganglion to supply the anterior part of the hard palate behind the incisor teeth. Fibres conveying taste impulses from the palate probably pass via the palatine nerves to the pterygopalatine ganglion, and travel through it without synapsing to join the nerve of the pterygoid canal and the greater petrosal nerve to the facial ganglion, where their cell bodies are situated. The central processes of these neurones traverse the sensory root of the facial nerve (nervus intermedius) to pass to the gustatory nucleus in the nucleus of the tractus solitarius. Parasympathetic postganglionic secretomotor fibres from the pterygopalatine ganglion run with the nerves to supply the palatine mucous glands.
© 2008 Elsevier
TONGUE The tongue is a highly muscular organ of deglutition, taste and speech. It is partly oral and partly pharyngeal in position, and is attached by its muscles to the hyoid bone, mandible, styloid processes, soft palate and the pharyngeal wall. It has a root, an apex, a curved dorsum and an inferior surface. Its mucosa is normally pink and moist, and is attached closely to the underlying muscles. The dorsal mucosa is covered by numerous papillae, some of which bear taste buds. Intrinsic muscle fibres are arranged in a complex interlacing pattern of longitudinal, transverse, vertical and horizontal fasciculi and this allows great mobility. Fasciculi are separated by a variable amount of adipose tissue which increases posteriorly. The root of the tongue is attached to the hyoid bone and mandible, and between them it is in contact inferiorly with geniohyoid and mylohyoid. The dorsum (posterosuperior surface) is generally convex in all directions at rest. It is divided by a V-shaped sulcus terminalis into an anterior, oral (presulcal) part which faces upwards, and a posterior, pharyngeal (postsulcal) part which faces posteriorly. The anterior part forms about two-thirds of the length of the tongue. The two limbs of the sulcus terminalis run anterolaterally to the palatoglossal arches from a median depression, the foramen caecum, which marks the site of the upper end of the embryonic thyroid diverticulum. The oral and pharyngeal parts of the tongue differ in their mucosa, innervation and developmental origins.
ORAL (PRESULCAL) PART (Figs 33.4, 33.5) The presulcal part of the tongue is located in the floor of the oral cavity. It has an apex touching the incisor teeth, a margin in contact with the gums and teeth, and a superior surface (dorsum) related to the hard and soft palates. On each side, in front of the palatoglossal arch, there are four or five vertical folds, the foliate papillae, which represent vestiges of larger papillae found in many other mammals. The dorsal mucosa has a longitudinal median sulcus and is covered by filiform, fungiform and circumvallate papillae. The mucosa on the inferior (ventral) surface is smooth, purplish and reflected onto the oral floor and gums: it is connected to the oral floor anteriorly by the lingual frenulum. The deep lingual vein, which is visible, lies lateral to the frenulum on either side. The plica fimbriata, a fringed mucosal ridge directed anteromedially towards the apex of the tongue, lies lateral to the vein. This part of the tongue develops from the lingual swellings of the mandibular arch and from the tuberculum impar.
PHARYNGEAL (POSTSULCAL) PART (Fig. 33.4)
Figure 33.4 Dorsum of the tongue, with adjoining palatoglossal and palatopharyngeal arches, and epiglottis. The palatine tonsils lie in the tonsillar recesses on either side.
Figure 33.5 Dissection of the inferior surface of the tongue, also showing the sublingual glands and submandibular duct openings. On the right side the mucous membrane has been removed and the inferior longitudinal muscle has been divided and partially resected.
The postsulcal part of the tongue constitutes its base and lies posterior to the
palatoglossal arches. Although it forms the anterior wall of the oropharynx, it is described here for convenience. Its mucosa is reflected laterally onto the palatine tonsils and pharyngeal wall, and posteriorly onto the epiglottis by a median and two lateral glossoepiglottic folds which surround two depressions or valleculae. The pharyngeal part of the tongue is devoid of papillae, and exhibits low elevations. There are underlying lymphoid nodules which are embedded in the submucosa and collectively termed the lingual tonsil. The ducts of small seromucous glands open on the apices of these elevations. The postsulcal part of the tongue develops from the hypobranchial eminence. On the rare occasions that the thyroid gland fails to migrate away from the tongue during development it remains in the postsulcal part of the tongue as a functioning lingual thyroid gland. page 584 page 585
MUSCLES OF THE TONGUE The tongue is divided by a median fibrous septum, attached to the body of the hyoid bone. There are extrinsic and intrinsic muscles in each half, the former extending outside the tongue and moving it bodily, the latter wholly within it and altering its shape. The extrinsic musculature consists of four pairs of muscles namely genioglossus, hyoglossus, styloglossus (and chondroglossus) and palatoglossus. The intrinsic muscles are the bilateral superior and inferior longitudinal, the transverse and the vertical. Genioglossus (Fig. 33.6)
Genioglossus is triangular in sagittal section, lying near and parallel to the midline. It arises from a short tendon attached to the superior genial tubercle behind the mandibular symphysis, above the origin of geniohyoid. From this point it fans out backwards and upwards. The inferior fibres of genioglossus are attached by a thin aponeurosis to the upper anterior surface of the hyoid body near the midline (a few fasciculi passing between hyoglossus and chondroglossus to blend with the middle constrictor of the pharynx). Intermediate fibres pass backwards into the posterior part of the tongue, and superior fibres ascend forwards to enter the whole length of the ventral surface of the tongue from root to apex, intermingling with the intrinsic muscles. The muscles of opposite sides are separated posteriorly by the lingual septum. Anteriorly they are variably blended by decussation of fasciculi across the midline. The attachment of the genioglossi to the genial tubercles prevents the tongue from sinking back and obstructing respiration, therefore anaesthetists pull the mandible forward to obtain the full benefit of this connection.
Figure 33.6 Dissection showing the muscles of the tongue and pharynx. Note that palatoglossus is not shown here, but is depicted in Fig. 35.3.
Vascular supply Genioglossus is supplied by the sublingual branch of the lingual artery and the submental branch of the facial artery. Innervation Genioglossus is innervated by the hypoglossal nerve. Actions Genioglossus brings about the forward traction of the tongue to protrude its apex from the mouth. Acting bilaterally, the two muscles depress the central part of the tongue, making it concave from side to side. Acting unilaterally, the tongue diverges to the opposite side. Hyoglossus (Fig. 33.6)
Hyoglossus is thin and quadrilateral, and arises from the whole length of the
greater cornu and the front of the body of the hyoid bone. It passes vertically up to enter the side of the tongue between styloglossus laterally and the inferior longitudinal muscle medially. Fibres arising from the body of the hyoid overlap those from the greater cornu. Relations
page 585 page 586
Hyoglossus is related at its superficial surface to the digastric tendon, stylohyoid, styloglossus and mylohyoid, the lingual nerve and submandibular ganglion, the sublingual gland, the deep part of the submandibular gland and duct, the hypoglossal nerve and the deep lingual vein. By its deep surface it is related to the stylohyoid ligament, genioglossus, the middle constrictor and the inferior longitudinal muscle of the tongue, and the glossopharyngeal nerve. Posteroinferiorly it is separated from the middle constrictor by the lingual artery. This part of the muscle is in the lateral wall of the pharynx, below the palatine tonsil. Passing deep to the posterior border of hyoglossus are, in descending order: the glossopharyngeal nerve, stylohyoid ligament and lingual artery. Vascular supply Hyoglossus is supplied by the sublingual branch of the lingual artery and the submental branch of the facial artery. Innervation Hyoglossus is innervated by the hypoglossal nerve. Action Hyoglossus depresses the tongue. Chondroglossus
Sometimes described as a part of hyoglossus, this muscle is separated from it by some fibres of genioglossus, which pass to the side of the pharynx. It is c.2 cm long, arising from the medial side and base of the lesser cornu and the adjoining part of the body of the hyoid. It ascends to merge into the intrinsic musculature between the hyoglossus and genioglossus muscles. A small slip occasionally springs from the cartilago triticea and enters the tongue with the posterior fibres of the hyoglossus muscle. Vascular supply, innervation and action These are similar to those described for hyoglossus. Styloglossus (Fig. 33.6)
Styloglossus is the shortest and smallest of the three styloid muscles. It arises from the anterolateral aspect of the styloid process near its apex, and from the styloid end of the stylomandibular ligament. Passing downwards and forwards, it divides at the side of the tongue into a longitudinal part, which enters the tongue dorsolaterally to blend with the inferior longitudinal muscle in front of hyoglossus, and an oblique part, overlapping hyoglossus and decussating with it. Vascular supply Styloglossus is supplied by the sublingual branch of the lingual artery. Innervation Styloglossus is innervated by the hypoglossal nerve. Action Styloglossus draws the tongue up and backwards. Stylohyoid ligament (Fig. 33.6)
The stylohyoid ligament is a fibrous cord which extends from the tip of the styloid process to the lesser cornu of the hyoid bone. It gives attachment to some fibres of styloglossus and the middle constrictor of the pharynx and is closely related to the lateral wall of the oropharynx. Below it is overlapped by hyoglossus. The ligament is derived embryologically from the second branchial arch. It may be partially calcified. Palatoglossus
Palatoglossus is closely associated with the soft palate in function and innervation, and is described with the other palatal muscles (p. 628). Intrinsic muscles (Fig. 33.7) Superior longitudinal
The superior longitudinal muscle constitutes a thin stratum of oblique and longitudinal fibres lying beneath the mucosa of the dorsum of the tongue. It extends forwards from the submucous fibrous tissue near the epiglottis and from the median lingual septum to the lingual margins. Some fibres are inserted into the mucous membrane. Inferior longitudinal
The inferior longitudinal muscle is a narrow band of muscle close to the inferior lingual surface between genioglossus and hyoglossus. It extends from the root of the tongue to the apex. Some of its posterior fibres are connected to the body of the hyoid bone. Anteriorly it blends with styloglossus. Transverse
The transverse muscles pass laterally from the median fibrous septum to the submucous fibrous tissue at the lingual margin, blending with palatopharyngeus. Vertical
The vertical muscles extend from the dorsal to the ventral aspects of the tongue in the anterior borders. Vascular supply
The intrinsic muscles are supplied by the lingual artery. Innervation
All intrinsic lingual muscles are innervated by the hypoglossal nerve. Actions
page 586 page 587
Figure 33.7 Cross-section through the tongue, the mouth and the body of the mandible opposite the first molar tooth.
The intrinsic muscles alter the shape of the tongue. Thus, contraction of the superior and inferior longitudinal muscles tend to shorten the tongue, but the former also turns the apex and sides upwards to make the dorsum concave, while the latter pulls the apex down to make the dorsum convex. The transverse muscle narrows and elongates the tongue while the vertical muscle makes it flatter and wider. Acting alone or in pairs and in endless combination, the intrinsic muscles give the tongue precise and highly varied mobility, important not only in alimentary function but also in speech.
VASCULAR SUPPLY AND LYMPHATIC DRAINAGE OF THE TONGUE Lingual artery (Fig. 33.8)
The tongue and the floor of the mouth are supplied chiefly by the lingual artery, which arises from the anterior surface of the external carotid artery (Figs 31.7, 31.16, 31.14). It passes between hyoglossus and the middle constrictor of the pharynx to reach the floor of the mouth accompanied by the lingual veins and the glossopharyngeal nerve. At the anterior border of hyoglossus, the lingual artery bends sharply upwards. It is covered by the mucosa of the tongue and lies between genioglossus medially and the inferior longitudinal muscle laterally. Near the tip of the tongue it anastomoses with its contralateral fellow. The branches of the lingual artery form a rich anastomotic network, which supplies the musculature of the tongue, and a very dense submucosal plexus. Named branches of the lingual artery in the floor of the mouth are the dorsal lingual, sublingual and deep lingual arteries. Dorsal lingual arteries
The dorsal lingual arteries are usually two or three small vessels. They arise medial to hyoglossus and ascend to the posterior part of the dorsum of the tongue. The vessels supply its mucous membrane, and the palatoglossal arch, tonsil, soft palate and epiglottis. They anastomose with their contralateral fellows. Sublingual artery
The sublingual artery arises at the anterior margin of hyoglossus. It passes forward between genioglossus and mylohyoid to the sublingual gland, and supplies the gland, mylohyoid and the buccal and gingival mucous membranes. One branch pierces mylohyoid and joins the submental branches of the facial artery. Another branch courses through the mandibular gingivae to anastomose with its contralateral fellow. A single artery arises from this anastomosis and enters a small foramen (lingual foramen) on the mandible, situated in the midline on the posterior aspect of the symphysis immediately above the genial tubercles. Deep lingual artery
The deep lingual artery is the terminal part of the lingual artery and is found on the inferior surface of the tongue near the lingual frenum. In addition to the lingual artery, the tonsillar and ascending palatine branches of the facial and ascending pharyngeal arteries also supply tissue in the root of the tongue. In the region of the valleculae, epiglottic branches of the superior laryngeal artery anastomose with the inferior dorsal branches of the lingual artery.
Figure 33.8 Dissection of the right half of the tongue from the medial side, exposing the end of the second part and the beginning of the third part of the left lingual artery and adjoining structures, in an edentulous subject.
Lingual veins
The veins draining the tongue follow two routes. Dorsal lingual veins drain the dorsum and sides of the tongue, join the lingual veins accompanying the lingual artery between hyoglossus and genioglossus, and empty into the internal jugular vein near the greater cornu of the hyoid bone. The deep lingual vein begins near the tip of the tongue and runs back just beneath the mucous membrane on the inferior surface of the tongue. It joins a sublingual vein from the sublingual salivary gland near the anterior border of hyoglossus and forms the vena comitans nervi hypoglossi, which run back with the hypoglossal nerve between mylohyoid and hyoglossus to join the facial, internal jugular or lingual vein. Lymphatic drainage (Fig. 33.9)
The mucosa of the pharyngeal part of the dorsal surface of the tongue contains many lymphoid follicles aggregated into dome-shaped groups, the lingual tonsils. Each group is arranged around a central deep crypt, or invagination, which opens onto the surface epithelium. The ducts of mucous glands open into the bases of the crypts. Small isolated follicles also occur beneath the lingual mucosa. The lymphatic drainage of the tongue can be divided into three main regions, namely
marginal, central and dorsal. The anterior region of the tongue drains into marginal and central vessels, and the posterior part of the tongue behind the circumvallate papillae drains into the dorsal lymph vessels. The more central regions may drain bilaterally. Marginal vessels
Marginal vessels from the apex of the tongue and the lingual frenulum area descend under the mucosa to widely distributed nodes. Some vessels pierce mylohyoid as it contacts the mandibular periosteum to enter either the submental or anterior or middle submandibular nodes, or else to pass anterior to the hyoid bone to the jugulo-omohyoid node. Vessels arising in the plexus on one side may cross under the frenulum to end in contralateral nodes. Efferent vessels of median submental nodes pass bilaterally. Some vessels pass inferior to the sublingual gland and accompany the companion vein of the hypoglossal nerve to end in jugulodigastric nodes. One vessel often descends further to reach the jugulo-omohyoid node, and passes either superficial or deep to the intermediate tendon of digastric.
page 587 page 588
Figure 33.9 Lymphatic drainage of the tongue. Removal of the sternocleidomastoid has exposed the whole chain of deep cervical lymph nodes. (From The lymphatics of the tongue with particular reference to the removal of lymphatic glands in cancer of the tongue, Jamieson JK, Dobson JF, 8: 80-87, 1920, © British Journal of Surgery Ltd. Reproduced with permission. Permission is granted by John Wiley and Sons Ltd on behalf of the BJSS Ltd.)
Vessels from the lateral margin of the tongue cross the sublingual gland, pierce mylohyoid and end in the submandibular nodes. Others end in the jugulodigastric or jugulo-omohyoid nodes. Vessels from the posterior part of the lingual margin traverse the pharyngeal wall to the jugulodigastric lymph nodes. Central vessels
The regions of the lingual surface draining into the marginal or central vessels are
not distinct. Central lymphatic vessels ascend between the fibres of the two genioglossi; most pass between the muscles and diverge to the right or left to follow the lingual veins to the deep cervical nodes, especially the jugulodigastric and jugulo-omohyoid nodes. Some pierce mylohyoid to enter the submandibular nodes. Dorsal vessels
Vessels draining the postsulcal region and the circumvallate papillae run posteroinferiorly. Those near the median plane may pass bilaterally. They turn laterally, join the marginal vessels and all pierce the pharyngeal wall, passing around the external carotid arteries to reach the jugulodigastric and juguloomohyoid lymph nodes. One vessel may descend posterior to the hyoid bone, perforating the thyrohyoid membrane to end in the jugulo-omohyoid node.
INNERVATION OF THE TONGUE The muscles of the tongue, with the exception of palatoglossus, are supplied by the hypoglossal nerve. Palatoglossus is supplied via the pharyngeal plexus (p. 630). The pathways for proprioception associated with the tongue musculature are unknown, but presumably may involve the lingual, glossopharyngeal or hypoglossal nerves, and the cervical spinal nerves which communicate with the hypoglossal nerve. The sensory innervation of the tongue reflects its embryological development. The nerve of general sensation to the presulcal part is the lingual nerve, which also carries taste sensation derived from the chorda tympani branch of the facial nerve. The nerve supplying both general and taste sensation to the postsulcal part is the glossopharyngeal nerve. An additional area in the region of the valleculae is supplied by the internal laryngeal branch of the vagus nerve. Lingual nerve (Fig. 30.7)
The lingual nerve is sensory to the mucosa of the floor of the mouth, mandibular lingual gingivae and mucosa of the presulcal part of the tongue (excluding the circumvallate papillae). It also carries postganglionic parasympathetic fibres from the submandibular ganglion to the sublingual and anterior lingual glands. The lingual nerve arises from the posterior trunk of the mandibular nerve in the infratemporal fossa (Figs 30.6, 30.7, 30.4, 30.8) where it is joined by the chorda tympani branch of the facial nerve and often by a branch of the inferior alveolar nerve. It then passes below the mandibular attachment of the superior pharyngeal constrictor and pterygomandibular raphe, closely applied to the periosteum of the medial surface of the mandible, until it lies opposite the distal (posterior) root of the third molar tooth, where it is covered only by the gingival mucoperiosteum. At this point it usually lies 2-3 mm below the alveolar crest and c.0.6 mm from the bone, but it sometimes lies above the alveolar crest. It next passes medial to the mandibular attachment of mylohyoid, which carries it progressively away from the mandible, and separates it from the alveolar bone covering the mesial root of the third molar tooth, and then passes downward and forward on the deep surface of mylohyoid to cross the lingual sulcus beneath the mucosa. In this position it lies on the deep portion of the submandibular gland. It passes below the submandibular duct which crosses it from medial to lateral, and curves upward, forward and medially to enter the tongue. Within the tongue the lingual nerve lies first on styloglossus and then the lateral surface of hyoglossus and genioglossus, before dividing into terminal branches that supply the overlying lingual mucosa. The lingual nerve is connected to the submandibular ganglion (Fig. 30.8) by two or three branches, and also forms connecting loops with twigs of the hypoglossal nerve at the anterior margin of hyoglossus.
The lingual nerve is at risk during surgical removal of (impacted) lower third molars, and after such operations up to 10% of patients may have symptoms of nerve damage, although these are usually temporary. The nerve is also at risk during operations to remove the submandibular salivary gland, because the duct must be dissected from the lingual nerve during these operations. Glossopharyngeal nerve
The glossopharyngeal nerve is distributed to the postsulcal part of the tongue and the circumvallate papillae. It communicates with the lingual nerve. The course of the glossopharyngeal nerve in the neck is described on page 555. Hypoglossal nerve (Fig. 31.21)
The course of the hypoglossal nerve in the neck is described on page 558. After crossing the loop of the lingual artery a little above the tip of the greater cornu of the hyoid, it inclines upwards and forwards on hyoglossus, passing deep to stylohyoid, the tendon of digastric and the posterior border of mylohyoid. Between mylohyoid and hyoglossus the hypoglossal nerve lies below the deep part of the submandibular gland, the submandibular duct and the lingual nerve, with which it communicates. It then passes onto the lateral aspect of genioglossus, continuing forwards in its substance as far as the tip of the tongue. It distributes fibres to styloglossus, hyoglossus and genioglossus and to the intrinsic muscles of the tongue. The special sensory innervation of the tongue
The sense of taste is dependent on scattered groups of sensory cells, the taste buds, which occur in the oral cavity and pharynx and are particularly plentiful on the lingual papillae of the dorsal lingual mucosa. Dorsal lingual mucosa
The dorsal mucosa is somewhat thicker than the ventral and lateral mucosae, is directly adherent to underlying muscular tissue with no discernible submucosa, and covered by numerous papillae. The dorsal epithelium consists of a superficial stratified squamous epithelium, which varies from non-keratinized, stratified squamous epithelium posteriorly, to fully keratinized epithelium overlying the filiform papillae more anteriorly. These features probably reflect the fact that the apex of the tongue is subject to greater dehydration than the posterior and ventral parts and is subject to more abrasion during mastication. The underlying lamina propria is a dense fibrous connective tissue, with numerous elastic fibres, and is continuous with similar tissue extending between the lingual muscle fasciculi. It contains numerous vessels and nerves from which the papillae are supplied, and also large lymph plexuses and lingual glands. Lingual papillae Lingual papillae are projections of the mucosa covering the dorsal surface of the tongue (Fig. 33.4). They are limited to the presulcal part of the tongue, produce its characteristic roughness and increase the area of contact between the tongue and the contents of the mouth. There are four principal types, named filiform, fungiform, foliate and circumvallate papillae, and all except the filiform papillae bear taste buds. Papillae are more visible in the living when the tongue is dry. Filiform papillae (Fig. 33.10) Filiform papillae are minute, conical or cylindrical projections which cover most of the presulcal dorsal area, and are arranged in diagonal rows that extend anterolaterally, parallel with the sulcus terminalis, except at the lingual apex where they are transverse. They have irregular cores of connective tissue and their epithelium, which is keratinized, may split into whitish fine secondary
processes. They appear to function to increase the friction between the tongue and food, and facilitate the movement of particles by the tongue within the oral cavity. Fungiform papillae (Fig. 33.10) Fungiform papillae occur mainly on the lingual margin but also irregularly on the dorsal surface, where they may occasionally be numerous. They differ from filiform papillae because they are larger, rounded and deep red in colour, this last reflecting their thin, non-keratinized epithelium and highly vascular connective tissue core. Each usually bears one or more taste buds on its apical surface. Foliate papillae Foliate papillae lie bilaterally in two zones at the sides of the tongue near the sulcus terminalis, each formed by a series of red, leaf-like mucosal ridges, covered by a non-keratinized epithelium. They bear numerous taste buds. Circumvallate papillae (Figs 33.11, 33.12)
Figure 33.10 Dorsal surface of the anterior tongue showing non-keratinized fungiform (left) and two keratinized filiform papillae (centre and right) with non-keratinized regions between. (By permission from Young B, Heath JW 2000 Wheater's Functional Histology. Edinburgh: Churchill Livingstone.)
page 588 page 589
Figure 33.11 Section through a circumvallate papilla. Serous glands (of von Ebner) empty via ducts into the base of the trench and numerous taste buds are contained within the stratified epithelium of the papillary wall (pale structures on the inner wall of the cleft, left side). (By permission from Young B, Heath JW 2000 Wheater's Functional Histology. Edinburgh: Churchill Livingstone.)
Circumvallate papillae are large cylindrical structures, varying in number from 8 to 12, which form a V-shaped row immediately in front of the sulcus terminalis on the dorsal surface of the tongue. Each papilla, 1-2 mm in diameter, is surrounded by a slight circular mucosal elevation (vallum or wall) which is separated from the papilla by a circular sulcus. The papilla is narrower at its base than its apex and the entire structure is generally covered with non-keratinized stratified squamous epithelium. Numerous taste buds are scattered in both walls of the sulcus, and small serous glands (of von Ebner) open into the sulcal base. Taste buds
Taste buds are microscopic barrel-shaped epithelial structures which contain chemosensory cells in synaptic contact with the terminals of gustatory nerves. They are numerous on all types of lingual papillae (except filiform papillae) particularly on their lateral aspects. Taste buds are not restricted to the papillae, and are scattered over almost the entire dorsal and lateral surfaces of the tongue and, rarely, on the epiglottis and lingual aspect of the soft palate. Each taste bud is linked by synapses at its base to one of three cranial nerves which carry taste, i.e. the facial, glossopharyngeal or vagus. They share some physiological features with neurones, for example action potential generation and synaptic transmission, and are therefore often referred to as paraneurones.
Figure 33.12 Circumvallate papilla. A, Scanning electron micrograph showing a circumvallate papilla surrounded by a trench. B, Section of circumvallate papilla showing pale barrel-shaped taste buds (B) in its walls. P, apical pore. (A, by kind permission from S Franey and by permission from Berkovitz BKB, Holland GR, Moxham BJ 2002 Oral Anatomy, Embryology and Histology, 3rd edn. Edinburgh: Mosby; B, by permission from Dr JB Kerr, Monash University, from Kerr JB 1999 Atlas of Functional Histology. London: Mosby.)
There is considerable individual variation in the distribution of taste buds in humans. They are most abundant on the posterior parts of the tongue, especially around the walls of the circumvallate papillae and their surrounding sulci, where there is an average of c.250 taste buds for each of the 8-12 papillae. Over 1000 taste buds are distributed over the sides of the tongue, particularly over the more posterior folds of the two foliate papillae, whereas they are rare, and sometimes even absent, on fungiform papillae (c.3 per papilla). Taste buds have been
described on the fetal epiglottis and soft palate but most disappear from these sites during postnatal development. Microstructure of taste buds
page 589 page 590
Each taste bud is a barrel-shaped cluster of 50-150 fusiform cells which lies within an oval cavity in the epithelium and converges apically on a gustatory pore, a 2 µm wide opening on the mucosal surface. The whole structure is about 70 µm in height by 40 µm across and is separated by a basal lamina from the underlying lamina propria. A small fasciculus of afferent nerve fibres penetrates the basal lamina and spirals around the sensory cells. Chemical substances dissolved in the oral saliva diffuse through the gustatory pores of the taste buds to reach the taste receptor cell membranes, where they cause membrane depolarization. Innervation of taste buds Individual nerve fibres branch to give a complex distribution of taste bud innervation. Each fibre may have many terminals, which may spread to innervate widely separated taste buds or may innervate more than one sensory cell in each bud. Conversely, individual buds may receive the terminals of several different nerve fibres. These convergent and divergent patterns of innervation may be of considerable functional importance. The gustatory nerve for the anterior part of the tongue, excluding the circumvallate papillae, is the chorda tympani, which travels via the lingual nerve. In most individuals, taste fibres run in the chorda tympani to cell bodies in the facial ganglion, but occasionally they diverge to the otic ganglion, which they reach via the greater petrosal nerve. Taste buds in the inferior surface of the soft palate are supplied mainly by the facial nerve, through the greater petrosal nerve, pterygopalatine ganglion and lesser palatine nerve: they may also be supplied by the glossopharyngeal nerve. Taste buds in the circumvallate papillae, postsulcal part of the tongue and in the palatoglossal arches and the oropharynx are innervated by the glossopharyngeal nerve, and those in the extreme pharyngeal part of the tongue and epiglottis receive fibres from the internal laryngeal branch of the vagus. Each taste bud receives two distinct classes of fibre: one branches in the periphery of the bud to form a perigemmal plexus, the other forms an intragemmal plexus within the bud itself which innervates the bases of the receptor cells. The perigemmal fibres contain various neuropeptides including calcitonin gene-related peptide (CGRP) and substance P, and appear to represent free sensory endings. Intragemmal fibres branch within the taste bud and each forms a series of synapses. Taste discrimination Gustatory receptors detect four main categories of taste sensation, classified as salty, sweet, sour and bitter; other taste qualities have been suggested, including metallic, and umami (Japanese: taste typified by monosodium glutamate). Although it is commonly stated that particular areas of the tongue are specialized to detect these different tastes, evidence indicates that all areas of the tongue are responsive to all taste stimuli. Each afferent nerve fibre is connected to widely separated taste buds and may respond to several different chemical stimuli. Some respond to all four classic categories, others to fewer or only one. Within a particular class of tastes, receptors are also differentially sensitive to a wide range of similar chemicals. Moreover, taste buds alone are able to detect only a rather restricted range of chemical substances in aqueous solution. It is difficult to separate the perceptions of taste and smell, because the oral and nasal cavities are continuous. Indeed, much of what is perceived as taste is the result of
airborne odorants from the oral cavity which pass through the nasopharynx to the olfactory area above it. Perceived sensations of taste are the results of the processing (presumably central) of a complex pattern of responses from particular areas of the tongue. Autonomic innervation of the tongue
The parasympathetic innervation of the various glands of the tongue is from the chorda tympani branch of the facial nerve which synapses in the submandibular ganglion: postganglionic branches are distributed to the lingual mucosa via the lingual nerve. The postganglionic sympathetic supply to lingual glands and vessels arises from the carotid plexus and enters the tongue through plexuses around the lingual arteries. Isolated nerve cells, perhaps postganglionic parasympathetic neurones, have been reported in the postsulcal region: presumably they innervate glandular tissue and vascular smooth muscle.
© 2008 Elsevier
TEETH INTRODUCTION AND TERMINOLOGY Humans have two generations of teeth: the deciduous (primary) dentition and the permanent (secondary) dentition. Teeth first erupt into the mouth at about 6 months after birth and all the deciduous teeth have erupted by 3 years of age. The first permanent teeth appear by 6 years, and thence the deciduous teeth are exfoliated one by one to be replaced by their permanent successors. A complete permanent dentition is present when the third molars erupt at or around the age of 18-21 years. In the complete deciduous dentition there are 20 teeth, 5 in each jaw quadrant. In the complete permanent dentition there are 32 teeth, 8 in each jaw quadrant. There are three basic tooth forms in both dentitions: incisiform, caniniform and molariform. Incisiform teeth (incisors) are cutting teeth, and have thin, blade-like crowns. Caniniform teeth (canines) are piercing or tearing teeth, and have a single, stout, pointed, cone-shaped crown. Molariform teeth (molars and premolars) are grinding teeth and possess a number of cusps on an otherwise flattened biting surface. Premolars are bicuspid teeth that are restricted to the permanent dentition and replace the deciduous molars. The tooth-bearing region of the jaws can be divided into four quadrants, the right and left maxillary and mandibular quadrants. A tooth may thus be identified according to the quadrant in which it is located (e.g. a right maxillary tooth or a left mandibular tooth). In both the deciduous and permanent dentitions, the incisors may be distinguished according to their relationship to the midline. Thus, the incisor nearest the midline is the central (first) incisor and the incisor that is more laterally positioned is termed the lateral (second) incisor. The permanent premolars and the permanent and deciduous molars can also be distinguished according to their mesiodistal relationships. The molar most mesially positioned is designated the first molar, and the one behind it is the second molar. In the permanent dentition, the tooth most distally positioned is the third molar. The mesial premolar is the first premolar, and the premolar behind it is the second premolar. The terminology used to indicate tooth surfaces is shown in Fig. 33.13. The aspect of teeth adjacent to the lips or cheeks is termed labial or buccal, that adjacent to the tongue being lingual (or palatal in the maxilla). Labial and lingual surfaces of an incisor meet medially at a mesial surface and laterally at a distal surface, terms which are also used to describe the equivalent surfaces of premolar and molar (postcanine) teeth. On account of the curvature of the dental arch, mesial surfaces of postcanine teeth are directed anteriorly and distal surfaces are directed posteriorly. Thus, the point of contact between the central incisors is the datum point for mesial and distal. The biting or occlusal surfaces of postcanine teeth are tuberculated by cusps which are separated by fissures forming a pattern characteristic of each tooth. The biting surface of an incisor is the incisal edge.
TOOTH MORPHOLOGY (Figs 33.13, 33.14) There are two incisors, a central and a lateral, in each half jaw or quadrant. In labial view, the crowns are trapezoid, the maxillary incisors (particularly the central) are larger than the mandibular. The biting or incisal edges initially have three tubercles or mamelons, which are rapidly removed by wear. In mesial or distal view their labial profiles are convex while their lingual surfaces are concavo-convex (the convexity near the cervical margin is caused by a low ridge or cingulum, which is prominent only on upper incisors). The roots of incisors are single and rounded in maxillary teeth, but flattened mesiodistally in mandibular teeth. The upper lateral incisor may be congenitally absent or may have a reduced form (peg-shaped lateral incisor). Behind each lateral incisor is a canine tooth with a single cusp (hence the American term cuspid) instead of an incisal edge. The maxillary canine is stouter and more pointed than the mandibular canine. The canine root, which is the longest of any tooth, produces a bulge (canine eminence) on the alveolar bone externally, particularly in the upper jaw. Although canines usually have single roots, that of the lower may sometimes be bifid.
Distal to the canines are two premolars, each with a buccal and lingual cusp (hence the term bicuspid). The occlusal surfaces of the maxillary premolars are oval (the long axis is buccopalatal) and a mesiodistal fissure separates the two cusps. In buccal view, premolars resemble the canines but are smaller. The maxillary first premolar usually has two roots (one buccal, one palatal) but may have one, and very rarely three, roots (two buccal and one palatal). The maxillary second premolar usually has one root. The occlusal surfaces of the mandibular premolars are more circular or more square than those of the upper premolars. The buccal cusp of the mandibular first premolar towers above the lingual cusp to which it is connected by a ridge separating the mesial and distal occlusal pits. In the mandibular second premolar a mesiodistal fissure usually separates a buccal from two smaller lingual cusps. Each lower premolar has one root, but very rarely the root of the first is bifid. Lower second premolars fail to develop in about 2% of individuals. page 590 page 591
Figure 33.13 A, The permanent teeth of the upper dental arch: occlusal aspect. B, The permanent teeth of the lower dental arch: occlusal aspect. C, Terminology employed for the identification of teeth according to their location in the lower jaw. The same terminology is employed for the teeth in the upper jaw
Figure 33.14 The permanent upper and lower teeth of the right side: labial and buccal surfaces.
Posterior to the premolars are three molars whose size decreases distally. Each has a large rhomboid (upper jaw) or rectangular (lower jaw) occlusal surface with four or five cusps. The maxillary first molar has a cusp at each corner of its occlusal surface and the mesiopalatal cusp is connected to the distobuccal by an oblique ridge. A smaller cusplet or tubercle (cusplet of Carabelli) usually appears on the mesiopalatal cusp (most commonly in Caucasian races). The tooth has three widely separated roots, two buccal and one palatal. The smaller maxillary second molar has a reduced or occasionally absent distopalatal cusp. Its three roots show varying degrees of fusion. The maxillary third molar, the smallest, is very variable in form. It usually has three cusps (the distopalatal being absent) and commonly the three roots are fused. The mandibular first molar has three buccal and two lingual cusps on its rectangular occlusal surface, the smallest cusp being distal. The cusps of this tooth are all separated by fissures. It has two widely separated roots, one mesial and one distal. The smaller mandibular second molar is like the first, but has only four cusps (it lacks the distal cusp of the first molar) and its two roots are closer together. The mandibular third molar is smaller still and, like the upper third molar, is variable in form. Its crown may resemble that of the lower first or second molar and its roots are frequently fused. As it erupts anterosuperiorly, the third molar is often impacted against the second molar, which produces food packing and inflammation, both indications for surgical removal. The maxillary third molar erupts posteroinferiorly and is rarely impacted. One or more third molars (upper or lower) fail to develop in up to 30% of individuals. Impacted mandibular third molars
In many subjects there is a disproportion between the size of the teeth and the size of the jaws such that there is insufficient space for all the teeth to erupt. As the third mandibular molar teeth (the wisdom teeth) are the last to erupt they are often impeded in their eruption and either become impacted against the distal aspect of the second molar or remain unerupted deeply within the jaw bone. If the tooth is completely covered by bone and mucosa it is very unlikely to cause any symptoms, and the subject remains unaware of their presence unless the teeth are seen on a routine dental radiograph. Very rarely the surrounding dental follicle may undergo cystic degeneration which can 'hollow out' the jaw, usually the mandible, to a considerable degree. The developing cyst may displace the tooth as it expands and the tooth may end up as far away as the condylar neck or coronoid process. More commonly, the erupting wisdom tooth erupts partially before impacting against the distal aspect of the second molar. When this occurs, symptoms are common due to recurrent soft tissue infection around the partially erupted tooth. This condition is known as pericoronitis and if the infecting organism is virulent, the infection may rapidly spread into the adjacent tissue spaces as described elsewhere. It is for this reason that so many wisdom teeth are removed in
adolescents and young adults. The surgery itself requires considerable skill as the lingual nerve passes across the surface of the periosteum lingually, separated from the tooth only by a cortical plate of bone no thicker than an egg shell. Damage to this nerve results in altered sensation to the ipsilateral side of the tongue. The root apices of the impacted tooth often lie immediately above the inferior alveolar canal, and removal of the tooth can result in damage to the underlying nerve and artery. Maxillary third molars are only rarely impacted. page 591 page 592
Deciduous teeth (Figs 33.15, 33.16)
Figure 33.15 The deciduous upper and lower teeth of the right side: labial and buccal surfaces.
Figure 33.16 A, The upper deciduous dentition (note the channels (arrows) leading to the developing permanent teeth). B, The lower deciduous dentition. (By permission from Berkovitz BKB, Moxham BJ 1994 Color Atlas of the Skull. London: Mosby.)
The incisors, canine and premolars of the permanent dentition replace two deciduous incisors, a deciduous canine and two deciduous molars in each jaw quadrant. The deciduous incisors and canine are shaped like their successors but are smaller and whiter and become extremely worn in older children. The deciduous second molars resemble permanent molars rather than their successors, the premolars. Each second deciduous molar has a crown which is almost identical to that of the posteriorly adjacent first permanent molar. The upper first deciduous molar has a triangular occlusal surface (its rounded 'apex' is palatal) and a fissure separates a double buccal cusp from the palatal cusp. The lower first deciduous molar is long and narrow, and its two buccal cusps are separated from its two lingual cusps by a zigzagging mesiodistal fissure. Like permanent molars, upper deciduous molars have three roots and lower deciduous molars have two roots. These roots diverge more than those of permanent teeth because each developing premolar tooth crown is accommodated directly under the crown of its deciduous predecessor. The roots of deciduous teeth are progressively resorbed by osteoclast-like cells (odontoclasts) prior to being shed. Eruption of teeth (Fig. 33.17)
Information on the sequence of development and eruption of teeth into the oral cavity is important in clinical practice and also in forensic medicine and archaeology. The tabulated data provided in Table 33.2 are largely based on European-derived populations and there is evidence of ethnic variation. When a permanent tooth erupts, about two-thirds of the root is formed and it takes about another three years for the root to be completed. For deciduous teeth, root completion is more rapid (Table 33.2). The developmental stages of initial calcification and crown completion are less affected by environmental influences than eruption, the timing of which may be modified by several factors such as early tooth loss and severe malnutrition. Figure 33.18 shows the panoramic appearance of the dentition seen with orthopantomograms at the time of birth, 3, 6!, 10 and 16 years of age. Dental alignment and occlusion
It is possible to bring the jaws together so that the teeth meet or occlude in many positions. When opposing occlusal surfaces meet with maximal 'intercuspation' (i.e. maximum contact), the teeth are said to be in centric occlusion (Fig. 33.19). In this position the lower teeth are normally opposed symmetrically and lingually with respect to the upper. Some important features of centric occlusion in a normal (idealized) dentition may be noted. Each lower postcanine tooth is slightly in front of its upper equivalent and the lower canine occludes in front of the upper. Buccal cusps of the lower postcanine teeth lie between the buccal and palatal cusps of the upper teeth. Thus, the lower postcanine teeth are slightly lingual and mesial to their upper equivalents. Lower incisors bite against the palatal surfaces of upper incisors, the latter normally obscuring about one-third of the crowns of the lower. This vertical overlap of incisors in centric occlusion is the overbite. The extent to which upper incisors are anterior to lowers is termed the overjet. In the most habitual jaw position, the resting posture, the teeth are slightly apart, the gap between them being the free-way space or interocclusal clearance. During mastication, especially with lateral jaw movements, the food is comminuted, which facilitates the early stages of digestion. The ideal occlusion is a rather subjective concept. If there is an ideal occlusion, it can only presently be defined in broad functional terms. Therefore, the occlusion can be considered 'ideal' when the teeth are aligned such that the masticatory loads are within physiological range and act through the long axes of as many teeth in the arch as possible; mastication involves alternating bilateral jaw movements (and not habitual, unilateral biting preferences as a result of adaptation to occlusal interference); lateral jaw movements occur without undue mechanical interference; in the rest position of the jaw, the gap between teeth (the freeway space) is correct for the individual concerned; the tooth alignment is aesthetically pleasing to its possessor. Variations from the ideal occlusion may be termed malocclusions (although these
could be regarded as normal for they are more commonly found in the population: c.75% of the population in the USA have some degree of occlusal 'disharmony'). However, the majority of malocclusions should be regarded as anatomical variations rather than abnormalities for they are rarely involved in masticatory dysfunction or pain, although they may be aesthetically displeasing. Variations in tooth number, size and form page 592 page 593
Figure 33.17 Development of the deciduous (blue) and permanent (yellow) teeth. (Modified with permission from Schour I, Massler M 1941 The development of the human dentition. J Am Dent Assoc 28: 1153-1160.)
Table 33-2. Chronology of the human dentition First evidence of calcification Crown Root (weeks in utero for completed Eruptiom completed Dentition Tooth deciduous teeth) (months) (months) (years) i1 14 1! 10 (8-12) 1! Deciduous i2 16 2! 11 (9-13) 2 Upper C 17 9 19 (163" 22) m1 15! 6 16 (13-19) 2! m2 19 11 29 (25-33) 3 i1 14 2! 8 (6-10) 1! i2 16 3 13 (101! Deciduous 16) lower C 17 9 20 (173" 23) m1 15! 5! 16 (14-18) 2" m2 18 10 27 (23-31) 3
l1 Permanent l2 upper C P1 P2 22"yrs M1 at birth M2 2!-3 yrs M3 7-9 yrs l1 Permanent l2 lower C P1 P2 2"2!yrs M1 at birth M2 2!-3 yrs M3 8-10 yrs
3-4 months 10-12 months 4-5 months 1!-1#yrs 6-7 yrs
4-5 yrs 7-8 yrs 4-5 yrs 8-9 yrs 6-7 yrs 11-12 yrs 5-6 yrs 10-11 yrs 10-12 yrs 12-14
2!-3 yrs
6-7 yrs
9-10
7-8 yrs
12-13 yrs
14-16
12-16 yrs
17-21 yrs
18-25
3-4 months 3-4 months 4-5 months 1#-2 yrs 6-7 yrs
4-5 yrs 6-7 yrs 4-5 yrs 7-8 yrs 6-7 yrs 9-10 yrs 5-6 yrs 10-12 yrs 1-12 yrs 13-14
2!-3 yrs
6-7 yrs
9-10
7-8 yrs
11-13 yrs
14-15
12-16 yrs
17-21 yrs
18-25
10 11 13-15 12-13
9 10 12-14 12-13
page 593 page 594
(Modified with permission from Ash MM 1993 Dental Anatomy, Physiology and Occlusion. Philadelphia: WB Saunders.)
Figure 33.18 Orthopantomogram of the dentition at birth. B, Orthopantomogram of the dentition at 2!years. C, Orthopantomogram of the dentition at 6!years. D, Orthopantomogram of the dentition at 10 years. E, Orthopantomogram of the dentition at 16 years. (B-F, by permission from Berkovitz BKB, Holland GR, Moxham BJ 2002 Oral Anatomy, Embryology and Histology, 3rd edn. Edinburgh: Mosby; D-F, also by kind permission from Eric Whaites.)
The incidence of variation in number and form, which is often related to race, is rare in deciduous teeth but not uncommon in the permanent dentition. One or more teeth may fail to develop, a condition known as hypodontia. Conversely, additional or supernumerary teeth may form, producing hyperdontia. The third permanent molar is the most frequently missing tooth: in one study one or more third molars failed to form in 32% of Chinese, 24% of English Caucasians and 2.5% of West Africans. In declining order of incidence, other missing teeth are maxillary lateral incisors, maxillary or mandibular second premolars, mandibular central incisors and maxillary first premolars. Hyperdontia affects the maxillary arch much more commonly than the mandibular dentition. The extra teeth are usually situated on the palatal aspect of the permanent incisors or distal to the molars. More rarely, additional premolars develop. Although supernumerary teeth in the incisor region are often small with simple conical crowns, they may impede the eruption of the permanent incisors. A supernumerary tooth situated between the central incisors is known as a
mesiodens. Teeth may be unusually large (macrodontia) or small (microdontia). For example, the crowns of maxillary central incisors may be abnormally wide mesiodistally; in contrast, a common variant of the maxillary lateral incisor has a small, peg-shaped crown. Epidemiological studies reveal that hyperdontia tends to be associated with macrodontia and hypodontia with microdontia, the most severely affected individuals representing the extremes of a continuum of variation. Together with family studies, this indicates that the causation is multifactorial, combining polygenic and environmental influences. Some variations in the form of teeth, being characteristic of race, are of anthropological and forensic interest. Mongoloid dentitions tend to have shovelshaped maxillary incisors with enlarged palatal marginal ridges. The additional cusp of Carabelli is commonly found on the mesiopalatal aspect of maxillary first permanent or second deciduous molars in Caucasian but rarely in Mongoloid dentitions. In African races the mandibular second permanent molar often has five rather than four cusps.
GENERAL ARRANGEMENT OF DENTAL TISSUES (Figs 33.20, 33.21, 33.22) A tooth (Fig. 33.20) consists of a crown, covered by very hard translucent enamel and a root covered by yellowish bone-like cementum. These meet at the neck or cervical margin. A longitudinal ground section (Fig. 32.21) reveals that the body of a tooth is mostly dentine (ivory) with an enamel covering up to about 2 mm thick, while the cementum is much thinner. The dentine surrounds a central pulp cavity, expanded at its coronal end into a pulp chamber and narrowed in the root as a pulp canal, opening at or near its tip by an apical foramen, occasionally multiple. The pulp is a connective tissue, continuous with the peridontal ligament via the apical foramen. It contains vessels for the support of the dentine and sensory nerves. page 594 page 595
Figure 33.19 Lateral view of the dentition in centric occlusion.
Figure 33.20 An extracted upper right canine tooth viewed from its mesial aspect, showing its principal parts. Note the root covered by cement (partially removed), and the curved cervical margin, convex towards the cusp of the tooth.
The root is surrounded by alveolar bone, its cementum separated from the osseous socket (alveolus) by the connective tissue of the periodontal ligament, c.0.2 mm thick (Fig. 33.23). Coarse bundles of collagen fibres, embedded at one end in cementum, cross the periodontal ligament to enter the osseous alveolar wall. Near the cervical margin, the tooth, periodontal ligament and adjacent bone are covered by the gingiva. On its internal surface the gingiva is attached to the tooth surface by the junctional epithelium, a zone of profound clinical importance because just above it is a slight recess, the gingival sulcus. As the sulcus is not necessarily self-cleansing, dental plaque may accumulate in it and this predisposes to periodontal disease. Enamel
Enamel is an extremely hard and rigid material which covers the crowns of teeth. It is a heavily mineralized cell secretion, containing 95-96% by weight crystalline apatites (88% by volume) and less than 1% organic matrix. The organic matrix comprises mainly unique enamel proteins, amelogenins and non-amelogenins such as enamelins, tuftelins. Although comprising a very small percentage of the weight and volume of enamel, the organic matrix permeates the whole of enamel. As its formative cells are lost from the surface during tooth eruption, enamel is incapable of further growth. Repair is limited to the remineralization of minute carious lesions.
page 595 page 596
Figure 33.21 A ground section of a young (permanent) lower first premolar tooth sectioned in the buccolingual longitudinal plane, photographed with transmitted light. The enamel striae are incremental lines of enamel growth (compare with Fig. 33.31). Within the dentine the lines of the dentinal tubules are visible, forming S-shaped curves in the apical region but straighter in the root.
Figure 33.22 Longitudinal section of a tooth and its environs.
Enamel reaches a maximum thickness of 2.5 mm over cusps and thins at the cervical margins. It is composed of closely packed enamel prisms or rods. In longitudinal section, enamel prisms extend from close to the enamel-dentine junction to within c.20 µm of the surface, where they are generally replaced by prismless (non-prismatic, aprismatic) enamel (Fig. 33.24). In cross-section the prisms are mainly horse-shoe shaped and are arranged in rows that are staggered such that the tails of the prisms in one row lie between the heads of the prism in the row above (prism pattern 3) (Fig. 33.25) and the tails are directed rootwards. The appearance of prism boundaries results from sudden changes in crystallite orientation. Prisms have a diameter of c.5 µm, and are packed with flattened hexagonal hydroxyapatite crystals, far larger than those found in the
other collagenous-based mineralized tissues. page 596 page 597
Figure 33.23 Demineralized section of a tooth with its root attached to the surrounding bone by the periodontal ligament. A, Alveolar bone; C, root of tooth lined by cementum; arrow, peridontal space. (By kind permission from Dr D Lunt.)
Figure 33.24 SEM of acid-etched outer enamel (A) showing enamel prisms, each (c.5 µm wide. A layer of prismless enamel (B) is evident on the surface. (By kind)permission from Professor D Whittaker.)
Two types of incremental lines are visible in enamel, short-term and long-term. At intervals of c.4 µm along its length, each prism is crossed by a line, probably reflecting diurnal swelling and shrinking in diameter during its growth. This shortterm daily growth line is known as a cross striation (Fig. 33.26). The longer term incremental lines pass from the enamel-dentine junction obliquely to the surface, where they end in shallow furrows, perikymata, visible on newly erupted teeth. Each line, known as an enamel stria, represents a period of 7-8 days enamel growth (Fig. 33.27). A prominent striation, the neonatal line, is formed in teeth whose mineralization spans birth (see above). Neonatal lines are present in the enamel and dentine of teeth mineralizing at the time of birth (all the deciduous teeth and the first permanent molars (see Fig. 33.13) and are therefore of
forensic importance, indicating that an infant has survived for a few days after birth. They reflect a disturbance in mineralization during the first few days after birth. Dentine
Figure 33.25 Ground cross-section of enamel showing cross-sectional keyhole (fish scale) appearance of enamel prisms (pattern 3). (By permission from Berkovitz BKB, Holland GR, Moxham BJ 2002 Oral Anatomy, Embryology and Histology, 3rd edn. Edinburgh: Mosby.)
Figure 33.26 Ground longitudinal section of enamel viewed with phase contrast showing prisms (vertical lines) and cross-striations (horizontal lines). (By permission from Berkovitz BKB, Holland GR, Moxham BJ 2002 Oral Anatomy, Embryology and Histology, 3rd edn. Edinburgh: Mosby.)
page 597 page 598
Figure 33.27 Ground longitudinal section of enamel showing enamel striae (arrows).
Viewed between crossed polarizing filters. (By kind permission from Dr AD Beynon.)
Dentine is a yellowish avascular tissue which forms the bulk of a tooth. It is a tough and compliant composite material, with a mineral content of c.70% dry weight (largely crystalline hydroxyapatite with some calcium carbonate ) and 20% organic matrix (type I collagen, glycosaminoglycans and phosphoproteins). Its conspicuous feature is the regular pattern of microscopic dentinal tubules, c.3 µm in diameter, which extend from the pulpal surface to the enamel-dentine junction. The tubules show lateral and terminal branching near the enamel-dentine junction (Fig. 33.28) and may project a short distance into the enamel (enamel spindles). Each tubule encloses a single cytoplasmic process of an odontoblast whose cell body lies in a pseudostratified layer which lines the pulpal surface. Processes are believed to extend the full thickness of dentine in newly erupted teeth, but in older teeth they may be partly withdrawn and occupy only the pulpal third, while the outer regions contain probably only extracellular fluid. The diameter of the dentine tubule is narrowed by deposition of peritubular dentine. This is different from normal dentine (intertubular dentine) because it is more highly mineralized and lacks a collagenous matrix. Peritubular dentine can therefore be identified by microradiography (Fig. 33.29). In time, it may completely fill the tubule, a process which gives rise to translucent dentine and which commences in the apical region of the root. The outermost zone (10-20 µm) of dentine differs in the crown and the root. In the crown it is referred to as mantle dentine and differs in the orientation of its collagen fibres. In the root, the peripheral zone presents a granular layer - with less overall mineral - beyond which is a hyaline layer which lacks a tubular structure and may function to produce a good bond between the cementum and dentine. Dentine is formed slowly throughout life, and so there is always an unmineralized zone of predentine at the surface of the mineralized dentine, adjacent to the odontoblast layer at the periphery of the pulp. Biochemical changes within the mineralizing matrix mean that predentine stains differently to the matrix of the mineralized dentine. The predentine-dentine border is generally scalloped, because dentine mineralizes both linearly and as microscopic spherical aggregates of crystals (calcospherites). Dentine, like enamel, is deposited incrementally, and exhibits both short- and long-term incremental lines. Long-term lines are known as Andresen lines and are c.20 µm apart: they represent increments of about 6-10 days (Fig. 33.30). Daily incremental lines (von Ebner lines) are c.4 µm apart. Where mineralization spans birth (i.e. all deciduous teeth and usually the first permanent molars) a neonatal line is formed in dentine similar to that which is seen in enamel, and it signals the abrupt change in both environment and nutrition which occurs at birth. Primary dentine formation proceeds at a steady but declining rate as first the crown and then the root is completed. This slow and intermittent deposition of dentine (regular secondary dentine) continues throughout life and further reduces the size of the pulp chamber. The presence of the odontoblast process means that dentine is a vital tissue. It responds to adverse external stimuli - such as rapidly advancing caries, excessive wear or tooth breakage - by forming poorly mineralized dead tracts, in which the odontoblasts of the affected region die and the tubules remain empty (tertiary dentine). A dead tract may be sealed from the pulp by a thin zone of sclerosed dentine and the deposition of irregular (tertiary) dentine by newly differentiated pulp cells (Fig. 33.31).
Figure 33.28 Ground longitudinal section of dentine showing branching of dentine tubules near the enamel-dentine junction (arrow). (By permission from Berkovitz BKB, Holland GR, Moxham BJ 2002 Oral Anatomy, Embryology and Histology, 3rd edn. Edinburgh: Mosby.)
Figure 33.29 Microradiograph of transversely sectioned dentinal tubules surrounded by a more radiopaque and therefore more mineralized zone of peritubular dentine. (By permission from Berkovitz BKB, Holland GR, Moxham BJ 2002 Oral Anatomy, Embryology and Histology, 3rd edn. Edinburgh: Mosby.)
Figure 33.30 Ground longitudinal section of dentine viewed in polarized light showing alternate light and dark bands representing long-period incremental lines (Andresen lines). The bands are approximately orientated at right angles to the direction of the dentinal tubules (arrows). (By permission from Berkovitz BKB, Holland GR, Moxham BJ 2002 Oral Anatomy, Embryology and Histology, 3rd edn. Edinburgh: Mosby.)
page 598 page 599
Figure 33.31 Longitudinal ground section of an incisor tooth.
Dental pulp
Dental pulp provides the nutritive support for the synthetic activity of the odontoblast layer. It is a well-vascularized, loose connective tissue, enclosed by dentine and continuous with the periodontal ligament via apical and accessory foramina. Several thin-walled arterioles enter by the apical foramen and run longitudinally within the pulp to an extensive subodontoblastic plexus. Blood flow rate per unit volume of tissue is greater in the pulp than in other oral tissues, and tissue fluid pressures within the pulp appear to be unusually high. As well as typical connective tissue cells, pulp uniquely contains the cell bodies of odontoblasts whose long processes occupy the dentinal tubules. Pulp also has dendritic antigen-presenting cells. Approximately 60% of pulpal collagen is type I, and the bulk of the remainder is type III. As dentine deposition increases with age, the pulp recedes until the whole of the crown may be removed without accessing the pulp. Dental pulp is extensively innervated by unmyelinated postganglionic vasoconstrictor sympathetic nerve fibres from the superior cervical ganglion, which enter with the arterioles, and by myelinated (A$) and unmyelinated (C) sensory nerve fibres from the trigeminal ganglion, which traverse the pulp longitudinally and ramify as a plexus (Raschkow's plexus) beneath the odontoblast layer (Fig. 33.22). Here, any myelinated nerve fibres lose their myelin sheaths and continue into the odontoblast layer, and some enter the dentinal tubules, especially the region beneath the cusps. Stimulation of dentine, whether by thermal, mechanical or osmotic means, evokes a pain response. Pulp nerves release numerous neuropeptides. Cementum
Cementum is a bone-like tissue which covers the dental roots, and is c.50% by weight mineralized (mainly hydroxyapatite crystals). However, unlike bone, cementum is avascular and lacks nerves. The cementum generally overlaps the enamel slightly, although it may meet it end on. Occasionally, the two tissues may fail to meet, in which case dentine is exposed in the mouth. If the exposed dentinal tubules remain patent then the teeth may be sensitive to stimuli such as cold water. In older teeth, the root may become exposed in the mouth as a consequence of occlusal drift and gingival recession, and cementum is often abraded away by incorrect tooth brushing and dentine exposed. Like bone, cementum is perforated by Sharpey's fibres, which represent the attachment bundles of collagen fibres in the periodontal ligament (extrinsic fibres). New layers of cementum, which are deposited incrementally throughout life to compensate for tooth movements, incorporate new Sharpey's fibres. Incremental lines are irregularly spaced. The first cement to be formed is thin (up to 200 µm), acellular and contains only extrinsic fibres. Cementum formed later towards the root apex is produced more rapidly and contains cementocytes lying in lacunae joined by canaliculi. The latter are mainly directed towards their source of
nutrients from the periodontal ligament. This cementum contains both extrinsic fibres derived from the periodontal ligament and intrinsic fibres of cementoblastic origin which lie parallel to the surface. Varying arrangements of layering between cellular and acellular cement occur. With increasing age, cellular cement may reach a thickness of a millimetre or more around the apices and at the branching of the roots, where it compensates for the loss of enamel by attrition. Cementum is not remodelled but small areas of resorption with evidence of repair may be seen. Forensic anatomy of teeth
In forensic medicine, dental evidence is valuable in identification of individuals, especially following mass disasters; estimation of age at death of skeletonized remains; establishing guilt in cases of criminal injury by biting. If teeth have been restored, extracted or replaced by a denture, an individual will have a virtually unique dentition which may have been recorded by the dentist in the form of charts, radiographs or plaster casts. Teeth are the most indestructible bodily structures and can provide an identification when trauma or fire has rendered the face unrecognizable. Moreover, the chronology of crown development, eruption and root formation can be used to estimate age until the third molar is completed at about 21 years. The method is even applicable to the fetus because the weight of mineralized tissue in teeth is closely related to age from about 22 weeks' gestation until birth. The time taken for a crown to form can be calculated from ground sections with considerable accuracy by counting the number of daily cross striations from the neonatal line. For permanent teeth, the time taken for the crown to form can be calculated by counting the number of the enamel striae. The age of adult teeth can be estimated from factors such as wear of the crown, reduction in size of the pulp and increase in thickness of cement in the apical half of the root. However, probably the most useful single characteristic is the amount of translucent dentine in the root, which is proportional to age. Such estimations are within 5-7 years of the chronological age and likely to be closer to the true age than those derived from skeletal changes. Periodontal ligament
The principal functions of the periodontal ligament are to support the teeth, generate the force of tooth eruption and provide sensory information about tooth position and forces to facilitate reflex jaw activity. The periodontal ligament is a dense fibrous connective tissue c.0.2 mm wide which contains cells associated with the development and maintenance of alveolar bone (osteoblasts and osteoclasts) and of cementum (cementoblasts and odontoclasts). It also contains a network of epithelial cells (epithelial cell rests) which are embryological remnants of an epithelial root sheath. They have no evident function but may give rise to dental cysts. The majority of collagen fibres of the periodontal ligament are arranged as variously oriented dense fibre bundles that connect alveolar bone and cementum and which may help to resist movement in specific directions (Fig. 33.32). About 80% of the collagen in the periodontal ligament is type I, most of the remainder is type III. The rate of turnover of collagen is probably the highest of any site in the body, for reasons which are as yet unclear. A very small volume of fibres are oxytalan fibres. The periodontal ligament has a rich nerve and blood supply. The nerves are both autonomic (for the vasculature) and sensory (for pain and proprioception). The majority of proprioceptive nerve endings appear to be Ruffini-like endings (p. 62). The blood vessels tend to lie towards the bone side of the periodontal ligament and the capillaries are fenestrated. Tissue fluid pressures appear to be high. Alveolar bone
page 599 page 600
Figure 33.32 Decalcified longitudinal section of a tooth showing groups of periodontal ligament fibres (the alveolar crest fibres and the horizontal fibres) in the region of the alveolar crest. van Gieson stain. (By permission from Berkovitz BKB, Holland GR, Moxham BJ 2002 Oral Anatomy, Embryology and Histology, 3rd edn. Edinburgh: Mosby.)
Figure 33.33 Anterior part of the left side of the mandible, with the superficial bone removed on the buccal side to show the roots of a number of teeth, some of which have also been sectioned vertically. Note the cortical plate of compact bone lining the sockets of the teeth (the lamina dura of radiographs: see Fig. 33.35), and the flat table of bone surmounting the interdental bone septa. In this specimen the inferior alveolar canal is widely separated from the roots of the teeth, a variable condition.
That part of the maxilla or mandible which supports and protects the teeth is known as alveolar bone. An arbitrary boundary at the level of the root apices of the teeth separates the alveolar processes from the body of the mandible or the maxilla (Fig. 33.33). Like bone in other sites, alveolar bone functions as a mineralized supporting tissue, gives attachment to muscles, provides a framework for bone marrow and acts as a reservoir for ions, especially calcium. It is dependent on the presence of teeth for its development and maintenance, and requires functional stimuli to maintain bone mass. Where teeth are congenitally absent, as for example in anodontia, it is poorly developed, and it atrophies after tooth extraction. The alveolar tooth-bearing portion of the jaws consists of outer and inner alveolar plates. The individual sockets are separated by plates of bone termed the interdental septa, while the roots of multi-rooted teeth are divided by interradicular septa septas. The compact layer of bone which lines the tooth socket has been called either the cribriform plate, on account of its content of vascular
(Volkmann's) canals which pass from the alveolar bone into the periodontal ligament, or bundle bone, because numerous bundles of Sharpey's fibres pass into it from the periodontal ligament (Fig. 33.34). In clinical radiographs, the bone lining the alveolus commonly appears as a continuous dense white line about 0.5-1 mm thick, the lamina dura (Fig. 33.35). However, this appearance gives a misleading impression of the density of alveolar bone: the X-ray beam passes tangentially through the socket wall, and so the radio-opacity of the lamina dura is an indication of the quantity of bone the beam has passed through, rather than the degree of mineralization of the bone. Superimposition also obscures the Volkmann's canals. Chronic infections of the dental pulp spread into the periodontal ligament, which leads to resorption of the lamina dura around the root apex. The presence of a continuous lamina dura around the apex of a tooth therefore usually indicates a healthy apical region (except in acute infections where resorption of bone has not yet begun). On the labial and buccal aspects of upper teeth, the two cortical plates usually fuse, and there is very little trabecular bone between them, except where the buccal bone thickens over the molar teeth near the root of the zygomatic arch. It is easier and more convenient to extract upper teeth by fracturing the buccal than the palatal plate. Anteriorly in the lower jaw, labial and lingual plates are thin, but in the molar region the buccal plate is thickened as the external oblique line. Near the lower third molar, the lingual bone is much thinner than the buccal and it is mechanically easier to remove this tooth, when impacted, via the lingual plate, although it is important to realize that the lingual nerve is here exposed to damage.
VASCULAR SUPPLY AND LYMPHATIC DRAINAGE OF THE TEETH AND SUPPORTING STRUCTURES Arterial supply of the teeth
Figure 33.34 Decalcified section of a root of a tooth showing Sharpey's fibres from the periodontal ligament entering alveolar bone (A). The Sharpey's fibres in bone are seen to be thicker, but less numerous, than those entering the cementum (B) on the tooth surface. van Gieson stain. (By permission from Berkovitz BKB, Holland GR, Moxham BJ 2002 Oral Anatomy, Embryology and Histology, 3rd edn. Edinburgh: Mosby.)
Figure 33.35 Bite-wing radiograph of teeth and surrounding bone. Note the different radiopacities of enamel and dentine. In a healthy tooth, such as the first molar illustrated here, the lamina dura is complete and appears as a radiopaque line. In the case of the adjacent second molar tooth in which the bulk of the crown has been lost due to dental caries, an abscess has formed at the base of the tooth and as a result the lamina dura has lost its continuity. (By kind permission from Ms Nadine White.)
The main arteries to the teeth and their supporting structures are derived from the maxillary artery, a terminal branch of the external carotid artery. The upper teeth are supplied by branches from the superior alveolar arteries and the lower teeth by branches from the inferior alveolar arteries. Superior alveolar arteries
The upper jaw is supplied by posterior, middle and anterior superior alveolar (dental) arteries. The posterior superior alveolar artery usually arises from the third part of the maxillary artery in the pterygopalatine fossa. It descends on the infratemporal surface of the maxilla, and divides to give branches that enter the alveolar canals to supply molar and premolar teeth, adjacent bone and the maxillary sinus, and other branches that continue over the alveolar process to supply the gingivae. The middle and anterior superior alveolar arteries are branches from the infraorbital artery. page 600 page 601
The infraorbital artery often arises with the posterior superior alveolar artery. It enters the orbit posteriorly through the inferior orbital fissure and runs in the infraorbital groove and canal with the infraorbital nerve. When the small middle superior alveolar artery is present it runs down the lateral wall of the maxillary sinus and forms anastomotic arcades with the anterior and posterior vessels, terminating near the canine tooth. The anterior superior alveolar artery curves through the canalis sinuosus to supply the upper incisor and canine teeth and the mucous membrane in the maxillary sinus. The canalis sinuosus swerves laterally from the infraorbital canal and inferomedially below it in the wall of the maxillary sinus, following the rim of the anterior nasal aperture, between the alveoli of canine and incisor teeth and the nasal cavity. It ends near the nasal septum where its terminal branch emerges. The canal may be up to 55 mm long. Inferior alveolar artery
The inferior alveolar (dental) artery, a branch of the maxillary artery, descends in the infratemporal fossa posterior to the inferior alveolar nerve. Here, it lies between bone laterally and the sphenomandibular ligament medially. Before entering the mandibular foramen it gives off a mylohyoid branch, which pierces the sphenomandibular ligament to descend with the mylohyoid nerve in its groove on the inner surface of the ramus of the mandible (Fig. 30.7). The mylohyoid artery ramifies superficially on the muscle and anastomoses with the submental branch of the facial artery. The inferior alveolar artery then traverses the mandibular canal with the inferior alveolar nerve to supply the mandibular molars and premolars and divides into the incisive and mental branches near the first premolar. The incisive branch continues below the incisor teeth (which it supplies) to the midline, where it anastomoses with its fellow, although few anastomotic vessels
cross the midline. In the canal the arteries supply the mandible, tooth sockets and teeth via branches which enter the minute hole at the apex of each root to supply the pulp. The mental artery leaves the mental foramen and supplies the chin and anastomoses with the submental and inferior labial arteries. Near its origin the inferior alveolar artery has a lingual branch, which descends with the lingual nerve to supply the lingual mucous membrane. The pattern of branching of the inferior alveolar artery reflects that of the nerves of the same name. Arterial supply of periodontal ligaments
The periodontal ligaments supporting the teeth are supplied by dental branches of alveolar arteries. One branch enters the alveolus apically and sends two or three small rami into the dental pulp through the apical foramen, and other rami into the periodontal ligament. Interdental arteries ascend in the interdental septa, sending branches at right angles into the periodontal ligament, and terminate by communicating with gingival vessels that also supply the cervical part of the ligament. The periodontal ligament therefore receives its blood from three sources: from the apical region; ascending interdental arteries; descending vessels from the gingivae. These vessels anastomose with each other, which means that when the pulp of a tooth is removed during endodontic treatment, the attachment tissues of the tooth remain vital. Venous drainage of the teeth
Veins accompanying the superior alveolar arteries drain the upper jaw and teeth anteriorly into the facial vein, or posteriorly into the pterygoid venous plexus. Veins from the lower jaw and teeth collect either into larger vessels in the interdental septa or into plexuses around the root apices and thence into several inferior alveolar veins. Some of these veins drain through the mental foramen to the facial vein, others drain via the mandibular foramen to the pterygoid venous plexus. Lymphatic drainage of the teeth
The lymph vessels from the teeth usually run directly into the ipsilateral submandibular lymph nodes. Lymph from the mandibular incisors, however, drains into the submental lymph nodes. Occasionally, lymph from the molars may pass directly into the jugulodigastric group of nodes.
INNERVATION OF THE TEETH (Fig. 33.36) The regional supply to the teeth and gingivae is shown in Table 33.1. The teeth in the upper jaw are supplied by the superior alveolar nerves while those in the lower jaw are supplied by the inferior alveolar nerve. Superior alveolar nerves
Figure 33.36 Longitudinal demineralized section of a tooth stained with a silver impregnation technique. Note the horizontal nerve trunk (top) within the pulp, with fine
nerve fibres, one of which (A) passes between the odontoblasts (B) lining the surface of the predentine (C).
The teeth in the upper jaw are supplied by the three superior alveolar (dental) nerves (Fig. 30.6). These arise from the maxillary nerve in the pterygopalatine fossa or in the infraorbital groove and canal. The posterior superior alveolar (dental) nerve leaves the maxillary nerve in the pterygopalatine fossa and runs anteroinferiorly to pierce the infratemporal surface of the maxilla, descending under the mucosa of the maxillary sinus. After supplying the lining of the sinus the nerve divides into small branches which link up as the molar part of the superior alveolar plexus, supplying twigs to the molar teeth. It also supplies a branch to the upper gingivae and the adjoining part of the cheek. The middle superior alveolar (dental) nerve arises from the infraorbital nerve as it runs in the infraorbital groove, and runs downwards and forwards in the lateral wall of the maxillary sinus. It ends in small branches which link up with the superior dental plexus, supplying small rami to the upper premolar teeth. This nerve is variable, and it may be duplicated or triplicated or absent. The anterior superior alveolar (dental) nerve leaves the lateral side of the infraorbital nerve near the midpoint of its canal and traverses the canalis sinuosus in the anterior wall of the maxillary sinus. It curves first under the infraorbital foramen, then passes medially towards the nose and finally turns downwards and divides into branches to supply the incisor and canine teeth. It assists in the formation of the superior dental plexus and it gives off a nasal branch, which passes through a minute canal in the lateral wall of the inferior meatus to supply the mucous membrane of the anterior area of the lateral wall as high as the opening of the maxillary sinus, and the floor of the nasal cavity. It communicates with the nasal branches of the pterygopalatine ganglion and finally emerges near the root of the anterior nasal spine to supply the adjoining part of the nasal septum. Inferior alveolar (dental) nerve (Figs 30.4-30.7, 30.6)
The course of the inferior alveolar nerve in the infratemoral fossa is described on page 523. Just before entering the mandibular canal the inferior alveolar nerve gives off a small mylohyoid branch which pierces the sphenomandibular ligament and enters a shallow groove on the medial surface of the mandible following a course roughly parallel to its parent nerve. It passes below the origin of mylohyoid to lie on the superficial surface of the muscle, between it and the anterior belly of digastric, both of which it supplies. It gives a few filaments to supply the skin over the point of the chin. In the mandibular canal, the inferior alveolar nerve runs downward and forward, generally below the apices of the teeth until below the first and second premolars where it divides into terminal incisive and mental branches. The incisive branch continues forward in a bony canal or in a plexiform arrangement, giving off branches to the first premolar, canine and incisor teeth, and the associated labial gingivae. The lower central incisor teeth receive a bilateral innervation, fibres probably crossing the midline within the periosteum to re-enter the bone via numerous canals in the labial cortical plate. page 601 page 602
The mental nerve passes upward, backward and outward to emerge from the mandible via the mental foramen between and just below the apices of the premolar teeth (Fig. 30.6). It immediately divides into three branches, two of which pass upward and forward to form an incisor plexus labial to the teeth, supplying the gingiva (and probably the periosteum). From this plexus and the dental branches, fibres turn downwards and then lingually to emerge on the lingual surface of the mandible on the posterior aspect of the symphysis or opposite the premolar teeth, probably communicating with the lingual or mylohyoid nerve. The third branch of the mental nerve passes through the intermingled fibres of depressor anguli oris and platysma to supply the skin of the lower lip and chin. Branches of the mental nerve also communicate with terminal filaments of the mandibular branch of the facial nerve. Variations in the fascicular organization of the inferior alveolar nerve are clinically important when extracting impacted third molars. It may appear as a single bundle lying a few millimetres below the roots of the teeth, or it may lie much
lower, and almost reach the lower border of the bone, so that it gives off a variable number of large rami which pass anterosuperiorly towards the roots before dividing to supply the teeth and interdental septa. Only rarely is it plexiform. The nerve may lie on the lingual or buccal side of the mandible (slightly more commonly on the buccal side). Even when the third molar tooth is in a normal position, the nerve may be so intimately related to it that it grooves its root. Exceptionally the nerve may be similarly related to the second molar. Nerves may pass from the substance of temporalis to enter the mandible through the retromolar fossa, where they communicate with branches of the inferior alveolar nerve. Foramina occur in c.10% of retromolar fossae and infiltration in this region can abolish sensation which occasionally remains after an inferior alveolar nerve block. Similarly, branches from the buccal, mylohyoid and lingual nerves may enter the mandible and provide additional routes of sensory transmission from the teeth. Thus, even when the inferior alveolar nerve has been anaesthetized correctly, pain may still be experienced by a patient when undergoing dental cavity preparation. Pain sensation in teeth
The teeth are supplied by nociceptors that generate pain sensation of a very high order. The mechanism underlying this sensitivity is of considerable clinical significance and is controversial. Currently, the most widely accepted view is that fluid movements through the dentine tubules stimulate nerve endings at the periphery of the dental pulp (hydrodynamic hypothesis). Local analgesia
It is technically possible to achieve profound regional anaesthesia by depositing local anaesthetic solution adjacent to the trigeminal nerve trunks or their branches within the infratemporal fossa (p. 523). These injections can either be performed transorally - posterior superior alveolar nerve block, maxillary nerve block, inferior alveolar nerve block, lingual nerve block and mandibular nerve block - or more rarely by an external route through the skin of the face - maxillary nerve block, inferior alveolar nerve block and mandibular nerve block. In the case of the mandible, the anterior teeth can be anaesthetized by simple diffusion techniques as the bone is relatively thin. However, this is not adequate for the cheek teeth due to the increased thickness of the bone. In this case, the inferior alveolar nerve has to be anaesthetized before it enters the inferior alveolar canal. The needle has to be placed within the pterygomandibular space to achieve a successful inferior alveolar nerve block. The lingual nerve is also usually blocked as it lies close to the inferior alveolar nerve. Because of the other structures within the infratemporal fossa it is vitally important that the operator has a detailed knowledge of the anatomy in this region to understand, and therefore try to avoid, the complications that may arise. Any damage to blood vessels in the infratemporal fossa, generally the pterygoid venous plexus, can lead to haematoma formation. In extreme cases, bleeding can track through the inferior orbital fissure resulting in a retrobulbar haematoma, which can result in loss of visual acuity or blindness. Intravascular injection of local anaesthetic solution (which usually contains adrenaline (epinephrine )) can have profound systemic effects and for this reason an aspirating syringe is always used to check that the needle has not entered a vessel prior to injection. If the needle is placed too medially it may enter medial pterygoid, while if directed too laterally it may penetrate temporalis. In either case, there will be lack of anaesthesia followed later by trismus. If the needle is placed too deeply, anaesthetic solution may cause a temporary Bell's palsy due to loss of conduction from the facial nerve in the region of the parotid gland. Finally, if the needle is not sterile, infection of the pterygomandibular space may ensue, which could spread to other important tissue spaces (p. 525). Diffusion of anaesthetic solution through the inferior orbital fissure could give temporary orbital symptoms such as paralysis of lateral rectus due to anaesthesia of the abducens nerve.
© 2008 Elsevier
THE SALIVARY GLANDS Salivary glands are compound, tubuloacinar exocrine glands (p. 34) whose ducts open into the oral cavity. They secrete saliva, a fluid which lubricates food to assist deglutition, moistens the buccal mucosa, which is important for speech, and provides an aqueous solvent necessary for taste and a fluid seal for sucking and suckling. They also secrete digestive enzymes, e.g. salivary amylase and antimicrobial agents e.g. IgA, lysozyme and lactoferrin, into saliva. Conditions where there is a significant decrease in the production of saliva (xerostomia) may result in periodontal inflammation and dental caries. An illustration of the position of the major salivary glands and their ducts is shown in Fig. 33.37. The major salivary glands are the paired parotid, submandibular and sublingual glands. In addition, there are numerous minor salivary glands scattered throughout the oral mucosa and submucosa. Approximately 0.5 litres of saliva is secreted per day. Salivary flow rates are c.0.3 ml/min when unstimulated, and rise to 1.5-2 ml/min when stimulated. Flow rate is negligible during sleep. In the unstimulated state, the parotid gland contributes c.20%, the submandibular gland c.65%, and the sublingual and minor salivary glands c.15% of the daily output of saliva. When stimulated, the parotid contribution rises to 50%.
Parotid gland (Fig. 33.37) The parotid gland is the largest salivary gland. It is almost entirely serous. The parotid duct runs through the cheek and drains into the mouth opposite the maxillary second permanent molar tooth. The parotid gland is situated in front of the external ear and is described in detail in relation to the face (p. 515).
Submandibular salivary gland (Fig. 33.37) The submandibular gland is irregular in shape and about the size of a walnut. It consists of a larger superficial and a smaller deep part, continuous with each other around the posterior border of mylohyoid. It is a seromucous (but predominantly serous) gland.
SUPERFICIAL PART OF THE SUBMANDIBULAR GLAND The superficial part of the gland is situated in the digastric triangle where it reaches forward to the anterior belly of digastric and back to the stylomandibular ligament, by which it is separated from the parotid gland. Above, it extends medial to the body of the mandible. Below, it usually overlaps the intermediate tendon of digastric and the insertion of stylohyoid. This part of the submandibular gland presents inferior, lateral and medial surfaces, and is partially enclosed between two layers of deep cervical fascia that extend from the greater cornu of the hyoid bone. The superficial layer is attached to the lower border of the mandible and covers the inferior surface of the gland. The deep layer is attached to the mylohyoid line on the medial surface of the mandible and covers the medial surface of the gland. The inferior surface, covered by skin, platysma and deep fascia, is crossed by the facial vein and the cervical branch of the facial nerve. Near the mandible the submandibular lymph nodes are in contact with the gland and some may be embedded within it. The lateral surface is related to the submandibular fossa on the medial surface of the body of the mandible and the mandibular attachment of medial pterygoid. The facial artery grooves its posterosuperior part, lies at first deep to the gland and then emerges between its lateral surface and the mandibular attachment of the medial pterygoid to reach the lower border of the mandible. page 602 page 603
Figure 33.37 The salivary glands of the left side. The cranial region of the superficial part of the submandibular gland has been excised and the cut mylohyoid has been turned down to expose a portion of the deep part of the gland.
The medial surface is related anteriorly to mylohyoid, from which it is separated by the mylohyoid nerve and vessels and branches of the submental vessels. More posteriorly, it is related to styloglossus, the stylohyoid ligament and the glossopharyngeal nerve, which separate it from the pharynx. In its intermediate part the medial surface is related to hyoglossus, from which it is separated by styloglossus, the lingual nerve, submandibular ganglion, hypoglossal nerve and deep lingual vein (sequentially from above down). Below, the medial surface is related to the stylohyoid muscle and the posterior belly of digastric.
DEEP PART OF THE SUBMANDIBULAR GLAND The deep part of the gland extends forwards to the posterior end of the sublingual gland. It lies between mylohyoid inferolaterally, hyoglossus and styloglossus medially, the lingual nerve superiorly, and the hypoglossal nerve and deep lingual vein inferiorly.
VASCULAR SUPPLY AND LYMPHATIC DRAINAGE The arteries supplying the gland are branches of the facial and lingual arteries. The lymph vessels drain into the deep cervical group of lymph nodes (particularly the jugulo-omohyoid node), interrupted by the submandibular nodes.
INNERVATION The secretomotor supply to the submandibular gland is derived from the submandibular ganglion. This is a small, fusiform body which lies on the upper part of hyoglossus. There are additional ganglion cells in the hilum of the gland. Like the ciliary, pterygopalatine and otic ganglia, the submandibular is a peripheral parasympathetic ganglion. It is superior to the deep part of the submandibular gland and inferior to the lingual nerve, and is suspended from the latter by anterior and posterior filaments (Fig. 30.6). Though related to the lingual nerve, the ganglion is connected functionally with the facial nerve and its chorda tympani branch. As with the other cranial parasympathetic ganglia, there are three roots associated with the submandibular ganglion (Fig. 30.8). The motor, parasympathetic root is the posterior filament which connects it to the lingual nerve. This conveys preganglionic fibres from the superior salivatory nucleus which travel in the facial, chorda tympani and lingual nerves to the ganglion, where they synapse. The postganglionic fibres are secretomotor to the submandibular and sublingual salivary glands. Some fibres may also reach the parotid gland. The sympathetic root is derived from the plexus on the facial artery. It consists of postganglionic fibres from the superior cervical ganglion which traverse the submandibular ganglion without synapsing. They are vasomotor to
the blood vessels of the submandibular and sublingual glands. Five or six branches from the ganglion supply the submandibular gland and its duct. Other fibres pass through the anterior filament that connects the submandibular gland to the lingual nerve and are carried to the sublingual and anterior lingual glands. Sensory fibres are derived from the lingual nerve.
SUBMANDIBULAR DUCT The submandibular duct is c.5 cm long and has a thinner wall than the parotid duct. It begins from numerous tributaries in the superficial part of the gland and emerges from the medial surface of this part of the gland behind the posterior border of mylohyoid. It traverses the deep part of the gland, passes at first up and slightly back for c.5 mm, and then forwards between mylohyoid and hyoglossus. It next passes between the sublingual gland and genioglossus to open in the floor of the mouth on the summit of the sublingual papilla at the side of the frenulum of the tongue (Fig. 33.3). It lies between the lingual and hypoglossal nerves on hyoglossus, but, at the anterior border of the muscle, it is crossed laterally by the lingual nerve, terminal branches of which ascend on its medial side. As the duct traverses the deep part of the gland it receives small tributaries draining this part of the gland. Like the parotid gland, the duct system of the submandibular gland can be visualized by sialography (Fig. 33.38).
page 603 page 604
Figure 33.38 Sialogram showing a normal submandibular duct (large arrow). Unusually, a sublingual duct is also evident (small arrow). (By kind permission from Dr N Drage.)
Sublingual salivary gland (Fig. 33.37) The sublingual gland is the smallest of the main salivary glands: each gland is narrow, flat, shaped like an almond, and weighs c.4 g. The sublingual gland lies on mylohyoid, and is covered by the mucosa of the floor of the mouth, which is raised as a sublingual fold (Fig. 33.3). The anterior end of the contralateral sublingual gland lies in front, and the deep part of the submandibular gland lies behind. The mandible above the anterior part of the mylohyoid line, the sublingual fossa, is lateral, and genioglossus is medial, separated from the gland by the lingual nerve and submandibular duct. The sublingual glands are seromucous, but predominantly mucous.
VASCULAR SUPPLY, INNERVATION AND LYMPHATIC DRAINAGE The arterial supply is from the sublingual branch of the lingual artery and the submental branch of the facial artery. Innervation is via the submandibular ganglion. Lymphatic drainage is to the submental nodes.
SUBLINGUAL DUCTS The sublingual gland has 8-20 excretory ducts. Smaller sublingual ducts open, usually separately, from the posterior part of the gland onto the summit of the sublingual fold (a few sometimes open into the submandibular duct). Small rami from the anterior part of the gland sometimes form a major sublingual duct (Bartholin's duct), which opens with, or near to, the orifice of the submandibular
duct. This duct may be visualized occasionally in a submandibular sialogram (Fig. 33.38). Ranula
If the ducts draining any salivary gland become obstructed, the gland itself is at risk of developing a retention cyst where the retained secretions dilate the gland itself rather like a balloon. This phenomenon is seen mostly in the minor salivary glands which line the lips and oral cavity. Trauma such as persistent lip biting results in scarring of the overlying oral mucosa and obstruction of the small drainage duct. When trauma occurs in the floor of the mouth and obstructs the drainage duct/s of the sublingual gland, the resulting retention cyst is known as a ranula. (Ranula is the Latin name for a frog and is used here because the tense cystic swelling is said to resemble the throat of a croaking frog.) A ranula usually presents as a large tense bluish swelling anteriorly in the floor of the mouth just to one side of the midline, which often displaces the tongue. Occasionally the developing retention cyst herniates through a midline dehiscence where the two mylohyoid muscles fail to meet in the midline anteriorly, and then the ranula may present as a submental swelling or as a combined submental and floor of mouth swelling. The treatment of a ranula is excision of the sublingual gland responsible.
Minor salivary glands The minor salivary glands of the mouth include the labial, buccal, palatoglossal, palatal and lingual glands. The labial and buccal glands contain both mucous and serous elements. The palatoglossal glands are mucous glands and are located around the pharyngeal isthmus. The palatal glands are mucous glands and occur in both the soft and hard palates. The anterior and posterior lingual glands are mainly mucous. The anterior glands are embedded within muscle near the ventral surface of the tongue and open by means of four or five ducts near the lingual frenum and the posterior glands are located in the root of the tongue. The deep posterior lingual glands are predominantly serous. Serous glands (of Von Ebner) occur around the circumvallate papillae, their secretion is watery, and they probably assist in gustation by spreading taste stimuli over the taste buds and then washing them away.
Microstructure of the salivary glands Salivary glands have numerous lobes composed of smaller lobules separated by dense connective tissue which is continuous with the capsule of the gland, and contains excretory (collecting) ducts, blood vessels, lymph vessels, nerve fibres and small ganglia. Each lobule has a single duct, whose branches terminate at dilated secretory 'endpieces', which are tubular or acinar in shape (Fig. 33.39). Their primary secretion is modified as it flows through intercalated, striated and excretory ducts into one or more main ducts which discharge saliva into the oral cavity. They contain a variable amount of intralobular adipose tissue: adipocytes are particularly numerous in the parotid gland. The secretory 'endpieces' of the human parotid gland are almost exclusively serous acini (Fig. 33.40): mucous tubules or acini are rare. In the submandibular gland, secretory units are predominantly serous acini, with some mucous tubules and acini (Fig. 33.41). Mucous tubules are often associated with groups of serous cells at their blind ends, appearing as crescent-shaped serous demilunes in routine histological preparations. However this appears to be a fixation artefact, as tissue prepared by rapid freezing methods lacks serous demilunes and the serous secretory cells align with mucous cells around a common lumen. In the sublingual gland mucous tubules and acini predominate (Fig. 33.42), but serous cells also occur, as acini or as serous demilunes. Serous cells are approximately pyramidal in shape. Their nuclei vary in shape and position, but are more rounded and situated less basally than in mucous cells. Apically, the cytoplasm is filled by proteinaceous secretory (zymogen) granules with high amylase activity. Additionally, serous cells secrete kallikrein, lactoferrin and lysozyme, an antibacterial enzyme whose synthesis has been localized in particular to the serous demilunes of the submandibular and sublingual glands, and which is important in the defence against oral pathogens. In the human parotid and submandibular glands, zymogen granules also show a positive periodic acid-Schiff staining reaction, which indicates the presence of polysaccharides, and some texts refer to these cells as seromucous. Mucous cells are cylindrical and have flattened, basal nuclei. Their apical cytoplasm is typically packed with large, pale-staining and electron-translucent secretory
droplets.
DUCTS Intercalated, striated (both intralobular) and extralobular collecting ducts lead consecutively from the secretory endpieces. The lining cells of intercalated ducts are flat nearest the secretory endpiece, but become cuboidal. They function primarily as a conduit for saliva but, together with the striated ducts, may also modify its content of electrolytes and secrete immunoglobulin A. Striated ducts (Fig. 33.40) are lined by a low columnar epithelium and are so-called because their lining cells have characteristic basal striations. The latter are regions of highly infolded basal plasma membrane, between which lie columns of vertically aligned mitochondria. The nuclei are consequently displaced by the basal striations from a typical ductal basal position to a central or even apical location. Infolding of the basal plasma membrane, and local abundance of mitochondria, are typical features of epithelial cells which actively transport electrolytes. Here, the cells transport potassium and bicarbonate into saliva: they produce a hypotonic saliva by reabsorbing sodium and chloride ions in excess of water. Striated ducts modify electrolyte composition and secrete immunoglobulin A, lysozyme and kallikrein. Immunoglobulin A is produced by subepithelial plasma cells and transported transcytotically across the epithelial barrier to be secreted, once it has been dimerized by epithelial secretory component, into the saliva. This is also a function of serous acinar cells and other secretory epithelia, notably the lactating breast (Chapter 58). The intralobular ductal system of the sublingual gland is less well-developed than that of the parotid and submandibular glands. Collecting ducts are metabolically relatively inert conduits which run within interlobular connective tissue septa in the glands. They transport saliva to the main duct which opens onto the mucosal surface of the buccal cavity. The lining epithelium of collecting ducts varies. It may be pseudostratified columnar, stratified cuboidal or columnar in the larger ducts, and has a distinct basal layer. It becomes a stratified squamous epithelium near the buccal orifice.
MYOEPITHELIAL CELLS Myoepithelial cells (Fig. 33.39) are contractile cells associated with secretory endpieces and with much of the ductal system. They lie between the basal lamina and the epithelial cells proper. They extend numerous cytoplasmic processes around serous acini and are often termed basket cells. Myoepithelial cells associated with ducts are more fusiform in shape, and are aligned along the length of the duct. Their cytoplasm contains abundant actin microfilaments which mediate contraction under the control of both sympathetic and parasympathetic stimulation. The outflow of saliva is thus accelerated through reduction in the luminal volume of secretory endpieces and ducts, contributing to the secretory pressure. page 604 page 605
page 605 page 606
Figure 33.39 Architecture of a generalized salivary gland including ultrastructural details. Solid black arrows indicate the direction of transport of salivary components and the open white arrow the direction of salivary flow. The innervation of the ducts, secretory units and arterioles in a generalized salivary gland is shown.
Figure 33.40 A section through the parotid gland, with deeply stained secretory acini surrounding two striated ducts in transverse section, accompanied by small venules. An intercalated duct is seen in longitudinal section in the bottom right hand corner. (Photograph by Sarah-Jane Smith.)
Figure 33.41 The mixed secretory units of the submandibular gland. Deeply stained serous acini (above) surround a striated duct (top left). Below are pale-staining mucous tubules, some with serous demilunes. Caps of serous cells appear, artifactually, to form crescentic extensions to the tubule, as seen at the bottom of the field. (Photograph by Sarah-Jane Smith.)
Figure 33.42 Mucous acini in the sublingual gland. Secretory cells are filled with palestaining mucinogen-containing vesicles and nuclei are displaced basally. (By permission from Kierszenbaum AL 2002 Histology and Cell Biology. St Louis: Mosby.)
CONTROL OF SALIVARY GLAND ACTIVITY The observed wide and rapid variation in the composition, quantity and rate of salivary secretion in response to various stimuli suggests an elaborate control mechanism. Secretion may be continuous, but at a low resting level, and may also occur spontaneously. It is mainly a response to the drying of the oral and pharyngeal mucosae. A rapid increase can be superimposed on the resting level, e.g. during mastication or when stimulated by the autonomic innervation. The controlled variation in the activity of the many types of salivary effector cells (serous, seromucous and mucous secretory cells, myoepithelial cells, epithelial cells of all the ductal elements and the smooth muscle of local blood vessels) affects the quantity and quality of saliva. There is no clear evidence that circulating hormones evoke secretion directly at physiological levels, but they may alter the response of glandular cells to neural stimuli. The control of salivation depends on reflex nerve impulses. The afferent inputs to the reflex arc pass to brain stem salivatory centres, especially from taste and mechanoreceptors in the mouth. A variety of other sensory modalities in and around the mouth are also involved, e.g. smell, for certain aspects of submandibular secretion in man. The afferent input is integrated centrally by the salivatory centres, which are themselves influenced by higher centres. The latter may provide facilitatory or inhibitory influences, which presumably explains why the mouth becomes dry under stress. The efferent drive to the glands passes via parasympathetic and sympathetic outputs from the centres. Relatively little is known about the connections of the preganglionic parasympathetic neurones in the salivatory centres, and virtually nothing is known about either the central location of the sympathetic preganglionic neurones, or the output pathways. No peripheral inhibitory mechanisms exist in the glands. The typical pattern of innervation is shown in Fig. 33.39, but details vary in different glands, and with age. Only the more constant features are illustrated and described here. Cholinergic nerves often accompany ducts and arborize freely around secretory endpieces, but adrenergic nerves usually enter glands along arteries and ramify with them. The main secretomotor nerves are predominantly non-myelinated axons: the few myelinated axons that have been seen are presumably either preganglionic efferent or visceral afferent. Within the glands the nerve fibres intermingle, such that cholinergic and adrenergic axons often lie in adjacent invaginations of one Schwann cell. Secretion and vasoconstriction are mediated via separate sympathetic axons. A single parasympathetic axon may, through serial en passant terminals, induce vasodilatation, secretion and myoepithelial contraction. Secretory endpieces are usually the most densely innervated structures in the gland: individual cells often have both cholinergic and adrenergic innervation. Secretion of water and electrolytes, which creates the foundation for the volume of saliva secreted, is the outcome of a complex set of processes which is largely induced by parasympathetic impulses. Secretion of protein is an ongoing constitutive process wherever it occurs. The regulated exocytosis of prepackaged proteins, which is the principle source of protein secretion into saliva, depends on the relative levels of activity in sympathetic and parasympathetic fibres. page 606 page 607
The ductal elements of salivary glands can markedly modify the composition of saliva. They are less densely innervated than secretory endpieces but their activity is also under neural control. Adrenal aldosterone promotes resorption of sodium and release of potassium into saliva by striated ductal cells, as it does in kidney tubules. Myoepithelial contraction is stimulated mainly by adrenergic innervation, but there may be an additional role for cholinergic axons.
© 2008 Elsevier
TISSUE SPACES AROUND THE JAWS (Fig. 33.43) The dissemination of infection in soft tissues is influenced by the natural barriers presented by bone, muscle and fascia. However, the tissue spaces around the jaws are primarily defined by muscles, principally mylohyoid, buccinator, masseter, medial pterygoid, superior constrictor and orbicularis oris. None of these 'spaces' is actually empty and they should merely be regarded as potential spaces that are normally occupied by loose connective tissue. It is only when inflammatory products destroy the loose connective tissue that a definable space is produced. The spaces are paired except for the submental, sublingual and palatal spaces.
POTENTIAL TISSUE SPACES AROUND THE LOWER JAW (Fig. 33.43) The important potential tissue spaces of the lower jaw are the submental; submandibular; sublingual; buccal; submasseteric; parotid; pterygomandibular; peripharyngeal and peritonsillar spaces.
Figure 33.43 Potential tissue spaces around the jaws. A, Coronal section showing the sublingual and submandibular spaces in the floor of the mouth and the possible routes for the spread of infections from periapical dental abscesses (left). B, Horizontal section through the mandibular molar region showing the associated tissue spaces. C, Inferior view of the floor of the mouth (suprahyoid region of the neck) showing the position of the submandibular and sublingual tissue spaces. (B and C, by permission from Berkovitz BKB, Moxham BJ 2002 Head and Neck Anatomy. London: Martin Dunitz.)
The submental and submandibular spaces are located below the inferior border of the mandible beneath mylohyoid, in the suprahyoid region of the neck. The submental space lies beneath the chin in the midline, between the mylohyoid muscles and the investing layer of deep cervical fascia. It is bounded laterally by the two anterior bellies of the digastric muscles. The submental space communicates posteriorly with the two submandibular spaces. The submandibular space is situated between the anterior and posterior bellies of the digastric muscle and communicates with the sublingual space around the posterior free border of mylohyoid. The sublingual space lies in the floor of the mouth, above the
mylohyoid muscles, and is continuous across the midline: it communicates with the submandibular spaces over the posterior free borders of the mylohyoid muscles. The remaining tissue spaces are illustrated in Fig. 33.43B. The buccal space is located in the cheek, on the lateral side of buccinator. The submasseteric spaces are a series of spaces between the lateral surface of the ramus of the mandible and masseter: they are formed because the fibres of masseter have multiple insertions onto most of the lateral surface of the ramus. The pterygomandibular space lies between the medial surface of the ramus of the mandible and medial pterygoid, and the parotid space lies behind the ramus of the mandible, in and around the parotid gland. The parapharyngeal space is bounded by the superior constrictor of the pharynx and the medial surface of medial pterygoid. It is restricted to the infratemporal region of the head and the suprahyoid region of the neck and communicates with the retropharyngeal space, which itself extends into the retrovisceral space in the lower part of the neck (the tissue spaces of the neck are described on p. 542 and of the pharynx on p. 626). The peritonsillar space lies around the palatine tonsil between the pillars of the fauces, and is part of the intrapharyngeal space. It is bounded by the medial surface of the superior constrictor of the pharynx and its mucosa. page 607 page 608
POTENTIAL TISSUE SPACES AROUND THE UPPER JAW The tissue spaces of the upper jaw are usually associated with spread of infection from the teeth. They are the canine (infraorbital), palatal and infratemporal spaces. The canine (infraorbital) space associated with the canine fossa lies between the levator labii superioris and zygomaticus muscles. The palatal space is not truly a tissue space in the hard palate, as the mucosa there is firmly bound to the periosteum. However, inflammation can strip away some of this periosteum to produce a well-circumscribed abscess. The infratemporal space is the upper extremity of the pterygomandibular space. It is closely related to the maxillary tuberosity and therefore the upper molars.
DENTAL ABSCESS (Fig. 33.43) Abscesses developing in relation to the apices of roots ultimately penetrate the surrounding bone where it is thinnest. The position of the resultant swelling in the soft tissues is largely determined by the relationship between muscle attachments and the sinus (the path taken by the infected material) in the bone. Thus, in the lower incisor region, because the labial bone is thin, abscesses generally appear as a swelling in the labial sulcus, above the attachment of mentalis. The abscess may open below mentalis, when it will point beneath the chin. If an abscess from a lower postcanine tooth opens below the attachment of buccinator, the swelling is in the neck; if it opens above, the swelling is in the buccal sulcus. If an abscess opens lingually above mylohyoid, the swelling is in the lingual sulcus; if it is below, the swelling is in the neck. Third molar abscesses tend to track into the neck rather than the mouth, because mylohyoid ascends posteriorly. Apart from canine teeth, which have long roots, abscesses on upper teeth usually open buccally below, rather than above, the attachment of buccinator. Because its root apex is occasionally curved towards the palate, abscesses of the upper lateral incisors may track into the palatal submucosa. Abscesses of upper canines often open facially just below the orbit. Here the swelling may obstruct drainage in the angular part of the facial vein which has no valves, and it is therefore possible for infected material to travel via the angular and ophthalmic veins into the cavernous sinus. Abscesses on the palatal roots of upper molars usually open on the palate. The superficial lamina of deep cervical fascia opposes the spread of abscesses towards the surface, and pus beneath it tends to migrate laterally. If the pus is in the anterior triangle, it may find its way into the mediastinum, anterior to the pretracheal lamina, but because the fascia here is so thin it more often approaches the surface and 'points' above the sternum. Pus behind the prevertebral lamina may extend laterally and point in the posterior triangle, or it may perforate the lamina and the buccopharyngeal fascia to bulge into the pharynx as a retropharyngeal abscess. Upper second premolars and first and second molars are related to the maxillary sinus. When this is large, the root apices of these teeth may be separated from its cavity solely by the lining mucosa. Sinus infections may stimulate the nerves entering the teeth, simulating toothache. Upper first premolars and third molars
may be closely related to the maxillary sinus. With loss of teeth, alveolar bone is extensively resorbed. Thus in the edentulous mandible the mental nerve, originally inferior to premolar roots, may lie near the crest of the bone. In the edentulous maxilla, the sinus may enlarge to approach the oral surface of the bone. Occasional bony prominences termed the torus mandibularis, torus maxillaris and torus palatinus, may lie lingual to the lower premolars or molars, the upper molars. They in the midline of the palate and may need surgical removal before satisfactory dentures can be fitted. Severe systemic infections during the time the teeth are developing may lead to faults in enamel, which are visible as horizontal lines (cf. Harris's growth lines).
© 2008 Elsevier
34 NECK AND UPPER AERODIGESTIVE TRACT Development of the face and neck FACE Facial development starts during the 4th postovulatory week and is completed at about 18 years of age. Although the most obvious changes in morphology occur up to stage 23 (8 weeks) of development, significant changes in the proportion of the face occur at puberty. The growth and fusion of five main prominences or processes form the face: the midline frontonasal process and bilateral maxillary and mandibular processes from the first pharyngeal arch. Several cell lines are associated with facial development. They are embryonic surface ectoderm, which covers all of the outer, and much of the inner, surfaces of the frontonasal, maxillary and mandibular processes; neural crest from the diencephalic and mesencephalic regions of the neural tube migrates to form the mesenchyme of the frontonasal process, while that from the rostral hindbrain (rhombomeres 1 and 2) migrates to the maxillary and mandibular processes; angiogenic mesenchyme, which develops in and extends from the original aortic arch arteries; paraxial mesenchyme, which provides all the skeletal muscle within the face; prechordal mesenchyme, which gives rise to the extraocular muscles of the eye (Chapter 43); neural plate, from which the afferent components of the cranial nerves extend, the efferent components being provided by neural crest and placodes (Chapter 26).
Figure 34.1 The contribution of the first arch and frontonasal process to the development of the face.
The diencephalic and mesencephalic neural crest mesenchyme proliferates and migrates rostrally and laterally between the ectoderm and prosen-cephalon to form the frontonasal process (Fig. 34.1). Migration begins well before neural tube closure and is completed afterwards. The olfactory placodes within the surface ectoderm remain connected to the neural tube, causing the migrating crest to stream and accumulate around them so that they appear to be displaced to the bottom of olfactory pits. The elevations formed around the pits during stages 14 and 15 are termed medial and lateral nasal processes. By stage 16 the medial processes have moved closer together and project caudally beyond the lateral processes. Internally, the medial processes project into the roof of the stomodeum to form the premaxillary fields. The frontonasal process gives rise to the forehead, nose, philtrum of the upper lip, premaxilla and upper incisor teeth. The surface facial contribution of the frontonasal process, which extends over the supraorbital and glabellar regions, includes the upper eyelid and conjunctiva and the external aspects of the nose. Internally the epithelial contribution includes the nasal vestibule, the nasal mucosa of the conchae and paranasal sinuses and the olfactory epithelium (see p. 610). The mandibular processes approach each other and fuse in the midline superior to the pericardial bulge at stage 12. Viewed from its ventral aspect, the maxillary process a somewhat triangular elevation which arises from the cranial aspect of the dorsal region of each mandibular process from stage 14. Each maxillary process grows in a ventral direction and fuses with the lateral nasal process, although the two are initially separated by a nasomaxillary groove (naso-optic
furrow) (Fig. 34.1). As the opposed margins of the lateral nasal and maxillary processes grow together they establish continuity between the side of the future nose and the cheek (Fig. 34.1). The ectoderm along the boundary between them does not entirely disappear. It gives rise to a solid cellular rod, which at first develops as a linear surface elevation, the nasolacrimal ridge, and then sinks into the mesenchyme. Its caudal end proliferates to connect with the caudal part of the lateral nasal wall, while its cranial extremity later connects with the developing conjunctival sac. The solid rod becomes canalized to form the nasolacrimal duct (Fig. 34.2B). The maxillary processes thus give rise to part of the cheek, maxilla, zygoma, lateral portions of the upper lip, hard and soft portions of the secondary palate. page 609 page 610
Figure 34.2 A, Primitive palate of a human embryo in the seventh week. The figure shows the anterior part of the roof of the mouth; large parts of the left lateral nasal prominence and the left maxillary prominence have been removed to expose the left primitive nasal cavity. B, Oblique coronal section through the head of a human
embryo 23 mm long. The nasal cavities communicate freely with the cavity of the mouth. C, Coronal section through the nasal cavity of a human embryo 28 mm long. (A, from a model by K Peter; C, after Kollman.)
Figure 34.3 The parts of the adult face which are derived from the nasal elevations, and the maxillary and mandibular prominences.
The surface facial contribution of the maxillary process extends from the supratragic point to the lateral angles of eye and mouth, includes the lower eyelid and conjunctiva, and follows the paranasal line of the nasolacrimal duct, finally including a controversial amount of the upper lip. Internally the epithelial contribution includes upper and lower surfaces of the palate, the lining of the maxillary sinus which opens into the nasal cavity, the buccal epithelium lining the upper lip and gums, the invagination of the parotid salivary gland, and the upper teeth from the molars to the canines. The mandibular process gives rise to part of the cheek and the mandible. Fusion between the adjacent surfaces of the mandibular and maxillary processes progressively reduces the relatively wide primitive mouth (stomodeal fissure). At the same time, the epithelial and connective tissues of the cheek enlarge. This proceeds from the para-otic region to the angle of the definitive oral fissure. The external ectoderm over the mandibular process becomes the skin of the face (Fig. 34.3), and it also takes part in the formation of the tragus of the auricle. Its surface facial contribution is roughly triangular: the apex includes the tragus, the upper border extends to the lateral angle of the mouth and the free border of lower lip, and its lower border curves to follow the principal submandibular flexure line of the neck. Internally the mandibular epithelial surface includes the buccal lining of the lower lip and gums, the invagination of the submandibular salivary gland, all the lower teeth and the epithelial surface of the anterior two-thirds of the tongue.
© 2008 Elsevier
FACE Facial development starts during the 4th postovulatory week and is completed at about 18 years of age. Although the most obvious changes in morphology occur up to stage 23 (8 weeks) of development, significant changes in the proportion of the face occur at puberty. The growth and fusion of five main prominences or processes form the face: the midline frontonasal process and bilateral maxillary and mandibular processes from the first pharyngeal arch. Several cell lines are associated with facial development. They are embryonic surface ectoderm, which covers all of the outer, and much of the inner, surfaces of the frontonasal, maxillary and mandibular processes; neural crest from the diencephalic and mesencephalic regions of the neural tube migrates to form the mesenchyme of the frontonasal process, while that from the rostral hindbrain (rhombomeres 1 and 2) migrates to the maxillary and mandibular processes; angiogenic mesenchyme, which develops in and extends from the original aortic arch arteries; paraxial mesenchyme, which provides all the skeletal muscle within the face; prechordal mesenchyme, which gives rise to the extraocular muscles of the eye (Chapter 43); neural plate, from which the afferent components of the cranial nerves extend, the efferent components being provided by neural crest and placodes (Chapter 26).
Figure 34.1 The contribution of the first arch and frontonasal process to the development of the face.
The diencephalic and mesencephalic neural crest mesenchyme proliferates and migrates rostrally and laterally between the ectoderm and prosen-cephalon to form the frontonasal process (Fig. 34.1). Migration begins well before neural tube closure and is completed afterwards. The olfactory placodes within the surface ectoderm remain connected to the neural tube, causing the migrating crest to stream and accumulate around them so that they appear to be displaced to the bottom of olfactory pits. The elevations formed around the pits during stages 14 and 15 are termed medial and lateral nasal processes. By stage 16 the medial processes have moved closer together and project caudally beyond the lateral processes. Internally, the medial processes project into the roof of the stomodeum to form the premaxillary fields. The frontonasal process gives rise to the forehead, nose, philtrum of the upper lip, premaxilla and upper incisor teeth. The surface facial contribution of the frontonasal process, which extends over the supraorbital and glabellar regions, includes the upper eyelid and conjunctiva and the external aspects of the nose. Internally the epithelial contribution includes the nasal vestibule, the nasal mucosa of the conchae and paranasal sinuses and the olfactory epithelium (see p. 610). The mandibular processes approach each other and fuse in the midline superior to the pericardial bulge at stage 12. Viewed from its ventral aspect, the maxillary process a somewhat triangular elevation which arises from the cranial aspect of the dorsal region of each mandibular process from stage 14. Each maxillary process grows in a ventral direction and fuses with the lateral nasal process, although the two are initially separated by a nasomaxillary groove (naso-optic furrow) (Fig. 34.1). As the opposed margins of the lateral nasal and maxillary processes grow together they establish continuity between the side of the future nose and the cheek (Fig. 34.1). The ectoderm along the boundary between them
does not entirely disappear. It gives rise to a solid cellular rod, which at first develops as a linear surface elevation, the nasolacrimal ridge, and then sinks into the mesenchyme. Its caudal end proliferates to connect with the caudal part of the lateral nasal wall, while its cranial extremity later connects with the developing conjunctival sac. The solid rod becomes canalized to form the nasolacrimal duct (Fig. 34.2B). The maxillary processes thus give rise to part of the cheek, maxilla, zygoma, lateral portions of the upper lip, hard and soft portions of the secondary palate. page 609 page 610
Figure 34.2 A, Primitive palate of a human embryo in the seventh week. The figure shows the anterior part of the roof of the mouth; large parts of the left lateral nasal prominence and the left maxillary prominence have been removed to expose the left primitive nasal cavity. B, Oblique coronal section through the head of a human embryo 23 mm long. The nasal cavities communicate freely with the cavity of the mouth. C, Coronal section through the nasal cavity of a human embryo 28 mm long. (A, from a model by K Peter; C, after Kollman.)
Figure 34.3 The parts of the adult face which are derived from the nasal elevations, and the maxillary and mandibular prominences.
The surface facial contribution of the maxillary process extends from the supratragic point to the lateral angles of eye and mouth, includes the lower eyelid and conjunctiva, and follows the paranasal line of the nasolacrimal duct, finally including a controversial amount of the upper lip. Internally the epithelial contribution includes upper and lower surfaces of the palate, the lining of the maxillary sinus which opens into the nasal cavity, the buccal epithelium lining the upper lip and gums, the invagination of the parotid salivary gland, and the upper teeth from the molars to the canines. The mandibular process gives rise to part of the cheek and the mandible. Fusion between the adjacent surfaces of the mandibular and maxillary processes progressively reduces the relatively wide primitive mouth (stomodeal fissure). At the same time, the epithelial and connective tissues of the cheek enlarge. This proceeds from the para-otic region to the angle of the definitive oral fissure. The external ectoderm over the mandibular process becomes the skin of the face (Fig. 34.3), and it also takes part in the formation of the tragus of the auricle. Its surface facial contribution is roughly triangular: the apex includes the tragus, the upper border extends to the lateral angle of the mouth and the free border of lower lip, and its lower border curves to follow the principal submandibular flexure line of the neck. Internally the mandibular epithelial surface includes the buccal lining of the lower lip and gums, the invagination of the submandibular salivary gland, all the lower teeth and the epithelial surface of the anterior two-thirds of the tongue.
© 2008 Elsevier
NASAL CAVITY The apex of the maxillary process extends beyond the lateral nasal process, and crosses the caudal end of the olfactory pit to meet and fuse with the premaxillary elevation that develops at the extremity of the frontonasal process. This closes off the lower or caudal edge of the olfactory pit, and the upper part of the opening is thus defined as the primitive external naris. The growth of the surrounding mesenchyme leads to a deepening of the pit to become a primitive nasal cavity, or nasal sac, the epithelial wall of which is contiguous with the epithelium of the stomodeal roof. The area of contact becomes more extensive as growth continues, and ultimately forms a thin layer, the oronasal membrane (Fig. 34.2A), which later disappears. Thereafter, the primitive nasal cavity communicates with the stomodeum through a primitive internal naris (choana), which at this stage is still situated well forward (ventrally) in the stomodeal roof. The nasal cavity thus acquires a floor through the fusion of the nasal and maxillary processes. At this stage the two external nares are still widely separated by an area derived from the frontonasal process, however, this separation is progressively reduced by the fusion of the premaxillary mesenchyme from the two sides. The primitive nasal cavities remain separated but become much more extensive with time. The nasal septum gradually extends backwards and downwards, and fuses with the maxillary palatal shelves to leave a free edge that reaches as far as the attachment of Rathke's pouch to the roof of the buccal cavity. On each side of the nasal septum, in a ventral or anterior position just above the primitive palate, placodal ectoderm is invaginated to form a pair of small diverticula which extend dorsally and cranially into the septum. These are the vomeronasal organs (Fig. 34.2C), auxiliary olfactory organs whose openings are close to the junction between the two premaxillae and the maxillae. They appear to be rudimentary in humans. A number of elevations appear on the lateral wall of each nasal cavity, which will develop into the superior, middle and inferior conchae. The paranasal sinuses appear in late fetal and early postnatal life as diverticula which gradually invade the frontal, ethmoid and sphenoid bones. These sinuses and the nasolacrimal duct, which is formed at the line of fusion of the maxillary and lateral nasal processes, all terminate in the lateral wall of the nasal cavity.
OLFACTORY EPITHELIUM page 610 page 611
The olfactory nerve fibre bundles (fila olfactoria) develop from a proortion of the placodal cells which line the olfactory pits (p. 610). The cells proliferate and give rise to olfactory receptor cells whose central processes are the axons of the olfactory nerve which grow into the overlying olfactory bulbs. The earliest pioneer neurites are devoid of glial ensheathment and cross a mesenchyme-filled gap between the placode and the superjacent brain. Olfactory axons subsequently become enclosed in the cytoplasmic processes of specialized ensheathing cells derived from the rostral neural crest. Within the olfactory bulb, the terminals of the olfactory axons ramify to establish rudimentary olfactory glomeruli. The remaining placodal cells, probably amplified by rostral neural crest cells, differentiate into columnar supporting (sustentacular) cells, rounded basal cells and, by
invagination, the flattened duct-lining and polyhedral acinar cells of the glands of Bowman. Basal infiltration by lymphocytes is a relatively late event.
PHARYNGOTYMPANIC TUBE The ventral end of the first pouch becomes obliterated, but its dorsal end persists and deepens as the head enlarges. It remains close to the ectoderm of the dorsal end of the first cleft and, together with the adjoining lateral part of the pharynx and dorsal part of the second pharyngeal pouch, constitutes the tubotympanic recess. The recess forms the tympanic cavity, the pharyngotympanic tube and their extensions. The dorsal end of the third arch is also considered to take part in the formation of the floor of the pharyngotympanic tube.
ADENOID AND TUBAL TONSILS A number of focal proliferations of endoderm become invaded by lymphoid tissue. The adenoid or pharyngeal tonsil develops in the posterior midline of the nasopharynx, and the tubal tonsil develops close to the pharyngeal opening of the pharyngotympanic tube (p. 647).
© 2008 Elsevier
BUCCAL CAVITY The buccal cavity develops mainly from ventral growth of the upper pharyngeal arches. The rostral growth of the embryo and the formation of the head fold cause the pericardial area and buccopharyngeal membrane to come to lie on the ventral surface of the embryo (p. 198). Further expansion of the forebrain dorsally, and bulging of the pericardium ventrally, together with enlargement of the facial processes laterally, means that the buccopharyngeal membrane becomes depressed at the base of a hollow, the stomodeum or primitive buccal cavity (Figs 34.1, 34.4). At the end of the fourth week (stage 12) the buccopharyngeal membrane breaks down so that communication is established between the stomodeum and cranial end of the foregut (future naso- and oropharynx respectively). No vestige of the membrane is evident in the adult, and this embryonic communication should not be confused with the permanent oropharyngeal isthmus. The pharyngeal arches grow in a ventral direction and lie progressively between the stomodeum and pericardium. With the fusion of the mandibular processes and the development of the maxillary processes (Fig. 34.1), the opening of the stomodeum assumes a pentagonal form, bounded cranially by the frontonasal process, caudally by the mandibular processes and laterally by the maxillary processes (Figs 34.1, 34.4). The inward growth and fusion of the maxillary palatine processes (Fig. 34.2) divides the stomodeum into a nasal and a buccal part. A shallow groove appears along the free margins of the prominences that bound the mouth cavity, and the ectoderm in its floor thickens and invades the underlying mesenchyme which divides into a medial dental lamina and a lateral vestibular lamina. The central cells of the latter degenerate and the furrow deepens. It is now termed the labiogingival groove or sulcus. The inner wall of the groove contributes to the formation of the alveolar processes of the maxillae and the mandible and their gingivae, while its outer wall forms the lips and cheeks. Ectoderm/mesenchymal interactions produce structures in the wall of the oral cavity (mucous glands, salivary glands, teeth and taste buds), non-keratinized buccal epithelium, and the keratinized layer of skin.
Figure 34.4 The head of a human embryo in the sixth week: ventral aspect. (From a model by K Peter.)
The site of the original buccopharyngeal membrane may be visualized in the older fetus and adult as a zone posterior to the palatoglossal arch which passes from the junction of the anterior two-thirds and posterior one-third of the tongue to the roof of the nasopharynx, just anterior to the entrance to the auditory tube, and then on to the sella turcica, which contains the ectodermally derived adenohypophysis. The thyroid gland and the auditory tubes are the most anterior endodermal derivatives from the upper part of the primitive foregut which arise posterior to the zone. All first arch structures innervated by the trigeminal nerve are anterior to this zone, and permit investigation by touch from the tongue or digits without concern. Structures immediately posterior to the palatoglossal arch
are innervated by the third arch nerve, the glossopharyngeal, and inappropriate touch is associated with initiation of the gag reflex. In the neonate the oral cavity is only potential with the mouth closed. Three spaces are formed in the oral cavity during suckling. There is a median space between the tongue and hard palate which bifurcates posteriorly to produce channels on each side of the approximated soft palate and epiglottis. Two lateral spaces, the lateral arcuate cavities, are formed between the tongue medially and the cheeks laterally: the upper and lower gums situated in these spaces do not touch during suckling. Each cheek is supported by a mass of subcutaneous fat, the suctorial pad, which lies between buccinator and masseter. The larynx is elevated so that its opening is directed into the nasopharynx - i.e. above the level of the spaces - as fluid passes to the pharynx during suckling. This ensures that babies can breathe while suckling. It was thought for many years that neonates preferentially breathe through the nose, resorting to mouth breathing only if the nasal passage is obstructed. Studies have shown that full-term infants are able to establish oral breathing in the presence of nasal occlusion of a mean duration of 7.8 seconds.
PALATE Once the primitive nasal cavities have been defined the ventral part of the roof of the oral cavity can be regarded as the primitive palate (median palatine prominence; Fig. 34.2A). It is formed by the premaxillary regions and maxillary processes which become confluent and establish continuity with the primitive nasal septum. As the head grows in size, the region of mesenchyme between the forebrain and oral cavity increases greatly by cellular proliferation, and the nasal cavities deepen and extend towards the forebrain. At the same time, they also extend dorsally from the primitive choanae as two deep, narrow grooves in the oral roof (Figs 34.2, 34.5) which are separated by a partition. The grooves and the partition deepen together, and the latter becomes the nasal septum, continuous rostrally with the primitive nasal septum (Fig. 34.2B). The broad dorsocaudal border of the nasal septum is at first in contact with the dorsum of the developing tongue (Fig. 34.2B), and the right and left nasal cavities communicate freely with the mouth except ventrally where the nasal floor has already been established by the primitive palate. page 611 page 612
Figure 34.5 A, Coronal sections through developing head showing formation,
elevation and fusion of the palatal processes and the nasal septum. B, View of the roof of the mouth showing fusion of the premaxilla and the maxillary palatal shelves.
During stage 17 (41 days) the internal aspects of the maxillary prominences produce palatine processes (shelves), which grow caudally and contribute to the development of the linguogingival sulci. They are separated from each other by the tongue. A coronal section through the head at stage 20 shows the maxillary palatine processes contiguous with the sides of the tongue and bent into a vertical position on each side of it (Figs 34.2B, 34.5). With further growth, the mandibular region and the tongue are carried forwards, ventrally, and the lingual tip passes round to the caudal surface of the primitive palate. At stage 23 (56-57 days) the palatine processes rapidly elevate, and assume a horizontal position which allows them to grow towards each other and thus to fuse (Figs 34.2C, 34.5): this process occurs from before backwards. The change of position, which occurs very rapidly, is caused by region-specific synthesis and accumulation of hyaluronan within the extracellular matrix of the palatal process mesenchyme. Hyaluronan binds more than 1000 times its own weight of water, causing swelling and expansion of the palatal shelves. The process is aided by the alignment of collagen fibrils and palatal mesenchymal cells in the shelves - the latter contract in response to acetylcholine and serotonin, which they secrete, thus regulating the elevation of the shelves - and by the epithelium, which restrains the swelling. Once these forces are in concert and exceed the resistance factors, the palatal shelves will mechanically elevate. This occurs during a period of continuous growth in head height but almost no growth in head width. This latter factor is important, because if palatal shelf elevation is delayed so that it occurs during a period of growth in facial width, the unfused processes are unable to touch physically and cleft palate may result. Other factors affecting palatal closure are the growth in length of the first arch cartilage, which allows the tongue to lower into the developing mandible; and the change in position of the maxilla relative to the anterior cranial base. This is maintained at about 84° during weeks 9 and 10, and has the effect of lifting the head and upper jaw upwards from the mandible which facilitates withdrawal of the tongue from between the palatal shelves and creates space for them to elevate. Mouth opening, tongue protrusion and hiccup movements have also been noted at this time, and it may be that these movements and their associated pressure changes assist palatal shelf elevation (Ferguson 1990). Generally, in female embryos palatal shelf elevation occurs 7 days later than in males, and congenital cleft palate is more common in female embryos. After elevation, the palatine processes grow medially along the inferior borders of the primitive choanae, and unite with them, and with the margins of the median palatine prominence, except over a small area in the midline where a nasopalatine canal maintains connection between the nasal and oral cavities for some time and marks the future position of the incisive fossa. The plates which form the early (primitive) palate are sometimes known as median palatine processes, the maxillary contributions being then named the lateral palatine processes. As the medial borders of the maxillary palatine processes fuse with each other and with the free border of the nasal septum, the nasal and oral cavities are progressively separated and the tongue is excluded from the nasal cavity. The nasal cavities extend dorsally until the choanae reach their final position. Slightly later the dorsomedial extremities of the palatine processes, which extend dorsally beyond the choanae, fuse together rostrocaudally to form the future epithelia and connective tissues of the soft palate (Fig. 34.2C). There is a later upgrowth of myogenic mesenchyme from the third and other pharyngeal arches into the palate and around the caudal margins of the auditory tube, along a line corresponding to the definitive palatopharyngeal arches. In the neonate the hard palate is only slightly arched, and it is usually corrugated by five or six irregular transverse folds (rugae) which assist gripping of the nipple when suckling. The epiglottis is high and makes direct contact with the soft palate.
TEETH AND GUMS page 612 page 613
Fusion of the mandibular prominences forms the early lower oral margin, and the oral margin is completed by the fusion of the frontonasal process with the maxillary prominences. At this stage there is no distinction between the lips and gums. The lower gingivae form as mesenchymal swellings which are separated from the lip by a labiogingival sulcus and from the tongue by a linguogingival sulcus. The upper gingivae are similarly separated from the lip by a labiogingival sulcus; however, they remain at the border of the palate. The surfaces of the
gingivae differentiate into an oral mucosa that is contiguous with the mucosa lining the lips and cheeks. A series of proliferative epithelial loci signal the site of future tooth development along the ridges of each gum. Teeth form from a series of epithelial/mesenchymal interactions along the dental lamina (Fig. 34.6). In 27 mm embryos individual dental laminae expand into ectodermal (dental) sacs surrounded by vascular mesenchyme. The ectoderm proliferates to form an enamel organ which surrounds a local portion of first arch neural crest mesenchyme, the dental papilla: collectively this unit constitutes a tooth bud or germ. The enamel organ initially forms a cap over the dental papilla and then later it expands into a bell shape. The inner layer is tightly adherent to the dental papilla and separated from the outer layer by accumulated glycosaminoglycans. The inner cells of the enamel organ differentiate into ameloblasts, and the underlying mesenchymal cells into odontoblasts. Deposition of extracellular matrix onto the adjoining basal lamina produces the tooth. The neural crest dental papilla mesenchyme specifies the type of tooth produced, i.e. incisor or molar; however, the mandibular epithelium is essential and specific for the development of teeth. Recombination of mandibular epithelium and second arch mesenchyme results in tooth development, but second arch epithelium and mandibular mesenchyme does not. Both the deciduous and permanent teeth develop in the same manner. The permanent teeth develop in accessional positions from the lingual aspects of the existing tooth germs. The tooth germs for the 12 permanent molar teeth develop from posterior extensions of the dental laminae on each side of both jaws. Calcification begins in both deciduous and permanent teeth before birth. The deciduous teeth have well-developed crowns by full term, whereas the permanent teeth remain as tooth buds.
Figure 34.6 Reciprocal tissue interaction in mammalian tooth development. The sequence of epithelial/mesenchymal interactions involved in the development of teeth in the embryonic mouse. (From a model by K Peter.)
SALIVARY GLANDS The salivary glands arise bilaterally as a result of epithelial/mesenchymal interactions between the ectodermal epithelial lining of the buccal cavity and the subjacent neural crest mesenchyme. The parotid gland can be recognized in human embryos at stage 15 as an elongated furrow running dorsally from the angle of the mouth between the mandibular and maxillary prominences. The
groove, which is converted into a tube, loses its connection with the epithelium of the mouth, except at its ventral end, and grows dorsally into the substance of the cheek. The tube persists as the parotid duct and its blind end proliferates in the local mesenchyme to form the gland. Subsequently the size of the oral fissure is reduced by partial fusion between the maxillary and mandibular prominences and the duct opens thereafter on the inside of the cheek at some distance from the angle of the mouth. In the neonate the parotid gland is rounded and lies between masseter and the ear. During infancy and early childhood, the growing gland covers the parotid duct. The submandibular gland is identifiable in 13 mm human embryos as an epithelial outgrowth into the mesenchyme from the floor of the linguogingival groove. It increases rapidly in size, and gives off numerous branching processes which later acquire lumina. At first the connection of the submandibular outgrowth with the floor of the mouth lies at the side of the tongue, but the edges of the groove in which it opens come together, from behind forwards, and form the tubular part of the submandibular duct. As a result, the orifice of the duct is shifted forwards till it is below the tip of the tongue, close to the median plane. The sublingual gland arises in 20 mm embryos as a number of small epithelial thickenings in the linguogingival groove and on the lateral side of the groove: the groove later closes to form the submandibular duct. Each thickening canalizes separately, and so many of the multiple sublingual ducts open separately on the summit of the sublingual fold, while others join the submandibular duct. The topography of the submandibular and sublingual glands is the same as in the adult.
TONGUE page 613 page 614
A small median elevation, the tuberculum impar or median tongue bud, appears in the floor of the pharynx before the pharyngeal arches meet ventrally, and it subsequently becomes incorporated in the anterior part of the tongue. A little later two oval mandibular or lingual swellings appear on the inner aspect of the mandibular processes. They meet each other in front, and caudally they converge on the tuberculum impar, with which they fuse (Fig. 34.7). A sulcus forms along the ventral and lateral margins of this elevation and deepens, internal to the future alveolar process of the mandible, to form the linguogingival groove, while the elevation constitutes the anterior or buccal (presulcal) part of the tongue. Caudal to the tuberculum impar, a second median elevation, the hypobranchial eminence (copula of His), forms in the floor of the pharynx, and the ventral ends of the fourth, third and, later, second, pharyngeal arches converge into it. A transverse groove separates its caudal part to delineate the epiglottis. Ventrally it approaches the presulcal tongue rudiment, and spreads in the form of a V, to form the posterior or pharyngeal part of the tongue. During this sequence of events, the third arch elements grow over and bury the elements of the second arch, thereby excluding it from the tongue. The mucous membrane of the pharyngeal part of the tongue therefore receives its sensory supply from the glossopharyngeal nerve, which is the nerve of the third arch. In the adult, the union of the anterior and posterior parts of the tongue approximately corresponds to the angulated sulcus terminalis, which has its apex at the foramen caecum, a blind depression produced at the time of fusion of the constituent parts of the tongue, but also marking the site of ingrowth of the median rudiment of the thyroid gland. The tongue initially consists of a mass of mesenchyme covered on its surface by first arch ectoderm and third arch endoderm. During the second month it is invaded by occipital myotomes which migrate from the lateral aspects of the myelencephalon. They pass ventrally round the pharynx to reach its floor accompanied by the hypoglossal nerve. The composite character of the tongue is revealed by its sensory innervation. The anterior, buccal part is innervated by the lingual nerve, derived from the posttrematic nerve of the first arch (mandibular nerve) and by the chorda tympani, which is often held to be the pretrematic nerve to the first arch. The posterior, pharyngeal part of the tongue is innervated by the glossopharyngeal, the nerve of the third arch, and the root of the tongue, near the epiglottis, is innervated by the vagus. The sulcus terminalis cannot be distinguished earlier than the 52 mm stage according to some observers. The vallate papillae appear at about the same time, and increase in number until the 170 mm stage. Serial reconstructions suggest
that the territory of the glossopharyngeal nerve extends considerably beyond these papillae. Lymphoid tissue similar to the palatine tonsils usually develops on the surface of the posterior part of the tongue, and is called the lingual tonsil. In the neonate, the tongue is short and broad, and its entire surface lies within the oral cavity (Fig. 11.5). The posterior third of the tongue descends into the neck during the first postnatal year, and by the fourth or fifth year the tongue forms part of the anterior wall of the pharynx.
Figure 34.7 A, The floor of the pharynx of a human embryo at the beginning of the sixth week. B, The floor of the pharynx of a human embryo, about 6 weeks old. (From a model by K Peter.)
TONSILS The palatine tonsils develop from the ventral parts of the second pharyngeal pouches (Fig. 34.8). The endoderm lining these pouches grows into the surrounding mesenchyme as a number of solid buds, which are excavated by degeneration and shedding of their central cells, forming tonsillar fossae and crypts. Lymphoid cells accumulate around the crypts and become grouped as lymphoid follicles. A slit-like intratonsillar cleft extends into the upper part of the tonsil and is possibly a remnant of the second pharyngeal pouch. The paired palatine tonsils are situated slightly higher in the tonsillar fossae in the neonate than in the adult. Each descends in position during the second and third postnatal year, but definitive lymph nodes appear after birth. The palatine tonsils begin to atrophy from the fifth year and involution is often complete by puberty.
ANOMALIES OF FACIAL DEVELOPMENT Congenital malformations produced by arrest of development and/or failure of fusion of components in the formation of the face and palate are not uncommon and variations in the degree of severity of the anomaly produced are seen. Failure of local fusion of one maxillary process with the corresponding premaxillary region, leading to a persistent fissure between the philtrum and lateral part of the upper lip on that side, is called a cleft lip. If the palatal shelves fail to fuse across
the midline and with the nasal septum a cleft palate is produced. Failure of fusion of the maxillary process with its adjacent lateral nasal process will lead to a cleft face. The further growth of the face during the fetal period has received little attention, although this period is by no means characterized entirely by incremental growth. It is during fetal life that human facial proportions develop (Fig. 34.9). The facial and cranial parts display different patterns of growth, though each influences the other. Cleft lip page 614 page 615
Figure 34.8 A, Primitive pharynx situated between neural tube and pericardial cavity. B, Ventral aspect of the endoderm of the pharynx showing three pharyngeal pouches. The areas of contact of the pharyngeal endoderm with the surface ectoderm are shown as flattened surfaces. C, Ventral and dorsal diverticuli of the third and fourth pharyngeal pouches and midline thyroid gland at 6 weeks. D, Thymus, thyroid and parathyroid glands at 7 weeks. E, Thymus, thyroid and parathyroid glands at 7.5 weeks. (Redrawn by permission from Hamilton WJ, Boyd JD, Mossman HW 1962 Human Embryology: Prenatal Development of Form and Function. Cambridge: W Heffer & Sons.)
Figure 34.9 Much of the postnatal growth of the skull is concerned with development of the viscerocranium. This diagram shows that with the height of the cranial vault expressed as similar in newborn and adult skulls (lines a - - - b) the facial skeleton increases particularly during childhood and puberty.
A lack of normal nasolabial muscle attachments is a consequence of uni- or bilateral cleft lip. This results in functional abnormalities that lead to underlying skeletal malformations. The three functional groups of superficial facial muscles nasolabial (transverse nasalis, levator labii superioris and levator labii superioris alequae nasi), bilabial (orbicularis oris) and labiomental (depressor anguli oris) are all displaced inferiorly. The absence of the correct insertion of the transverse muscles of the nose and of orbicularis oris on the medial aspect of the cleft onto
the tissues around the anterior nasal spine and nasal septum and, most importantly, to the contralateral muscles, is responsible for the deviation of the anterior border of the nasal septum to the non-cleft side. A further consequence is the underdevelopment of the incisor-bearing part of the maxilla. These abnormalities in turn influence the mucocutaneous tissues, which results in the displacement of the skin of the nostril to the upper part of the lip, retraction of the labial skin and abnormalities of the soft tissues on either side of the mucocutaneous junction (Fig. 34.10A,B). page 615 page 616
Figure 34.10 Unrepaired clefts of lip and palate. A, Bilateral cleft lip. B, Bilateral cleft lip and cleft palate. C, Cleft palate. D, Cleft lip and palate. (A-C, by permission from O'Doherty NJ 1975 Atlas of the Newborn. 5 Congenital Abnormality. London: Pharma Books. D, by kind permission from Dr BAW Brown and by permission from Berkovitz BKB, Holland GR, Moxham BJ 2002 Oral Anatomy, Embryology and Histology, 3rd edn. Edinburgh: Mosby.)
As a consequence of the inadequate muscular support, the skin of the nasal floor, the nasal sill, the vestibule and the base of the columella - the lowest, mobile, part of the septal cartilage - drift inferiorly and lie in the region normally occupied by the upper lip. The skin of the lip on both sides of the cleft has an abnormal domelike appearance because the underlying orbicularis oris is not continuous across the cleft but remains bunched up in the lip. The cartilages of the nose on the side of the cleft adopt a characteristic twisted appearance. The base is more inferiorly positioned, the cartilage is flattened and distended and the cartilaginous nasal septum is deviated to the non-cleft side. These deformities are the direct result of abnormal muscle attachments that alter the functional matrix. The underlying maxilla is also affected. The greater segment is displaced toward the non-cleft side and the minor segment toward the cleft side. Again this is due to insufficient stimuation by normally functioning overlying musculature. The cleft in the lip is normally closed postnatally between the age of 5 and 6 months. Several techniques have been described for lip closure, and they differ most obviously in the description of the details of the skin incisions. The underlying principle is to restore the muscle insertions to their normal position so that functional extracellular matrix can influence the subsequent growth and development. Each of the three muscle groups - nasolabial, bilabial and labiomental - must be separately identified and reattached. Provided this is achieved, subsequent growth and development should proceed normally. The previously displaced maxillary segments will spontaneously correct their position: bone grafting may be undertaken at the age of 9-10 years. Cleft palate
Many varieties of cleft palate have been observed (Fig. 34.10C,D). The commonest type is unilateral, where only one side of the nasal cavity is in communication with the mouth and the extent of the cleft is variable. In the mildest forms, only the soft palate, and sometimes just the uvula, is cleft. Very rarely, palatopharyngeal incompetence is due to muscle hypoplasia, particularly of the musculus uvulae, and a submucous cleft may be revealed clinically as a Vshaped notch in the midline of the soft palate during function. Such examples of arrested development may be associated with disturbances in embryonic nutrition during the second and the third months of gestation and the grosser varieties, where there is protrusion of the premaxillary region, with associated forwards extension of the nasal septum, are usually coupled with malformations in other regions of the body. The timing of surgical closure of a cleft palate is critical. All surgical techniques for palatal closure involve raising and mobilizing flaps from adjacent structures. This inevitably results in the formation of scar tissue which subsequently inhibits growth and development of the palate and upper jaw. Many techniques use flaps raised from the vomer to close the nasal layer and lateral palatal flaps based posteriorly on the greater palatine arteries. Unfortunately the use of vomerine flaps results in a reduction of vertical growth of the maxilla whereas the mobilization of palatal mucoperiosteal flaps results in a reduction of transverse growth. For these reasons, palatal closure is delayed for as long as possible. However, if such surgery is delayed beyond 18 months of age, when the infant
first begins to develop speech sounds, the child might never develop normal speech. Whatever technique is used to close a cleft palate, it is important to reestablish normal muscle function by approximating the palatoglossus and tensor palatini muscles in the midline whilst avoiding scarring as much as possible (Markus et al 1993). Cleft face page 616 page 617
Facial cleft is a rare malformation which follows failure of fusion between the maxillary process and the lateral nasal process. Here the nasolacrimal duct persists as an open furrow, a condition which is usually associated with ipsilateral cleft lip. The palatine processes may fail to fuse with each other and the nasal septum to variable degrees. In its severest form fusion is wholly lacking, and this produces a wide fissure between the palatine processes through which the nasal septum is visible. On each side the premaxillary parts of the palate are separated from the maxillary palatine processes by clefts which are continuous ventrally with bilateral clefts in the upper lip. In such cases the philtrum is a separate entity that is continuous cranially and dorsally with the nasal septum. The floor of the nasal cavity is deficient throughout its extent and the choanae are incomplete. Midline anomalies such as median cleft lip (true hare lip), cleft nose and cleft lower jaw are rarely encountered. More common are minor degrees of cleft chin and micrognathia.
© 2008 Elsevier
NECK PARATHYROID GLAND The parathyroid glands develop from interactions between the third and fourth pharyngeal pouch endoderm and local cranial (vagal) neural crest mesenchyme. The third pharyngeal pouch has dorsal and ventral sites of proliferation (Fig. 34.8). Bilaterally, the epithelium on the dorsal aspect of the pouch and in the region of its duct-like connection with the cavity of the pharynx becomes differentiated as the primordium of the inferior parathyroid glands (parathyroid III). Although the connection between the pouches and the pharynx is soon lost, the connection between the dorsal parathyroids III and the ventral thymic rudiments persists for some time, and the former passes caudally with the developing thymus. The superior parathyroid glands (parathyroid IV) develop from the dorsal recess of the fourth pharyngeal pouches. They come into relation with, and appear to be anchored by, the lateral lobes of the thyroid gland and thus remain cranial to the parathyroid glands derived from the third pouches. The cardiac neural crest mesenchyme provides the connective tissue elements whereas invading angiogenic mesenchyme gives rise to the vasculature which includes fenestrated capillaries and lymphatics. In the neonate the parathyroid glands are as variable in size and position as they are in the adult. They double in size between birth and puberty. Parathyroid hormone is produced from the 12th week of development.
THYMUS The thymus gland is formed from the ventral part of the third pharyngeal pouch on each side (Fig. 34.8). It cannot be recognized prior to the differentiation of the inferior parathyroid glands at stage 16, but thereafter it is represented by two elongated diverticula which soon become solid cellular masses and grow caudally into the surrounding cardiac (vagal) neural crest mesenchyme. Ventral to the aortic sac the two thymic rudiments meet but do not fuse, and they are subsequently united by connective tissue only. The connection with the third pouch is soon lost, but the stalk may persist for some time as a solid, cellular cord. As the thymus proliferates and descends the local cardiac (vagal) neural crest mesenchyme controls the pattern and development of the gland. Defective development of cardiac neural crest which affects the heart and peripheral neural ganglia also results in thymic deficiencies as seen in the DiGeorge and Pierre Robin syndromes. Crest mesenchyme forms connective tissue septa which produce the lobulated architecture of the gland. Angiogenic mesenchyme, including lymphoid stem cells, invades this local mesenchyme and by 10 weeks, over 95% of the cells in the gland belong to the T cell lineage, with a few developing erythroblasts and B lymphocytes. Hassall's corpuscles are also present. By 12 weeks, the mesenchymal septa, blood vessels and nerves have reached the newly differentiating medulla, which allows the entry of macrophage lineage precursors. Macrophages and interdigitating cells are first seen at 14 weeks. Granulopoiesis occurs in the perivascular spaces. By 17 weeks, the
thymus appears fully differentiated, and after this time it produces the main type of thymocyte that is present throughout life (designated TdT+). Thymic tissue may also develop from the ventral recess of the fourth pharyngeal pouch in a proportion of embryos, when it is usually found close to the thyroid gland in close association with the superior parathyroid gland. An ectodermal contribution to the thymus, probably of placodal origin, occurs in some mammals but a similar contribution in man is conjectural.
CAUDAL PHARYNGEAL COMPLEX The most caudal endodermal invaginations of the pharynx are the fourth pharyngeal pouch, elements of the transitory fifth pharyngeal pouch and the ultimobranchial body (Fig. 34.8). Collectively these diverticuli are termed the caudal pharyngeal complex, and they are connected to the pharynx via the pharyngobranchial duct. They are surrounded by cardiac (vagal) neural crest and by the tissues of the developing thyroid gland. The cells of the ultimobranchial body, the lowest of the pharyngeal pouches, become incorporated into the lateral thyroid lobes, and give rise to the 'C' or parafollicular cells of the thyroid gland. Ccell hyperplasia is associated with medullary carcinoma, and has been reported within the neck in what are presumed to be remnants of the ultimobranchial body.
THYROID GLAND The thyroid gland is a midline derivative of the pharynx. It is first identifiable in embryos of c.20 somites as a median thickening of endoderm lying in the floor of the pharynx between the first and second pharyngeal pouches and immediately dorsal to the aortic sac. This area is later invaginated to form a median diverticulum which appears late in the fourth week in the furrow immediately caudal to the median tongue bud (Fig. 34.8). It grows caudally as a tubular duct. The tip of the duct bifurcates and the tissue mass subsequently divides into a series of double cellular plates, from which the isthmus and the lateral lobes of the thyroid gland are developed. The primary thyroid follicles differentiate by reorganization and proliferation of the cells of these plates. Secondary follicles subsequently arise by budding and subdivision. These primary and secondary endodermal cells are the progenitors of the follicular parenchyma proper. The parafollicular or C cells of the thyroid gland are derived from the ultimobranchial body. The original diverticulum, its bifurcation and generations of follicles invade the cardiac neural crest mesenchyme. The latter gives rise to the connective tissue capsule, interlobular septa and perifollicular investments which carry the main neurovascular and lymphatic supply to the gland. The median diverticulum is connected to the pharynx by the thyroglossal duct. The site of its initial connection with the endodermal floor of the mouth is marked by the foramen caecum, whence it extends caudally in the midline ventral to the primordium of the hyoid bone (behind which it later forms a recurrent loop). The distal part of the duct may differentiate into the pyramidal lobe and levator muscle - or suspensory fibrous band - of the thyroid. The remainder becomes fragmented and disappears, although the lingual part is often identifiable until late in fetal life.
Occasionally parts of the midline thyroglossal duct persist and may occur in lingual, suprahyoid, retrohyoid, or infrahyoid positions. They may form aberrant masses of thyroid tissue, cysts, fistulae or sinuses, usually in the midline. A lingual thyroid situated at the junction of the buccal and pharyngeal parts of the tongue is not uncommon, but nodules of glandular tissue may also be found other than in the midline, e.g. laterally placed posterior to sternocleidomastoid, and, on occasion, below the level of the thyroid isthmus. The thyroid gland is relatively large in the neonate (Fig. 11.4), where it has a long narrow isthmus connecting lobes which do not yet contact the upper part of the trachea. The gland attains half the adult size by 2 years postnatally. Colloid is present in the gland from 3 months' gestation and thyroxin is present by 4.5 months' gestation. UPDATE Date Added: 20 September 2005 Publication Services, Inc. Abstract: Thyroid hemiagenesis: a report of three cases and a review of the literature. Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=15909495&query_hl=11 Thyroid hemiagenesis: a report of three cases and review of the literature. Buyukdereli G, Guney IB, Kibar M et al: Ann Nucl Med 19(2):147-150, 2005.
© 2008 Elsevier
VASCULAR SUPPLY OF THE FACE, SCALP AND NECK ARTERIAL SUPPLY page 617 page 618
Figure 34.11 The origins of the main cranial arteries. (After Padget DH 1948 The development of cranial arteries in the human embryo. Contrib Embryol Carnegie Inst Washington 32: 205-261, by permission.)
The arteries supplying the face, scalp and neck arise from the early aortic arch arteries (p. 1042) which develop in a craniocaudal sequence. The first and second aortic arch arteries supply embryonic structures which are dwindling in size as the lower arches are established. When the first and second aortic arch arteries begin to regress, the supply to the corresponding arches is derived from a transient ventral pharyngeal artery, which grows from the aortic sac. It terminates by dividing into mandibular and maxillary branches. Later the stapedial artery develops from the dorsal stem of the second arch artery. It passes through the condensed mesenchymal site of the future ring of the stapes to anastomose with the cranial end of the ventral pharyngeal artery and so annexes its terminal distribution. The fully developed stapedial artery possesses three branches, mandibular, maxillary and supraorbital, which follow the divisions of the trigeminal nerve (Fig. 34.11). The mandibular and maxillary branches diverge from a common stem. When the external carotid artery emerges from the base of the third arch it incorporates the stem of the ventral pharyngeal artery, and its maxillary branch communicates with the common trunk of origin of the maxillary and mandibular branches of the stapedial artery and annexes these vessels. The proximal part of the common trunk persists as the root of the middle meningeal artery. More distally the meningeal artery is derived from the proximal part of the
supraorbital artery. The maxillary branch becomes the infraorbital artery and the mandibular branch forms the inferior alveolar artery. When the definitive ophthalmic artery differentiates as a branch from the terminal part of the internal carotid artery, it communicates with the supraorbital branch of the stapedial artery which distally becomes the lacrimal artery. The latter retains an anastomotic connection with the middle meningeal artery. The dorsal stem of the original second arch artery remains as one or more caroticotympanic branches of the internal carotid artery.
VENOUS DRAINAGE The primary vessels consist of a close-meshed capillary plexus drained on each side by the precardinal vein, which is at first continuous cranially with a transitory primordial hindbrain channel that lies on the neural tube medial to the cranial nerve roots. This is soon replaced by the primary head vein which runs caudally from the medial side of the trigeminal ganglion, lateral to the facial and vestibulocochlear nerves and the otocyst, then medial to the vagus nerve, to become continuous with the precardinal vein. A lateral anastomosis subsequently brings it lateral to the vagus nerve. The ventral pharyngeal vein drains the mandibular and hyoid arches into the common cardinal vein. As the neck elongates, its termination is transferred to the cranial part of the precardinal vein which later becomes the internal jugular vein. The ventral pharyngeal vein receives tributaries from the face and tongue and becomes the linguofacial vein. As the face develops, the primitive maxillary vein extends its drainage into the territories of supply of the ophthalmic and mandibular divisions of the trigeminal nerve, including the pterygoid and temporal muscles, and it anastomoses with the linguofacial vein over the lower jaw. This anastomosis becomes the facial vein which receives a strong retromandibular vein from the temporal region, and drains through the linguofacial vein into the internal jugular. The stem of the linguofacial vein is now the lower part of the facial vein, whilst the dwindling connection of the facial with the primitive maxillary becomes the deep facial vein. The external jugular vein develops from a tributary of the cephalic vein from the tissues of the neck and anastomoses secondarily with the anterior facial vein. At this stage the cephalic vein forms a venous ring around the clavicle by which it is connected with the caudal part of the precardinal vein. The deep segment of the venous ring forms the subclavian vein and receives the definitive external jugular vein. The superficial segment of the venous ring dwindles, but may persist in adult life. REFERENCES Ferguson MWF 1991 The orofacial region. In: Wigglesworth JS, Singer DB (eds) Textbook of Fetal and Perinatal Pathology, Chapter 22. Oxford: Blackwell Scientific. Markus AF, Smith WP, Delaire J 1993 Primary closure of cleft palate: a functional approach. Br J Oral Maxillofac Surg 31: 71-7. Medline Similar articles Full article
© 2008 Elsevier
35 NECK AND UPPER AERODIGESTIVE TRACT Pharynx The pharynx is a 12-14 cm long musculomembranous tube shaped like an inverted cone. It extends from the cranial base to the lower border of the cricoid cartilage (the level of the sixth cervical vertebra), where it becomes continuous with the oesophagus. The width of the pharynx varies constantly because it is dependent on muscle tone, especially of the constrictors: at rest the pharyngooesophageal junction is closed as a result of tonic closure of the cricopharyngeal sphincter, and during sleep muscle tone is low and the dimensions of the pharynx are markedly decreased (which may give rise to snoring and sleep apnoea). The pharynx is limited above by the posterior part of the body of the sphenoid and the basilar part of the occipital bone, and it is continuous with the oesophagus below. Behind, it is separated from the cervical part of the vertebral column and the prevertebral fascia that covers longus colli and capitis by loose connective tissue in the prevertebral space. The muscles of the pharynx are the three constrictors and the three elevators. The constrictors may be thought of as three overlapping cones which arise from structures at the sides of the head and neck and pass posteriorly to insert into a midline fibrous band, the pharyngeal raphe. The arterial supply to the pharynx is derived from branches of the external carotid artery, particularly the ascending pharyngeal artery, but also from the ascending palatine and tonsillar branches of the facial artery; the maxillary artery (greater palatine, pharyngeal and artery of the pterygoid canal), and from dorsal lingual branches of the lingual artery. The pharyngeal veins begin in a plexus external to the pharynx, receive meningeal veins and a vein from the pterygoid canal, and usually end in the internal jugular vein. Lymphatic vessels from the pharynx and cervical oesophagus pass to the deep cervical nodes, either directly or through the retropharyngeal or paratracheal nodes. The motor and sensory innervation is principally via branches of the pharyngeal plexus. The pharynx lies behind, and communicates with, the nasal, oral and laryngeal cavities via the nasopharynx, oropharynx and laryngopharynx respectively (Fig. 35.1). Its lining mucosa is continuous with that lining the pharyngotympanic tubes, nasal cavity, mouth and larynx.
© 2008 Elsevier
NASOPHARYNX BOUNDARIES The nasopharynx lies above the soft palate and behind the posterior nares, which allow free respiratory passage between the nasal cavities and the nasopharynx (Figs 35.1, 35.2). The nasal septum separates the two posterior nares, each of which measures c.25 mm vertically and 12 mm transversely. Just within these openings lie the posterior ends of the inferior and middle nasal conchae. The walls of the nasopharynx are rigid (except for the soft palate) and its cavity is therefore never obliterated, unlike the cavities of the oro- and laryngopharynx. The nasal and oral parts of the pharynx communicate through the pharyngeal isthmus which lies between the posterior border of the soft palate and the posterior pharyngeal wall. Elevation of the soft palate and constriction of the palatopharyngeal sphincter close the isthmus during swallowing. The nasopharynx has a roof, a posterior wall, two lateral walls and a floor. The roof and posterior wall form a continuous concave slope that leads down from the nasal septum to the oropharynx. It is bounded above by mucosa overlying the posterior part of the body of the sphenoid, and further back by the basilar part of the occipital bone as far as the pharyngeal tubercle. Further down, the mucosa overlies the pharyngobasilar fascia and the upper fibres of the superior constrictor, and behind these, the anterior arch of the atlas. A lymphoid mass, the adenoid, lies in the mucosa of the upper part of the roof and posterior wall in the midline. The lateral walls of the nasopharynx display a number of important surface features. On either side each receives the opening of the pharyngotympanic tube (also termed the auditory or Eustachian tube), situated 10-12 mm behind and a little below the level of the posterior end of the inferior nasal concha (Fig. 35.1). The tubal aperture is approximately triangular in shape, and is bounded above and behind by the tubal elevation which consists of mucosa overlying the protruding pharyngeal end of the cartilage of the pharyngotympanic tube. A vertical mucosal fold, the salpingopharyngeal fold, descends from the tubal elevation behind the aperture (Fig. 35.1) and covers salpingopharyngeus in the wall of the pharynx (Fig. 35.3), and a smaller salpingopalatine fold extends from the anterosuperior angle of the tubal elevation to the soft palate in front of the aperture. As levator veli palatini enters the soft palate it produces an elevation of the mucosa immediately below the tubal opening (Fig 35.3). A small mass of lymphoid tissue, the tubal tonsil, lies in the mucosa immediately behind the opening of the pharyngotympanic tube. Further behind the tubal elevation there is a variable depression in the lateral wall, the pharyngeal recess (fossa of Rosenmüller). The floor of the nasopharynx is formed by the upper surface of the soft palate.
MICROSTRUCTURE OF NON-TONSILLAR NASOPHARYNX The nasopharyngeal epithelium anteriorly is ciliated, pseudostratified respiratory in type (p. 31, Fig. 3.6), with goblet cells. The ducts of mucosal and submucosal seromucous glands open onto its surface. Posteriorly the respiratory epithelium changes to non-keratinized stratified squamous epithelium which continues into the oropharynx and laryngopharynx. The transitional zone between the two types of epithelium consists of columnar epithelium with short microvilli instead of cilia. Superiorly this zone meets the nasal septum, laterally it crosses the orifice of the pharyngotympanic tube, and it passes posteriorly at the union of the soft palate and the lateral wall. There are numerous mucous glands around the tubal orifices.
INNERVATION Much of the mucosa of the nasopharynx behind the pharyngotympanic tube is supplied by the pharyngeal branch of the pterygopalatine ganglion which traverses the palatovaginal canal with a pharyngeal branch of the maxillary artery. The maxillary nerve is thought to transmit the principal sensory supply from the pharyngotympanic tube and middle ear cavity, presumably through its pharyngeal branch.
ADENOID OR NASOPHARYNGEAL TONSIL (Fig. 35.4)
The adenoid, or nasopharyngeal tonsil, is a median mass of mucosa-associated lymphoid tissue (MALT, p. 77) situated in the roof and posterior wall of the nasopharynx. At its maximal size (during the early years of life) it is shaped like a truncated pyramid, often with a vertically oriented median cleft, so that its apex points towards the nasal septum and its base at the junction of the roof and posterior wall of the nasopharynx. page 619 page 620
Figure 35.1 Median sagittal section through the head and neck. Where it passes through the brain, the section passes slightly to the right of the median plane but, below the base of the skull, including the nasal cavity, it passes slightly to the left of the median plane.
The free surface of the nasopharyngeal tonsil is marked by folds that radiate forwards and laterally from a median blind recess, the pharyngeal bursa (bursa of Luschka), which extends backwards and up. The recess marks the rostral end of the embryological notochord. The number and position of the folds and of the deep fissures which separate them vary. A median fold may pass forwards from the pharyngeal bursa towards the nasal septum, or instead a fissure may extend forwards from the bursa, dividing the nasopharyngeal tonsil into two distinct halves which reflect its paired developmental origins (Fig 35.4). The prenatal origins and growth of the nasopharyngeal tonsil are described on pages 611 and 647. After birth it initially grows rapidly, but usually undergoes a degree of involution and atrophy from the age of 8-10 years (although hypoplasia may still occur in adults up to the seventh decade). Relative to the volume of the nasopharynx, the size of the tonsil is largest at 5 years, which may account for the frequency of nasal breathing problems in preschool children, and the incidence of adenoidectomy in this age group.
Vascular supply and lymphatic drainage
The arterial supply of the nasopharyngeal tonsil is derived from the ascending pharyngeal and ascending palatine arteries, the tonsillar branches of the facial artery, the pharyngeal branch of the maxillary artery and the artery of the pterygoid canal. In addition, a nutrient or emissary vessel to the neighbouring bone, the basisphenoid artery, which is a branch of the inferior hypophysial arteries, supplies the bed of the nasopharyngeal tonsil and is a possible cause of persistent postadenoidectomy haemorrhage in some patients. Numerous communicating veins drain the nasopharyngeal tonsil into the internal submucous and external pharyngeal venous plexuses. They emerge from the deep lateral surface of the tonsil and join the external palatine (paratonsillar) veins (Fig. 35.6), and pierce the superior constrictor either to join the pharyngeal venous plexus, or to unite to form a single vessel that enters the facial or internal jugular vein. They may also connect with the pterygoid venous plexus. Microstructure of the nasopharyngeal tonsil (Fig. 35.5)
The adenoid is covered laterally and inferiorly mainly by ciliated respiratory epithelium which contains scattered small patches of non-keratinized stratified squamous epithelium. Its superior surface is separated from the periosteum of the sphenoid and occipital bones by a connective tissue hemicapsule to which the fibrous framework of the tonsil is anchored. The latter consists of a mesh of collagen type III (reticular) fibres which supports a lymphoid parenchyma similar to that in the palatine tonsil. The nasopharyngeal epithelium lines a series of mucosal folds around which the lymphoid parenchyma is organized into follicles and extrafollicular areas. Internally, the tonsil is subdivided into four to six lobes by connective tissue septa, which arise from the hemicapsule and penetrate the lymphoid parenchyma. Seromucous glands lie within the connective tissue, and their ducts extend through the parenchyma to reach the nasopharyngeal surface. Functions of the nasopharyngeal tonsil
The nasopharyngeal tonsil forms part of the circumpharyngeal lymphoid ring (Waldeyer's ring), and therefore presumably contributes to the defence of the upper respiratory tract. The territories served by its lymphocytes are uncertain, but may include the nasal cavities, nasopharynx, pharyngotympanic tubes and the middle and inner ears. page 620 page 621
Figure 35.2 The pharyngeal musculature, exposed from the posterior aspect.
Adenoidectomy
Surgical removal of the adenoids is commonly performed to clear nasopharyngeal obstruction and as part of the treatment of chronic secretory otitis media. A variety of methods are employed including suction diathermy, suction microdebridement and most commonly blind curettage. When using the latter it is important to avoid hyperextension of the cervical spine as this throws the arch of the atlas into prominence and may result in damage to the prevertebral fascia and anterior spinal ligaments, with resultant infection and cervical instability. Extreme lateral curettage can result in damage to the tubal orifice and excessive bleeding because the vasculature is denser laterally. Pharyngotympanic tube (Fig. 38.4)
The pharyngotympanic tube connects the tympanic cavity to the nasopharynx and allows the passage of air between these spaces in order to equalize the air pressure on both aspects of the tympanic membrane. It is c.36 mm long and descends anteromedially from the tympanic cavity to the nasopharynx at an angle of c.45° with the sagittal plane and 30° with the horizontal (these angles increase with age and elongation of the skull base). It is formed partly by cartilage and fibrous tissue and partly by bone. The cartilaginous part, c.24 mm long, is formed by a triangular plate of cartilage, the greater part being in the posteromedial wall of the tube.Its apex is attached by fibrous tissue to the circumference of the jagged rim of the bony part of the tube and its base is directly under the mucosa of the lateral nasopharyngeal wall, forming a tubal elevation behind the pharyngeal orifice of the tube (Fig. 35.1). The upper part of the cartilage is bent laterally and downwards, producing a broad
medial lamina and narrow lateral lamina. In transverse section it is hook-like and incomplete below and laterally, where the canal is composed of fibrous tissue. The cartilage is fixed to the cranial base in the groove between the petrous part of the temporal bone and the greater wing of the sphenoid, and ends near the root of the medial pterygoid plate. The cartilaginous and bony parts of the tube are not in the same plane, the former descending a little more steeply than the latter. The diameter of the tube is greatest at the pharyngeal orifice, least at the junction of the two parts (the isthmus) and increases again towards the tympanic cavity. The bony part, c.12 mm long, is oblong in transverse section, with its greater dimension in the horizontal plane. It starts from the anterior tympanic wall and gradually narrows to end at the junction of the squamous and petrous parts of the temporal bone, where it has a jagged margin for the attachment of the cartilaginous part. The carotid canal lies medially. page 621 page 622
Figure 35.3 Median sagittal section of the head, showing a dissection of the interior of the pharynx, after the removal of the mucous membrane. The bodies of the cervical vertebrae have been removed and the cut posterior wall of the pharynx then retracted dorsolaterally. Palatopharyngeus is reflected dorsally to show the cranial fibres of the inferior constrictor; the dorsum of the tongue is pulled ventrally to display a part of styloglossus in the angular interval between the mandibular and the lingual fibres of origin of the superior constrictor.
Figure 35.4 Appearance of a nasopharyngeal tonsil following adenoidectomy by curettage. Rostral surface is to the left; surface folds radiate forward from a median recess (arrowhead). In this example, the impression left by contact with the left Eustachian cushion is evident laterally (arrow). B, Transnasal endoscopic view of adenoid. 1, adenoid (in posterior naris); 2, inferior concha (posterior view); 3, posterior end of nasal septum. (A, specimen provided by Professor MJ Gleeson, ENT Department, GKT School of Medicine, London.) (B, by kind permission from Mr Simon A Hickey.)
The mucosa of the pharyngotympanic tube is continuous with the nasopharyngeal and tympanic mucosae. It is lined by a ciliated columnar epithelium and is thin in the osseous part but thickened by mucous glands in the cartilaginous part. Near the pharyngeal orifice there is a variable, but sometimes considerable, lymphoid mass, the tubal tonsil. UPDATE Abstract: Imaging of the patulous Eustachian tube.
Date Added: 08 July 2005
Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=15513527&query_hl=9 Imaging of the patulous Eustachian tube. At birth the pharyngotympanic tube is about half its adult length, it is more horizontal and its bony part is relatively shorter but much wider. The pharyngeal orifice is a narrow slit, level with the palate and without a tubal elevation. Relations Salpingopharyngeus is attached to the inferior part of the cartilaginous tube near its pharyngeal opening. Posteromedially are the petrous part of the temporal bone and levator veli palatini, which arises partly from the medial lamina of the tube. Anterolaterally tensor veli palatini separates the tube from the otic ganglion, the mandibular nerve and its branches, the chorda tympani nerve and the middle meningeal artery. Some fibres of tensor veli palatini are attached to the lateral
lamina of the cartilage and to the fibrous part, and these fibres are sometimes referred to as dilator tubae. The pharyngotympanic tube is opened during deglutition but the mechanism is uncertain. Dilator tubae, aided by salpingopharyngeus, may be responsible. Levator veli palatini elevates the cartilaginous part of the pharyngotympanic tube, and so might allow passive opening by releasing tension on the cartilage. Vascular supply Arteries to the pharyngotympanic tube arise from the ascending pharyngeal and middle meningeal arteries and from the artery of the pterygoid canal. The veins of the pharyngotympanic tube usually drain to the pterygoid venous plexus.
© 2008 Elsevier
OROPHARYNX BOUNDARIES (Figs 35.1, 35.2) page 622 page 623
Figure 35.5 Transverse section of a nasopharyngeal tonsil. Numerous lymphoid follicles (F) are covered on their nasopharyngeal surface by respiratory epithelium (E) with folds and deep crypts (C). Seromucous glands (G) and connective tissue septa (S) penetrate the lymphoid tissue. (Provided by N Kirkpatrick, Division of Anatomy and Cell Biology, GKT School of Medicine, London; photograph by Sarah-Jane Smith.)
The oropharynx extends from the soft palate to the upper border of the epiglottis (Fig. 35.1). It opens into the mouth through the oropharyngeal isthmus, demarcated by the palatoglossal arch, and faces the pharyngeal aspect of the tongue. Its lateral wall consists of the palatopharyngeal arch and palatine tonsil (Fig. 33.2). Posteriorly, it is level with the body of the second, and upper part of the third, cervical vertebrae (Fig. 35.1).
SOFT PALATE The soft palate is a mobile flap suspended from the posterior border of the hard palate, sloping down and back between the oral and nasal parts of the pharynx. (Fig. 35.1). The boundary between the hard and soft palate is readily palpable
and may be distinguished by a change in colour, the soft palate being a darker red with a yellowish tint. The soft palate is a thick fold of mucosa enclosing an aponeurosis, muscular tissue, vessels, nerves, lymphoid tissue and mucous glands. In most individuals two small pits, the fovea palatini, one on each side of the midline, may be seen: they represent the orifices of ducts from some of the minor mucous glands of the palate. In its usual relaxed and pendant position, the anterior (oral) surface of the soft palate is concave, and has a median raphe. The posterior aspect is convex and continuous with the nasal floor, the anterosuperior border is attached to the posterior margin of the hard palate, and the sides blend with the pharyngeal wall. The inferior border is free, and hangs between the mouth and pharynx. A median conical process, the uvula, projects downwards from its posterior border (Fig. 33.2). Taste buds also occur on the oral aspect of the soft palate. The anterior third of the soft palate contains little muscle and consists mainly of the palatine aponeurosis. This region is less mobile and more horizontal than the rest of the soft palate and is the chief area acted upon by tensor veli palatini. A small bony prominence, produced by the pterygoid hamulus, can be felt just behind and medial to each upper alveolar process, in the lateral part of the anterior region of the soft palate. The pterygomandibular raphe - a tendinous band between buccinator and the superior constrictor - passes downwards and outwards from the hamulus to the posterior end of the mylohyoid line. When the mouth is opened wide, this raphe raises a fold of mucosa that marks internally the posterior boundary of the cheek, and is an important landmark for an inferior alveolar nerve block. Palatine aponeurosis
A thin, fibrous, palatine aponeurosis strengthens the soft palate, and is composed of the expanded tendons of the tensor veli palatini muscles. It is attached to the posterior border and inferior surface of the hard palate behind any palatine crests, and extends medially from behind the greater palatine foramina. It is thick in the anterior two-thirds of the soft palate but very thin further back. Near the midline it encloses the musculus uvulae. All the other palatine muscles are attached to the aponeurosis. Pillars of fauces (Fig. 33.2)
The lateral wall of the oropharynx presents two prominent folds, the pillars of the fauces. The anterior fold, or palatoglossal arch, runs from the soft palate to the side of the tongue and contains palatoglossus. The posterior fold, or palatopharyngeal arch, projects more medially and passes from the soft palate to merge with the lateral wall of the pharynx. It contains palatopharyngeus. A triangular tonsillar fossa (tonsillar sinus) lies on each side of the oropharynx between the diverging palatopharyngeal and palatoglossal arches, and contains the palatine tonsil. Vascular supply
The arterial supply of the soft palate is usually derived from the ascending palatine branch of the facial artery. Sometimes this is replaced or supplemented
by a branch of the ascending pharyngeal artery which descends forwards between the superior border of the superior constrictor and levator veli palatini, and accompanies the latter to the soft palate. The veins of the soft palate usually drain to the pterygoid venous plexus. Innervation
The secretomotor supply to most of the mucosa of the soft palate travels via the lesser palatine nerve in postganglionic branches from the pterygopalatine ganglion. The lesser palatine nerve also contains sensory fibres including those supplying taste buds in the oral surface of the soft palate, which travel through the pterygopalatine ganglion without synapsing to access the greater petrosal nerve (a branch of the facial nerve). Postganglionic secretomotor parasympathetic fibres may pass to the posterior parts of the soft palate from the otic ganglion (which receives preganglionic fibres via the lesser petrosal branch of the glossopharyngeal nerve). Gag reflex
The gag reflex is discussed on page 233. Uvulopalatopharyngoplasty
The pharyngeal airway is kept patent in the patient who is awake by the combined dilating action of genioglossus, tensor veli palatini, geniohyoid and stylohyoid, which act to counter the negative pressure generated in the lumen of the pharynx during inspiration. The tone in the muscles is reduced during sleep, but is also affected by alcohol and other sedatives, hypothyroidism and a variety of neurological disorders. If the dilator muscle tone is insufficient, the walls of the pharynx may become apposed. Intermittent pharyngeal obstruction may cause snoring, and complete obstruction may cause apnoea, hypoxia and hypercarbia which lead to arousal and sleep disturbance. Surgical techniques involving reduction in the length of the soft palate, removal of the tonsils and plicating of the tonsillar pillars can be used to raise the intrinsic dilating tone in the pharyngeal wall and to reduce the bulk of (and to stiffen) the soft palate. This will reduce the tendency of the soft palate to vibrate and generate noise during periods of incipient collapse of the pharynx. An alternative treatment is to deliver air to the pharynx at above atmospheric pressure via a closely fitting facemask, thus inflating the pharynx and countering its tendency to collapse.
PALATINE TONSIL (Figs 35.6, 35.7) page 623 page 624
Figure 35.6 Coronal section through the left palatine tonsil.
The right and left palatine tonsils form part of the circumpharyngeal lymphoid ring. Each tonsil is an ovoid mass of lymphoid tissue situated in the lateral wall of the oropharynx. Size varies according to age, individuality and pathological status (tonsils may be hypertrophied and/or inflamed). It is therefore difficult to define the normal appearance of the palatine tonsil. For the first 5 or 6 years of life the tonsils increase rapidly in size. They usually reach a maximum at puberty when they average 20-25 mm in vertical, and 10-15 mm in transverse, diameters, and they project conspicuously into the oropharynx. Tonsillar involution begins at puberty, when the reactive lymphoid tissue begins to atrophy, and by old age only a little tonsillar lymphoid tissue remains. The long axis of the tonsil is directed from above, downwards and backwards. Its medial, free, surface usually presents a pitted appearance. The pits, 10-20 in number, lead into a system of blind-ending, often highly branching, crypts which extend through the whole thickness of the tonsil and almost reach the connective tissue hemicapsule. In a healthy tonsil the openings of the crypts are fissure-like and the walls of the crypt lumina are collapsed so that they are in contact with each other. The human tonsil is polycryptic. The branching crypt system reaches its maximum size and complexity during childhood. The mouth of a deep intratonsillar cleft (recessus palatinus), opens in the upper part of the medial surface of the tonsil (Fig. 35.6). It often erroneously called the supratonsillar fossa, and yet it is not situated above the tonsil but within its substance. The mouth of the cleft is semilunar, curving parallel to the convex dorsum of the tongue in the parasagittal plane. The upper wall of the recess contains lymphoid tissue which extends into the soft palate as the pars palatina of the palatine tonsil. After the age of 5 years this embedded part of the tonsil diminishes in size. From
the age of 14, there is a tendency for the whole tonsil to involute, and for the tonsillar bed to flatten out. During young adult life a mucosal fold termed the plica triangularis stretches back from the palatoglossal arch down to the tongue. It is infiltrated by lymphoid tissue and frequently represents the most prominent (anteroinferior) portion of the tonsil. It rarely persists into middle age. The lateral or deep surface of the tonsil spreads downwards, upwards and forwards. Inferiorly, it invades the dorsum of the tongue, superiorly, the soft palate, and, anteriorly, it may extend for some distance under the palatoglossal arch. This deep, lateral aspect is covered by a layer of fibrous tissue, the tonsillar hemicapsule, separable with ease for most of its extent from the underlying muscular walls of the pharynx which is formed here by the superior constrictor, with styloglossus on its lateral side (Fig. 35.6). Anteroinferiorly the hemicapsule adheres to the side of the tongue and to palatoglossus and palatopharyngeus. In this region the tonsillar artery, a branch of the facial, pierces the superior constrictor to enter the tonsil, accompanied by venae comitantes. An important and sometimes large vein, the external palatine or paratonsillar vein, descends from the soft palate lateral to the tonsillar hemicapsule before piercing the pharyngeal wall (Fig. 35.6). Haemorrhage from this vessel from the upper angle of the tonsillar fossa may complicate tonsillectomy. The muscular wall of the tonsillar fossa separates the tonsil from the ascending palatine artery, and, occasionally, from the tortuous facial artery itself, which may lie near the pharyngeal wall at the lower tonsillar level. The internal carotid artery lies c.25 mm behind and lateral to the tonsil. Microstructure of the palatine tonsil
page 624 page 625
Figure 35.7 A, Transverse section through a whole palatine tonsil, showing many secondary follicles, oropharyngeal surface epithelium and the connective tissue hemicapsule. B, Reticulated epithelium from a crypt of a palatine tonsil, immunostained to show numerous interdigitating cells and macrophages. Note the close contacts between these cells and infiltrating lymphocytes (arrows). (A, by permission from Young B, Heath JW 2000 Wheater's Functional Histology. Edinburgh: Churchill Livingstone.) (B, provided by M Perry, Division of Anatomy and Cell Biology, GKT School of Medicine, London.)
Each tonsil is a mass of lymphoid tissue associated with the oropharyngeal mucosa and fixed in its position, unlike most other examples of mucosaassociated lymphoid tissue (p. 77). It is covered on its oropharyngeal aspect by non-keratinized stratified squamous epithelium. The whole of the tonsil is supported internally by a delicate meshwork of fine collagen type III (reticulin) fibres which are condensed in places to form more robust connective tissue septa that also contain elastin. These septa partition the tonsillar parenchyma, and merge at their ends with the dense irregular fibrous hemicapsule on the deep aspect of the tonsil and with the lamina propria on the pharyngeal surface. Blood vessels, lymphatics and nerves branch or join within the connective tissue condensations. The hemicapsule forms its lateral boundary with the oropharyngeal wall, and with the mucosa which covers its highly invaginated free surface (Fig. 35.7). The 10-20 crypts formed by invagination of the free surface mucosa are narrow tubular epithelial diverticula which often branch within the tonsil and frequently are packed with plugs of shed epithelial cells, lymphocytes and bacteria, which may calcify. The epithelium lining the crypts is mostly similar to that of the oropharyngeal surface, i.e. stratified squamous, but there are also patches of reticulated epithelium, which is much thinner. This has a complex structure which is of great importance in the immunological function of the tonsil. Reticulated epithelium Reticulated epithelium lacks the orderly laminar structure of stratified squamous epithelium. Its base is deeply invaginated in a complex manner so that the epithelial cells, with their slender branched cytoplasmic processes, provide a coarse mesh to accommodate the infiltrating lymphocytes and macrophages. The basal lamina of this epithelium is discontinuous. Although the oropharyngeal surface is unbroken, the epithelium may become exceedingly thin in places, so that only a tenuous cytoplasmic layer separates the pharyngeal lumen from the underlying lymphocytes. Epithelial cells are held together by small desmosomes, anchored into bundles of keratin filaments. Interdigitating dendritic cells (antigenpresenting cells, APCs) (p. 80) are also present. The intimate association of epithelial cells and lymphocytes facilitates the direct transport of antigen from the external environment to the tonsillar lymphoid cells, i.e. reticulated epithelial cells are functionally similar to the microfold (M) cells of the gut. The total surface area of the reticulated epithelium is very large because of the complex branched nature of the tonsillar crypts, and has been estimated at 295 cm2 for an average palatine tonsil. Lymph nodes elsewhere depend on indirect antigen delivery through afferent lymphatic vessels (p. 75), but these are absent from the tonsil (although it is drained by efferent lymphatics). Tonsillar lymphoid tissue There are four lymphoid compartments in the palatine tonsils. Lymphoid follicles, many with germinal centres (p. 74), are arranged in rows roughly parallel to neighbouring connective tissue septa. Their size and cellular content varies in proportion to the immunological activity of the tonsil. The mantle zones of the follicles, each with closely packed small lymphocytes, form a dense cap, always situated on the side of the follicle nearest to the mucosal surface. These cells are
the products of B-lymphocyte proliferation within the germinal centres. Extrafollicular, or T-lymphocyte areas contain a specialized microvasculature including high endothelial venules (HEVs) (p. 143), through which circulating lymphocytes enter the tonsillar parenchyma. The lymphoid tissue of the reticulated crypt epithelium contains predominantly IgG- and IgA-producing B lymphocytes (including some mature plasma cells), T lymphocytes and antigenpresenting cells, APCs (p. 80). There are numerous capillary loops in this subsurface region. Vascular supply and lymphatic drainage
The arterial blood supply to the palatine tonsil is derived from branches of the external carotid artery. The principal artery is the tonsillar artery, which is a branch of the facial, or sometimes the ascending palatine, artery. It ascends between medial pterygoid and styloglossus, perforates the superior constrictor at the upper border of styloglossus (Fig. 35.6) and ramifies in the tonsil and posterior lingual musculature. Additional small tonsillar branches may be derived from the ascending pharyngeal artery; the dorsal lingual branches of the lingual artery (supplying the lower part of the palatine tonsil); the greater palatine branch of the maxillary artery (supplying the upper part of the tonsil), and the ascending palatine branch of the facial artery. Arteries enter the deep surface of the tonsil, branch within the connective tissue septa, narrow to become arterioles and then give off capillary loops into the follicles, interfollicular areas and the cavities within the base of the reticulated epithelium. The capillaries rejoin to form venules, many with high endothelium, and the veins return within the septal tissues to the hemicapsule as tributaries of the pharyngeal drainage. The tonsillar artery and its venae comitantes often lie within the palatoglossal fold, and may haemorrhage if this fold is damaged during surgery. Unlike lymph nodes, the tonsils do not possess afferent lymphatics or lymph sinuses. Instead, dense plexuses of fine lymphatic vessels surround each follicle and form efferent lymphatics which pass towards the hemicapsule, pierce the superior constrictor and drain to the upper deep cervical lymph nodes directly (especially the jugulodigastric nodes) or indirectly through the retropharyngeal lymph nodes. The former are typically enlarged in tonsillitis, when they project beyond the anterior border of sternocleidomastoid and are palpable superficially 1-2 cm below the angle of the mandible: when enlarged, they represent the most common swelling in the neck. Innervation
The palatine tonsil region receives its nerve supply through tonsillar branches of the maxillary nerve and the glossopharyngeal nerve. The maxillary nerve fibres pass through, but do not synapse in, the pterygopalatine ganglion, and are distributed through the lesser palatine nerves. The latter, together with the tonsillar branches of the glossopharyngeal nerve, form a plexus around the tonsil. From this plexus (the circulus tonsillaris), nerve fibres are also distributed to the soft palate and the region of the oropharyngeal isthmus. The tympanic branch of the glossopharyngeal nerve supplies the mucous membrane lining the tympanic cavity. Infection, malignancy and postoperative inflammation of the tonsil and tonsil fossa may therefore be accompanied by pain referred to the ear.
Tonsillectomy
Surgical removal of the pharyngeal tonsils is commonly performed to prevent recurrent acute tonsillitis or to treat airway obstruction by hypertrophied or inflamed palatine tonsils. Occasionally the tonsil may be removed to treat an acute peritonsillar abscess, which is a collection of pus between the superior constrictor and the tonsillar hemicapsule. Many methods have been employed, the commonest being dissection in the plane of the fibrous hemicapsule followed by ligation or electrocautery to the vessels divided during the dissection. The nerve supply to the tonsil is so diffuse that tonsillectomy under local anaesthesia is performed successfully by local infiltration rather than by blocking the main nerves. Surgical access to the glossopharyngeal nerve may be achieved by separating the fibres of superior constrictor. Waldeyer's ring
Waldeyer's ring is a circumpharyngeal ring of mucosa-associated lymphoid tissue which surrounds the openings into the digestive and respiratory tracts. It is made up anteroinferiorly by the lingual tonsil, laterally by the palatine and tubal tonsils, and posterosuperiorly by the nasopharyngeal tonsil and smaller collections of lymphoid tissue in the inter-tonsillar intervals.
© 2008 Elsevier
LARYNGOPHARYNX BOUNDARIES (Figs 35.1, 35.2) The laryngopharynx (known clinically as the hypopharynx) extends from the superior border of the epiglottis, where it is delineated from the oropharynx by the lateral glossoepiglottic folds, to the inferior border of the cricoid cartilage, where it becomes continuous with the oesophagus. The laryngeal inlet lies in its incomplete anterior wall, and the posterior surfaces of the arytenoid and cricoid cartilages lie below this opening. Piriform fossa
A small piriform fossa lies on each side of the laryngeal inlet, bounded medially by the aryepiglottic fold and laterally by the thyroid cartilage and thyrohyoid membrane. Branches of the internal laryngeal nerve lie beneath its mucous membrane. At rest, the laryngopharynx extends posteriorly from the lower part of the third cervical vertebral body to the upper part of the sixth. During deglutition it may be elevated considerably by the hyoid elevators. Inlet of larynx (Fig. 35.2)
The obliquely-sloping inlet of the larynx lies in the anterior part of the laryngopharynx. This inlet is bounded above by the epiglottis, below by the arytenoid cartilages of the larynx, and laterally by the aryepiglottic folds. Below the inlet, the anterior wall of the laryngopharynx is formed by the posterior surface of the cricoid cartilage.
© 2008 Elsevier
PHARYNGEAL FASCIA The two named layers of fascia in the pharynx are the pharyngobasilar and the buccopharyngeal fascia. The fibrous layer that supports the pharyngeal mucosa is thickened above the superior constrictor to form the pharyngobasilar fascia (Fig. 35.8). It is attached to the basilar part of the occipital bone and the petrous part of the temporal bone medial to the pharyngotympanic tube, and to the posterior border of the medial pterygoid plate and pterygomandibular raphe. Inferiorly, it diminishes in thickness, but is strengthened posteriorly by a fibrous band attached to the pharyngeal tubercle of the occipital bone which descends as the median pharyngeal raphe of the constrictors. This fibrous layer is really the epimysial covering of the muscles and their aponeurotic attachment to the base of the skull. The thinner external part of the epimysium is the buccopharyngeal fascia, which covers the superior constrictor and passes forwards over the pterygomandibular raphe to cover buccinator.
© 2008 Elsevier
PHARYNGEAL TISSUE SPACES
Figure 35.8 Muscles of the pharynx: posterior view. (From Schaefer EA, Symington J, Bryce TH (eds) 1915 Quain's Anatomy, 11th edn. London: Longmans, Green, with permission from Pearson Education.)
Pharyngeal tissue spaces can be subdivided into peripharyngeal and intrapharyngeal spaces. The anterior part of the peripharyngeal space is formed by the submandibular and submental spaces, posteriorly by the retropharyngeal space and laterally by the parapharyngeal spaces. The retropharyngeal space is an area of loose connective tissue which lies behind the pharynx and anterior to the prevertebral fascia, and extends upwards to the base of the skull and downwards to the retrovisceral space in the infrahyoid part of the neck. Each parapharyngeal space passes laterally around the pharynx and is continuous with
the retropharyngeal space. However, unlike the retropharyngeal space, it is a space which is restricted to the suprahyoid region. It is bounded medially by the pharynx; laterally by the pterygoid muscles - here it is part of the infratemporal fossa - and the sheath of the parotid gland; superiorly by the base of the skull; and is limited inferiorly by suprahyoid structures, particularly the sheath of the submandibular gland. An intrapharyngeal space potentially exits between the inner surface of the constrictor muscles and the pharyngeal mucosa. Infections in this space are either restricted locally or spread through the pharynx into the retropharyngeal or parapharyngeal spaces. The peritonsillar space is an important part of the intrapharyngeal space: it lies around the palatine tonsil between the pillars of the fauces. Infections in the intratonsillar space usually spread up or down the intrapharyngeal space, or through the pharynx into the parapharyngeal space. UPDATE Date Added: 30 May 2006 Abstract: Anatomy of the roof of the parapharyngeal space Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=15112970&query_hl=9&itool=pubmed_docsum Roof of the parapharyngeal space: Defining its boundaries and clinical implications. Maheshwar AA, Kim EY, Pensak ML, Keller JT: Ann Otol Rhinol Laryngol 113:283-288, 2004. page 626 page 627
SPREAD OF INFECTION Infection which spreads into the parapharyngeal space will produce pain, and trismus. There may be swelling in the oropharynx extending up to the uvula, displacing it to the contralateral side, and dysphagia. Posterior spread from the parapharyngeal space into the retropharyngeal space will produce bulging of the posterior pharyngeal wall, dyspnoea and nuchal rigidity. Involvement of the carotid sheath may produce symptoms due to thrombosis of the internal jugular vein and cranial nerve symptoms involving the IX-XII nerves. If the infection continues to spread unchecked, mediastinitis will ensue. A virulent infection in the retropharyngeal space may spread through the prevertebral fascia into the underlying prevertebral space. Infection in this tissue space may descend into the thorax and even below the diaphragm and results in chest pain, severe dyspnoea and retrosternal discomfort. Pharyngeal infection from mucosally associated lymph tissues such as the palatine tonsil, or as a result of a penetrating injury (e.g. from an ingested foreign body), may result in the spread of infection into the tissue spaces of the neck adjacent to the pharynx. This is an extremely dangerous situation since there is potential for rapid spread throughout the neck and, more dangerously, to the superior mediastinum to cause overwhelming life-threatening infection.
PARAPHARYNGEAL SPACE TUMOURS Tumours may develop a priori in the parapharyngeal tissue space and remain asymptomatic for some time. When they do present it may be with a diffuse pattern of symptoms, which are often the result of effects of compression on the lower cranial nerves, e.g. dysarthria, resulting from impairment of tongue movements secondary to hypoglossal nerve damage; dysphagia, with overspill and aspiration of ingested material into the airway, resulting from loss of sensory information from the distribution of the pharyngeal plexus nerves, the vagus and cranial accessory; motor dysfunction of the pharynx and larynx resulting from loss of motor innervation via the pharyngeal plexus, and the recurrent laryngeal branch of the vagus to the intrinsic muscles of the larynx.
© 2008 Elsevier
MUSCLES OF SOFT PALATE AND PHARYNX The muscles of the palate and pharynx are levator veli palatini, tensor veli palatini, palatoglossus, palatopharyngeus, musculus uvulae, the superior, middle, and inferior constrictors, salpingopharyngeus and stylopharyngeus. With the exception of tensor veli palatini, which is supplied by the motor branch of the mandibular division of the trigeminal through the nerve to medial pterygoid, and stylopharyngeus, which is supplied by the glossopharyngeal nerve, the muscles are supplied by the cranial part of the accessory nerve via the pharyngeal plexus.
LEVATOR VELI PALATINI (Fig. 35.3, 35.8, 35.9, 35.10) Levator veli palatini arises by a small tendon from a rough area on the inferior surface of the petrous part of the temporal bone, in front of the lower opening of the carotid canal. Additional fibres arise from the inferior aspect of the cartilaginous part of the pharyngotympanic tube and from the vaginal process of the tympanic bone. At its origin the muscle is inferior rather than medial to the pharyngotympanic tube and only crosses medial to it at the level of the medial pterygoid plate. It passes medial to the upper margin of the superior constrictor and in front of salpingopharyngeus. Its fibres spread in the medial third of the soft palate between the two strands of palatopharyngeus to attach to the upper surface of the palatine aponeurosis as far as the midline, where they interlace with those of the contralateral muscle. Thus the two levator muscles form a sling above and just behind the palatine aponeurosis. Vascular supply The blood supply of levator veli palatini is derived from the ascending palatine branch of the facial artery and the greater palatine branch of the maxillary artery. Innervation Levator veli palatini is innervated from the cranial part of the accessory nerve via the pharyngeal plexus. Actions
Figure 35.9 Muscles of the left half of the soft palate and adjoining part of the pharyngeal wall in sagittal section. Part of levator veli palatini is cut sagittally. (Based on a dissection by the late James Whillis, Department of Anatomy, GKT School of Medicine, London.)
The primary role of the levator veli palatini muscles is to elevate the almost vertical posterior part of the soft palate and pull it slightly backwards. During swallowing, the soft palate is at the same time made rigid by the contraction of the tensor veli palatini muscles and touches the posterior wall of the pharynx, thus separating the nasopharynx from the oropharynx. By additionally pulling on the lateral walls of the nasopharynx posteriorly and medially, the muscles also narrow that space. Levator veli palatini has little or no effect on the pharyngotympanic tube, although it might allow passive opening.
TENSOR VELI PALATINI (Figs 33.6, 35.8, 35.10) Tensor veli palatini arises from the scaphoid fossa of the pterygoid process and posteriorly from the medial aspect of the spine of the sphenoid bone. Between these two sites it is attached to the anterolateral membranous wall of the pharyngotympanic tube (including its narrow isthmus where the cartilaginous medial two-thirds meets the bony lateral one-third). Some fibres may be continuous with those of tensor tympani. Inferiorly, the fibres converge on a
delicate tendon that turns medially around the pterygoid hamulus to pass through the attachment of buccinator to the palatine aponeurosis and the osseous surface behind the palatine crest on the horizontal plate of the palatine bone. There is a small bursa between the tendon and the pterygoid hamulus. The muscle is thin and triangular and lies lateral to the medial pterygoid plate, pharyngotympanic tube and levator veli palatini. Its lateral surface contacts the upper and anterior part of medial pterygoid, the mandibular, auriculotemporal and chorda tympani nerves, the otic ganglion and the middle meningeal artery. Vascular supply The blood supply of tensor veli palatini is derived from the ascending palatine branch of the facial artery and the greater palatine branch of the maxillary artery. Innervation The motor innervation of tensor veli palatini is derived from the mandibular nerve via the nerve to medial pterygoid, and reflects the development of the muscle from the first branchial arch. Actions
page 627 page 628
Figure 35.10 Muscles of the pharynx.
Acting together the tensor veli palatini muscles tauten the soft palate, principally its anterior part, and depress it by flattening its arch. Acting unilaterally, the muscle pulls the soft palate to one side. Although contraction of both muscles will slightly depress the anterior part of the soft palate, it is often assumed that the increased rigidity aids palatopharyngeal closure. However, it is now believed that a primary role of the tensor is to open the pharyngotympanic tube, for example during deglutition and yawning. In this way the muscle equalizes air pressure between the middle ear and nasopharynx.
MUSCULUS UVULAE (Fig. 35.2) Musculus uvulae arises from the posterior nasal spine of the palatine bone and the superior surface of the palatine aponeurosis, and lies between the two laminae of the aponeurosis. It runs posteriorly above the sling formed by levator veli palatini and inserts beneath the mucosa of the uvula. The two sides of the muscle are united along most of its length. Vascular supply The blood supply of musculus uvulae is derived from the ascending palatine branch of the facial artery and the descending palatine branch of the maxillary artery. Innervation The nerve supply to musculus uvulae is derived from the cranial part of the accessory nerve via the pharyngeal plexus. Actions By retracting the uvular mass and thickening the middle third of the soft palate, musculus uvulae aids levator veli palatini in palatopharyngeal closure. The two muscles run at right angles to each other and their contraction raises a 'levator eminence' which helps seals off the nasopharynx.
PALATOGLOSSUS (Fig. 35.3) Palatoglossus is narrower at its middle than at its ends. Together with its overlying mucosa it forms the palatoglossal arch or fold (see Fig. 33.2). It arises from the oral surface of the palatine aponeurosis where it is continuous with its fellow. It extends forwards, downwards and laterally in front of the palatine tonsil to the side of the tongue. Some of its fibres spread over the dorsum of the tongue, others pass deeply into its substance to intermingle with fibres of the intrinsic transverse muscle. Vascular supply Palatoglossus receives its blood supply from the ascending palatine branch of the facial artery and from the ascending pharyngeal artery. Innervation Palatoglossus is supplied by the cranial part of the accessory nerve via the pharyngeal plexus, and is therefore unlike all the other muscles of the tongue, which are supplied by the hypoglossal nerve. Actions Palatoglossus elevates the root of the tongue and approximates the palatoglossal arch to its contralateral fellow, thus shutting off the oral cavity from the oropharynx.
PALATOPHARYNGEUS (Figs 35.3, 35.9) Palatopharyngeus and its overlying mucosa form the palatopharyngeal arch (see Fig. 33.2). Within the soft palate palatopharyngeus is composed of two fasciculi which are attached to the upper surface of the palatine aponeurosis; they lie in
the same plane but are separated from each other by levator veli palatini. The thicker anterior fasciculus arises from the posterior border of the hard palate as well as the palatine aponeurosis, where some fibres interdigitate across the midline. The posterior fasciculus is in contact with the mucosa of the pharyngeal aspect of the palate, and joins the posterior band of the contralateral muscle in the midline. The two layers unite at the posterolateral border of the soft palate, and are joined by fibres of salpingopharyngeus. Passing laterally and downwards behind the tonsil, palatopharyngeus descends posteromedial to and in close contact with stylopharyngeus, to be attached with it to the posterior border of the thyroid cartilage. Some fibres end on the side of the pharynx, attached to pharyngeal fibrous tissue and others cross the midline posteriorly, decussating with those of the contralateral muscle. Palatopharyngeus thus forms an incomplete internal longitudinal muscular layer in the wall of the pharynx. Passavant's muscle (palatopharyngeal sphincter) The existence of Passavant's muscle remains controversial. It has been described as a part of the superior constrictor and palatopharyngeus muscles. An alternative view holds that it is a distinct palatine muscle that arises from the anterior and lateral parts of the upper surface of the palatine aponeurosis, lies lateral to levator veli palatini, blends internally with the upper border of the superior constrictor and encircles the pharynx as a sphincter-like muscle (Fig. 35.9). Whatever its origin, when it contracts, it forms a ridge (Passavant's ridge) when the soft palate is elevated. The change from columnar, ciliated, 'respiratory' epithelium to stratified, squamous epithelium that takes place on the superior aspect of the soft palate occurs along the line of attachment of the palatopharyngeal sphincter to the palate. The muscle is hypertrophied in cases of complete cleft palate. Vascular supply Palatopharyngeus receives its arterial supply from the ascending palatine branch of the facial artery, the greater palatine branch of the maxillary artery and the pharyngeal branch of the ascending pharyngeal artery. Innervation Palatopharyngeus is innervated by the cranial part of the accessory nerve via the pharyngeal plexus. Actions Acting together, the palatopharyngei pull the pharynx up, forwards and medially, and thus shorten it during swallowing. They also approximate the palatopharyngeal arches and draw them forwards.
SUPERIOR CONSTRICTOR (Figs 33.6, 35.6, 35.8, 35.9, 35.10) page 628 page 629
The superior constrictor is a quadrilateral sheet of muscle and is thinner than the other two constrictors. It is attached anteriorly to the pterygoid hamulus (and sometimes to the adjoining posterior margin of the medial pterygoid plate), the posterior border of the pterygomandibular raphe, the posterior end of the mylohyoid line of the mandible, and, by a few fibres, to the side of the tongue.
The fibres curve back into a median pharyngeal raphe which is attached superiorly to the pharyngeal tubercle on the basilar part of the occipital bone Relations The upper border of the superior constrictor is separated from the cranial base by a crescentic interval which contains levator veli palatini, the pharyngotympanic tube and an upward projection of pharyngobasilar fascia. The lower border is separated from the middle constrictor by stylopharyngeus and the glossopharyngeal nerve (Fig. 35.10). Anteriorly the pterygomandibular raphe separates the superior constrictor from buccinator, and posteriorly the superior constrictor lies on the prevertebral muscles and fascia, from which it is separated by the retropharyngeal space. The ascending pharyngeal artery, pharyngeal venous plexus, glossopharyngeal and lingual nerves, styloglossus, middle constrictor, medial pterygoid, stylopharyngeus, and the stylohyoid ligament all lie laterally, and palatopharyngeus, the tonsillar capsule and the pharyngobasilar fascia lie internally. Vascular supply The arterial supply of the superior constrictor is derived mainly from the pharyngeal branch of the ascending pharyngeal artery and the tonsillar branch of the facial artery. Innervation The superior constrictor is innervated by the cranial part of the accessory nerve from the pharyngeal plexus. Actions The superior constrictor constricts the upper part of the pharynx.
MIDDLE CONSTRICTOR (Figs 33.6, 35.3, 35.8, 35.10) The middle constrictor is a fan-shaped sheet attached anteriorly to the lesser cornu of the hyoid and the lower part of the stylohyoid ligament (the chondropharyngeal part of the muscle), and to the whole of the upper border of the greater cornu of the hyoid (the ceratopharyngeal part). The lower fibres descend deep to the inferior constrictor to reach the lower end of the pharynx, the middle fibres pass transversely and the superior fibres ascend and overlap the superior constrictor. All fibres insert posteriorly into the median pharyngeal raphe. Relations The glossopharyngeal nerve and stylopharyngeus pass through a small gap between the middle and superior constrictors, and the internal laryngeal nerve and the laryngeal branch of the superior thyroid artery pass between the middle and inferior constrictors. The prevertebral fascia and longus colli and longus capitis are posterior, the superior constrictor, stylopharyngeus and palatopharyngeus are internal, and the carotid vessels, pharyngeal plexus of nerves and some lymph nodes are lateral. Near its hyoid attachment, the middle constrictor lies deep to hyoglossus, from which it is separated by the lingual artery. Vascular supply
The arterial supply of the middle constrictor is derived mainly from the pharyngeal branch of the ascending pharyngeal artery and the tonsillar branch of the facial artery. Innervation The middle constrictor is innervated by the cranial part of the accessory nerve from the pharyngeal plexus. Actions The middle constrictor constricts the middle part of the pharynx during swallowing.
INFERIOR CONSTRICTOR (Figs 33.6, 35.3, 35.8, 35.10) The inferior constrictor is the thickest of the three constrictor muscles, and is usually described in two parts, cricopharyngeus and thyropharyngeus. Cricopharyngeus arises from the side of the cricoid cartilage between the attachment of cricothyroid and the articular facet for the inferior thyroid cornu. Thyropharyngeus arises from the oblique line of the thyroid lamina, a strip of the lamina behind this, and by a small slip from the inferior cornu. Some additional fibres arise from a tendinous cord that loops over cricothyroid. Both cricopharyngeus and thyropharyngeus spread posteromedially to join the contralateral muscle. Thyropharyngeus is inserted into the median pharyngeal raphe and its upper fibres ascend obliquely to overlap the middle constrictor, however cricopharyngeus blends with the circular oesophageal fibres around the narrowest part of the pharynx. Relations The buccopharyngeal fascia is external, the prevertebral fascia and muscles are posterior, the thyroid gland, common carotid artery and sternothyroid are lateral, and the middle constrictor, stylopharyngeus, palatopharyngeus and the fibrous lamina are internal. The internal laryngeal nerve and laryngeal branch of the superior thyroid artery reach the thyrohyoid membrane by passing between the inferior and middle constrictors. The external laryngeal nerve descends on the superficial surface of the muscle, just behind its thyroid attachment, and pierces its lower part. The recurrent laryngeal nerve and the laryngeal branch of the inferior thyroid artery ascend deep to its lower border to enter the larynx. Vascular supply The arterial supply of the inferior constrictor is derived mainly from the pharyngeal branch of the ascending pharyngeal artery and the muscular branches of the inferior thyroid artery. Innervation Both parts of the inferior constrictor are supplied by the cranial part of the accessory nerve from the pharyngeal plexus. Cricopharyngeus is also supplied by the recurrent laryngeal nerve and the external branch of the superior laryngeal nerve. Actions Thyropharyngeus constricts the lower part of the pharynx. Cricopharyngeus acts as a sphincter at the junction of the laryngopharynx and the oesophagus.
Hypopharyngeal diverticula (Fig. 35.11)
The pharyngeal mucosa that lies between cricopharyngeus and thyropharyngeus is relatively unsupported by pharyngeal muscles and is called the dehiscence of Killian. A delay in the relaxation of cricopharyngeus, which can occur when the swallowing mechanism becomes discoordinated, generates a zone of elevated pressure adjacent to the mucosa in the dehiscence. The result is the development of a pulsion diverticulum (a pouch of prolapsing mucosa), which breaches the thin muscle wall adjacent to the sixth cervical vertebra and expands, usually a little to the left side, into the parapharyngeal potential space. This may trap portions (or all) of the passing food bolus, resulting in regurgitation of old food, aspiration pneumonia, halitosis and weight loss. Treatment may involve open excision or inversion of the pouch to prevent it filling, coupled with division of the circular fibres of cricopharyngeus, to prevent the build-up of pressure in the region and recurrence of the pouch.
SALPINGOPHARYNGEUS (Fig. 35.3) Salpingopharyngeus arises from the inferior part of the cartilage of the pharyngotympanic tube near its pharyngeal opening and passes downwards within the salpingopharyngeal fold to blend with palatopharyngeus. Vascular supply Salpingopharyngeus receives its arterial supply from the ascending palatine branch of the facial artery, the greater palatine branch of the maxillary artery and the pharyngeal branch of the ascending pharyngeal artery. Innervation Salpinopharyngeus is innervated through the cranial part of the accessory nerve from the pharyngeal plexus. Actions Salpingopharyngeus elevates the pharynx, and may also assist tensor veli palatini to open the cartilaginous end of the pharyngotympanic tube during swallowing.
STYLOPHARYNGEUS (Figs 33.6, 35.3, 35.8) Stylopharyngeus is a long slender muscle, cylindrical above and flat below. It arises from the medial side of the base of the styloid process, descends along the side of the pharynx and passes between the superior and middle constrictors to spread out beneath the mucous membrane. Some fibres merge into the constrictors and the lateral glossoepiglottic fold, while others join fibres of palatopharyngeus and are attached to the posterior border of the thyroid cartilage. The glossopharyngeal nerve curves round the posterior border and the lateral side of stylopharyngeus, and passes between the superior and middle constrictors to reach the tongue. page 629 page 630
Figure 35.11 Hypopharyngeal diverticulae. Posterior (A) and lateral (B) views showing Zenker's diverticulum arising from the dehiscence of Killian between cricopharyngeus and thyropharyngeus.
Vascular supply
Stylopharyngeus receives its arterial supply from the pharyngeal branch of the ascending pharyngeal artery. Innervation Stylopharyngeus is innervated by the glossopharyngeal nerve. Actions Stylopharyngeus elevates the pharynx and larynx.
© 2008 Elsevier
PHARYNGEAL PLEXUS Almost all of the nerve supply to the pharynx, whether motor or sensory, is derived from the pharyngeal plexus which lies on the external surface of the pharynx, especially on the middle constrictor. The plexus is formed by the pharyngeal branches of the glossopharyngeal and vagus nerves with contributions from the superior cervical sympathetic ganglion. Mixed nerves leave the plexus and ascend or descend external to the superior and inferior constrictors before branching within the muscular layer and mucosa. The muscles of the pharynx - with the exception of stylopharyngeus, which is supplied by the glossopharyngeal nerve - are supplied from the pharyngeal plexus by the pharyngeal branch of the vagus. This branch emerges from the upper part of the inferior vagal ganglion and consists chiefly of filaments from the cranial accessory nerve. It passes between the external and internal carotid arteries to reach the upper border of the middle pharyngeal constrictor. It gives off a minute filament, the ramus lingualis vagi, which joins the hypoglossal nerve as it curves round the occipital artery. The inferior constrictor also receives contributions from the external and recurrent laryngeal nerves.
© 2008 Elsevier
ANATOMY OF SWALLOWING (DEGLUTITION) Swallowing is initiated reflexly when food or liquid stimulates sensory nerves in the oropharynx. In man, c.600 swallows are reputed to occur in each 24-hour period, but of these, only some 150 relate to feeding. The remaining swallows, which occur less frequently at night, are unconscious 'empty' swallows that appear to relate primarily to the clearance of saliva from the mouth. However, swallowing in the adult human has traditionally been studied in relation to food swallowing (solid or liquid) carried out on command. Such swallows have been divided into three phases, usually described as the first or oral, the second or pharyngeal and the third or oesophageal, phases. Over the years, the exact dividing line between the first and second phases has changed slightly, and is defined primarily by convention.
ORAL PHASE The oral phase is traditionally described as a voluntary action in which a bolus of food is moved from the oral cavity up to or through the fauces. Transport of the bolus through the mouth is accomplished by first forming a shallow midline gutter along the tongue to accommodate the bolus, and then by elevating the tongue and the floor of that midline gutter from before backwards. The gutter is probably formed by the co-contraction of the styloglossi and the genioglossi. At this stage a posterior oral seal exists which is associated with elevation of the posterior tongue. Emptying of the longitudinal gutter may involve contraction of hyoglossus and some intrinsic lingual muscles but there is a simultaneous elevation of the anterior and mid tongue, the hyoid bone and the floor of the mouth, which is produced by co-contraction of muscles such as mylohyoid, geniohyoid and stylohyoid. Elevation is accompanied by a relaxation of the posterior oral seal and a forward movement of the posterior tongue which is followed by bilateral contraction of the palatoglossi. The overall effect is of a cam-like action of the tongue which sweeps or squeezes the bolus towards the fauces where the pharyngeal aperture is initially increased and then closed.
PHARYNGEAL PHASE The pharyngeal stage is considered to be reflex and involves the pharynx changing from being an air channel (between the posterior nares and laryngeal inlet) to being a food channel (from the fauces to the upper end of the oesophagus). This complex action requires a brief cessation of respiratory movements and closure of the airway at two levels. At the upper level, a seal is produced by activation of the superior pharyngeal constrictor and contraction of a subset of palatopharyngeal fibres forming a variable, ridge-like, structure (Passavant's ridge) to which the soft palate is elevated. From an evolutionary perspective, the ridge represents the remnant of a sphincter which formed around a larynx which was situated higher in the neck: a high laryngeal position, by comparison with the human adult, is the norm in other mammals and in the human infant. Interestingly, the pharyngeal ridge becomes hypertrophic in an infant with a cleft palate, presumably in an attempt to produce a seal to the nasal airway. At the lower level, in the normal adult, the seal of the airway at the
laryngeal inlet is produced by closure of the glottis. The inlet is further protected by raising and tipping the laryngeal inlet forward under the bulge of the posterior tongue and by the flexing of the epiglottis over the laryngeal inlet as the bolus passes over it. page 630 page 631
In this second stage, the three pharyngeal constrictor muscles undergo a sequential contraction which is usually interpreted as the driving force which propels the bolus towards the oesophagus. However, recent evidence that the head of the bolus moves faster than the wave of pharyngeal contraction suggests that, at least in some situations, the kinetic energy imparted to the bolus as it is expelled from the mouth into the pharynx may be sufficient to carry it through the pharynx. The function of sequential contraction of the pharyngeal constrictor muscles may then be to facilitate subsequent pharyngeal clearance.
OESOPHAGEAL PHASE The third or oesophageal stage involves the relaxation of cricopharyngeus (the upper oesophageal sphincter) to allow the bolus to enter the oesophagus. Once in the oesophagus, the bolus is propelled by sequential waves of contractions of the oesophageal musculature down to the lower oesophageal sphincter, which opens momentarily to allow the bolus to enter the stomach.
CENTRAL PATTERN GENERATION The patterning and timing of striated muscle contraction in the first, second and in the early part of the third, stages of swallowing are generated at a brain stem level in a network of neural circuits which form a central pattern generator (CPG). In contrast, the patterns of activation in the smooth muscle of the lower part of the oesophagus are generated locally in intramural plexi driven by vagal autonomics. The CPG is activated by nerve impulses travelling via descending pathways from the motor cortex, and ascending pathways from peripheral sensory nerves, particularly the superior laryngeal nerve, which innervates the valleculae, epiglottis and part of the larynx. Within the CPG, the nucleus of the tractus solitarius receives the descending and peripheral afferent influences. Motor neurones collectively innervate pharyngeal striated muscle and bilaterally represent the outflow from the CPG.
SWALLOWING IN THE NEONATE (Fig. 35.12) The functional and anatomical differences between the human neonate and the adult suggest that the classical description of the divisions of a swallow should not be applied uncritically to all aspects of swallowing in the human infant. In the adult, the tip of the epiglottis is significantly lower than the inferior edge of the soft palate. In the neonate, the epiglottis may extend above the soft palate so that the laryngeal airway is in direct continuity with the posterior nares, and a potential space is formed between the soft palate above, the epiglottis behind and the tongue anteroinferiorly. In other mammals with an oropharyngeal anatomy similar to that of the human infant, up to 14 cycles of tongue movement or oral phases cause the accumulation of food in this space. Subsequent emptying of the space
is a single event followed by movement of the bolus down the oesophagus. The ratio of accumulation cycles to swallow events in the human neonate is c.1.5:1, which is lower than in other mammals, but still implies some temporary accumulation. In the case of a liquid bolus, accumulated material may be passed laterally to the epiglottis through the piriform fossae rather than over the flexed epiglottis, although it is not known whether this happens in the human infant.
Figure 35.12 Sagittal section of the head of a neonate. Note the relatively high position of the larynx, the opening being at the level of the soft palate (A). B, epiglottis. (By permission from Berkovitz BKB, Holland GR, Moxham BJ 2002 Oral Anatomy, Embryology and Histology, 3rd edn. Edinburgh: Mosby.)
The change towards the adult anatomy and co-ordination of the phases of swallowing starts a few months after birth. Differential growth in length of the human pharynx causes the larynx to take up its low adult position and the epiglottis to lose contact with the soft palate. The adult anatomy does not allow any significant accumulation of food to occur immediately anterior to the epiglottis so that the transport of food through the fauces has to bear a 1:1 relationship to pharyngeal and oesophageal transit.
PHARYNGOSTOMY AND EPIGLOTTOPEXY Loss of control of the pharyngeal phase of swallowing, e.g. due to neurological disease or ablative head and neck surgery, may result in aspiration of food, especially fluids, leading to pneumonia. This problem may be addressed surgically by pharyngostomy or epiglottopexy. In pharyngostomy, a fine bore feeding tube is passed into the lower oesophagus or stomach via the nose or anterior abdominal wall. Alternatively, a tube may be passed through the cervical skin, fascia and platysma directly into the piriform fossa. This is achieved by passing a curved forcep into the piriform fossa and pushing it laterally, displacing the contents of the carotid sheath and tenting up the platysma and cervical skin from the inside. By cutting down on to the forcep it is possible to grasp the feeding
tube and pull it into the piriform fossa prior to feeding it on into the oesophagus. The tract formed by such a puncture epithelializes and may be used for long term alimentation. In epiglottopexy, the neck and the pharynx are opened to expose the laryngeal inlet, the aryepiglottic folds are denuded of mucosa to encourage their adhesion, and the epiglottis is sutured down to the aryepiglottic folds to shield the laryngeal inlet. The resultant compromise of the airway can be offset by the creation of an alternative airway via a tracheostomy. The pharyngeal wall and cervical skin are reconstituted by suturing. REFERENCES Bluestone CD 1998 Anatomy and physiology of the Eustachian tube. In: Cummings CW et al (eds) Otolaryngology Head and Neck Surgery. 3rd edition. St Louis: Mosby: 3003-25. Freelander E 1992 Blood supply of the human levator and tensor veli palatini muscles. Clin Anat 5: 34-44. Graney DO, Retruzzelli GJ, Myers EW 1998 Anatomy. In: Cummings CW et al (eds) Otolaryngology Head and Neck Surgery. 3rd edition. St Louis: Mosby: 1327-48. Provides a concise account of the anatomy of the pharynx, highlighting features of clinical relevance. Sade J (ed) 1989 Basic Aspects of the Eustachian Tube and Middle Ear Disease. Geneva: Kugler and Ghedini. Thexton A 1998 Some aspects of swallowing. In: Harris M, Edgar M, Meghji S (eds) Clinical Oral Science. Oxford: Wright: 150-66. Thexton AJ, Crompton AW 1998 The control of swallowing. In: Linden RWA (ed) The Scientific Basis of Eating. Taste and Smell, Salivation, Mastication and Swallowing and their Dysfunctions. Frontiers of Oral Biology Series, vol 9. Basel: Karger: 168-222. Wood-Jones I 1940 The nature of the soft palate. J Anat 77: 147-70. Describes the structure of the soft palate and its movements during swallowing. page 631 page 632
© 2008 Elsevier
36 NECK AND UPPER AERODIGESTIVE TRACT Larynx The larynx is an air passage, a sphincter and an organ of phonation. It extends from the tongue to the trachea. It projects ventrally between the great vessels of the neck and is covered anteriorly by skin, fasciae and the hyoid depressor muscles (Fig. 31.21) Above, it opens into the laryngopharynx and forms its anterior wall while below it continues into the trachea (Figs 35.1, 35.2). It is mobile on deglutition. At rest it lies opposite the third to sixth cervical vertebrae in adult males, although it is somewhat higher in children and adult females. In infants between 6 and 12 months, the tip of the epiglottis (the highest part of the larynx) is a little above the junction of the dens and body of the axis vertebra. Until puberty the male and female larynges are similar in size. After puberty, the male larynx enlarges considerably in comparison with that of the female: all the cartilages increase in size, the thyroid cartilage projects in the anterior midline of the neck, and its sagittal diameter nearly doubles. The male thyroid cartilage continues to increase in size until 40 years of age, after which no further growth occurs. The average measurements of the larynx in European adults are: Length Transverse diameter Sagittal diameter
In males 44 mm 43 mm 36 mm
In females 36 mm 41 mm 26 mm
© 2008 Elsevier
SKELETON OF THE LARYNX (Figs 36.1, 36.2, 36.3, 36.4) The skeletal framework of the larynx is formed by a series of cartilages interconnected by ligaments and fibrous membranes, and moved by a number of muscles. The hyoid bone is attached to the larynx: it is usually regarded as a separate structure with distinctive functional roles, and is described on page 541. The laryngeal cartilages are the single cricoid, thyroid and epiglottic cartilages, and the paired arytenoid, cuneiform, corniculate and tritiate cartilages. In relation to the surface anatomy of the larynx, the levels of the laryngeal cartilages worth noting are: C3 (level of body of hyoid and its greater cornu); C3-4 junction (level of upper border of thyroid cartilage and bifurcation of common carotid artery); C4-5 junction (level of thyroid cartilage); C6 (level of cricoid cartilage). The corniculate, cuneiform, tritiate and epiglottic cartilages and the apices of the arytenoid are composed of elastic fibrocartilage, with little tendency to calcify. The thyroid, cricoid and the greater part of the arytenoid cartilages consist of hyaline cartilage and may undergo mottled calcification as age advances, commencing about the twenty-fifth year in the thyroid cartilage and somewhat later in the cricoid and arytenoids. By the sixty-fifth year these cartilages commonly appear patchily dense in radiographs.
EPIGLOTTIS (Figs 36.3, 36.4, 36.5)
Figure 36.1 Anterolateral view of the laryngeal cartilages and ligaments.
The epiglottis is a thin leaf-like plate of elastic fibrocartilage, which projects obliquely upwards behind the tongue and hyoid body, and in front of the laryngeal inlet. Its free end, which is broad and round, and occasionally notched in the midline, is directed upwards. Its attached part, or stalk (petiolus), is long and narrow and is connected by the elastic thyroepiglottic ligament to the back of the laryngeal prominence of the thyroid cartilage just below the thyroid notch (Fig. 36.5). Its sides are attached to the arytenoid cartilages by aryepiglottic folds (which contain the aryepiglottic muscle). Its free upper anterior surface is covered by mucosa (the epithelium is non-keratinized stratified squamous), which is reflected onto the pharyngeal aspect of the tongue and the lateral pharyngeal walls as a median glossoepiglottic, and two lateral glossoepiglottic, folds. There is a depression, the vallecula, on each side of the median fold. The lower part of its anterior surface, behind the hyoid bone and thyrohyoid membrane, is connected to the upper border of the hyoid by an elastic hyoepiglottic ligament (Fig. 36.5),
and separated from the thyrohyoid membrane by adipose tissue, which constitutes the clinically important pre-epiglottic space. The smooth posterior surface is transversely concave and vertically concavo-convex, and is covered by ciliated respiratory mucosa: its lower projecting part is called the tubercle. The cartilage is posteriorly pitted by small mucous glands (Fig. 36.4). It is perforated by branches of the internal laryngeal nerve and fibrous tissue, which means that the posterior, i.e. laryngeal, surface of the epiglottis is in continuity through these perforations with the pre-epiglottic space. Functions of the epiglottis (See also p. 630.)
During deglutition the hyoid bone moves upwards and forwards, and the epiglottis is bent posteriorly as a result of pressure from the base of the tongue and contraction of the aryepiglottic muscles. The food bolus slips over its anterior surface as it bends back over the laryngeal inlet. The bolus then splits to pass into the piriform fossae, which constitute the lateral food passages. The epiglottis is not essential to swallowing, which occurs with minimal aspiration even if the epiglottis is destroyed by disease, nor is it essential for respiration or phonation. page 633 page 634
Figure 36.2 Lateral view of the laryngeal cartilages and ligaments.
Figure 36.3 Posterior view of the laryngeal cartilages and ligaments.
THYROID CARTILAGE (Figs 36.1, 36.2, 36.3, 36.4) The thyroid cartilage is the largest of the laryngeal cartilages. It consists of two quadrilateral laminae whose anterior borders fuse along their inferior two-thirds at a median angle, forming the subcutaneous laryngeal prominence ('Adam's apple'). This projection is most distinct at its upper end, and is well marked in men but scarcely visible in women. Above, the laminae are separated by a V-shaped superior thyroid notch or incisure. Posteriorly the laminae diverge, and their posterior borders are prolonged as slender horns, the superior and inferior cornua. A shallow ridge, the oblique line, curves downwards and forwards on the external surface of each lamina: it runs from the superior thyroid tubercle lying a little anterior to the root of the superior cornu, to the inferior thyroid tubercle on the inferior border of the lamina. Sternothyroid, thyrohyoid, and thyropharyngeus (part of the inferior pharyngeal constrictor) are attached to this line (Fig. 36.6). The internal surface of the lamina is smooth. Above and behind, it is slightly concave and covered by mucosa. The thyroepiglottic ligament, the paired vestibular and vocal ligaments, and the thyroarytenoid, thyroepiglottic and vocalis muscles are all attached to the inner surface of the cartilage, in the angle between the laminae. The superior border of each lamina is concave behind and convex in front, and the thyrohyoid membrane is attached along this edge (Figs 36.1, 36.2, 36.3). The inferior border of each lamina is concave behind and nearly straight in front, and the two parts are separated by the inferior thyroid tubercle (Fig. 36.2). Anteriorly, the thyroid cartilage is connected to the cricoid cartilage by the anterior (median) cricothyroid ligament, which is a thickened portion of the cricothyroid membrane. The anterior border of each thyroid lamina fuses with its partner at an angle of c.90° in men and c.120° in women. The shallower angle in men is associated with the larger laryngeal prominence, the greater length of the vocal cords, and the resultant deeper pitch of the voice. The posterior border is thick and rounded and receives fibres of stylopharyngeus and palatopharyngeus. The superior cornu, which is long and narrow, curves upwards, backwards and medially, and ends in a conical apex to which the lateral thyrohyoid ligament is attached. The inferior cornu is short and thick, and curves down and slightly anteromedially. On the medial surface of its lower end there is a small oval facet for articulation with the side of the cricoid cartilage: it is usually only well-defined in c.20% of cases. A narrow, rhomboidal, flexible strip, the intrathyroid cartilage, lies between the two laminae, and is joined to them by fibrous tissue, during infancy.
CRICOID CARTILAGE (Fig. 36.4) The cricoid cartilage is attached below to the trachea, and articulates with the thyroid cartilage and the two arytenoid cartilages by synovial joints. It forms a complete ring around the airway, the only laryngeal cartilage to do so (Fig. 36.4).
It is smaller, but thicker and stronger, than the thyroid cartilage, and has a narrow curved anterior arch, and a broad, flatter posterior lamina. Cricoid arch
The cricoid arch is vertically narrow in front (5-7 mm in height), and widens posteriorly towards the lamina. Cricothyroid is attached to the external aspect of its front and sides, and cricopharyngeus (part of the inferior pharyngeal constrictor) is attached behind cricothyroid. The arch is palpable below the laryngeal prominence, from which it is separated by a depression containing the resilient cricovocal membrane. Cricoid lamina
The cricoid lamina is approximately quadrilateral in outline, and 2-3 cm in vertical dimension. It bears a posterior median vertical ridge. The two fasciculi of the longitudinal layer of oesophageal muscle fibres (muscularis externa) are attached by a tendon to the upper part of the ridge. Posterior cricoarytenoid attaches to a shallow depression on either side of the ridge. A discernible circular synovial facet, facing posterolaterally, sometimes marks the junction of the lamina and arch: it indicates the site where the cricoid articulates with the inferior thyroid cornu. The inferior border of the cricoid is horizontal, and joined to the first tracheal cartilage by the cricotracheal ligament (Fig. 36.1). The superior border runs obliquely up and back, and gives attachment anteriorly to the thick median part of the cricothyroid membrane, and laterally to the membranous parts of the cricothyroid membrane (Fig. 36.1) and lateral cricoarytenoid. The posterosuperior aspect of the lamina presents a shallow median notch, on each side of which is a smooth, oval, convex facet, directed upwards and laterally, for articulation with the base of an arytenoid cartilage. The internal surface of the cricoid cartilage is smooth and lined by mucosa. Subglottic stenosis
Congenital malformation of the cricoid cartilage may result in severe narrowing of the subglottic airway and respiratory obstruction. A similar situation may result from trauma and scarring following prolonged endotracheal intubation for the purposes of ventilation of premature babies on intensive care units. page 634 page 635
Figure 36.4 Cartilages of the larynx: thyroid (A), arytenoid (B), cricoid (C), epiglottis (D). The attachments of the vestibular ligaments (above) and the vocal ligaments (below) are shown in A, posterior aspect. Note the pitted surface of the epiglottis (D).
ARYTENOID CARTILAGE (Figs 36.3, 36.4) The paired arytenoid cartilages articulate with the lateral parts of the superior border of the cricoid lamina. Each is pyramidal, and has three surfaces, two processes, a base and an apex. The posterior surface, which is triangular, smooth and concave, is covered by transverse arytenoid. The anterolateral surface is convex and rough, and bears, near the apex of the cartilage, an elevation from which a crest curves back, down and then forwards to the vocal process. The lower part of this crest separates two depressions (foveae). The upper is triangular, and the vestibular ligament is attached to it. The lower is oblong, and vocalis and lateral cricoarytenoid are attached to it. The medial surface is narrow, smooth, and flat. It is covered by mucosa and its lower edge forms the lateral boundary of the intercartilaginous part of the rima glottidis. The base is concave, with a smooth surface for articulation with the lateral part of the upper border of the cricoid lamina. Its round, prominent lateral angle, or muscular process, projects backwards and laterally: it gives attachment to posterior cricoarytenoid behind, and lateral cricoarytenoid in front. The vocal ligament is attached to its pointed anterior angle (the vocal process), which projects horizontally forward. The apex curves backwards and medially and articulates with the corniculate cartilage.
CORNICULATE CARTILAGES (Figs 36.3, 36.5) The corniculate cartilages are two conical nodules of elastic fibrocartilage which articulate with the apices of the arytenoid cartilages, prolonging them posteromedially. They lie in the posterior parts of the aryepiglottic mucosal folds, and are sometimes fused with the arytenoid cartilages.
CUNEIFORM CARTILAGES (Fig. 36.5) The cuneiform cartilages are two small elongated, club-like nodules of elastic fibrocartilage, one in each aryepiglottic fold anterosuperior to the corniculate cartilages and are visible as whitish elevations through the mucosa.
TRITIATE CARTILAGES (CARTILAGO TRITICEA) (Fig. 36.3)
The tritiate cartilages are two small nodules of elastic cartilage which are situated one on either side above the larynx within the posterior free edge of the thyrohyoid membrane, about halfway between the superior cornu of the thyroid cartilage and the tip of the greater cornua of the hyoid bone. Their functions are unknown, although they may serve to strengthen this connection. Calcification of laryngeal cartilages
The thyroid, cricoid, and most of the arytenoid, cartilages consist of hyaline cartilage, and may therefore become calcified. This process normally starts at about 18 years of age. Initially it involves the lower and posterior part of the thyroid cartilage, and it subsequently spreads to involve the remaining cartilages. Calcification of the arytenoid cartilage starts at the base. The degree and frequency of calcification of the thyroid and cricoid cartilages appear to be less in females. There is some evidence to suggest that a predilection for tumour invasion may be enhanced by calcification of the laryngeal cartilages. UPDATE Date Added: 27 September 2005 Publication Services, Inc. Abstract: Ossification of laryngeal cartilages on lateral cephalometric radiographs. Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=15825782&query_hl=6 Ossification of laryngeal cartilages on lateral cephalometric radiographs. Mupparapu M, Vuppalapati A: Angle Orthod 75(2):196-201, 2005. page 635 page 636
Figure 36.5 Sagittal section of the left side of the larynx showing laryngeal membranes.
Figure 36.6 Lateral view of the articulated thyroid and cricoid cartilage, showing the position of the muscular attachments (blue lines) and cricothyroid fibres.
The tip and upper portion of the vocal process of the arytenoid cartilage consists of non-calcifying, elastic cartilage. This may have considerable functional significance: the vocal process may bend at the elastic cartilage during adduction and abduction, and the two arytenoid cartilages will contact mainly at their 'elastic' superior portions during adduction.
© 2008 Elsevier
JOINTS CRICOTHYROID JOINT The joints between the inferior cornua of the thyroid cartilage and the sides of the cricoid cartilage are synovial. Each is enveloped by a capsular ligament strengthened posteriorly by fibrous bands (Figs 36.2, 36.3). Both capsule and ligaments are rich in elastin fibres. The cricoid rotates on the inferior cornua around a transverse axis which passes transversely through both cricothyroid joints, and, to a limited extent, it also glides in different directions on the thyroid cornua.
CRICOARYTENOID JOINT A pair of synovial joints exists between the facets on the lateral parts of the upper border of the lamina of the cricoid cartilage and the bases of the arytenoids. Each joint is enclosed by a capsular ligament and strengthened by a ligament that, although traditionally called the posterior cricoarytenoid ligament, is largely medial in position. The cricoarytenoid joints permit the arytenoids to rotate about an oblique axis (dorso-medio-cranial to ventro-latero-caudal), by which each vocal process swings laterally or medially, thereby increasing or decreasing the width of the rima glottidis. They also permit a gliding movement, by which the arytenoids approach or recede from one another, the direction and slope of their articular surfaces imposing a forward and downward movement on lateral gliding. The movements of gliding and rotation are associated, i.e. medial gliding occurs with medial rotation and lateral gliding with lateral rotation. The posterior cricoarytenoid ligaments limit forward movements of the arytenoid cartilages on the cricoid cartilage. It has been suggested that the 'rest' position of the cricoarytenoid ligament is a major determinant of the position of a denervated vocal cord. UPDATE Date Added: 27 September 2005 Publication Services, Inc. Abstract: Anatomy of the cricothyroid articulation: differences between men and women. Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=15825579&query_hl=9 Anatomy of the cricothyroid articulation: differences between men and women. Ximenes Filho JA, Bohadana SC, Perazzio AF et al: Ann Otol Rhinol Laryngol 114(3):250-252, 2005.
ARYTENOCORNICULATE JOINTS Synovial or cartilaginous joints link the arytenoid and corniculate cartilages.
INNERVATION OF THE CRICOTHYROID, CRICOARYTENOID AND ARYTENOCORNICULATE JOINTS
The cricothyroid, cricoarytenoid and arytenocorniculate joints are innervated chiefly by branches of the recurrent laryngeal nerves, which arise independently or from branches of the nerve to the laryngeal muscles. Numerous lamellated (Pacinian) corpuscles, Ruffini corpuscles and free nerve endings occur in the capsules of the laryngeal joints.
© 2008 Elsevier
SOFT TISSUES The skeletal framework of the larynx is interconnected by ligaments and fibrous membranes, of which the thyrohyoid, cricothyroid, quadrangular and cricovocal membranes are the most significant. The thyrohyoid membrane is external to the larynx, whereas the paired quadrangular and cricovocal membranes are internal. All the membranes are composed of fibroelastic tissue. The named ligaments are the median (anterior) cricothyroid ligament, the hyoepiglottic and thyroepiglottic ligaments and the cricotracheal ligament.
EXTRINSIC LIGAMENTS AND MEMBRANES Thyrohyoid membrane
The thyrohyoid membrane is a broad, fibroelastic layer, attached below to the superior border of the thyroid cartilage lamina and the front of its superior cornua, and above to the superior margin of the body and greater cornua of the hyoid (Figs 36.1, 36.2, 36.3). It thus ascends behind the concave posterior surface of the hyoid, separated from its body by a bursa which facilitates the ascent of the larynx during swallowing. Its thicker part is the median thyrohyoid ligament. The more lateral, thinner, parts are pierced by the superior laryngeal vessels and internal laryngeal nerves (Fig. 36.2). Externally it is in contact with thyrohyoid and omohyoid and the body of the hyoid bone. Its inner surface is related to the epiglottis and the piriform fossae of the pharynx. The round, cord-like, elastic lateral thyrohyoid ligaments form the posterior borders of the thyrohyoid membrane, and connect the tips of the superior thyroid cornua to the posterior ends of the greater hyoid cornua (Fig. 36.2). Hyo- and thyroepiglottic ligaments
The epiglottis is attached to the hyoid bone and thyroid cartilage by the extrinsic hyoepiglottic and intrinsic thyroepiglottic ligaments respectively. Cricotracheal ligament
The cricotracheal ligament unites the lower cricoid border to the first tracheal cartilage, and is thus continuous with the perichondrium of the trachea (Fig. 36.1). page 636 page 637
INTRINSIC LIGAMENTS AND MEMBRANES (Figs 36.5, 36.7) The fibroelastic membrane of the larynx lies within the cartilaginous skeleton of the larynx, beneath the laryngeal mucosa. It is interrupted on both sides of the larynx by a horizontal cleft between the vestibular and vocal ligaments. Its upper part, the quadrangular membrane, extends between the arytenoid cartilages and the sides of the epiglottis. Its lower part forms the cricovocal membrane, which connects the thyroid, cricoid and arytenoid cartilages. Quadrangular membrane
Each quadrangular membrane passes from the lateral margin of the epiglottis to the arytenoid cartilage on its own side. It is often poorly defined. The upper and
lower borders of the membrane are free. The upper border slopes posteriorly to form the aryepiglottic ligament, which constitutes the central component of the aryepiglottic fold. It is less defined in its upper portion. Posteriorly it passes through the fascial plane of the oesophageal suspensory ligament, and helps to form a median corniculopharyngeal ligament which extends into the submucosa adjacent to the cricoid cartilage. This ligament may exert vertical traction. The lower border forms the vestibular fold. The cuneiform cartilages lie within the aryepiglottic folds. Cricothyroid ligament and cricovocal membrane
The cricothyroid ligament is composed mainly of elastic tissue. It consists of two parts: an anterior part, the anterior (median) cricothyroid ligament, and a lateral part, the cricovocal membrane. Cricothyroid membrane and anterior (median) cricothyroid ligament The cricothyroid membrane passes upwards from the upper border of the cricoid cartilage to the lower border of the thyroid cartilage. Anteriorly, it is thickened to form the anterior (median) cricothyroid ligament, which is broader below and narrower above. Cricovocal membrane This membrane is sometimes called the conus elasticus or the lateral cricothyroid ligament; such terminology ignores the fact that the cricovocal membrane is attached to the arytenoid cartilage as well as to the thyroid cartilage, and that it shows a thickened ligament only where it becomes the vocal ligament.
Figure 36.7 Superior view of laryngeal cartilages together with cricothyroid, quadrangular, and related ligaments and membranes.
The cricovocal membrane is thinner than the anterior cricothyroid ligament. It arises beneath the cricothyroid membrane from the inner surface of the cricoid cartilage, near its lower margin. It passes upwards beneath the lower border of the thyroid cartilage and is attached anteriorly to the inner surface of the angle of the thyroid cartilage (just below its midpoint) and posteriorly to the tip of the vocal process of the arytenoid cartilage. Between these attachments, the upper edge of the cricovocal membrane is free, horizontally aligned and thickened to form the vocal ligament, which underlines the mucosa-covered vocal cord. The cricovocal membrane is covered internally by mucosa and externally by lateral cricoarytenoid and thyroarytenoid.
MICROSTRUCTURE OF THE LARYNX The laryngeal mucosa is continuous with that of the pharynx above and the trachea below. It lines the entire inner surface of the larynx including the ventricle and saccule and is thickened over the vestibular folds where it is the chief component. Over the vocal cords it is thinner, and is firmly attached to the underlying vocal ligaments. It is loosely adherent to the anterior surface of the epiglottis, but firmly attached to its anterior surface and the floor of the valleculae. On the aryepiglottic folds it is reinforced by a considerable amount of fibrous connective tissue, and adheres closely to the laryngeal surfaces of the cuneiform and arytenoid cartilages. The laryngeal epithelium is mainly a ciliated, pseudostratified respiratory epithelium where it covers the inner aspects of the larynx, including the posterior surface of the epiglottis, and it provides a ciliary clearance mechanism shared with most of the respiratory tract (Chapter 62). However, the vocal cords are covered by non-keratinized, stratified squamous epithelium, an important variation which protects the tissue from the effects of the considerable mechanical stresses acting on the surfaces of the vocal cords. The exterior surfaces of the larynx which merge with the laryngopharynx and oropharynx (including the anterior surface of the epiglottis), are subject to the abrasive effects of swallowed food, and are also covered by non-keratinized stratified squamous epithelium. The laryngeal mucosa has numerous mucous glands, especially over the epiglottis, where they pit the cartilage, and along the margins of the aryepiglottic folds anterior to the arytenoid cartilages, where they are known as the arytenoid glands. Many large glands in the saccules of the larynx secrete periodically over the vocal cords during phonation. The free edges of these folds are devoid of glands and their stratified epithelium is vulnerable to drying and requires the secretions of neighbouring glands. Hoarseness due to excessive speaking is due to partial temporary failure of this secretion. Taste buds, like those in the tongue (p. 588), occur on the posterior epiglottic surface, aryepiglottic folds and less often in other laryngeal regions.
© 2008 Elsevier
LARYNGEAL CAVITY (Figs 36.8, 36.9) The laryngeal cavity space extends from the laryngeal inlet (from the pharynx) down to the lower border of the cricoid cartilage, where it continues into the trachea. It is partially divided into upper and lower parts by paired upper and lower mucosal folds which project into its lumen. There is a middle part between the two sets of folds. The upper folds are the vestibular (ventricular or false vocal) folds, and the median aperture which they guard is the rima vestibuli. The lower pair are the (true) vocal cords, and the fissure between the latter is the rima glottidis or glottis. The vocal cords are the primary source of phonation, whereas the vestibular folds normally do not contribute directly to sound production. The clinical term supraglottis refers to that part of the larynx which lies above the glottis. It includes the laryngeal ventricle, vestibular folds, the laryngeal surface of the epiglottis, arytenoid cartilages and the laryngeal aspects of the aryepiglottic folds.
SUPRAGLOTTIS (UPPER PART) The upper part of the laryngeal cavity contains the laryngeal inlet or aditus, the aryepiglottic fold and the laryngeal ventricle. Laryngeal inlet or aditus page 637 page 638
Figure 36.8 Sagittal section showing the interior aspect of the left half of the larynx.
Figure 36.8 Sagittal section showing the interior aspect of the left half of the larynx.
Figure 36.9 Coronal section through the larynx and the cranial end of the trachea: posterior aspect.
The upper part of the laryngeal cavity is entered by the laryngeal inlet or 'aditus laryngis', which is the aperture between the larynx and pharynx. This faces backwards and somewhat upwards, because the anterior wall of the larynx is much longer than the posterior (and slopes downwards and forwards in its upper part because of the oblique inclination of the epiglottis). The inlet is bounded anteriorly by the upper edge of the epiglottis, posteriorly by the transverse mucosal fold between the two arytenoids (posterior commissure), and on each side by the edge of a mucosal ridge, the aryepiglottic fold, which runs between the side of the epiglottis and the apex of the arytenoid cartilage. The midline groove between the two corniculate tubercles is termed the interarytenoid notch. Aryepiglottic fold
The aryepiglottic fold contains ligamentous and muscular fibres. The ligamentous fibres represent the free upper border of the quadrangular membrane (Figs 36.5, 36.7). The muscle fibres are continuations of the oblique arytenoids. The posterior part of the aryepiglottic fold has two oval swellings, one above and in front, the other behind and below. These swellings mark the positions of the underlying cuneiform and corniculate cartilages respectively. They are separated by a shallow vertical furrow which is continuous below with the opening of the laryngeal ventricle.
Laryngeal introitus
This clinical term denotes the space between the laryngeal inlet and vestibular folds. It is wide above, narrow below and higher anteriorly than posteriorly. Its anterior wall is formed by the posterior surface of the epiglottis, the lower part of which (the epiglottic tubercle) bulges backwards a little. Its lateral walls, which are higher in front and shallow behind, are the medial surfaces of the aryepiglottic folds. Its posterior wall is the interarytenoid mucosa, above the ventricular folds.
MIDDLE PART The middle part of the laryngeal cavity is the smallest, and extends from the rima vestibuli above to the rima glottidis below. On each side it contains the vestibular folds, the ventricle and the saccule of the larynx. Vestibular folds and ligaments
The narrow vestibular ligament represents the thickened lower border of the quadrangular membrane (Fig. 36.7). It is fixed in front to the thyroid angle below the epiglottic cartilage and behind to the anterolateral surface of the arytenoid cartilage above its vocal process. With its covering of mucosa, it is termed the vestibular (ventricular or false vocal) fold (Figs 36.8, 36.9). The presence of a loose vascular mucosa lends the vestibular folds a pink appearance in vivo, as they lie above and lateral to, the vocal cords. Ventricle (sinus) of larynx
On each side of the larynx, a slit between the vestibular and vocal cords opens into a fusiform recess called the laryngeal ventricle (Figs 36.8, 36.9). The ventricle extends upwards into the laryngeal wall lateral to the vestibular fold. Saccule of larynx
The ventricle of the larynx opens into the saccule of the larynx (Fig. 36.9), a pouch which ascends forwards from the ventricle, between the vestibular fold and thyroid cartilage, and occasionally reaches the upper border of the cartilage. It is conical, and curves slightly backwards; 60-70 mucous glands, sited in the submucosa, open onto its luminal surface. The orifice of the saccule is guarded by a delicate fold of mucosa, the ventriculosaccular fold. The saccule has a fibrous capsule. This is continuous below with the vestibular ligament, and is covered medially by a few muscular fasciculi from the apex of the arytenoid cartilage, which pass forwards between the saccule and vestibular mucosa into the aryepiglottic fold. Laterally the saccule is separated from the thyroid cartilage by the thyroepiglottic muscle, which compresses the saccule, expressing its secretion onto the vocal cords, which lack glands, to lubricate and protect them against desiccation and infection. Saccules occasionally protrude through the thyrohyoid membrane. Laryngocoele
The laryngeal ventricle may on occasion become pathologically enlarged due to obstruction of the ventricular aditus by inflammation, by scarring or by a tumour. As the sealed cavity of the ventricle contains the laryngeal saccule, an expanding
mucus-filled cyst is formed. This cyst (laryngocoele) may expand into the paraglottic space and extend superiorly to expand the aryepiglottic fold and reach the vallecula (internal laryngocoele). Acute respiratory obstruction may result especially if the contents of the cyst become infected. The cyst may also expand through the thyrohyoid membrane at the point of entry of the internal laryngeal neurovascular bundle to present as a lump in the neck overlying the thyrohyoid membrane (external laryngocoele). Vocal cords (folds) and ligaments page 638 page 639
The free thickened upper edge of the cricovocal membrane forms the vocal ligaments (Fig. 36.7). It stretches back on either side from the midlevel of the thyroid angle to the vocal processes of the arytenoids. When covered by mucosa, it is termed the vocal fold or vocal cord (cord is the preferred clinical term) (Figs 36.8, 36.9). The vocal cords form the anterolateral edges of the rima glottidis and are concerned with sound production. The mucosa overlying the vocal ligament is thin and lies directly on the vocal ligament, and so the vocal cord appears pearly white in vivo. It is loosely attached to the ligaments: oedema fluid readily collects in this potential space in disease. Known as Reinke's space, it extends along the length of the free margin of the vocal ligament and a little way onto the superior surface of the cord. The site where the vocal cords meet anteriorly is known as the anterior commissure. Fibres of the vocal ligament here pass through the thyroid cartilage to blend with the overlying perichondrium, forming Broyles' ligament. The latter contains blood vessels and lymphatics and is therefore a potential route for the escape of malignant tumours from the larynx. Each vocal ligament is composed of a band of yellow elastic tissue related laterally to vocalis. Reinke's oedema
The mucous membrane is loosely attached throughout the larynx and can accommodate considerable swelling which may compromise the airway in acute infections. At the edge of the true vocal cords the mucosal covering is tightly bound to the underlying ligament so that oedema fluid does not pass between the upper and lower compartments of the vocal cord mucosa. Any tissue swelling above the vocal cord exaggerates the potential space deep to the mucosa (Reinke's space), causing accumulation of extracellular fluid and flabby swelling of the vocal cords (Reinke's oedema). Smoking and vocal abuse may initiate such changes. Vocal cord nodules
Aberrant muscle balance during phonation may cause initial contact during vocal cord apposition to occur at a point at the junction of the anterior third and the posterior two-thirds of the vocal ligament. Excessive trauma at this point, for example when singing with poor technique or forcing the voice may produce subepithelial haemorrhage or bruising, and subsequent pathological changes such as subepithelial scarring ('singer's nodes' or 'clergyman's nodes'). Rima glottidis
The rima glottidis or glottis is the fissure between the vocal cords anteriorly and
the arytenoid cartilages posteriorly (Fig. 36.10). It is bounded behind by the mucosa passing between the arytenoid cartilages at the level of the vocal cords. The rima glottidis is customarily divided into two regions: an anterior intermembranous part, which makes up about three-fifths of its anteroposterior length and is formed by the underlying vocal ligament, and a posterior intercartilaginous part which is formed by the vocal processes of the arytenoid cartilages. The average sagittal diameter of the glottis in adult males is 23 mm and in adult females 17 mm. It is the narrowest part of the larynx. Its width and shape vary with the movements of the vocal cords and arytenoid cartilages during respiration and phonation.
SUBGLOTTIS (LOWER PART) The lower part of the laryngeal cavity, or the subglottis, extends from the vocal cords to the lower border of the cricoid. In transverse section it is elliptical above and wider and circular below, and is continuous with the trachea. Its walls are lined by respiratory mucosa, and supported by the cricothyroid ligament above and the cricoid cartilage below.
LARYNGOSCOPIC EXAMINATION (Fig. 36.12) The laryngeal inlet, the structures around it and the cavity of the larynx can all be inspected using fibreoptic endoscopy, either through the mouth (Fig. 36.11A) or nasopharynx (Fig. 36.11B). The epiglottis is seen foreshortened, but its tubercle is visible. From the epiglottic margins the aryepiglottic folds can be traced posteromedially and the cuneiform and corniculate elevations recognized. The pink vestibular folds and pearly white vocal cords are visible and, when the rima glottidis is wide open, the anterior arch of the cricoid cartilage, the tracheal mucosa and cartilages may be seen. The piriform fossae can also be inspected.
LARYNGEAL OBSTRUCTION AND TRAUMA
Figure 36.10 Different positions of the vocal cords and arytenoid cartilages. A, Position of rest in quiet respiration. The intermembranous part of the rima glottidis is triangular and the intercartilaginous part is rectangular in shape. Key: 1, intermembranous part of glottis; 2, intercartilagenous part of glottis. B, Forced inspiration. Both parts of the rima glottidis are triangular in shape. C, Abduction of the vocal cords. The arrows indicate the lines of pull of the posterior cricoarytenoid muscles. The abducted vocal cords and the abducted, retracted and laterally rotated arytenoid cartilages are shown in dotted outline. Both parts of the rima glottidis are triangular. D, Adduction of the vocal cords. The arrows indicate the lines of pull of the lateral cricoarytenoid muscles. The adducted vocal cords and the medially rotated arytenoid cartilages are shown in dotted outlines. E, Closure of the rima glottidis. The arrows indicate the line of pull of the transverse arytenoid muscle. Both the vocal cords and the arytenoid cartilages are adducted (dotted lines), but there is no rotation of the latter.
Large foreign bodies may obstruct the laryngeal inlet or rima glottidis and suffocation may ensue, while smaller ones can enter the trachea or bronchi, or lodge in the laryngeal ventricle and cause reflex closure of the glottis with consequent suffocation. Inflammation of the upper larynx, e.g. secondary to infection or the effects of smoke inhalation, may swell the mucosa by effusion of fluid into the abundant, loose submucous tissue (oedema of the supraglottis). The effusion does not involve or extend below the vocal cords, because the mucosa adheres directly to the vocal ligaments without the intervention of submucous tissue. Laryngotomy below the vocal cords through the cricothyroid ligament or
tracheotomy may be necessary to restore a free airway. The mucosa of the upper larynx is highly sensitive, and contact with foreign bodies excites immediate coughing. Suicidal wounds are usually made through the thyrohyoid membrane, damaging the epiglottis, superior thyroid vessels, external and internal carotid arteries and internal jugular veins. Less frequently they are above the hyoid with damage to the lingual muscles and lingual and facial vessels.
© 2008 Elsevier
THE INFANT LARYNX page 639 page 640
Figure 36.11 Laryngoscopic approaches. A shows the oral and B the nasopharyngeal approaches to the visualization of the larynx.
The infant larynx differs markedly from the adult larynx. Its cavity is short and funnel-shaped, and is about one-third the size of the adult, although it is proportionately larger, and this has two main consequences. First, its lumen is disproportionately narrower than the adult, and second, it lies higher in the neck than the adult larynx. At rest the upper border of the epiglottis is at the level of the second or third cervical vertebra, and when the larynx is elevated it reaches the level of the first cervical vertebra. This high position is associated with the ability
of the infant to use its nasal airway to breathe while suckling. The epiglottis is Xshaped, with a furled petiole, and the laryngeal cartilages are softer and more pliable than the adult larynx (which may predispose to airway collapse in inspiration, which leads to the clinical picture of laryngomalacia). The thyroid cartilage is shorter and broader than in the adult and lies closer to the hyoid bone in the neonate. This means that the thyrohyoid ligament is relatively short. Neither the superior notch nor the laryngeal prominence are as marked as they are in the adult. The cricoid cartilage is the same shape as in the adult. The vocal cords are 4-4.5 mm long, which is relatively shorter than in either childhood or the adult.
Figure 36.12 Laryngeal folds viewed in abduction, as seen through a laryngoscope. (By permission from Berkovitz BKB, Moxham BJ 2002 Head and Neck Anatomy. London: Martin Dunitz.)
The mucosa of the supraglottis is more loosely attached than it is in the adult larynx and it exhibits multiple submucosal glands. Inflammation of the supraglottis will therefore rapidly result in gross oedema. The mucosa is also lax in the subglottis, which is the narrowest part of the infant larynx: it measures 3.5 mm in diameter in neonates. Swelling at this point rapidly results in severe respiratory obstruction. Unlike the adult, the neonatal subglottic cavity extends posteriorly as well as inferiorly, which is important to consider when passing an endotracheal tube. The ventricle of the larynx is small, whereas the saccule of the larynx is often considerably larger than it is in adult life. By about the third year, sexual differences become apparent in the larynx. It is larger in boys, but the angle between the thyroid laminae is more pronounced in girls. At puberty these changes increase, and there is greater enlargement of the male larynx.
UPDATE Date Added: 05 October 2005 Publication Services, Inc. Abstract: Dimensions of the neonatal cricothyroid membrane - how feasible is a surgical cricothyroidotomy? Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=15828992&query_hl=9 Dimensions of the neonatal cricothyroid membrane - how feasible is a surgical cricothyroidotomy? Navsa N, Tossel G, Boon JM: Paediatr Anaesth 15(5):402406, 2005. UPDATE Date Added: 27 September 2005 Publication Services, Inc. Abstract: Developmental changes of laryngeal dimensions in un-paralyzed, sedated children. Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=12502977&query_hl=8 Developmental changes of laryngeal dimensions in un-paralyzed, sedated children. Litman RS, Weissend EE, Shibata D et al: Anesthesiology 98(1):41-45, 2003.
© 2008 Elsevier
THE PARALUMENAL SPACES A number of potential spaces or compartments can be identified in and around the larynx. The three most commonly considered are the pre-epiglottic, the paraglottic and the subglottic spaces. They are not closed compartments and their existence does not preclude the spread of tumours. Knowledge of the anatomy of these spaces, and the potential pathways of spread of tumours from them, has significantly influenced the surgical approach to disease in this region.
THE PRE-EPIGLOTTIC SPACE The pre-epiglottic space might be expected to lie anterior to the epiglottis, but in reality, it also extends beyond the lateral margins of the epiglottis, which gives it the form of a horse-shoe. It is primarily filled with adipose tissue and appears to contain no lymph nodes. Its upper boundary is formed by the weak hyoepiglottic membrane, which is strengthened medially as the median hyoepiglottic ligament. Its anterior boundary is the thyrohyoid membrane, which is strengthened medially as the median thyrohyoid ligament. Its lower boundary is the thyroepiglottic ligament, which is continuous laterally with the quadrangular membrane behind. Its upper lateral border is the greater cornu of the hyoid bone. Inferolaterally, the pre-epiglottic space is in continuity with the paraglottic space and it is often invaded from the latter by the laryngeal saccule. It is also in continuity with the mucosa of the laryngeal surface of the epiglottis via multiple perforations in the cartilage of the epiglottis. It is through these perforations that malignancies of the laryngeal surface of the epiglottis may invade the fat and areolar tissue of the pre-epiglottic space.
THE PARAGLOTTIC SPACE page 640 page 641
The paraglottic space is a region of adipose tissue which contains the internal laryngeal nerve, the laryngeal ventricle and part, or all, of the laryngeal saccule. It is bounded laterally by the thyroid cartilage and thyrohyoid membrane. Superomedially, it is usually continuous with the pre-epiglottic space, although it may be partitioned from it by a fibrous septum. The cricovocal membrane lies inferomedially, and the mucosa of the piriform fossa lies posteriorly. The lower border of the thyroid cartilage is inferior. Anteroinferiorly, there are deficiencies in the paramedian gap at the side of the anterior cricothyroid ligament, and posteroinferiorly adipose tissue extends towards the cricothyroid joint. Some authorities exclude thyroarytenoid from the paraglottic space. Supraglottic tumours may spread into the paraglottic space and reach the subglottis, or extend beyond the limits of the larynx. Ventricular tumours may obstruct mucous outflow from the saccule and cause its expansion within the paraglottic space to form a secondary laryngocoele: the tumour itself may also spread transglottically, and thereby fix the vocal cord either by invasion of the cricoarytenoid joint or by damaging the recurrent laryngeal nerve. Fixation of the
vocal cord is a good indicator of a tumour within the paraglottic space. The proximity of the mucosa at the piriform fossa makes its removal in surgery mandatory for such disease.
THE SUBGLOTTIC SPACE The subglottic space is bounded laterally by the cricovocal membrane, medially by the mucosa of the subglottic region and above by the undersurface of Broyle's ligament in the midline. It is continuous below with the inner surface of the cricoid cartilage and its mucosa.
© 2008 Elsevier
MUSCLES The muscles of the larynx may be divided into extrinsic and intrinsic groups. The extrinsic muscles connect the larynx to neighbouring structures and are responsible for moving it vertically during phonation and swallowing. They include the infrahyoid strap muscles, i.e. thyrohyoid, sternothyroid and sternohyoid, and the inferior constrictor muscle of the pharynx. Two of the three elevator muscles of the pharynx, i.e. stylo- and palatopharyngeus, are also connected directly to the thyroid cartilage, mainly to the posterior aspect of the thyroid lamina and cornu. The role of the extrinsic muscles during respiration appears variable. Thus, the larynx has been seen to rise, descend or barely move, during inspiration. The extrinsic muscles can affect the tone and pitch of the voice by raising or lowering the larynx, and geniohyoid elevates and anteriorly displaces the larynx, particularly during deglutition. The intrinsic muscles are the cricothyroid, posterior and lateral cricoarytenoid, transverse and oblique arytenoid, aryepiglotticus, thyroarytenoid and its subsidiary part, vocalis, and thyroepiglotticus: all are confined to the larynx in their attachments, and all but the transverse arytenoid are paired. Whereas most of the intrinsic muscles lie internally, under cover of the thyroid cartilage or the mucosa, the cricothyroids appear on the outer aspect of the larynx. The intrinsic laryngeal muscles may be placed in three groups according to their main actions. The posterior and lateral cricoarytenoids and oblique and transverse arytenoids vary the dimensions of the rima glottidis. The cricothyroids, posterior cricoarytenoids, thyroarytenoids and vocales regulate the tension of the vocal ligaments. The oblique arytenoids, aryepiglottic and thyroepiglottic muscles modify the laryngeal inlet. Bilateral pairs of muscles usually act in concert with each other. Neuromuscular spindles have been found in all human laryngeal muscles, the maximum number (23) being found in the transverse arytenoid. The control of phonation requires very considerable neuromuscular co-ordination, and effective proprioception would appear to be essential to this capacity. The mass of muscle related to adduction far outweighs that related to abduction. In this context, it is of interest to note that histological examination of normal larynges revealed evidence of some degenerative changes in the posterior cricoarytenoid muscle, the single muscle associated with abduction, but none in the remaining muscles.
INTRINSIC MUSCLES Oblique arytenoid and the aryepiglottic muscle
The oblique arytenoids lie superficial to the transverse arytenoid, and are sometimes considered to be part of it. They cross each other obliquely at the back of the larynx, each extending from the back of the muscular process of one arytenoid cartilage to the apex of the opposite one. Some fibres continue laterally round the arytenoid apex into the aryepiglottic fold, forming the aryepiglottic
muscle (Fig. 36.13). Vascular supply Oblique arytenoid receives its blood supply from the laryngeal branches of the superior and inferior thyroid arteries.
Figure 36.13 The muscles of the larynx (most of the left lamina of the thyroid cartilage has been removed): left lateral aspect.
Innervation Oblique arytenoid is innervated by the recurrent laryngeal nerve. Actions The oblique arytenoids and aryepiglottic muscles act as a sphincter of the laryngeal inlet by adducting the aryepiglottic folds, and approximating the arytenoid cartilages to the tubercle of the epiglottis. Their poor development limits their capacity to act as a sphincter of the inlet. Transverse (inter) arytenoid (Fig. 35.2)
Transverse arytenoid is a single, unpaired muscle which bridges the gap at the back of the larynx between the two arytenoid cartilages and fills their posterior concave surfaces. It is attached to the back of the muscular process and adjacent lateral border of both arytenoids.
Vascular supply Transverse arytenoid receives its blood supply from the laryngeal branches of the superior and inferior thyroid arteries. Innervation Transverse arytenoid is innervated by the recurrent laryngeal nerves. It also receives branches from the internal laryngeal nerve, although there is debate as to whether these branches contain any distinct motor input. The nerves form a dense, but highly variable, plexus. Actions Transverse arytenoid pulls the arytenoid cartilages towards each other, closing the posterior, intercartilaginous, part of the rima glottidis (Fig. 36.10E). This action is accomplished by drawing the arytenoids upwards along the sloping shoulders of the cricoid lamina, without rotation. Posterior cricoarytenoid (Fig. 35.2) page 641 page 642
Posterior cricoarytenoid arises from the posterior surface of the cricoid lamina (Fig. 36.4). Its fibres ascend laterally and converge to insert on the back of the muscular process of the ipsilateral arytenoid cartilage. The highest fibres run almost horizontally, the middle obliquely, and the lowest are almost vertical: some reach the anterolateral surface of the arytenoid cartilage. An additional strip of muscle, ceratocricoid, is occasionally seen in relation to the lower border of posterior cricoarytenoid, arising from the cricoid cartilage and inserting on to the posterior aspect of the inferior horn of the thyroid cartilage. Vascular supply Posterior cricoarytenoid receives its blood supply from the laryngeal branches of the superior and inferior thyroid arteries. Innervation Posterior cricoarytenoid is innervated by the recurrent laryngeal branch of the vagus. Actions The posterior cricoarytenoids are the only laryngeal muscles which open the glottis, rotating the arytenoid cartilages laterally around an axis passing through the cricoarytenoid joints, and thus separating the vocal processes and the attached vocal cords (Fig. 36.10C). They also pull the arytenoids backwards, assisting the cricothyroids to tense the vocal cords. The most lateral fibres draw the arytenoid cartilages laterally, and so the rima glottidis becomes triangular when the posterior cricoarytenoid muscles contract. Lateral cricoarytenoid
Lateral cricoarytenoid is attached anteriorly to the upper border of the cricoid arch. It ascends obliquely backwards to be attached to the front of the muscular process of the ipsilateral arytenoid cartilage (Fig. 36.13). Vascular supply
Lateral cricoarytenoid receives its blood supply from the laryngeal branches of the superior and inferior thyroid arteries. Innervation Lateral cricoarytenoid is innervated by the recurrent laryngeal nerve via a single branch which forms a homogeneous nerve plexus located in the middle of the muscle. This suggests that lateral cricoarytenoid acts as a single unit, unlike the other intrinsic muscles of the larynx. Actions Lateral cricoarytenoid rotates the arytenoid cartilage in a direction opposite to that of posterior cricoarytenoid, and so closes the rima glottidis (Fig. 36.10D). Cricothyroid (Fig. 36.6)
Cricothyroid is attached anteriorly to the external aspect of the arch of the cricoid cartilage. Its fibres pass backwards and diverge into two groups, a lower 'oblique' part which slants backwards and laterally to the anterior border of the inferior cornu of the thyroid; and a superior 'straight' part which ascends more steeply backwards to the posterior part of the lower border of the thyroid lamina. The medial borders of the paired cricothyroids are separated anteriorly by a triangular gap which is occupied by the anterior cricothyroid ligament. Vascular supply Cricothyroid is supplied by the cricothyroid artery, a branch of the superior thyroid artery, which crosses high on the cricothyroid ligament to communicate with its contralateral fellow. Innervation Unlike the other intrinsic muscles of the larynx, cricothyroid is innervated not by the recurrent laryngeal nerve, but by the external branch of the superior laryngeal nerve. Actions The cricothyroids stretch the vocal ligaments by tilting the thyroid cartilage forwards and downwards on the cricoid. Because the arytenoid cartilages are anchored to the cricoid lamina, the sagittally directed rotation of the thyroid cartilage increases the distance between their vocal processes and the anterior angle of the thyroid, and so lengthens, and affects tension in, the vocal ligaments. Thyroarytenoid and vocalis (Figs 36.9, 36.13)
Thyroarytenoid is a broad, thin muscle, lying lateral to the vocal cord, cricovocal membrane, laryngeal ventricle and saccule. It is attached anteriorly to the lower half of the angle of the thyroid cartilage, and to the cricothyroid ligament. Its fibres pass backwards, laterally and upwards to the anterolateral surface of the arytenoid cartilage. Its lower and deeper fibres form a band which, in a coronal section, appears as a triangular bundle, and is attached to the lateral surface of the vocal process and to the inferior impression on the anterolateral surface of the arytenoid cartilage. This bundle, the vocalis muscle, is parallel with and just lateral to the vocal ligament. It is said to be thicker behind than in front, because many deeper fibres start from the vocal ligament and so do not extend to the thyroid
cartilage. Others consider that all its fibres loop and intertwine as they pass from the thyroid to the arytenoid cartilage. A few fibres extend along the wall of the ventricle from the lateral margin of the arytenoid cartilage to the side of the epiglottis. The superior thyroarytenoid, a small muscle which is not always present, lies on the lateral surface of the main mass of the thyroarytenoid, and extends obliquely from the thyroid angle to the muscular process of the arytenoid cartilage (Fig. 36.13). Vascular supply Thyroarytenoid receives its arterial blood supply from the laryngeal branches of the superior and inferior thyroid arteries. Innervation All parts of thyroarytenoid are supplied by the recurrent laryngeal nerve. It also receives a communicating branch from the external laryngeal nerve, although it is not clear whether such branches carry motor or sensory fibres. Actions The thyroarytenoids draw the arytenoid cartilages towards the thyroid cartilage, thereby shortening and relaxing the vocal ligaments. At the same time, they rotate the arytenoids medially to approximate the vocal cords. In addition, they can rotate the arytenoid cartilages medially and so aid closure of the rima glottidis. Relaxation of the posterior parts of the vocal ligaments by the vocalis muscles, combined with tension in the anterior parts of the ligaments, is responsible for raising the pitch of the voice. Vocalis can change the timbre of the voice by affecting the mass of the vocal cords. Thyroepiglotticus
Many of the fibres of thyroarytenoid are prolonged into the aryepiglottic fold, where some terminate, and others continue to the epiglottic margin as the thyroepiglottic muscle (Fig. 36.13). The thyroepiglottics can widen the inlet of the larynx by their action on the aryepiglottic folds.
© 2008 Elsevier
VASCULAR SUPPLY AND LYMPHATIC DRAINAGE (Fig. 36.14)
page 642 page 643
Figure 36.14 Arterial supply and lymphatic drainage of the larynx. (Adapted from Oxford Textbook of Functional Anatomy, Vol 3 Head and Neck, MacKinnon P, Morris J (eds), 1990. By permission of Oxford University Press.)
The blood supply of the larynx is derived mainly from the superior and inferior laryngeal arteries. Rich anastomoses exist between the corresponding contralateral laryngeal arteries and between the ipsilateral laryngeal arteries. The superior laryngeal arteries supply the greater part of the tissues of the larynx, from the epiglottis down to the level of the vocal cords, including the majority of the laryngeal musculature. The inferior laryngeal artery supplies the region around cricothyroid, while its posterior laryngeal branch supplies the tissue around posterior cricoarytenoid.
SUPERIOR LARYNGEAL ARTERY The superior laryngeal artery is normally derived from the superior thyroid artery, a branch of the external carotid artery, as this artery passes down towards the upper pole of the thyroid gland. However, in c.15% of cases, it arises directly from the external carotid artery between the origins of the superior thyroid and lingual arteries. The superior laryngeal artery runs down towards the larynx, with the internal branch of the superior laryngeal nerve lying above it. It enters the larynx by penetrating the thyrohyoid membrane and divides into a number of branches
which supply the larynx from the tip of the epiglottis down to the inferior margin of thyroarytenoid. It supplies the larynx, and anastomoses with its fellow and the inferior laryngeal branch of the inferior thyroid artery. The cricothyroid artery arises from the superior thyroid artery and may contribute to the supply of the larynx. It follows a variable course either superficial or deep to sternothyroid. If superficial, it may be accompanied by branches of the ansa cervicalis, and if deep, it may be related to the external laryngeal nerve. It can anastomose with the artery of the opposite side and with the laryngeal arteries.
INFERIOR LARYNGEAL ARTERY The inferior laryngeal artery is smaller than the superior laryngeal artery. It is a branch of the inferior thyroid artery, which arises from the thyrocervical trunk of the subclavian artery. It ascends on the trachea with the recurrent laryngeal nerve, enters the larynx at the lower border of the inferior constrictor, just behind the cricothyroid articulation, and supplies the laryngeal muscles and mucosa. It anastomoses with its contralateral fellow, and with the superior laryngeal branch of the superior thyroid artery. A posterior laryngeal artery of variable size has been described as a regular feature that arises as an internal branch of the inferior thyroid artery.
SUPERIOR AND INFERIOR LARYNGEAL VEINS Venous return from the larynx occurs via superior and inferior laryngeal veins which run parallel to the laryngeal arteries. They are tributaries of the superior and inferior thyroid veins respectively. The superior thyroid vein drains into the internal jugular vein and the inferior thyroid vein usually into the left brachiocephalic vein.
LYMPHATIC DRAINAGE (Fig. 36.14) The lymph vessels draining the supraglottic part of the larynx above the vocal cords accompany the superior laryngeal artery, pierce the thyrohyoid membrane, and end in the upper deep cervical lymph nodes, often bilaterally. The supraglottic lymphatics also communicate with those at the base of the tongue. The vocal cords with their firmly bound mucosa and paucity of lymphatics provide a clear demarcation between the upper and lower areas of the larynx. Below the vocal cords, some of the lymph vessels pass through the cricovocal membrane to reach the prelaryngeal (Delphian) and/or pretracheal lymph nodes. Others run with the inferior laryngeal artery to join the lower deep cervical nodes. Spread of supra- and subglottic tumours
The upper deep cervical lymph nodes act as pathways of spread for malignant tumours of the supraglottic larynx: up to 40% of these tumours will have undergone such spread at the time of clinical presentation. The glottis is very poorly endowed with lymphatic vessels, which means that 95% of malignant tumours confined to the glottis will present with a change in voice or airway obstruction but will not show signs of spread to adjacent lymph nodes at presentation. Tumours of the subglottic larynx will often spread to the
paratracheal lymph node chain prior to clinical presentation. However, the presenting symptoms may be voice change and airway obstruction rather than a mass in the neck, because the paratracheal lymph nodes occupy a deep seated position in the root of the neck and so their enlargement may remain occult.
© 2008 Elsevier
INNERVATION (Fig. 36.15) The larynx is innervated by the internal and external branches of the superior laryngeal nerve, the recurrent laryngeal nerve and sympathetic nerves. The internal laryngeal nerve is sensory, the external laryngeal nerve is motor, and the recurrent laryngeal nerve is mixed. The internal laryngeal nerve is sensory down to the vocal cords, the recurrent laryngeal nerve is sensory below the vocal cords, and there is overlap between the territories innervated by the two nerves at the vocal cords. All the intrinsic muscles of the larynx are supplied by the recurrent laryngeal nerve except for cricothyroid, which is supplied by the external laryngeal nerve. The detailed course of the vagus in the neck is described on page 556.
SUPERIOR LARYNGEAL NERVE The superior laryngeal nerve arises from the middle of the inferior vagal ganglion, and in its course receives one or more communications from the superior cervical sympathetic ganglion: most frequently, the connection is with the external laryngeal nerve. The superior laryngeal nerve divides into two branches, a smaller external and a larger internal branch, c.1.5 cm below the ganglion: rarely both branches may arise from the ganglion. Internal laryngeal nerve
The internal laryngeal nerve passes forwards c.7 mm before piercing the thyrohyoid membrane, usually at a higher level than the superior thyroid artery. It splits into superior, middle and inferior branches on entering the larynx. The superior branch supplies the mucosa of the piriform fossa. The large middle branch is distributed to the mucosa of the ventricle, specifically the quadrangular membrane, and therefore probably conveys the afferent component of the cough reflex. The inferior ramus is mainly distributed to the mucosa of the ventricle and subglottic cavity. On the medial wall of the piriform fossa, descending branches give twigs to the interarytenoid muscle and share communicating branches with the recurrent laryngeal nerve. The precise nature and function of these communicating nerves have yet to be determined. External laryngeal nerve
page 643 page 644
Figure 36.15 Nerve supply to the larynx. (Adapted from Oxford Textbook of Functional Anatomy, Vol 3 Head and Neck, MacKinnon P, Morris J (eds), 1990. By permission of Oxford University Press.)
The external laryngeal nerve continues downwards and forwards on the lateral surface of the inferior constrictor to which it contributes some small branches. In c.30% of cases, the nerve is located within the fibres of the constrictor muscle. It passes beneath the attachment of sternothyroid to the oblique line of the thyroid cartilage and supplies cricothyroid. A communicating nerve continues from the posterior surface of cricothyroid, crosses the piriform fossa and enters thyroarytenoid where it anastomoses with branches from the recurrent laryngeal nerve. It has been suggested that these communicating branches may provide both additional motor components to thyroarytenoid and sensory fibres to the mucosa in the region of the subglottis. The close relationship of the external laryngeal nerve to the superior thyroid artery puts the nerve at potential risk when the artery is clamped during thyroid lobectomy. The external laryngeal nerve is potentially at risk where it is either particularly close to the artery (in c.20% of cases), or where, instead of crossing the superior thyroid vessels c.1 cm or more above the superior pole of the gland, it actually passes below this point (in c.20% of cases).
RECURRENT LARYNGEAL NERVE The upper part of the recurrent laryngeal nerve has a close but variable relationship to the inferior thyroid artery: it may pass in front of, behind, or parallel to, the artery. The nerve enters the larynx either by passing deep to (in two-thirds
of cases) or between (in one-third of cases) the fibres of cricopharyngeus at its attachment to the lateral aspect of the cricoid cartilage. It supplies cricopharyngeus as it passes. At this point, the nerve is in intimate proximity to the posteromedial aspect of the thyroid gland. The main trunk divides into two or more branches, usually below the lower border of the inferior constrictor, although branching may occur higher up. The anterior branch is mainly motor and sometimes called the inferior laryngeal nerve, and the posterior branch is mainly sensory. The inferior laryngeal nerve passes posterior to the cricothyroid joint and its ligament. In this region, more often than not it may be covered by fibres of posterior cricoarytenoid. The first ramus of the main motor branch of the recurrent laryngeal nerve innervates posterior cricoarytenoid. It continues and innervates interarytenoid and then lateral cricoarytenoid, before it terminates in thyroarytenoid. Communications exist between the superior laryngeal nerve and its branches and the recurrent laryngeal nerve, and a communicating branch reaches the superior laryngeal nerve via the ansa Galeni. The recurrent laryngeal nerve does not always lie in a protected position in the tracheo-oesophageal groove, but may be slightly anterior to it (more often on the right), and it may be markedly lateral to the trachea at the level of the lower part of the thyroid gland. On the right the nerve is as often anterior to, or posterior to, or intermingled with, the terminal branches of the inferior thyroid artery. On the left the nerve is usually posterior to the artery, though occasionally anterior to it. The nerve may supply extralaryngeal branches to the larynx, which arise before it passes behind the inferior thyroid cornu. Outside its capsule the thyroid gland has a distinct covering of pretracheal fascia which splits into two layers at the posterior border of the gland (p. 542). One layer covers the entire medial surface of its lobe and, at or just above the isthmus, has a conspicuous thickening, the lateral ligament of the thyroid gland, which attaches the gland to the trachea and the lower part of the cricoid cartilage. The other layer is posterior; it passes behind the oesophagus and pharynx and is attached to the prevertebral fascia (p. 542). In this way, a compartment is formed on each side, lateral to the trachea and oesophagus, and the recurrent laryngeal nerve and terminal parts of the inferior thyroid artery lie in the fat of this compartment. The nerve may be lateral or medial to the lateral ligament of the thyroid gland, or sometimes may be embedded in it. An unusual anomaly that is of relevance to laryngeal pathology and surgery is the so-called 'non-recurrent' laryngeal nerve. In this condition, which has a frequency of between 0.3-1%, only the right side is affected and it is always associated with an abnormal origin of the right subclavian artery from the aortic arch on the left side. The right recurrent laryngeal nerve arises directly from the vagus nerve trunk high up in the neck and enters the larynx close to the inferior pole of the thyroid gland. If unrecognized, it may be susceptible to injury during surgery, as well as potentially being compressed by small tumours of the thyroid gland.
AUTONOMIC SUPPLY TO THE LARYNX
Parasympathetic, secretomotor fibres run with both the superior and recurrent laryngeal nerves to mucous glands throughout the larynx. Postganglionic sympathetic fibres run to the larynx with its blood supply, and have their origin in the superior and middle cervical ganglia.
VAGAL NERVE LESIONS AND RECURRENT LARYNGEAL NERVE PARALYSIS Unilateral complete palsy of the recurrent laryngeal nerve (more commonly on the left side due to its increased length) leads to isolated paralysis of all the laryngeal muscles on the affected side with the exception of cricothyroid (supplied by the external laryngeal nerve). The patient may be asymptomatic or have a hoarse, breathy voice. The hoarseness may be permanent or may become less severe with time as the opposite cord develops the ability to hyperadduct and appose the paralysed cord and thus close the glottis during phonation and coughing. Clinically, the position of the vocal cord in the acute phase after section of the recurrent laryngeal nerve is very variable. Stridor is more common after bilateral lesions but by no means the rule; indeed the cords may be sufficiently abducted for there to be little problem with airway obstruction, although the voice is always weaker in this situation. With chronic lesions the cords lie more widely separated, which leads to a weakened voice but a more secure upper airway. Variation in the position of paralysed vocal cords in more chronic lesions is probably more related to the degree of associated atrophy and fibrosis of paraglottic muscles than to the relative degrees of weakness and denervation of the apposing adductor and abductor muscle groups. For many years conventional wisdom was that movements of abduction were affected more than those of adduction when the recurrent laryngeal nerve was partially lesioned (Semon's law). However, it is now recognized that it is difficult to predict the effect that partial lesions of the recurrent laryngeal nerve will have on vocal cord position. Modern studies of human recurrent laryngeal nerve anatomy show that the fibres are randomly arranged in the nerve, which probably contributes to the difficulty in regaining co-ordinated vocal cord movement despite careful microsurgical re-anastomosis of a nerve that has been cut, e.g. after trauma or during laryngeal transplantation. Involvement of the superior laryngeal nerve in addition to the recurrent laryngeal nerve suggests a lesion proximal to the inferior (nodose) ganglion. This results in paralysis of all laryngeal musculature (including cricothyroid). The affected cord is paralysed and lies in the so-called 'cadaveric' position halfway between abduction and adduction. If the lesion is unilateral the voice is weak and hoarse, but if it is bilateral phonation is almost absent, the vocal pitch cannot be altered and the cough is weak and ineffective. UPDATE Date Added: 20 September 2005 Publication Services, Inc. Abstract: Laryngeal involvement by differentiated thryroid carcinoma. Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?
cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=15972190&query_hl=10 Laryngeal involvement by differentiated thyroid carcinoma. Zbaren P, Nuyens M, Thoeny HC et al: Am J Surg 190(1):153-155, 2005. There is debate as to the effect of lesions of the external laryngeal nerve. Complete section is most likely during the ligation of the vessels forming the vascular pedicle of the thyroid gland during thyroid lobectomy. This often but not always causes a weakening of the voice and mild hoarseness, and sometimes the effect is not really noticeable. Bilateral lesions produce these effects more often than not.
© 2008 Elsevier
ANATOMY OF SPEECH During its evolution, the larynx has developed a complex musculoskeletal structure and refined mechanism of neuromuscular control. These mechanisms allow the larynx to modify the expiratory stream to produce highly complex patterns of sound of varying loudness, frequency and duration, in addition to its sphincteric functions for the protection and control of respiratory activities. The ability to execute these complex movements depends largely on the cerebral hemispheres in which specific parts are involved in the motor aspects of language, such as speech and writing, and sensory manifestations of language, including reading and understanding the spoken word.
SPEECH PRODUCTION The production of any sound requires a source of energy: for the human voice this is the momentum of the expired air. Structures which can vibrate are also required to produce noise (phonation) in the majority of speech sounds in English. A series of resonators that can interrupt, dampen or amplify certain sound frequencies modify the exhaled airstream into its final form (articulation). page 644 page 645
During speech, it is necessary to exert a force sufficient to produce a pressure in the order of 7 cm H2O beneath the vocal cords (subglottal pressure) to make them vibrate. The subglottal pressure can be increased up to 60 cm H2O in loud speech or singing. Variations in subglottal pressure mainly affect the loudness but can also influence the pitch of the voice. Vibration of air in the vocal tract superior to the larynx is due primarily to the activity of the vocal cords. Resonance in the larynx above the vocal cords and in the pharynx, mouth and nasal cavity selectively amplifies or dampens certain harmonics. Interruption of airflow through the resonators by articulatory organs, (the lips, tongue, teeth and soft palate), modifies the egressive airstream to produce distinct units of speech (phonemes).
MUSCULAR CONTROL OF THE AIRSTREAM The expiratory force used in speech is produced by the controlled relaxation of the respiratory muscles to counteract passive elastic recoil of the lungs and thoracic wall. After a rapid inspiration, the intercostal muscles are relaxed slowly to prolong the expiratory phase of respiration as long as the desired utterance. The internal intercostal muscles contract towards the end of the utterance to maintain subglottal pressure as lung air volume nears its resting expiratory level. These muscles also make small contractions to vary the expiratory force and hence the loudness of individual words or phrases for emphasis. The anterior abdominal muscles which are used in prolonged and forced expiration, and in some subjects at the end of quiet respiration, may be involved in speech, especially in singing, shouting and in attempts to speak without the pause necessary for inspiration. Contrary to popular belief, the diaphragm plays little part in regulation of expiratory force. Unlike the intercostal muscles, the diaphragmatic musculature is sparsely supplied with muscle spindles, and therefore control of the diaphragm is poorly regulated: minute changes can be effected more
successfully using the intercostal and anterior abdominal muscles. The pulmonic airstream is the source of energy in normal speech. However, after removal of the larynx, e.g. following laryngeal cancer, patients can be taught to swallow air, store it in a segment of the oesophagus and use it as the energy source (oesophageal speech). Speech in these circumstances tends to have a belching quality and may be badly phrased. Laryngectomy patients always produce phrases that are shorter than normal, and so prostheses incorporating valves and surgical shunts are often inserted to provide a larger egressive airstream by diverting air from the respiratory tract into the oesophagus.
PHONATION The default position of the rima glottidis is open to maintain patency of the airway during respiration. In quiet respiration, the anterior intermembranous part of the rima glottidis is triangular when viewed from above (see Fig. 36.10). Its apex is anterior and its base posterior, and it is represented by an imaginary line c.8 mm long connecting the anterior ends of the arytenoid vocal processes. The intercartilaginous part between the medial surfaces of the arytenoids is rectangular as the two vocal processes lie parallel to each other. During forced respiration, the rima glottidis is widened and the vocal cords are fully abducted to increase the airway. The arytenoid cartilages rotate laterally, and this moves their vocal processes apart, and converts the rima glottidis into a diamond shape in which both intermembranous and intercartilaginous parts are triangular. The greatest width of the rima glottidis is at the point of the attachments of the vocal cords to the vocal processes. Phonation is necessary for all vowels and a number of consonants in spoken English. Preparatory to phonation, both the intermembranous and intercartilaginous parts of the glottis are adducted, reducing the space between the vocal cords to a linear chink. This position of the vocal cords is achieved by adduction and medial rotation of the arytenoids at the cricoarytenoid joints. Contraction of the lateral cricoarytenoids produces rotatory movements and the transverse arytenoids bring about gliding. The mucous membrane covering the interarytenoid muscles, the interarytenoid fold, intrudes into the larynx when these muscles adduct the arytenoids, and so aids closure of the intercartilaginous part of the rima glottidis. The vocal cords are also tensed, an essential prerequisite for vibration. If sufficient subglottal pressure is generated below the vocal cords after closure and tensing of the glottis, they will be forced apart. The fall in pressure due to their separation and the continued adductor muscle activity result in closure of the glottis. Closure is assisted by the suction effect of the fall in pressure (the Bernoulli effect). During the rapid opening and closing of the glottis, the vocal cords exhibit longitudinal and transverse waves due to their structure. These add harmonics to the fundamental frequency of the tone produced. The fundamental frequency of the human voice is determined by the resting length of the vocal cords. It varies with age and sex. The frequency range of human speech is from 60-500Hz with an average of c.120Hz in males, 200Hz in females and 270Hz in children. Variations in frequency (pitch changes) during an utterance are determined by the complex interrelationships between length,
tension and thickness of the vocal cords. It must be emphasized that one of these variables cannot be altered without affecting the other two parameters to some extent. Gross changes to the vocal cords demonstrate the effects of these variables. Inflamed and swollen vocal cords are much thicker than normal and result in a hoarse voice. At puberty, growth of the thyroid cartilage in males lengthens the vocal cords and lowers the fundamental frequency: the voice 'breaks' as a result. During panic, the vocal cords may be tensed, which means that the cry for help is a high-pitched squeak. Pitch is altered by increasing length. At first sight this may seem counterintuitive, but, as the vocal cords are lengthened, there will be a consequent thinning and change in tension. Although an analogy is often drawn between the vocal cords and vibrating strings, a better analogy is a rubber band: if a rubber band is lengthened, the tension will increase, but the thickness will decrease. The vocal cords may be lengthened by up to 50% of their resting length. It is likely that the initial pitch setting is achieved by action of the cricothyroids, and that fine adjustments can then be made using the vocalis muscles. Paralysis of both cricothyroids, which is usually associated with damage to the origins of the superior laryngeal nerve in the vagal nuclei in brainstem stroke, results in permanent hoarseness and inability to vary the pitch of the voice. It is important to remember that once the vocal cords are set in motion they will deviate from their original setting as they vibrate. Auditory feedback of the sounds produced is used to make minute compensatory adjustments to length, tension and thickness to maintain a constant pitch. Any lengthening of the vocal cords tends to thin them. The thickness can be increased by the vocalis part of thyroarytenoid. Because of its attachment to the vocal ligament, vocalis shortens and relaxes the vocal cords while increasing their thickness. Changes in tension of the vocal cords are produced by the same muscles that change their length, namely cricothyroid, posterior cricoarytenoid and vocalis, probably acting isometrically. In whispering, the intermembranous glottis is closed, but the intercartilaginous part remains widely patent, so that air escapes freely and phonation ceases.
FREQUENCY CHARACTERISTICS OF SPEECH page 645 page 646
The sound produced by the mechanism described above is not a pure tone because several harmonics at multiples of the fundamental frequency will also be generated. Harmonics give a note of a particular frequency its defining characteristics. An 'A' played on an oboe or violin is immediately recognizable because of the different harmonics generated by the design of the instrument. In the human vocal tract, the fundamental frequency and its harmonics are transmitted to the column of air which extends from the vocal cords to the exterior, mainly through the mouth. A significant air stream also passes through the nasal cavities during articulation of the nasal consonants /m/, /n/ and /E/ ('!g'), as in 'mincing' (/mNnsNE/), when the soft palate is depressed to allow air into the nasopharynx. The supraglottal tract acts as a selective resonator but unlike, for example, an organ pipe, it is variable in length, shape and volume. These
parameters may be altered by the muscles of the pharynx, soft palate, fauces, tongue, cheeks and lips. The relative positions of the upper and lower teeth, which are determined by the degree of opening and protrusion or retraction of the mandible, also have an effect. In addition, the tension of the walls of the column can be altered, especially in the pharynx. The result is that the fundamental frequency (pitch) and harmonics produced by the passage of air through the glottis are modified by changes in the supraglottal tract. Harmonics may be amplified, or dampened. The fundamental frequency and its associated harmonics may also be raised or lowered by appropriate elevation or depression of the hyoid bone and the larynx as a unit by the selective actions of the extrinsic laryngeal musculature, namely, the inferior pharyngeal constrictor, the infrahyoid and suprahyoid muscles. Effectively, these movements shorten or lengthen the resonating column, and to some extent also alter the geometry of the walls of the air passages. Analysis of the human voice shows that it has a very similar pattern of harmonics for all fundamental frequencies, determined by the vocal tract acting as a selective filter and resonator. This is necessary for the maintenance of a constant quality of voice without which intelligibility would be lost. For example, recorded speech played back without its harmonics is completely unintelligible. Each human voice is unique and recognizable as belonging to a particular person because of its special characteristics. Indeed, it has been suggested that the unique frequency spectrum of each individual voice could be used for personal identification.
ARTICULATION During articulation the egressive airstream is given a rapidly changing specific quality by the articulatory organs, the tongue, palate, teeth, lips and nasal cavity. The discipline of phonetics primarily deals with the way in which speech sounds are produced, and consequently with the analysis of the mode of production of phonemes by the vocal apparatus. In order to analyse the way in which the articulators are used in different speech sounds, words are broken down into units called phonemes, which are defined as the minimal sequential contrastive units used in any language. The human vocal tract can produce many more phonemes than are employed in any one language. Not all languages have the same phonemes. Even within the same language, the phonemes can vary in different parts of the same country and in other countries where that language is also spoken. Anyone who has tried to learn a foreign language knows how difficult it can be to reproduce phonemes that are not used in their native speech because such phonemes require unfamiliar positioning of the speech organs. A native speaker of any language can quickly recognize the origins of anyone attempting to use their language as a second language. The second language speaker will usually speak it with an accent characteristic of their own first language because they are using the familiar configurations of their vocal tract for each phoneme instead of the correct positioning.
PRODUCTION OF VOWELS All vowel sounds require phonation by vibration of the vocal cords. Each vowel
sound has its own characteristic higher harmonics (frequency spectrum) because the pharyngeal and oral cavities act as selective resonators to amplify or dampen different harmonics. These frequencies are always higher multiples of the fundamental frequencies and are called formants. The sounds of the different vowels are determined by the shape and size of the mouth, and the positions of the tongue and lips are the most important variables. The tongue may be placed high or low (close and open vowels), or further forwards or back (front and back vowels) and the lips may be rounded or spread.
PRODUCTION OF CONSONANTS Consonants may be defined as speech sounds that are determined by the closure or narrowing of some part of the vocal tract to stop or perturb the airflow. If a consonant is sounded without vibration of the vocal cords, it is defined as unvoiced, while if phonation is a component, it is voiced. There are many pairs of unvoiced and voiced consonants formed by using exactly the same parts of the speech organs, for example (with the voiceless consonant first in each case) /p/ and /b/, /t/ and /d/, /k/ and /g/, /f/ and /v/. Probably the most graphical example is /s/ and /z/. If the larynx is loosely palpated while making a sustained unvoiced 'ssssss' sound, no vibration is felt but if the 'ssssss' is commuted into a prolonged voiced 'zzzzzz' then vibration in the larynx should be readily detectable. The position of the tongue and other articulators is exactly the same for both /s/ and /z/, only the presence or absence of phonation making a difference between them. Consonants are further classified according to the parts of the speech organs involved, i.e. lips, tongue, teeth and palate, and their positions. For example, /p/ and /b/ are bilabial, the upper and lower lips being approximated to produce them; whereas /f/ and /v/ are labiodental, the lower lip being raised to contact the upper incisor teeth. For the purposes of phonetic analysis, the articulatory organs can be subdivided into a number of regions. For example, the tongue is divided into the tip, anterior edge, the front part of the dorsum, the centre and back parts of the remaining dorsum, and a most posterior part (the root). In some cases, these bear no obvious relationship to the anatomical parts of the tongue but they are useful in describing the part of the dorsum of the tongue contacting other areas of the mouth. Consonants may be further subclassified by their mechanism of production. A plosive is defined as a consonant requiring sudden explosive release of air e.g. /p/ and /b/. A fricative is a rustling of the breath due to friction as the air column passes through a considerable narrowing of the oral cavity. During articulation of /f/ and /v/, for example, the lower lip is approximated to the upper teeth, without complete closure. An affricate is defined as a plosive followed by a fricative as /ch/ in chain; the 'ch' component is phonetically a plosive /t/ followed by fricative /sh/. A full phonetic description of the phoneme /b/ is a voiced bilabial plosive. REFERENCES Berkovitz BKB, Moxham BJ, Hickey S 2000 The anatomy of the larynx. In: Ferlito A (ed) Diseases of the Larynx. London: Chapman and Hall: 25-44. Durham FC, Harrison TS 1962 The surgical anatomy of the superior laryngeal nerve. Surg Gynecol Obstet
118: 33-44. Erkki A, Pitkanen R, Suominen H 1987 Observations on the structure and the biomechanics of the cricothyroid articulation. Acta Otolaryngol (Stockh) 103: 117-26. Friedman M, Toriumi DM, Grybauskas V, Katz A 1986 Nonrecurrent laryngeal nerves and their clinical significance. Laryngoscope 96: 87-90. Medline Similar articles Full article Kirchner JA, Carter D 1987 Intralaryngeal barriers to the spread of cancer. Acta Otolaryngol (Stockh) 103: 503-13 Munir Turk L, Hogg DA 1993 Age changes in the human laryngeal cartilages. Clin Anat 6: 154-62. Provides data concerning the distribution of areas of calcification that occur with age within the cartilages of the larynx. Pracy R 1983 The infant larynx. J Laryngol Otol 97: 933-47. Draws attention to features that differ from the adult condition and that may have clinical significance. Medline Similar articles Reidenbach MM 1995 Normal topography of the conus elasticus. Anatomical bases for the spread of laryngeal cancer. Surg Radiol Anat 17: 107-11. Medline Similar articles Full article Reidenbach MM 1996a The periepiglottic space: topographic relations and histological organisation. J Anat 188: 173-82. Medline Similar articles Reidenbach MM 1996b The paraglottic space and transglottic cancer: anatomic considerations. Clin Anat 9: 244-51. Medline Similar articles Full article Reidenbach MM 1998 Subglottic region: normal topography and possible clinical implications. Clin Anat 11:9-21. The four papers by Reidenbach are based on serial reconstruction of sections derived from plastinated material. Medline Similar articles Full article Sato I, Shimada K 1995 Arborization of the inferior laryngeal nerve and internal nerve on the posterior surface of the larynx. Clin Anat 8: 379-87. Discusses the possible clinical significance of connections between the inferior (recurrent) laryngeal nerve and the internal laryngeal nerve. Medline Similar articles Full article Welsh LW, Welsh JJ, Rizzo TA 1983 Laryngeal spaces and lymphatics: current anatomic concepts. Ann Otol Rhinol Laryngol Suppl 105: 19-31. Describes the 'tissue spaces' and lymphatic drainage of the larynx and their importance in determining the route of spread of tumours. Medline Similar articles
© 2008 Elsevier
37 NECK AND UPPER AERODIGESTIVE TRACT Development of the pharynx, larynx and oesophagus PHARYNX Pharyngeal endoderm is in contact with and develops from, mesenchyme and epithelia from neural crest, paraxial mesenchyme of the somitomeres, somites, lateral plate mesenchyme (which at this level is unsplit), cleft ectoderm, general endothelium and the outflow tract of the heart. The mechanism of formation of the pharynx, which is complex, is known to be intimately related to the development of the viscerocranium (Chapter 26) and laryngeal cartilages (Chapters 26, 37): it is likely that some of the processes are regulated by Hox gene expression (Fig. 26.5). The distal foregut, and the midgut and hindgut consist of a serous or adventitial layer, a layer of splanchnopleuric mesenchyme - which is derived from the splanchnopleuric coelomic epithelium - and an inner endodermal epithelium. In contrast, the proximal foregut contains a mixture of striated muscle in the upper pharynx, which blends with the smooth muscle of the gut wall within the territory of middle third of the oesophagus. A novel mesenchymal population which develops in a rostrocaudal and lateromedial sequence has been identified at the interface between the endoderm and the paraxial mesenchyme of the somitomeres and occipital somites. It starts as a sparse layer, becomes denser prior to the formation of endothelial networks, and ultimately forms a fenestrated mesenchymal monolayer between developing blood vessels and the endoderm. Later it expands between the notochord and the roof of the foregut, and may participate in the formation of pharyngeal and oesophageal smooth muscle and connective tissues.
NASOPHARYNX The nasopharynx represents the most rostral portion of the original stomodeum derived from endoderm. The dorsal part of the ectodermal aspect of the first (maxillomandibular) arch contributes to the formation of the lateral wall of the nasopharynx in front of the orifice of the pharynogtympanic tube. This region is the zone of transition of ectoderm to endoderm which passes from the junction of the anterior two-thirds and posterior one-third of the tongue to the developing sella turcica in the sphenoid bone.
OROPHARYNX The development of the palate subdivides the primitive pharynx so that the original arches and pouches are widely separated. The site of the second arch is partly indicated by the palatoglossal arch; however, the forward growth of the third arch obliterates the middle portion of the second arch and separates its dorsal and ventral ends. The second pharyngeal pouch is represented by the intratonsillar cleft, around which the tonsil develops. The third arch forms the lateral glossoepiglottic fold. The ventral ends of the fourth arches fuse with the caudal part of the hypobranchial eminence and so contribute to the formation of the
epiglottis. The adjoining portion becomes connected to the arytenoid swelling and may be identified in the aryepiglottic fold.
LARYNGOPHARYNX After the caudal part of the hypobranchial eminence has separated from the pharyngeal (posterior) part of the tongue (Figs 26.2, 34.7), it is in continuity with two linear ridges which appear in the ventral wall of the pharynx. Together these structures form an inverted U, sometimes regarded as an independent formation, the furcula (of His). The ridges have been identified as the sixth arches, and are placed very obliquely owing to the shortness of the pharyngeal floor compared with the greater extent of the roof. They are carried downwards on the ventral wall of the foregut and bound the median laryngotracheal groove, from which the lower part of the larynx, the trachea, bronchi and lungs are developed.
LYMPHOID TISSUE IN THE PHARYNX The pharyngeal endoderm gives rise to a series of lymphoid organs, namely, the adenoid (pharyngeal tonsil), lateral pharyngeal lymphoid bands, tubal tonsil, lingual tonsils and palatine tonsils, and the thymus (p. 617).
NEONATAL PHARYNX In the neonate the pharynx is one-third of the relative length in the adult. The nasopharynx is a narrow tube which curves gradually to join the oropharynx without any sharp junctional demarcation. An oblique angle is formed at this junction by 5 years of age and in the adult the nasopharynx and oropharynx join at almost a right angle.
© 2008 Elsevier
PHARYNX Pharyngeal endoderm is in contact with and develops from, mesenchyme and epithelia from neural crest, paraxial mesenchyme of the somitomeres, somites, lateral plate mesenchyme (which at this level is unsplit), cleft ectoderm, general endothelium and the outflow tract of the heart. The mechanism of formation of the pharynx, which is complex, is known to be intimately related to the development of the viscerocranium (Chapter 26) and laryngeal cartilages (Chapters 26, 37): it is likely that some of the processes are regulated by Hox gene expression (Fig. 26.5). The distal foregut, and the midgut and hindgut consist of a serous or adventitial layer, a layer of splanchnopleuric mesenchyme - which is derived from the splanchnopleuric coelomic epithelium - and an inner endodermal epithelium. In contrast, the proximal foregut contains a mixture of striated muscle in the upper pharynx, which blends with the smooth muscle of the gut wall within the territory of middle third of the oesophagus. A novel mesenchymal population which develops in a rostrocaudal and lateromedial sequence has been identified at the interface between the endoderm and the paraxial mesenchyme of the somitomeres and occipital somites. It starts as a sparse layer, becomes denser prior to the formation of endothelial networks, and ultimately forms a fenestrated mesenchymal monolayer between developing blood vessels and the endoderm. Later it expands between the notochord and the roof of the foregut, and may participate in the formation of pharyngeal and oesophageal smooth muscle and connective tissues.
NASOPHARYNX The nasopharynx represents the most rostral portion of the original stomodeum derived from endoderm. The dorsal part of the ectodermal aspect of the first (maxillomandibular) arch contributes to the formation of the lateral wall of the nasopharynx in front of the orifice of the pharynogtympanic tube. This region is the zone of transition of ectoderm to endoderm which passes from the junction of the anterior two-thirds and posterior one-third of the tongue to the developing sella turcica in the sphenoid bone.
OROPHARYNX The development of the palate subdivides the primitive pharynx so that the original arches and pouches are widely separated. The site of the second arch is partly indicated by the palatoglossal arch; however, the forward growth of the third arch obliterates the middle portion of the second arch and separates its dorsal and ventral ends. The second pharyngeal pouch is represented by the intratonsillar cleft, around which the tonsil develops. The third arch forms the lateral glossoepiglottic fold. The ventral ends of the fourth arches fuse with the caudal part of the hypobranchial eminence and so contribute to the formation of the epiglottis. The adjoining portion becomes connected to the arytenoid swelling and may be identified in the aryepiglottic fold.
LARYNGOPHARYNX
After the caudal part of the hypobranchial eminence has separated from the pharyngeal (posterior) part of the tongue (Figs 26.2, 34.7), it is in continuity with two linear ridges which appear in the ventral wall of the pharynx. Together these structures form an inverted U, sometimes regarded as an independent formation, the furcula (of His). The ridges have been identified as the sixth arches, and are placed very obliquely owing to the shortness of the pharyngeal floor compared with the greater extent of the roof. They are carried downwards on the ventral wall of the foregut and bound the median laryngotracheal groove, from which the lower part of the larynx, the trachea, bronchi and lungs are developed.
LYMPHOID TISSUE IN THE PHARYNX The pharyngeal endoderm gives rise to a series of lymphoid organs, namely, the adenoid (pharyngeal tonsil), lateral pharyngeal lymphoid bands, tubal tonsil, lingual tonsils and palatine tonsils, and the thymus (p. 617).
NEONATAL PHARYNX In the neonate the pharynx is one-third of the relative length in the adult. The nasopharynx is a narrow tube which curves gradually to join the oropharynx without any sharp junctional demarcation. An oblique angle is formed at this junction by 5 years of age and in the adult the nasopharynx and oropharynx join at almost a right angle.
© 2008 Elsevier
LARYNX The larynx is probably formed from the lower two pharyngeal arches, although the exact contribution of the sixth arch is still not clear. It lies at the cranial end of the laryngotracheal groove, where it is bounded laterally by the ventral ends of the sixth arches and ventrally by the caudal part of the hypobranchial eminence (Figs 34.7A&B, 26.2). Paired arytenoid swellings appear in the ventral ends of the sixth arches, one on each side of the cranial end of the groove. As they enlarge they approximate to each other and to the caudal part of the hypobranchial eminence where the epiglottis develops. The opening into the larynx, at first a vertical slit, is converted into a T-shaped cleft by the enlargement of the arytenoid swellings. The vertical limb of the T lies between the two swellings and its horizontal limb lies between them and the epiglottis. The arytenoid swellings differentiate into the arytenoid and corniculate cartilages (Fig. 26.7), and the ridges that join them to the epiglottis become the definitive aryepiglottic folds within which the cuneiform cartilages differentiate from the epiglottis. The thyroid cartilage develops from the ventral ends of the cartilages of the fourth, or fourth and fifth, pharyngeal arches. The cartilage appears as two lateral plates, each chondrified from two centres and united in the midventral line by a fibrous membrane within which an additional centre of chondrification develops. The cricoid cartilage arises from two cartilaginous centres, which soon unite ventrally, gradually extend and ultimately fuse on the dorsal surface of the tube as the cricoid lamina (p. 634).
© 2008 Elsevier
OESOPHAGUS page 647 page 648
An anterior midline lung bud diverticulum first appears at stage 12. By stage 14 the cells of the splanchnopleuric mesenchyme that surrounds the developing trachea and oesophagus are sufficiently diverged in their inductive ability to promote the characteristics of the two different tubes. During stage 17 the oesophageal epithelium is surrounded by a wide submucosal zone and muscular coats can be distinguished. The oesophagus can be distinguished from the stomach at stage 13 (embryo 5 mm). It elongates during successive stages and its absolute length increases more rapidly than that of the embryo as a whole. The oesophagus is invested by splanchnopleuric mesenchyme cranially posterior to the developing trachea, and more caudally between the developing lungs and pericardioperitoneal canals posterior to the pericardium. Caudal to the pericardium, the pregastric segment of the oesophagus acquires a short thick dorsal meso-oesophagus from splanchnopleuric mesenchyme, and a short ventral mesogastrium from the cranial stratum of the mesenchyme of the septum transversum (Ch. 90). The oesophagus has a limited relationship to primary coelomic epithelium. However, it is related to secondary extensions from the primary coelom via the para-oesophageal right and left pneumatoenteric recesses (Fig. 90.7), the oblique sinus of the pericardium, and, in the lower thorax, to the mediastinal pleura. By stage 15 (week 5), the mucosa consists of two layers of cells: their proliferation never occludes the lumen. The mucosa is lined with a ciliated epithelium at 10 weeks, and a stratified squamous epithelium at the end of the 5th month. Occasional patches of ciliated epithelium have been described at birth. Circular muscle can be seen at stage 15, but longitudinal muscle has not been identified until stage 21. Neuroblasts can be demonstrated at relatively early developmental stages, thus myenteric plexuses display cholinesterase activity by 9.5 weeks and ganglion cells differentiate by 13 weeks. The oesophagus may be capable of peristalsis in the first trimester. At birth the oesophagus extends 8-10 cm from the cricoid cartilage to the gastric cardiac orifice. It commences and ends 1-2 vertebrae higher than in the adult, and extends from between the fourth to the sixth cervical vertebrae to the level of the ninth thoracic vertebra. Its average diameter is 5 mm and it possesses the constrictions seen in the adult. The narrowest constriction is at its junction with the pharynx, where the inferior pharyngeal constrictor functions to constrict the lumen, and in this region it may be traumatized with instruments or catheters. Peristalsis along the oesophagus and at the lower oesophageal sphincter is immature at birth, which probably accounts for the frequent regurgitation of food that occurs in the newborn period. The pressure at the lower oesophageal sphincter approaches that of the adult at 36 weeks of age.
© 2008 Elsevier
38 EAR AND AUDITORY AND VESTIBULAR APPARATUS External and middle ear The ear can be subdivided into the external, middle and internal ear, all of which are associated with, or lie within, the temporal bone on the lateral aspect of the skull. Each ear is a distance receptor for the collection, conduction, modification, amplification and analysis of complex waves of sound. It also contains the receptors for hearing and balance.
© 2008 Elsevier
EXTERNAL EAR The external ear consists of the auricle, or pinna, and the external acoustic meatus. The auricle projects from the side of the head to collect sound waves, and the meatus leads inwards from the auricle to conduct vibrations to the tympanic membrane. These structures do not act merely as a simple ear-trumpet: they are the first of a series of stimulus modifiers in the auditory apparatus.
Skin The skin of the auricle continues into the external acoustic meatus and covers the external surface of the tympanic membrane. It is thin, has no dermal papillae, and is closely adherent to the cartilaginous and osseous parts of the tube. Inflammation here is therefore very painful. The thick subcutaneous tissue of the cartilaginous part of the meatus contains numerous ceruminous glands which secrete ear wax, or cerumen. Their coiled tubular structure resembles that of sweat glands. The secretory cells are columnar when active, but cuboidal when quiescent. They are covered externally by myoepithelial cells. Ducts open either on to the epithelial surface or into the nearby sebaceous gland of a hair follicle. Cerumen prevents the maceration of meatal skin by trapped water. Overproduction or accumulation of wax may completely block the meatus or obstruct the vibration of the tympanic membrane. Although ceruminous glands and hair follicles are largely limited to the cartilaginous meatus, a few small glands and fine hairs also occur in the roof of the lateral part of the osseous meatus. The warm humid environment of the relatively enclosed meatal air aids the mechanical responses of the tympanic membrane.
Auricle (pinna) (Fig. 38.1) The lateral surface of the auricle is irregularly concave, faces slightly forwards and displays numerous eminences and depressions. Its prominent curved rim, or helix, usually bears posterosuperiorly a small auricular tubercle, which is quite pronounced around the sixth month of intrauterine life. The antihelix is a curved prominence, parallel and anterior to the posterior part of the helix: it divides above into two crura which flank a depressed triangular fossa. The curved depression between the helix and antihelix is the scaphoid fossa. The antihelix encircles the deep, capacious concha of the auricle, which is incompletely divided by the crus or anterior end of the helix. The conchal area above this, the cymba conchae, overlies the suprameatal triangle of the temporal bone, which can be felt through it, and which overlies the mastoid antrum. The tragus is a small curved flap below the crus of the helix and in front of the concha: it projects posteriorly, partly overlapping the meatal orifice. The antitragus is a small tubercle opposite the tragus and separated from it by the intertragic incisure or notch. Below it is the lobule, composed of fibrous and adipose tissues. It is soft, unlike the majority of the auricle which is supported by elastic cartilage and is firm. The cranial surface of the auricle presents elevations which correspond to the depressions on its lateral surface, and after which they are named (e.g. eminentia conchae, eminentia fossae triangularis).
Figure 38.1 Lateral surface of auricle. (By permission from Berkovitz BKB, Moxham BJ 2002 Head and Neck Anatomy. London: Martin Dunitz.)
Cryptotia Cryptotia or 'pocket ear' is an abnormality of the auricle where the upper part of the auricle is buried beneath the temporal skin. It is fairly easily mobilized into a more normal position surgically, and the posterior surface of the pinna is covered with skin from the post-auricular temporal skin. Stahl's deformity Stahl's deformity is a congenital deformity of the auricle where the helix is flattened and the upper crus of the antihelix is duplicated, which produces a ridge of cartilage running from the antihelix to the rim of the helix. This causes a pointing of the ear and a reversal of the normal concavity of the scaphoid fossa. It is easily corrected in the first 6 weeks of life by the application of external moulds, but thereafter the cartilage may be too stiff and a formal surgical correction may be necessary. Pre-auricular sinus
page 649 page 650
The auricular tubercles, the embryological precursors of the auricle, arise around the dorsal end of the embryonic first branchial cleft from the first and second branchial arches. They fuse to form the auricle and surround the dorsal end of the first branchial cleft from which the external acoustic meatus arises. Sinuses and cysts are often found just anterior to the root of the helix, near to the point of fusion of the tubercles derived from the first branchial arch and those derived from the second branchial arch. There is debate as to whether the abnormalities are epithelial inclusions between the tubercles or remnants of the first branchial cleft. The sinuses may be simple pits or complex branching sinuses, and they occasionally extend deeply towards the external acoustic meatus so that they lie
close to the facial nerve. Clinically they may become chronically infected and require surgical excision: this may be technically demanding surgery given the close proximity to the facial nerve.
CARTILAGINOUS FRAMEWORK OF THE AURICLE The auricle is a single thin plate of elastic fibrocartilage covered by skin, its surface moulded by eminences and depressions (Fig. 38.2). It is connected to the surrounding parts by ligaments and muscles, and is continuous with the cartilage of the external acoustic meatus. There is no cartilage in the lobule or between the tragus and the crus of the helix, where the gap is filled by dense fibrous tissue. Anteriorly, where the helix curves upwards, there is a small cartilaginous projection, the spine of the helix. Its other extremity is prolonged inferiorly as the tail of the helix and it is separated from the antihelix by the fissura antitragohelicina. The cranial aspect of the cartilage bears the eminentia conchae and eminentia scaphae, which correspond to the depressions on the lateral surface. The two eminences are separated by a transverse furrow, the sulcus antihelicis transversus, which corresponds to the inferior crus of the antihelix on the lateral surface. The eminentia conchae is crossed by an oblique ridge, the ponticulus, for the attachment of auricularis posterior. There are two fissures in the auricular cartilage, one behind the crus helicis and another in the tragus.
LIGAMENTS OF THE AURICLE There are two sets of ligaments associated with the auricle. Extrinsic ligaments connect the auricle with the temporal bone, and intrinsic ligaments connect individual auricular cartilages. There are two extrinsic ligaments, anterior and posterior. The anterior ligament extends from the tragus and the spine of the helix to the root of the zygomatic process of the temporal bone. The posterior ligament passes from the posterior surface of the concha to the lateral surface of the mastoid process. The chief intrinsic ligaments are first, a strong fibrous band which passes from the tragus to the helix completing the meatus anteriorly and forming part of the boundary of the concha; and second, a band which passes between the antihelix and the tail of the helix. Less prominent bands also exist on the cranial aspect of the auricle.
AURICULAR MUSCLES Extrinsic auricular muscles connect the auricle to the skull and scalp and move the auricle as a whole, and intrinsic auricular muscles connect the different parts of the auricle.
Figure 38.2 The cranial surface of the left auricular cartilage.
EXTRINSIC MUSCLES The extrinsic auricular muscles are the auriculares anterior, superior and posterior. The auricularis anterior, the smallest of the three, is a thin fan of pale fibres, which arises from the lateral edge of the epicranial aponeurosis: its fibres
converge to insert into the spine of the helix. The auricularis superior, the largest of the three, is also thin and fan-shaped and converges from the epicranial aponeurosis via a thin, flat tendon to attach to the upper part of the cranial surface of the auricle. The auricularis posterior consists of two or three fleshy fasciculi which arise by short aponeurotic fibres from the mastoid part of the temporal bone and insert into the ponticulus on the eminentia conchae. Vascular supply The arterial supply of the extrinsic auricular muscles is derived mainly from the posterior auricular artery. Innervation Auriculares anterior and superior are supplied by temporal branches of the facial nerve and auricularis posterior is supplied by the posterior auricular branch of the facial nerve. Actions In man these muscles have very little obvious effect. Auricularis anterior draws the auricle forwards and upwards; auricularis superior elevates the auricle slightly; auricularis posterior draws the auricle back. Despite the paucity of auricular movement, auditory stimuli may evoke patterned responses from these small muscles and electromyography can detect the 'crossed acoustic response' which is elicited by this means in investigative clinical neurology.
INTRINSIC MUSCLES The intrinsic auricular muscles are helicis major and minor, tragicus, antitragicus, transversus auriculae and obliquus auriculae. Helicis major is a narrow vertical band on the anterior margin of the helix, which passes from its spine to its anterior border, where the helix is about to curve back. Helicis minor is an oblique fasciculus, which covers the crus helicis. Tragicus is a short, flattened, vertical band on the lateral aspect of the tragus. Antitragicus passes from the outer part of the antitragus to the tail of the helix and the antihelix. Transversus auriculae, on the cranial aspect of the auricle, consists of scattered fibres, partly tendinous, partly muscular, which extend between the eminentia conchae and the eminentia scaphae. Obliquus auriculae, also on the cranial aspect of the auricle, consists of a few fibres which extend from the upper and posterior parts of the eminentia conchae to the eminentia scaphae. Vascular supply The intrinsic auricular muscles are supplied by branches of the posterior auricular and superficial temporal arteries. Innervation The intrinsic auricular muscles on the lateral aspect of the auricle are innervated by the temporal branches of the facial nerve, and those on the cranial aspect of the auricle are innervated by the posterior auricular branch of the facial nerve. Actions The intrinsic muscles modify auricular shape minimally, if at all, in most human ears. However, rare individuals can modify the shape and position of their external ears.
VASCULAR SUPPLY AND LYMPHATIC DRAINAGE Arteries The posterior auricular branch of the external carotid artery supplies three or four branches to the cranial surface of the auricle: twigs from these arteries reach the lateral surface, some through fissures in the cartilage, others round the margin of the helix. The posterior auricular artery ascends between the parotid gland and the styloid process to the groove between the auricular cartilage and mastoid process. The auricle is also supplied by anterior auricular branches of the superficial temporal artery, which are distributed to its lateral surface, and by a branch from the occipital artery.
Veins Auricular veins correspond to the arteries of the auricle. Arteriovenous anastomoses are numerous in the skin of the auricle and are thought to important in the regulation of core temperature. Lymphatic drainage Auricular lymphatics drain into the parotid lymph nodes, especially the node in front of the tragus; the upper deep cervical lymph nodes; and the mastoid lymph nodes. page 650 page 651
INNERVATION The sensory innervation of the auricle is complex and not fully determined. This is perhaps because the external ear represents an area where skin originally derived from a branchial region meets skin originally derived from a postbranchial region. The sensory nerves involved are: the great auricular nerve, which supplies most of the cranial surface and the posterior part of the lateral surface (helix, antihelix, lobule); the lesser occipital nerve, which supplies the upper part of the cranial surface; the auricular branch of the vagus, which supplies the concavity of the concha and posterior part of the eminentia; the auriculotemporal nerve, which supplies the tragus, crus of the helix and the adjacent part of the helix; and the facial nerve, which together with the auricular branch of the vagus probably supplies small areas on both aspects of the auricle, in the depression of the concha, and over its eminence. The details of the cutaneous innervation derived from the facial nerve require further clarification. It is possible that, as the auricular branch of the vagus traverses the temporal bone and crosses the facial canal c.4 mm above the stylomastoid foramen, it contributes an ascending branch to the facial nerve, which carries it to the pinna.
External acoustic meatus (Figs 38.3, 38.4) The external acoustic meatus extends from the concha to the tympanic membrane. Its length is c.2.5 cm from the floor of the concha and c.4 cm from the tragus. It has two structurally different parts; the lateral third is cartilaginous and the medial two-thirds is osseous. It forms an S-shaped curve, directed at first medially, anteriorly and slightly up (pars externa), then posteromedially and up (pars media) and lastly anteromedially and slightly down (pars interna). It is oval in section, its greatest diameter is obliquely inclined posteroinferiorly at the external orifice, but is nearly horizontal at its medial end. There are two constrictions, one near the medial end of the cartilaginous part, the other, the isthmus, in the osseous part c.2 cm from the bottom of the concha. The tympanic membrane, which closes its medial end, is obliquely set and consequently the floor and the anterior wall of the meatus are longer than its roof and posterior wall. The lateral, cartilaginous part is c.8 mm long. It is continuous with the auricular cartilage and attached by fibrous tissue to the circumference of the osseous part. This meatal cartilage is deficient posterosuperiorly, and the gap is occupied by a sheet of collagen. Two or three deep fissures exist in its anterior part-the fissures of Santorini.
Figure 38.3 The external and middle regions of the left ear: anterior aspect.
The osseous part is c.16 mm long, and is narrower than the cartilaginous part. It is directed anteromedially and slightly downwards, with a slight posterosuperior convexity. Its medial end is smaller than the lateral end and it terminates obliquely. The anterior wall projects medially c.4 mm beyond the posterior and is marked, except above, by a narrow tympanic sulcus, to which the perimeter of the tympanic membrane is attached. Its lateral end is dilated and mostly rough for the attachment of the meatal cartilage. The anterior, inferior, and most of the posterior, parts of the osseous meatus are formed by the tympanic element of the temporal bone, which in the fetus is only a tympanic ring. The posterosuperior region is formed by the squamous part of the temporal bone. Relations of the meatus The condylar process of the mandible lies anterior to the meatus and is partially separated from the cartilaginous part by a small portion of the parotid gland. A blow on the chin may cause the condyle to break into the meatus. The middle cranial fossa lies above the osseous meatus and the mastoid air cells are posterior to it, separated from the meatus only by a thin layer of bone. Its deepest part is situated below the epitympanic recess, and is anteroinferior to the mastoid antrum: the lamina of bone which separates it from the antrum is only 1-2 mm thick and provides the 'transmeatal approach' of aural surgery. Vasculature and lymphatic drainage The arterial supply of the external acoustic meatus is derived from the posterior auricular artery, the deep auricular branch of the maxillary artery and the auricular branches of the superficial temporal artery. Associated veins drain into the external jugular and maxillary veins and the pterygoid plexus. The lymphatics drain into those associated with the pinna. Innervation The sensory innervation of the external acoustic meatus is derived from the auriculotemporal branch of the mandibular nerve-which supplies the anterior and superior walls-and the auricular branch of the vagus-which supplies the posterior and inferior walls. The facial nerve may also contribute via its communication with the vagus nerve. UPDATE Date Added: 01 December 2004 Abstract: Branching patterns of the facial nerve and its communication with the auriculotemporal nerve.
Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=15368081 Branching patterns of the facial nerve and its communication with the auriculotemporal nerve.
EXTERNAL SURGICAL APPROACHES TO THE MIDDLE EAR Surgical access to the middle ear can be achieved via a variety of routes. Provided the external acoustic meatus is wide enough, the tympanic membrane can be elevated by incising the skin of the bony meatus circumferentially and then peeling the deep segment off the bone until the fibrous anulus of the tympanic membrane is visualized. This can then be elevated from the tympanic groove and the middle ear mucosa incised to allow the tympanic membrane to be reflected forwards. This approach, called the permeatal or tympanotomy approach, is used for stapedectomy, and limited middle ear work such as tympanic neurectomy for drooling. page 651 page 652
Figure 38.4 The left auditory apparatus as if viewed through a semi-transparent temporal bone. Compare with Fig. 38.9. Note the bend (genu) in the facial nerve at the site of the geniculate ganglion.
If the external acoustic meatus is too narrow to allow adequate visualization of the middle ear, or if access is required to the mastoid aditus and antrum, it is necessary to displace the superficial soft tissues. There are two main external approaches to the middle ear: the endaural approach and the postauricular approach. The endaural approach involves making an incision in the notch between the
tragus and the helix. This is carried down to expose the lower margin of temporalis-which can be used to harvest a strong fascial graft for reconstructionand the bone of the bony external acoustic meatus. The cartilaginous meatus is separated from the bony meatus and reflected laterally as a conchomeatal flap. The bony meatus can then be widened by drilling away bone, so that access to the middle ear and the mastoid antrum and air sinuses is achieved without damaging the ossicular chain or the facial nerve. The postauricular approach involves making an incision in the post-auricular skin down to temporalis and the periosteum of the mastoid process, dividing the posterior auricular muscles on the way. Grafts can be harvested from the temporalis fascia. The periosteum is incised and elevated to expose the bony external acoustic meatus from behind. The skin over the junction of the bony and cartilaginous meatus is incised so as to allow the cartilage of the auricle and meatus to be swung forward on its blood supply and so expose the bony meatus and mastoid process. Access can then be gained by drilling and elevating a tympanomeatal skin flap as described for the endaural approach.
© 2008 Elsevier
MIDDLE EAR The essential function of the middle ear is to transfer energy efficiently from relatively weak vibrations in the elastic, compressible air in the external acoustic meatus, to the incompressible fluid around the delicate receptors in the cochlea. Mechanical coupling between the two systems must match their resistance to deformation or 'flow', i.e. their impedance, as closely as possible. Thus aerial waves of low amplitude and low force per unit area arrive at the tympanic membrane, which has 15-20 times the area of the stapedial footplate in contact with perilymph in the inner ear. In this manner, the force per unit area generated by the footplate is increased by a similar amount, while the amplitude of vibration is almost unchanged. Protective mechanisms are incorporated into the design of the tympanic cavity. These include: the presence of the pharyngotympanic tube to equalize pressure on both sides of the delicate tympanic membrane; the shape of the articulations between the ossicles; and the reflex contractions of stapedius and tensor tympani in response to sounds of fairly high intensity, which prevent damage due to sudden or excessive excursions of the ossicles. The tympanic cavity or middle ear (Figs 38.5, 38.6, 38.7) is an irregular, laterally compressed space in the petrous part of the temporal bone. It is lined with mucous membrane and filled with air, which reaches it from the nasopharynx via the pharyngotympanic tube. It contains three small bones, the malleus, incus and stapes, which collectively are called the auditory ossicles. They form an articulated chain connecting the lateral and medial walls of the cavity, and transmit the vibrations of the tympanic membrane across the cavity to the cochlea. UPDATE Date Added: 12 July 2005 Abstract: Imaging microscopy of the middle and inner ear using CT microscopy Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=15495168&query_hl=9 Imaging microscopy of the middle and inner ear: Part I: CT microscopy. The space within the middle ear can be subdivided into the tympanic cavity proper, opposite the tympanic membrane, and an epitympanic recess (the attic), above the level of the membrane, which contains the upper half of the malleus and most of the incus. Including the recess, the vertical and anteroposterior diameters of the cavity are each c.15 mm; the transverse diameter is c.6 mm superiorly and 4 mm inferiorly, but opposite the umbo it is only 2 mm. The cavity is bounded laterally by the tympanic membrane and medially by the lateral wall of the internal ear. It communicates posteriorly with the mastoid antrum and with the mastoid air cells. Anteriorly it communicates with the nasopharynx via the pharyngotympanic tube (Figs 38.3, 38.4). UPDATE Date Added: 19 July 2005 Abstract: Anatomic limitations of posterior exposure of sinus tympani. Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=15467617&query_hl=6 Anatomic limitations of posterior exposure of sinus tympani.
page 652 page 653
Figure 38.5 Oblique section through the left temporal bone, showing the medial wall of the middle ear. Compare with Fig. 38.9.
Figure 38.6 Oblique vertical section through the left temporal bone, to show roof and lateral wall of the middle ear and the mastoid antrum.
The tympanic cavity and mastoid antrum, auditory ossicles and structures of the internal ear are all almost fully developed at birth and subsequently alter little. In fetuses the cavity contains a gelatinous tissue which has practically disappeared by birth, when it is filled by a fluid which is absorbed when air enters via the pharyngotympanic tube. The tympanic cavity is a common site of infection, which usually spreads to it from the nose and pharynx along the pharyngotympanic tube. UPDATE Date Added: 23 November 2005 Publication Services, Inc. Abstract: Acoustics of the human middle-ear air space. Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=16158643&query_hl=10 Acoustics of the human middle-ear air space. Stepp CE, Voss SE. Acoust Soc Am. 2005 Aug;118(2):861-71.
Boundaries of tympanic cavity ROOF OF TYMPANIC CAVITY (Fig. 38.6)
Figure 38.7 The lateral wall of the left tympanic cavity.
A thin plate of compact bone, the tegmen tympani, separates the cranial and tympanic cavities, and forms much of the anterior surface of the petrous temporal bone. The tegmen tympani is prolonged posteriorly as the roof the mastoid antrum and anteriorly it covers the canal for tensor tympani. In youth, the unossified petrosquamosal suture may allow the spread of infection from the tympanic cavity
to the meninges. In adults, veins from the tympanic cavity traverse this suture to reach the superior petrosal or petrosquamous sinus and thus may also transmit infection to these structures. Longitudinal fractures of the middle cranial fossa almost always involve the tympanic roof, accompanied by the rupture of the tympanic membrane or a fractured roof of the osseous external acoustic meatus. Such injuries usually cause bleeding from the ear, with escape of cerebrospinal fluid if the dura mater has been torn.
FLOOR OF TYMPANIC CAVITY page 653 page 654
The floor of the tympanic cavity is a narrow, thin, convex plate of bone which separates the cavity from the superior bulb of the internal jugular vein. The bone may be patchily deficient, in which case the tympanic cavity and the vein are separated only by mucous membrane and fibrous tissue. Alternatively, the floor is sometimes thick and may contain some accessory mastoid air cells. A small aperture for the tympanic branch of the glossopharyngeal nerve lies near the medial wall.
LATERAL WALL OF TYMPANIC CAVITY (Figs 38.6, 38.7) The lateral wall consists mainly of the tympanic membrane, but also contains the ring of bone to which the membrane is attached. There is a deficiency or notch in the upper part of this ring, close to which are the small openings of the anterior and posterior canaliculi for the chorda tympani and the petrotympanic fissure. The posterior canaliculus for the chorda tympani is situated in the angle between the posterior and lateral walls of the tympanic cavity just behind the tympanic membrane, and level with the upper end of the handle of the malleus. It leads into a minute canal that descends in front of the facial canal and ends in it c.6 mm above the stylomastoid foramen. The canaliculus transmits the chorda tympani and a branch of the stylomastoid artery to the tympanic cavity. The petrotympanic fissure opens just above and in front of the ring of bone to which the tympanic membrane is attached. It is a mere slit c.2 mm in length, and contains the anterior process and anterior ligament of the malleus. It transmits the anterior tympanic branch of the maxillary artery to the tympanic cavity. The anterior canaliculus for the chorda tympani opens at the medial end of the petrotympanic fissure: the chorda tympani leaves the tympanic cavity through this canaliculus.
TYMPANIC MEMBRANE (Figs 38.6, 38.7, 38.8, 38.12)
Figure 38.8 Left tympanic membrane. A, External aspect as seen through a speculum. Note that a bright cone of light is seen in the anteroinferior quadrant of the membrane when it is illuminated. B, Viewed in auroscope. (B, by kind permission from Mr Simon A Hickey.)
The tympanic membrane separates the tympanic cavity from the external acoustic meatus. It is thin, semi-transparent, and almost oval, though somewhat broader above than below. It lies obliquely, at an angle of c.55° with the meatal floor. Its longest, anteroinferior diameter is from 9 to 10 mm and its shortest is from 8 to 9 mm. Most of its circumference is a thickened fibrocartilaginous ring or anulus which is attached to the tympanic sulcus at the medial end of the meatus. This sulcus is deficient superiorly. Two bands, the anterior and posterior malleolar folds, pass from the ends of the notch to the lateral process of the malleus. The small triangular part of the membrane, the pars flaccida, lies above these folds and is lax and thin. The major part of the tympanic membrane, the pars tensa, is taut. The handle of the malleus is firmly attached to the internal surface of the tympanic membrane as far as its centre, which projects towards the tympanic cavity. The inner surface of the membrane is thus convex and the point of greatest convexity is termed the umbo. Although the membrane as a whole is convex on its inner surface, its radiating fibres are curved with their concavities directed inwards.
MICROSTRUCTURE Histologically, the tympanic membrane is composed of three strata: an outer cuticular layer, an intermediate fibrous layer and an inner mucous layer. The cuticular stratum is continuous with the thin skin of the meatus. It is keratinized, stratified squamous in type, devoid of dermal papillae and hairless. Its subepithelial tissue is vascularized and may develop a few peripheral papillae. Ultrastructurally, it is c.10 cells thick and has two zones, a superficial layer of non-nucleated squames, and a deep zone which resembles the epidermal stratum spinosum. There are numerous desmosomes between cells, the deepest of which lie on a continuous basal lamina, but lack epithelial pegs and hemidesmosomes. The cells of this stratum have a propensity for lateral migration and differentiation not shared with any other stratified squamous epithelia in the body. The fibrous stratum consists of an external layer of radiating fibres which diverge from the handle of the malleus, and a deep layer of circular fibres, which are plentiful peripherally, but sparse and scattered centrally. Ultrastructurally the filaments are c.10 nm in diameter, and are linked at 25 nm intervals. They have a distinctive amino acid composition, and may consist of a protein peculiar to the tympanic membrane. Small groups of collagen fibrils appear at 11 weeks in utero, interspersed with small bundles of elastin microfibrils. Older specimens contain more typically cross-banded collagen fibrils and an amorphous elastin component. The fibrous stratum is replaced by loose connective tissue in the pars flaccida. UPDATE Date Added: 13 September 2005 Publication Services, Inc. Abstract: Smooth muscle in the annulus fibrosus of the tympanic membrane in bats, rodents, insectivores, and humans. Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=15668036&query_hl=6 Smooth muscle in the annulus fibrosus of the tympanic membrane in bats, rodents, insectivores, and humans. Henson MM, Madden VJ, Rask-Andersen H et al: Hear Res. Feb;200(1-2):29-37, 2005. page 654 page 655
The mucous stratum is a part of the mucosa of the tympanic cavity, and is thickest near the upper part of the membrane. It consists of a single layer of very flat cells, with overlapping interdigitating boundaries. There are desmosomes and tight junctions between adjacent cells. Their cytoplasm contains only a few organelles: the luminal surfaces of these apparently metabolically inert cells have a few irregular microvilli and are covered by an amorphous electron-dense material. Ciliated columnar cells are absent.
INNERVATION OF THE TYMPANIC MEMBRANE The tympanic membrane is mainly innervated by the auriculotemporal nerve, and appears to perceive only pain. There is a minor, inconstant, overlapping sensory supply from the seventh, ninth and tenth cranial nerves. The auricular branch of the vagus arises from the superior vagal ganglion and is joined soon after by a ramus from the inferior ganglion of the glossopharyngeal nerve. It passes behind the internal jugular vein and enters the mastoid canaliculus on the lateral wall of the jugular fossa. It traverses the temporal bone and crosses the facial canal c.4 mm above the stylomastoid foramen. At this point it supplies an ascending branch to the facial nerve. Fibres of the nervus intermedius may pass to the auricular branch of the vagus here, which may explain the cutaneous vesiculation that sometimes accompanies geniculate herpes. The auricular branch then traverses the tympanomastoid fissure, and divides into two rami. One joins the posterior auricular nerve. The other branch is distributed to the skin of part of the cranial surface of the auricle, the posterior wall and floor of the external acoustic meatus, and to the adjoining part of the outer surface of the tympanic membrane. The auricular branch therefore contains somatic afferent nerve fibres, which probably terminate in the spinal trigeminal nucleus. Stimulation of the vagus nerve-as in syringing the ear-can have a reflex reaction on heart rate.
PHARYNGOTYMPANIC TUBE BLOCKAGE IN CHILDREN The pharyngotympanic tube serves to ventilate the middle ear, exchanging nasopharyngeal air with the air in the middle ear which has been altered in its composition as a result of transmucosal gas exchange with the haemoglobin in the blood vessels of the mucosa. It also carries mucus from the middle ear cleft to the nasopharynx as a result of ciliary transport. In children the pharyngotympanic tube is relatively narrow and is prone to obstruction as a result of mucosal swelling following infection and allergy. Obstruction of the tube results in a relative vacuum being created in the middle ear secondary to transmucosal gas exchange, and this in turn promotes mucosal secretion and the formation of a middle ear effusion. Because of the collapsibility of the pharyngotympanic tube, the vacuum thus created can overcome the distending effect of the muscles of the tube and 'lock' the tube shut. The resultant persistent middle ear effusion can cause hearing loss by splinting the tympanic membrane and impeding its vibration. It can also provide an ideal environment for the proliferation of bacteria with the result that an acute otitis media may develop. It is possible to relieve the vacuum and unlock the tube, and then remove the effusion by myringotomy, i.e. surgically creating a hole in the tympanic membrane. This hole will generally heal rapidly and it is common practice to insert a flanged ventilation tube-a grommet or tympanostomy tube-to keep the hole open. Migration of the outer squamous layer of the tympanic membrane eventually displaces the tube and the myringotomy heals.
OTITIS MEDIA
Acute otitis media usually arises as a result of ascent of infection from the nasopharynx via the pharyngotympanic tube to the middle ear cleft. From there it may extend to the mastoid aditus and antrum. Swelling secondary to the infection may result in the closure of both exits from the middle ear, i.e. the pharyngotympanic tube and the aditus, with subsequent accumulation of pus under pressure, which causes lateral bulging and inflammation of the tympanic membrane. The latter may burst releasing mucopurulent discharge into the external acoustic meatus, which results in a release of the pressure in the middle ear and a diminution in the levels of pain. After a brief period the discharge dries up, and for the most part the resultant perforation of the tympanic membrane heals. Normal ventilation and drainage of mucus from the middle ear is restored once the swelling in the pharyngotympanic tube resolves. On occasion the process will fail to produce a perforation of the tympanic membrane and the inflammatory exudates will not drain. The immune defence system sterilizes the exudates of organisms, resulting in a sterile mucoid effusion-secretory otitis media-that may cause protracted deafness because its relatively incompressible nature prevents free vibration of the tympanic membrane.
MYRINGOPLASTY Where pathological perforation of the tympanic membrane occurs and fails to heal there may be hearing impairment and a tendency to infection as a result of contamination with organisms from the external acoustic meatus. This may cause a chronic suppurative discharge. Myringoplasty is a surgical procedure that uses a connective tissue scaffold or graft to support healing of the perforation. The commonest technique involves the elevation of the tympanic anulus and the placement of a piece of fibrous connective tissue, e.g. part of the fibrous deep fascia which invests the lateral surface of temporalis or the perichondrium of the tragal cartilage, onto the undersurface of the tympanic membrane to close the perforation. The healed edges of the perforation are stripped of epithelium to encourage healing and scar formation. The fibrous tissue supports the healing tympanic membrane and may in part be incorporated into the repair. Once the perforation is healed, the vibratory function of the tympanic membrane is usually restored to normal.
MEDIAL WALL OF TYMPANIC CAVITY (Figs 38.5, 38.9) The medial wall of the tympanic cavity is also the lateral boundary of the internal ear. Its features are the promontory, fenestra vestibuli (oval window), fenestra cochleae (round window) and the facial prominence. The promontory is a rounded prominence furrowed by small grooves which lodge the nerves of the tympanic plexus. It lies over the lateral projection of the basal turn of the cochlea. A minute spicule of bone frequently connects the promontory to the pyramidal eminence of the posterior wall. The apex of the cochlea lies near the medial wall of the tympanic cavity, anterior to the promontory. A depression behind the promontory, the sinus tympani, indicates the position of the ampulla of the posterior semicircular canal. The fenestra vestibuli (fenestra ovalis) is a kidney-shaped opening situated above and behind the promontory, and leading from the tympanic cavity to the vestibule of the inner ear. Its long diameter is horizontal and its convex border is directed upwards. It is occupied by the base of the stapes, the circumference of which is attached to the margin of the fenestra by an anular ligament.
The fenestra cochlea (fenestra rotunda) is situated below and a little behind the fenestra vestibuli, from which it is separated by the posterior part of the promontory. It lies completely under the overhanging edge of the promontory in a deep hollow or niche, and is placed very obliquely. In dried specimens it opens anterosuperiorly from the tympanic cavity into the scala tympani of the cochlea. In life, it is closed by the secondary tympanic membrane, which is somewhat concave towards the tympanic cavity and convex towards the cochlea, the membrane being bent so that its posterosuperior one-third forms an angle with its anteroinferior two-thirds. The membrane has three layers: an external layer which is derived from the tympanic mucosa; an internal layer from the cochlear lining membrane; and an intermediate, fibrous layer. The prominence of the facial nerve canal indicates the position of the upper part of the bony canal which contains the facial nerve. This canal crosses the medial tympanic wall, just above the fenestra vestibuli, and then curves down into the posterior wall of the cavity. Its lateral wall may be partly deficient.
POSTERIOR WALL OF TYMPANIC CAVITY (Figs 38.5, 38.6, 38.7, 38.9) The posterior wall of the tympanic cavity is wider above than below. Its main features are the aditus to the mastoid antrum, the pyramid, and the fossa incudis. The aditus to the mastoid antrum is a large irregular aperture which leads back from the epitympanic recess into the upper part of the mastoid antrum. A rounded eminence on the medial wall of the aditus, above and behind the prominence of the facial nerve canal, corresponds with the position of the lateral semicircular canal. page 655 page 656
Figure 38.9 Oblique section through the left temporal bone, to show the medial wall of the middle ear. The cochlea and the semicircular canals are in blue. Note the relationship of the first coil of the cochlea to the promontory and the closeness of the facial nerve canal and the lateral semicircular canal to the medial wall of the aditus.
The pyramidal eminence is situated just behind the fenestra vestibuli and in front of the vertical part of the facial nerve canal. It is hollow and contains stapedius. Its summit projects towards the fenestra vestibuli and is pierced by a small aperture which transits the tendon of stapedius. The cavity in the pyramidal eminence is prolonged down and back in front of the facial nerve canal, and communicates with the latter by an aperture through which a small branch of the facial nerve passes to stapedius. The fossa incudis is a small depression in the lower and posterior part of the epitympanic recess. It contains the short process of the incus, which is fixed to the fossa by ligamentous fibres.
MASTOID ANTRUM (Figs 38.4, 38.5, 38.6, 38.7, 38.9) The mastoid antrum is an air sinus in the petrous part of the temporal bone. Its topographical relations are of considerable surgical importance. In the upper part of its anterior wall is an opening, the aditus to the mastoid antrum, which leads back from the epitympanic recess. The lateral semicircular canal lies medial to the aditus. The medial wall of the antrum is related to the posterior semicircular canal (Fig. 38.9). The sigmoid sinus is posterior, and may be separated from the antrum by mastoid air cells. The roof is formed by the tegmen tympani, and lies below the middle cranial fossa and temporal lobe of the brain. The floor has several openings which communicate with the mastoid air cells. Anteroinferior is the descending part of the facial nerve canal. The lateral wall of the antrum, which
offers the usual surgical approach to the cavity, is formed by the postmeatal process of the squamous part of the temporal bone. This is only 2 mm thick at birth but increases at a rate of c.1 mm a year, to attain a final thickness of 12-15 mm. In adults, the lateral wall of the antrum corresponds to the suprameatal triangle on the outer surface of the skull, which is palpable through the cymba conchae. The superior side of the suprameatal triangle, the supramastoid crest, is level with the floor of the middle cranial fossa. The anteroinferior side, which forms the posterosuperior margin of the external acoustic meatus, indicates approximately the position of the descending part of the facial nerve canal. The posterior side, formed by a posterior vertical tangent to the posterior margin of the external acoustic meatus, is just anterior to the sigmoid sinus. UPDATE Abstract: Anatomic variations of the sigmoid sinus.
Date Added: 19 July 2005
Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=15365535&query_hl=8 Anatomic variations of the sigmoid sinus. The adult capacity of the mastoid antrum is variable, but on average is c.1 ml and its general diameter c.10 mm. Unlike other air sinuses, it is present at birth, and is indeed then almost adult in size, although it is at a higher level relative to the external acoustic meatus than in adults. In the very young, the thinness of the lateral antral wall and the absence of the mastoid process, means that the stylomastoid foramen and emerging facial nerve are very superficially situated.
MASTOID AIR CELLS (Figs 38.5, 38.9) The mastoid air cells vary considerably in number, form and size. Usually they interconnect and are lined by a mucosa with squamous non-ciliated epithelium, continuous with that in the mastoid antrum and tympanic cavity. They may fill the mastoid process, even to its tip, and some may be separated from the sigmoid sinus and posterior cranial fossa only by extremely thin bone, which is occasionally deficient. Some may lie superficial to, or even behind the sigmoid sinus and others may occur in the posterior wall of the descending part of the facial nerve canal. Those in the squamous part of the temporal bone may be separated from deeper cells in the petrous part by a plate of bone in the line of the squamomastoid suture (Korner's septum). Sometimes they extend very little into the mastoid process, in which case the process consists largely of dense bone or trabecular bone containing bone marrow. Varieties of the mastoid process occur. The three types most commonly described are pneumatized, with many air cells; sclerotic, with few or none; and mixed, which contains both air cells and bone marrow. Mastoid air cells may extend beyond the mastoid process into the squamous part of the temporal bone above the supramastoid crest; into the posterior root of the zygomatic process of the temporal bone; into the osseous roof of the external acoustic meatus just below the middle cranial fossa; and into the floor of the tympanic cavity very close to the superior jugular bulb. Rarely, a few may excavate the jugular process of the occipital bone. An important group may extend medially into the petrous part of the temporal bone, even to its apex, related to the pharyngotympanic tube, carotid canal, labyrinth and abducens nerve. However some investigators maintain that these are not continuous with the mastoid cells, but grow independently from the tympanic cavity. The extensions of the mastoid air cells described above are pathologically important
since infection may spread to the structures around them. Though the mastoid process antrum is well-developed at birth, the mastoid air cells are merely minute antral diverticula at this stage. As the mastoid develops in the second year, the cells gradually extend into it and by the fourth year they are well formed, although their greatest growth occurs at puberty. In c.20% of skulls the mastoid process has no air cells at all. Innervation The mastoid air cells are innervated by a meningeal branch of the mandibular nerve.
MASTOIDITIS page 656 page 657
Mastoiditis occurs as a result of the spread of a bacterial infection from the tympanic cavity via the aditus to the mastoid antrum and associated mastoid air cells. This dangerous condition may spread from the antrum to surrounding structures and cause life-threatening infection. In particular the infection may spread through the tegmen tympani to the dura mater of the middle cranial fossa, to cause an extradural collection. This in turn may cause necrosis of the adjacent dura mater and infection may spread to form a subdural empyema in the subarachnoid space, or an abscess in the substance of the adjacent temporal lobe. Bacterial meningitis may also develop. Similar spread may be seen into the posterior cranial fossa. In both sites the infection may prove fatal. Infection may spread laterally through the cortical bone of the lateral aspect of the mastoid process to form a subperiosteal postauricular abscess (Bezold's abscess), or through the cortical bone of the mastoid tip of the mastoid process to the attachment of the posterior belly of digastric and sternocleidomastoid, which stimulates painful muscular contraction and torticollis.
ANTERIOR WALL OF TYMPANIC CAVITY (Figs 38.3, 38.6) The inferior, larger area of the anterior wall is narrowed by the approximation of the medial and lateral walls of the cavity. It is a thin lamina and forms the posterior wall of the carotid canal. It is perforated by the superior and inferior caroticotympanic nerves and the tympanic branch or branches of the internal carotid. The canals for tensor tympani, and the osseous part of the pharyngotympanic tube open above it. The canal for tensor tympani is superior to that for the pharyngotympanic tube. Both canals incline downwards and anteromedially to open in the angle between the squamous and petrous parts of the temporal bone, and are separated by a thin, osseous septum. The canal for tensor tympani and the bony septum runs posterolaterally on the medial tympanic wall, and ends immediately above the fenestra vestibuli. Here, the posterior end of the septum is curved laterally to form a pulley, the processus trochleariformis (cochleariformis). The tendon of tensor tympani turns laterally over the pulley before attaching to the upper part of the handle of the malleus. UPDATE Date Added: 20 September 2005 Publication Services, Inc. Abstract: Finite element analysis of active Eustachian tube function. Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=15047672&query_hl=9 Finite element analysis of active Eustachian tube function. Ghadiali SN, Banks J,
Swarts JD: J Appl Physiol 97(2):648-654, 2004.
Auditory ossicles (Fig. 38.10) A chain of three mobile ossicles, the malleus, incus and stapes, transfers sound waves across the tympanic cavity from the tympanic membrane to the fenestra vestibuli. The malleus is attached to the tympanic membrane and the base of the stapes is attached to the rim of the fenestra vestibuli. The incus is suspended between them, and articulates with both bones.
MALLEUS (Figs 38.3, 38.6, 38.7, 38.10) The malleus is the largest of the ossicles, and is shaped somewhat like a mallet. It is 8-9 mm long and has a head, neck, handle (manubrium) and anterior and lateral processes. The head is the large upper end of the bone and is situated in the epitympanic recess. It is ovoid in shape. It articulates posteriorly with the incus, and is covered elsewhere by mucosa. The cartilaginous articular facet for the incus is narrowed near its middle and consists of a larger upper part and a smaller lower part, orientated almost at right angles to each other. Opposite the constriction the lower margin of the facet projects in the form of a process, the spur of the malleus. The neck is the narrowed part below the head and inferior to this is an enlargement from which the processes project. The handle of the malleus is connected by its lateral margin with the tympanic membrane (Figs 38.6, 38.7, 38.8). It is directed downwards, medially and backwards. It decreases in size towards its free end, which is curved slightly forwards and is flattened transversely. Near the upper end of its medial surface there is a slight projection to which the tendon of tensor tympani is attached. The anterior process is a delicate bony spicule, directed forwards from the enlargement below the neck. It is connected to the petrotympanic fissure by ligamentous fibres. In fetal life this is the longest process of the malleus and is continuous in front with Meckel's cartilage. The lateral process is a conical projection from the root of the handle of the malleus. It is directed laterally and is attached to the upper part of the tympanic membrane and, via the anterior and posterior malleolar folds, to the sides of the notch in the upper part of the tympanic sulcus (Fig. 38.8). Ossification
Figure 38.10 The left ear ossicles.
The cartilaginous precursor of the malleus originates as part of the dorsal end of Meckel's cartilage. With the exception of its anterior process, the malleus ossifies from a single endochondral centre which appears near the future neck of the bone in the fourth month in utero. The anterior process ossifies separately in dense connective tissue and joins the rest of the bone at about the sixth month of fetal life.
INCUS (Figs 38.3, 38.6, 38.7, 38.10) The incus is shaped less like an anvil-from which it is named-than a premolar tooth, with its two diverging roots. It has a body and two processes. The body is somewhat cubical but laterally compressed. On its anterior surface it has a saddle-shaped facet for articulation with the head of the malleus. The long process, rather more than half the length of the handle of the malleus, descends almost vertically, behind and parallel to the handle. Its lower end bends medially and ends in a rounded lenticular process, the medial surface of which is covered with cartilage and articulates with the head of the stapes. The short process, somewhat conical, projects backwards and is attached by ligamentous fibres to the fossa incudis in the lower and posterior part of the epitympanic recess.
Ossification The incus has a cartilaginous precursor continuous with the dorsal extremity of Meckel's cartilage. Ossification often spreads from a single centre in the upper part of its long process in the fourth fetal month; the lenticular process may have a separate centre.
STAPES (Figs 38.3, 38.9, 38.10) page 657 page 658
The stapes is also known as the stirrup. It has a head, neck, two limbs and a base. The head (caput) is directed laterally and has a small cartilaginous facet for the lenticular process of the incus. The neck is the constricted part supporting the head, and the tendon of stapedius is attached to its posterior surface. The limbs (crura) diverge from the neck and are connected at their ends by a flattened oval plate, the base, which forms the footplate of the stapes. The base is attached to the margin of the fenestra vestibuli by a ring of fibres (the anular ligament). The anterior limb is shorter and less curved than the posterior. Ossification The stapes is preformed in the perforated dorsal moiety of the hyoid arch cartilage of the fetus. Ossification starts from a single endochondral centre, which appears in the base in the fourth fetal month and then gradually spreads through the limbs of the stapes to reach the head. At birth the auditory ossicles have reached an advanced state of maturity.
LIGAMENTS OF AUDITORY OSSICLES The ossicles are connected to the tympanic walls by ligaments. There are three for the malleus and one each for the incus and stapes. Some of these are mere mucosal folds which carry blood vessels and nerves to and from the ossicles and their articulations, and others contain a central, strong band of collagen fibres. The anterior ligament of the malleus stretches from the neck of the malleus, just above the anterior process, to the anterior wall of the tympanic cavity near the petrotympanic fissure. Some of its collagen fibres traverse this fissure to reach the spine of the sphenoid, and others continue into the sphenomandibular ligament. The latter, like the anterior malleolar ligament, is derived from the perichondrial sheath of Meckel's cartilage. The anterior malleolar ligament may contain muscle fibres, called laxator tympani or musculus externus mallei. The lateral ligament of the malleus is a triangular band which stretches from the posterior part of the border of the tympanic incisure to the head of the malleus. The superior ligament of the malleus connects the head of the malleus to the roof of the epitympanic recess. The posterior ligament of the incus connects the end of its short process to the fossa incudis. The superior ligament of the incus is little more than a mucosal fold passing from the body of the incus to the roof of the epitympanic recess. The vestibular surface and rim of the stapedial base are covered with hyaline cartilage. The cartilage encircling the base is attached to the margin of the fenestra vestibuli by a ring of elastic fibres, the anular ligament of the base of the stapes. The posterior part of this ligament is much narrower than the anterior part: it acts as a kind of hinge on which the stapedial base moves when stapedius contracts and during acoustic oscillation.
ARTICULATION OF AUDITORY OSSICLES The articulations are typical synovial joints. The incudomalleolar joint is saddle shaped. The incudostapedial joint is a ball and socket articulation. Their articular surfaces are covered with articular cartilage, and each joint is enveloped by a capsule rich in elastic tissue and lined by synovial membrane.
MOVEMENTS OF THE AUDITORY OSSICLES The handle of the malleus faithfully follows all movements of the tympanic membrane. The malleus and incus rotate together around an axis which runs from the short process and posterior ligament of the incus to the anterior ligament of the malleus. When the tympanic membrane and handle of the malleus move inwards (medially), the long process of the incus moves in the same direction and pushes the stapedial base towards the labyrinth and the perilymph contained within the labyrinth. The movement of the perilymph causes a compensatory outward bulging of the secondary tympanic membrane-which closes the fenestra cochleae. These events are reversed when the tympanic membrane moves outwards. However, if its movement is considerable, the incus does not follow the full outward excursion of the malleus, and merely glides on it at the incudomalleolar joint, thus preventing a dislocation of the base of the stapes from the fenestra vestibuli. When the handle of the malleus is carried medially, the spur at the lower margin of the head of the malleus locks the incudomalleolar joint, and this necessitates an inward movement of the long process of the incus. The joint is unlocked again when the handle of the malleus is carried outwards. The three bones together act as a bent lever so that the stapedial base does not move in the fenestra vestibuli like a piston, but rocks on a fulcrum at its anteroinferior border, where the anular ligament is thick. The rocking movement around a vertical axis, which is like a swinging door, is said to occur only at moderate intensities of sound. With loud, low-pitched sounds, the axis becomes horizontal, and the upper and lower margins of the stapedial base oscillate in opposite directions around this central axis, thus preventing excessive displacement of the perilymph.
OTOSCLEROSIS AND STAPEDECTOMY Otosclerosis is a hereditary localized disease of bone derived from the embryonic otic capsule in which lamellar bone is replaced by woven bone of greater thickness and vascularity. The position of the focus of new bone determines its effect on the function of the ear. Where it occurs around the footplate of the stapes it may fix the footplate to the margin of the oval window and prevent it moving. This prevents the passage of vibrations of the tympanic membrane passing through the ossicular chain to the inner ear, which results in clinical hearing loss. Complete deafness does not result, because vibrations can still pass directly to the cochlea via the bones of the skull-albeit in a markedly less efficient manner. UPDATE Date Added: 27 September 2005 Publication Services, Inc. Abstract: Development of the stapes and associated structures in human embryos. Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=16050903&query_hl=5 Development of the stapes and associated structures in human embryos.
Rodríguez-Vázquez JF: J Anat 207(2):165-174, 2005. Stapedectomy is a surgical procedure designed to bypass the fixation of the stapes footplate caused by otosclerosis. The tympanic membrane is temporarily elevated for access to the middle ear and, under microscopic control, the incudostapedial joint is disarticulated using microinstruments. The limbs of the stapes and stapedius are then divided, usually with microscissors or a laser, and the superstructure of the stapes removed. A small hole is then made in the fixed footplate of the stapes using a microdrill or laser to expose the fluids of the inner ear. A small graft of connective tissue, usually fat or derived from vein, is used to seal the hole with a flexible membrane. A small piston usually made of plastic and wire is crimped onto the long process of the incus and placed in the perforation in the stapes footplate. The tympanic membrane is then returned. The connection between the tympanic membrane and the inner ear is thus reconstituted and hearing restored.
Muscles of tympanic cavity TENSOR TYMPANI (Figs 38.3, 38.6) Tensor tympani is a long slender muscle which occupies the bony canal above the osseous part of the pharyngotympanic tube, from which it is separated by a thin bony septum. It arises from the cartilaginous part of the pharyngotympanic tube and the adjoining region of the greater wing of the sphenoid, as well as from its own canal. It passes back within its canal, and ends in a slim tendon which bends laterally round the pulley-like processus trochleariformis and finally attaches to the handle of the malleus, near its root. Vascular supply Tensor tympani receives its arterial blood supply from the superior tympanic branch of the middle meningeal artery. Innervation Tensor tympani is innervated by a branch of the nerve to medial pterygoid-a ramus of the mandibular nerve-which traverses the otic ganglion without interruption to reach the muscle. Actions Tensor tympani draws the handle of the malleus medially, and so tenses the tympanic membrane and helps to damp sound vibrations: its action also pushes the base of the stapes more tightly into the fenestra vestibuli.
STAPEDIUS Stapedius arises from the wall of a conical cavity in the pyramidal eminence on the posterior wall of the tympanic cavity, and from its continuation anterior to the descending part of the facial nerve canal. Its minute tendon emerges from the orifice at the apex of the pyramid and passes forwards to attach to the posterior surface of the neck of the stapes (Fig. 38.9). The muscle is of an asymmetrical bipennate form, and contains numerous small motor units, each of only six to nine muscle fibres. A few neuromuscular spindles exist near the myotendinous junction. UPDATE Date Added: 13 September 2005 Publication Services, Inc. Abstract: High-resolution X-ray computed tomographic scanning of the human
stapes footplate. Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=15903201&query_hl=5 High-resolution X-ray computed tomographic scanning of the human stapes footplate. Hagr AA, Funnell WR, Zeitouni AG et al: J Otolaryngol. Aug;33(4):21721, 2004. Vascular supply Stapedius receives its arterial blood supply from branches of the posterior auricular, anterior tympanic and middle meningeal arteries. Innervation Stapedius is supplied by a branch of the facial nerve which is given off in the facial canal. page 658 page 659
Actions Stapedius helps to damp down excessive sound vibrations. It opposes the action of tensor tympani that pushes the stapes more tightly into the fenestra vestibuli. Paralysis of stapedius results in hyperacusis. Stapedial and tensor tympani reflex (p. 342) When noises are loud, and immediately before speaking, a reflex contraction of stapedius and tensor tympani occurs which helps to damp down the movement of the ossicular chain before vibrations reach the internal ear. The afferent pathways involve the auditory component of the eighth cranial nerve, and higher centres prior to speech. The efferent pathway involves the facial nerve (stapedius) and the mandibular nerve (tensor tympani).
Tympanic mucosa The mucosa of the tympanic cavity is continuous with that of the pharynx, via the pharyngotympanic tube. It covers the ossicles, muscles and nerves in the cavity, and forms the inner layer of the tympanic membrane and the outer layer of the secondary tympanic membrane. It also spreads into the mastoid antrum and air cells. It forms several vascular folds which extend from the tympanic walls to the ossicles: one descends from the roof of the cavity to the head of the malleus and the upper margin of the body of the incus and a second surrounds stapedius. Other folds invest the chorda tympani nerve and tensor tympani. The folds separate off saccular recesses which give the interior of the tympanic cavity a somewhat honeycombed appearance. The superior recess of the tympanic membrane lies between the neck of the malleus and the pars flaccida. The anterior and posterior recesses of the tympanic membrane, formed by the mucosa around the chorda tympani, lie anterior and posterior respectively to the handle of the malleus. The tympanic mucosa is pale, thin and slightly vascular. It has a ciliated columnar epithelium, except over the posterior part of the medial wall, the posterior wall, and often parts of the tympanic membrane and the auditory ossicles, where the cells are flatter and non-ciliated. Near the pharyngotympanic tube, goblet cells are numerous; otherwise there are no mucous glands. The mastoid antrum and air cells are lined by flat, non-ciliated epithelium. The epithelium is closely attached to periosteum, and forms a mucoperiosteum. It has surfactant on its surface.
CHOLESTEATOMA Cholesteatoma is the name given to keratinizing squamous epithelium within the middle ear. There is debate as to how such epithelium comes to be in the middle ear. Theories include development from embryological cell rests, metaplasia from inflamed mucoperiosteum, and aberrant migration of squamous epithelium either through a perforation in the tympanic membrane-usually in the pars flaccida or posterosuperior pars tensa-or within an area of tympanic membrane atelectasis where the tympanic membrane becomes adherent to the medial wall of the tympanic cavity. It is likely that all of these may occur at some time. A feature of cholesteatoma that is poorly understood is its ability to erode bone, through activating osteoclasts. This allows the epithelium to proliferate and invade, destroying the temporal bone and carrying infection to the soft tissues. Thus, cholesteatoma can cause: deafness through damage to the ossicles and inner ear; problems with balance through damage to the vestibule and semicircular canals; facial palsy through ischaemia and necrosis of the facial nerve; and intracranial sepsis. Treatment involves microsurgical dissection of the invading sac of epithelium with preservation of these delicate structures wherever possible.
Vascular supply and lymphatic drainage of the tympanic cavity A number of arteries supply the walls and contents of the tympanic cavity. Three, namely the deep auricular, anterior tympanic and stylomastoid arteries, are larger than the others. The deep auricular branch of the first part of the maxillary artery often arises with the anterior tympanic artery. It ascends in the parotid gland behind the temporomandibular joint, pierces the cartilaginous or osseous wall of the external acoustic meatus and supplies its cuticular lining, the exterior of the tympanic membrane and the temporomandibular joint. The anterior tympanic branch of the first part of the maxillary artery ascends behind the temporomandibular joint and enters the tympanic cavity through the petrotympanic fissure. It ramifies on the interior of the tympanic membrane, and forms a vascular circle around it with the posterior tympanic branch of the stylomastoid artery. It also anastomoses with twigs of the artery of the pterygoid canal and caroticotympanic branches of the internal carotid artery in the mucosa of the tympanic cavity. The stylomastoid branch of the occipital or posterior auricular arteries supplies the posterior part of the tympanic cavity and mastoid air cells. It also enters the stylomastoid foramen to supply the facial nerve and semicircular canals. In the young, its posterior tympanic branch forms a circular anastomosis with the anterior tympanic artery. The smaller arteries supplying the tympanic cavity include: the petrosal branch of the middle meningeal artery, which enters through the hiatus for the greater petrosal nerve; the superior tympanic branch of the middle meningeal artery, which traverses the canal for tensor tympani; an inferior tympanic branch from the ascending pharyngeal artery, which traverses the tympanic canaliculus-together with the tympanic branch of the glossopharyngeal nerve-to supply the medial wall of the tympanic cavity; a branch from the artery of the pterygoid canal, which accompanies the pharyngotympanic tube; and a tympanic branch or branches from the internal carotid artery, which is given off in the carotid canal and
perforates the thin anterior wall of the tympanic cavity. The mastoid air cells and dura mater are also supplied by a mastoid branch from the occipital artery. This is small in size and sometimes absent. When present, it enters the cranial cavity via the mastoid foramen near the occipitomastoid suture. In early fetal life a stapedial artery traverses the stapes. The veins from the tympanic cavity terminate in the pterygoid venous plexus and the superior petrosal sinus. A small group of veins runs medially from the mucosa of the mastoid antrum through the arch formed by the anterior semicircular canal, and emerges onto the posterior surface of the petrous temporal bone at the subarcuate fossa. These veins drain into the superior petrosal sinus and are the remains of the large subarcuate veins of childhood. They represent a potential route for the spread of infection from the mastoid antrum to the meninges. Lymphatic vessels of the tympanic and antral mucosae drain to the parotid or upper deep cervical lymph nodes. Vessels of the tympanic end of the pharyngotympanic tube probably end in the deep cervical nodes.
Innervation of the tympanic cavity TYMPANIC PLEXUS The nerves that constitute the tympanic plexus ramify on the surface of the promontory on the medial wall of the tympanic cavity. They are derived from the tympanic branch of the glossopharyngeal nerve (Fig. 38.11) and the caroticotympanic nerves. The former arises from the inferior ganglion of the glossopharyngeal nerve, and reaches the tympanic cavity via the tympanic canaliculus for the tympanic nerve. The superior and inferior caroticotympanic nerves are postganglionic sympathetic fibres which are derived from the carotid sympathetic plexus. They traverse the wall of the carotid canal to join the plexus. The tympanic plexus supplies branches to the mucosa of the tympanic cavity, pharyngotympanic tube and mastoid air cells. It sends a branch to the greater petrosal nerve via an opening anterior to the fenestra vestibuli. The lesser petrosal nerve, which may be regarded as the continuation of the tympanic branch of the glossopharyngeal nerve, traverses the tympanic plexus. It occupies a small canal below that for tensor tympani. It runs past, and receives a connecting branch from, the geniculate ganglion of the facial nerve. The lesser petrosal nerve emerges from the anterior surface of the temporal bone via a small opening lateral to the hiatus for the greater petrosal nerve and then traverses the foramen ovale or the small canaliculus innominatus to join the otic ganglion (Fig. 38.11). Postganglionic secretomotor fibres leave this ganglion in the auriculotemporal nerve to supply the parotid gland.
FACIAL NERVE (Fig. 38.11) page 659 page 660
Figure 38.11 The facial nerve. A, Course of the facial nerve and its branches through
the temporal bone; the vestibulocochlear nerve has been omitted. B, A plan of the intrapetrous section of the facial nerve, its branches and communications. The course of the taste fibres from the mucous membrane of the palate and from the anterior presulcal part of the tongue is represented by the blue lines. (A, by permission from Hall-Craggs ECB 1986 Anatomy as a Basis for Clinical Medicine, 2nd edn. Baltimore: Urban and Schwarzenberg.)
The facial nerve enters the temporal bone through the internal acoustic meatus accompanied by the vestibulocochlear nerve. At this point the motor root, which supplies the muscles of the face, and the nervus intermedius, which contains sensory fibres concerned with the perception of taste and parasympathetic (secretomotor) fibres to various glands, are separate components. They merge within the meatus. At the end of the meatus, the facial nerve enters its own canal, the facial canal, which runs across the medial wall and down the posterior wall of the tympanic cavity to the stylomastoid foramen (Fig. 38.9). As the nerve enters the facial canal, there is a bend which contains the geniculate ganglion (Figs 38.4, 38.11). UPDATE Date Added: 07 September 2005 Publication Services, Inc. Abstract: Facial canal anatomy in patients with mandibulofacial dysostosis: comparison with respect to the severities of microtia and middle ear deformity. Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=16015188&query_hl=12 Facial canal anatomy in patients with mandibulofacial dysostosis: comparison with respect to the severities of microtia and middle ear deformity. Takegoshi H, Kaga K, Chihara Y et al: Otol Neurotol. Jul;26(4):803-8, 2005. The branches which arise from the facial nerve within the temporal bone can be divided into those which come from the geniculate ganglion and those which arise within the facial canal. The main branch from the geniculate ganglion is the greater (superficial) petrosal nerve. It is a branch of the nervus intermedius. The greater petrosal nerve passes anteriorly, receives a branch from the tympanic plexus and traverses a hiatus on the anterior surface of the petrous part of the temporal bone. It enters the middle cranial fossa and runs forwards in a groove on the bone above the lesser petrosal nerve. It passes beneath the trigeminal ganglion to reach the foramen lacerum. Here it is joined by the deep petrosal nerve from the internal carotid sympathetic plexus, to become the nerve of the pterygoid canal (Vidian's nerve). The greater petrosal nerve contains parasympathetic fibres destined for the pterygopalatine ganglion, and taste fibres from the palate. The nerve to stapedius arises from the facial nerve in the facial nerve canal behind the pyramidal eminence of the posterior wall of the tympanic cavity. It passes forwards through a small canal to reach the muscle.
page 660 page 661
Figure 38.12 Chorda tympani nerve crossing the tympanic membrane. (By kind permission from Mr Simon A Hickey.)
The chorda tympani (Fig. 38.12, Fig. 30.7) leaves the facial nerve c.6 mm above the stylomastoid foramen and runs anterosuperiorly in a canal to enter the tympanic cavity via the posterior canaliculus. It then curves anteriorly in the substance of the tympanic membrane between its mucous and fibrous layers (Fig. 38.6), crosses medial to the upper part of the handle of the malleus to the anterior wall, where it enters the anterior canaliculus (Fig. 38.7) It exits the skull at the petrotympanic fissure and its further course is described on page 525. It contains parasympathetic fibres which supply the submandibular and sublingual salivary glands via the submandibular ganglion (Fig. 30.8) and taste fibres from the anterior two-thirds of the tongue. The geniculate ganglion also communicates with the lesser petrosal nerve.
BELL'S PALSY Bell's palsy is the name given to a lower motor neurone palsy of the facial nerve which occurs spontaneously and without obvious cause. It is characterized by a flaccid paralysis of the ipsilateral muscles of facial expression; decreased lacrimation in the ipsilateral eye (which is controlled by neurones in the greater petrosal nerve); and hyperacusis or decreased tolerance of loud noises in the ipsilateral ear due to paralysis of stapedius. Its cause remains the subject of speculation, but recent MRI studies suggest that it may be the result of viral neuronitis either in the bony first part of the facial canal (labyrinthine segment) at
the apex of the internal auditory canal, or in the adjacent brain stem. In the majority of cases spontaneous full recovery occurs after a few weeks.
DEHISCENCES OF FACIAL NERVE CANAL The facial nerve may be somewhat variable in its anatomical course through the temporal bone. It may split into two or three strands, or pass a few millimetres posteriorly to its second bend, before it turns inferiorly posterior to the fossa incudis-a position where it is particularly vulnerable during surgical exploration of the mastoid antrum. It may be dehiscent, particularly in its second part, when it occasionally overhangs the stapes, or run inferior to the stapes superstructure, a position which renders it vulnerable during surgery to the stapes. The motor fibres to the face may be carried through the chorda tympani, which is then enlarged. When this occurs, the distal facial nerve dwindles to a fibrous strand in a narrowed stylomastoid foramen. In chronic bone disease in the tympanic cavity, the facial nerve may be exposed in its canal. Inflammation may lead to facial paralysis of the infranuclear or lower motor neurone type. UPDATE Date Added: 07 September 2005 Publication Services, Inc. Absract: Fallopian canal dehiscences: a survey of clinical and anatomical findings. Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=15592859&query_hl=13 Fallopian canal dehiscences: a survey of clinical and anatomical findings. Di Martino E, Sellhaus B, Haensel et al: Eur Arch Otorhinolaryngol. Feb. 262(2):1206, 2005. REFERENCES Anderson SD 1976 The intratympanic muscles. In: Scientific Foundations of Otolaryngology. Hinchcliffe R (ed) Heinemann, London. pp 257-280. Anson BJ, Donaldson JA 1976 The Surgical Anatomy of the Temporal Bone and Ear. Saunders, Philadelphia. Baily CM 1997 Surgical anatomy of the skull base. In: Scott Brown's Otolaryngology. Sixth ed. Vol 1. Kerr GA (ed). Butterworth Heinemann, London. Chapter 15. pp. 1-15. Bluestone C, Klein J 2002 Otitis media, atelectasis and Eustachian tube dysfunction. In: Pediatric Otolaryngology. Bluestone CD, Stool SE (eds) 4th edition. Vol. 1. Saunders, Philadelphia. pp 474-686. Couter RT 1980 A Colour Atlas of Temporal Bone Surgical Anatomy. Wolfe Medical, London. Glassock (III) ME, Shambaugh GE 1990 Surgery of the Ear. 4th edition. Saunders, Philadelphia. Grey P 1995 The clinical significance of the communicating branches of the somatic sensory supply of the middle and external ear. J Laryngol Otol 109: 1141-1145. Medline Similar articles Honjo I 1988 Eustachian Tube and Middle Ear Diseases. Springer-Verlag, Berlin. Phelps PD, Lloyd GAS 1990 Diagnostic Imaging of the Ear. 2nd edition. Springer-Verlag, Berlin. Wright A 1997 Anatomy and ultrastructure of the human ear. In: Scott Brown's Otolaryngology. Sixth ed. Vol 1. Kerr GA (ed). Butterworth Heinemann, London. Chapter 1. pp. 1-50. page 661 page 662
© 2008 Elsevier
39 EAR AND AUDITORY AND VESTIBULAR APPARATUS Inner ear The inner ear contains the organ of hearing, the cochlea, and the organs of balance, the utricle, saccule and semicircular canals. It consists of the bony (osseous) labyrinth, a series of interlinked cavities in the petrous temporal bone, and the membranous labyrinth of interconnected membranous sacs and ducts that lie within the bony labyrinth. The gap between the internal wall of the bony labyrinth and the external surface of the membranous labyrinth is filled with perilymph, a clear fluid with an ionic composition similar to that of other extracellular fluids, i.e. low in potassium ions and high in sodium and calcium. The membranous labyrinth contains endolymph, a fluid with an ionic composition more like that of cytosol, i.e. high in potassium ions and low in sodium and calcium. Moreover, the endolymphatic compartment is c.80 mV more positive than the perilymphatic compartment. These differences in ionic composition and potential are essential to the primary function of the inner ear, because they provide the driving force for mechanotransduction, the process by which sensory hair cells convert the vibrations set up in the inner ear fluids by sound or head movements into electrical signals that are transmitted via the vestibulocochlear nerve to the vestibular and cochlear nuclei in the brain stem. UPDATE Date Added: 13 September 2005 Publication Services, Inc. Abstract: Image fusion of CT and MRI for the visualization of the auditory and vestibular system. Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=15817422&query_hl=8 Image fusion of CT and MRI for the visualization of the auditory and vestibular system. Seemann MD, Beltle J, Heuschmid M et al: Eur J Med Res. Feb 28;10(2):47-55, 2005. UPDATE Date Added: 13 September 2005 Publication Services, Inc. Abstract: Imaging microscopy of the middle and inner ear. Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=15495168&query_hl=7 Imaging microscopy of the middle and inner ear. Lane JI, Witte RJ, Driscoll CL et al: Clin Anat. Nov;17(8):607-12, 2004.
© 2008 Elsevier
BONE TEMPORAL BONE: INTERNAL ACOUSTIC MEATUS The internal acoustic meatus is separated from the internal ear at its lateral fundus by a vertical plate divided unequally by a transverse crest (Fig. 39.1). The canal for the facial nerve passes above and anterior to the crest. Posterior to the crest, the superior vestibular area contains openings for nerves to the utricle and anterior and lateral semicircular ducts. Below the crest, an anterior cochlear area contains a spiral of small holes, the tractus spiralis foraminosus, which encircles the central cochlear canal. Behind this, the inferior vestibular area contains openings for saccular nerves, and most posteroinferior, a single hole (foramen singulare) admits the nerve to the posterior semicircular duct. UPDATE Date Added: 14 June 2005 Abstract: Course of the meatus acusticus internus as criterion for sex differentiation. Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=15567614&query_hl=11 Course of the meatus acusticus internus as criterion for sex differentiation.
LABYRINTH Bony labyrinth
Figure 39.1 The fundus of the left internal acoustic meatus, exposed by a section through the petrous part of the left temporal bone nearly parallel to the line of its superior border.
The bony labyrinth consists of the vestibule, semicircular canals and cochlea, which are all cavities lined by periosteum; they contain the membranous labyrinth (Figs 38.4, 38.9, 39.2, 39.3, 39.4). The bony labyrinth consists of bone that is more dense and harder than other parts of the petrous bone, and it is therefore possible, particularly in young skulls, to dissect it out from the petrous temporal bone.
Figure 39.2 The left bony labyrinth: lateral aspect.
page 663 page 664
Figure 39.3 The interior of the left bony labyrinth.
Figure 39.4 The membranous labyrinth (blue) projected onto the bony labyrinth. The arrows indicate the direction of pressure waves in the cochlea.
Vestibule
The vestibule is the central part of the bony labyrinth and lies medial to the tympanic cavity, posterior to the cochlea and anterior to the semicircular canals (Figs 38.4, 39.2). It is somewhat ovoid in shape but flattened transversely, and measures c.5 mm from front to back and vertically, and c.3 mm across. In its lateral wall is the opening of the oval window (fenestra vestibuli) into which the base of the stapes inserts, and to which the base of the stapes is attached by an anular ligament (Fig. 38.9). Anteriorly, on the medial wall, is a small spherical recess that contains the saccule; it is perforated by several minute holes, the macula cribrosa media, which transmit fine branches of the vestibular nerve to the saccule (Figs 39.3, 39.4). Behind the recess is an oblique vestibular crest, the anterior end of which forms the vestibular pyramid. This crest divides below to enclose a small depression, the cochlear recess, which is perforated by vestibulocochlear fascicles as they pass to the vestibular end of the cochlear duct. Posterosuperior to the vestibular crest, in the roof and medial wall of the vestibule, is the elliptical recess (Fig. 39.3), which contains the utricle. The pyramid and adjoining part of the elliptical recess are perforated by a number of holes, the macula cribrosa superior. The holes in the pyramid transmit the nerves to the utricle and those in the recess transmit the nerves to the ampullae of the superior and lateral semicircular canals (Fig. 38.4, Fig. 39.3). The region of the pyramid and elliptical recess corresponds to the superior vestibular area in the internal acoustic meatus (Fig. 39.1). The vestibular aqueduct opens below the elliptical recess. It reaches the posterior surface of the petrous bone and contains one or more small veins and part of the membranous labyrinth, the endolymphatic duct (Fig. 39.4). In the posterior part of the vestibule are the five openings of the semicircular canals; in its anterior wall is an elliptical opening that leads into the scala vestibuli of the cochlea. UPDATE Date Added: 07 August 2006 Abstract: Endolymphatic duct status during middle fossa dissection of the internal auditory canal Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? db=pubmed&cmd=Retrieve&dopt=AbstractPlus&list_uids=16540891&query_hl=12&itool=pubmed_docsum Endolymphatic duct status during middle fossa dissection of the internal auditory canal: a human temporal bone radiographic study. Drew BR, Semaan MT, Hsu DP, et al: Laryngoscope 116:370-374, 2006. Semicircular canals
The three semicircular canals, superior (anterior), posterior and lateral (horizontal), are located posterosuperior to the vestibule (Figs 38.4, 38.9, 39.2, 39.3, 39.4). They are compressed from side to side and each forms approximately two-thirds of a circle. They are unequal in length, but similar in
diameter along their lengths, except where they bear a terminal swelling, an ampulla, which is almost twice the diameter of the canal. The superior (anterior) semicircular canal is 15-20 mm long. It is vertical in orientation and lies transverse to the long axis of the petrous temporal bone under the anterior surface of its arcuate eminence. The eminence may not accurately coincide with this semicircular canal, but may instead be adapted to the occipitotemporal sulcus on the inferior surface of the temporal lobe of the brain. The ampulla at the anterior end of the canal opens into the upper and lateral part of the vestibule. Its other end unites with the upper end of the posterior canal to form the crus commune (common limb), which is c.4 mm long, and opens into the medial part of the vestibule. The posterior semicircular canal is also vertical but curves backwards almost parallel with the posterior surface of the petrous bone. It is 18-22 mm long and its ampulla opens low in the vestibule, below the cochlear recess where the macula cribrosa inferior transmits nerves to it. Its upper end joins the crus commune. The lateral (horizontal) canal is 12-15 mm long and its arch runs horizontally backwards and laterally. Its anterior ampulla opens into the upper and lateral angle of the vestibule, above the oval window and just below the ampulla of the superior canal; its posterior end opens below the opening of the crus commune. The two lateral semicircular canals of the two ears are often described as being in the same plane and the anterior canal of one side as being almost parallel with the opposite posterior canal. However, measurements of the angular relations of the planes of the semicircular osseous canals in 10 human skulls led Blanks et al (1975) to suggest that the planes of the three ipsilateral canals are not completely perpendicular to each other. The angles were measured as: horizontal/anterior 111.76 ± 7.55°, anterior/posterior 86.16 ± 4.72°, posterior/horizontal 95.75 ± 4.66°. The planes of similarly orientated canals of the two sides also showed some departure from being parallel: left anterior/right posterior 24.50 ± 7.19°, left posterior/right anterior 23.73 ± 6.71°, left horizontal/right horizontal 19.82 ± 14.93°. The same observers (Curthoys et al 1977) also measured the dimensions and radii of the canals. The means for the radii of the osseous canals were found to be as follows: horizontal 3.25 mm, anterior 3.74 mm, posterior 3.79 mm. The diameters of the osseous canals are c.1 mm (minor axis) and 1.4 mm (major axis). The membranous ducts within them are much smaller, but are also elliptical in transverse section, and have major and minor axes of 0.23 and 0.46 mm (Fig. 39.5). Representative means for ampullary dimensions are as follows: length 1.94 mm, height 1.55 mm. Phylogenetic studies suggest that the arc sizes of the semicircular canals in humans and other primates may be functionally linked to sensory control of body movements. The angulation and dimensions of the canals may be related to locomotor behaviour and possibly to agility, or more specifically to the frequency spectra of natural head movements (see review by Spoor & Zonneveld, 1998). Cochlea page 664 page 665
Figure 39.5 Transverse section through the left posterior semicircular canal and duct of an adult man. (After JK Milne Dickie.)
The cochlea (from the Greek cochlos for snail) is the most anterior part of the labyrinth, lying in front of the vestibule (Figs 38.4, 38.9, 39.2, 39.3, 39.4, 39.6). It is c.5 mm from base to apex, and 9 mm across its base. Its apex, or cupula, points towards the anterosuperior area of the medial wall of the tympanic cavity (Figs 38.4, 39.6). Its base faces the bottom of the internal acoustic meatus and is perforated by numerous apertures for the cochlear nerve. The cochlea has a conical central bony core, the modiolus, and a spiral canal runs around it. A delicate osseous spiral lamina (or ledge) projects from the modiolus, partially dividing the canal (Fig. 39.7). Within this bony spiral lies the membranous cochlear duct, attached to the modiolus at one edge and to the outer cochlear wall by its other edge. There are therefore three longitudinal channels within the cochlea. The middle canal (the cochlear duct or scala media) is blind, and ends at the apex of the cochlea; its flanking channels communicate with each other at the modiolar apex at a narrow slit, the helicotrema (Fig. 39.4). Two elastic membranes form the upper and lower bounds of the scala media. One is Reissner's membrane, the thin vestibular membrane that separates the scala media from the scala vestibuli. The other is the basilar membrane, which forms the partition between the scala media and the scala tympani. The organ of Corti, the sensory epithelium responsible for hearing, sits on the inner surface of the basilar membrane (p. 671). At the base of the scala vestibuli is the oval window (fenestra vestibuli), which leads onto the vestibular cavity but is sealed by the footplate of the stapes. The scala tympani is separated from the tympanic cavity by the secondary tympanic membrane at the round window (fenestra cochleae). The central cochlear core, the modiolus, has a broad base near the lateral end of the internal acoustic meatus, where it corresponds to the spiral tract (tractus spiralis foraminosus) (Fig. 39.6). There are several openings in this area for the fascicles of the cochlear nerve: those for the first 1! turns run through the small holes of the spiral tract, and those for the apical turn run through the hole that forms the centre of the tract. Canals from the spiral tract go through the modiolus and open in a spiral sequence into the base of the osseous spiral lamina. Here the small canals enlarge and fuse to form Rosenthal's canal, a spiral canal in the modiolus which follows the course of the osseous spiral lamina and contains the spiral ganglion (Fig. 39.7). The main tract continues through the centre of the modiolus to the cochlear apex. The osseous cochlear canal spirals for about 2" turns around the modiolus and is c.35 mm long. At its first turn, the canal bulges towards the tympanic cavity where it underlies the promontory. At the base of the cochlea, the canal is c.3 mm in diameter but it becomes progressively reduced in diameter as it spirals apically to end at the cupula. In addition to the round and oval windows, which are the two
main openings at it base, the canal has a third smaller opening for the cochlear aqueduct or canaliculus. The latter is a minute funnel-shaped canal that runs to the inferior surface of the petrous temporal bone; it transmits a small vein to the inferior petrosal sinus and connects the subarachnoid space to the scala tympani.
page 665 page 666
Figure 39.6 Horizontal section through the left temporal bone. (Drawn from a section prepared at the Ferens Institute and lent by the late J Kirk.)
Figure 39.7 Section through the second turn of the cochlea seen in Fig. 39.6. The
Figure 39.7 Section through the second turn of the cochlea seen in Fig. 39.6. The modiolus is to the left. Mallory's stain.
The osseous or primary spiral lamina is a ledge that projects from the modiolus into the osseous canal like the thread of a screw (Fig. 39.7). It is attached to the inner edge of the basilar membrane and ends in a hook-shaped hamulus at the cochlear apex, partly bounding the helicotrema (Fig. 39.4), which is an opening connecting the scala tympani and scala vestibuli. From Rosenthal's canal, many tiny canals, the habenula perforata, radiate through the osseous lamina to its rim; they carry fascicles of the cochlear nerve to the organ of Corti (Fig. 39.8). A secondary spiral lamina projects inwards from the outer cochlear wall towards the osseous spiral lamina and is attached to the outer edge of the basilar membrane. It is most prominent in the lower part of the first turn: the gap between the two laminae increases progressively towards the cochlear apex, which means that the basilar membrane is wider at the apex of the cochlea than at the base. Microstructure of the bony labyrinth
Figure 39.8 Whole-mount preparation of the organ of Corti from a human cochlea, stained with osmium to show the distribution of tissues, including the myelinated axons. (Provided by H Felix, M Gleeson and L-G Johnsson, ENT Department, University of Zurich and GKT School of Medicine, London.)
The wall of the bony labyrinth is lined by fibroblast-like perilymphatic cells and extracellular fibres (Fig. 39.5). The morphology of the cells varies in different parts of the labyrinth. Where the perilymphatic space is narrow, as in the cochlear aqueduct, the cells are reticular or stellate in form, and give off sheet-like cytoplasmic extensions that cross the space. Where the space is wider, as in the scalae vestibuli and tympani of the cochlea and much of the vestibule, the perilymphatic cells on the periosteum and the external surface of the membranous labyrinth are extremely flat, and resemble a squamous epithelium. Elsewhere, on parts of the perilymphatic surface of the basilar membrane, the cells are cuboidal. Bundles of collagen fibres are closely related to the periosteal and labyrinthine aspects of these cells. Composition of inner ear fluids
The space between the bony and membranous labyrinths is filled with perilymph (Figs 39.4, 39.5). The membranous labyrinth is filled with endolymph, a fluid produced by the marginal cells of the stria vascularis and the dark cells of the vestibule (see review by Wangemann & Schacht 1996) (Figs 39.7, 39.9). Whatever their relative contributions, endolymph probably circulates in the labyrinth; it enters the endolymphatic sac, where it is transferred into the adjacent vascular plexus via the specialized epithelium of the sac. Pinocytotic removal of fluid may also occur in other labyrinthine regions. page 666 page 667
page 667 page 668
Figure 39.9 Structure of the cochlear organ of Corti and stria vascularis, showing the arrangement of the various types of cell and their overall innervation. The organization of the inner and outer hair cells and their synaptic connections are depicted below. Afferent nerve terminals are coloured green and efferent fibres purple.
Perilymph was initially considered to be an ultrafiltrate of plasma because of its low protein content. However, it more closely resembles cerebrospinal fluid in ionic composition, particularly in the scala tympani. Its composition is not precisely the same in both cochlear scalae: concentrations of potassium, glucose , amino acids and proteins are greater in the scala vestibuli. This has led to the suggestion that perilymph in the scala vestibuli is derived from plasma via the endothelial boundary of the cochlear blood vessels, whereas the perilymph in the scala tympani contains some cerebrospinal fluid derived from the subarachnoid spaces via the cochlear canaliculus. However, the lack of significant bulk flow suggests that perilymph homeostasis is predominantly locally regulated. Perilymph contains approximately 5 mM K + , 150 mM Na + , 120 mM Cl - and 1.5 mM Ca 2+. Endolymph contains greater K + (150 mM) and Cl - (130 mM) concentrations and lower Na + (2 mM) and Ca 2+ (20µM) concentrations than perilymph. The major differences in ionic composition between the two fluids are important for the function of the inner ear. Displacements of the stereociliary bundles of the sensory cells activate relatively non-specific cationic channels in the stereociliary tips, which allows an influx of cations, particularly K + and Ca 2+, from the endolymph. Hair cells also possess K + channels activated by membrane voltage or intracellular Ca 2+ concentrations, and these allow efflux of K + into the perilymph which bathes their basal and lateral membranes. In addition, synaptic transmission at the base and sides of hair cells depends on the influx of Ca 2+ from the perilymph through voltage-dependent calcium channels in order to release neurotransmitter. Membranous labyrinth
The membranous labyrinth is separated from the periosteum by a space that contains perilymph and a web-like network of fine blood vessels (Figs 39.4, 39.5). It can be divided into two major regions, the vestibular apparatus and the cochlear duct. The vestibular apparatus consists of three membranous semicircular canals which communicate with the utricle, a membranous sac leading into a smaller chamber, the saccule, via the utriculosaccular duct. This Y-shaped duct has a side branch to the endolymphatic duct, which passes to the endolymphatic sac, a small but functionally important expansion situated under the dura of the petrous temporal bone. From the saccule, a narrow canal, the ductus reuniens, leads to the base of the cochlear duct. These various ducts and sacs form a closed system of inter-communicating channels. Endolymph is resorbed into the cerebrospinal fluid from the endolymphatic sac, which therefore provides the site for the drainage of endolymph for the entire membranous labyrinth. The terminal fibres of the vestibular nerve are connected to five distinct areas of specialized sensory epithelium (two maculae and three crests) in the walls of the membranous labyrinth. Maculae are flat plaques of sensory hair cells surrounded by supporting cells, and are found in the utricle and saccule. The crests are ridges bearing sensory hair cells and supporting cells. They are found in the walls of the ampullae near the utricular openings of the three semicircular canals, one for each canal. Utricle
The utricle is the larger of the two major vestibular sacs. It is an irregular, oblong, dilated sac that occupies the posterosuperior region of the vestibule (Fig. 39.4), and contacts the elliptical recess (where it is a blind-ended pouch) and the area inferior to it.
Figure 39.10 Section of the utricular macula from a guinea pig, showing the relative positions of the hair cells and supporting cell nuclei. Semi-thin resin section, toluidine
blue stain. (The inner ear is extremely vulnerable to hypoxia and situated in one of the hardest bones in the body, which means that well-fixed human tissue is rarely obtained for histology. Guinea pigs are one of the most frequently used animal models of human hearing and their inner ear ultrastructure is very similar.) (Provided by RM Walsh, DN Furness and CM Hackney, MacKay Institute of Communication and Neuroscience, Keele University.)
page 668 page 669
Figure 39.11 A, Morphological organization of the saccular and utricular maculae and the relationship of their hair cells to the otolithic membrane. The utricular macula has been tilted in the plane of the page to emphasize that it lies horizontally, whereas the saccular macula lies vertically when the head is in an upright position. Note the different shapes of the maculae, the position of the striola as indicated by a curved line in each case, and the different orientations of their stereociliary bundles. The arrows indicate the excitatory direction of deflection. B, Scanning electron micrograph of a fracture of a utricular macula (guinea pig) showing a Type I hair cell (left) and a Type II hair cell (right). C, The differing innervation patterns of the two types of hair cell. (Provided by DN Furness, MacKay Institute of Communication and Neuroscience, Keele University.)
The macula of the utricle (or utriculus) is a specialized area of neurosensory epithelium lining the membranous wall, and is the largest of the vestibular sensory areas (Fig. 39.10). It is triangular or heart-shaped in surface view and lies horizontally with its long axis orientated anteroposteriorly and its sharp angle pointing posteriorly (Fig. 39.11). It is flat except at the anterior edge, where it is
gently folded in on itself, and it measures c.2.8 mm long by 2.2 mm wide. The mature form of the macula is reached early in development, but in the adult a bulge is often present on the anterolateral border; there is sometimes an indentation at the anteromedial border. The epithelial surface is covered by the otolithic membrane (statoconial membrane), a gelatinous structure in which many small crystals, the otoconia (otoliths, statoliths), are embedded. A curved ridge, the 'snowdrift line', runs along the length of the otolithic membrane. It corresponds to a narrow crescent of underlying sensory epithelium termed the striola, c.0.13µm wide. The density of sensory hair cells in this strip of epithelium is c.20% less than in the rest of the macula. The striola is convex laterally and runs from the medial aspect of the anterior margin in a posterior direction towards, but not reaching, the posterior pole. The part of the macula medial to the striola is called the pars interna and is slightly larger than the pars externa, which is lateral to it. The significance of this area is that the sensory cells are functionally and anatomically polarized towards it (Fig. 39.11). The macula in each utricle is approximately horizontal when the head is in its normal position. Linear acceleration of the head in any horizontal plane will result in the otolithic membrane lagging behind the movement of the membranous labyrinth as a result of the inertia produced by its mass. The membrane thus maximally stimulates one group of hair cells by deflecting their bundles towards the striola whilst inhibiting others by deflecting their bundles away from it. Hence each horizontal movement of the head will produce a specific pattern of firing in the utricular efferents. Saccule
The saccule (or sacculus) is a slightly elongated, globular sac lying in the spherical recess near the opening of the scala vestibuli of the cochlea (Fig. 39.11). As already noted, it is connected to the utricle and endolymphatic duct by the utriculosaccular duct, and to the cochlea by the ductus reuniens, which leaves inferiorly to open into the base of the cochlear duct (Fig. 39.4). The saccular macula is an almost elliptical structure, 2.6 mm long and 1.2 mm at its widest point. Its long axis is orientated anteroposteriorly but, in contrast to the utricular macula, the saccular macula lies in a vertical plane on the wall of the saccule. Its elliptical shape is very slightly distorted by a small anterosuperior bulge. Like the utricular macula, it is covered by an otolithic (statoconial) membrane and possesses a striola, c.0.13 mm wide, which extends along its long axis as an S-shaped strip about which the sensory cells are functionally and anatomically polarized (Fig. 39.11). The part of the macula above the striola is termed the pars interna, and that below it, the pars externa. The operation of the saccule is similar to that of the utricle. However, because of its vertical orientation, the saccule is particularly sensitive to linear acceleration of the head in the vertical plane, and is therefore a major gravitational sensor when the head is in an upright position. It is also particularly sensitive to movement along the anteroposterior axis. Semicircular canals
The lateral, superior and posterior semicircular ducts follow the course of their osseous canals. Throughout most of their length they are securely attached, by much of their circumference, to the osseous walls (p. 462). They are approximately one-quarter of the diameter of their osseous canals (Fig. 39.5). The medial ends of the superior and posterior canals fuse to form a single common duct, the crus commune, before entering the utricle. The lateral end of each canal is dilated to form an ampulla, which lies within the ampulla of the osseous canal. The short segment of duct between the ampullae and utricle is the crus ampullaris. The membranous wall of each ampulla contains a transverse elevation (septum transversum) on the central region of which is a sensory area, the ampullary crest (crista). This is a saddle-shaped ridge that lies transversely across the duct. It is broadly concave on its free edge along most of its length and has a concave gutter (planum semilunatum) at either end between the ridge and the duct wall. Sectioned across the ridge, the crests of the lateral and anterior semicircular canals have smoothly rounded corners; the posterior crest is more angular. A vertical plate of gelatinous extracellular material, the cupula, is attached along the free edge of the crest (Fig. 39.12). It projects far into the lumen of the ampulla, so that movements of endolymph within the duct readily deflect the cupula and therefore also the stereocilia of the sensory cells that are inserted into its base. The three semicircular canals thus detect angular accelerations during tilting or
turning movements of the head in any direction. Microstructure of the vestibular system
The maculae and crests detect the orientation of the head with respect to gravity and changes in head movement by means of the mechanosensitive hair cells that are interspersed among the non-sensory supporting cells in their sensory epithelia. These hair cells are in contact with afferent and efferent endings of vestibular nerve fibres by synapses at their base. The entire epithelium lies on a bed of thick, fibrous connective tissue containing myelinated vestibular nerve fibres and blood vessels. The axons lose their myelin sheaths as they perforate the basal lamina of the sensory epithelium. There are two types of sensory hair cell in the vestibular system, Type I and Type II.
Figure 39.12 Section of an ampullary crest from a 6-month-old human fetus. (Drawn from a section prepared at the Ferens Institute and lent by EW Walls, Professor Emeritus, University College London.)
page 669 page 670
Figure 39.13 Scanning electron micrograph of a stereociliary bundle from the utricle (guinea pig). The stereocilia are arranged in rows of increasing height towards the tallest element, the kinocilium. Deflection in the direction of the kinocilium results in depolarization of the hair cell. (Provided by DN Furness, MacKay Institute of Communication and Neuroscience, Keele University.)
Type I vestibular sensory cells measure c.25µm in length, with a free surface of 6-7µm in diameter. The basal part of the cell does not reach the basal lamina of the epithelium. Each cell is typically bottle-shaped, with a narrow neck and a rather broad, rounded basal portion containing the nucleus (Fig 39.9, 39.11). The apical surface is characterized by 30-50 stereocilia (large, regular microvilli c.0.25µm across) and a single kinocilium (with the typical '9 + 2' arrangement of microtubules characteristic of true cilia). The kinocilium is considerably longer than the stereocilia, and may attain 40µm, whereas the stereocilia are of graded lengths. They are characteristically arranged in regular rows behind the kinocilium in descending order of height, the longest being next to the kinocilium (Fig. 39.13). The kinocilium emerges basally from a typical basal body, with a centriole immediately beneath it. Close to the inner surface of their basal two-thirds, every cell contains numerous synaptic ribbons with associated synaptic vesicles. The postsynaptic surface of an afferent nerve ending encloses the greater part of the sensory cell body in the form of a cup (chalice or calyx) (Fig. 39.9). Efferent nerve fibres make synapses with the external surface of the calyx, rather than directly with the sensory cell. There is much greater variation in the sizes of Type II sensory cells (Fig. 39.14). Some are up to 45µm long, and almost span the entire thickness of the sensory epithelium, whereas others are shorter than Type I cells. They are mostly cylindrical, but otherwise resemble Type I cells in their contents and the presence of an apical kinocilium and stereocilia. However, their kinocilia and stereocilia tend to be shorter and less variable in length. The most striking difference between Type I and II cells is their efferent nerve terminals: Type II cells receive several efferent nerve boutons containing a mixture of small clear and dense-core vesicles around their bases. Afferent endings are small expansions rather than chalices.
Figure 39.14 A, Transmission electron micrograph of human Type I vestibular hair cell (vr) bearing an apical group of stereocilia (st) seen in a vertical section through the macula. Note that the hair cell is bottle-shaped, and that much of it is enclosed in the calyceal ending (c) of an afferent nerve terminal. Abbreviation: sc, supporting cells. B, Transmission electron micrograph of human Type II vestibular hair cell: a bouton-type afferent nerve terminal is in contact with the basal part. (Provided by H Felix, M Gleeson and L-G Johnsson, ENT Department, University of Zurich and GKT School of Medicine, London.)
Each sensory cell is structurally and functionally polarized (Fig. 39.13). Deflection of the hair bundle towards the kinocilium results in depolarization of the hair cell, and increases the rate of neurotransmitter release from its base. Deflection away from the kinocilium hyperpolarizes the hair cell and reduces the release of neurotransmitter. The hair cells have specific orientations within each sensory organ (Fig. 39.13). In the maculae, they are arranged symmetrically on either side of the striola. In the utricle, the kinocilia are positioned on the side of the sensory cell nearest to the striola. In the saccule, they are furthest from it. In the ampullary crests, the cells are orientated with their rows of stereocilia at right angles to the long axis of the semicircular duct. In the lateral crest the kinocilia are on the side towards the utricle, whereas in the anterior and posterior crests they are away from it. These different arrangements are important functionally, because any given acceleration of the head maximally depolarizes one group of hair cells and maximally inhibits a complementary set, thus providing a unique representation of the magnitude and orientation of any movement (for further details, see Furness 2002). The Type I and II sensory cells are set within a matrix of supporting cells that reach from the base of the epithelium to its surface, and form rosettes round the sensory cells, as seen in surface view. Although their form is irregular, they can easily be recognized by the position of their nuclei, which tend to lie below the level of sensory cell nuclei and just above the basal lamina (Fig. 39.10). The apices of the supporting cells are attached by tight junctions to neighbouring cells to produce the reticular lamina, a composite layer which forms a plate that is relatively impermeable to cations other than via the mechanosensitive transduction channels of the hair cells. The otolithic membrane is a layer of extracellular material with a complex structure. It can be divided into two strata. The external layer is composed of otoliths or otoconia, which are barrel-shaped crystals of calcium carbonate with angular ends, up to 30µm long, and heterogeneous in distribution (Figs 39.9, 39.10, 39.11, 39.13). They are attached to a more basal gelatinous layer into which the stereocilia and kinocilia of the sensory cells are inserted. The gelatinous material consists largely of glycosaminoglycans associated with fibrous protein. UPDATE Date Added: 17 April 2006 Abstract: Spatial and temporal coding by vestibular semicircular canal afferents Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=15845995&query_hl=56&itool=pubmed_docsum Determinants of spatial and temporal coding by semicircular canal afferents. Highstein SM, Rabbitt RD, Holstein GR, Boyle RD: J Neurophysiol 93:2359-2370, 2005. Epley's manoeuvre page 670 page 671
Benign paroxysmal positional vertigo is a condition in which a sensation of rotation with associated nystagmus is induced by adopting a particular position (with the abnormal ear dependent). It is believed that calcium carbonate crystals from the otoliths become freed from their hair cells and, in certain positions, drop into the ampulla of the posterior semicircular canal, possibly becoming adherent to the cupula and rendering it gravity-sensitive. In certain positions the alignment of the axis of the posterior semicircular canal with gravity results in the displacement of the cupula and the activation of the vestibulo-ocular reflex, resulting in compensatory nystagmoid eye movements in response to apparent head movements. Epley's canalith repositioning procedure relies on the adoption of a series of body postures designed to allow the aberrant crystals (or canaliths) to float out of the posterior semicircular canal and to stick to the wall of the vestibule. Cure rates in excess of 80% have been recorded and the procedures have largely superseded surgical procedures designed to denervate the ampulla of the posterior semicircular canal (cingular neurectomy) or obliterate the canal completely. Endolymphatic duct and sac
The endolymphatic duct runs in the osseous vestibular aqueduct and becomes dilated distally to form the endolymphatic sac. This is a structure of variable size,
which may extend through an aperture on the posterior surface of the petrous bone to end between the two layers of the dura on the posterior surface of the petrous temporal bone near the sigmoid sinus (Fig. 39.4). The surface cells throughout the entire endolymphatic duct resemble those lining the nonspecialized parts of the membranous labyrinth, and consist of squamous or low cuboidal epithelium. The epithelial lining and subepithelial connective tissue become more complex where the duct dilates to form the endolymphatic sac. An intermediate or rugose segment and a distal sac can be distinguished. In the intermediate segment, the epithelium consists of light and dark cylindrical cells. Light cells are regular in form, and have numerous long surface microvilli with endocytic invaginations between them and large clear vesicles in their apical region. In contrast, dark cells are wedge-shaped, and have a narrow base, few apical microvilli and dense, fibrillar cytoplasm. The endolymphatic sac has important roles in the maintenance of vestibular function. Endolymph produced elsewhere in the labyrinth is absorbed in this region, probably mainly by the light cells. Damage to the sac, or blockage of its connection to the rest of the labyrinth, causes endolymph to accumulate; this produces hydrops, which affects both vestibular and cochlear function. The epithelium is also permeable to leukocytes, including macrophages which can remove cellular debris from the endolymph, and to various cells of the immune system that contribute antibodies to this fluid. A unique positive electrical potential exists in the endolymphatic spaces, varying from +77 mV in the cochlear duct near the stria vascularis to c.+44 mV in the utricle. This is additional to the normal resting potentials of the receptor cells, so that there is a very considerable potential difference across their membranes. This undoubtedly contributes to the extreme sensitivity to mechanical deformation of the labyrinthine sensory receptors. Cochlear duct
The cochlear duct is a spiral tube that runs within the bony cochlea (Figs 39.4, 39.6, 39.7, 39.8). The osseous spiral lamina projects for part of the distance between the modiolus and the outer wall of the cochlea and is attached to the inner edge of the basilar membrane. The endosteum of the outer wall is thickened to form a spiral cochlear ligament, which projects inwards as a triangular basal crest attached to the outer rim of the basilar membrane. Immediately above this is a concavity, the external spiral sulcus (sulcus spiralis externus), above which the thick, highly vascular periosteum projects as a spiral prominence. Above the prominence is a specialized, thick epithelial layer, the stria vascularis. A second, thinner vestibular membrane, Reissner's membrane, extends from the thickened endosteum on the osseous spiral lamina to the outer wall of the cochlea, where it is attached above the stria. Reissner's membrane consists of two layers of squamous epithelial cells separated by a basal lamina. The side facing the scala vestibuli bears flattened perilymphatic cells, with tight junctions between them, creating a diffusion barrier. The endolymphatic side is lined by squamous epithelial cells; these are also joined by tight junctions and are involved in ion transport. The canal thus enclosed between the scala tympani and the scala vestibuli is the cochlear duct (Fig. 39.7). It is triangular in cross-section throughout the length of the cochlea. Its closed upper end, the lagaena, is attached to the cupula. The lower end of the duct turns medially, narrowing into the ductus reuniens, and connects with the saccule (Fig. 39.4). The organ of Corti, the sensory epithelium of the cochlea, sits upon the basilar membrane. The apices of the sensory hair cells and supporting cells it contains are joined by tight junctions to form the reticular lamina. The diffusion barriers which line the cochlear duct ensure that the apices of the sensory hair cells are bathed by endolymph, whereas their lateral and basal regions are bathed in perilymph. The stria vascularis lies on the outer wall of the cochlear duct, above the spiral eminence (Fig. 39.7). It has a special stratified epithelium containing a dense intraepithelial capillary plexus and three cell types: superficial marginal, dark or chromophil cells; intermediate light, or chromophobe cells; and basal cells. The endolymphatic surface consists only of the apices of marginal cells. The intermediate and basal cells lie deeper within the stria and send cytoplasmic processes towards the surface, between the deeper parts of the marginal cells. The long descending cytoplasmic processes of the marginal dark cells and the
ascending processes of the intermediate and basal cells envelop the intraepithelial capillaries. The stria vascularis is involved in ion transport, and it helps to produce the unusual ionic composition of endolymph. It is the source of the large positive endocochlear electrical potential, maintenance of which is directly dependent upon adequate oxygenation of the epithelial cells, which is provided by the intraepithelial capillary plexus. The osseous spiral lamina consists of two plates of bone, between which are canals for the cochlear nerve filaments. On the upper plate, the periosteum is thickened to form the spiral limbus (limbus laminae spiralis) (Fig. 39.7). It ends externally in the internal spiral sulcus, which in section is shaped like a C. Its upper part, the overhanging limbic edge, is the vestibular labium and the lower tapering part is the tympanic labium which is perforated by small holes (the habenula perforata) for branches of the cochlear nerve (Fig. 39.8). The upper surface of the vestibular labium is crossed at right angles by furrows, separated by numerous elevations, the auditory teeth (dentes acustici) (Fig. 39.9). The limbus is covered by a layer that appears superficially to be squamous epithelium; however, only the cells over the 'teeth' are flat, and those in the furrows are flaskshaped interdental cells. The epithelium is continuous with the epithelium in the internal spiral sulcus and on the inferior surface of Reissner's membrane. During development the interdental cells secrete some of the material that forms the tectorial membrane. Basilar membrane
The basilar membrane stretches from the tympanic lip of the osseous spiral lamina to the basal crest of the spiral ligament (Figs 39.7, 39.9). It consists of two zones. The thin zona arcuata stretches from the spiral limbus to the bases of the outer pillar cells and supports the organ of Corti. It is composed of compact bundles of small (8-10 nm diameter) collagenous filaments, mainly radial in orientation. The outer thicker zona pectinata starts beneath the bases of the outer pillar cells and is attached to the crista basilaris. The basilar membrane is trilaminar in the zona pectinata, but the upper and lower layers fuse at its attachment to the crista basilaris. The length of the basilar membrane is c.35 mm; its width increases from 0.21 mm basally to 0.36 mm at its apex, accompanied by corresponding narrowing of the osseous spiral lamina and a decrease in the thickness of the basal crest. The lower or tympanic surface of the basilar membrane is covered by a layer of vascular connective tissue and elongated perilymphatic cells. One vessel, the spiral vessel (vas spirale), is larger; it lies immediately below the tunnel of Corti. Organ of Corti page 671 page 672
Figure 39.15 Scanning electron micrograph of a portion of the organ of Corti (guinea pig) dissected to expose the outer row of outer hair cells and their attendant Deiters' cells with narrow phalangeal processes. The stereociliary bundles of two rows of outer hair cells are visible above the reticular lamina. (Provided by DN Furness, MacKay Institute of Communication and Neuroscience, Keele University.)
The organ of Corti consists of a series of epithelial structures that lie on the zona arcuata of the basilar membrane (Figs 39.7, 39.9, 39.15, 39.16). The more
central of these structures are two rows of cells, the internal and external pillar cells. The bases of the pillar cells are expanded, and rest contiguously on the basilar membrane, but their rod-like cell bodies are widely separated. The two rows incline towards each other and come into contact again at the heads of the pillars, enclosing between them and the basilar membrane the tunnel of Corti, which has a triangular cross-section (Fig. 39.9). Internal to the inner pillar cells is a single row of inner hair cells. External to the outer pillar cells are three or four rows of outer hair cells. The bases of the outer hair cells are cupped by supporting cells called Deiters' cells, except for a gap where cochlear axons synapse with them. The apical ends of the hair cells and apical processes of the supporting cells form a regular mosaic called the reticular lamina, which is covered by the tectorial membrane, a gel-like structure projecting from the spiral limbus. A narrow gap separates the tectorial membrane from the reticular lamina except where the apical stereocilia of the outer hair cells project to make contact with it. In addition to the tunnel of Corti, other intercommunicating spaces, the spaces of Nuel, surround the outer hair cells. This entire intercommunicating complex of spaces of Nuel and tunnel of Corti is filled with perilymph, which diffuses through the matrix of the basilar membrane. The fluid in these spaces is also sometimes called the cortilymph; it is possible that minor alterations in perilymphatic composition occur within it, because it is exposed to the activities of synaptic endings and specialized excitable cells. Each pillar cell has a base or crus, an elongated scapus (rod) and an upper end or caput (head) (Fig. 39.9); each crus and caput are in contact, but the scapi are separated by the tunnel of Corti. Electron microscopy shows many microtubules, 30 nm in diameter, arranged in linked parallel bundles of 2000 or more in the scapus, originating in the crus and diverging above the scapus to terminate in the head region. The nucleus is situated in the foot-like expansion resting on the basal lamina. There are almost 6000 internal pillar cells. Their bases rest on the basilar membrane near the tympanic lip of the internal spiral sulcus, and their bodies form an angle of c.60° with the basilar membrane. Their heads resemble the proximal end of the ulna, with deep concavities for the heads of the outer pillar cells, which they overhang. There are almost 4000 external pillar cells. They are longer and more oblique than the internal pillar cells, and form an angle of c.40° with the basilar membrane. Their heads fit into the concavities on the heads of the inner pillar cells and project externally as thin processes that contact the processes of the Deiters' cells. The distances between the bases of the internal and external pillar cells increase from the cochlear base to its apex, whereas the angles they make with the basilar membrane diminish. Cochlear hair cells are the sensory transducers of the cochlea: collectively they detect the amplitude and frequency of the sound waves that enter the cochlea (Fig. 39.9). All cochlear hair cells have a common pattern of organization. They are elongated cells with a group of modified apical microvilli or stereocilia (which contain parallel arrays of actin filaments) and a group of synaptic contacts with cochlear axons at their rounded bases (Figs 39.16, 39.17, 39.18). The inner hair cells form a single row along the inner edge of inner pillar cells (and the spiral tunnel), whereas the outer hair cells are arranged in three or, in some regions of the human cochlea, in four or even five rows, interspersed with supporting cells (Fig. 39.16). These two groups have distinctive roles in sound reception; the differences in their detailed structure reflect this functional divergence. There are c.3500 inner hair cells and c.12,000 outer hair cells. The two sets of hair cells lean towards each other apically at about the same angles as the neighbouring inner and outer pillar cells. The geometric arrangement of these cells is very precise, and this pattern is closely related to the sensory performance of the cochlea.
page 672 page 673
Figure 39.16 Scanning electron micrographs of the surface of an organ of Corti from a human cochlea. A, Low-power view showing a single row of inner hair cells (left) and three rows of outer hair cells, with additional rows in places. B, Higher magnification of stereociliary bundles of outer hair cells. (Provided by H Felix, M Gleeson and L-G Johnsson, ENT Department, University of Zurich and GKT School of Medicine, London.)
Figure 39.17 Scanning electron micrograph of the surface of the organ of Corti, the reticular lamina (guinea pig). Three rows of V-shaped stereociliary bundles can be seen protruding from the apices of the outer hair cells. They are separated from the single row of inner hair cells (which have relatively linear stereociliary bundles) by the
apices of the inner pillar cells. (Provided by DN Furness, MacKay Institute of Communication and Neuroscience, Keele University.)
Figure 39.18 Scanning electron micrograph of stereociliary bundle of one outer hair cell (guinea pig) showing three rows of stereocilia increasing in height. Deflection of the stereocilia in the direction of the tallest row results in depolarization of the hair cell. Microvilli can be seen on the surface of Deiters' cells (front right). (Provided by DN Furness, MacKay Institute of Communication and Neuroscience, Keele University.)
The inner hair cells are pear-shaped and slightly curved; the narrower end is directed towards the surface of the organ of Corti and the wider basal end is positioned some distance above the inner end of the basilar membrane (Fig. 39.9). The inner hair cells are surrounded by inner border cells and by inner phalangeal cells, which are attached externally to the heads of the inner pillar cells. The flat apical surface of each inner hair cell is elliptical when viewed from above, its long axis directed in the direction of the row of hair cells (Fig. 39.17). The breadth of the apex exceeds that of the inner pillar cells, so that each inner hair cell is related to more than one inner pillar cell. The apex bears 50-60 stereocilia, arranged in several ranks of progressively ascending height, the tallest on the strial side. The tips of the shorter rows are connected diagonally to the sides of the adjacent taller stereocilia by thin filaments called tip links; each stereocilium is also connected to all its neighbours by lateral links. The length of a stereociliary row varies along the length of the cochlea, being longest at the apex and shortest at its base. The stereociliary bases insert into a transverse lamina of dense fibrillar material, the cuticular plate, which lies immediately beneath the apical surface of each inner hair cell. The cuticular plate includes a small aperture containing a basal body. During development, a kinocilium containing microtubules is anchored here, a condition which persists in vestibular hair cells. At its base, each inner hair cell forms ten or more synaptic contacts with afferent endings, each being marked by a presynaptic structure similar to the ribbon synapses (see Fig. 4.8C) of the retina. Occasionally, an efferent synapse makes direct contact with a hair cell base, but these are usually presynaptic to the terminal expansions of afferent endings, rather than to the hair cell itself. Outer hair cells are long cylindrical cells which are nearly twice as tall as the inner hair cells (Figs 39.9, 39.11, 39.16). There is a gradation of length: the outermost row is longest in any one cochlear region, and those of the cochlear apex are taller than those of the base. They are surrounded by the apical or phalangeal processes of the Deiters' cells or, on the internal side of the inner row, by the heads of the outer pillar cells. The stereocilia, which may number up to 100 per cell, are arranged in three rows of graded heights; the longest is on the outer side. The rows are arranged in the form of a V or W depending on cochlear region, the points of the angles directed externally. The stereocilia are also graded in length
according to cochlear region: those of the cochlear base are shortest. Like those of inner hair cells, the stereocilia possess tip links and other filamentous connections with their neighbours, and are inserted at their narrow bases into a cuticular plate. The tallest stereocilia are embedded in shallow impressions on the underside of the tectorial membrane. The rounded nucleus is positioned near the base of the cell. Below the nucleus are a few ribbon-like synapses associated with afferent endings of the cochlear nerve. The latter are fewer in number and smaller than the cluster of efferent boutons that contact the base of the cell. The neurotransmitter at the afferent synapse in both inner and outer hair cells is probably glutamate, whereas that of the efferent endings is acetylcholine, although other neurotransmitters or neuromodulators have been demonstrated. Deiters' or phalangeal cells lie between the rows of outer hair cells. Their expanded bases lie on the basilar membrane and their apical ends partially envelop the bases of the outer hair cells (Fig. 39.9, 39.15). Each has a finger-like (phalangeal) process that extends diagonally upwards between the hair cells to the reticular membrane, where it forms a plate-like expansion that fills the gaps between hair cell apices. Five or six rows of columnar supporting cells or external limiting cells, such as Hensen's cells and Claudius' cells lie external to the Deiters' cells (Fig. 39.9). The apices of the hair cells and supporting cells that form the reticular lamina are linked by tight junctions, desmosomes and extensive gap junctions which couple them electrically (Fig. 39.9). This arrangement is significant for two reasons. The reticular lamina creates a highly impermeable barrier to the passage of ions other than via the mechanotransducer channels in the stereociliary membranes. It also forms a rigid support between the apices of the hair cells, coupling them mechanically to the movements of the underlying basilar membrane, which causes lateral shearing movements between the stereocilia and the overlying tectorial membrane. If there is hair cell loss as a result of trauma such as excessive noise or ototoxic drugs, the supporting cells expand rapidly to fill the gap, disturbing the regular pattern of the reticular lamina (phalangeal scars), but restoring its function. Tectorial membrane
The tectorial membrane overlies the sulcus spiralis internus and organ of Corti and is a stiff, gelatinous plate (Figs 39.7, 39.9). It contains collagen types II, V and IX, interspersed with glycoproteins (tectorins), which contribute approximately half of the total protein. In transverse section, the tectorial membrane has a characteristic shape. The underside is nearly flat and the upper surface is convex, and it is thin on the modiolar side where it is attached to the vestibular labium of the spiral limbus. Its outer part forms a thickened ridge, overhanging the edge of the reticular lamina. The lower surface is relatively smooth, except where the stereocilia of the outer hair cells are embedded in the membrane, leaving a pattern of W- or V-shaped impressions, an S-shaped ridge called Hensen's stripe, which projects towards the stereocilia of the inner hair cells. The interdental cells of the spiral limbus are believed to secrete the membrane.
© 2008 Elsevier
VASCULAR SUPPLY ARTERIES The inner ear is principally supplied by the labyrinthine artery. The stylomastoid branch of either the occipital artery or the posterior auricular artery also supplies the semicircular canals. Labyrinthine artery
page 673 page 674
The labyrinthine artery arises from the basilar artery, or sometimes from the anterior inferior cerebellar artery. It divides at the bottom of the internal auditory meatus into cochlear and vestibular branches. The cochlear branch divides into 12-14 twigs, which traverse the canals in the modiolus and are distributed as a capillary plexus to the spiral lamina, basilar membrane, stria vascularis and other cochlear structures. Vestibular arterial branches supply the utricle, saccule and semicircular ducts.
VEINS The veins draining the vestibule and semicircular canals accompany the arteries. They receive the cochlear veins at the base of the modiolus, and form the labyrinthine vein, which ends in the posterior part of the superior petrosal sinus or in the transverse sinus. A small vessel from the basal cochlear turn traverses the cochlear canaliculus to join the internal jugular vein. For details of the microvasculature of the cochlea of man and other mammals, see Axelsson (1988).
© 2008 Elsevier
INNERVATION UPDATE Date Added: 01 December 2004 Abstract: Connections between the facial, vestibular and cochlear nerve bundles within the internal auditory canal. Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=15255963 Connections between the facial, vestibular and cochlear nerve bundles within the internal auditory canal.
VESTIBULOCOCHLEAR NERVE (Fig. 39.19) The vestibulocochlear nerve emerges from the pontocerebellar angle. It courses through the posterior cranial fossa to enter the petrous temporal bone via the internal acoustic meatus, where it divides into an anterior trunk, the cochlear nerve, and a posterior trunk, the vestibular nerve. Both contain the centrally directed axons of bipolar neurones, the cell bodies of which are situated close to their peripheral terminals, together with a smaller number of efferent fibres that arise from brain stem neurones and terminate on cochlear and vestibular sensory cells. In audiological practice, it is important to distinguish between intratemporal and intracranial lesions. It is relevant to note therefore that this surgical distinction does not correlate with the precise anatomical description of peripheral and central portions of the auditory and vestibular systems. Clinically, the term 'peripheral auditory lesion' is used to describe lesions peripheral to the spiral ganglion, and the term 'peripheral vestibular disturbance' includes lesions of the vestibular ganglion and the entire vestibular nerve. Furthermore, the intratemporal portion of the vestibulocochlear nerve in humans consists of two histologically distinct portions: a central glial zone adjacent to the brain stem, and a peripheral or non-glial zone. In the glial zone the axons are supported by central neuroglia, whereas in the non-glial zone they are ensheathed by Schwann cells. The nonglial zone extends into the cerebellopontine angle medial to the internal acoustic meatus in more than 50% of human vestibulocochlear nerves. The central pathways of the vestibular and cochlear nerves are described in Chapter 24 (pp. 436, 436). Intratemporal vestibular nerve
The maculae and crests are innervated by dendrites of bipolar neurones in the vestibular (Scarpa's) ganglion situated in the trunk of the nerve within the lateral end of the internal auditory meatus (Fig. 39.20). The peripheral processes of the ganglion cells are aggregated into definable nerves, each with a specific distribution (Fig. 39.1). The main nerve divides at and within the ganglion into superior and inferior divisions, which are connected by an isthmus. The superior division, the larger of the two, passes through the small holes in the superior vestibular area to supply the ampullary crests of the lateral and anterior semicircular canals via the lateral and anterior ampullary nerves, respectively. A secondary branch of the lateral ampullary nerve supplies the macula of the utricle; however, the greater part of the utricular macula is innervated by the utricular nerve, which is a separate branch of the superior division. Another branch of the superior division, Voit's nerve, supplies part of the saccule. The inferior division of the vestibular nerve passes through small holes in the inferior vestibular area to supply the remainder of the saccule and the posterior ampullary crest via saccular and singular branches, respectively; the latter passes through the foramen singulare. Occasionally, a very small supplementary or accessory branch innervates the posterior crest; it is probably a vestigial remnant of the crista neglecta, an additional area of sensory epithelium found in some other mammals but seldom in man.
Figure 39.19 Human vestibulocochlear nerve, in transverse section. On the left, the cochlear nerve (seen as a comma-shaped profile) abuts the inferior division of the vestibular nerve (right). The singular nerve is a separate fascicle between the superior and inferior divisions of the vestibular nerve. (Provided by H Felix, M Gleeson and LG Johnsson, ENT Department, University of Zurich and GKT School of Medicine, London.)
Figure 39.20 A portion of a human vestibular ganglion, showing neuronal perikarya, myelinated axons and small blood vessels. Toluidine blue stained resin section. (Provided by H Felix, M Gleeson and L-G Johnsson, ENT Department, University of Zurich and GKT School of Medicine, London.)
Afferent and efferent cochlear fibres are also present in the inferior division of the vestibular nerve, but leave at the anastomosis of Oort to join the main cochlear nerve (see review by Warr 1992). Another anastomosis, the vestibulofacial anastomosis, is situated more centrally between the facial and vestibular nerves, and is the point at which fibres originating in the intermediate nerve pass from the vestibular nerve to the main trunk of the facial nerve. UPDATE Date Added: 08 July 2005 Abstract: Anatomohistologic study of von Oort's vestibulocochlear anastomosis. Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=15545928&query_hl=8 Anatomohistologic study of von Oort's vestigulocochlear anastomosis. Full article in French. There are approximately 20,000 fibres in the vestibular nerve, of which 12,000 travel in the superior division and 8000 travel in the inferior division. The distribution of fibre diameters is bimodal, with peaks at 4µm and 6.5µm. The smaller fibres go mainly to the Type II hair cells and the larger fibres tend to supply the Type I hair cells. In addition to the afferents, efferent and autonomic fibres have been identified. Efferent fibres synapse exclusively with the afferent calyceal terminals around Type I cells and usually with the afferent boutons on
Type II cells, although a few are in direct contact with the cell bodies of Type II cells. The autonomic fibres do not contact vestibular sensory cells, but terminate beneath the sensory epithelia. Two distinct sympathetic components have been identified in the vestibular ganglion: a perivascular adrenergic system derived from the stellate ganglion, and a blood vessel-independent system derived from the superior cervical ganglion. Vestibular ganglion
page 674 page 675
The cell bodies of the bipolar neurones that contribute to the vestibular nerve vary considerably in size: their circumferences range from 45 to 160µm (Felix et al 1987). No topographically ordered distribution relating to size has been found. The cell bodies are notable for their abundant granular endoplasmic reticulum, which in places forms Nissl bodies, and prominent Golgi complexes (Fig. 39.20). They are covered by a thin layer of satellite cells and are often arranged in pairs, closely abutting each other so that only a thin layer of endoneurium separates the adjacent coverings of satellite cells. This arrangement has led to speculation that ganglion cells may affect each other directly by electrotonic spread (ephaptic transmission: see Felix et al 1987) (Fig. 39.20). Anatomy of balance and posture
The vestibular system provides precise information about the orientation of the head in three-dimensional space and its rate and direction of movement. It consists of two otolithic organs, the utricle and the saccule, which detect linear acceleration due to gravitational pull and the direction of other linear accelerations, and three semicircular canals, which detect angular accelerations and hence head rotations. The vestibular labyrinths on each side of the head are arranged symmetrically with respect to each other, so ensuring that any movement of the head will result in a unique pattern of nerve input to the brain. The stereocilia in the apical hair bundles of the mechanosensitive hair cells in each of these organs are embedded in an overlying accessory gel-like structure, the otolithic membrane (in the utricle and the saccule) and the cupula (in the semicircular canals). Their apical surfaces are bathed in endolymph: tight junctional complexes between the apices of the hair cells and their adjacent supporting cells separate the endolymph from the perilymph that bathes their basolateral surfaces. Deflection of the stereocilia (caused by displacements of their overlying accessory membranes by fluid movements in the membranous labyrinth) produces either an increased or decreased rate of opening of the mechanotransduction channels at their tips, depending on whether they are deflected towards or away from the tallest row. This in turn results in signals travelling along the vestibular portion of the vestibulocochlear nerve to the brain for comparison with visual and somatosensory signals, which also signal the position of the head in space (for a more detailed account, see Furness 2002). Intratemporal cochlear nerve
The cochlear nerve connects the organ of Corti to the cochlear and related nuclei of the brain stem. The cochlear nerve lies inferior to the facial nerve throughout the internal acoustic meatus. It becomes intimately associated with the superior and inferior divisions of the vestibular nerve, which are situated in the posterior compartment of the canal, and leaves the internal acoustic meatus in a common fascicle (Fig. 39.6). There are approximately 30-40 000 nerve fibres in the human cochlear nerve (for review, see Nadol 1988). Their fibre diameter distribution is unimodal, and ranges from 1 to 11µm, with a peak at 4-5µm. Functionally, the nerve contains both afferent and efferent somatic fibres, together with adrenergic postganglionic sympathetic fibres from the cervical sympathetic system. Afferent cochlear innervation
The afferent fibres are myelinated axons with bipolar cell bodies that lie in the spiral ganglion in the modiolus (Fig. 39.21). There are two types of ganglion cell: most (90-95%) are large Type I cells, the remainder are smaller Type II cells (see reviews by Nadol 1988, Eybalin 1993). I cells contain a prominent spherical nucleus, abundant ribosomes and many mitochondria; in many mammals (although possibly not in humans) they are surrounded by myelin sheaths. In contrast, Type II cells are smaller, always unmyelinated, and have a lobulated
nucleus. The cytoplasm of Type II cells is enriched with neurofilaments, but has fewer mitochondria and ribosomes than Type I cells. Each inner hair cell is in synaptic contact with the unbranched peripheral processes of approximately 10 Type I ganglion cells. The processes of Type II ganglion cells diverge within the organ of Corti and innervate more than 10 outer hair cells. The peripheral and central processes of Type I ganglion cells are relatively large in diameter and are myelinated, whereas those of Type II are smaller and unmyelinated. The peripheral processes of both types of cell radiate from the modiolus into the osseous spiral lamina, where the Type I axons lose their myelin sheaths before entering the organ of Corti through the habenula perforata. Three distinct groupings of afferent fibres have been identified: inner radial, basilar and outer spiral fibres (Fig. 39.22).
Figure 39.21 Transmission electron micrograph showing several Type II ganglion cells and nerve fibres in a human spiral ganglion. Note the absence of myelin from the surrounding sheaths of the ganglion cells. (Provided by H Felix, M Gleeson and L-G Johnsson, ENT Department, University of Zurich and GKT School of Medicine, London.)
Inner radial fibres The inner radial fibre group consists of the majority of afferent fibres. They run directly in a radial direction to the inner hair cells, each of which receives endings from several of these fibres. Basilar fibres Basilar fibres are afferent to the outer hair cells and take an independent spiral course, turning towards the cochlear apex near the bases of the inner hair cells. They run for a distance of about five pillar cells before turning radially again and crossing the floor of the tunnel of Corti, often diagonally, to form part of the outer spiral bundle. Outer spiral bundles The afferent fibres of the bundles of the outer spiral group course towards the basal part of the cochlea, continually branching off en route to supply several outer hair cells. The outer spiral bundles also contain efferent fibres (see below). Efferent cochlear fibres
The efferent nerve fibres in the cochlear nerve are derived from the olivocochlear system (see reviews by Warr 1992, Guinan 1996). Within the modiolus, the efferent fibres form the intraganglionic spiral bundle, which may be one or more discrete groups of fibres situated at the periphery of the spiral ganglion (Fig. 39.22). There are two main groups of olivocochlear efferents: lateral and medial. The lateral efferents come from small neurones in and near the lateral superior olivary nucleus and arise mainly, but not exclusively, ipsilaterally. They are organized into inner spiral fibres that run in the inner spiral bundle before terminating on the afferent axons that supply the inner hair cells. The medial efferents originate from larger neurones in the vicinity of the medial superior olivary nucleus, and the majority arise contralaterally. They are myelinated and cross the tunnel of Corti to synapse with the outer hair cells mainly by direct contact with their bases, although a few synapse with the afferent terminals. The efferent innervation of the outer hair cells decreases along the organ of Corti from cochlear base to apex, and from the first (inner) row to the third. The efferents use acetylcholine, !-aminobutyric acid (GABA), or both as their neurotransmitter. They may also contain other neurotransmitters and neuromodulators. page 675 page 676
Figure 39.22 The innervation of the organ of Corti. The ganglion cells that give rise to the sensory nerve fibres include those related to the inner hair cells (dark green) and others innervating the outer hair cells (light green). Efferent fibres are depicted in purple. There is a great contrast between the convergent afferent innervation of the inner hair cells (c.10 fibres to each cell) and the divergent supply of the outer hair cells (one afferent fibre to c.10 cells). This illustration is a simplified view of the complex innervation of the organ of Corti (see the text for further details).
Activity of the medial efferents inhibits cochlear responses to sound: the strength of the activity grows slowly with increasing sound level. They are believed to
modulate the micromechanics of the cochlea by altering the mechanical responses of the outer hair cells, thus changing their contribution to frequency sensitivity and selectivity. The lateral efferents related to the inner hair cells also respond to sound. They appear to modify transmission through their postsynaptic action on inner hair cell afferents. The cholinergic fibres may excite the radial fibres, whilst those containing GABA may inhibit them, although their role is less well understood than that of the medial efferents (see review by Guinan 1996). Autonomic cochlear innervation
Autonomic nerve endings appear to be entirely sympathetic. Two adrenergic systems have been described within the cochlea: a perivascular plexus derived from the stellate ganglion and a blood vessel-independent system derived from the superior cervical ganglion. Both systems travel with the afferent and efferent cochlear fibres and seem to be restricted to regions away from the organ of Corti. The sympathetic nervous system may cause primary and secondary effects in the cochlea by remotely altering the metabolism of various cell types and by influencing the blood vessels and nerve fibres with which it makes contact. Anatomy of auditory reception page 676 page 677
Figure 39.23 The principal activities of the peripheral auditory apparatus. For clarity, the cochlea is depicted as though it had been uncoiled. The points of maximal stimulation of the basilar membrane by high frequency (blue) and low frequency (red) vibrations, together with their transmission pathways through the external and middle ear, are also indicated.
Sounds waves entering the external ear are converted into electrical signals in the cochlear nerve by the peripheral auditory system (Fig. 39.23). Vibrations in the air column in the external acoustic meatus cause a comparable set of vibrations in the tympanic membrane and auditory ossicles. The chain of ossicles acts as a lever which increases the force per unit area at the round window by 1.2 times, whilst the reduction in size of the round window compared with the tympanic membrane increases the force per unit area of the oscillating surface a further 17 times. This overcomes the inertia of the cochlear fluids and produces in them
pressure waves that are conducted almost instantaneously to all parts of the basilar membrane. The latter varies continuously in width, mass and stiffness from the basal to the apical end of the cochlea. Each part of the basilar membrane vibrates, but only the region tuned to a specific frequency will respond maximally to a pure tone entering the ear. A wave of mechanical motion, the travelling wave, is propagated along the basilar membrane to the position where it responds maximally and then dies away again. With increasing frequency, the locus of maximum amplitude moves progressively from the apical to the basal end of the cochlea. The pattern of vibrations in the basilar membrane thus varies with the intensity and frequency of the acoustic waves reaching the perilymph. Because of the arrangement of the hair cells on the basilar membrane, these oscillations generate a largely transverse shearing force between the outer hair cells and the overlying tectorial membrane (in which the apices of the hair cell stereocilia are embedded). This movement depends on the mechanical properties of the entire organ of Corti, including its cytoskeleton, which stiffens this structure. The inner hair cell stereocilia, which probably do not touch the tectorial membrane although they come very close to it, are likely to be stimulated by local movements of the endolymph. Displacement of the stereociliary bundle of a hair cell opens mechanotransducer channels near the tips of its stereocilia, and this allows potassium and calcium ions from the endolymph to enter the hair cell (see overview by Fettiplace 2002). This induces a depolarizing receptor potential and the release of neurotransmitter onto the cochlear afferents at the base of the cell. In this way a specific group of auditory axons is activated at the position of maximal basilar membrane vibration. The mechanical behaviour of the basilar membrane is responsible for a rather broad discrimination between different frequencies (passive tuning, see overview by Ashmore 2002), but fine frequency discrimination in the cochlea appears to be related to physiological differences between the hair cells. Individual tuning of hair cells may result from differences in shape, stereociliary length, or possibly variations in the molecular composition of sensory membranes, and may have a role in cochlear amplification (active tuning). The activity of the outer hair cells appears to play an important part in regulating inner hair cell sensitivity at specific frequencies. Outer hair cells can change length when stimulated electrically at frequencies of many thousands of cycles per second. The rapidity of these changes in length indicates a novel type of motile mechanism, which is believed to depend on conformational changes in proteins located in the plasma membrane of the cells. When the membrane potential of the outer hair cells changes, they generate forces along their axes. When the mechanotransducer channels open, they are thought to oppose the viscous forces which tend to damp down the vibration of the cochlear partition, and adjust the mechanics of the organ of Corti on a cycle-by-cycle basis. Alternatively they may alter the mechanics of the partition more slowly under the influence of the efferent pathway. At a particular frequency, an increase in the intensity of stimulus is signalled by an increase in the rate of discharge in individual cochlear axons. At greater intensities it is signalled by the number of activated cochlear axons (recruitment). The respective roles of the two groups of hair cells have been much debated, particularly since differences in their innervation and physiological behaviour have become apparent. Because of their rich afferent supply, inner hair cells are believed to be the major source of auditory signals in the cochlear nerve. Some evidence for this view is based on the finding that animals treated with antibiotics that are specifically toxic to outer hair cells are still able to hear, but their sensitivity and frequency discrimination is impaired. Some electrical responses of the cochlea can be recorded with extracellular electrodes. The most significant is the endolymphatic potential, a steady potential recordable between the cochlear duct and the scala tympani, which is caused by the different ionic compositions of their fluids. As the resting potential of hair cells is c.70 mV (negative inside) and the endolymphatic potential is positive in the cochlear duct, the total transmembrane potential across the apices of hair cells is 150 mV. This is a greater resting potential than is found anywhere else in the body, and provides the driving force for mechanotransduction and for the cochlear amplifier.
page 677 page 678
Under stimulation by sound, a rapid oscillatory cochlear microphonic potential can be recorded. It matches the frequency of the stimulus and movements of the basilar membrane precisely, and appears to depend on fluctuations in the conductance of hair cell membranes, probably of the outer hair cells. At the same time, an extracellular summating potential develops, a steady direct current shift related to the (intracellular) receptor potentials of the hair cells. Cochlear nerve fibres then begin to respond with action potentials, which are also recordable from the cochlea. Intracellular recording of auditory responses from inner hair cells has confirmed that these cells resemble other receptors: their steady receptor potentials are related in size to the amplitude of the acoustic stimulus. At the same time, afferent axons are stimulated by synaptic action at the bases of the inner hair cells. They fire more rapidly as the vibration of the basilar membrane increases in amplitude, up to a threshold that depends on the sensitivity of the specific nerve fibre involved. Each inner hair cell is contacted by axons with response thresholds that range from 0 decibels sound pressure level (dB SPL), the approximate threshold of human hearing, to those which respond to intensities in excess of 100dB SPL; the loudest sound tolerable is around 120dB SPL. Each axon responds most sensitively to the frequency represented by its particular cochlear location, its characteristic frequency (Fig. 39.23). Deafness
Two causes of deafness are usually distinguished: conductive hearing loss and sensorineural hearing loss. Conductive hearing loss may result from trauma to the external or middle ears, blockage of the external auditory meatus, or disruption of the tympanic membrane (e.g. by intense sounds or extreme pressure changes). It may also result from chronic inflammation of the tympanic membrane (e.g. by a cholesteatoma, which may also damage the ossicles); from an infection of the middle ear (otitis media with effusion), which produces a fluid build-up in the normally air-filled middle ear and so impedes the movements of the ossicles; or from otosclerosis, an inappropriate thickening of bone around the footplate of the stapes. Sensorineural hearing loss refers to loss or damage of the sensory hair cells or their innervation. The sensory cells of the inner ear are particularly vulnerable to mechanical trauma produced by high intensity noise and to changes in their physiological environment caused by infection or hypoxia. Changes in their ionic environment rapidly lead to degenerative processes that result in hair cell loss, often by apoptosis, and produce either hearing loss or vestibular dysfunction. These changes can be induced by drugs such as the aminoglycoside antibiotics, some diuretics, and certain anticancer drugs. A decrease in cochlear sensitivity, presbyacusis, almost invariably occurs with age: hair cells at the high frequency end of the cochlea tend to be lost first. Ménière's disease is a distressing disorder of the inner ear characterized by episodes of hearing loss, tinnitus and vertigo. Histological examination of an affected ear reveals endolymphatic hydrops (swelling of the endolymphatic spaces), suggesting poor drainage of the endolymph via the endolymphatic sac.
SURGICAL APPROACHES TO THE INNER EAR The inner ear may be approached surgically from various directions. The promontory that overlies the basal turn of the cochlea and the oval window may be opened via the middle ear (after elevating the tympanic membrane). The lateral semicircular canal may be opened via the aditus (after widening the bony external acoustic meatus and removing the incus). The arcuate eminence may be opened to give access to the superior semicircular canal via the floor of the middle cranial fossa. The posterior semicircular canal may be opened deep to the mastoid segment of the Fallopian canal via the mastoid process (after drilling away the overlying air cells). All of these approaches are usually reserved for destructive operations on the labyrinth to treat intractable vertigo. The round window niche and its membrane may be approached via a posterior tympanotomy. In this procedure, the mastoid air cells are removed to allow access to the bony triangle bounded above by the fossa of incus, superficially by the chorda tympani, and deeply by the descending portion of the facial nerve. This bone is drilled away carefully to expose the facial recess of the tympanic cavity
and the round window niche. Using this access, the stimulating electrode of a multichannel intracochlear implant can be passed into the scala tympani of the cochlea so that it lies against the spiral lamina and can stimulate the adjacent fibres of the cochlear nerve. The endolymphatic sac may be approached after exenteration of the mastoid air cells by elevating the cortical bone of the anterolateral wall of the posterior cranial fossa, anterior to the sigmoid venous sinus and posterior to the posterior semicircular canal (below a line extended from the axis of the lateral semicircular canal). Access to the sac is required in some operations that aim to control vertigo secondary to Ménière's disease. The internal acoustic meatus may be approached, at a cost to hearing, by drilling away the entire bony labyrinth via the posterior cranial fossa (after craniectomy in the occipital region and retraction of the cerebellum), or via the middle cranial fossa (after a temporal craniotomy and retraction of the dura of the middle fossa and the temporal lobe). These approaches are usually used to access tumours of the cerebellopontine angle and internal acoustic meatus. UPDATE Date Added: 07 September 2005 Shanida Helena Nataraja, PhD (Dianthus Medical Limited) Update: Superior petrosal triangle as anatomic landmark for subtemporal middle fossa orientation. A number of different anatomic landmarks can be used for orientation during middle fossa dissection, but considerable inter-patient variability means that these landmarks are often not readily identifiable in clinical practice. This study explored a method of identifying the head of the malleus and the use of this anatomic landmark during a middle fossa approach. In Part I, ten preserved human adult cadaveric temporal bones were dissected, using a dissecting microscope and an electric drill, in order to define the relationship between various middle fossa landmarks and the head of the malleus. In Part II, temporal bones were dissected from five fresh whole cadaveric heads. Using the specimens from Part I, the mean (± SD) distance from the root of the zygoma to the head of the malleus was 18.7 ±1.7 mm; from the head of the malleus to the foramen spinosum (FS) was 19.2 ± 1.0 mm; from the root of the zygoma to the FS was 30.1 ± 1.17 mm; and from the head of the malleus to the acoustic porus (AP) was 19.8 ± 0.5 mm. Using the specimens from Part II, the intersection of two 19 mm arcs (one from the root of the zygoma and the other from the FS) was found to locate the head of malleus within 2.5 ± 2.4 mm. Furthermore, the intersection of a reference line drawn between the root of the zygoma and the head of the malleus was 1.0 (0.9) mm away from the center apex of the internal auditory canal (IAC). The data from Part I reveal a constant relationship between the root of the zygoma, the head of the malleus, and AP, which is described here as the superior petrosal triangle. The data from Part II indicate that a reference line between the root of the zygoma and the head of the malleus can be used to identify both the IAC and the AP. Miller RS, Pensak ML: The superior petrosal triangle as a constant anatomical landmark for subtemporal middle fossa orientation. Laryngoscope 113(8):1327-1331, 2003. REFERENCES Ashmore J 2002 The mechanics of hearing. In: Roberts D (ed) Signals and Perception: The Fundamentals of Human Sensation. Basingstoke and New York: Palgrave Macmillan: 3-16. Axelsson A 1988 Comparative anatomy of cochlear blood vessels. Am J Otolaryngol 9: 278-90. Medline Similar articles Blanks RH, Curthoys IS, Markham CH 1975 Planar relationships of the semicircular canals in man. Acta Otolaryngol 80: 185-96. Medline Similar articles Curthoys IS, Blanks RH, Markham CH 1977 Semicircular canal radii of curvature (R) in cat, guinea pig and man. J Morphol 115: 1-15. Eybalin M 1993 Neurotransmitters and neuromodulators of the mammalian cochlea. Physiol Rev 73: 30973. Medline Similar articles Felix H, Hoffman V, Wright A, Gleeson MJ 1987 Ultrastructural findings on human Scarpa's ganglion. Acta Otolaryngol Suppl 436: 85-92. Medline Similar articles Fettiplace R 2002 The transformation of sound stimuli into electrical signals. In: Roberts D (ed) Signals and Perception: The Fundamentals of Human Sensation. Basingstoke and New York: Palgrave Macmillan: 1728. Furness DN 2002 The vestibular system. In: Roberts D (ed) Signals and Perception: The Fundamentals of Human Sensation. Basingstoke and New York: Palgrave Macmillan: 77-90. Guinan J Jr 1996 Physiology of olivocochlear efferents. In: Dallos P, Popper AN, Fay RR (eds) The Cochlea. New York: Springer Verlag: 435-502. Comprehensive description of the efferent innervation of the cochlea and its function Nadol JB 1988 Comparative anatomy of the cochlea and auditory nerve in mammals. Hear Res 34: 253-
66. Medline
Similar articles
Full article
Spoor F, Zonneveld F 1998 Comparative review of the human bony labyrinth. Am J Phys Anthropol Suppl 27: 211-51. Wangemann P, Schacht J 1996 Homeostatic mechanisms in the cochlea. In: Dallos P, Popper AN, Fay RR (eds) The Cochlea. New York: Springer Verlag: 130-5. Reviews what is known about inner ear fluids (perilymph and endolymph), how they are produced and what their functional significance might be. Warr WB 1992 Organization of olivocochlear efferent systems in mammals. In: Webster DB, Popper AN, Fay RR (eds) Mammalian Auditory Pathway: Neuroanatomy. New York: Springer Verlag: 410-48.
© 2008 Elsevier
40 EAR AND AUDITORY AND VESTIBULAR APPARATUS Development of the ear INNER EAR The rudiments of the internal ears appear shortly after those of the eyes as two patches of thickened surface epithelium, the otic placodes, lateral to the hindbrain. The early otic epithelium, which is derived from the otic placode, initiates and then suppresses chondrogenesis in the surrounding periotic mesenchyme. Sonic hedgehog protein, fibroblast growth factors and transforming growth factor beta have all been shown to be active in the early stages of otic capsule development in the mouse (Frenz et al 1994). Each otic placode invaginates as an otic pit while at the same time giving cells to the statoacoustic (vestibulocochlear) ganglion (Fig. 14.2). The mouth of the pit then closes to form an otocyst (auditory or otic vesicle) (Fig. 40.1). The otocyst is initially piriform, but a vertical infolding of its wall progressively marks off a tubular diverticulum on the medial side, which differentiates into the ductus and saccus endolymphaticus. The latter both communicate via the ductus with the remainder of the vesicle, the utriculosaccular chamber, which is placed laterally. Three compressed diverticula emerge as disc-like evaginations from the dorsal part of this chamber. The central parts of their walls coalesce and disappear and their peripheral portions persist as the semicircular ducts. The anterior duct is completed first, and the lateral last. A medially directed evagination arises from the ventral part of the utriculosaccular chamber and coils progressively as the cochlear duct. Its proximal extremity becomes constricted and forms the ductus reuniens. The central part of the chamber now represents the membranous vestibule, which becomes divided into a small ventral saccule and a larger utricle. This is achieved mainly by horizontal infolding that extends from the lateral wall of the vestibule towards the opening of the ductus endolymphaticus until only a narrow utriculosaccular duct remains between saccule and utricle. The duct becomes acutely bent on itself: its apex is continuous with the ductus endolymphaticus. During this period the membranous labyrinth rotates so that its long axis, which was originally vertical, becomes more or less horizontal. Cells derived from the otocyst not only contribute placodal cells to the vestibulocochlear ganglion, but also differentiate into specialized paraneuronal hair cells of the utricle, saccule, ampullae of the semicircular ducts, and organ of Corti; various specialized sustentacular cells and the unique epithelia of the stria vascularis and endolymphatic sac, and cells from which the general epithelial lining of the membranous labyrinth develops. The periotic mesenchyme surrounding the various parts of the epithelial labyrinth is converted into a cartilaginous otic capsule which ossifies to form most of the
bony labyrinth of the internal ear apart from the modiolus and osseous spiral lamina. For a time the cartilaginous capsule is incomplete which means that the cochlear, vestibular and facial ganglia are exposed in the gap between its canalicular and cochlear parts. They are soon covered by an outgrowth of cartilage, and the facial nerve becomes enclosed by a growth of cartilage from the cochlear to the canalicular part of the capsule. Perilymphatic spaces develop in the embryonic connective tissue between the cartilaginous capsule and the epithelial wall of the labyrinth. The rudiment of the periotic cistern or vestibular perilymphatic space can be seen in an embryo of from 30-40 mm in length in the reticulum between the saccule and the fenestra vestibuli. The scala tympani develops opposite the fenestra cochleae and is followed later by the scala vestibuli. The two scalae gradually extend along each side of the ductus cochlearis, and when they reach the tip of the ductus a communication, the helicotrema, opens between them. The modiolus and the osseous spiral lamina of the cochlea are not preformed in cartilage but ossify directly from connective tissue. The rudiment of the eighth nerve appears in the fourth week as the vestibulocochlear ganglion, which lies between the otocyst and the wall of the hindbrain. At first it is fused with the ganglion of the facial nerve (acousticofacial ganglion) but later the two separate. The cells of the vestibulocochlear ganglion are mainly derived from the placodal ectoderm. The ganglion divides into vestibular and cochlear parts, each associated with the corresponding division of the eighth nerve. Ganglionic neurones, which remain bipolar throughout life, are unusual in that many of their somata become enveloped in thin myelin sheaths. Their peripheral processes provide the afferent innervation of the labyrinthine hair cells, which also become associated with the outgrowing axons of the olivocochlear bundle - from cells of the superior olivary complexes in the pons.
© 2008 Elsevier
INNER EAR The rudiments of the internal ears appear shortly after those of the eyes as two patches of thickened surface epithelium, the otic placodes, lateral to the hindbrain. The early otic epithelium, which is derived from the otic placode, initiates and then suppresses chondrogenesis in the surrounding periotic mesenchyme. Sonic hedgehog protein, fibroblast growth factors and transforming growth factor beta have all been shown to be active in the early stages of otic capsule development in the mouse (Frenz et al 1994). Each otic placode invaginates as an otic pit while at the same time giving cells to the statoacoustic (vestibulocochlear) ganglion (Fig. 14.2). The mouth of the pit then closes to form an otocyst (auditory or otic vesicle) (Fig. 40.1). The otocyst is initially piriform, but a vertical infolding of its wall progressively marks off a tubular diverticulum on the medial side, which differentiates into the ductus and saccus endolymphaticus. The latter both communicate via the ductus with the remainder of the vesicle, the utriculosaccular chamber, which is placed laterally. Three compressed diverticula emerge as disc-like evaginations from the dorsal part of this chamber. The central parts of their walls coalesce and disappear and their peripheral portions persist as the semicircular ducts. The anterior duct is completed first, and the lateral last. A medially directed evagination arises from the ventral part of the utriculosaccular chamber and coils progressively as the cochlear duct. Its proximal extremity becomes constricted and forms the ductus reuniens. The central part of the chamber now represents the membranous vestibule, which becomes divided into a small ventral saccule and a larger utricle. This is achieved mainly by horizontal infolding that extends from the lateral wall of the vestibule towards the opening of the ductus endolymphaticus until only a narrow utriculosaccular duct remains between saccule and utricle. The duct becomes acutely bent on itself: its apex is continuous with the ductus endolymphaticus. During this period the membranous labyrinth rotates so that its long axis, which was originally vertical, becomes more or less horizontal. Cells derived from the otocyst not only contribute placodal cells to the vestibulocochlear ganglion, but also differentiate into specialized paraneuronal hair cells of the utricle, saccule, ampullae of the semicircular ducts, and organ of Corti; various specialized sustentacular cells and the unique epithelia of the stria vascularis and endolymphatic sac, and cells from which the general epithelial lining of the membranous labyrinth develops. The periotic mesenchyme surrounding the various parts of the epithelial labyrinth is converted into a cartilaginous otic capsule which ossifies to form most of the bony labyrinth of the internal ear apart from the modiolus and osseous spiral lamina. For a time the cartilaginous capsule is incomplete which means that the cochlear, vestibular and facial ganglia are exposed in the gap between its canalicular and cochlear parts. They are soon covered by an outgrowth of cartilage, and the facial nerve becomes enclosed by a growth of cartilage from the
cochlear to the canalicular part of the capsule. Perilymphatic spaces develop in the embryonic connective tissue between the cartilaginous capsule and the epithelial wall of the labyrinth. The rudiment of the periotic cistern or vestibular perilymphatic space can be seen in an embryo of from 30-40 mm in length in the reticulum between the saccule and the fenestra vestibuli. The scala tympani develops opposite the fenestra cochleae and is followed later by the scala vestibuli. The two scalae gradually extend along each side of the ductus cochlearis, and when they reach the tip of the ductus a communication, the helicotrema, opens between them. The modiolus and the osseous spiral lamina of the cochlea are not preformed in cartilage but ossify directly from connective tissue. The rudiment of the eighth nerve appears in the fourth week as the vestibulocochlear ganglion, which lies between the otocyst and the wall of the hindbrain. At first it is fused with the ganglion of the facial nerve (acousticofacial ganglion) but later the two separate. The cells of the vestibulocochlear ganglion are mainly derived from the placodal ectoderm. The ganglion divides into vestibular and cochlear parts, each associated with the corresponding division of the eighth nerve. Ganglionic neurones, which remain bipolar throughout life, are unusual in that many of their somata become enveloped in thin myelin sheaths. Their peripheral processes provide the afferent innervation of the labyrinthine hair cells, which also become associated with the outgrowing axons of the olivocochlear bundle - from cells of the superior olivary complexes in the pons.
© 2008 Elsevier
MIDDLE EAR (TYMPANIC CAVITY) AND PHARYNGOTYMPANIC TUBE The pharyngotympanic tube and tympanic cavity are extensions of the early pharynx and develop from a hollow, the tubotympanic recess. This lies between the first and third pharyngeal arches, and has a floor which consists of the second arch and its limiting pouches. The forward growth of the third arch causes the inner part of the recess to narrow to form the tubal region, and also excludes the inner part of the second arch from this portion of the floor. The more lateral part of the recess develops into the tympanic cavity and its floor forms the lateral wall of the tympanic cavity up to approximately the level at which the chorda tympani branches off from the facial nerve. The lateral wall of the tympanic cavity contains first and second arch elements. The first arch territory is limited to that part in front of the anterior process of the malleus, and the second arch forms the outer wall behind this and also turns on to the posterior wall to include the tympanohyal region. The tubotympanic recess at first lies inferolateral to the cartilaginous otic capsule, but as the capsule enlarges the spatial relationship alters and the tympanic cavity becomes anterolateral. A cartilaginous process grows from the lateral part of the capsule to form the tegmen tympani and it curves caudally to form the lateral wall of the pharyngotympanic tube. In this way, the tympanic cavity and the proximal part of the pharyngotympanic tube become included in the petrous region of the temporal bone. During the sixth or seventh month the mastoid antrum appears as a dorsal expansion of the tympanic cavity. The malleus develops from the dorsal end of the ventral mandibular (Meckel's) cartilage and the incus from the dorsal cartilage of the first arch (p. 450), which is probably homologous to the quadrate bone of birds and reptiles. The stapes stems mainly from the dorsal end of the cartilage of the second (hyoid) arch, first as a ring (anulus stapes) encircling the small stapedial artery (p. 617). The primordium of the stapedius muscle appears close to the artery and facial nerve at the end of the second month, and at almost the same time the tensor tympani begins to appear near the extremity of the tubotympanic recess. At first the ossicles are embedded in the mesenchymal roof of the tympanic cavity, later they become covered by the mucosa of the middle ear cavity which becomes air filled after birth.
© 2008 Elsevier
EXTERNAL EAR page 679 page 680
Figure 40.1 A-F The stages in the development of the membranous labyrinth from the otocyst, at the embryonic stages and viewed from the aspects indicated. Note also the relationship of the vestibular (orange) and cochlear (yellow) parts of the vestibulocochlear nerve.
The external acoustic meatus develops from the dorsal end of the hyomandibular or first pharyngeal groove. Close to its dorsal extremity this groove extends inwards as a funnel-shaped primary meatus from which the cartilaginous part and a small area of the roof of the osseous meatus are developed. A solid epidermal plug extends inwards from the tube along the floor of the tubotympanic recess, and the cells in the centre of the plug subsequently degenerate to produce the inner part of the meatus (secondary meatus). The epidermal stratum of the
tympanic membrane is formed from the deepest ectodermal cells of the epidermal plug, and the fibrous stratum is formed from the mesenchyme between the meatal plate and the endodermal floor of the tubotympanic recess. The development of the auricle is initiated by the appearance of six hillocks which form round the margins of the dorsal portion of the hyomandibular groove at the 4 mm stage. Of the six, three are on the caudal edge of the mandibular arch and three on the cranial edge of the hyoid arch. These hillocks appear from stage 15. They tend to be less obvious prior to that stage and, of those on the mandibular arch only the most ventral, which subsequently forms the tragus, can be identified throughout earlier stages. The rest of the auricle is formed in the mesenchyme of the hyoid arch, which extends forwards round the dorsal end of the remains of the hyomandibular groove, forming a keel-like elevation which is the forerunner of the helix. The contribution made by the mandibular arch to the auricle is greatest at the end of the second month, and it becomes relatively reduced as growth continues until eventually the area of skin supplied by the mandibular nerve extends little above the tragus. The lobule is the last part of the auricle to develop. REFERENCE Frenz DA, Liu W, Williams JD, Hatcher V, Galinovic-Schwartz V, Flanders KC, Van de Water TR 1994 Induction of chondrogenesis: requirement for synergistic interaction of basic fibroblast growth factor and transforming growth factor-beta. Development 120: 415-24. Medline Similar articles
© 2008 Elsevier
41 THE BONY ORBIT AND PERIPHERAL AND ACCESSORY VISUAL APPARATUS The orbit and its contents EYELIDS (Fig. 41.1) The eyelids (palpebrae) are two thin moveable folds which cover the anterior surface of the eye. They protect the eye, by their closure, from trauma or from excessive light. The act of blinking maintains a thin film of tears over the cornea. The upper eyelid is larger and more mobile than the lower eyelid, and contains an elevator muscle, levator palpebrae superioris. When the eyelids are open, the upper lid just overlaps the upper part of the cornea, whereas the lower lid lies just below the cornea. An elliptical space, the palpebral fissure, is left between their margins. When the eyelids are closed, the upper lid moves down to cover the whole of the cornea. The eyelids are covered by skin externally and by conjunctiva internally. The skin of the eyelids is thin and almost translucent. The supporting framework of each eyelid is formed by dense fibrous tissue arranged as a tarsal plate and an orbital septum attaching the plate to the orbital margin. The main muscle within the eyelids is orbicularis oculi. Each lid margin exhibits a small elevation, the lacrimal papilla, approximately onesixth of the way along from the medial canthus of the eye. There is a small opening, the punctum lacrimale, in the centre of the papilla. The margin of the eyelid lateral to the lacrimal papilla bears the eyelashes and is termed the ciliary part of the eyelid. The margin of the eyelid medial to the lacrimal papilla lacks eyelashes and is termed the lacrimal part of the eyelid. The lid margin for both the upper and lower eyelids exhibits a 'grey line' which corresponds to the mucocutaneous junction. The eyelashes lie in front of this line, and the openings of the tarsal glands (meibomian glands) lie behind it. The tarsal glands are seen as a series of parallel, faint yellow lines arranged perpendicular to the lid margins when the eyelids are everted.
Figure 41.1 The eyelids and eyeball. (By permission from Berkovitz BKB, Moxham BJ 2002 Head and Neck Anatomy. London: Martin Dunitz.)
Eyelashes are short, thick, curved hairs, arranged in double or triple rows. The upper, which are more numerous and longer, curve upwards, while those in the lower lid curve down so that upper and lower lashes do not interlace when the lids are closed. The medial and lateral angles of the eye are referred to as the medial (inner) and lateral (outer) canthi. The lateral canthus is relatively featureless. The medial canthus is c.2 mm lower than the lateral canthus: this distance is increased in some asiatic groups. It is separated from the eyeball by a small triangular space, the lacrimal lake (lacus lacrimalis), in which a small, reddish body called the lacrimal caruncle is situated. The caruncle contains sebaceous and sweat glands, and sometimes accessory lacrimal glands. It represents an area of modified skin containing some fine hairs, and is mounted on the plica semilunaris, a fold of conjunctiva which is believed by some to be a vestige of the nictitating membrane of other animals. In oriental Asians, a semilunar fold of skin termed the epicanthus passes from the medial end of the upper eyelid to the lower eyelid and obscures the caruncle. UPDATE Abstract: Development of the semilunar plica
Date Added: 10 April 2006
Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=15255295&query_hl=1&itool=pubmed_docsum The structure of the human semilunar plica at different stages of its development-A morphological and morphometric study. Arends G, Schramm U: Ann Anat 186:195-207, 2004. The lower eyelid can be everted to reveal its conjunctiva up to the point where it is reflected from the eyelid onto the sclera, i.e. at the inferior fornix. The upper eyelid is less easily everted. UPDATE Date Added: 26 April 2006 Abstract: Anatomical differences between the Asian and non-Asian eyelid Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=15167728&query_hl=16&itool=pubmed_docsum Microscopic anatomy of the lower eyelid in Asians. Lim WK, Rajendran K, Choo CT et al: Ophthal Plast Reconstr Surg 20:207-211, 2004. The eyelids are demarcated from the adjacent facial skin by the superior and inferior palpebral furrows. Additional furrows appear with age just beyond the inferior orbital margins, e.g. a nasojugal furrow medially and a malar furrow laterally. UPDATE Date Added: 02 May 2006 Abstract: Orbicularis oculi myocutaneus flap for reconstruction of inner canthus defect Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=15308401&query_hl=22&itool=pubmed_docsum The orbicularis oculi myocutaneus flap in the repair of the medial canthal region: A new strategy for canthal resurfacing. Stagno d'Alcontres F, D'Amico E, Colonna MR et al: Br J Plast Surg 57:540-542, 2004.
STRUCTURE OF EYELIDS From its facial surface inwards each eyelid consists of skin, subcutaneous connective tissue, fibres of the palpebral part of orbicularis oculi, submuscular connective tissue, the tarsus with its tarsal glands and orbital septum, and palpebral conjunctiva. The upper lid also contains the aponeurosis of levator palpebrae superioris (Fig. 41.2). The skin is extremely thin and is continuous at the palpebral margins with the conjunctiva. The subcutaneous connective tissue is very delicate, seldom contains any adipose tissue and lacks elastic fibres.
The palpebral fibre bundles of orbicularis oculi are thin and pale and run parallel with the palpebral fissure. Deep to them is the submuscular connective tissue, a loose fibrous layer which in the upper lid is continuous with the subaponeurotic layer of the scalp. Effusions of blood or pus at this level can therefore pass down from the scalp into the upper eyelid. The main nerves lie in the submuscular layer, which means that local anaesthetics should be injected deep to orbicularis oculi. UPDATE Date Added: 11 January 2005 Shanida Helena Nataraja, PhD (Dianthus Medical Limited) Update: Anatomical study of the lower positioned transverse ligament The lower positioned transverse ligament (LPTL) is a structure in the upper eyelid, distinct from the higher positioned transverse ligament (WL) that determines fissure height. It is therefore crucial that the integrity of this structure is preserved during upper eyelid surgery. Kakizaki et al. report the anatomical findings of a study of ten postmortem orbits with eyelids. These orbits were taken from five Japanese cadavers (average age at death was 78; range 75-81). Both the LPTL and the WL were observed in the upper eyelid of all orbits. Additionally, a subtype of LPTL, robust ligamentous tissue (RLT), was observed in four eyelids. The LPTL was seen to arise from the anterior surface of the trochlea, run inferolaterally on the levator aponeurosis (beside the line of fusion of the orbital septum and the levator aponeurosis), and then continue to the periosteum of the lateral orbital rim just above the lateral horn of the levator aponeurosis. The WL was seen to arise form the superior aspect of the trochlea and disperse to the capsule of the lacrimal gland. Demarcation between the LPTL and WL on the trochlea was unclear, lending further support to the notion that both structures probably originate from the same optical septum-derived organ. The origin of the histological differences between these two structures (LPTL contains fewer elastic fibers) remains unknown. If the LPTL is mistaken for the WL during blepharoptosis surgery, the LPTL becomes fixed in such a way that there is insufficient elevation of the eyelid. The observation that 40% of eyelids were contain a RLT is of interest because this structure reduces eyelid flexibility and restricts opening movements. RLT incision is therefore necessary in blepharoptosis surgery or double fold operations to ensure free eyelid opening. Kakizaki H, Zako M, Nakano T, Iwaki M, Mito H. Anatomical study of the lower-positioned transverse ligament.Br J Plast Surg. 2004;57(4):370-2.
CONJUNCTIVA (Fig. 41.2) The conjunctiva is a transparent mucous membrane which covers the internal palpebral surfaces and is reflected over the sclera, where its epithelium becomes continuous with that of the cornea. page 681 page 682
Figure 41.2 The upper eyelid and anterior segment of the eye: sagittal section. (Provided by the late Gordon L Ruskell, Department of Optometry and Visual Science, The City University, London.)
The palpebral conjunctiva is very vascular, and has a dense subepithelial layer of capillaries. It contains mucosa-associated lymphoid tissue (MALT - Chapter 5) especially at the orbital edges of the tarsi to which it is closely adherent. At the free palpebral margins the conjunctiva is continuous with the skin, the lining epithelium of the ducts of the tarsal glands, and with the lacrimal canaliculi and lacrimal sac (p. 685). There is therefore continuity with the nasolacrimal duct and nasal mucosa which is important in the spread of infection. The line of reflection of the conjunctiva from the lids to the eyeball, the conjunctival fornix, is subdivided into superior and inferior fornices. The ducts of the lacrimal gland open into the lateral part of the superior fornix (Fig. 41.3). The bulbar conjunctiva is loosely connected to the eyeball over the exposed sclera, is thin and transparent, and is only slightly vascular. The epithelium of the palpebral conjunctiva at the margin of the lids is nonkeratinized stratified squamous, 10-12 cells thick. About 2 mm from each margin there is often a subtarsal groove in which foreign bodies frequently lodge: here the epithelium has two or three layers, and consists of columnar and flat surface cells. These persist throughout most of the palpebral conjunctiva. Near the fornices in the orbital conjunctiva the cells are taller, and a trilaminar conjunctival epithelium covers much of the anterior, exposed surface of the sclera. It thickens closer to the corneoscleral junction and then changes to stratified squamous epithelium typical of the cornea. Mucous-secreting goblet cells are scattered in the conjunctival epithelium. They are most frequent on each side of the fornix, but absent from the exposed surfaces of the bulbar conjunctiva and the corneoscleral junction. page 682 page 683
Figure 41.3 The orbital glands responsible for the secretion of the tear film.
TARSAL PLATES The two tarsi (Fig. 41.2) are thin, elongated, crescent-shaped plates of firm, dense fibrous tissue c.2.5 cm long. There is one in each eyelid to provide support and determine eyelid form. Each tarsal plate is convex forwards, and conforms to the configuration of the anterior surface of the eye. The free ciliary border is straight and adjacent to the eyelash follicles. The orbital, border is convex and attached to the orbital septum. The superior tarsus, the larger of the two, is semi-oval, c.10 mm in height centrally. Its inferior edge is parallel to, and c.2 mm from, the lid margin. The deepest fibres of the aponeurosis of levator palpebrae superioris are attached to its anterior surface, and the fibrous extension of the superior tarsal muscle is attached to its upper margin (Fig. 41.4). The smaller inferior tarsus is narrower, and c.4 mm in vertical height. The tarsal plates are connected to the margins of the orbit by the orbital septum and by the medial and lateral palpebral (canthal) ligaments. The medial palpebral ligament passes from the medial ends of the two tarsal plates to the anterior lacrimal crest and the frontal process of the maxilla. It splits at its insertion into the tarsal plates to surround the lacrimal canaliculi, and lies in front of the nasolacrimal sac and the orbital septum. The lateral palpebral ligament is relatively poorly developed. It passes from the lateral ends of the tarsal plates to a small tubercle on the zygomatic bone within the orbital margin and is more deeply situated than the medial palpebral ligament. It lies beneath the orbital septum and the lateral palpebral raphe of orbicularis oculi. Each tarsal plate is associated with a thin lamina of smooth muscle. Opposite the equator of the eye the superior tarsal muscle passes from the inferior face of levator palpebrae superioris to a fibrous extension which projects to the upper margin of the superior tarsus. There is a corresponding but less prominent inferior tarsal muscle in the lower eyelid which unites the inferior tarsus to the anterior expansion of the fused fascial sheath of inferior rectus and inferior oblique. The smooth muscles assist in the elevation of the upper, and depression of the lower, eyelids. They may adjust the size of the aperture of the open eye according to mood and other factors.
page 683 page 684
Figure 41.4 The tarsi, their ligaments and the orbital septum: anterior aspect.
Tarsal glands (Figs 41.2, 41.3) are modified sebaceous glands embedded in the tarsi, and may be visible through the conjunctiva when the eyelids are everted. They are yellow and arranged in a single row of c.25 in the upper lid, and fewer in the lower lid. They occupy the full tarsal height, and are therefore longer centrally where the tarsi are higher. Each gland consists of a straight tube with many lateral diverticula, and opens by a minute orifice on the free palpebral margin. It is enclosed by a basement membrane, and is lined at its orifice by stratified epithelium and elsewhere by a layer of polyhedral cells. The oily secretion of the tarsal glands spreads over the margins of the eyelids, and so an oily layer is drawn over the tear film as the palpebral fissure opens after a blink, reducing evaporation and contributing to tear film stability. The presence of the oily, hydrophobic secretions of tarsal glands along the margins of the eyelids also inhibits the spillage of tears onto the face.
ORBITAL SEPTUM The orbital septum (Fig. 41.4) is a weak membranous sheet which is attached to the orbital rim where it becomes continuous with the periosteum. It passes inwards into each eyelid and blends with the tarsal plates, and, in the upper eyelid, with the superficial lamella of levator palpebrae superioris. The orbital septum is thickest laterally, where it lies in front of the lateral palpebral ligament. It passes behind the medial palpebral ligament and nasolacrimal sac, but in front of the pulley of superior oblique. The orbital septum is pierced above by levator palpebrae superioris and below by the ligament from inferior rectus. The lacrimal, supratrochlear, infratrochlear and supraorbital nerves and vessels pass through the septum from the orbit on the way to the face and scalp.
VASCULAR SUPPLY AND LYMPHATIC DRAINAGE The eyelids are supplied by the lateral and medial palpebral arteries, both of which have superior and inferior branches which anastomose to form arcades: the superior arteries supply the upper eyelid, and the inferior arteries supply the lower eyelid. The lateral palpebral arteries, usually two, are given off by the lacrimal
branch of the ophthalmic artery. The medial palpebral arteries arise directly from the ophthalmic artery below the trochlea, and descend behind the nasolacrimal sac to enter the eyelids, where each bifurcates. Their branches course laterally along the tarsal edges to form superior and inferior arcades which are completed by anastomoses with branches of the supraorbital and zygomatico-orbital arteries (superior arch) and the lateral palpebral artery (both arches). The inferior arch links with the facial artery to supply the mucosa of the nasolacrimal duct. The eyelids are also supplied by branches of the infraorbital, facial, transverse facial and superficial temporal arteries. The veins which drain the eyelids are larger and more numerous than the arteries. They pass either superficially to veins on the face and forehead, or deeply to the ophthalmic veins within the orbit. Bulbar conjunctival veins pass to the orbital surfaces of the rectal muscles and join the superior or inferior ophthalmic vein. The lymph vessels which drain the eyelids and conjunctiva commence in a superficial plexus beneath the skin, and in a deep plexus in front of and behind the tarsi. These plexuses communicate with one another and medial and lateral sets of vessels arise from them. The lymph vessels of the lateral set drain the whole thickness of the lateral part of the upper and lower lids and all the ocular conjunctiva. They pass laterally from the lateral commissure to end in the superficial and deep parotid lymph nodes and also in the deep parotid lymph nodes. The lymph vessels of the medial set drain the skin over the medial part of the upper eyelid, the whole thickness of the medial half of the lower lid, and the caruncula lacrimalia. They follow the course of the facial vein and end in the submandibular group of lymph nodes. UPDATE Date Added: 28 December 2004 Shanida Helena Nataraja, PhD (Dianthus Medical Limited) Update: Patterns of regional and distant metastasis in patients with eyelid and periocular squamous cell carcinoma The frequency and location of both regional and distant lymph node metastasis have recently been established in patients with squamous cell carcinoma (SCC) of the eyelid and periocular skin. Eligible patients were identified through a search of the Tumor Registry database at the University of Texas MD Anderson Cancer Center. One hundred and eleven patients (89 men; 22 women) who had been treated between 1952 and 2002 for eyelid and periocular skin SCC, and for whom at least 6 months of follow-up data were available, were assessed. The clinical records of these patients were reviewed retrospectively and the following data extracted: age at diagnosis, gender, location of lesion, treatment modalities, patterns of regional nodal and distant metastasis, local recurrence, perineural invasion, and overall survival. The age of diagnosis ranged from 31 to 91 years (median: 64 years). The most common site of SCC was the lower eyelid (54 patients; 49%), followed by the medial canthus (40 patients; 36%) and the upper eyelid (25 patients; 23%). Treatment modalities received included wide local excision in 96 patients (87%), radiotherapy in 7 patients (6%), and primary exenteration owing to extensive tumor in 7 patients (6%). Local recurrence occurred in 41 patients (37%), and 27 patients (24%) had regional node metastasis and 7 patients (6%) had distant node metastasis during the study. Nine patients (8%) had perineural invasion. These findings are largely consistent with previously reported observations. However, the incidence of regional node metastasis (24%) appears to be higher than previously demonstrated (range: 1021%): the authors acknowledge that this may have been influenced by selection bias; there is a much higher proportion of patients with advanced disease at the University of Texas MD Anderson Cancer Center than normally seen in other clinical settings. Nevertheless, a high incidence of regional node metastasis warrants careful surveillance of the regional nodes. The authors propose that the technique of sentinel node biopsy, a technique that can be used to identify patients with a high risk for regional metastasis (i.e. lesions wider than 2 cm,
those that are locally recurrent, or those with perineural invasion), should be explored in greater depth, and its role as a possible adjunct to the current management of eyelid or periocular SCC established. Faustina M, Diba R, Ahmadi MA, Gutstein BF, Esmaeli B. Patterns of regional and distant metastasis in patients with eyelid and periocular squamous cell carcinoma. Ophthalmology. 2004;111(10):1930-2. Medline Similar articles
INNERVATION (Fig. 41.4) The cutaneous innervation of the eyelids comes from both the ophthalmic and maxillary divisions of the trigeminal nerve. The upper eyelid is supplied mainly by the supraorbital branch of the frontal nerve. Additional contributions come from the lacrimal nerve, the supratrochlear branch of the frontal nerve, and the infratrochlear branch of the nasociliary nerve. The nerve supply to the lower eyelid is principally from the infraorbital branch of the maxillary nerve, with small contributions from the lacrimal and infratrochlear nerves. The bulbar conjunctiva is supplied by the ophthalmic division of the trigeminal nerve. Autonomic fibres are probably vasomotor in function.
ECTROPION Ectropion describes the rolling out of the lower eyelid so that it is no longer in contact with the cornea. It is most commonly a senile development but there are also cicatricial and paralytic forms. It causes overspill of tear fluid (epiphora), instability of the tear film, and chronic conjunctivitis resulting from exposure. If the condition persists, the conjunctiva may become dry, thickened and unsightly; in severe cases, drying of the cornea may cause loss of vision. In the senile form ectropion is caused by reduced tension in orbicularis oculi, and in the paralytic form it occurs because orbicularis oculi is unable to contract as a consequence of facial nerve palsy. The condition may warrant treatment by full thickness shortening of the lid.
ENTROPION Entropion describes the inversion of the eyelids and is largely a problem of later years (the involutional or senile form). Other forms are caused by spasm of the orbicularis oculi (spastic form); by cicatricial contraction of the palpebral conjunctiva, or as a congenital disorder. Connective tissue changes associated with age relax the tension in the lower eyelid in involutional entropion. The tarsal plate becomes thinned, atrophic and unstable, and its attachments to inferior rectus slacken, causing the lid to turn inwards. The lashes abrade the cornea (trichiasis) producing discomfort and in severe untreated cases the cornea becomes inflamed and there is loss of transparency. Spastic entropion is due to an acute spasm of orbicularis oculi, often induced by irritation of the eye, and may be an additional factor in involutional entropion. The tarsal plates of the eyelids stop short of the margins, which are therefore less firm, but they contain the ciliary or marginal part of orbicularis oculi, which may be responsible for the inversion in spastic entropion.
© 2008 Elsevier
EYELIDS (Fig. 41.1) The eyelids (palpebrae) are two thin moveable folds which cover the anterior surface of the eye. They protect the eye, by their closure, from trauma or from excessive light. The act of blinking maintains a thin film of tears over the cornea. The upper eyelid is larger and more mobile than the lower eyelid, and contains an elevator muscle, levator palpebrae superioris. When the eyelids are open, the upper lid just overlaps the upper part of the cornea, whereas the lower lid lies just below the cornea. An elliptical space, the palpebral fissure, is left between their margins. When the eyelids are closed, the upper lid moves down to cover the whole of the cornea. The eyelids are covered by skin externally and by conjunctiva internally. The skin of the eyelids is thin and almost translucent. The supporting framework of each eyelid is formed by dense fibrous tissue arranged as a tarsal plate and an orbital septum attaching the plate to the orbital margin. The main muscle within the eyelids is orbicularis oculi. Each lid margin exhibits a small elevation, the lacrimal papilla, approximately onesixth of the way along from the medial canthus of the eye. There is a small opening, the punctum lacrimale, in the centre of the papilla. The margin of the eyelid lateral to the lacrimal papilla bears the eyelashes and is termed the ciliary part of the eyelid. The margin of the eyelid medial to the lacrimal papilla lacks eyelashes and is termed the lacrimal part of the eyelid. The lid margin for both the upper and lower eyelids exhibits a 'grey line' which corresponds to the mucocutaneous junction. The eyelashes lie in front of this line, and the openings of the tarsal glands (meibomian glands) lie behind it. The tarsal glands are seen as a series of parallel, faint yellow lines arranged perpendicular to the lid margins when the eyelids are everted.
Figure 41.1 The eyelids and eyeball. (By permission from Berkovitz BKB, Moxham BJ 2002 Head and Neck Anatomy. London: Martin Dunitz.)
Eyelashes are short, thick, curved hairs, arranged in double or triple rows. The upper, which are more numerous and longer, curve upwards, while those in the lower lid curve down so that upper and lower lashes do not interlace when the lids are closed.
The medial and lateral angles of the eye are referred to as the medial (inner) and lateral (outer) canthi. The lateral canthus is relatively featureless. The medial canthus is c.2 mm lower than the lateral canthus: this distance is increased in some asiatic groups. It is separated from the eyeball by a small triangular space, the lacrimal lake (lacus lacrimalis), in which a small, reddish body called the lacrimal caruncle is situated. The caruncle contains sebaceous and sweat glands, and sometimes accessory lacrimal glands. It represents an area of modified skin containing some fine hairs, and is mounted on the plica semilunaris, a fold of conjunctiva which is believed by some to be a vestige of the nictitating membrane of other animals. In oriental Asians, a semilunar fold of skin termed the epicanthus passes from the medial end of the upper eyelid to the lower eyelid and obscures the caruncle. UPDATE Abstract: Development of the semilunar plica
Date Added: 10 April 2006
Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=15255295&query_hl=1&itool=pubmed_docsum The structure of the human semilunar plica at different stages of its development-A morphological and morphometric study. Arends G, Schramm U: Ann Anat 186:195-207, 2004. The lower eyelid can be everted to reveal its conjunctiva up to the point where it is reflected from the eyelid onto the sclera, i.e. at the inferior fornix. The upper eyelid is less easily everted. UPDATE Date Added: 26 April 2006 Abstract: Anatomical differences between the Asian and non-Asian eyelid Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=15167728&query_hl=16&itool=pubmed_docsum Microscopic anatomy of the lower eyelid in Asians. Lim WK, Rajendran K, Choo CT et al: Ophthal Plast Reconstr Surg 20:207-211, 2004. The eyelids are demarcated from the adjacent facial skin by the superior and inferior palpebral furrows. Additional furrows appear with age just beyond the inferior orbital margins, e.g. a nasojugal furrow medially and a malar furrow laterally. UPDATE Date Added: 02 May 2006 Abstract: Orbicularis oculi myocutaneus flap for reconstruction of inner canthus defect Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=15308401&query_hl=22&itool=pubmed_docsum The orbicularis oculi myocutaneus flap in the repair of the medial canthal region: A new strategy for canthal resurfacing. Stagno d'Alcontres F, D'Amico E, Colonna MR et al: Br J Plast Surg 57:540-542, 2004.
STRUCTURE OF EYELIDS From its facial surface inwards each eyelid consists of skin, subcutaneous connective tissue, fibres of the palpebral part of orbicularis oculi, submuscular connective tissue, the tarsus with its tarsal glands and orbital septum, and palpebral conjunctiva. The upper lid also contains the aponeurosis of levator palpebrae superioris (Fig. 41.2). The skin is extremely thin and is continuous at the palpebral margins with the conjunctiva. The subcutaneous connective tissue is very delicate, seldom contains any adipose tissue and lacks elastic fibres. The palpebral fibre bundles of orbicularis oculi are thin and pale and run parallel with the palpebral fissure. Deep to them is the submuscular connective tissue, a loose fibrous layer which in the upper lid is continuous with the subaponeurotic layer of the scalp. Effusions of blood or pus at this level can therefore pass down from the scalp into the upper eyelid. The main nerves lie in the submuscular
layer, which means that local anaesthetics should be injected deep to orbicularis oculi. UPDATE Date Added: 11 January 2005 Shanida Helena Nataraja, PhD (Dianthus Medical Limited) Update: Anatomical study of the lower positioned transverse ligament The lower positioned transverse ligament (LPTL) is a structure in the upper eyelid, distinct from the higher positioned transverse ligament (WL) that determines fissure height. It is therefore crucial that the integrity of this structure is preserved during upper eyelid surgery. Kakizaki et al. report the anatomical findings of a study of ten postmortem orbits with eyelids. These orbits were taken from five Japanese cadavers (average age at death was 78; range 75-81). Both the LPTL and the WL were observed in the upper eyelid of all orbits. Additionally, a subtype of LPTL, robust ligamentous tissue (RLT), was observed in four eyelids. The LPTL was seen to arise from the anterior surface of the trochlea, run inferolaterally on the levator aponeurosis (beside the line of fusion of the orbital septum and the levator aponeurosis), and then continue to the periosteum of the lateral orbital rim just above the lateral horn of the levator aponeurosis. The WL was seen to arise form the superior aspect of the trochlea and disperse to the capsule of the lacrimal gland. Demarcation between the LPTL and WL on the trochlea was unclear, lending further support to the notion that both structures probably originate from the same optical septum-derived organ. The origin of the histological differences between these two structures (LPTL contains fewer elastic fibers) remains unknown. If the LPTL is mistaken for the WL during blepharoptosis surgery, the LPTL becomes fixed in such a way that there is insufficient elevation of the eyelid. The observation that 40% of eyelids were contain a RLT is of interest because this structure reduces eyelid flexibility and restricts opening movements. RLT incision is therefore necessary in blepharoptosis surgery or double fold operations to ensure free eyelid opening. Kakizaki H, Zako M, Nakano T, Iwaki M, Mito H. Anatomical study of the lower-positioned transverse ligament.Br J Plast Surg. 2004;57(4):370-2.
CONJUNCTIVA (Fig. 41.2) The conjunctiva is a transparent mucous membrane which covers the internal palpebral surfaces and is reflected over the sclera, where its epithelium becomes continuous with that of the cornea. page 681 page 682
Figure 41.2 The upper eyelid and anterior segment of the eye: sagittal section. (Provided by the late Gordon L Ruskell, Department of Optometry and Visual Science, The City University, London.)
The palpebral conjunctiva is very vascular, and has a dense subepithelial layer of capillaries. It contains mucosa-associated lymphoid tissue (MALT - Chapter 5) especially at the orbital edges of the tarsi to which it is closely adherent. At the free palpebral margins the conjunctiva is continuous with the skin, the lining epithelium of the ducts of the tarsal glands, and with the lacrimal canaliculi and lacrimal sac (p. 685). There is therefore continuity with the nasolacrimal duct and nasal mucosa which is important in the spread of infection. The line of reflection of the conjunctiva from the lids to the eyeball, the conjunctival fornix, is subdivided into superior and inferior fornices. The ducts of the lacrimal gland open into the lateral part of the superior fornix (Fig. 41.3). The bulbar conjunctiva is loosely connected to the eyeball over the exposed sclera, is thin and transparent, and is only slightly vascular. The epithelium of the palpebral conjunctiva at the margin of the lids is nonkeratinized stratified squamous, 10-12 cells thick. About 2 mm from each margin there is often a subtarsal groove in which foreign bodies frequently lodge: here the epithelium has two or three layers, and consists of columnar and flat surface cells. These persist throughout most of the palpebral conjunctiva. Near the fornices in the orbital conjunctiva the cells are taller, and a trilaminar conjunctival epithelium covers much of the anterior, exposed surface of the sclera. It thickens closer to the corneoscleral junction and then changes to stratified squamous epithelium typical of the cornea. Mucous-secreting goblet cells are scattered in the conjunctival epithelium. They are most frequent on each side of the fornix, but absent from the exposed surfaces of the bulbar conjunctiva and the corneoscleral junction. page 682 page 683
Figure 41.3 The orbital glands responsible for the secretion of the tear film.
TARSAL PLATES The two tarsi (Fig. 41.2) are thin, elongated, crescent-shaped plates of firm, dense fibrous tissue c.2.5 cm long. There is one in each eyelid to provide support and determine eyelid form. Each tarsal plate is convex forwards, and conforms to the configuration of the anterior surface of the eye. The free ciliary border is straight and adjacent to the eyelash follicles. The orbital, border is convex and attached to the orbital septum. The superior tarsus, the larger of the two, is semi-oval, c.10 mm in height centrally. Its inferior edge is parallel to, and c.2 mm from, the lid margin. The deepest fibres of the aponeurosis of levator palpebrae superioris are attached to its anterior surface, and the fibrous extension of the superior tarsal muscle is attached to its upper margin (Fig. 41.4). The smaller inferior tarsus is narrower, and c.4 mm in vertical height. The tarsal plates are connected to the margins of the orbit by the orbital septum and by the medial and lateral palpebral (canthal) ligaments. The medial palpebral ligament passes from the medial ends of the two tarsal plates to the anterior lacrimal crest and the frontal process of the maxilla. It splits at its insertion into the tarsal plates to surround the lacrimal canaliculi, and lies in front of the nasolacrimal sac and the orbital septum. The lateral palpebral ligament is relatively poorly developed. It passes from the lateral ends of the tarsal plates to a small tubercle on the zygomatic bone within the orbital margin and is more deeply situated than the medial palpebral ligament. It lies beneath the orbital septum and the lateral palpebral raphe of orbicularis oculi. Each tarsal plate is associated with a thin lamina of smooth muscle. Opposite the equator of the eye the superior tarsal muscle passes from the inferior face of levator palpebrae superioris to a fibrous extension which projects to the upper margin of the superior tarsus. There is a corresponding but less prominent inferior tarsal muscle in the lower eyelid which unites the inferior tarsus to the anterior expansion of the fused fascial sheath of inferior rectus and inferior oblique. The smooth muscles assist in the elevation of the upper, and depression of the lower, eyelids. They may adjust the size of the aperture of the open eye according to mood and other factors.
page 683 page 684
Figure 41.4 The tarsi, their ligaments and the orbital septum: anterior aspect.
Tarsal glands (Figs 41.2, 41.3) are modified sebaceous glands embedded in the tarsi, and may be visible through the conjunctiva when the eyelids are everted. They are yellow and arranged in a single row of c.25 in the upper lid, and fewer in the lower lid. They occupy the full tarsal height, and are therefore longer centrally where the tarsi are higher. Each gland consists of a straight tube with many lateral diverticula, and opens by a minute orifice on the free palpebral margin. It is enclosed by a basement membrane, and is lined at its orifice by stratified epithelium and elsewhere by a layer of polyhedral cells. The oily secretion of the tarsal glands spreads over the margins of the eyelids, and so an oily layer is drawn over the tear film as the palpebral fissure opens after a blink, reducing evaporation and contributing to tear film stability. The presence of the oily, hydrophobic secretions of tarsal glands along the margins of the eyelids also inhibits the spillage of tears onto the face.
ORBITAL SEPTUM The orbital septum (Fig. 41.4) is a weak membranous sheet which is attached to the orbital rim where it becomes continuous with the periosteum. It passes inwards into each eyelid and blends with the tarsal plates, and, in the upper eyelid, with the superficial lamella of levator palpebrae superioris. The orbital septum is thickest laterally, where it lies in front of the lateral palpebral ligament. It passes behind the medial palpebral ligament and nasolacrimal sac, but in front of the pulley of superior oblique. The orbital septum is pierced above by levator palpebrae superioris and below by the ligament from inferior rectus. The lacrimal, supratrochlear, infratrochlear and supraorbital nerves and vessels pass through the septum from the orbit on the way to the face and scalp.
VASCULAR SUPPLY AND LYMPHATIC DRAINAGE The eyelids are supplied by the lateral and medial palpebral arteries, both of which have superior and inferior branches which anastomose to form arcades: the superior arteries supply the upper eyelid, and the inferior arteries supply the lower eyelid. The lateral palpebral arteries, usually two, are given off by the lacrimal
branch of the ophthalmic artery. The medial palpebral arteries arise directly from the ophthalmic artery below the trochlea, and descend behind the nasolacrimal sac to enter the eyelids, where each bifurcates. Their branches course laterally along the tarsal edges to form superior and inferior arcades which are completed by anastomoses with branches of the supraorbital and zygomatico-orbital arteries (superior arch) and the lateral palpebral artery (both arches). The inferior arch links with the facial artery to supply the mucosa of the nasolacrimal duct. The eyelids are also supplied by branches of the infraorbital, facial, transverse facial and superficial temporal arteries. The veins which drain the eyelids are larger and more numerous than the arteries. They pass either superficially to veins on the face and forehead, or deeply to the ophthalmic veins within the orbit. Bulbar conjunctival veins pass to the orbital surfaces of the rectal muscles and join the superior or inferior ophthalmic vein. The lymph vessels which drain the eyelids and conjunctiva commence in a superficial plexus beneath the skin, and in a deep plexus in front of and behind the tarsi. These plexuses communicate with one another and medial and lateral sets of vessels arise from them. The lymph vessels of the lateral set drain the whole thickness of the lateral part of the upper and lower lids and all the ocular conjunctiva. They pass laterally from the lateral commissure to end in the superficial and deep parotid lymph nodes and also in the deep parotid lymph nodes. The lymph vessels of the medial set drain the skin over the medial part of the upper eyelid, the whole thickness of the medial half of the lower lid, and the caruncula lacrimalia. They follow the course of the facial vein and end in the submandibular group of lymph nodes. UPDATE Date Added: 28 December 2004 Shanida Helena Nataraja, PhD (Dianthus Medical Limited) Update: Patterns of regional and distant metastasis in patients with eyelid and periocular squamous cell carcinoma The frequency and location of both regional and distant lymph node metastasis have recently been established in patients with squamous cell carcinoma (SCC) of the eyelid and periocular skin. Eligible patients were identified through a search of the Tumor Registry database at the University of Texas MD Anderson Cancer Center. One hundred and eleven patients (89 men; 22 women) who had been treated between 1952 and 2002 for eyelid and periocular skin SCC, and for whom at least 6 months of follow-up data were available, were assessed. The clinical records of these patients were reviewed retrospectively and the following data extracted: age at diagnosis, gender, location of lesion, treatment modalities, patterns of regional nodal and distant metastasis, local recurrence, perineural invasion, and overall survival. The age of diagnosis ranged from 31 to 91 years (median: 64 years). The most common site of SCC was the lower eyelid (54 patients; 49%), followed by the medial canthus (40 patients; 36%) and the upper eyelid (25 patients; 23%). Treatment modalities received included wide local excision in 96 patients (87%), radiotherapy in 7 patients (6%), and primary exenteration owing to extensive tumor in 7 patients (6%). Local recurrence occurred in 41 patients (37%), and 27 patients (24%) had regional node metastasis and 7 patients (6%) had distant node metastasis during the study. Nine patients (8%) had perineural invasion. These findings are largely consistent with previously reported observations. However, the incidence of regional node metastasis (24%) appears to be higher than previously demonstrated (range: 1021%): the authors acknowledge that this may have been influenced by selection bias; there is a much higher proportion of patients with advanced disease at the University of Texas MD Anderson Cancer Center than normally seen in other clinical settings. Nevertheless, a high incidence of regional node metastasis warrants careful surveillance of the regional nodes. The authors propose that the technique of sentinel node biopsy, a technique that can be used to identify patients with a high risk for regional metastasis (i.e. lesions wider than 2 cm,
those that are locally recurrent, or those with perineural invasion), should be explored in greater depth, and its role as a possible adjunct to the current management of eyelid or periocular SCC established. Faustina M, Diba R, Ahmadi MA, Gutstein BF, Esmaeli B. Patterns of regional and distant metastasis in patients with eyelid and periocular squamous cell carcinoma. Ophthalmology. 2004;111(10):1930-2. Medline Similar articles
INNERVATION (Fig. 41.4) The cutaneous innervation of the eyelids comes from both the ophthalmic and maxillary divisions of the trigeminal nerve. The upper eyelid is supplied mainly by the supraorbital branch of the frontal nerve. Additional contributions come from the lacrimal nerve, the supratrochlear branch of the frontal nerve, and the infratrochlear branch of the nasociliary nerve. The nerve supply to the lower eyelid is principally from the infraorbital branch of the maxillary nerve, with small contributions from the lacrimal and infratrochlear nerves. The bulbar conjunctiva is supplied by the ophthalmic division of the trigeminal nerve. Autonomic fibres are probably vasomotor in function.
ECTROPION Ectropion describes the rolling out of the lower eyelid so that it is no longer in contact with the cornea. It is most commonly a senile development but there are also cicatricial and paralytic forms. It causes overspill of tear fluid (epiphora), instability of the tear film, and chronic conjunctivitis resulting from exposure. If the condition persists, the conjunctiva may become dry, thickened and unsightly; in severe cases, drying of the cornea may cause loss of vision. In the senile form ectropion is caused by reduced tension in orbicularis oculi, and in the paralytic form it occurs because orbicularis oculi is unable to contract as a consequence of facial nerve palsy. The condition may warrant treatment by full thickness shortening of the lid.
ENTROPION Entropion describes the inversion of the eyelids and is largely a problem of later years (the involutional or senile form). Other forms are caused by spasm of the orbicularis oculi (spastic form); by cicatricial contraction of the palpebral conjunctiva, or as a congenital disorder. Connective tissue changes associated with age relax the tension in the lower eyelid in involutional entropion. The tarsal plate becomes thinned, atrophic and unstable, and its attachments to inferior rectus slacken, causing the lid to turn inwards. The lashes abrade the cornea (trichiasis) producing discomfort and in severe untreated cases the cornea becomes inflamed and there is loss of transparency. Spastic entropion is due to an acute spasm of orbicularis oculi, often induced by irritation of the eye, and may be an additional factor in involutional entropion. The tarsal plates of the eyelids stop short of the margins, which are therefore less firm, but they contain the ciliary or marginal part of orbicularis oculi, which may be responsible for the inversion in spastic entropion.
© 2008 Elsevier
SOFT TISSUE Certain named regions of soft connective tissue within the orbit have anatomical and clinical significance. These include the orbital septum, canthal ligaments, fascial sheath of the eye, suspensory ligament, periosteum and orbital fat.
ORBITAL SEPTUM This is described on page 683.
FASCIAL SHEATH OF THE EYEBALL (Fig. 41.5) A thin fascial membrane, the vagina bulbi (fascia bulbi or Tenon's capsule), envelops the eyeball from the optic nerve to the corneoscleral junction, separates it from the orbital fat, and forms a socket. Its ocular aspect is loosely attached to the sclera by delicate bands of episcleral connective tissue. Posteriorly, the fascia is traversed by ciliary vessels and nerves. It fuses with the sclera and with the sheath of the optic nerve around its entrance to the eyeball. Attachment to the sclera is strongest in this position and again anteriorly, just behind the corneoscleral junction at the limbus. The fascial sheath is perforated by the tendons of the extraocular muscles and is reflected on to each as a tubular sheath, the muscular fascia. The sheath of superior oblique reaches the fibrous trochlea (pulley) of the muscle. The sheaths of the four recti muscles are very thick anteriorly but are reduced posteriorly to a delicate perimysium. Just before they blend with the vagina bulbi, the thick sheaths of adjacent recti become confluent and form a fascial ring. Expansions from the sheaths are important for the attachments they make. Those from the medial and lateral rectus muscles are triangular and strong, and are attached respectively to the lacrimal and zygomatic bones. As they may limit the actions of the two recti, they are termed the medial and lateral check ligaments (Figs 41.5, 41.6). Other extraocular muscles have less substantial check ligaments: the capacity of any of them to actually limit movement has been questioned. The sheath of inferior rectus is thickened on its underside and blends with the sheath of inferior oblique. These two, in turn, are continuous with the fascial ring noted earlier and therefore with the sheaths of the medial and lateral recti. As the latter are attached to the orbital walls by check ligaments, a continuous fascial band, the suspensory ligament of the eye, is slung like a hammock below the eye. The suspensory ligament provides sufficient support for the eye such that, even when the maxilla forming the floor of the orbit is removed, the eye will retain its position. page 684 page 685
Figure 41.5 A, The orbital fascia in sagittal section. B, Scheme of the orbital fascia in horizontal section. (After Whitnall SE 1932 Anatomy of the Human Orbit, 2nd edn. London: Oxford University Press. By permission of Oxford University Press.)
The thickened fused sheath of inferior rectus and inferior oblique also has an anterior expansion into the lower eyelid, where, augmented by some fibres of orbicularis oculi (the inferior tarsal muscle), it attaches to the inferior tarsus. Contraction of inferior rectus in downward gaze therefore also draws the lid downward. The sheath of levator palpebrae superioris also thickens anteriorly, and just behind the aponeurosis it fuses inferiorly with the sheath of superior rectus. It extends forward between the two muscles and attaches to the upper fornix of the conjunctiva. Other extensions of the fascial sheath of the eye pass medially and laterally and attach to the orbital walls, forming the transverse ligament of the eye. This structure is of uncertain significance, but presumably plays a part in drawing the fornix upwards in gaze elevation and it may act as a fulcrum for levator
movements. Other numerous finer fasciae form radial septa which extend from the vagina bulbi and the muscle sheaths to the periosteum of the orbit, and so provide compartments for orbital fat. The orbital septum is the most anteriorly placed. Many of the fasciae contain smooth muscle cells. The ocular and orbital fasciae are arranged to assist in the location of the eye within the orbit without obstructing the activities of the extraocular muscles, except possibly in the extremes of rotation. They also prevent the gross displacement of orbital fat, for this could interfere with the accurate positioning of the two eyes in binocular vision. The periosteum of the orbit is only loosely attached to bone. Behind, it is united with the dura mater of the optic nerve and in front it is continuous with the periosteum of the orbital margin, where it gives off a stratum contributing to the orbital septum. It also attaches to the trochlea, and, as the lacrimal fascia, forms the roof and lateral wall of the sulcus for the nasolacrimal sac (p. 687).
ORBITAL FAT (Fig. 41.7) The spaces between the main structures of the orbit are occupied by orbital fat. This is particularly the case between the optic nerve and the cone of muscles (see Fig. 41.8). The fat helps to stabilize the position of the eyeball and acts as a socket within which the eyeball can rotate. Conditions resulting in an increased overall volume of orbital fat, e.g. hyperthyroidism, may thrust the eyeball forwards (exophthalmos). UPDATE Abstract: Adipose body of the orbit
Date Added: 13 June 2006
Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=11948952&query_hl=2&itool=pubmed_docsum Adipose body of the orbit. Wolfram-Gabel R, Khan JL: Clin Anat 15:186-192, 2002.
© 2008 Elsevier
NASOLACRIMAL APPARATUS (Figs 41.3, 41.9) The nasolacrimal apparatus consists of the lacrimal gland, which secretes a complex fluid (tears) and whose excretory ducts convey fluid to the surface of the eye; the paired lacrimal canaliculi; the lacrimal sac, and the nasolacrimal duct, by which the fluid is collected and conveyed into the nasal cavity.
LACRIMAL GLAND The lacrimal gland consists of an orbital part and a palpebral part. They are continuous posterolaterally around the concave lateral edge of the aponeurosis of levator palpebrae superioris. The orbital part, about the size and shape of an almond, lodges in the lacrimal fossa on the medial aspect of the zygomatic process of the frontal bone, just within the orbital margin. It lies above levator palpebrae superioris and, laterally, above lateral rectus. Its lower surface is connected to the sheath of levator palpebrae superioris and its upper surface is connected to the orbital periosteum. Its anterior border is in contact with the orbital septum and its posterior border is attached to the orbital fat. The palpebral part, about one-third the size of the orbital part is subdivided into two or three lobules and extends below the aponeurosis of levator palpebrae superioris into the lateral part of the upper lid, where it is attached to the superior conjunctival fornix. It is visible through the conjunctiva when the lid is everted. The ducts of the lacrimal gland, about six in number, open into the superior fornix. Those from the orbital part (four or five) penetrate the aponeurosis of levator palpebrae superioris to join those from the palpebral part. Excision of the palpebral part is therefore functionally equivalent to the total removal of the gland. Many small accessory lacrimal glands occur in or near the fornix. They are more numerous in the upper eyelid. Their presence may explain why the conjunctiva does not dry up after extirpation of the main lacrimal gland. Vascular supply The lacrimal gland receives its arterial blood supply from the lacrimal branch of the ophthalmic artery. It may also receive blood from the infraorbital artery. Venous drainage is into the superior ophthalmic vein. Innervation The lacrimal gland is innervated by secretomotor parasympathetic fibres from the pterygopalatine ganglion which may reach the gland via zygomatic and lacrimal branches of the maxillary nerve or pass directly to the gland. Preganglionic parasympathetic fibres travel to the ganglion in the greater petrosal nerve, which arises from the facial nerve at the geniculate ganglion. UPDATE Date Added: 24 October 2006 Helen E Wiggett, PhD (Dianthus Medical Limited) Update: Distribution of pterygopalatine ganglion efferents to the lacrimal gland It is assumed that in humans pterygopalatine efferents of the lacrimal secretomotor nerve pathway pass to the lacrimal gland via the zygomatic and lacrimal nerves, although this is not certain. To determine precisely the postganglionic pathway in humans, the orbit and pterygopalatine fossa was removed as a whole from nine human cadavers, cut coronally into slabs, and embedded in resin. Serial sections were cut and stained to allow the nerve pathway to be traced. Nine to 15 rami orbitals passed dorsally from the pterygopalatine ganglia and either were directed dorsocranially destined for the cavernous sinus plexus or continued dorsally to the orbit. One group of 4-8 rami (35-90 µm) that entered the orbit passed dorsally adjacent to the lateral wall of the orbit joining the filaments of the retro-orbital plexus at the apex. The junctions occurred close to the lacrimal nerve, with one or more plexus branches joining the nerve or transiently penetrating its perineurium. It was unclear whether any rami continued in the nerve distal to the apparent junction. The rami mostly contained unmyelinated nerve fibers, which is consistent with them being postganglionic fibers, and because they were derived from the pterygopalatine ganglion, they were presumed to be parasympathetic. From the retro-orbital plexus, 5-10 rami lacrimales, which were smaller in diameter than the rami orbitals and most of the branches of the retro-orbital plexus, accompanied the lacrimal nerve to the lacrimal gland entering the posterior margin of the gland. No rami passed from the pterygopalatine ganglion to the lacrimal gland without first joining the retro-orbital
plexus. An accessory ophthalmic artery, a branch of the middle meningeal artery, was interposed along the pathway taken by the rami lacrimales entering the orbit through the superior orbital fissure. In most preparations, the arteries were 200350 µm in diameter and contained 2-3 rami in their adventitia or adjacent loose connective tissue that were similar in size to the rami orbitals and continued to the lacrimal gland as supplementary rami lacrimales. In two preparations, the arteries were much larger (diameters of 850 µm and 1100 µm; larger than the ophthalmic artery) and contained 6 rami within their wall, which passed deep within the orbit with their arteries to the junction with the ophthalmic artery. In conclusion, this study shows for the first time in humans that pterygopalatine efferents pass to the lacrimal gland via the retro-orbital plexus without using the ophthalmic nerve and its branches as conduits, as previously suggested. This is similar to the pathway in monkeys. These results indicate that clinical measures to reduce lacrimation by severing ophthalmic branches are not likely to be effective. Ruskell GL: Distribution of pterygopalatine ganglion efferents to the lacrimal gland in man. Exp Eye Res 78(3):329-335, 2004.
Microstructure
page 685 page 686
Figure 41.6 The eye viewed anteriorly showing the extraocular muscles.
Figure 41.7 Two coronal sections through the left orbit (viewed from in front: right is lateral) cut through planes passing (A) 5 mm behind the posterior pole of the globe of the eye, and (B) 4.6 mm in front of its posterior pole (the lens which is visible, has been displaced backwards. atc, adipose tissue compartments; eb, ethmoid bone; fn, frontal nerve; frb, frontal bone; iov, inferior ophthalmic vein; ir, inferior rectus; lb, lacrimal bone; lacg, lacrimal gland; lr, lateral rectus; ls, nasolacrimal sac; m, maxilla (bone); ms, maxillary sinus; mr, medial rectus; ncn, nasociliary nerve; oo, orbicularis oculi; opn, optic nerve; sb, sphenoid bone; lps, levator palpebrae superioris; som, superior oblique; sov, superior ophthalmic vein; sr, superior rectus; zyg, zygomatic bone. (By permission from Kornneef 1977.)
The lacrimal gland is lobulated and tubuloacinar in form (p. 34). Its secretory units are acini (Fig. 41.10) and resemble those of the salivary glands. The secretion is a watery fluid with an electrolyte content similar to that of plasma and contains, amongst other components, a bacteriocidal enzyme, lysozyme. Other components of tears include the proteins lactoferrin, IgA, and tear-specific prealbumen, as well as some major serum proteins (IgM, IgG, transferrin and serum albumen). It has been suggested that there are two types of secretion from the orbital glands, basic and reflex. The former is a constant slow baseline secretion, and the latter occurs in response to stimulation, e.g. neural stimulation. The content of the tears varies considerably in the non-stimulated and stimulated states.
LACRIMAL CANALICULI (Figs 41.1, 41.9, 41.11) There is one lacrimal canaliculus in each lid, c.10 mm long. Each commences at a puncta lacrimalia. The superior canaliculus, smaller and shorter than the inferior, at first ascends, and then bends at an acute angle, and passes medially and downwards to reach the nasolacrimal sac. The inferior canaliculus first descends and turns almost horizontally to the sac. At their angles the canaliculi are dilated into ampullae. The mucosa lining the ducts has a non-keratinized stratified squamous epithelium lying on a basement membrane, outside which is a lamina propria rich in elastic fibres (the ducts are therefore easily dilated when probed), and a layer of skeletal muscle fibres continuous with the lacrimal part of orbicularis oculi. At the base of each lacrimal papilla the muscle fibres are circularly arranged in the form of a sphincter.
NASOLACRIMAL SAC Figs 41.9, 41.11) The lacrimal sac is the closed upper end of the nasolacrimal duct. It is c.12 mm long and lies in a fossa adjacent to the lacrimal groove in the anterior part of the medial wall of the orbit. The sac is bounded in front by the anterior lacrimal crest of the maxilla and behind by the posterior lacrimal crest of the lacrimal bone. Its closed upper end is laterally flattened and its lower part is rounded and merges into the duct. The lacrimal canaliculi open into its lateral wall near its upper end. page 686 page 687
Figure 41.8 Coronal sections through the two orbits: posterior aspect. On the left side the plane of the section is more posterior and passes behind the eyeball.
Figure 41.9 Coronal section through the left half of the nasal cavity (anterior aspect) to show the relation of the lacrimal passages to the maxillary and ethmoidal sinuses and the inferior nasal concha. The mucous membrane is coloured. (After Whitnall SE 1932 Anatomy of the Human Orbit, 2nd edn. London: Oxford University Press. By permission of Oxford University Press.)
A layer of lacrimal fascia, continuous with the orbital periosteum, passes between the lacrimal crest of the maxilla and the lacrimal bone, and forms a roof and lateral wall to the lacrimal fossa. There is a plexus of minute veins between the fascia and the nasolacrimal sac. The fascia separates the sac from the medial palpebral ligament in front and the lacrimal part of orbicularis oculi behind. The lower half of the lacrimal fossa is related medially to the anterior part of the middle meatus, and the upper half to the anterior ethmoidal sinuses. The nasolacrimal sac has a fibroelastic wall. It is lined internally by mucosa which is continuous with the conjunctiva through the lacrimal canaliculi, and with the nasal mucosa through the nasolacrimal duct.
NASOLACRIMAL DUCT (Figs 41.9, 41.11) The nasolacrimal duct is c.18 mm long, and descends from the lacrimal sac to open anteriorly in the inferior meatus of the nose at an expanded orifice. A fold of
mucosa (plica lacrimalis) forms an imperfect valve just above its opening (ostium lacrimalis). The duct runs down an osseous canal formed by the maxilla, lacrimal bone and inferior nasal concha (Fig. 41.12). It is narrowest in the middle and is directed downwards, backwards and a little laterally. The mucosa of the nasolacrimal sac and the nasolacrimal duct has a bilaminar columnar epithelium, which is ciliated in places. Around the duct there is a rich plexus of veins forming erectile tissue: engorgement of these veins may obstruct the duct.
ANATOMY OF TEARS Lacrimal fluid enters the conjunctival sac at its superolateral angle and, by capillarity and blinking, is carried across the eye to the lacus lacrimalis, mainly between the lower palpebral margin and the eyeball. From the lacus it enters the lacrimal canaliculi. Contraction of orbicularis oculi presses the puncta lacrimalia more firmly into the lacus and capillary attraction draws the secretion into the lacrimal sac. Sudden dilatation of the sac, produced by the lacrimal part of orbicularis oculi during blinking, probably aids this. Normally the tarsal secretions prevent the tears from overflowing the lid margins and also cover the capillary film of fluid on the cornea and sclera with a film of oil, which delays evaporation. Tear film
The tear film, i.e. the interface between the external environment and the ocular surface has a number of functions. It forms a smooth refracting surface over the corneal surface; lubricates the eyelids; maintains an optimal environment for the epithelia of the cornea and conjunctiva; dilutes and washes away noxious substances; provides an antibacterial system for the ocular surface; serves as an entry pathway for polymorphonuclear leukocytes. The film consists of three layers. The outer layer is a lipid layer c.0.1µm thick. It floats on an intermediate, aqueous layer, c.7µm thick, which contains electrolytes, water, IgA and proteins, many of them antibacterial enzymes. The inner layer is mucous which covers the cornea and conjunctiva. Each layer of the tear film is secreted by a different set of orbital glands. The mucous layer is secreted by goblet cells, and it may also contain glycoproteins (possibly mucins) secreted by the stratified squamous epithelial cells of the cornea and conjunctiva. The aqueous layer is secreted by the main and accessory lacrimal glands (Fig. 41.3), and may receive a contribution from the corneal epithelial cells. The lipid layer is secreted by the tarsal glands. UPDATE Date Added: 21 December 2004 Shanida Helena Nataraja, PhD (Dianthus Medical Limited) Update: The tear film lipid layer The tear film lipid layer was the subject of a series of articles in the March 2004 issue of Experimental Eye Research. Selected papers are summarised below. Bron et al. review the functional aspects of the tear film lipid layer which is primarily composed of meibomian lipid, secreted by meibomian glands (see McCulley et al. below) embedded within the tarsal plates. Not only do these lipids provide a smooth optical surface for the cornea, but they also reduce the rate of evaporation from the eye. Delivery of these lipids to the reservoirs located on the skin of each lid margin is the result of a steady secretory process and the delivery of a small amount of lipids in every blink. The formation of the lipid film is a complex process: lipids are delivered to the tear film in the up-phase of the blink and the lipid film spreads rapidly upwards over the aqueous surface of the tear film. Estimates of tear film lipid layer thickness vary considerably (range 13-100 nm); however, researchers agree that the layer demonstrates remarkable stability. This stability is largely the result of the high, non-Newtonian viscosity and low surface tension conferred by the formation of complexes between the meibomian lipids and proteins in the aqueous phase, such as tear lipocalin. Yokoi et al. review different non-invasive methods used to assess the tear film, and their application to the study of tear film physiology and dry eye. Although several diagnostic techniques are available to assess the tear film and diagnose dry eye, most of these techniques are invasive and therefore modify the parameter they are designed to measure. There are, however, three main noninvasive techniques currently in use: meniscometry; interferometry; and meibometry. Meniscometry allows the measurement of the tear meniscus radius, thereby providing an indirect measure of tear volume. Interferometry can be used evaluate dry eye severity and so provide a direct measure of the thickness and fluidity of the lipid layer. Meibometry allows the measurement of the amount of meibomian lipid on the lip margin. These modern, non-invasive, methods are more appropriate methods for the assessment of the thin and transparent layers
of the preocular tear film. McCulley et al. review the growing body of evidence implicating the meibomian glands of the eyelids in the pathogenesis of a number of different ocular surface disorders, such as chronic blepharitis and dry eye. The mechanisms regulating the amount and type of lipid produced by the meibomian glands remain unclear. However, there is evidence that androgen and estrogen receptors present in the meibomian glands play an important role, as do other factors, such as stem cells and neurologic stimulants. Changes in both meibomian gland morphology and the lipid composition of the secreted fluid are thought to underlie the pathophysiology of chronic blepharitis. Not only do a large proportion of chronic blepharitis patients exhibit marked changes in meibomian gland structure, but they also show distinct differences in the lipid components of the tear film, including decreased amounts of polar and non-polar lipids. A better understanding of meibomian gland function, including the effects of hormones, gland drop-out, lipid composition, and microflora, will provide further insights into the pathophysiology of some ocular diseases. Nagymihalyi et al. examined the influence of eyelid temperature on the delivery of meibomian oil. Meibometry measurements were taken from all subjects in a stepwise manner: Step 1 - eyelids were cleaned and a baseline measurement taken, the volunteers were asked to blink 10 times at their normal rate, and then a second meibometry measurement was taken; Step 2 - closed eyelids were heated for 5 min using a 250 W infrared lamp, and the procedures described in Step 1 were then repeated; Step 3 - closed eyelids were cooled for 5 min using an icepack, and the procedures described in Step 1 were then repeated. A total of 20 female subjects (age: 68 years, range: 52-84) were enrolled into the study. No significant difference was found between the 10-blink recovery rate of the left and right eyes, and the eyes were therefore considered to be independent. The mean 10-blink recovery reading was 154 ± 12 IU (mean ± SEM) at room temperature (24°C), 203 ±15 IU with a 4.9 ± 0.3°C temperature increase, and 108 ± 15 IU with a 7.6 0.4C temperature decrease. This study suggests that external heating or cooling of the eye can affect the delivery of the meibomian gland secretion, most probably through its effects on the viscosity and flow characteristics of the oil. 1) Bron AJ, Tiffany JM, Gouveia SM, Yokoi N, Voon LW. Functional aspects of the tear film lipid layer. Exp Eye Res. 2004;78(3):347-60. Medline Similar articles 2) Yokoi N, Komuro A. Non-invasive methods of assessing the tear film. Exp Eye Res. 2004;78(3):399407. Medline Similar articles 3) McCulley JP, Shine WE. The lipid layer of tears: dependent on meibomian gland function. Exp Eye Res. 2004;78(3):361-5. Medline Similar articles 4) Nagymihalyi A, Dikstein S, Tiffany JM. The influence of eyelid temperature on the delivery of meibomian oil. Exp Eye Res. 2004;78(3):367-70. Medline Similar articles
LACRIMATION REFLEX (Fig. 41.13) page 687 page 688
Figure 41.10 Organization of the secretory units in the lacrimal gland.
The lacrimation reflex is stimulated by irritation of the conjunctiva and cornea. The afferent limb of the reflex involves branches of the ophthalmic nerve, with an additional contribution from the infraorbital nerve if the conjunctiva of the lower eyelid is involved. Impulses enter the brain and spread by interneurones to activate parasympathetic neurones in the superior salivatory centre (associated with the facial nerve) and sympathetic neurones in the upper thoracic spinal cord. The efferent pathway to the lacrimal gland involves the greater petrosal nerve, which carries parasympathetic preganglionic secretomotor fibres, and the deep petrosal nerve, which carries postganglionic sympathetic fibres: the parasympathetic fibres relay in the pterygopalatine ganglion, the sympathetic fibres pass through without synapsing. Lacrimation may also occur in response to emotional triggers.
© 2008 Elsevier
BONY ORBIT (Figs 41.14, 41.15, 41.16) The upper part of the facial skeleton contains two orbital cavities. Each cavity is pyramidal, and has a base at the orbital opening and a long, posteromedially directed axis. The major structures which occupy the orbit are the eye and optic nerve; the extraocular muscles; the oculomotor, trochlear and abducent nerves; the ophthalmic and maxillary divisions of the trigeminal nerve; the ciliary parasympathetic ganglion; the ophthalmic vessels and the nasolacrimal apparatus. All of these are contained within, and supported by, a considerable quantity of orbital fat. Each orbit has a roof, floor, medial and lateral walls, a base and apex. Roof of the orbit (superior wall)
The roof of the orbit is formed chiefly by the thin orbital plate of the frontal bone. It is gently concave on its orbital aspect, which separates the orbital contents and the brain in the anterior cranial fossa. Anteromedially it contains the frontal sinus and it displays a small trochlear fovea or spine where the trochlea (pulley) for superior oblique is attached. Anterolaterally there is a shallow lacrimal fossa which houses the orbital part of the lacrimal gland. The common tendinous ring of the four recti is attached to the bone near the superior, medial and lower margins of the orbital opening of the optic canal. Posteriorly, the roof of the orbit includes a part of the inferior aspect of the lesser wing of the sphenoid. The suture between these bones is almost horizontal. The optic canal lies between the roots of the lesser wing, and is bounded medially by the body of the sphenoid. Medial wall of the orbit (Fig. 41.16)
The medial wall of the orbit is extremely thin except posteriorly. It curves inferolaterally into the floor of the orbit. The vertical lacrimal groove which houses the nasolacrimal sac lies anteriorly: it opens below into the inferior meatus of the lateral wall of the nasal cavity via the nasolacrimal canal. The floor of the groove separates the orbital and nasal cavities anteriorly, but more posteriorly the ethmoidal sinuses intervene. The medial wall is limited in front by the anterior lacrimal crest on the frontal process of the maxilla, to which orbicularis oculi and lacrimal fascia are attached. The maxillolacrimal suture lies behind the lacrimal crest in the floor of the lacrimal groove. The upper opening of the nasolacrimal groove is completed laterally by the lacrimal hamulus, which curves anteromedially to the lower part of the anterior lacrimal crest. The lacrimal part of orbicularis oculi and lacrimal fascia are attached to the posterior lacrimal crest of the lacrimal groove, which is mostly formed by the lacrimal bone, and they bridge the groove. Posteriorly the orbital surface of the lacrimal bone is flat, and it articulates by a vertical suture with the orbital plate of the ethmoid labyrinth. The frontolacrimal and lacrimal maxillary sutures limit the medial wall in front. page 688 page 689
Figure 41.11 Radiograph of the lacrimal drainage pathway, demonstrated by the injection of radio-opaque tracer into the lacrimal duct. PL, puncta lacrimalia; ILC, inferior lacrimal canaliculus; LS, lacrimal sac; ND, nasolacrimal duct; NS, nasal septum. (Provided by TD Hawkins, Addenbrooke's Hospital, Cambridge; photograph by Sarah-Jane Smith.)
Figure 41.12 The medial wall of the nasolacrimal canal is formed by the maxilla and the articulation of the descending process of the lacrimal bone with the lacrimal process of the inferior nasal concha.
Figure 41.13 Lacrimation reflex. (Redrawn from MacKinnon P, Morris J 1990 Oxford Textbook of Functional Anatomy, Vol 3. Head and Neck. Oxford: Oxford University Press. By permission of Oxford University Press.)
Figure 41.14 Horizontal section through the left orbit and nasal cavity viewed from above.
The orbital plate of the ethmoid bone contributes most to the remainder of the medial wall. It is almost rectangular, and very thin, and forms the lateral walls of the ethmoidal sinuses. Above, it articulates with the medial edge of the orbital plate of the frontal bone at a suture which is interrupted by anterior and posterior ethmoidal foramina. The posterior foramen may be absent; occasionally there is a middle ethmoidal foramen. The anterior ethmoidal canal transmits its vessels and nerves into the anterior cranial fossa and also to the anterior nasal mucosa at the lateral edge of the cribriform plate. Below, the orbital plate of the ethmoid articulates with the medial edge of the orbital surface of the maxilla and posteriorly with the orbital process of the palatine bone. Posteriorly, it articulates with the body of the sphenoid, which forms the medial wall of the orbit posteriorly, separated from the orbital roof by the optic canal. Floor of the orbit (inferior wall) (Figs 41.14, 41.15, 41.16)
The floor of the orbit is mostly formed by the maxilla and, anterolaterally, by the zygomatic bone. Posteromedially the small triangular orbital process of the palatine bone joins the medial wall. page 689 page 690
Figure 41.15 Lateral wall of left orbit. (By permission from Berkovitz BKB, Moxham BJ 1989 A Colour Atlas of the Skull. London: Mosby-Wolfe.)
Figure 41.16 Medial wall of left orbit. (By permission from Berkovitz BKB, Moxham BJ 1989 A Colour Atlas of the Skull. London: Mosby-Wolfe.)
The floor is thin and largely roofs the maxillary sinus. Not quite horizontal, it ascends a little laterally. Anteriorly it curves into the lateral wall. Posteriorly it is separated from the lateral wall by the inferior orbital fissure, which connects the orbit posteriorly to the pterygopalatine fossa, and more anteriorly to the infratemporal fossa. The medial lip is notched by the infraorbital groove. The latter passes forwards and sinks into the floor to become the infraorbital canal which opens on the face at the infraorbital foramen. Infraorbital groove, canal and foramen contain the infraorbital nerve and vessels. The infraorbital foramen is sometimes double (even multiple), accessory foramina are usually smaller and recorded at incidences of 2-18% in various populations. Lateral wall of the orbit (Fig. 41.15)
The lateral wall is formed by the orbital surface of the greater wing of the sphenoid posteriorly and by the frontal process of the zygomatic bone anteriorly. The bones meet at the sphenozygomatic suture. The zygomatic surface contains the openings of minute canals for the zygomaticofacial and zygomaticotemporal nerves, the former near the junction of the floor and lateral wall, the latter at a slightly higher level, sometimes near the suture. The lateral wall is the thickest wall of the orbit, especially posteriorly where it separates the orbit from the middle cranial fossa. Anteriorly the lateral wall separates the orbit and the infratemporal fossa. The lateral wall and roof are continuous anteriorly but are separated posteriorly by the superior orbital fissure. This lies between the greater wing (below) and lesser wing (above) of the sphenoid, and communicates with the middle cranial fossa. It tapers laterally but widens at its medial end, its long axis descending posteromedially. Where the fissure begins to widen, its inferolateral edges shows a projection, often a spine, for the lateral attachment of the common tendinous ring. An infraorbital sulcus which runs from the superolateral end of the superior orbital fissure towards the orbital floor, has been described, sometimes associated with an anastomosis between the middle meningeal and infraorbital arteries. Apex of the orbit
The apex of the orbit lies near the medial end of the superior orbital fissure and contains the optic canal.
ORBITAL FISSURES AND OPTIC CANAL The superior and inferior orbital fissures and the optic canal open into the orbit and transmit important nerves and vessels. Superior orbital fissure
The superior orbital fissure connects the cranial cavity with the orbit. It is bounded medially by the body of the sphenoid, above by the lesser wing of the sphenoid, below by the medial margin of the orbital surface of the greater wing, and laterally, between the greater and lesser wings of the sphenoid, by the frontal bone. It transmits the oculomotor, trochlear and abducens nerves, branches of the ophthalmic nerve and the ophthalmic veins. Inferior orbital fissure (Fig. 41.14)
The inferior orbital fissure is bounded above by the greater wing of the sphenoid, below by the maxilla and the orbital process of the palatine bone, and laterally by the zygomatic bone or zygomaticomaxillary suture. The maxilla and sphenoid often meet at the anterior end of the fissure, and exclude the zygomatic bone.
The inferior orbital fissure connects the orbit with the pterygopalatine and infratemporal fossae. It transmits the infraorbital and zygomatic branches of the maxillary nerve and accompanying vessels; orbital rami from the pterygopalatine ganglion; and a connection between the inferior ophthalmic vein and pterygoid venous plexus. Anteromedially, lateral to the lacrimal hamulus, a small maxillary depression may mark the attachment of inferior oblique. Optic canal
The lesser wing of the sphenoid is connected to the body of the sphenoid by a thin, flat anterior root and a thick, triangular posterior root. The optic canal lies between these roots and connects the orbit to the middle cranial fossa. It contains the optic nerve and ophthalmic artery. Common tendinous ring (Fig. 41.17)
Many important structures pass through the superior orbital fissure and optic foramen at the apex of the orbit. Their disposition is best understood by referring to the origin of the four recti muscles from a fibrous ring called the common tendinous ring. This ring surrounds the optic canal and encloses part of the superior orbital fissure. Since the optic nerve and ophthalmic artery enter the orbit via the optic canal, they lie within the ring. The superior and inferior divisions of the oculomotor nerve, nasociliary branch of the ophthalmic nerve, and abducens nerve also enter the orbit within the common tendinous ring, but via the superior orbital fissure. The trochlear nerve, and the frontal and lacrimal branches of the ophthalmic nerve enter the orbit through the superior orbital fissure but lie outside the common tendinous ring. Structures entering the orbit through the inferior orbital fissure obviously lie outside the common tendinous ring. page 690 page 691
Figure 41.17 The common tendinous ring with its muscle origins superimposed, and the relative positions of the nerves entering the orbital cavity through the superior orbital fissure and optic canal. Note that the attachments of levator palpebrae superioris and superior oblique lie external to the common tendinous ring but are attached to it. The ophthalmic veins frequently pass through the ring. The recurrent meningeal artery, a branch of the ophthalmic artery, is often conducted from the orbit to the cranial cavity through its own foramen. (Based mainly on the data of Whitnall SE 1932 Anatomy of the Human Orbit, 2nd edn. London: Oxford University Press, and Koornneef (1977). Provided by the late Gordon L Ruskell, Department of Optometry and Visual Science, The City University, London.)
© 2008 Elsevier
EXTRAOCULAR MUSCLES There are seven extraocular or extrinsic muscles associated with the eye. Levator palpebrae superioris is an elevator of the upper eyelid, while the other six, i.e. four recti (superior, inferior, medial and lateral), and two obliqui (superior and inferior), are capable of moving the eye in almost any direction.
LEVATOR PALPEBRAE SUPERIORIS (Figs 41.4, 41.8, 41.17, 41.18) Levator palpebrae superioris is a thin, triangular muscle which arises from the inferior aspect of the lesser wing of the sphenoid, above and in front of the optic canal, and separated from it by the attachment of superior rectus. It has a short narrow tendon at its posterior attachment, and broadens gradually, then more sharply as it passes anteriorly above the eyeball. The muscle ends in front in a wide aponeurosis. Some of its tendinous fibres pass straight into the upper eyelid to attach to the anterior surface of the tarsus, while the rest radiate and pierce orbicularis oculi to pass to the skin of the upper eyelid. The connective tissue coats of the adjoining surfaces of levator palpebrae superioris and superior rectus are fused. Where the two muscles separate to reach their anterior attachments, the fascia between them forms a thick mass to which the superior conjunctival fornix is attached: this is usually described as an additional attachment of levator palpebrae superioris. Traced laterally, the aponeurosis of the levator passes between the orbital and palpebral parts of the lacrimal gland to a tubercle (Whitnall's tubercle) on the zygomatic bone, just within the orbital margin. Traced medially, it loses its tendinous nature as it passes closely over the reflected tendon of superior oblique, and continues on to the medial palpebral ligament as loose strands of connective tissue.
Figure 41.18 A dissection of the left orbit, viewed from in front, to show the origins of the orbital muscles and the relative positions of the nerves of the orbit.
Vascular supply Levator palpebrae superioris receives its arterial supply both directly from the ophthalmic artery and indirectly from its supraorbital branch. Innervation Levator palpebrae superioris is supplied by a branch of the superior division of the oculomotor nerve which enters the inferior surface of the muscle. Sympathetic fibres to the smooth muscle component of levator palpebrae superioris are derived from the plexus surrounding the internal carotid artery. These nerve fibres may join the oculomotor nerve in the cavernous sinus and pass forward in its superior branch. Actions
Levator palpebrae superioris elevates the upper eyelid. During this process the lateral and medial parts of its aponeurosis are stretched and thus limit its action: the elevation is also said to be checked by the orbital septum. 'Check' mechanisms abound in the orbit, but there is little direct evidence that connective tissue structures thus implicated do function in this manner. Elevation of the eyelid is opposed by the palpebral part of orbicularis oculi. Levator palpebrae superioris is linked to superior rectus by a check ligament, thus there is elevation of the upper eyelid when the gaze of the eye is directed upwards. The position of the eyelids depends on reciprocal tone in orbicularis oculi and levator palpebrae superioris, and on the degree of ocular protrusion. In the opened position the margin of the inferior eyelid usually crosses the eyeball level with the lower edge of the circumference of the iris, the upper eyelid covering about half of the width of the upper part of the iris. The eyes are closed by movements of both lids, produced by the contraction of the palpebral part of orbicularis oculi and relaxation of levator palpebrae superioris. In looking upwards, the levator contracts and the upper lid follows the ocular movement. At the same time, the eyebrows are also usually raised by the frontal parts of occipitofrontalis to diminish their overhang. The lower lid lags behind ocular movement, so that more sclera is exposed below the cornea and the lid is bulged a little by the lower part of the elevated eye. When the eye is depressed both lids move, the upper retains its normal relation to the eyeball and still covers about a quarter of the iris. The lower lid is depressed because the extension of the thickened fascia of rectus inferior and obliquus inferior pull on its tarsus as the former contracts. The palpebral fissures are widened in states of fear or excitement by contraction of the smooth muscle of the superior and inferior tarsal muscles, as a result of increased sympathetic activity. Lesions of the sympathetic supply result in drooping of the upper eyelid (ptosis), as seen in Horner's syndrome.
THE FOUR RECTI (Figs 41.18, 41.19, 41.20) page 691 page 692
Figure 41.19 The muscles of the left orbit, lateral view. (Provided by the late Gordon L Ruskell, Department of Optometry and Visual Science, The City University, London.)
The four recti are approximately strap-shaped; each has a thickened middle part which thins gradually to a tendon. They are attached posteriorly to a common tendinous ring around the superior, medial and inferior margins of the optic canal
(Fig. 41.17). This continues laterally across the inferior and medial parts of the superior orbital fissure and is attached to a tubercle or spine on the margin of the greater wing of the sphenoid. The tendon is closely adherent to the dural sheath of the optic nerve medially and to the surrounding periosteum. Inferior rectus, part of medial rectus and the lower fibres of lateral rectus are attached to the lower part of the ring and superior rectus, part of medial rectus and the upper fibres of lateral rectus are attached to the upper part. A second small tendinous slip of lateral rectus is attached to the orbital surface of the greater wing of the sphenoid, lateral to the common tendinous ring. The relationship of structures associated with the common tendinous ring is described on page 690. Each rectus muscle passes forwards, in the position implied by its name, to be attached anteriorly by a tendinous expansion into the sclera, posterior to the margin of the cornea. Superior rectus
Superior rectus is slightly larger than the other recti muscles. It arises from the upper part of the common tendinous ring, above and lateral to the optic canal. Some fibres also arise from the dural sheath of the optic nerve. The fibres pass forwards and laterally (at an angle of c.25° to the median plane of the eye in the primary position) to insert into the upper part of the sclera c.8 mm from the limbus. The insertion is slightly oblique, the medial margin more anterior than the lateral margin. Vascular supply Superior rectus receives its arterial supply both directly from the ophthalmic artery and indirectly from its supraorbital branch. Innervation Superior rectus is supplied by the superior division of the oculomotor nerve which enters the inferior surface of the muscle. Actions Superior rectus moves the eye so that the cornea is directed upwards (elevation) and medially (adduction). To obtain upward movement alone, the muscle must function with inferior oblique. Superior rectus also causes intorsion of the eye (i.e. medial rotation). Because a check ligament extends from the muscle to levator palpebrae superioris, elevation of the cornea also results in elevation of the upper eyelid. For more detailed discussion of its actions, see page 694. Inferior rectus
Inferior rectus arises from the common tendinous ring, below the optic canal. It runs along the orbital floor in a similar direction to superior rectus (i.e. forwards and laterally) and inserts obliquely into the sclera below the cornea, c.6.5 mm from the limbus. Vascular supply Inferior rectus receives its arterial supply from the ophthalmic artery and from the infraorbital branch of the maxillary artery. Innervation Inferior rectus is innervated by a branch of the inferior division of the oculomotor nerve which enters the superior surface of the muscle. Actions The principal activity of inferior rectus is to move the eye so that the cornea is directed downwards (depression). Inferior rectus also causes the cornea to deviate medially. To obtain downward movement alone, inferior rectus must function with superior oblique. Inferior rectus is responsible for extorsion of the eye (i.e. lateral rotation). A ligament passes from the muscle to the inferior tarsal plate of the eyelid, and this causes the lower eyelid to be depressed when inferior rectus contracts. For more detailed discussion of its actions, see page 694. Medial rectus
Medial rectus is slightly shorter than the other recti muscles, but is said to be the
strongest. It arises from the medial part of the common tendinous ring, and also from the dural sheath of the optic nerve, and passes horizontally forwards along the medial wall of the orbit, below superior oblique. Medial rectus inserts into the medial surface of the sclera, c.5.5 mm from the limbus and slightly anterior to the other recti muscles. Vascular supply Medial rectus receives its arterial supply from the ophthalmic artery. Innervation Medial rectus is supplied by a branch from the inferior division of the oculomotor nerve which enters the lateral surface of the muscle. Actions Medial rectus moves the eye so that the cornea is directed medially (adducted). The two medial recti muscles acting together are responsible for convergence. For more detailed discussion of its actions, see page 694. Lateral rectus
Lateral rectus arises from the lateral part of the common tendinous ring and bridges the superior orbital fissure. Some fibres also arise from a spine on the greater wing of the sphenoid. The muscle passes horizontally forward along the lateral wall of the orbit to insert into the lateral surface of the sclera, c.7 mm from the limbus. page 692 page 693
page 693 page 694
Figure 41.20 The geometrical basis of ocular movements. A, The relationship between the orbital and ocular axes, with the eyes in the primary position, where the visual axes are parallel. B and C, The ocular globe in anterior and posterior views to show conventional geometry. A.M.S.Q., anterior medial superior quadrant; P.L.I.Q., posterior lateral inferior quadrant. D, The orbits from above showing the medial and lateral recti and the superior rectus (left) and the inferior rectus (right), indicating turning moments primarily around the vertical axis. E, Superior (left) and inferior (right) oblique muscles showing turning moments primarily around the vertical and also anteroposterior axes. F, Lateral view to show the actions of the superior and inferior recti around the horizontal axis. G, Lateral view to show the action of the superior and inferior oblique muscles around the anteroposterior axis. H, Anterior view to show the medial rotational movement of the superior and inferior recti around the vertical axis.
Conventionally the 12 o'clock position indicated is said to be intorted (superior rectus) or extorted (inferior rectus) as indicated by the small arrows on the cornea. I, Anterior view to show the torsional effects of the superior oblique (intorsion) and inferior oblique (extorsion) around the anteroposterior axis, as indicated by the small arrows on the cornea.
Vascular supply Lateral rectus receives its arterial supply from the ophthalmic artery directly and/or from its lacrimal branch. Innervation Lateral rectus receives its nerve supply from the abducens nerve by branches which enter the medial surface of the muscle. Actions Lateral rectus moves the eye so that the cornea is directed laterally (abducted). For more detailed discussion of its actions, see page 694.
SUPERIOR OBLIQUE (Figs 41.6, 41.8, 41.17, 41.18, 41.19) Superior oblique is a fusiform muscle which arises from the body of the sphenoid superomedial to the optic canal and the tendinous attachment of the superior rectus. It passes forwards to end in a round tendon which plays through a fibrocartilaginous loop, the trochlea, attached to the trochlear fossa of the frontal bone. Tendon and trochlea are separated by a delicate synovial sheath. Having passed through the trochlea, the tendon descends posterolaterally and inferior to superior rectus, and is attached to the sclera in the superolateral part of the posterior quadrant behind the equator, between the superior and lateral recti. Vascular supply Superior oblique receives its arterial supply directly from the ophthalmic artery and indirectly from its supraorbital branch. Innervation Superior oblique is supplied by the trochlear nerve which enters the superior surface of the muscle. Actions Because of its insertion into the posterior part of the eyeball, contraction of superior oblique elevates the back of the eye, which results in depression of the cornea (particularly with the eye in the adducted position). Superior oblique moves the eye laterally and also causes intorsion. For more detailed discussion of its actions, see page 694.
INFERIOR OBLIQUE (Figs 41.6, 41.18, 41.19) Inferior oblique is a thin, narrow muscle near the anterior margin of the floor of the orbit. It arises from the orbital surface of the maxilla lateral to the nasolacrimal groove and ascends posterolaterally, at first between inferior rectus and the orbital floor, and then between the eyeball and lateral rectus. It is inserted into the lateral part of the sclera behind the equator of the eyeball, in the inferolateral part of the posterior quadrant between the inferior and lateral recti, near to, but slightly posterior to, the attachment of superior oblique. The muscle broadens and thins, and, in contrast to the other extraocular muscles, it has a barely discernible tendon at its scleral attachment. Vascular supply Inferior oblique receives its arterial supply from the ophthalmic artery and from the infraorbital branch of the maxillary artery. Innervation Inferior oblique is innervated by a branch of the inferior division of the oculomotor nerve which enters the orbital surface of the muscle. Actions Because of its insertion into the posterior part of the eye, contraction of inferior oblique depresses the back of the eye, which results in elevation of the cornea
(particularly in the adducted position). The muscle moves the eye laterally and also causes extorsion. For more detailed discussion of its actions, see page 694.
MINOR MUSCLES OF THE EYELIDS Several smooth muscles are associated with the orbit, although they are not directly attached to the eyeball. Orbitalis, the orbital muscle of Müller, lies at the back of the orbit and spans the infraorbital fissure. Its functions are uncertain, but its contraction may possibly produce a slight forward protrusion of the eyeball. The superior and inferior tarsal muscles are small muscle laminae inserted into the upper and lower eyelids. They are described in more detail with the tarsal plates (p. 683). Since they are composed of smooth muscles, all three minor muscles receive a sympathetic innervation from the superior cervical ganglion via the internal carotid plexus. They are affected by dysfunction of the sympathetic innervation, e.g. in Horner's syndrome, which means that the upper eyelid droops (ptosis).
MOVEMENTS OF THE EYES Ocular movements are frequently accompanied by movements of the head, which can be likened to the coarse adjustment of an optical instrument, whereas the finer adjustments are made by the ocular musculature. Actions of extraocular muscles (Figs 41.20, 41.21)
Levator palpebrae superioris elevates the upper lid, and its antagonist is the palpebral part of orbicularis oculi. The degree of elevation of the lid, which, apart from blinking, is maintained for long periods during waking hours, is a compromise between ensuring an adequate exposure of the optical media and controlling the amount of incident light. In very bright sunshine, the latter can be reduced by lowering the upper lid, and so limiting glare. Electrically, the levator discharges steadily for a given fixation, but with increasing rates with upward lid position, and relaxes during closure of the palpebral fissure. The role of the superior tarsal muscle is less clear. Its tonus is related to sympathetic nerve activity, since ptosis is a consequence of impairment of the sympathetic nerve supply to the head. The six extraocular muscles all rotate the eyeball in directions dependent upon the geometrical relation between their osseous and global attachments (Fig. 41.20A,D,E): these are altered by the ocular movements themselves. For convenience each muscle is considered in isolation. However, it is essential to appreciate that any movement of an eyeball alters the tension and/or length in all six muscles. Because they form more obvious groupings as antagonists or synergists, it is useful to consider the four recti and two obliques as separate groups, remembering always that they act in concert. It has been suggested that the extrinsic ocular muscles collectively position the eyeball in the orbital cavity and prevent anteroposterior movements of the eyeball, other than a slight retraction during blinks, because the recti exert a posterior traction while the obliques pull the eyeball to some degree anteriorly. They may be assisted by various 'check ligaments' (p. 684). A simplified description of the actions of the extraocular muscles is summarized in Fig. 41.21.
page 694 page 695
Figure 41.21 Simplified summary of the actions of the extraocular muscles.
Of the four recti, the medial and lateral exert comparatively straightforward forces on the eyeball. Being approximately horizontal, when the visual axis is in its primary position, directed to the horizon, they rotate the eye medially (adduction) or laterally (abduction) about an imaginary vertical axis (Fig. 41.20D). They are antagonists. The visual axis can be swept through a horizontal arc by reciprocal adjustment of their lengths. When, as is usual, both eyes are involved, the four medial and lateral recti can either adjust both visual axes in a conjugate movement from point to point at infinity, their axes remaining parallel, or they can converge or diverge the axes to or from nearer or more distant objects of attention in the visual field. The medial and lateral recti do not rotate the eye around its transverse axis and so cannot elevate or depress the visual axes as gaze is transferred from nearer to more distant objects or the reverse. This movement requires the superior and inferior recti (aided by the two oblique muscles). It must be remembered that the orbital axis does not correspond with the visual axis in its primary position but diverges from it at an angle of c.23° (the value varies between individuals, and depends on the angle between the orbital axes and the median plane) (Fig. 41.20A). Thus the simple rotation caused by an isolated superior rectus, analysed with reference to the three hypothetical ocular axes, appears more complex, being primarily elevation (transverse axis), and secondarily a less powerful medial rotation (vertical axis) and slight intorsion (anteroposterior axis) in which the midpoint of the upper rim of the cornea (often referred to as '12 o'clock') is rotated medially towards the nose. These actions, compounded as a single, simple rotation, are easily appreciated when it is seen that the direction of traction of superior rectus runs in a posteromedial direction from its attachment in front, which is anterior to the equator and superior to the cornea, to its osseous attachment near the orbital apex (Fig. 41.20D,H). Inferior rectus pulls in a similar direction to superior rectus, but rotates the visual axis downwards about the transverse axis. It rotates the eye medially on a vertical axis but its action around the anteroposterior axis extorts the eye, i.e. rotates it so that the corneal '12 o'clock' point turns laterally. The combined, equal contractions of the superior and inferior recti therefore rotate the eyeball medially, since their effects around the transverse and anteroposterior axes are opposed. In binocular
movements they assist the medial recti in converging the visual axes, and by reciprocal adjustment they can elevate or depress the visual axes. As the eyeball is rotated laterally, the lines of traction of the superior and inferior recti approach the plane of the anteroposterior ocular axis (Fig. 41.20H), and so their rotational effects about this and the vertical ocular axis diminish. In abduction to c.23°, they become almost purely an elevator and depressor respectively of the visual axis. Superior oblique acts on the eye from the trochlea, and, since the attachment of inferior oblique is for practical purposes vertically below this, both muscles approach the eyeball at the same angle, being attached in approximately similar positions in the superior and inferior posterolateral ocular quadrants (Fig. 41.20I). Superior oblique elevates the posterior aspect of the eyeball, and inferior oblique depresses it, which means that the former rotates the visual axis downwards and the latter rotates it upwards, and both movements occur around the transverse axis. When the eye is in the primary position, the obliquity of both muscles means that they pull in a direction posterior to the vertical axis and both therefore rotate the eye laterally around this axis. With regard to the anteroposterior axis, in isolation, superior oblique intorts the eye and inferior oblique extorts it. Like the superior and inferior recti, therefore, the two obliques have a common turning movement around the vertical axis but they are opposed forces in respect of the other two. Acting in concert they could therefore assist the lateral rectus in abducting the visual axis, as in divergence of the eyes in transferring attention from near to far. Again, like the superior and inferior recti, the directions of traction of the oblique muscles also vary with ocular position, such that they become more nearly a pure elevator and a depressor as the eye is adducted. Ocular rotations are for the most part under voluntary control, whereas torsional movements cannot be voluntarily initiated. When the head is tilted in a frontal plane, reflex torsions occur. Any small lapse in the concerted adjustment of both eyes produces diplopia. Movements that shift or stabilize gaze
The role of eye movements is to bring the image of objects of visual interest onto the fovea of the retina and to hold the image steady in order to achieve and maintain the highest level of visual acuity. Several types of eye movement are required to ensure that these conditions are met. Moreover, the movements of both eyes must be near perfectly matched to achieve the benefits of binocularity. Both volitional and reflexive movements are involved and may be so classified. Alternatively, they may be grouped into those movements that shift gaze as visual interest changes, and those that stabilize gaze by maintaining a steady image on the retina. They have distinct characteristics, and are generated by different neural mechanisms in response to different stimuli, but share a common final motor pathway. Movements that shift gaze are of three types, saccades, vergence, and vestibular-generated rapid changes in fixation. In so-called 'fixation' of a focus of attention, whether uniocular or binocular, the visual axis is not 'fixed' in a perfectly steady manner but undergoes minute, but observable flicking (of a few minutes or even seconds of arc) across the true line of fixation. These microsaccades are rapid and surprisingly complex. When interest changes to another feature of the visual scene, the eyes execute a fast or saccadic movement to take up fixation. If the required rotation is small the saccade is accurate, whereas small supplementary corrective saccades are needed if the shift is substantial. In addition to visually evoked saccades, they may occur in response to other extroreceptive stimuli, e.g. auditory, tactile, or centrally evoked. They may be volitional or reflex. As an example of the latter, in reading a line of print the eyes make three or four jerky saccades rather than following the line smoothly: the line is usefully imaged only when the eye is stationary, consequently little of the line is seen by the centre of the fovea. The term saccade, a French word of obscure origin meaning a 'jerk on the reins', was introduced by Dodge in 1903 for the swings of fixation observed in subjects reading a line of print. In general, reaction times and movements are measured in microseconds; amplitude varies from seconds of arc to many degrees, with an accuracy of 0.2° or better, and the velocity of a large saccade may reach 500°
per second. The speed of saccades is assured by an initial contraction of the appropriate muscles which is slightly excessive. The necessary deceleration when the target is fixated is apparently largely dependent on the viscoelasticity of the extraocular muscles and orbital soft tissues, and not on antagonistic muscular activity. Vergence is a relatively slow movement permitting maintenance of single binocular vision of close objects. The eyes converge towards the midline between the two eyes to achieve imagery of the object on both foveas. The view of the object at the two eyes is not quite the same and the disparity is used to assess depth. Additionally the pupils constrict and the eyes accommodate to achieve sharp focused images. These three activities constitute the near reflex. The vestibular apparatus induces a variety of reflex eye movements to compensate for the potentially disruptive effects on vision caused by head and body movement. Receptors of the semicircular canals respond to active or passive rotational (angular) accelerations of the head. When the body makes substantial rotational movements a vestibulo-ocular reflex generates a cycle of responses involving both the shifting and stabilizing of gaze. Body rotation is matched by counter-rotation of the eyes so that gaze direction is unaltered and clear vision is maintained. Physical constraint limits the rotation to 30° or less and is followed by a rapid saccadic movement of the eyes, a physiological nystagmus, to another object in the visual scene and the cycle is repeated. Consequently vision is clear throughout most of the cycle while the image is stationary, but at the cost of no useful vision during the brief periods of the saccades. The reflex is efficient and rapid. Such speed could not be generated by the visual system which is slow relative to the short latency of vestibular receptors. Other vestibular generated reflexes, which induce compensatory eye movements to stabilize gaze, are activated during brief head movements. When the head is sharply rotated in any direction, the eyeball rotates by an equal amount in the opposite direction as a consequence of the stimulation of semicircular canal cristae (angular acceleration), and gaze is undisturbed. Brief rotational movements are commonly combined with translational (linear acceleration) movements monitored by otolith organs. For example, a linear displacement occurs in walking as the head bobs vertically with each stride, and a rotational displacement occurs as the head rolls, invoking otolith and canal responses respectively to stabilize the retinal image. Head perturbations induced by the vibrations of, for example, an idling bus engine, may generate a vertical linear displacement alone. Vestibular disease incurring the loss of the rapid, fine compensatory eye movements in locomotion destabilizes the retinal image, blurs vision and may render locomotion intolerable. page 695 page 696
The otoliths also respond to the pull of gravity, generating static vestibulo-ocular reflexes associated with head tilt. When static otolith orientation is changed, e.g. when the head is tilted upwards or downwards, the eyes counter-rotate to maintain fixation of the horizontal meridian. Lateral tilt towards a shoulder generates a torsional counter-rotation of the eyes, a movement which cannot be made voluntarily. The torsional tilt reflex, equal and opposite in direction by the two eyes, is fully compensatory over 40° or so in afoveate animals, but in man it is vestigial: it is fractionally compensatory and varies in extent between individuals. Because the foveal image is unaffected by torsional movements, the subject is unaware of any visual penalty. Pursuit eye movements are used to track a moving object of visual interest, maintaining the image approximately on the fovea. They are usually preceded by a saccade to capture the image but, unlike saccades, they are slow and motivated by vision. If the angular shift required to track the moving object is large or is moving swiftly, the initial saccade is frequently inaccurate and one or more small corrective saccades are made before tracking begins. Because the stimulus is visual, the pursuit system response is subject to a relatively long latency (c.100 msec): the limitation in performance this imposes may be offset by a predictive capacity when object movement follows a regular pattern, and the eye
movements adjust in anticipation to speed and direction. The optokinetic response is another visually mediated reflex which stabilizes retinal imagery during rotational movement. As the visual scene changes the eyes follow, holding the retinal image steady until the eyes shift rapidly in the opposite direction to another area of the visual scene. The full field of vision, rather than small objects within it, is the stimulus, and the alternating slow and fast phases of movement generated describes optokinetic nystagmus. This reflex functions in collaboration with the rotational vestibulo-ocular reflex. In sustained rotations of the body, the vestibulo-ocular reflex fades because of the mechanical arrangements of the semicircular canals. In darkness the reflex, which is initially compensatory, loses velocity, and after c.45 seconds the eyeballs become stationary. With a visual input, the reflex is sustained by the optokinetic response. Because the reflex is already initiated, the relative delay of visual input is overcome. The integration of the two systems is served by an accessory optic system projection to the vestibular nuclei via the inferior olive and cerebellum. The usual method of evoking optokinetic nystagmus in the laboratory or clinic is to present a horizontally moving pattern of vertical black-on-white stripes while the head of the subject is held stationary. Saccadic activity is almost ever-present in human vision. Thus both visual axes are endlessly and rapidly transferred to new points of interest in any part of the total visual field. Binocular gaze is very frequently made to travel routes of the most variable complexity in examining objects of some extent in the field, and both visual axes must be maintained with sufficient accuracy to avoid diplopia. Binocular movements involving convergence are markedly slower than conjugate movements, which presumably reflects the greater complexity of neural control that these movements require (and the speed of contraction of ciliaris must be a factor). Most human visual activity concerns targets of regard near enough to demand convergence and hence a neuronal intermediation of greater flexibility. Since the prime purpose is the clear perception of a 'target', it is not surprising that the visual input is itself utilized in continuous feedback to achieve the correct aiming of visual axes. Continual movements of the eyeball appear to be essential for vision to occur. Retinal and more central neural networks appear to be designed primarily to detect transient events such as movements rather than static, maintained stimuli. Indeed, images which are essentially static, such as those due to retinal blood vessels, are not detectable unless the shadows they cast on photoreceptors are made to move, e.g. by shifting narrow-angle illumination with an ophthalmoscope. The central control of conjugate gaze is discussed in Section 2.
© 2008 Elsevier
VASCULAR SUPPLY AND LYMPHATIC DRAINAGE ARTERIES WITHIN THE ORBIT The main vessel supplying orbital structures is the ophthalmic artery. Its terminal branches anastomose on the face and scalp with those of the facial, maxillary and superficial temporal arteries, thereby establishing connections between the external and internal carotid arteries. In addition to the ophthalmic artery, the infraorbital branch of the maxillary artery, and possibly the recurrent meningeal artery, supply orbital structures. Ophthalmic artery (Fig. 41.22)
Figure 41.22 Distribution of the branches of the ophthalmic artery viewed from above. The artery has several anastomoses with branches of the external carotid artery, e.g. the middle meningeal with the recurrent meningeal, the facial (angular) with the frontal or dorsal nasal, the superficial temporal with the supraorbital (inconstant). (By permission from MacKinnon P, Morris J 1990 Oxford Textbook of Functional Anatomy, Vol 3. Head and Neck. Oxford: Oxford University Press. By permission of Oxford University Press.)
The ophthalmic artery leaves the internal carotid artery as it quits the cavernous sinus medial to the anterior clinoid process. It enters the orbit by the optic canal, inferolateral to the optic nerve. For a short distance it is then lateral to the optic nerve and medial to the oculomotor and abducens nerves, the ciliary ganglion
and lateral rectus. It crosses between the optic nerve and superior rectus to reach the medial wall of the orbit, running between superior oblique and medial rectus. At the medial end of the upper eyelid, it divides into supratrochlear and dorsal nasal branches. As it crosses the optic nerve with the nasociliary nerve, it is separated from the frontal nerve by superior rectus and levator palpebrae superioris. Its terminal branch accompanies the infratrochlear nerve. In c.15% of subjects the ophthalmic artery crosses below the optic nerve. It has the following branches: central artery of the retina, lacrimal artery, muscular branches, ciliary arteries, supraorbital artery, posterior ethmoidal artery, anterior ethmoidal artery, meningeal branch, medial palpebral arteries, supratrochlear artery, dorsal nasal artery. Many of the branches of the ophthalmic artery accompany sensory nerves of the same name and have a similar distribution. Central artery of the retina
The small central artery of the retina is the first branch. It begins below the optic nerve and for a short distance it lies in the dural sheath of the nerve. It enters the inferomedial surface of the nerve c.1.25 cm behind the eye, and runs to the retina along its axis. Its further distribution is described on page 716. Muscular branches
Muscular branches frequently spring from a common trunk to form superior and inferior groups, most of which accompany branches of the oculomotor nerve. The inferior, more constant, contains most of the anterior ciliary arteries. Other muscular vessels branch from the lacrimal and supraorbital arteries or from the trunk of the ophthalmic artery. page 696 page 697
Ciliary arteries There are three groups of ciliary arteries: long and short posterior, and anterior. Long posterior ciliary arteries, usually two, pierce the sclera near the optic nerve, pass anteriorly along the horizontal meridian and join the greater arterial circle in the iris. About seven short posterior ciliary arteries pass close to the optic nerve to reach the eyeball where they divide into 15-20 branches. They pierce the sclera around the optic nerve to supply the choroid, and anastomose with twigs of the central retinal artery at the optic disc. Anterior ciliary arteries arise from muscular branches of the ophthalmic artery. They reach the eyeball on the tendons of the recti, form a circumcorneal subconjunctival vascular zone, and pierce the sclera near the sclerocorneal junction to end in the greater arterial circle of the iris. Lacrimal artery
The lacrimal artery is a large branch which usually leaves the ophthalmic artery near its exit from the optic canal, although it occasionally arises before the ophthalmic artery enters the orbit. It accompanies the lacrimal nerve along the upper border of lateral rectus, supplies and traverses the lacrimal gland, and ends in the eyelids and conjunctiva as the lateral palpebral arteries. The latter run medially in the upper and lower lids and anastomose with the medial palpebral arteries.The lacrimal artery gives off one or two zygomatic branches. One reaches the temporal fossa via the zygomaticotemporal foramen, and anastomoses with the deep temporal arteries. The other reaches the cheek by the zygomaticofacial foramen, and anastomoses with transverse facial and zygomatico-orbital arteries. A recurrent meningeal branch, usually small, passes back via the lateral part of the superior orbital fissure to anastomose with a middle
meningeal branch: it is sometimes large, replacing the lacrimal artery, and becomes a greater contributor to the orbital blood supply. Supraorbital artery
The supraorbital artery leaves the ophthalmic artery where it crosses the optic nerve, ascends medial to superior rectus and levator palpebrae superioris, meets the supraorbital nerve and runs with it between the periosteum and levator palpebrae superioris to the supraorbital foramen (or notch). It passes through the foramen and divides into superficial and deep branches which supply the skin, muscles and frontal periosteum, and anastomose with the supratrochlear artery, and the frontal branch of the superficial temporal artery and its contralateral fellow. It supplies superior rectus and levator palpebrae superioris, and sends a branch across the trochlea to the medial canthus. At the supraorbital margin it often sends a branch to the diploe of the frontal bone and may also supply the mucoperiosteum in the frontal sinus. Posterior ethmoidal artery
The posterior ethmoidal artery runs through the posterior ethmoidal canal and supplies the posterior ethmoidal air sinuses. It enters the cranium, sends a meningeal branch to the dura mater and nasal branches which descend into the nasal cavity via the cribriform plate, and anastomoses with branches of the sphenopalatine artery. Anterior ethmoidal artery
The anterior ethmoidal artery passes with its accompanying nerve through the anterior ethmoidal canal to supply ethmoidal and frontal air sinuses. Entering the cranium, it gives off a meningeal branch to the dura mater and nasal branches which descend into the nasal cavity with the anterior ethmoidal nerve. It runs in a groove on the deep surface of the nasal bone to supply the lateral nasal wall and septum. A terminal branch appears on the nose between the nasal bone and the upper nasal cartilage. Meningeal branch
A meningeal branch, usually small, passes back through the superior orbital fissure to the middle cranial fossa, and anastomoses with the middle and accessory meningeal arteries. It is sometimes large when it becomes a major contributor to the orbital blood supply. Medial palpebral arteries
The medial palpebral arteries are described on page 510. Supratrochlear artery
The supratrochlear artery is a terminal branch of the ophthalmic artery. It leaves the orbit superomedially with the supratrochlear nerve, ascends on the forehead to supply the skin, muscles and pericranium, and anastomoses with the supraorbital artery and with its contralateral fellow. Dorsal nasal artery
The dorsal nasal artery is the other terminal branch of the ophthalmic artery, and emerges from the orbit between the trochlea and medial palpebral ligament. It gives a branch to the upper part of the nasolacrimal sac and then divides into two branches. One branch joins the terminal part of the facial artery, and the other runs along the dorsum of the nose, supplies its outer surface and anastomoses
with its contralateral fellow and the lateral nasal branch of the facial artery. Infraorbital branch of the maxillary artery
The infraorbital branch of the maxillary artery enters the orbit through the posterior part of the inferior orbital fissure. It passes along the infraorbital groove of the maxilla in the floor of the orbit before entering the infraorbital canal, and comes out onto the face through the infraorbital foramen. Whilst in the infraorbital groove, it gives off branches which supply inferior rectus and inferior oblique, the nasolacrimal sac and, inconstantly, the lacrimal gland.
VEINS WITHIN THE ORBIT The veins draining the orbit are the superior and inferior ophthalmic veins and the infraorbital vein. The veins of the eyeball mainly drain into the vortex veins. Superior and inferior ophthalmic veins (Figs 41.7, 41.17, 41.23)
The superior and inferior ophthalmic veins link the facial and intracranial veins. They are devoid of valves. The superior ophthalmic vein forms posteromedial to the upper eyelid from two tributaries which connect anteriorly with the facial and supraorbital veins. It runs with the ophthalmic artery, lying between the optic nerve and superior rectus, and receives the corresponding tributaries, the two superior vortex veins of the eyeball, and the central vein of the retina. The central vein of the retina sometimes drains directly into the cavernous sinus, although it still gives a communicating branch to the superior ophthalmic vein. The superior ophthalmic vein may also receive the inferior ophthalmic vein. It traverses the superior orbital fissure usually above the common tendinous ring of the recti muscles and ends in the cavernous sinus.
page 697 page 698
Figure 41.23 The veins of the left orbit: lateral aspect. Note that the eyeball is shown at c.50% of real size, relative to the orbit, to reveal the veins.
The inferior ophthalmic vein begins in a network near the anterior region of the orbital floor and medial wall. It runs backwards on inferior rectus and across the inferior orbital fissure, and then either joins the superior ophthalmic vein or passes through the superior orbital fissure - within or below the common tendinous ring of
the recti muscles - to drain directly into the cavernous sinus. The inferior ophthalmic vein receives tributaries from inferior rectus and inferior oblique, the nasolacrimal sac and the eyelids. It also receives the two inferior vortex veins of the eyeball. The inferior ophthalmic vein communicates with the pterygoid venous plexus by a branch which passes through the inferior orbital fissure. It may also communicate with the facial vein across the inferior margin of the orbit. The infraorbital vein
The infraorbital vein runs with the infraorbital nerve and artery in the floor of the orbit, and passes backwards through the inferior orbital fissure into the pterygoid venous plexus. It drains structures in the floor of the orbit and communicates with the inferior ophthalmic vein. The infraorbital vein may communicate with the facial vein on the face. Central retinal vein
The central retinal vein first traverses the optic nerve, and receives branches which drain the optic nerve, including a central vein which drains forwards. It then leaves the nerve to pursue a short course in the subarachnoid space before entering the cavernous sinus or the superior ophthalmic vein.
LYMPHATIC DRAINAGE Lymphatic vessels other than those draining the conjunctiva have not been identified.
© 2008 Elsevier
INNERVATION Somatic and autonomic motor and somatic sensory nerves are found in the orbit. The motor nerves are the oculomotor, trochlear and abducens nerves, and they supply the extraocular muscles. Parasympathetic fibres from the oculomotor nerve (via the ciliary ganglion) supply sphincter pupillae and ciliaris, and from the facial nerve (via the pterygopalatine ganglion) supply the lacrimal gland and choroid. Sympathetic fibres supply dilator pupillae. Both sympathetic and parasympathetic nerves supply the arteries. The sensory nerves within the orbit are the optic, ophthalmic and maxillary nerves, although the maxillary nerve and most ophthalmic branches only pass through the orbit to supply the face and jaws.
OCULOMOTOR NERVE (Figs 41.17, 41.18, 41.24) The oculomotor nerve is the third cranial nerve. It is the main source of innervation to the extraocular muscles and also contains parasympathetic fibres which relay in the ciliary ganglion.
Figure 41.24 The nerves of the left orbit: superior aspect.
The oculomotor nerve emerges at the midbrain, on the medial side of the crus of the cerebral peduncle. It passes along the lateral dural wall of the cavernous sinus where it divides into superior and inferior divisions which run beneath the trochlear and ophthalmic nerves. The two divisions of the oculomotor nerve enter the orbit through the superior orbital fissure, within the common tendinous ring of the recti muscles, separated by the nasociliary branch of the ophthalmic nerve. The superior division of the oculomotor nerve passes above the optic nerve to enter the inferior (ocular) surface of superior rectus. It supplies this muscle and gives off a branch which runs to supply levator palpebrae superioris. The inferior division of the oculomotor nerve divides into three branches, medial, central and lateral. The medial branch passes beneath the optic nerve to enter the lateral (ocular) surface of medial rectus. The central branch runs downwards and forwards to enter the superior (ocular) surface of inferior rectus. The lateral branch travels forwards on the lateral side of inferior rectus to enter the orbital
surface of inferior oblique. The lateral branch also communicates with the ciliary ganglion to distribute parasympathetic fibres to sphincter pupillae and ciliaris.
TROCHLEAR NERVE (Fig. 41.24) The trochlear nerve is the fourth cranial nerve and is the only cranial nerve which emerges from the dorsal surface of the brain. It passes from the midbrain onto the lateral surface of the crus of the cerebral peduncle and runs through the lateral dural wall of the cavernous sinus. It then crosses the oculomotor nerve and enters the orbit through the superior orbital fissure, above the common tendinous ring of the recti muscles and levator palpebrae superioris, and medial to the frontal and lacrimal nerves. The trochlear nerve travels but a short distance to enter the superior (orbital) surface of superior oblique, which is its sole target.
ABDUCENS NERVE (Figs 41.24, 41.25) The abducens nerve is the sixth cranial nerve, and emerges from the brain stem between the pons and the medulla oblongata. It is related to the cavernous sinus, but unlike the oculomotor, trochlear, ophthalmic and maxillary nerves, which merely invaginate the lateral dural wall, it passes through the sinus itself, lying lateral to the internal carotid artery. The abducens nerve enters the orbit through the superior orbital fissure, within the common tendinous ring of the recti muscles (Fig. 41.17), at first below, and then between, the two divisions of the oculomotor nerve and lateral to the nasociliary nerve. It passes forwards to enter the medial (ocular) surface of lateral rectus, which is its sole target.
OPHTHALMIC NERVE The ophthalmic nerve, a division of the trigeminal nerve, travels through the orbit to supply targets that are primarily in the upper part of the face. It arises from the trigeminal ganglion in the middle cranial fossa and passes forwards along the lateral dural wall of the cavernous sinus. It gives off three main branches, the lacrimal, frontal and nasociliary nerves, just before it reaches the superior orbital fissure. Lacrimal nerve (Figs 41.18, 41.24, 41.25)
The lacrimal nerve enters the orbit through the superior orbital fissure, above the common tendinous ring of the recti muscles, and lateral to the frontal and trochlear nerves. It passes forwards along the lateral wall of the orbit on the superior border of lateral rectus, and travels through the lacrimal gland and the orbital septum to supply conjunctiva and skin covering the lateral part of the upper eyelid. The lacrimal nerve communicates with the zygomatic branch of the maxillary nerve, and so parasympathetic fibres associated with the pterygopalatine ganglion might be conveyed to the lacrimal gland. Frontal nerve (Figs 41.18, 41.24, 41.25)
The frontal nerve is the largest branch of the ophthalmic nerve. It enters the orbit through the superior orbital fissure, above the common tendinous ring of the recti muscles, and lies between the lacrimal nerve laterally and the trochlear nerve medially. It passes forwards on levator palpebrae superioris, towards the rim of the orbit: about halfway along this course it divides into the supraorbital and supratrochlear nerves. page 698 page 699
Figure 41.25 The nerves of the left orbit and the ciliary ganglion: lateral aspect.
The supraorbital nerve is the larger of the terminal branches of the frontal nerve. It continues forwards along levator palpebrae superioris and leaves the orbit through the supraorbital notch or foramen to emerge onto the forehead. The supraorbital nerve supplies the mucous membrane which lines the frontal sinus, skin and conjunctiva covering the upper eyelid, and skin over the forehead and scalp. The postganglionic sympathetic fibres which innervate the sweat glands of the supraorbital area probably travel in the supraorbital nerve, having entered the ophthalmic nerve through its communication with the abducens nerve within the cavernous sinus. The supratrochlear nerve runs medially above the pulley for superior oblique. It gives a descending branch to the infratrochlear nerve and ascends onto the forehead through the frontal notch to supply skin and conjunctiva covering the upper eyelid, and skin over the forehead. Nasociliary nerve (Figs 41.18, 41.24, 41.25)
The nasociliary nerve is intermediate in size between the frontal and lacrimal nerves, and is more deeply placed in the orbit, which it enters through the common tendinous ring, lying between the two rami of the oculomotor nerve. It crosses the optic nerve with the ophthalmic artery and runs obliquely below superior rectus and superior oblique to reach the medial orbital wall. Here, as the anterior ethmoidal nerve, it passes through the anterior ethmoidal foramen and canal and enters the cranial cavity. It runs forwards in a groove on the upper surface of the cribriform plate beneath the dura mater and descends through a slit lateral to the crista galli into the nasal cavity, where it occupies a groove on the internal surface of the nasal bone and gives off two internal nasal branches (p. 573). The medial internal nasal nerve supplies the anterior septal mucosa, and the lateral internal nasal nerve supplies the anterior part of the lateral nasal wall. The anterior ethmoidal nerve emerges, as the external nasal nerve (p. 512), at
the lower border of the nasal bone, and descends under the transverse part of nasalis to supply the skin of the nasal ala, apex and vestibule. The nasociliary nerve has connections with the ciliary ganglion and has long ciliary, infratrochlear and posterior ethmoidal branches. The ramus communicans to the ciliary ganglion usually branches from the nerve as it enters the orbit lateral to the optic nerve. It is sometimes joined by a filament from the internal carotid sympathetic plexus or from the superior ramus of the oculomotor nerve as it enters the posterosuperior angle of the ganglion. Two or three long ciliary nerves branch from the nasociliary nerve as it crosses the optic nerve (Fig. 41.25). They accompany the short ciliary nerves and pierce the sclera near the attachment of the optic nerve. Running forwards between sclera and choroid, they supply the ciliary body, iris and cornea and are said to contain postganglionic sympathetic fibres for the dilator pupillae from neurones in the superior cervical ganglion. An alternative pathway for the supply of the dilator pupillae is via the sympathetic root associated with the ciliary ganglion. The posterior ethmoidal nerve leaves the orbit by the posterior ethmoidal foramen and supplies the ethmoidal and sphenoidal sinuses.
MAXILLARY NERVE The maxillary nerve is a sensory division of the trigeminal nerve. Most of the branches from the maxillary nerve arise in the pterygopalatine fossa. It gives rise to the zygomatic and infraorbital nerves that pass into the orbit through the inferior orbital fissure and two others that pass through the pterygopalatine ganglion without synapsing and are distributed to the nose, palate and pharynx. Zygomatic nerve
The zygomatic nerve is located close to the base of the lateral wall of the orbit. It soon divides into two branches, the zygomaticotemporal and the zygomaticofacial nerves, which run for only a short distance in the orbit before passing onto the face through the lateral wall of the orbit. They may either enter separate canals within the zygomatic bone or the zygomatic nerve itself may enter the bone before dividing. The zygomaticotemporal nerve exits the zygomatic bone on its medial surface, and pierces the temporal fascia to supply the skin over the temple. It also gives a branch to the lacrimal nerve which may carry parasympathetic fibres to the lacrimal gland (Figs 41.25, 30.6). The zygomaticofacial nerve leaves the zygomatic bone on its lateral surface to supply skin overlying the prominence of the cheek. Infraorbital nerve (Fig. 41.25)
The infraorbital nerve initially lies in the infraorbital groove on the floor of the orbit. As it approaches the rim of the orbit it runs into the infraorbital canal through which it passes to emerge onto the face at the infraorbital foramen. The infraorbital nerve supplies the skin of the lower eyelid, possibly the conjunctiva, and skin over the upper jaw, and also provides the middle and anterior superior alveolar nerves. Orbital branches of pterygopalatine ganglion
Several rami orbitales arise dorsally from the pterygopalatine ganglion and enter the orbit through the inferior orbital fissure. Branches leave the orbit through the posterior ethmoidal air sinus. There is strong experimental evidence from studies of animals including monkeys that postganglionic parasympathetic branches pass directly to the lacrimal gland, ophthalmic artery and choroid. page 699 page 700
OPTIC NERVE (Figs 41.17, 41.18) The optic nerve is the second cranial nerve. It arises from the optic chiasma on the floor of the diencephalon and enters the orbit through the optic canal, accompanied by the ophthalmic artery. It changes its shape from being flattened at the chiasma to rounded as it passes through the optic canal. In the orbit it passes forwards, laterally and downwards, and pierces the sclera at the lamina cribrosa, slightly medial to the posterior pole. It has a somewhat tortuous course within the orbit to allow for movements of the eyeball. It is surrounded by extensions of the three layers of meninges. The optic nerve has important relationships with other orbital structures. As it leaves the optic canal, it lies superomedial to the ophthalmic artery, and is separated from lateral rectus by the oculomotor, nasociliary and abducens nerves, and sometimes by the ophthalmic veins. The optic nerve is closely related to the origins of the four recti muscles, whereas more anteriorly, where the muscles diverge, it is separated from them by a substantial amount of orbital fat. Just beyond the optic canal, the ophthalmic artery and the nasociliary nerve cross the optic nerve to reach the medial wall of the orbit. The central artery of the retina enters the substance of the optic nerve about halfway along its length. Near the back of the eye, it becomes surrounded by long and short ciliary nerves and vessels.
CILIARY GANGLION (Figs 41.25, 41.26) The ciliary ganglion is a parasympathetic ganglion which is concerned functionally with the motor innervation of certain intraocular muscles. It is a small, flat, reddish-grey swelling, 1-2 mm in diameter, connected to the nasociliary nerve, and located near the apex of the orbit in loose fat c.1 cm in front of the medial end of the superior orbital fissure. It lies between the optic nerve and lateral rectus, usually lateral to the ophthalmic artery. Its neurones, which are multipolar, are larger than in typical autonomic ganglia; a very small number of more typical neurones are also present. Its connections or roots enter or leave it posteriorly. Eight to ten delicate filaments, termed the short ciliary nerves, emerge anteriorly from the ganglion arranged in two or three bundles, the lower being larger. They run forwards sinuously with the ciliary arteries, above and below the optic nerve, and divide into 15-20 branches that pierce the sclera around the optic nerve and run in small grooves on the internal scleral surface. They convey parasympathetic, sympathetic and sensory fibres between the eyeball and the ciliary ganglion: only the parasympathetic fibres synapse in the ganglion.
Figure 41.26 The ciliary ganglion, with its roots and branches of distribution. Red, sympathetic fibres; heavy black, parasympathetic fibres; blue, sensory (cerebrospinal) fibres. Alternative pathways are given for the sympathetic fibres.
The parasympathetic root, derived from the branch of the oculomotor nerve to the inferior oblique, consists of preganglionic fibres from the Edinger-Westphal
nucleus, which relay in the ganglion. Postganglionic fibres travel in the short ciliary nerves to the sphincter pupillae and ciliaris. More than 95% of these fibres supply ciliaris, which is much the larger muscle in volume. The sympathetic root contains fibres from the plexus around the internal carotid artery within the cavernous sinus. These postganglionic fibres, derived from the superior cervical ganglion, form a fine branch which enters the orbit through the superior orbital fissure inside the common tendinous ring of the recti muscles. The fibres either pass directly to the ganglion, or join the nasociliary nerve and travel to the ganglion in its sensory root. Either way, they traverse the ganglion without synapsing to emerge into the short ciliary nerves. They are distributed to the blood vessels of the eyeball. Sympathetic fibres innervating dilator pupillae may sometimes travel via the short ciliary nerves (rather than the more usual route via the ophthalmic, nasociliary and long ciliary nerves). The sensory fibres which pass through the ciliary ganglion are derived from the nasociliary nerve. They enter the short ciliary nerves and carry sensation from the cornea, the ciliary body and the iris. REFERENCES Hayreh SS 1942 The ophthalmic artery. III Branches. Br J Ophthalmol 46: 212-46. Jones LT 1964 The anatomy of the upper eyelid and its relation to ptosis surgery. Am J Ophthalmol 57: 943-59. Medline Similar articles Knop E, Knop N 2002 A functional unit for ocular surface immune defence formed by the lacrimal gland, conjunctiva and lacrimal drainage system. Adv Exp Med Biol 506B: 635-44. Koornneef L 1977 Spatial Aspects of Orbital Musculo-fibrous Tissue in Man. Amsterdam: Swestsa Zeitlinger. Leigh RJ, Zee DS 1999 The Neurology of Eye Movement. 3rd edition. Oxford: Oxford University Press. Ruskell GL 1975 Nerve terminals and epithelial cell variety in the human lacrimal gland. Cell Tiss Res 138: 121-36.
© 2008 Elsevier
42 The eye The eyeball, the peripheral organ of vision, is situated in a skeletal cavity, the orbit, the walls of which help to protect it from injury. The orbit also has a more fundamental role in the visual process itself, in providing a rigid support and direction to the eye and in forming the sites of attachment for its external muscles. This setting permits the accurate positioning of the visual axis under neuromuscular control, and determines the spatial relationship between the two eyes - essential for binocular vision and conjugate eye movements. The eyeball is embedded in orbital fat, separated from it by a thin fascial sheath (Chapter 41). It is composed of the segments of two spheres of different radii. The anterior segment, part of the smaller sphere, is transparent and forms c.7% of the surface of the whole globe. It is more prominent than the posterior segment, which is part of a larger sphere and opaque, and forms the remainder of the globe. The anterior segment is bounded by the cornea and the lens, and is incompletely subdivided into anterior and posterior chambers by the iris. These chambers are continuous through the pupil. The anterior chamber is slightly overlapped by the sclera peripherally. The angle between the iris and cornea therefore forms an annulus of greater diameter than the limbus, the junction between the sclera and cornea. The difference between these two varies from 1 to 2 mm, the angle being deeper above and below than at the sides of the eyeball. The posterior chamber lies between the posterior surface of the iris and the anterior aspect of the lens and its supporting ligament, the zonule, and is triangular in section. The apex of the triangle is the point where the iris touches the lens, and the base, or zonular region, extends among the collagenous bundles of the zonule, sometimes even into a retrozonular space between the zonule and the vitreous humour in the posterior segment of the eyeball. The posterior segment consists of the parts of the eye posterior to the zonule and lens. The anterior pole is the centre of the anterior (corneal) curvature, and the posterior pole is the centre of its posterior (scleral) curvature; a line joining these two points forms the optic axis. (By the same convention, the eye has an equator, equidistant between the poles: any circumferential line joining the poles is a meridian.) The optic axes of the two eyes, in their primary position, are parallel and do not correspond with the orbital axes, which diverge anterolaterally at a marked angle to each other (Chapter 41). The optic nerves follow the orbital axes and are therefore not parallel; each enters its eye c.3 mm medial (nasal) to the posterior pole. The ocular vertical diameter (23.5 mm) is rather less than the transverse and anteroposterior diameters (24 mm); the anteroposterior diameter at birth is c.17.5 mm and at puberty 20-21 mm; it may vary considerably in myopia (c.29 mm) and in hypermetropia (c.20 mm). In females all diameters are on average slightly less than in the male.
© 2008 Elsevier
OCULAR FIBROUS TISSUE The eye has three layers enclosing its contents. From the outer surface these are a fibrous layer, which consists of the sclera behind and the cornea in front; a vascular, pigmented layer which consists of (from behind forwards) the choroid, ciliary body and iris, collectively termed the uveal tract; and a neural layer, known as the retina. The fibrous layer of the eyeball (Fig. 42.1) has an opaque posterior sclera and a transparent anterior cornea. Together these form the protective enclosing capsule of the eye, a semi-elastic structure which when made turgid by intraocular pressure, determines with great precision the optical geometry of the visual apparatus. The sclera also provides attachments for the extraocular muscles which rotate the eye, its smooth external surface rotating easily on the adjacent tissues of the orbit. The cornea admits light, refracts it towards a retinal focus, and plays an important role in the image-processing mechanism.
Sclera (Fig. 42.1) The sclera, so named from its relatively hard consistency, is a dense layer. When distended by intraocular pressure it maintains the shape of the eyeball. It is thickest (c.1.1 mm) posteriorly, near the optic nerve entry point, and is thinnest (0.4 mm) at the equator and at the attachments of the recti (Chapter 41). Its external surface is covered by a delicate episcleral lamina of loose fibrous tissue which contains sparse blood vessels and is in contact with the inner surface of the fascial sheath of the eyeball. The anterior part is covered by conjunctiva which is reflected onto it from the deep surfaces of the eyelids. The scleral internal surface is attached to the choroid by a delicate fibrous layer, the suprachoroid lamina, which contains numerous fibroblasts and melanocytes. Anteriorly, it is attached to the ciliary body by the lamina supraciliaris. Posteriorly, the sclera is pierced by the optic nerve and is continuous with the fibrous nerve sheath and hence with the dura mater. The sclera has the appearance of a perforated plate, the lamina cribrosa sclerae, where the nerve pierces it (Fig. 42.2): the optic nerve fascicles pass through these minute orifices. The central retinal artery and vein pass through a larger, central aperture. Numerous small apertures transmit the ciliary vessels and nerves through the sclera close to the perimeter of the cribriform plate. Just behind the equator four larger apertures transmit the venae vorticosae. The lamina cribrosa is the weakest part of the sclera and bulges outwards in the condition of a cupped disc when intraocular pressure is raised chronically as in glaucoma. Anteriorly, the sclera is directly continuous with the cornea at the corneoscleral junction or limbus (Fig. 42.3). Near the internal surface of the sclera there is an annular endothelial canal, the sinus venosus sclerae (canal of Schlemm), at this junction. In section, the canal appears as an oval cleft, whose outer wall grooves the sclera. Posteriorly, the cleft extends as far as a rim of scleral tissue, the scleral spur, which in section forms a triangle with its apex directed forwards. The sinus may be double or multiple in part of its course. Its inner wall, adjoining the aqueous chamber, consists of loose trabecular tissue continuous anteriorly with the posterior limiting lamina and endothelium of the cornea. There are spaces among its fibres through which aqueous humour filters from the anterior chamber to the sinus, draining peripherally to the anterior ciliary veins. Most of the fibres of the trabecular tissue mentioned above are attached to the anterior, external aspect of the scleral spur. Of the remainder, most are continuous with meridional fibres of the ciliary muscle, some of which attach to the posterior internal aspect of the scleral spur. The iridocorneal angle of the anterior chamber is bordered anteriorly by trabecular tissue and the scleral spur, and posteriorly by the periphery of the iris. page 701 page 702
Figure 42.1 The organization of the eye, viewed from above. In this illustration the left eye and part of the lower eyelid are depicted in horizontal section and also cut away to show internal structure.
The sclera is composed of dense collagenous tissue mixed with occasional elastic fibres and interspersed with flat fibroblasts. Collagen forms 75% of the dry scleral weight, a factor which is important in the regulation of intraocular pressure. Fibre bundles are arranged circumferentially around the optic disc, and around the orifices of the lamina cribrosa. Elsewhere on the external surface of the sclera, fibres are arranged mostly in a reticular manner. The fibres of the tendons of the recti muscles intersect scleral fibres at right angles at their attachments, and then interlace deeper in the sclera. Collagen fibres of the scleral spur are orientated circularly, and there is an increased incidence of elastic fibres here. Scleral vessels are few and mainly disposed in the episcleral lamina, especially close to the limbus. The sclera provides passage for nerves of the cornea and vascular autonomic nerves, but its own innervation is sparse.
Cornea (Figs 42.3, 42.4, 42.5) The cornea is the anterior, projecting, transparent part of the external tunic. Its
tear film cover is the major site of refraction of light which enters the eye. Convex anteriorly, it projects from the sclera as a dome-shaped elevation forming c.7% of the external tunic area. Its curvature changes very little after the first year or so of life. Since it is more curved (radius (r) = 6.8-8.5 mm, averaging 7.8 mm) than the sclera (r = 11.5 mm), a slight sulcus sclerae marks the corneoscleral junction. Corneal thickness is c.0.7 mm close to the corneoscleral junction, and 0.5-0.6 mm at its centre. Viewed from in front, the corneal perimeter is slightly elliptical, and its transverse diameter is a little greater than its vertical. Its posterior perimeter is circular and is more extensive than the anterior surface in its vertical axis, because in section the corneoscleral junction is slightly oblique above and below. The corneal diameter is c.11.7 mm on its posterior aspect; anteriorly it is 11.7 mm horizontally and 10.6 mm vertically. Of the total 60 dioptres refractive power of the eye, the cornea provides c.40 dioptres. page 702 page 703
Figure 42.2 Exit of the human optic nerve from the eyeball, showing the distribution of collagenous (blue) and neuroglial (magenta) tissues. Sep, septa of collagenous connective tissue carried into the nerve from the pia mater and dividing the nerve fibres into numerous fascicles; Gl.M, astroglial membrane separating nerve fibres from connective tissue; Gl.C, astrocytes and oligodendrocytes among the fibres in their fascicles; Du, Ar, Pia, dura, arachnoid and pia maters respectively. (1a) is the internal limiting membrane of the retina, which is continuous with an astroglial membrane (of Elschnig) covering the optic disc (1b). An accumulation of astrocytes forms a central meniscus of Kuhnt in the centre of the disc (2). The anterior or so-called 'choroidal part' of the lamina cribrosa (6) is separated from the choroid by a spur of collagenous tissue (3). The 'border tissue of Jacoby' (4), which is largely astroglia, frequently extends beyond the choroid (5) to separate much of the retina from the 'retinal part' of the optic nerve head. The posterior part of the lamina cribrosa (7) contains collagenous tissue derived from the optic nerve septa and fenestrated sheets of collagen fibres continuous with those of the sclera. (By permission from Anderson DR, Hoyt W 1969 Ultrastructure of intraorbital portion of human and monkey optic nerve. Arch Ophthalmol 82: 506-530.)
Microscopically, the cornea consists of five layers arranged anteroposteriorly as
follows: corneal epithelium, which is continuous with the conjunctival epithelium; anterior limiting lamina; substantia propria; posterior limiting lamina; endothelium.
CORNEAL EPITHELIUM The corneal epithelium covers the anterior surface of the cornea and generally has five layers of cells. The deepest are columnar with flat bases and rounded apices, and large rounded or oval nuclei. Cells in the second layer are polyhedral and resemble those in the epidermal stratum spinosum. In the more superficial layers the cells become progressively flatter. However, unlike the cells of the epidermis, they contain flat nuclei, are not normally keratinized, and present a smooth, optically perfect surface. At the corneoscleral junction (limbus) the corneal epithelium merges with the limbal conjunctival epithelium, which thickens (up to 12 cells) and soon loses the regular surface of the cornea. It is of clinical significance that the cornea does not appear to possess epithelial stem cells. Cell replacement depends on the centripetal migration (from the edges of the cornea) of cells which are the progeny of mitotic limbal stem cells.
ANTERIOR LIMITING LAMINA The anterior limiting lamina lies behind the corneal epithelium. It contains a dense mass of collagen fibrils set in a matrix similar to that of the substantia propria. The lamina is 12µm thick and is readily distinguishable from the substantia propria because it contains no fibroblasts. It appears amorphous by light microscopy.
SUBSTANTIA PROPRIA OR STROMA page 703 page 704
Figure 42.3 Meridional section through the iridocorneal angle. The conjunctiva (left) was damaged in preparation. (Provided by the late Gordon L Ruskell, Department of Optometry and Visual Science, The City University, London.)
Figure 42.4 Normal human corneal epithelial cells. Parts of the outlines of several cells are visible; in the upper and lower cells, microvilli predominate but some microplicae are seen. The cells at the right of the field display predominantly microplicae, with only a few microvilli. (From Pfister RR, Burstein NL 1977 The normal and abnormal human corneal epithelial surface. Invest Ophthalmol Vis Sci 16: 614622. By permission from the Association for Research in Vision and Ophthalmology.)
The substantia propria or stroma forms the bulk of the cornea. It is a compact and transparent layer, composed of 200-250 sequential lamellae, each made up of fine parallel collagen fibrils mainly of type I collagen. Flat dendritic interconnecting fibroblasts form a coarse mesh between the lamellae. Alternate lamellae are typically orientated at large angles to each other (Fig. 42.5). Each lamella is c.2µm thick and of variable breadth (10-250µm, or, rarely, more). All fibrils in a given lamella have similar diameters, and are smaller in anterior lamellae than more posteriorly (a range of 21-65 nm). The dimensions of the fibrils are much smaller than the wavelength of light: this feature, and the regularity of their spacing, are the principal factors which determine stromal transparency (Hogan et al 1971).
POSTERIOR LIMITING LAMINA
Figure 42.5 Substantia propria of the human cornea; note the geometric precision of the alternation in direction of adjacent layers of collagen fibres. (By kind permission from John Marshall, Institute of Ophthalmology, London.)
The posterior limiting lamina covers the substantia propria posteriorly. It is thin and apparently homogeneous, and is regarded as the basement membrane of the endothelium. It is known to grow throughout life. It is 5µm thick at birth, and may increase to 17µm by the ninth decade. At the limbus of the cornea it disperses into the fibres of trabecular tissue which adjoin the inner wall of the scleral venous sinus. Aqueous humour drains from the eye through the iridocorneal angle (the 'filtration angle' in clinical terminology; Figs 42.3, 42.6). The trabecular spaces are interconnected, and it is believed that there is no impediment to flow from the anterior chamber to the inner wall of the sinus. The wall of the sinus is constructed of a continuous single thin endothelial layer. Passage of aqueous humour to the sinus probably occurs via giant pinocytotic vacuoles, which form on the inner face of the endothelium and discharge into the sinus at the outer face, and through intercellular clefts. Aqueous humour then passes through a plexus of fine intrascleral vessels which connect the sinus with anterior ciliary veins. Normally the sinus does not contain blood: pressure gradients prevent the reflux of blood even though the channels between the sinus and veins have no valves. In venous congestion, blood may enter the sinus, however, the continuous endothelial outer wall of the trabeculae prevents further reflux.
ENDOTHELIUM The corneal endothelium covers the posterior surface of the cornea and lines the spaces of the iridocorneal angle. It is a layer of polygonal, flattened cells: there are prominent interdigitations between adjacent cells.
VASCULAR SUPPLY The cornea contains neither blood nor lymphatic vessels: the capillaries of the conjunctiva and sclera end in loops near its periphery. page 704
page 705
Figure 42.6 The iridocorneal angle and adjoining structures, showing the proximity of the scleral venous sinus (a) to the pectinate ligaments (f). The trabecular meshwork of the latter is partly uveal, being continuous with the iris (h) and ciliary body (CB) and muscle (i). Anterior to the scleral spur (d), scleral trabecular tissue (c) is even closer to the scleral venous sinus. Aqueous fluid percolates through this trabecular region, reaching the lumen of the sinus through small apertures (b). The pectinate ligament diminishes as it approaches the corneal limbus (e) and in this junctional zone the posterior limiting membrane (of Descemet) also terminates (g). The endothelium of the anterior chamber (posterior corneal epithelium) is continuous with the endothelium of the trabeculae (j) at the limbus. (By permission from Hogan MJ, Alvarado JA, Weddell JE 1971 Histology of the Human Eye. Philadelphia: WB Saunders.)
INNERVATION The cornea is well innervated by numerous branches of the ophthalmic nerve which either form an annular plexus around the periphery of the cornea, or pass directly from the sclera and enter the substantia propria radially as 70-80 small groups of fibres. Upon entering the cornea, the few myelinated nerves lose their myelin sheaths. The nerves ramify throughout the corneal matrix in a delicate reticulum, and their terminal filaments form an intricate subepithelial plexus. Fine varicose axons from this plexus cross the anterior limiting membrane and form an intraepithelial plexus. There are no specialized end organs, the epithelial nerve fibres are devoid of Schwann cells, and they do not arborize.
© 2008 Elsevier
OCULAR VASCULAR TUNIC The vascular tunic, or uveal tract, consists of the choroid, ciliary body and iris, which collectively form a continuous structure. The choroid covers the internal scleral surface, and extends forwards to the ora serrata. The ciliary body continues forward from the choroid to the circumference of the iris, which is a circular diaphragm behind the cornea and in front of the lens. It presents an almost central aperture, the pupil.
Choroid (Figs 42.1, 42.7) The choroid is a thin, highly vascular, dark brown tissue which lines almost fivesixths of the eye posteriorly. It is pierced behind by the optic nerve where it is firmly adherent to the sclera. Elsewhere its external surface is loosely connected to the sclera by the suprachoroid layer (lamina fusca). Internally it is firmly attached to the pigmented layer of the retina, and at the optic disc it is continuous with the pia-arachnoid tissues around the optic nerve. Structurally, the choroid consists largely of a dense capillary plexus. The blood flow through the choroid is high, a feature probably associated with an intraocular pressure of 15-20 mmHg, which means that a venous pressure above 20 mmHg is required to maintain circulation. The cooling effect of the choroidal circulation on the retina may be important. Externally, the suprachoroid layer, c.30µm thick, is composed of delicate non-vascular lamellae, each one a network of fine collagen and elastic fibres, and stellate cells which contain dark brown granules. Ganglionic neurones and neural plexuses are present in the connective tissue. The choroid proper lies internal to the suprachoroid layer, which is partly scleral tissue, and consists of a number of layers. Although descriptions of these vary, the following are generally recognized: an external vascular layer of small arteries and veins, loose connective tissue and scattered pigment cells; a capillary layer; a thin, apparently structureless, basal layer. The vascular layer is sometimes subdivided on the basis of blood vessel calibre, which naturally decreases towards the capillary layer. It also contains the terminal branches of short posterior ciliary arteries (Fig. 42.1) which extend meridionally from their entry through the sclera near the optic disc. The veins are larger, and converge spirally onto four, or very occasionally more, principal vorticose veins. These pierce the sclera behind the equator to reach tributaries of the ophthalmic veins. The capillary layer is separated from the retina only by the thin basal layer of the choroid, and provides nutrition to the retina. It consists of a close meshwork of large vessels: the meshes widen slightly towards the ciliary body. The basal layer appears as a glassy, homogeneous layer (lamina vitrea), only 2-4µm thick, under the light microscope. It consists largely of an elastic fibre mesh. Its function is uncertain, but is thought to be related to the passage of fluid and solutes from the choroidal capillaries to the retina. It is derived from both the retina and the choroid. In advancing years, lipid may be deposited in this membrane, which impairs the exchange of gases, nutrients and metabolites between the choroidal blood and the outer layers of the retina, and causes degenerative disease in the photoreceptor layer of the neural retina. The choroidal pigment cells limit the passage of light through the sclera to the retina. More importantly, they may absorb light traversing the retina beyond the receptors, so preventing internal reflection within the vitreous body. The vessels of the choroid have a rich autonomic vasomotor supply, however, the sensory supply is at most very poor.
Ciliary body (Figs 42.8, 42.9) The ciliary body is directly continuous with the choroid behind, and with the iris in front. While all of these regions of the uveal tract share certain various features, they also exhibit regional differences related to variations in their function. The ciliary body is concerned with the suspension of the lens and with accommodation, and it contains an accumulation of muscle fibres which cause it to bulge internally. It is also the major source of aqueous fluid for the anterior segment of the eye, which its anterior aspect faces. Posteriorly it is contiguous with the vitreous humour, and it probably secretes some of the glycosaminoglycans (GAGs) of the vitreous body. The anterior and the long ciliary arteries meet in the ciliary body (Fig. 42.7), and so it is a highly vascular region. The major nerves to all the anterior tissues of the eyeball pass through the ciliary body. Externally, the ciliary body may be represented by a line which extends from c.1.5 mm posterior to the limbus of the cornea (corresponding also to the scleral spur) to a line c.7.5-8 mm posterior to this on the temporal side, and 6.5-7 mm on the nasal side. The ciliary body is thus slightly eccentric. It projects posteriorly from the scleral spur, which is its attachment, with a meridional width varying from 5.5 to 6.5 mm. Internally, it exhibits a posteriorly crenated or scalloped periphery, the ora serrata, where it is continuous with the choroid and retina. Anteriorly, it is confluent with the periphery of the iris, and externally bounds the iridocorneal angle of the anterior chamber. page 705 page 706
Figure 42.7 The vascular arrangements of the uveal tract. The long posterior ciliary arteries, one of which is visible (A), branch at the ora serrata (b) and feed the capillaries of the anterior part of the choroid. Short posterior ciliary arteries (C) divide rapidly to form the posterior part of the choriocapillaris. Anterior ciliary arteries (D) send recurrent branches to the choriocapillaris (e) and anterior rami to the major arterial circle (f). Branches from the circle extend into the iris (g) and to the limbus. Branches of the short posterior ciliary arteries (C) form an anastomotic circle (h) (of Zinn) round the optic disc, and twigs (i) from this join an arterial network on the optic nerve. The vorticose veins (J) are formed by the junctions (k) of suprachoroidal tributaries (l). Smaller tributaries are also shown (m, n). The veins draining the scleral venous sinus (o) join anterior ciliary veins and vorticose tributaries. (By permission from Hogan MJ, Alvarado JA, Weddell JE 1971 Histology of the Human Eye. Philadelphia: WB Saunders.)
Figure 42.8 The posterior aspect of the anterior half of the eyeball showing the termination of the neural retina at the ora serrata and the ciliary body. The definition of the serrations is less clear temporally (right). Note the lighter crests of the ciliary processes and the greater width of the ciliary body temporally. The lens has retained sufficient transparency to reveal the iris border and its crenated perimeter is due to tension imposed by the attached suspensory ligaments (unseen). (Provided by the late Gordon L Ruskell, Department of Optometry and Visual Science, The City University, London.)
Melanin in the deeper layer of its epithelium renders the ciliary body brown. It has an anterior plicated part, the corona ciliaris (pars plicata), which surrounds the base of the iris. Posterior to this is a smooth, annular orbiculus ciliaris (pars plana, ciliary ring). The orbiculus forms more than half of the meridional width of the ciliary body, and is 3.5-4 mm across. Its peripheral rim is the ora serrata, at which the optical or sensory part of the retina is suddenly reduced to two layers of epithelial cells, which extend over the whole ciliary body as the pars ciliaris retinae, and beyond this extend to the posterior surface of the iris. The corona ciliaris, a smaller annular region within the orbiculus, is ridged meridionally by 7080 ciliary processes radiating from the base of the iris. A minor ridge, or ciliary plica, lies in the valley between most of the processes. The crests of the processes are less pigmented, and this gives them the appearance of white (or light) striae, from which the name ciliary is derived. Fibres of the zonule (the suspensory ligament of the lens) extend into the valleys. They pass beyond the ciliary processes to fuse with the basal lamina of the superficial epithelial layer of the orbiculus ciliaris. Their sites of attachment are marked by striae which pass back from the valleys of the corona, across the orbiculus, almost as far as the apices of the dentate processes of the ora serrata (Fig. 42.9).
CILIARY EPITHELIUM page 706 page 707
Figure 42.9 The ciliary region seen from the ocular interior. Above is the periphery of the lens, attached by the fibres of the zonule (suspensory ligament) to the processes of the corona ciliaris (pars plicata) of the ciliary body (a). The orbiculus ciliaris or pars plana ciliaris (b) has a scalloped boundary, the ora serrata (c), which separates it from the retina (d). Flanking the 'bays' (e) of this are the dentate processes (f), with which linear ridges or striae (g) are continuous. These striae extend forwards between the main ciliary processes, providing an attachment for the longer zonular fibres. The posterior aspect of the iris shows radial (h) and circumferential (i) sulci. (By permission from Hogan MJ, Alvarado JA, Weddell JE 1971 Histology of the Human Eye. Philadelphia: WB Saunders.)
Figure 42.10 Transverse section through the choroid. (Provided by the late Gordon L Ruskell, Department of Optometry and Visual Science, The City University, London.)
The ciliary epithelium is bilaminar, and consists of two layers of simple epithelium which are derived embryonically from the two layers of the optic cup. The superficial layer consists of columnar cells over the orbiculus, and cuboidal cells over the ciliary processes: it becomes irregular and more flattened between the processes. These cells contain little or no pigment and are the sole anterior continuation of the neural retina. The pigment epithelium of the retina is continuous with the deeper layer of the ciliary epithelium, where the cells are approximately cuboidal and are loaded with pigment. The two layers are normally firmly united, but fluid may separate them pathologically. The pigment layer is united to the stroma of the ciliary body by its basal lamina, which continues back into the basal lamina of the choroid (Fig. 42.10). A basal lamina covers the free surface of the bilayer and is continuous with the internal limiting membrane of the retina. This arrangement is a consequence of the invagination of the optic cup during development.
CILIARY STROMA page 707 page 708
Figure 42.11 The ciliary muscle and its components. The meridional or longitudinal (1), radial or oblique (2), and circular or sphincteric (3) layers of muscle fibres are displayed by successive removal towards the ocular interior. The cornea and sclera have been removed, leaving the pectinate ligament (a), the scleral venous sinus (b), collecting venules (c) and scleral spur (d). The meridional fibres often display acutely angled junctions (e) and terminate in epichoroidal stars (f). The radial fibres meet at obtuse angles (g) and similar junctions, at even wider angles (h), occur in the circular stratum of the ciliaris. (By permission from Hogan MJ, Alvarado JA, Weddell JE 1971 Histology of the Human Eye. Philadelphia: WB Saunders.)
The ciliary stroma is composed largely of loose bundles of collagen, which form a considerable mass between the ciliary muscle and over lying processes, and extend into both of them. This inner layer of connective tissue contains numerous larger branches of the ciliary vessels. A dense reticulum of large capillaries is concentrated in the ciliary processes. Anteriorly, near the periphery of the iris, is the major arterial circle (Fig. 42.7), which is formed chiefly by long posterior ciliary branches of the ophthalmic artery. These enter the eye some distance behind the ocular equator and pass between the choroid and sclera to the ciliary body. Ciliary veins, also draining the iris, pass posteriorly to join the vorticose veins of the choroid.
CILIARY MUSCLE (Fig. 42.11) The ciliary muscle is a small annular mass of smooth muscle. Descriptions of the muscle generally recognize three main parts: meridional, radial or oblique, and circular or sphincteric. Most, perhaps almost all, ciliary muscle fibres are attached to the scleral spur. The outermost fibres are meridional or longitudinal, and pass posteriorly into the stroma of the choroid. The innermost fibres swerve acutely from the spur to run circumferentially as a sphincteric element near the periphery of the lens. Obliquely interconnecting radial fibres run between these two muscular strata, frequently forming an interweaving lattice. Myelinated and non-myelinated nerve fibres abound in the ciliary muscle and ciliary body. The former are postganglionic parasympathetic fibres from the ciliary ganglion which stimulate the ciliary muscle to contract. Sympathetic fibres are sparse: they have a very limited capacity to relax the muscle. On contraction the muscle is displaced radially towards the optic axis. All parts of the muscle act in concert, and tension on the zonular ligaments is relaxed which frees the lens to assume its accommodated shape.
Iris (Fig. 42.12) The iris is an adjustable diaphragm around a central aperture (slightly medial to true centre), the pupil, which controls the amount of light entering the eye (Fig. 42.1). Pupillary diameter varies from 1 to at least 8 mm, and has an even wider range under the influence of drugs. This gives an aperture range in excess of f20f2.5, and a ratio of 32:1 for the amount of light permitted to enter the eye. While this is not enough to save the retina from the effects of intense illumination, it moderates the great range of luminosity encountered in ordinary use, and preserves useful vision under highly variable conditions. (The pupillary diameters noted above and the average iridial diameter of c.12 mm are estimated through the cornea, which introduces a magnification factor of c.12%.) Pupillary constriction and dilatation are self-explanatory terms, for which miosis and mydriasis are used clinically, though they are more properly reserved for the extreme limits of contraction and dilatation. Loewenstein and Loewenfeld (1970) should be consulted for further information. Though the iris is named after the rainbow, its range of colour extends only from light blue to very dark brown. The colour often varies between the two eyes and even within the same iris (Fig. 42.12). It is a product of the combined effect of the iridial connective tissue and the pigment cells which selectively absorb and reflect different frequencies of light energy. Pigment is necessary to confine light transmission to the pupil and central lens, where optical aberrations are least. The concentration of melanocytes is the main factor which determines the hue of the iris. The distribution of pigment is often irregular, and this produces a flecked appearance. When pigment is largely absent, other than in the epithelial layers, which is the condition at birth, the colour is light blue. The iris is not a flat diaphragm because the lens causes it to bulge a little. This means that it is more accurately described as a shallow cone, truncated by the pupillary aperture. It is sited between the cornea and lens and immersed in aqueous fluid. It therefore partially divides the anterior segment into an anterior chamber, enclosed by the cornea and iris (which meet at the iridocorneal angle), and a confusingly termed posterior chamber, which lies between the iris and the lens. page 708 page 709
Figure 42.12 Composite view of the surfaces and internal strata of the iris. In a clockwise direction from above, the pupillary (A) and ciliary (B) zones are shown in successive segments. The first (brown iris) shows the anterior border layer and the openings of crypts (c). In the second segment (blue iris), the layer is much less prominent and the trabeculae of the stroma are more visible. The third segment shows the iridial vessels, including the major arterial circle (e) and the incomplete minor arterial circle (f). The fourth segment shows the muscle stratum, including the sphincter (g) and dilator (h) of the pupil. The everted 'pupillary ruff' of the epithelium on the posterior aspect of the iris (d) appears in all segments. The final segment, folded over for pictorial purposes, depicts this aspect of the iris, showing radial folds (i and j) and the adjoining ciliary processes (k). (By permission from Hogan MJ, Alvarado JA, Weddell JE 1971 Histology of the Human Eye. Philadelphia: WB Saunders.)
MICROSTRUCTURE The stroma of the iris is formed of fibroblasts, melanocytes and loose collagenous
matrix. The intercellular spaces appear to communicate freely with the anterior chamber, and interchange of fluid between the two may assist the considerable changes in thickness which occur during contraction and relaxation of the iris. There is no elastic tissue. The anterior surface of the iris does not possess a distinct epithelium, but is a modified superficial layer of the general stroma of the iris, formed mainly by an increased number of fibroblasts and melanocytes. At the periphery it blends with the pectinate ligament and the trabecular connective tissue of the iridocorneal angle (Hogan et al 1971), and at the pupillary rim it meets the epithelium of the posterior surface of the iris. This stroma contains the regional vessels and nerves. An aggregation of smooth muscle cells near the pupillary rim forms an annular contractile sphincter pupillae. The epithelial surface covering the iris posteriorly is a continuation of the bilaminar epithelium of the ciliary body formed from the two layers of the optic cup. The pupil, through which this epithelium curves for a short distance on to the anterior surface as the pigment ruff or 'border', therefore corresponds to the opening of the optic cup. The deeper of the two epithelial layers is confusingly termed the anterior epithelium, although it lies posterior to the stroma. Its cells are pigmented, as are those of the same layer in the ciliary epithelium. They give rise to the dilator pupillae, which like the sphincter has an unusual embryological origin, from the neural ectoderm of the optic cup which interacts with the local neural crest. Superficial (posterior) to this layer is a stratum of heavily pigmented cells, the socalled posterior epithelium, which is continuous with the internal non-pigmented, retinal, layer of the ciliary epithelium. The surface bears numerous delicate radial ridges at its free surface which facilitate the movement of aqueous humour from the posterior to the anterior chamber.
MUSCLES OF THE IRIS The muscles of the iris are sphincter and dilator pupillae.
SPHINCTER PUPILLAE Sphincter pupillae is a flat annulus of smooth muscle c.0.75 mm wide and 0.15 mm thick. Its densely packed fusiform muscle cells are often arranged in small bundles, as in the ciliary muscle, and pass circumferentially around the pupil (Fig. 42.12). Collagenous connective tissue lies in front of and behind the muscle fibres and is very dense posteriorly, where it binds the sphincter to the pupillary end of the dilator muscle. It is attached to the epithelial layer at the pupil margin. Small axons, mostly non-myelinated, ramify in the connective tissue between bundles.
DILATOR PUPILLAE Dilator pupillae is a thin layer which lies immediately anterior to the epithelium of the posterior surface of the iris. Its 'fibres' are the muscular processes of the anterior layer of this epithelium, whose cells are therefore myoepithelial. Myofilaments are present in both parts of these cells, but are more abundant in the fusiform basal muscular processes, which are c.4µm thick, 7µm wide and 60µm in length. They form a layer some 3-5 elements thick through most of the iris, from its periphery to the outer perimeter of the sphincter, which it slightly overlaps. Here the dilator thins out, and sends spurs to blend with the sphincter. Unlike the apical parts of the myoepithelial cells, these have a basal lamina and are joined by gap junctions like those between the sphincteric muscle cells. Nonmyelinated nerve axons pass between, and terminate on, their muscular processes.
VASCULAR SUPPLY OF THE IRIS (Figs 42.7, 42.12) The iris receives its blood supply from the long posterior and anterior ciliary arteries (Fig. 42.7). Both long ciliary arteries, on reaching the attached margin of the iris, divide into an upper and a lower branch. The branches anastomose with the corresponding contralateral arteries, and the anterior ciliary arteries, to form a vascular circle, the circulus arteriosus major. Vessels converge from this circle to the free margin of the iris, where they anastomose to form an incomplete circulus arteriosus minor: there is a view that these vessels are venous. The smaller arteries and veins are very similar in their structure and also share some peculiarities. They are often slightly helical, which may allow them to adapt to changes in iridial shape as the pupil varies in size. All of the vessels, including the capillaries, have a non-fenestrated endothelium and a prominent, often thick, basal lamina. There is no elastic lamina in the arteries or veins, and there are few smooth muscle cells, especially in the veins. Connective tissue in the tunica media is loose, whereas the adventitia is remarkably dense and collagenous, so that it appears to form almost a separate tube. The loose stratum of the media has been regarded as a lymph space, but this is improbable; it is c.7µm in width, and contains a matrix which is probably derived from the endothelial basal lamina (Hogan et al 1971). page 709 page 710
INNERVATION OF THE IRIS The iris is innervated largely by branches of the long ciliary rami of the nasociliary nerve and of the short ciliary rami of the ciliary ganglion (p. 700). The latter provide parasympathetic postganglionic myelinated axons which innervate the sphincter pupillae. They lose their myelin well before entering the muscle. The dilator is supplied with non-myelinated postganglionic fibres from the superior sympathetic ganglion; their routes are less well established. The internal carotid sympathetic plexus is said to send a branch via the ciliary ganglion, which reaches the eye in the short ciliary nerves: other fibres may travel in the long ciliary nerves. Both the sphincter and the dilator may have a double autonomic supply. An additional small fraction of dilator and sphincter muscle nerve endings have been identified as parasympathetic and sympathetic respectively in experimental animal studies, including those on primates. Although ganglion cells have been noted in the iris, almost all nerve fibres are probably postganglionic. They form a plexus around the periphery of the iris, from which small nerves and fibres extend to the two muscles, to vessels, and to the anterior border layer. Some fibres may be afferent and some are vasomotor: it is difficult to identify afferent nerve endings in the iris and ciliary body or to distinguish them from efferent autonomic endings.
PUPILLARY MEMBRANE In the fetus, the pupil is closed by a thin, vascular pupillary membrane. Its vessels are derived partly from those of the iridial margin, and partly from those of the lens capsule. They end in loops near the centre of the membrane, which is avascular. At about the sixth month of gestation, absorption of the membrane begins, from the centre towards the periphery, until at birth when only scattered fragments remain. Exceptionally the pupillary membrane may persist and interfere with vision.
© 2008 Elsevier
RETINA (Figs 42.13, 42.14, 42.15) The retina is the sensory neural layer of the eyeball. It is a most complex structure and should be considered as a special area of the brain, from which it is derived by outgrowth from the diencephalon (Chapter 14). It is dedicated to the detection and early analysis of visual information and is an integrated part of the much larger apparatus of visual analysis present in the thalamus, cortex and other areas of the central nervous system (Section 2).
page 710 page 711
Figure 42.13 Ophthalmoscopic photographs of the right human retina. A, Note dichotomous branching of vessels, arteries being brighter red and showing a more pronounced 'reflex' to light, as a pale stria along their length. The veins are also larger in calibre; more of them cross arteries superficially than is usual. The optic disc, around the entry of the vessels, is a light pink, with a surrounding zone of heavier pigmentation. Compare with Fig. 42.14A from the same Caucasian adult. B, Appearances in a heavily pigmented individual (African-origin adult), with a paler optic disc than in A. Note accentuation of the edge of the disc by retinal and choroidal pigmentation. The arteries cross the veins superficially in this retina. C, Normal macula of a young Caucasian subject. The vessels radiate from the centrally placed fovea. The macular branches of the central retinal artery are approaching from the right. The macula is largely free of vessels of macroscopic size, but the capillaries here form a particularly close network, except at the fovea. D, The region of the optic disc in an eye with poorly developed pigmentation. Three cilioretinal arteries are curving round the edge of the disc (two on the left, one on the right). Between the two cilioretinal arteries a single macular artery is apparent. Due to the depressed pigmentation choroidal vessels are also visible, especially veins; and on the left of the photograph two large vorticose venous tributaries can be seen.
The retina lies between the choroid externally and the vitreous body internally. It is thin, being thickest (0.56 mm) near the optic disc, diminishing to 0.1 mm anterior to the equator, and continuing at this thickness to the ora serrata. It also thins locally at the fovea of the macula. The retina is continuous with the optic nerve at the optic disc. Anteriorly, at the ora serrata (p. 705), a thin, non-neural prolongation of the retina extends forwards over the ciliary processes and iris as the ciliary and iridial parts of the retina respectively: they consist of pigmented and columnar epithelial layers only. The optic part of the retina extends from the optic disc to the ora serrata. It is soft, translucent and purple in the fresh, unbleached state, because of the presence of rhodopsin (visual purple), but soon becomes opaque and bleached when exposed to light. Near the centre of the retina there is a region 5-6 mm in diameter, which contains the macula lutea (Fig. 42.14C,D), an elliptical yellowish area, c.2 mm horizontally and 1 mm vertically. Its colour is due to the presence of xanthophyll derivatives. The macula lutea contains a central depression, the fovea centralis or foveola, with a diameter of c.0.4 mm, where visual resolution is highest (Fig. 42.15). Here, all elements except pigment epithelium cone photoreceptors are displaced laterally. The minute size of the foveola is the reason why the visual axes must be directed with great accuracy in order to achieve the most discriminative vision.
page 711 page 712
Figure 42.14 Fluorescence angiograms of the retina. These are produced by photography with a fundus camera at known periods of time following introduction of fluorescein into the circulation. A, Angiogram of the same retina as that appearing in Fig. 42.13A taken in 'mid-venous' phase. The arteries display an even fluorescence, but the veins appear striped, due to laminar flow. This appearance is the reverse of and not to be compared with the arterial 'reflex' which is seen in Fig. 42.13A. The background mottling is due to fluorescence from the choroidal vessels. B, Angiogram of the left optic disc, showing the major arteries and veins and also their smaller branches. Note particularly the radial pattern in the retinal capillaries. The laminar flow in the veins is less obvious than in Fig. 42.13A. C, Angiogram showing the macular region of a right eye. The main macular vessels are approaching from the right. The subject was an elderly person with considerable macular pigmentation, which masks fluorescence from the choroidal circulation. Compare with: D, Angiogram of the macula of a young subject (left eye) showing the macular capillaries in detail. Note the central avascular fovea. Compare with C.
Figure 42.15 Section through the fovea centralis. (By permission from Young B, Heath JW 2000 Wheater's Functional Histology. Edinburgh: Churchill Livingstone.)
About 3 mm medial (nasal) and 1 mm superior to the foveola, the optic nerve becomes continuous with the retina at the optic disc ('blind spot'). It is c.1.5 mm in diameter. The name 'optic papilla', often applied to the disc, is a misnomer, since almost all of a normal disc is level with the retina. Centrally it contains a shallow depression, where it is pierced by the central retinal vessels (Figs 42.1, 42.13, 42.14A,B). The disc is devoid of photoreceptors and is therefore insensitive to light. By ophthalmoscopy it is normally pink but it is much paler than the retina, and may be grey or almost white. In optic atrophy the capillary vessels disappear and the disc is then white.
MICROSTRUCTURE The retina is derived from the two layers of the invaginated optic vesicle (p. 245); the outer layer becomes the layer of pigment cells, the inner layer develops into a complex multilaminar structure of sensory and neural cells. Anteriorly, sensory and neural cells are absent from the retina as it approaches the ora serrata and merges into the ciliary body and iris (p. 705). The neural retina contains a variety of cell types. They include the photoreceptors (rod and cone cells), their first order neurones (bipolar cells) and the somata and axons of the second order neurones (ganglion cells); and two major classes of interneurones, the horizontal and amacrine cells. The retina also contains neuroglial elements and a rich vascular system, chiefly of capillaries. It is backed by specialized pigment epithelial cells.
LAYERS OF THE RETINA (Fig. 42.16) The retina is organized into layers or zones where distinctive components of its cells are clustered together or in register to form continuous strata. These layers extend uninterrupted throughout the photoreceptive retina except at the exit point of the optic nerve fibres at the optic disc, although certain layers are much reduced at the foveola where the photoreceptive elements predominate. The names given to the different layers reflect in part the components present within them, and also their position in the thickness of the retina. Conventionally, those structures furthest from the vitreous (i.e. towards the choroid) are designated as outer or external, and those towards the vitreous are inner or internal. Customarily, ten retinal layers are distinguished (Fig. 42.17), beginning at the
choroidal edge and passing towards the vitreous. These are: retinal pigment epithelium; layer of rods and cones (outer segments and inner segments); external limiting membrane; outer nuclear layer; outer plexiform layer (OPL); inner nuclear layer (INL); inner plexiform layer (IPL); ganglion cell layer; nerve fibre layer; internal limiting membrane. Some of these are subdivisible into substrata, and an innermost plexiform layer between layers 8 and 9 has also been demonstrated.
Figure 42.16 The retina of a 60-year-old male. The section was made close to the optic nerve head explaining the thickened nerve fibre layer. (Provided by the late Gordon L Ruskell, Department of Optometry and Visual Science, The City University, London.)
Rod and cone cells reach radially inwards from the rod and cone lamina through the outer nuclear layer, where they have their nuclei, to the outer plexiform layer in which they synapse with bipolar and horizontal cells. Bipolar cells possess dendrites in the outer plexiform layer, cell bodies and nuclei in the inner nuclear layer, and axons in the inner plexiform layer where they synapse with ganglion cell and amacrine cell dendrites. Horizontal cells have their dendrites and axons in the outer plexiform layer and their nuclei in the inner nuclear layer, while ganglion cells have their dendrites in the inner plexiform layer, their cell bodies in the ganglion cell layer, and their axons in the layer of nerve fibres (and within the optic nerve). Amacrine cell dendrites are mainly in the inner plexiform layer, although some (interplexiform cells) extend into the outer plexiform layer; amacrine cell dendrites are either situated in the inner nuclear layer or in the outer part of the ganglionic layer (displaced amacrines). Pigment cells lie behind the retina, and several types of retinal glial cell are distributed in distinctive locations
among its different layers. The composition of the different retinal layers is as follows: Layer 1: Pigment epithelium This is a simple low cuboidal epithelium which forms the back of the retina, and, therefore forms the boundary with the choroid, from which it is separated by a thick composite basal lamina. page 712 page 713
Figure 42.17 The layered arrangement of neuronal cell bodies in the retina and the interconnections of their processes in the intervening plexiform layers. Also shown are the two principal types of neuroglial cell in the retina; microglia are also present but not shown.
Layer 2: Rod and cone cell processes This contains the photoreceptive outer segments and the outer part of the inner segments of rod and cone cells. Layer 3: External limiting membrane This layer appears as a distinct line by light microscopy. It consists of a zone of intercellular junctions of the zonula adherens type (p. 7) between the processes of radial glial cells and photoreceptor processes. Layer 4: Outer nuclear layer This consists of several tiers of rod and cone cell bodies and their nuclei, the cone nuclei lying outermost. Mingled with these are the outer and inner fibres from the same cell bodies, directed outward to the bases of inner segments, and inwards towards the outer plexiform layer. Layer 5: Outer plexiform layer This is a region of complex synaptic arrangements between the processes of the cells whose cell bodies lie in the adjacent layers. The outer plexiform layer contains the synaptic processes of rod and cone cells, bipolar cells, horizontal cells, and some interplexiform cells (which in this account are grouped with the amacrines). Layer 6: Inner nuclear layer
This is composed of three nuclear strata. Horizontal cell nuclei form the outermost zone, then in sequence inwards, the nuclei and cell bodies of bipolar cells, radial glial cells, and the outer set of amacrine cells, including the interplexiform cells whose dendrites cross this layer. Layer 7: Inner plexiform layer This is divisible into three layers depending on the types of contact occurring. The outer or 'OFF' layer contains synapses between 'OFF' bipolar cells, ganglion cells and some amacrines; a middle or 'ON' layer contains synapses between the axons of 'ON' bipolars and the dendrites of ganglion cells and displaced amacrines; and an inner 'rod' layer contains synapses between rod bipolars and displaced amacrines. (Refer to Wässle & Boycott 1991 for an explanation of the 'OFF' and 'ON' cell designations.) Layer 8: Ganglion cell layer This layer contains the nuclei of the displaced amacrine cells. Its inner regions consist of the cell bodies, nuclei and initial segments of retinal ganglion cells of various classes. Layer 9: Nerve fibre layer This contains the unmyelinated axons of retinal ganglion cells. It forms a zone of variable thickness over the inner retinal surface, and is the only component of the retina at the point where the fibres pass into the nerve at the optic disc. The inner aspect of this layer contains the nuclei and processes of astrocytes which, together with radial glial cells, ensheath the nerve fibres. Between the nerve fibre layer and the ganglion cells there is another narrow innermost plexiform layer where neuronal processes make synaptic contact with the axon hillocks and initial segments of ganglion cells. Layer 10: Internal limiting membrane This is a glial boundary between the retina and the vitreous body. It is formed by the end feet of radial glial cells and astrocytes, and is separated from the vitreous body by a basal lamina.
CELLS OF THE RETINA Retinal pigment epithelial cells The retinal pigment epithelial cells are low cuboidal cells which form a single continuous layer extending from the periphery of the optic disc to the ora serrata, and continue from there into the ciliary epithelium (p. 705). They are flat in radial section, hexagonal or pentagonal in surface view and number about 4-6 million in the human retina. Their cytoplasm contains numerous melanin granules (Fig. 42.18). Apically (towards the rods and cones), the cells bear long (5-7µm) microvilli which contact or project between the outer ends of rod and cone processes. The tips of rod outer segments are deeply inserted into invaginations in the apical membrane. The attachments are unsupported by junctional complexes and are broken in the clinical condition of retinal detachment arising from trauma or disease processes. page 713 page 714
Figure 42.18 Unstained retinal pigment epithelium from a 40-year-old individual, seen
in surface view. (By kind permission from John Marshall, Institute of Ophthalmology, London.)
Pigment epithelial cells play a major role in the turnover of rod and cone photoreceptive components. Their cytoplasm contains the phagocytosed ends of rods and cones undergoing lysosomal destruction. The final products of this process are lipofuscin granules, which accumulate in these cells and add to their granular appearance. The failure of some part of this process may cause progressive loss of retinal function and eventual blindness, e.g. when enzyme deficiencies cause the build-up of shed, but undegraded, photoreceptor components within the retina. The epithelium also acts as an anti-reflection device and prevents the light bouncing back into the photoreceptive layer with consequent loss of image sharpness. This process is complex, because the energy absorbed could be dissipated as heat or generate free radicals, which are both potentially damaging products. Indeed, very intense light may damage the pigment cells and cause epithelial breakdown. The zone of tight junctions between the pigment cells allows the epithelium to function as an important blood-retinal barrier between the retina and the vascular system of the choroid. These guard the special ionic environment of the retina, and inhibit the entry of leukocytes into this immunologically sequestered compartment of the eye. Cone and rod cells (Fig. 42.19) The cone and rod cells are the retinal photoreceptor cells. They are long, radially orientated structures with a cylindrical photoreceptive portion at the end nearest the pigment epithelium and synaptic contacts at the other end, within the outer plexiform layer. Both types of cell have a similar organization, although details differ. From the external (choroidal) end inwards, the cells consist of outer and inner segments, a cell body containing the nucleus, and either a cone pedicle or a rod spherule (depending on cell type); this is an area of synaptic contact with adjacent bipolar and horizontal cells and with other cone or rod cells. The outer and inner segments together form a cone process or a rod process (it should be noted that the terms cone and rod are also often loosely applied to the whole cell); the cone process is wider, but tapers (hence the name), whereas the rod processes are cylindrical. The outer and inner segments are connected by a short cilium.
page 714 page 715
Figure 42.19 The major features of (A) a retinal rod cell and (B) a retinal cone cell. Note that the relative size of the pigment epithelial cells has been exaggerated for illustrative purposes.
Figure 42.20 Section of a human retinal rod showing the junction between the inner and outer segments. (By kind permission from G Vrenson and B Willekens, Ophthalmic Research Institute of Amsterdam.)
Cone cells are chiefly responsible for high spatial resolution and colour vision in good lighting conditions (photopic vision). Rod cells provide high monochromatic sensitivity to a much wider range of illumination down to much lower intensities (scotopic vision), although with relatively low spatial discrimination because of their different neural connections. Cone cells are of three types, red, green and blue, according to their maximum spectral sensitivities. They are highly concentrated in the centre of the retina (the fovea) where visual acuity is greatest, but they populate the whole retina, intermingled with rods, as far as its neural edge. Rods are excluded from the fovea. The total number of rods in the human retina has been estimated at 110-125 million, and of cones at 63-68 million (Österberg 1935). The outer segments of rods contain the photoreceptive protein rhodopsin (visual purple). Related photosensitive pigments with different absorption properties are present in cones. Photoreceptive pigments are incorporated into flattened membranous discs which form as deep infoldings of the plasma membrane and stack together within the photoreceptor outer segments (Fig. 42.20). They bud off as free discs within the outer segment of rods, where their turnover is rapid. New discs are generated at the proximal end closest to the soma, and shed at the distal end embedded in the pigment epithelium, where they are phagocytosed. Turnover appears to be less rapid in cones, where the discs retain continuity with the plasma membrane, and a more random insertion of disc components may occur. Cones are much narrower at the fovea where they closely resemble rods in size. Bipolar cells (Fig. 42.17) Bipolar cells are radially orientated neurones, each with one or more dendrites which synapse with cones or rods and horizontal cells and interplexiform cells in the outer plexiform layer. Their somata are located in the inner nuclear layer, and
axonal branches given off in the inner plexiform layer synapse with dendrites of ganglion cells or amacrine cells (Fig. 42.17). Cone bipolars are of three major types, namely midget, blue cone or diffuse, according to their connectivity and size. As their name implies, midget cone bipolars are small cells, each one a part of a single, one-to-one, channel from cone to ganglion cell; they are thought to mediate high spatial resolution. Blue cones have similar connectivity and selectively form part of a short wavelength mediating channel. The larger diffuse cone bipolars are connected to up to ten cones and are thought to signal luminosity rather than colour. Rod bipolars receive direct photoreceptive inputs from many rods and relate to ganglion cells indirectly via a synapse with amacrine cells. Horizontal cells (Fig. 42.17) Horizontal cells are inhibitory interneurones whose dendrites and axons extend within the outer plexiform layer, making synaptic contacts with the bases of cones and rods, and, via gap junctions at the tips of their dendrites, with each other. Their cell bodies lie in the outer part of the inner nuclear layer. Because of their interactions with photoreceptor cells and bipolar cells, horizontal cells create inhibitory surrounds. When illumination of a photoreceptor cluster with a point of light causes depolarization of synaptically connected 'ON' bipolars at its bright centre, horizontal cell dendrites cause inhibition at the edge of the illuminated area, thus sharpening contrast and maximizing spatial resolution. Amacrine cells (Fig. 42.17) Amacrine cells lack typical axons but their dendrites function as axons and dendrites, and make both incoming and outgoing synapses. Each neurone has a cell body either in the inner nuclear layer; near its boundary with the inner plexiform layer; or on the outer aspect of the ganglion cell layer, when they are known as displaced amacrine cells. The processes of amacrine neurones make scattered chemical synaptic contacts with the axons of bipolar cells; dendrites (and possibly axons) of ganglion cells; and the processes of other amacrine cells. They also receive numerous synapses from bipolar cells. Some amacrine cells form electrical synapses with bipolar cells. There are several classes of amacrine cell which variously serve a number of important functions in vision. One class (amacrine II cells) transmits signals from rod bipolars on to ganglion cells and is therefore an essential element in the rod pathway. Others appear to be important modulators of photoreceptive signals, and serve to adjust or maintain relative colour and luminosity inputs under changing light conditions, e.g. at different times of day. They are probably responsible for some of the complex forms of image analysis known to occur within the retina, e.g. directional movement detection. Ganglion cells (Fig. 42.17) Ganglion cells are the final common pathway neurones of the retina. Their dendrites are synaptically connected with processes of bipolar and amacrine neurones in the inner plexiform layer, and their axons likewise with neurones in the CNS. Their axons form the layer of nerve fibres on the inner surface of the retina. They turn tangentially to the optic disc, through which they leave the eye as fibres of the optic nerve, and the axons are subsequently distributed to various parts of the brain including the lateral geniculate nucleus, pretectal area and superior colliculus of the midbrain, the thalamic pulvinar and the accessory optic system. Ganglion cell bodies form a single stratum in most of the retina, but become progressively more numerous near the macula. They are ranked in about 10 rows in the macular area, and their number diminishes again towards the fovea, from which they are almost totally excluded. Ganglion cells are multipolar neurones, varying from 10-30µm or more in diameter. Their dendrites vary in number and branching patterns, and usually emerge opposite the axon. Numbers of ganglion cells in the human macular area reach 38,000/mm 2: they are more numerous in the nasal retina than the temporal, and in the superior retina compared with the inferior, although numbers vary considerably in different eyes. In total, each human retina has c.10 6 ganglion cells, each of which receives signals from large numbers of photoreceptor cells. Nerve fibre layer
In the nerve fibre layer, axons of ganglion cells converge on to the optic disc from the whole retina. They converge in a simple radial pattern from the medial (nasal) half of the retina. However, the macular area, inferolateral to the optic disc, complicates the course of the lateral (temporal) axons. Axons from the macula form a papillomacular fasciculus which passes almost straight to the disc. The more temporal fibres, which are more peripheral, swerve circumferentially above and below the macula to reach the disc. page 715 page 716
Axons of ganglion cells are almost always non-myelinated within the retina, which is an optical advantage because myelin is refractile. They are surrounded by the processes of radial glial cells and retinal astrocytes. A few small myelinated fibres may occur, but in general myelin sheaths usually only commence as the axons enter the optic disc to become the optic nerve. Retinal glial cells
Retinal glial cells are of three types, radial (Müller) cells, astrocytes and microglia. Radial glial cells form the predominant glial element of most of the retina. Retinal astrocytes are largely confined to the ganglion cell and nerve fibre layers. Microglial cells are scattered throughout the neural part of the retina in small numbers (Fig. 42.17). Radial glial cells span the entire thickness of the neural retina. They ensheath and separate the various photoreceptive and neural cells except at synaptic sites. They form the outer boundary of the retinal tissue at the level of the inner rod and cone segments, and the inner boundary at the internal limiting membrane. Their nuclei lie within the inner nuclear layer, and from this region a single thick fibre ascends radially, giving off complex lateral lamellae which branch among the processes of the outer plexiform layer. Apically the central process terminates in a surface from which microvilli project into the space between the rod and cone processes. Just beneath this area the radial glial cells form a line of dense zonulae adherentes with each other and with receptor inner segments, and so form the external limiting membrane. On the inner aspect of the retina, the main radial glial cell process expands in a terminal foot plate which contacts those of neighbouring radial glial cells and astrocytes and attaches to the internal limiting membrane. Like other neuroglia (Chapter 4), radial glia also contact blood vessels, especially capillaries, and their basal laminae fuse with those of the vascular smooth muscle in the media of larger vessels or of the endothelia lining capillaries. These extensive neuroglial cells form much of the total retinal volume, and almost totally fill the extracellular space between neural elements. Their functions appear to be similar to those of astrocytes, i.e. maintenance of the stability of the retinal extracellular environment by ionic transport; uptake of neurotransmitter; removal of debris; storage of glycogen; electrical insulation of receptors and neurones; mechanical support of the neural retina. The cell bodies of retinal astrocytes lie between the layer of nerve fibres and the internal limiting membrane, whilst their processes branch to form sheaths around ganglion cell axons. They are present only in regions of the retina that are vascularized, and are therefore absent from the fovea. Astrocytes contribute substantially to the glia limitans which surrounds the capillaries. Retinal microglia are scattered mainly within the inner plexiform layer. Their radiating branched processes spread mainly parallel to the retinal plane, and this gives them a starlike appearance when viewed microscopically from the surface of the retina. They can act as phagocytes, and their number increases in the injured retina. Internal limiting membrane The expanded end-feet of radial glial cells and astrocytes are separated from the vitreous body by a complex, rather thick (0.5µm) internal limiting membrane which is continuous with the internal limiting membrane of the ciliary body. The delicate collagen fibrils of the vitreous body blend with the glial basal lamina. The internal limiting membrane is involved in fluid exchange between the vitreous and the retina and, perhaps through the latter, with the choroid. It has various other functions including anchorage of retinal glial cells, and inhibition of cell migration into the vitreous body. Modifications in the macular area All the retinal layers are modified in the macular area, and to a marked degree in
the fovea, which is largely devoid of rod cells or processes. Approximately 2500 close-packed, elongated, very narrow cone cells lie in the floor of the fovea (foveola), an arrangement which favours photopic vision, and the high degree of spatial discrimination typical of foveal vision. The general displacement of the outer nuclear layer to the foveal periphery means that the internal processes of the photoreceptors are stretched out tangentially in the external plexiform layer, and consequently there are no cone pedicles or rod spherules in the central fovea and foveola. The inner nuclear layer is also displaced to the edge of the foveal depression, and the internal plexiform, ganglionic and nerve fibre layers are almost absent from the whole fovea. Therefore, even on the foveal wall, the retina is thinner and more transparent than elsewhere. Capillaries reach the foveal margin, but they only invade the ganglionic layer at its circumference, so that the fovea is normally devoid of all blood vessels.
© 2008 Elsevier
OPTIC DISC AND RETINAL BLOOD VESSELS The retina is placed between two sets of arteries and veins, the ciliary vessels of the choroid, and the branches of the central retinal vessels. It depends on both circulations, since neither is sufficient independently to maintain full visual activity in the retina. The choroidal circulation is described on page 716, and the orbital and intraneural parts of the central retinal vessels are described on page 705. The central retinal vessels enter and leave the retina at the optic disc, which will be described before the vessels are considered.
Optic disc (Figs 42.1, 42.2, 42.13, 42.14) The optic disc is the region where retinal tissues meet the neural and glial elements of the optic nerve and the connective tissues of the sclera and meninges. It is the exit point for the optic nerve fibres, and a point of entry and exit for the retinal circulation. It is the only site where anastomoses occur with other arteries (the posterior ciliary arteries). It is visible, by ophthalmoscopy, and is a region of much clinical importance, since it is here that the central vessels can be inspected directly: the only vessels so accessible in the whole body. Oedema of the disc (papilloedema) may be the first sign of raised intracranial pressure, which is transmitted into the subarachnoid space around the optic nerve and compresses the central retinal vein where it crosses the space. The optic disc is superomedial to the posterior pole of the eye, and so lies away from the visual axis. It is round or oval, usually c.1.6 mm in transverse diameter and 1.8 mm in vertical diameter, and its appearance is very variable (for details see Jonas et al 1988). In light-skinned subjects, the general retinal hue is a bright terracotta-red, with which the pale pink of the disc contrasts sharply; its central part is usually even paler and may be light grey. These differences are due in part to the degree of vascularization of the two regions, which is much less at the optic disc, and also to the total absence of choroidal or retinal pigment cells, since the retina is represented in the disc by little more than the internal limiting membrane. In subjects with strongly melanized skins, both retina and disc are darker (Fig. 42.13B). The optic disc does not project at all in many eyes, and rarely does it project sufficiently to justify the term papilla. It is usually a little elevated on its lateral side, where the papillomacular nerve fibres turn into the optic nerve. There is usually a slight depression where the retinal vessels traverse its centre.
RETINAL VASCULAR SUPPLY The central retinal artery enters the optic nerve as a branch of the ophthalmic artery, c.1.2 cm behind the eyeball. It travels in the optic nerve to its head, where its fascicles traverse the lamina cribrosa. At this level, which is usually not visible to ophthalmoscopy, the central artery divides into two equal branches, superior and inferior. After a few millimetres, these divide into superior and inferior nasal, and superior and inferior temporal, branches. Each of these four supplies its own 'quadrant' of the retina, although each territory is much more than a quadrant, since the branches ramify as far as the ora serrata. Corresponding retinal veins unite to form the central retinal vein. However, the courses of the venous and
arterial vessels do not correspond exactly, and arteries often cross veins, usually lying superficial to them. In severe hypertension the arteries may press on the veins and cause visible dilations distal to these crossings. Arterial pulsation is not visible by routine ophthalmoscopy without higher magnification. The branching of the artery is usually dichotomous, and equal rami diverge at angles of 45-60°. Smaller branches may leave singly and at right angles. Arteries and veins ramify in the nerve fibre layer, near the internal limiting membrane, which accounts for their clarity when seen through an ophthalmoscope (Figs 42.13, 42.14A,B). Arterioles pass deeper into the retina and may penetrate to the internal nuclear lamina, from which venules return to larger superficial veins. The question of whether or not the dense capillary bed is diffusely organized or layered is unsettled. Some lamination has been identified, most noticeably at the interface between the inner nuclear and outer plexiform layers. The structure of the blood vessels resembles that of vessels elsewhere, except that the internal elastic lamina is absent from the arteries, and muscle cells may appear in their adventitia. Capillaries have a non-fenestrated endothelium. page 716 page 717
Microcirculatory studies of the human retina in flat preparations, stained after trypsin digestion, have revealed many details of the capillary arrangement. It resembles that seen in renal glomeruli, i.e. a network of capillaries which connect individual arterioles and venules which are themselves devoid of anastomoses and arteriovenous shunts. Thus the territories of the arteries which supply a particular quadrant do not overlap, nor do the branches within a quadrant anastomose with each other. In consequence, a blockage in a retinal artery causes loss of vision in the corresponding part of the visual field. The only exception to this end-arterial pattern is in the vicinity of the optic disc. Here, the posterior ciliary arteries enter the eye near the disc (Figs 42.1, 42.7), and their rami not only supply the adjacent choroid, but also form an anastomotic circle in the sclera around the head of the optic nerve (Fig. 42.2). Branches from this ring join the pial arteries of the nerve, and small cilioretinal arteries from any arteries in this region may enter the eye and contribute to the retinal vasculature (Fig. 42.13D). Similarly, small retinociliary veins may sometimes also be present. Retinal capillaries do not pass towards the external surface of the retina beyond the inner nuclear lamina. They show regional differences in density, and are especially numerous in the macula, but absent from the central fovea. They become less numerous in the peripheral retina and are altogether absent from a zone c.1.5 mm wide which adjoins the ora serrata. Within the optic nerve, the central artery is innervated by sympathetic and parasympathetic fibres: the nerve supply does not extend to the retina.
© 2008 Elsevier
OCULAR REFRACTIVE MEDIA The components of the eye that transmit and refract light are the cornea, the aqueous humour, the lens and the vitreous body. Of these, only the refracting power of the lens can be varied. The cornea is described on page 702.
Aqueous humour To satisfy the requirements of vision the eye has its own circulatory system. Aqueous humour is secreted into the posterior chamber by the non-pigmented epithelium of the ciliary processes. It passes into the anterior chamber through the pupil and drains to the scleral venous sinus at the iridocorneal angle through the spaces of the trabecular tissue. It is responsible for maintaining the metabolism of the avascular transparent media, vitreous, lens and cornea, and it also maintains and regulates the relatively high intraocular pressure (c.17 mmHg), and hence the constancy of the ocular dimensions of the eyeball, via the balance between production and drainage. Depth of the anterior chamber may be assessed using slit-lamp biomicroscopy, and the filtration angle may be viewed directly by gonioscopy. Any interference with its drainage into the sinus increases intraocular pressure leading to the condition of glaucoma.
GLAUCOMA The characteristic physical sign of glaucoma is increased intraocular pressure, which is usually caused by obstruction of aqueous humour drainage at the iridocorneal angle (filtration angle). Increased formation of aqueous humour, or raised pressure in the veins draining the canal of Schlemm, are less commonly responsible. Although some drainage normally also occurs through uveoscleral channels, this alternative pathway cannot compensate adequately if the angle is blocked. Sustained raised pressure leads to defects in the visual field, and subsequently to blindness, either because of direct mechanical damage (particularly at the optic nerve head), or impairment of the blood supply, or both. Glaucoma may be either primary or secondary to a specific anomaly or disease of the eye. Secondary glaucoma
A variety of conditions may lead to secondary glaucoma: the following are the most common. Inflammatory glaucoma causes an increased pressure because the drainage channels become clogged by a turbid aqueous humour and by inflammatory exudates. As a later consequence of inflammation, exudates may form annular posterior synechiae, and these may attach the iris to the lens and prevent aqueous flow from the posterior to the anterior chamber. The root of the iris may attach to the cornea and block the angle: this may occur after perforation of the cornea, as a consequence of trauma or postoperatively, or following damage to the lens. Infantile glaucoma (buphthalmos) is an obstructive type which is usually the result of a failure in the development of the tissues of the iridocorneal angle. Primary glaucoma
The two distinct types of primary glaucoma are primary closed-angle and primary open-angle glaucoma. They differ in their clinical course and symptomatology. The closed-angle type characteristically affects hypermetropic eyes with a shallow anterior chamber where the angle is narrowed by the proximity of the root of the iris to the cornea. These features arise because the hypermetropic eye is generally small and the angle is likely to be narrowed further by the normal growth of the lens which presses the iris forward; lens development could explain the preponderance of the condition in the fifth and sixth decades, mostly in women. Attacks of raised pressure are usually sudden and subacute, and there is transient reduced vision and corneal oedema, followed later by persistent pressure instability. Without treatment, this ultimately leads to acute, painful congestive attacks and to permanent loss of vision.
Primary open-angle glaucoma is the commonest type, and the least readily diagnosed; there are no overt structural changes in the anterior segment of the eye. The disease is practically symptomless and slowly progressive, usually over several years. The outcome is similar to that of the other forms of glaucoma in that there is permanent congestion and blindness. Cellular changes in the trabecular meshwork reduce the facility of aqueous drainage, and this causes raised intraocular pressure, although the increment may not be maintained throughout the day. In the later stages a reduction in the visual field may be noticed.
Lens (Fig. 42.1, 42.21) The lens is a transparent, encapsulated, biconvex body, which lies between the iris and the vitreous body. Posteriorly, the lens contacts the hyaloid fossa (p. 719) of the vitreous body. Anteriorly, it forms a ring of contact with the free border of the iris, but further away from the axis of the lens the gap between the two increases to form the posterior chamber of the eye (p. 708). The lens is encircled by the ciliary processes, and is attached to them by the zonular fibres which issue mainly from the pars plana of the ciliary body. Collectively, the fibres form the zonule which holds the lens in place and transmits the forces which stretch the lens (except in visual accommodation). The lens has a characteristic shape (Fig. 42.1). Its anterior convexity is less steep, and has a greater radius of curvature, than the posterior, which has a more parabolic shape. The central points of these surfaces are the anterior and posterior poles; a line connecting these is the axis of the lens. The marginal circumference of the lens is its equator. In fetuses the lens is nearly spherical, has a slight reddish tinge, and is soft, such that it breaks up on application of the slightest pressure. A hyaloid artery from the central retinal artery traverses the vitreous body to the posterior pole of the lens, whence its branches spread as a plexus. This covers the posterior surface and is continuous round the capsular circumference with the vessels of the pupillary membrane and iris. In infants and adults the lens is avascular, colourless and transparent, but still quite soft in texture. In old age, the anterior surface becomes a little more curved, which pushes the iris forward slightly. It becomes less clear, with an amber tinge, and its nucleus is denser. In cataract, the lens gradually becomes opaque, causing blindness. The dimensions of the lens are optically and clinically important, but they change with age as a consequence of continuous growth. Its equatorial diameter at birth is 6.5 mm, increasing rapidly at first, then more slowly to 9.0 mm at 15 years of age, and even more gradually to reach 9.5 mm in the ninth decade. Its axial dimension increases from 3.5-4.0 mm at birth to 4.75-5.0 mm at age 95. The radii of curvature reduce throughout life; the anterior surface shows the greater change as the lens thickens (Brown 1974). Average adult radii of the anterior and posterior surfaces are 10 mm and 6 mm respectively; the reduction during accommodation occurs mainly at the anterior surface.
MICROSTRUCTURE OF THE LENS page 717 page 718
Figure 42.21 An adult human lens, fractured across to reveal its lamellar structure. Note that the small central part with a different fibre orientation may represent the embryonic nucleus; the adult nucleus cannot be distinguished from the cortex. The more steeply curved surface (below) is posterior. The different texture of the lens in the right part of this picture is caused by cutting prior to the fracture procedure. (Reprinted from Experimental Eye Research, Vrensen et al, Membrane architecture as a function of lens maturation. 54: 433-446, 1992, with permission from Elsevier.)
Figure 42.22 Section through the anterior layers of the lens. The thin capsule covers the single row of epithelial cells and the lens substance is composed of regularly stacked younger fibres and more densely stained and complex deeper fibres. (Provided by the late Gordon L Ruskell, Department of Optometry and Visual Science, The City University, London.)
The lens is derived from embryonic ectoderm (p. 198) and consists almost entirely of large numbers of stiff, very elongated, prismatic cells known as lens fibres, which are tightly packed together in a highly organized manner (Fig. 42.22). The anterior surface of the lens, as far as the equator, is covered by a layer of lens cells. The whole is surrounded by the lens capsule. The lens is avascular and devoid of nerve fibres or other structures which might affect its transparency. Its surface forms a very effective barrier against invasion by cells or elements of the immune system, and so creates an immunologically sequestered environment.
LENS FIBRES Each lens fibre is up to 12 mm long, depending on age and position in the lens
(related to their time of formation). Fibres near the surface at the equator are nucleated, and the nuclei form a short S-shaped bow which extends inwards from the surface, reflecting their sequence of formation from the superficial layer of anterior epithelial cells. The deeper fibres lose their nuclei. In cross-section, individual fibres are flattened hexagons measuring c.10µm by 2µm. They are tightly packed, and adjacent fibres are firmly adherent and interlocked by innumerable junctions of various kinds, resembling ball-and-socket or tongueand-groove joints, or close-fitting angular processes. Lens fibres are also in contact through desmosomes and numerous gap junctions. The cells contain crystallins, proteins which are responsible for the transparency and refractile properties, and for much of the elasticity, of the lens. At least two varieties coexist, ! and ". They occur in very high concentrations, and form up to 60% of the lens fibre mass. Variations in their concentration in different parts of the lens give rise to regional differences in refractive index, and these correct for spherical and chromatic aberrations which might otherwise occur in a homogeneous lens. Variations in lens fibre structure and composition make it possible to distinguish a softer cortical zone and a firmer central part, the nucleus (Fig. 42.21). Where sheaves of lens fibres contact each other at their ends, faint sutural lines are formed, which radiate out from the poles towards the equator (Fig. 42.23). In fetuses, the sutures on the anterior surface of the lens form a triradiate pattern (Fig. 42.23A), centred on the anterior pole and resembling the limbs of an upright letter Y; posteriorly, a similar, though inverted, sutural configuration is present. In adults, the sutures increase in number and complexity as a consequence of lens growth and other changes in the arrangement of lens fibres (Fig. 42.23B). The sutures represent the lines of linearly registered interlocking junctions between terminating lens fibres. If an extracted lens is hardened with fixatives and then broken open, the arrangement of lens fibres produces a striking, onion-like appearance, the fibres splitting into a series of concentric lamellae of varying thicknesses (Fig. 42.21). All lens fibres cross the equator (or the plane passing through it) where they are generated, and terminate on both an anterior and a posterior suture. Because of the curious growth pattern of the lens, fibres which start near the central axis of the lens anteriorly, terminate posteriorly on a suture near the periphery, and vice versa.
ANTERIOR LENS CELLS Anterior lens cells form a transparent layer of simple cuboidal epithelium over the anterior surface of the lens (Fig. 42.22). They divide and migrate to the equator where they transform into lens fibres. In surface view, anterior lens cells are polygonal. They are c.15µm across and slightly less in height, while in the central area they may be flattened to 6µm. They are tall and thin towards the equator, and mitosis is more frequent. Near the equator of the lens, they differentiate, i.e. they synthesize the characteristic lens fibre proteins, and undergo extreme elongation. As other cells follow suit, the earlier cells come to occupy a deeper position in the lens; they lose their nuclei as new recruits are added to the lens fibre population. Lens capsule page 718 page 719
Figure 42.23 The structure of the fetal (A) and adult (B) human lens, showing the major details of arrangement of the lens cells or fibres. The anterior (a) and posterior (b) triradiate sutures are shown in the fetal lens. Fibres pass from the apex of an arm of one suture to the angle between two arms at the opposite pole, as shown in the coloured segments. Intermediate fibres show the same reciprocal behaviour. The suture pattern becomes much more complex as successive strata are added to the exterior of the growing lens, and the original arms of each triradiate suture show secondary and tertiary dichotomous branchings. (By permission from Hogan MJ, Alvarado JA, Weddell JE 1971 Histology of the Human Eye. Philadelphia: WB Saunders.)
The lens capsule is a thick basement membrane which covers the outer surface of the lens (Fig. 42.22) and the anterior lens cell layer. It is thickest anteriorly (c.10µm), becomes thinner posteriorly and consists of various classes of collagen fibre (I, III and IV), as well as a range of glycosaminoglycans and glycoproteins, e.g. fibronectin and laminin. It is probably derived from the anterior lens cells and their fetal precursors. The capsule is elastic: during lens flattening it can stretch up to about 60% in circumference without tearing. Zonular fibres are inserted into the capsule in the region of the equator. They are composed of thin (4-7 nm) fibrils with hollow centres, and resemble fibrils associated with elastic connective tissue fibres. They possess some elasticity, although this decreases with age.
LENS REFRACTION The dioptric power of the lens is much less than that of the cornea. All ocular optical media have a refractive index close to that of water, but the corneal surface is in contact with air and most of the 60 dioptres of the refractive power of the eye are affected here. The value of the lens is its ability to vary its dioptric
power, which is dependent upon its capacity to change shape. It has a greater refractive index than the adjacent media, varying from 1.386 at its periphery to 1.406 at its core, and contributes c.17 dioptres to the total of the relaxed eye. Its range in dioptric power permits a further 12 dioptres in youth, but the available dioptric range decreases with age, being halved at 40 years and reduced to 1 dioptre or less at 60. Most young children show minor refractive errors modifying towards emmetropia in preschool and later years. For further information on physiological optics consult Bennett and Rabetts (1989).
DISORDERS OF REFRACTION In a relaxed eye, when the refracting structures are so related to its length that the retina receives a focused image of a distant object it is said to be emmetropic. A majority of eyes are emmetropic or nearly so and this state is maintained throughout life including the early years when growth demands a fine adjustment between cornea, lens power and eye length. But a large minority, have errors of refraction or ametropia that takes three different forms. These are: myopia, when the eye is too long for its refractive power; hyperopia, when it is too short; and astigmatism, when the refractive power of the eye is not the same in different meridians. Astigmatism usually occurs together with myopia and hyperopia. In myopia the image falls in front of the retina and if accommodation is attempted, blurring of the image is increased, however the myope has the advantage that close objects at an appropriate distance are conjugate to the retina and are therefore seen clearly. In hyperopia the image falls on an imaginary point behind the retina but the hyperope has the advantage that with accommodation, increasing the power of the eye, the image can be brought to a focus. This requires effort, perhaps sufficient to cause symptoms of asthenopia. No adjustments can be made to correct for astigmatism but the condition may induce an unrewarding effort to seek the best focus and again give rise to symptoms. There is a hereditable factor in ametropia and a relationship with the demands of close work in the young is widely accepted. These errors of refraction are amenable to correction using spectacle or contact lenses or less commonly by various forms of refractive surgery. There is a moderate deviation from emmetropia in the neonate including 1-3 dioptres of astigmatism in 50% and slight hyperopia. This is reduced after a year and mostly eliminated by school age when emmetropia predominates. Surveys show that in later years, mean hyperopia increases a fraction and myopia reduces. During the fifth decade the ability to change the power of the lens diminishes to an extent that neither the corrected ametrope nor the emmetrope is able to focus near objects clearly and reading spectacles become necessary. The disability increases gradually until the ability to accommodate is completely lost in the seventh decade. This condition, presbyopia, is treated using reading lenses. It is offset to a very limited extent by the reduction of the pupil aperture with age. This increases depth of focus, but at the cost of creating the further problem of requiring greater illumination. Other errors of refraction are the concomitants of eye disease, especially those which affect the cornea. Corneal curvature may be sufficiently altered as a residual defect of past disease to cause irregular astigmatism. This can be corrected conservatively only by using contact lenses which in effect provide a substitute for the irregular anterior corneal surface. In keratoconus, the cornea is thinned and steepened centrally, which distorts the refracting surface of the cornea to an extent that spectacle correction may not improve vision. Again contact lenses provide adequate treatment and in a large majority of cases keratoplasty does not become necessary. A dislocated lens, e.g. in Marfan's syndrome, disrupts the refractive status of the eye, and may not be amenable to spectacle correction.
Vitreous body page 719 page 720
The vitreous body fills the vitreous chamber, and occupies about four-fifths of the eyeball. It is hollowed in front as a deep concavity, the hyaloid fossa, which is adapted to the lens. It is colourless, consisting of c.99% water, but not entirely structureless. At its perimeter it has a gel-like consistency (100-300µm thick) and is firmly attached to the surrounding structures of the eye; nearer the centre it has a more liquid zone. Hyaluronan , in the form of long glycosaminoglycan chains, fills the whole vitreous. In addition, the peripheral gel or cortex contains a random loose network of type II collagen fibrils which are occasionally grouped into fibres. The cortex also contains scattered cells, the hyalocytes, which possess the characteristics of mononuclear phagocytes. They are responsible for the production of hyaluronan . Whilst they are normally in a resting state, they have the capacity to be actively phagocytic in inflammatory conditions. Hyalocytes are not present in the cortex bordering the lens. The liquid vitreous is absent at birth, appears first at 4 or 5 years, and increases to occupy half the vitreous space by the seventh decade. The cortex is most dense at the pars plana of the ciliary body adjacent to the ora serrata, where attachment is strongest, and this is often referred to as the base of the vitreous. Here the vitreous is thickened into a mass of radial (zonular) fibres which form the suspensory ligament of the lens. A narrow hyaloid canal (Fig. 42.1) runs from the optic nerve head to the central posterior surface of the lens. In the fetus this contains the hyaloid artery which normally disappears about 6 weeks before birth. It persists as a very delicate fibrous structure and is of no functional importance. REFERENCES Bennett AG, Rabbets RB 1989 Clinical Visual Optics, 2nd edn. London: Butterworth-Heinemann A detailed account of human ocular media in relation to physiological optics. Brown NP 1974 The change in lens curvature with age. Exp Eye Res 19: 175-183 Medline Similar articles Full article Hogan MJ, Alvarado JA, Weddell JE 1971 Histology of the Human Eye. Philadelphia: Saunders Jonas JB, Gusek GC, Naumann GO 1988 Optic disc, cup and neuroretinal rim size, configuration and correlations in normal eyes. Invest Ophthalmol Vis Sci 29: 1151-1158 A clinical template of a normal range. Medline Similar articles Loewenstein O, Loewenfeld IE 1969 The pupil. In: Davson H (ed) The eye, vol. 3, New York: Academic Press Osternerg GA 1935 Topography of the layers of the rods and cones in the human retina. Acta Ophthalmol Suppl 6 Wässle H, Boycott BB 1991 Functional architecture of the mammalian retina. Physiol Rev 71: 447-480 Unitary cell responses of the various retinal neurones and their connectivity. Medline Similar articles
© 2008 Elsevier
43 THE BONY ORBIT AND PERIPHERAL AND ACCESSORY VISUAL APPARATUS Development of the eye The development of the eye involves a series of interactions between neighbouring tissues in the head. These are the neurectoderm of the forebrain, which forms the sensory retina and accessory pigmented structures; the surface ectoderm, which forms the lens and cornea, and the intervening neural crest mesenchyme, which contributes to the fibrous coats of the eye. These interactions lead to the development of the potential to form optic vesicles throughout a broad anterior domain of neurectoderm. Subsequent interactions between mesenchyme and neurectoderm subdivide this region into bilateral domains at the future sites of the eyes. The parallel process of lens determination appears to depend on a brief period of inductive influence which spreads through the surface ectoderm from the rostral neural plate and elicits a lens-forming area of the head. Reciprocal interactions which are necessary for the complete development of both tissues take place as the optic vesicle forms and contacts the potential lens ectoderm (Saha et al 1992). Vascular tissue of the developing eye may form by local angiogenesis or vasculogenesis of angiogenetic mesenchyme. (Accounts of the development of the eye are given in O'Rahilly 1966 and 1983.)
© 2008 Elsevier
EMBRYONIC COMPONENTS OF THE EYE The first morphological sign of eye development is a thickening of the diencephalic neural folds at 22 days postovulation, when the embryo has 7-8 somites. This optic primordium extends on both sides of the neural plate and crosses the midline at the primordium chiasmatis. A slight transverse indentation, the optic sulcus, appears in the inner surface of the optic primordium on each side of the brain. During the period when the rostral neuropore closes, at about 24 days, the walls of the forebrain at the optic sulcus begin to evaginate, projecting laterally towards the surface ectoderm so that, by 25 days, the optic vesicles are formed. The lumen of each vesicle is continuous with that of the forebrain. Cells delaminate from the walls of the optic vesicle and, probably joined by head mesenchyme and cells derived from the mesencephalic neural crest, invest the vesicle in a sheath of mesenchyme. By 28 days, regional differentiation is apparent in each of the source tissues of the eye. The optic vesicle is visibly differentiated into its three primary parts. Thus, a thick-walled region marks the future optic stalk at the junction with the diencephalon; laterally, the tissue which will become the sensory retina forms a flat disc of thickened epithelium in close contact with the surface ectoderm; and the thin-walled part of the vesicle which lies between these regions will later form the pigmented layer of the retina. The area of surface ectoderm that is closely apposed to the optic vesicle also thickens to form the lens placode. The mesenchymal sheath of the vesicle begins to show signs of angiogenesis. Between 32 and 33 days postovulation, the lens placode and optic vesicle undergo coordinated morphogenesis. The lens placode invaginates, forming a pit which pinches off from the surface ectoderm to form the lens vesicle. The surface ectoderm reforms a continuous layer which will become the corneal epithelium. The lateral part of the optic vesicle also invaginates to form a cup, the inner layer of which - facing the lens vesicle - will become the sensory retina, and the outer layer becomes the pigmented retinal epithelium. As a result of these folding movements, what were the apical (luminal) surfaces of the two layers of the cup now face one another across a much reduced lumen, the intraretinal space. The pigmented layer becomes attached to the mesenchymal sheath, but the junction between the pigmented and sensory layers is less firm and is the site of pathological detachment of the retina. The two layers are continuous at the lip of the cup which, at the end of the third month, grows round the front of the lens and forms the pigmented iris. Between the base of the cup and the brain, the narrow part of the optic vesicle forms the optic stalk. The anteroventral surface of the vesicle and distal part of the stalk are also infolded, forming a wide groove - the choroid fissure - through which mesenchyme extends with an associated artery, the hyaloid artery. As growth proceeds, the fissure closes, and the artery is included in the distal part of the stalk. Failure of the optic fissure to close is a rare anomaly that is always accompanied by a corresponding deficiency in the choroid and iris (congenital coloboma).
© 2008 Elsevier
DIFFERENTIATION OF THE FUNCTIONAL COMPONENTS OF THE EYE The developments just described bring the embryonic components of the eye into the spatial relationships necessary for the passage, focusing, and sensing of light. The next phase of development involves further patterning and cell-type differentiation in order to develop the specialized structures of the adult organ. The optic cup becomes patterned, from the base to the rim, into regions with distinct functions (Fig. 43.1). The external stratum remains as a rather thin layer of cells which begin to acquire pigmented melanosomes and form the pigmented epithelium of the retina around 36 days. In a parallel process, which had already begun before invagination, the cells of the inner layer of the cup proliferate to form a thick epithelium. The inner layer forms neural tissue over the base and sides of the cup and non-neural tissue around the lip. The non-neural epithelium is further differentiated into the components of the prospective iris at the rim, and the ciliary body a little further back adjacent to the neural area. The development of this pattern is reflected in regional differences in the expression of various genes which encode transcriptional regulators and which are therefore likely to play key roles in controlling and coordinating development. Each of these genes is expressed prior to overt cell-type differentiation. For example, PAX6 is expressed in the prospective ciliary and iris regions of the optic cup. Individuals heterozygous for mutations in PAX6 lack an iris, which suggests a causal role for this gene in the development of the iris. The genes expressed in the eye are also active at a variety of other specific sites in the embryo, and this may, in part, account for the co-involvement of the eye and other organs in syndromes which result from single genetic lesions.
DEVELOPING NEURAL RETINA This comprises an outer nuclear zone and an inner marginal zone which is devoid of nuclei. Around 36 days the cells of the nuclear zone invade the marginal zone, and by 44 days the nervous stratum of the retina consists of inner and outer neuroblastic layers. The inner neuroblastic layer gives rise to the ganglion cells, the amacrine cells and the somata of the 'fibrous' sustentacular cells (of Muller); the outer neuroblastic layer is the source of the horizontal and rod-and-cone bipolar neurones and probably the rod-and-cone cells, which first appear in the central part of the retina. By the eighth month all the named layers of the retina can be identified. However, the retinal photoreceptor cells continue to form after birth, generating an array of increasing resolution and sensitivity. page 721 page 722
Figure 43.1 A-C, Section through the developing eyes of human embryos. A, 8 mm CR length. The thick nervous and the thinner pigmented layers of the retina and the developing lens are shown. Stained with haematoxylin and eosin. B, 13.2 mm CR length. C, 40 mm CR length. Note the layers of the retina, developing lens, pupillary membrane, cornea, conjunctival sac, anterior and posterior aqueous chambers, the developing vitreous body, and condensing circumoptic mesenchyme and the fused eyelids. Stained with haematoxylin and eosin. (A, from material loaned by Professor RJ Harrison; B, by permission from Streeter GL 1948 Developmental horizons in human embryos. Contrib Embryol Carnegie Inst Washington 32: 133-203.)
The divergent differentiation of the pigmented and sensory layers of the retina depends on interactions mediated by diffusible molecules. For example, soluble factors from the retina elicit the polarized distribution of plasma membrane proteins and the formation of tight junctions in the pigmented epithelium. Neural retinal differentiation appears to be mediated by fibroblast growth factors. However, the pigmented epithelium retains the potential to become neural retina and will do so if the embryonic retina is wounded.
OPTIC NERVE The optic nerve develops from the optic stalk. The centre of the optic cup, where the optic fissure is deepest, will later form the optic disc. Here the neural retina is continuous with the corresponding invaginated cell layer of the optic stalk; consequently the developing nerve fibres of the ganglion cells pass directly into the wall of the stalk and convert it into the optic nerve. The fibres of the optic nerve begin to acquire their myelin sheaths shortly before birth, but the process is not completed until some time later. The optic chiasma is formed by the meeting and partial decussation of the fibres of the two optic nerves in the ventral part of the lamina terminalis at the junction of the telencephalon with the diencephalon in the floor of the third ventricle. Beyond the chiasma, the fibres are continued backwards as the optic tracts, and pass principally to the lateral geniculate bodies and to the superior tectum.
CILIARY BODY
The ciliary body is a compound structure. Its epithelial components are the region of the inner layer of the retina between the iris and the neural retina and the adjacent outer layer of pigmented epithelium. The cells here differentiate in close association with the surrounding mesenchyme to form highly vascularized folds that secrete fluid into the globe of the eye. The inner surface of the ciliary body also forms the site of attachment of the lens, while the outer layer is associated with smooth muscle derived from mesenchymal cells in the choroid lying between the anterior scleral condensation and the pigmented ciliary epithelium.
IRIS The iris develops from the tip of the optic cup where the two layers remain thin and are associated with vascularized, muscular connective tissue. The muscles of the sphincter and dilator pupillae are unusual in being of neurectodermal origin, and develop from the cells of the pupillary part of the optic cup. The mature colour of the iris develops after birth and is dependent on the relative contributions of the pigmented epithelium on the posterior surface of the iris and the chromatophore cells in the mesenchymal stroma of the iris. If only epithelial pigment is present, the eye appears blue, whereas if there is an additional contribution from the chromatophores, the eye appears brown.
LENS The lens develops from the lens vesicle (Fig. 43.1A). Initially this is a ball of actively proliferating epithelium which encloses a clump of disintegrating cells, but by 37 days there is a discernible difference between the thin anterior (i.e. outward facing) epithelium and the thickened posterior epithelium. Cells of the posterior wall lengthen and fill the vesicle (Fig. 43.1B,C) and reduce the original cavity to a slit by about 44 days. The posterior cells become filled with a very high concentration of proteins (crystallins) which render them transparent. They also become densely packed within the lens as primary lens fibres. Cells at the equatorial region of the lens elongate and contribute secondary lens fibres to the body of the lens in a process which continues into adult life, sustained by continued proliferation of cells in the anterior epithelium. The polarity and growth of the lens appear to depend on the differential distribution of soluble factors which promote either cell division or lens fibre differentiation and are present in the anterior chamber and vitreous humour respectively. page 722 page 723
The developing lens is surrounded by a vascular mesenchymal condensation, the vascular capsule, the anterior part of which is named the pupillary membrane. The posterior part of the capsule is supplied by branches from the hyaloid artery, and the anterior part is supplied by branches from the anterior ciliary arteries. During the fourth month, the hyaloid artery gives off retinal branches. By the sixth month all of the vessels have atrophied except the hyaloid artery. The latter becomes occluded during the eighth month of intrauterine life, although its proximal part persists in the adult as the central artery of the retina. Atrophy of the hyaloid vasculature and of the pupillary membrane appears to be an active process of programmed tissue remodelling which is macrophage-dependent. The hyaloid canal, which carries the vessels through the vitreous, persists after the vessels
have become occluded. In the newly born child it extends more or less horizontally from the optic disc to the posterior aspect of the lens but when the adult eye is examined with a slit-lamp it can be seen to follow an undulating course, sagging downwards as it passes forwards to the lens. With the loss of its blood vessels the vascular capsule disappears and the lens becomes dependent for its nutrition on diffusion via the aqueous and vitreous humours. The lens remains enclosed in the lens capsule, a thickened basal lamina derived from the lens epithelium. Sometimes the pupillary membrane persists at birth, which gives rise to congenital atresia of the pupil.
VITREOUS BODY The vitreous body develops between the lens and the optic cup as a transparent, avascular gel of extracellular substance. The precise derivation of the vitreous remains controversial. The lens rudiment and the optic vesicle are at first in contact, but they draw apart after closure of the lens vesicle and formation of the optic cup, and remain connected by a network of delicate cytoplasmic processes. This network, derived partly from cells of the lens and partly from those of the retina, is the primitive vitreous body. At first these cytoplasmic processes are connected to the whole of the neuroretinal area of the cup, but later they become limited to the ciliary region where, by a process of condensation, they form the basis of the suspensory ligaments of the ciliary zonule. The vascular mesenchyme which enters the cup through the choroidal fissure and around the equator of the lens associates locally with this reticular tissue and thus contributes to the formation of the vitreous body.
AQUEOUS CHAMBER The aqueous chamber of the eye develops in the space between the surface ectoderm and the lens that is invaded by mesenchymal cells of neural crest origin. The chamber initially appears as a cleft in this mesenchymal tissue. The mesenchyme superficial to the cleft forms the substantia propria of the cornea, and that deep to the cleft forms the mesenchymal stroma of the iris and the pupillary membrane. Tangentially, this early cleft extends as far as the iridocorneal angle where communications are established with the sinus venosus sclerae. When the pupillary membrane disappears the cavity continues to form between the iris and the lens capsule as far as the zonular suspensory fibres. In this way the aqueous chamber is divided by the iris into anterior and posterior chambers, which communicate through the pupil. The walls of these chambers furnish both the sites of production of and the channels for circulation and reabsorption of the aqueous humour.
CORNEA The cornea is induced in front of the anterior chamber by the lens and optic cup. The corneal epithelium is formed from surface ectoderm and the epithelium of the anterior chamber is formed from mesenchyme. A regular array of collagen fibres is established between these two layers and these serve to reduce scattering of light entering the eye.
CHOROID AND SCLERA
The choroid and sclera differentiate as inner, vascular, and outer, fibrous, layers from the mesenchyme that surrounds the optic cup. The blood vessels of the choroid develop from the fifteenth week and include the vasculature of the ciliary body. The choroid is continuous with the internal sheath of the optic nerve which is pia-arachnoid mater, and the sclera is continuous with the outer sheath of the optic nerve, and thus with the dura mater.
© 2008 Elsevier
DIFFERENTIATION OF STRUCTURES AROUND THE EYE EXTRAOCULAR MUSCLES The extrinsic ocular muscles derive from prechordal mesenchyme which ingresses at the primitive node very early in development. The prechordal cells lie at the rostral tip of the notochordal process and remain mesenchymal after the notochordal process becomes epithelial and gains a basal lamina (Fig. 26.5). The prechordal mesenchyme migrates laterally towards the paraxial mesenchyme. Although this is a singular origin for muscle, the early myogenic properties of these cells have been demonstrated experimentally; moreover, if transplanted into limb buds, the cells are able to develop into muscle tissue (Wachtler & Jacob 1986). Early embryos develop bilateral premandibular, intermediate and caudal cavities in the head - previously described as preotic somites. The walls of the premandibular head cavities are lined by flat or cylindrical cells which do not exhibit the characteristics of a germinal epithelium. As the oculomotor nerve grows down to the level of the head cavity, a condensation of premuscle cells appears at its ventrolateral side which later subdivides into the blastemata of the different muscles that are supplied by the nerve. Similar events occur with respect to the intermediate head cavity (trochlear nerve and superior oblique), and the caudal head cavity (abducens nerve and lateral rectus) (Fig. 26.5). There is no doubt that the head cavities are formed by a mesenchymal/epithelial shift similar to that seen in the somites. However, the epithelial plate of the somite is a germinal centre which produces postmitotic myoblasts destined for epaxial regions, and migratory premitotic myoblasts destined for the limbs and body wall. The head cavities may serve a similar purpose if a mesenchyme/epithelial shift is part of a maturation process for putative myoblasts. However, it may not need to provide a centre for cell replication because premitotic myoblasts differentiated directly from the prechordal mesenchyme may form the premuscular masses.
EYELIDS The eyelids are formed as small cutaneous folds of surface ectoderm and neural crest mesenchyme (Fig. 43.1C). During the middle of the third month their edges come together and unite over the cornea to enclose the conjunctival sac, and they usually remain united until about the end of the sixth month. When the eyelids open, the conjunctiva which lines their inner surfaces and covers the white (scleral) region of the eye fuses with the corneal epithelium. The eyelashes and the lining cells of the tarsal (meibomian), ciliary and other glands which open onto the margins of the eyelids are all derived from the tarsal plate. Orbicularis oculi develops from skeletal myoblasts which invade the eyelids from the second pharyngeal arch. Levator palpebrae superioris develops from the prechordal mesenchyme and is attached to the upper eyelid by tendons derived from the neural crest. Smooth muscle also develops within the eyelids.
LACRIMAL APPARATUS The epithelium of the alveoli and ducts of the lacrimal gland arise as a series of tubular buds from the ectoderm of the superior conjunctival fornix. These buds are arranged in two groups: one forms the gland proper and the other forms its palpebral process (de la Cuadra-Blanco et al 2003). The lacrimal sac and nasolacrimal duct are derived from ectoderm in the nasomaxillary groove between the lateral nasal process and the maxillary process of the developing face (p. 609). This thickens to form a solid cord of cells, the nasolacrimal ridge, which sinks into the mesenchyme. During the third month the cord becomes canalized to form the nasolacrimal duct. The lacrimal canaliculi arise as buds from the cranial extremity of the cord which establish openings (puncta lacrimalia) on the margins of the lids. The inferior canaliculus isolates a small part of the lower eyelid to form the lacrimal caruncle and plica semilunaris. UPDATE Date Added: 12 April 2005 Abstract: Changing structure of human semilunar plica at different stages of development. Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=15255295 Changing structure of human semilunar plica at different stages of development. REFERENCES de la Cuadra-Blanco C, Peces-Pena MD, Mèrida-Velasco JR 2003 Morphogenesis of the human lacrimal gland. J Anat 203: 531-6. Medline Similar articles O'Rahilly R 1966 The early development of the eye in staged human embryos. Contrib Embryol Carnegie Inst 38: 1. O'Rahilly R 1983 The timing and sequence of events in the development of the human eye and ear during the embryonic period proper. Anat Embryol 168: 87-99. Medline Similar articles Full article Saha MS, Servetnick M, Grainger RM 1992 Vertebrate eye development. Curr Opin Genet Dev 2: 582-8. Reviews the interactions involved in eye development and discusses genes responsible for development of the eye. Medline Similar articles Full article Wachtler F, Jacob M 1986 Origin and development of the cranial skeletal muscles. Bibl Anat 29: 24-46. Medline Similar articles page 723 page 724
© 2008 Elsevier
SECTION 4 BACK AND MACROSCOPIC ANATOMY OF SPINAL CORD Andrew Williams (Lead Editor) Richard LM Newell (Editor) Patricia Collins (Embryology, Growth and Development) Critical reviewers: Michael A Adams (biomechanics), Paul Cartwright (chapter 44), Alison McGregor (biomechanics) page 725 page 726
© 2008 Elsevier
44 Surface anatomy of the back Most clinical disorders of the back present as low back pain with or without associated lower limb pain, so historically most attention has been paid to the anatomy of the lower (lumbosacral) back. In this chapter, the term 'the back' will include the whole of the posterior aspect of the trunk and of the neck. The whole of this region has great clinical importance but its anatomy has often been neglected. Recent understanding of the detailed topography of the bony and softtissue elements of the lower back owes much to the work of Bogduk (1997).
Figure 44.1 Back view of trunk. (Photograph by Sarah-Jane Smith.)
The soft tissues of the back of the trunk and neck include the skin and
subcutaneous fat, the underlying fascial layers, and the musculature. The deep, 'true' or epaxial muscles lie within compartments in their own fascial 'skeleton'. The bony framework to which the muscles and fasciae attach includes not only elements of the axial skeleton, i.e. the vertebral column and occiput, but also elements of the pectoral and pelvic girdles as well as the ribs. The occiput is described on p. 463, the scapula on p. 819, the ribs on p. 955 and the pelvis on p. 1421.
© 2008 Elsevier
SKELETAL LANDMARKS (Figs 44.1, 44.2, 44.3) VERTEBRAL SPINES
page 727 page 728
Figure 44.2 Back view of trunk, arms abducted. (Photograph by Sarah-Jane Smith.)
Figure 44.3 Back of trunk, oblique view. (Photograph by Sarah-Jane Smith.)
Table 44-1. Vertebral spines as landmarks for the viscera Spine Viscera C5 Cricoid cartilage, start of oesophagus C7 Apex of lung T3 Aorta reaches spine. Tracheal bifurcation T4 Aortic arch ends. Upper border of heart T8 Lower border of heart. Central tendon of diaphragm. T10 Lower border of lung. Cardia of stomach. Upper border of kidney
T12 L1 L2 L3 L4 L5
Lowest level of pleura. Pylorus Hilum of kidney. Renal arteries. Superior mesenteric artery Spinal cord terminates. Pancreas. Duodenojejunal flexure Lower border of kidney Bifurcation of aorta Inferior vena cava begins
In the midline a median furrow runs from the external occipital protuberance above to the natal cleft below. The furrow is most shallow in the lower cervical region and is deepest in the midlumbar zone. Inferiorly, it widens out into a flattened, triangular area, the apex of which lies at the start of the natal cleft and corresponds to the third sacral spine. Palpation of the median furrow reveals the sagittal curves of the spine: the cervical curve is convex forwards (lordosis) and extends from the first cervical to the second thoracic vertebra; the thoracic curvature is concave forwards (kyphosis) and extends from the second to the twelfth thoracic vertebra; the lumbar curvature is convex forwards and extends from the twelfth thoracic vertebra to the lumbosacral prominence. The external occipital protuberance is subcutaneous and can be felt and often seen, and it can be palpated without difficulty when it is approached from below. The inion is the point situated on this protuberance in the median plane. The tips of the spines of the cervical vertebrae are obscured by the overlying ligamentum nuchae. The tubercle on the posterior arch of the atlas is impalpable; the first bony prominence which is encountered when the finger is drawn downwards in the midline from the external occipital protuberance is the spine of the second cervical vertebra. The ligamentum nuchae terminates inferiorly at the spine of the seventh cervical vertebra, which is the highest, and sometimes the only, visible projection in this region (vertebra prominens). Immediately below this the spine of the first thoracic vertebra is palpable; it is usually more prominent than the seventh cervical vertebra. The spine of the second thoracic vertebra can also often be felt. The third thoracic spinous process is level with the spine of the scapula, and the seventh with the inferior scapular angle when the arm is by the side. The identification of the remaining thoracic spines is not easy, even in a thin subject when the trunk is fully flexed, because of the manner in which they overlap one another in the midthoracic region. In the upper and lower thoracic regions, the tips of the thoracic spines lie opposite the upper part of the body of the immediately subjacent vertebra. In the midthoracic region, they lie opposite the lower part of the vertebra below. The tip of the spine of each lumbar vertebra can usually be palpated without difficulty, especially if the trunk is flexed. Each lies opposite the inferior part of its own body. The body of the fourth lumbar vertebra is level with the summits of the iliac crests, so the fourth lumbar spine overlies the L4/5 interspace (a point useful in lumbar puncture). The second sacral spine is level with the posterior superior iliac spines. Sacral and coccygeal landmarks are described below (p. 728). Note that when the subject lies in the lateral position (on the side), the median soft-tissue furrow may not coincide with the median plane, especially in the
lumbar region in the obese. Careful palpation may be necessary to identify the spines, a point which is of particular importance during lumbar or epidural puncture in this position. Spinal levels of viscera
The palpable vertebral spines can be used as landmarks for the levels of the viscera. Some of the more important are shown in Table 44.1 (minor differences between the two sides are ignored).
SCAPULAR LANDMARKS The shape of the back in the upper thoracic region is determined largely by the scapula and the muscles which attach to it, especially trapezius. The relative prominence of the scapula and its muscles depends upon the state of contraction of the latter. Bony scapular landmarks are most evident when the arms hang by the sides. The scapula overlies the second to seventh ribs. Its superior angle is palpable beneath trapezius, and its inferior angle is just covered by latissimus dorsi. These angles are joined by the medial border of the scapula which runs vertically. The scapular spine runs subcutaneously and is easily palpable from its root medially to the acromion process laterally. The root of the spine of the scapula lies opposite the third thoracic spine and the inferior angle lies opposite the seventh thoracic spine when the arm is by the side.
POSTERIOR PELVIC AND SACROCOCCYGEAL BONY LANDMARKS page 728 page 729
At the lower part of the back the iliac crest can be palpated throughout its whole length, and can be traced backwards and upwards from the anterior superior iliac spine to its highest point, and then downwards and medially to the posterior superior iliac spine, which is overlain by a dimple in the skin. A line joining these sacral dimples passes through the second sacral spine and the body of the second sacral vertebra, and is a useful landmark for the lower end of the adult dural sac. There may also be a less prominent pair of dimples at the level of the L4/5 intervertebral disc. The posterior superior iliac spine lies over the centre of the sacroiliac joint. The supracristal plane joins the highest points of the iliac crest on each side. It passes through the body of the fourth lumbar vertebra and has been used as a clinical landmark when performing a lumbar puncture (p. 730) though difficulty in its definition reduces its reliability in this procedure (Broadbent et al 2000). The tip of the coccyx can be felt deeply near the centre of the natal cleft. As the examining finger passes cranially, the sacral cornua can be felt on either side: these demarcate the sacral hiatus, and form the landmark for performing a caudal anaesthetic block.
© 2008 Elsevier
MUSCULOTENDINOUS LANDMARKS (Figs 44.1, 44.2, 44.3) In the upper and middle cervical region the median furrow lies between the cylindrical prominences formed mainly by the semispinalis muscles, which are accentuated by neck extension against resistance. In the thoracic and lumbar regions, a broad elevation produced by the erector spinae muscle group extends for about one hand's breadth on either side of the median furrow and is present between the iliac crest and the twelfth rib. The lateral border of this elevation then crosses the ribs at their angles, passing medially as it ascends. The muscle group can be demonstrated by extending the back against resistance. Trapezius is a flat, triangular muscle which covers the back of the neck and shoulder. Together, the two trapezius muscles resemble a trapezium or quadrilateral in which two of the angles correspond to the shoulders, a third to the occipital protuberance and the fourth to the spine of the twelfth thoracic vertebra. The two muscles cover the back of the neck and shoulders like a monk's cowl, hence the ancient name of the trapezius was musculus cucullaris. The upper part of trapezius is demonstrated by elevation of the shoulders, or by extension and lateral flexion of the neck, against resistance. The lower fibres are best seen when the subject pushes both hands hard against a wall with the elbows extended. The anterosuperior border of the muscle forms the posterior boundary of the posterior triangle of the neck and can be seen in muscular subjects, especially during elevation of the shoulder against resistance. In a well-muscled subject the outline of latissimus dorsi can easily be traced, particularly if the arm is adducted against resistance. The triangle of auscultation lies between the upper border of latissimus dorsi, the lower lateral border of trapezius and the medial border of the scapula. The lumbar triangle, one of the sites of the rare primary lumbar hernia, lies inferiorly, between the lowermost outer border of latissimus dorsi, the posterior free border of the external oblique, and the iliac crest.
© 2008 Elsevier
SPINAL CORD AND ITS COVERINGS (Fig. 44.4) The surface relationships of the spinal cord and its coverings are of great clinical importance throughout life. During development the vertebral column elongates more rapidly than the spinal cord, which leads to an increasing discrepancy between the anatomical level of spinal cord segments and their corresponding vertebrae. At stage 23 the vertebral column and spinal cord are the same length, and the cord ends at the last coccygeal vertebra: this arrangement continues until the third fetal month. At birth the spinal cord terminates at the lower border of the second lumbar vertebra, and may sometimes reach the third lumbar vertebra. In the adult, the spinal cord is said to terminate at the level of the disc between the first and second lumbar vertebral bodies, which lies a little above the level of the elbow joint when the arm is by the side, and also lies approximately in the transpyloric plane (p. 1099). However, there is considerable variation in the level at which the spinal cord ends. It may end below this level in as many as 40% of subjects, or opposite the body of either the first or second lumbar vertebra: very occasionally it ends opposite the twelfth thoracic or even the third lumbar vertebra.
Figure 44.4 Contents of the vertebral canal in the lumbosacral region. Adapted from Mackintosh RR 1951 Lumbar Puncture and Spinal Analgesia. Edinburgh: E&S Livingstone. (Modified with permission from Mackintosh RR 1951 Lumbar Puncture and Spinal Analgesia. Edinburgh: E&S Livingstone.)
In estimating the vertebral levels of cord segments in the adult, a useful approximation is that in the cervical region the tip of a vertebral spinous process corresponds to the succeeding cord segment (i.e. the sixth cervical spine is opposite the seventh spinal segment); at upper thoracic levels the tip of a vertebral spine corresponds to the cord two segments lower (i.e. the fourth spine is level with the sixth segment), and in the lower thoracic region there is a
difference of three segments (i.e. the tenth thoracic spine is level with the first lumbar segment). The eleventh thoracic spine overlies the third lumbar segment and the twelfth is opposite the first sacral segment. In making this estimate by palpation of the vertebral spines, the relationship of the individual spines to their vertebral bodies should be remembered (p. 727). The dural sac (theca), and thus the subarachnoid space and its contained cerebrospinal fluid (CSF), usually extends to the level of the second segment of the sacrum. This corresponds to the line joining the sacral dimples located in the skin over the posterior superior iliac spines. Occasionally the dural sac ends as high as the fifth lumbar vertebra, and very rarely it may extend to the third part of the sacrum, in which case it is occasionally possible to enter the subarachnoid space inadvertently during the course of a sacral nerve block.
© 2008 Elsevier
UNDERLYING VISCERA The posterior surface markings of the abdominal viscera are described on page 1098.
© 2008 Elsevier
CLINICAL EXAMINATION Clinical examination of the back of the trunk and neck best follows the order of inspection, palpation and movement. The examination will be determined by the circumstances of presentation and by the history, and may include musculoskeletal, neurological and vascular observations. Information relevant to the neurological and vascular examination of the skin and material relating to spinal movements and deeper innervation are found below. Palpation of the region involves careful assessment of the bony and musculotendinous landmarks described above, looking in particular for asymmetry, deformity and tenderness. Note that, apart from the spines, most of the bony elements of the vertebrae and almost all of the intervertebral joints are not palpable from behind. In regions of lordosis (sagittal plane curves of the spine with anterior convexity, i.e. midcervical and mid- and lower lumbar), parts of the vertebral column can often be palpated anteriorly with care in well-relaxed, thin subjects.
© 2008 Elsevier
CLINICAL PROCEDURES ACCESS TO CEREBROSPINAL FLUID The safest approach to the cerebrospinal fluid (CSF) is to enter the lumbar cistern of the subarachnoid space in the midline, well below the level at which the spinal cord normally terminates (see p. 729). The fine needle employed is unlikely to damage the mobile nerve roots of the cauda equina. This procedure is called lumbar puncture. It is also possible to access the CSF by midline puncture of the cerebello-medullary cistern (cisterna magna) (Chapter 16): this is cisternal puncture.
LUMBAR PUNCTURE: ADULT Lumbar puncture in the adult may be performed with the patient either sitting or lying on the side on a firm flat surface. In each position, the lumbar spine must be flexed as far as possible, in order to separate the vertebral spines maximally and expose the ligamentum flavum in the interlaminar window (Fig. 44.5). A line is then taken between the highest points of the iliac crests: this line intersects the vertebral column just above the palpable spine of L4. With the spines now identified, the skin is anaesthetised and a needle is inserted between the spines of L3 and L4 (or L4 and L5). Exact identification of the level by palpation is difficult (Broadbent et al 2000). The soft tissues which the needle will ultimately traverse should also be anaesthetised, though care should be taken lest the injection of an excessive amount of local anaesthetic compromise appreciation of the structures being traversed. These include the subcutaneous fat, and supraspinous and interspinous ligaments down to the ligamentum flavum itself. The lumbar puncture needle may then be inserted in the midline or just to one side, and angled in the horizontal and sagittal planes sufficiently to pierce the ligamentum flavum in or very near the midline (Fig. 44.6).There is then a slight loss of resistance as the needle enters the epidural space, and careful advancement will next pierce the dura and arachnoid to release CSF.
LUMBAR PUNCTURE: NEONATE AND INFANT At full term (40 weeks) the spinal cord usually terminates somewhat lower than the adult level, sometimes reaching the body of L3. The supracristal plane intersects the vertebral column slightly higher (L3-4). By the second postnatal month the level of cord termination has usually reached its permanent position level with the body of the first lumbar vertebra. The lower end of the subarachnoid spine is found at sacral levels 1 or 2. These differences must be borne in mind when identifying the landmarks before undertaking lumbar puncture in the neonate and infant. A lumbar puncture is performed by placing the baby in a position, either lying or 'sitting', which gives maximum convex curvature to the lumbar spine. A needle with trocar is inserted into the back between the spines of the third and fourth lumbar vertebrae and into the subarachnoid space below the level of the conus medullaris. The space between L3 and L4 is approximately level with the iliac crests and it is usual to insert the needle and trocar into the intervertebral space
immediately above or below the iliac crests.
CISTERNAL PUNCTURE
Figure 44.5 The lumbar interlaminar window in extension and flexion.
page 730 page 731
Figure 44.6 The position of the needle in lumbar puncture.
In cisternal puncture, the cisterna magna (Chapter 16) is entered by midline puncture through the posterior atlanto-occipital membrane. Further details of this difficult specialist technique are beyond the scope of this book.
ACCESS TO THE EPIDURAL SPACE The epidural 'space' lies between the spinal dura and the wall of the vertebral canal (p. 778). It contains epidural fat and a venous plexus. Access to this space, usually in the lumbar region, is required for the administration of anaesthetic and analgesic drugs, and for endoscopy. The caudal route is used mainly for analgesic injections. Lumbar epidural
For access to the lumbar epidural space, the approach is as for lumbar puncture. The intention in epidural injection is to avoid dural puncture, so it is best to enter the epidural space in the midline posteriorly, where the depth of the space is greatest. Techniques for entering the epidural space rely on the appreciation of loss of resistance to injection of the chosen medium (usually air or saline) as the space is entered. There is very little distance between the ligamentum flavum and the underlying dura on either side of the median plane. Caudal epidural
The route of access to the caudal epidural space is via the sacral hiatus. The space is thus entered below the level of termination of the dural sac (S2). With the patient in the lateral position or lying prone over a pelvic pillow, the sacral hiatus is identified by palpation of the sacral cornua (p. 749) (Fig. 44.7). These are felt at the upper end of the natal cleft c.5 cm above the tip of the coccyx. Alternatively, the sacral hiatus may be identified by constructing an equilateral triangle based on a line joining the posterior superior iliac spines: the inferior apex of this triangle overlies the hiatus. After local anaesthetic infiltration, a needle is introduced at 45° to the skin, to penetrate the posterior sacrococcygeal ligament and enter the sacral canal. Once the canal is entered, the hub of the needle is lowered so that the needle may pass along the canal (Fig. 44.8). If the needle is angled too obliquely it will strike bone; if it is placed too superficially it will lie outside the canal. The latter malposition can be confirmed by careful injection of air while palpating the skin over the lower sacrum. Thoracic and cervical epidurals
It is possible to access the epidural space at thoracic and cervical levels, but the specialist techniques required are outside the scope of this book. The principles are the same as those for lumbar epidurals, but the special anatomy of the vertebral spines at the other levels requires the angle of approach to be modified.
Figure 44.7 Palpation of the sacral cornua for caudal epidural injection. With permission from Ellis H, Feldman SA 1997 Anatomy for Anaesthetists, 7th edn. Oxford: Blackwell Science. (By permission from Ellis H, Feldman S 1997 Anatomy for Anaesthetists, 7th edn. Oxford: Blackwell Science.)
Figure 44.8 Position of the needle in caudal epidural injection.
REFERENCES Bogduk N 1997 Clinical Anatomy of the Lumbar Spine and Sacrum, 3rd edn. Edinburgh: Churchill Livingstone. Broadbent CR, Maxwell WE, Ferrie R, Wilson DJ, Gawne-Cain M, Russell R 2000 Ability of anaesthetists to identify a marked lumbar interspace. Anaesthesia 55: 1122-6. Medline Similar articles Full article page 731 page 732
© 2008 Elsevier
45 The back SKIN The skin of the back of the trunk is thick and highly protective, but has low discriminatory sensation. The superficial fascia is thick and fatty in most areas of the back. Its attachment to the deeper fascial layers is strong in the midline, especially in the neck, but becomes weaker more laterally. The skin of the back of the neck is thicker than that of the front of the neck, but thinner than that of the back of the trunk. The quantity, texture and distribution of hair vary with sex, race and the individual, though well-defined hair tracts have been delineated (Fig. 45.1). Lines of skin tension (p. 173) run horizontally in the cervical and lumbosacral regions but form segments of two adjacent circles in the thoracic region (Fig. 45.2).
Figure 45.1 Hair tracts on the dorsal surface of the body. (By permission from Wood Jones F (ed) 1949 Buchanan's Manual of Anatomy, 8th edn. London: Baillière Tindall and Cox.)
CUTANEOUS INNERVATION AND DERMATOMES (Figs 46.14, 45.3) The skin of the back of the neck and trunk is innervated by the dorsal (posterior primary) rami of the spinal nerves (Fig. 46.14) where dorsal rami are covered in detail. In the cervical and upper thoracic regions (down to T6) skin is supplied by the medial branches of these rami, while in the lower thoracic, lumbar and sacral regions it is supplied by the lateral branches. The total area supplied by these dorsal rami is shown in Fig. 46.14. The spinal nerves involved include C2 to C5, T2 to L3, S2 to S4, and Co1. The pattern of their dermatomes is shown in Fig.
45.3. There is about half a segment of overlap between these cutaneous 'strips': the strips supplied by the dorsal rami do not correspond exactly to those served by ventral rami, and differ slightly in both width and position.
CUTANEOUS VASCULAR SUPPLY AND LYMPHATIC DRAINAGE (Figs 45.4, 45.5)
page 733 page 734
Figure 45.2 Lines of skin tension on the dorsum of the trunk and head. (From Kriassl CL, Plast Reconstruct Surg 8: 1-28, 1951. By permission from Lippincott Williams and Wilkins.)
Figure 45.3 Dermatomes on the dorsal surface of the body. The small diagram shows the regular arrangement of dermatomes in the upper and lower limbs of the embryo. (Adapted with permission from Moffat DB 1993 Lecture Notes on Anatomy, 2nd edn. Oxford: Blackwell Scientific.)
The skin of the back of the trunk receives its arterial blood supply mainly from musculocutaneous branches of posterior intercostal, lumbar and lateral sacral arteries, which all accompany the cutaneous branches of their respective dorsal rami. In addition, there is a supply from the dominant vascular pedicles of the superficial (extrinsic) back muscles. The skin over the scapula is supplied by branches of the suprascapular, dorsal scapular and subscapular arteries. The skin of the back of the neck is supplied mainly from the occipital and deep cervical arteries. The superficial cervical or transverse cervical artery supplies the skin of the lower part of the back of the neck (Cormack & Lamberty 1994).
Veins drain the skin of the back of the neck into tributaries of the occipital and deep cervical veins. The skin of the back of the trunk drains into the azygos system, via tributaries of the posterior intercostal and lumbar veins. Lymph from the skin of the back of the neck drains into occipital, lateral deep cervical and axillary nodes. From the back of the trunk, drainage is to the posterior (subscapular) axillary nodes and to the lateral superficial inguinal nodes.
© 2008 Elsevier
SKIN The skin of the back of the trunk is thick and highly protective, but has low discriminatory sensation. The superficial fascia is thick and fatty in most areas of the back. Its attachment to the deeper fascial layers is strong in the midline, especially in the neck, but becomes weaker more laterally. The skin of the back of the neck is thicker than that of the front of the neck, but thinner than that of the back of the trunk. The quantity, texture and distribution of hair vary with sex, race and the individual, though well-defined hair tracts have been delineated (Fig. 45.1). Lines of skin tension (p. 173) run horizontally in the cervical and lumbosacral regions but form segments of two adjacent circles in the thoracic region (Fig. 45.2).
Figure 45.1 Hair tracts on the dorsal surface of the body. (By permission from Wood Jones F (ed) 1949 Buchanan's Manual of Anatomy, 8th edn. London: Baillière Tindall and Cox.)
CUTANEOUS INNERVATION AND DERMATOMES (Figs 46.14, 45.3) The skin of the back of the neck and trunk is innervated by the dorsal (posterior primary) rami of the spinal nerves (Fig. 46.14) where dorsal rami are covered in detail. In the cervical and upper thoracic regions (down to T6) skin is supplied by the medial branches of these rami, while in the lower thoracic, lumbar and sacral regions it is supplied by the lateral branches. The total area supplied by these dorsal rami is shown in Fig. 46.14. The spinal nerves involved include C2 to C5, T2 to L3, S2 to S4, and Co1. The pattern of their dermatomes is shown in Fig.
45.3. There is about half a segment of overlap between these cutaneous 'strips': the strips supplied by the dorsal rami do not correspond exactly to those served by ventral rami, and differ slightly in both width and position.
CUTANEOUS VASCULAR SUPPLY AND LYMPHATIC DRAINAGE (Figs 45.4, 45.5)
page 733 page 734
Figure 45.2 Lines of skin tension on the dorsum of the trunk and head. (From Kriassl CL, Plast Reconstruct Surg 8: 1-28, 1951. By permission from Lippincott Williams and Wilkins.)
Figure 45.3 Dermatomes on the dorsal surface of the body. The small diagram shows the regular arrangement of dermatomes in the upper and lower limbs of the embryo. (Adapted with permission from Moffat DB 1993 Lecture Notes on Anatomy, 2nd edn. Oxford: Blackwell Scientific.)
The skin of the back of the trunk receives its arterial blood supply mainly from musculocutaneous branches of posterior intercostal, lumbar and lateral sacral arteries, which all accompany the cutaneous branches of their respective dorsal rami. In addition, there is a supply from the dominant vascular pedicles of the superficial (extrinsic) back muscles. The skin over the scapula is supplied by branches of the suprascapular, dorsal scapular and subscapular arteries. The skin of the back of the neck is supplied mainly from the occipital and deep cervical arteries. The superficial cervical or transverse cervical artery supplies the skin of the lower part of the back of the neck (Cormack & Lamberty 1994).
Veins drain the skin of the back of the neck into tributaries of the occipital and deep cervical veins. The skin of the back of the trunk drains into the azygos system, via tributaries of the posterior intercostal and lumbar veins. Lymph from the skin of the back of the neck drains into occipital, lateral deep cervical and axillary nodes. From the back of the trunk, drainage is to the posterior (subscapular) axillary nodes and to the lateral superficial inguinal nodes.
© 2008 Elsevier
FASCIAL LAYERS The main fascial layers in the axial and paraxial regions of the trunk and neck are the thoracolumbar fascia, the deep cervical fascia and the continuous prevertebral plane. The latter consists of the prevertebral, endothoracic, retroperitoneal and posterior part of the pelvic fasciae.
THORACOLUMBAR FASCIA (Figs 45.6, 68.1, 68.2, 68.3)
page 734 page 735
Figure 45.4 Areas of cutaneous arterial supply on the dorsum of the trunk. (Redrawn with permission from Cormack GC, Lamberty BGH 1994 The Arterial Anatomy of Skin Flaps. Edinburgh: Churchill Livingstone.)
Figure 45.5 Cutaneous lymphatics on the dorsum of the trunk. (By permission from Romanes GJ (ed) 1964 Cunningham's Textbook of Anatomy, 10th edn. London: Oxford University Press.)
The thoracolumbar (lumbodorsal) fascia covers the deep muscles of the back and the trunk. Above, it passes anterior to serratus posterior superior and is continuous with the superficial lamina of the deep cervical fascia on the back of the neck. In the thoracic region the thoracolumbar fascia provides a thin fibrous covering for the extensor muscles of the vertebral column and separates them from the muscles connecting the vertebral column to the upper extremity. Medially it is attached to the spines of the thoracic vertebrae, and laterally to the angles of the ribs. In the lumbar region the thoracolumbar fascia is in three layers. The posterior layer is attached to the spines of the lumbar and sacral vertebrae and to the supraspinous ligaments. The middle layer is attached medially to the tips of
the lumbar transverse processes and the intertransverse ligaments, below to the iliac crest, and above to the lower border of the twelfth rib and the lumbocostal ligament. The anterior layer covers quadratus lumborum and is attached medially to the anterior surfaces of the lumbar transverse processes behind the lateral part of psoas major; below, it is attached to the iliolumbar ligament and the adjoining part of the iliac crest; above, it forms the lateral arcuate ligament. The posterior and middle layers unite to form a tough raphe at the lateral margin of erector spinae, and at the lateral border of quadratus lumborum they are joined by the anterior layer to form the aponeurotic origin of transversus abdominis. At sacral levels, the posterior layer is attached to the posterior superior iliac spine and posterior iliac crest, and fuses with the underlying erector spinae aponeurosis. Bogduk (1997) describes two laminae in the posterior layer at lumbar levels, with varying orientation of the constituent collagen fibres relating to the biomechanical function of the fascia. The posterior and middle layers of the thoracolumbar fascia and the vertebral column together form an osteofascial compartment which encloses the erector spinae muscle group. The attachments of the fascia, especially those which give it continuity with the abdominal wall musculature, give it an important role in lifting, though the exact details of this role remain controversial. The fascia may also play an important role in load transfer between the trunk and the limbs: its tension is affected by the actions of latissimus dorsi, gluteus maximus, and the hamstrings. An erector spinae compartment syndrome may be one cause of low back pain.
DEEP CERVICAL FASCIA The investing layer of the deep cervical fascia forms the deep fascia of the posterior aspect of the neck. It attaches in the midline to the external occipital protuberance, the ligamentum nuchae and the spine of the seventh cervical vertebra, and splits to enclose trapezius on each side. Inferiorly the posterior part of the investing layer attaches with trapezius to the spine and acromion of the scapula.
© 2008 Elsevier
VERTEBRAL COLUMN (Figs 45.7, 45.8) The vertebral column is a curved linkage of individual bones or vertebrae. A continuous series of vertebral foramina runs through the articulated vertebrae posterior to their bodies, and collectively constitutes the vertebral canal, which transmits and protects the spinal cord and nerve roots, their coverings and vasculature. A series of paired lateral intervertebral foramina transmit the spinal nerves and their associated vessels between adjacent vertebrae. The linkages between the vertebrae include cartilaginous interbody joints and paired synovial facet (zygapophyseal) joints, together with a complex of ligaments and overlying muscles and fasciae. The muscles directly concerned with vertebral movements and attached to the column lie mainly posteriorly. Several large muscles producing major spinal movements lie distant from the column and without direct attachment to it, e.g. the anterolateral abdominal wall musculature. Movements of the column and the muscles concerned are described on page 769. The column as a whole receives its vascular supply and innervation according to general anatomical principles which are considered below. Vertebral column morphology is influenced externally by mechanical and environmental factors and internally by genetic, metabolic and hormonal factors. These all affect its ability to react to the dynamic forces of everyday life, such as compression, traction and shear. These dynamic forces can vary in magnitude and are much influenced by occupation, locomotion and posture. The adult vertebral column usually consists of 33 vertebral segments. Each presacral segment (except the first two cervical) is separated from its neighbour by a fibrocartilaginous intervertebral disc. The functions of the column are to support the trunk, to protect the spinal cord and nerves, and to provide attachments for muscles. It is also an important site of haemopoiesis throughout life. Its total length in males is c.70 cm and in females c.60 cm. The intervertebral discs contribute about a quarter of this length in young adults, though there is some diurnal variation in this contribution (p. 770). Approximately 8% of overall body length is accounted for by the cervical spine, 20% by the thoracic, 12% by the lumbar and 8% by the sacrococcygeal regions. Although the usual number of vertebrae is 7 cervical, 12 thoracic, 5 lumbar, 5 sacral and 4 coccygeal, this total is subject to frequent variability, and there have been reports of variation between 32 and 35 bones. The demarcation of groups by their morphological characteristics may be blurred: thus there may be thoracic costal facets on the seventh cervical vertebra, giving it the appearance of an extra thoracic vertebra; lumbar-like articular processes may be found on the lowest thoracic vertebra; the fifth lumbar vertebra may be wholly or partially incorporated into the sacrum. As a result of these changes in transition between vertebral types, there may be 23-25 mobile presacral vertebrae.
ANTERIOR ASPECT The anterior aspect of the column is formed by the anterior surfaces of the vertebral bodies and of the intervertebral discs (Fig. 45.7A). It has important anatomical relations at all levels, and should be considered in continuity. It forms part of several clinically significant junctional or transitional zones, including the prevertebral/retropharyngeal zone of the neck, the thoracic inlet, the diaphragm and the pelvic inlet. The anterior aspect of the column is covered centrally by the anterior longitudinal ligament, which forms a fascial plane with the prevertebral and endothoracic fascia and with the subperitoneal areolar tissue of the posterior abdominal wall. Infection and other pathological processes may spread along this fascial plane.
LATERAL ASPECT The lateral aspect of the vertebral column is arbitrarily separated from the posterior by articular processes in the cervical and lumbar regions and by transverse processes in the thoracic region (Fig. 45.7B). Anteriorly it is formed by the sides of vertebral bodies and intervertebral discs. The oval intervertebral foramina, behind the bodies and between the pedicles, are smallest at the cervical and upper thoracic levels, and increase progressively in size in the thoracic and upper lumbar regions. The lumbosacral (L5/S1) intervertebral
foramen is the smallest of the lumbar foramina. The foramina permit communication between the lumen of the vertebral canal and the paravertebral soft tissues (a 'paravertebral space' is sometimes described), which may be important in the spread of tumours and other pathological processes. The lateral aspects of the column have important anatomical relations, some of which vary considerably between the two sides.
POSTERIOR ASPECT The posterior aspect of the column is formed by the posterior surfaces of the laminae and spinous processes, their associated ligaments, and the facet joints (Fig. 45.7C). It is covered by the deep muscles of the back. page 735 page 736
Figure 45.6 Muscles and fasciae of the posterior abdominal wall. (By permission from Kiss F, Szentagothai J 1964 Atlas of Human Anatomy. Oxford: Pergamon Press.)
Structural defects of the posterior bony elements
Deformity and bony deficiency may occur at several sites within the posterior elements. The laminae may be wholly or partially absent, or the spinous process alone may be affected, even without overlying soft tissue signs (spina bifida occulta). A defect may occur in the part of the lamina between the superior and inferior articular processes (pars interarticularis): this condition is spondylolysis, and may be developmental or result from acute or fatigue fracture. If such defects are bilateral, the column becomes unstable at that level, and forward displacement of that part of the column above (cranial to) the defects may occur: this is spondylolisthesis. Abnormality of the laminar bone, or degenerative changes in the facet joints, may also lead to similar displacement in the absence of pars defects. The deformity of the vertebral canal resulting from severe spondylolisthesis may lead to neural damage. Much more rarely, bony defects may occur elsewhere in the posterior elements, e.g. in the pedicles. Detailed anatomical relations of all aspects of the vertebral column at the various
levels are best appreciated by the study of horizontal (axial) sections and images (Figs 45.35D and 45.54).
CURVATURES Embryonic and fetal curvatures
The embryonic body appears flexed. It has primary thoracic and pelvic curves which are convex dorsally. Functional muscle development leads to the early appearance of secondary cervical and lumbar spinal curvatures in the sagittal plane. The cervical curvature appears at the end of the embryonic period, and reflects the development of function in the muscles responsible for head extension, an important component of the 'grasp reflex'. Radiographic examination of human fetuses aged from 8 to 23 weeks shows that the secondary cervical curvature is almost always present. Lumbar flattening has also been identified as early as the eighth week. Ultrasound investigations support the role of movement in the development of these curvatures. The early appearance of the secondary curves is probably accentuated by postnatal muscular and nervous system development at a time when the vertebral column is highly flexible and is capable of assuming almost any curvature. Neonatal curvatures
In the neonate the vertebral column has no fixed curvatures (Figs 11.5, 11.6). It is particularly flexible and if dissected free from the body it can easily be bent (flexed or extended) into a perfect half circle. A slight sacral curvature can be seen which develops as the sacral vertebrae ossify and fuse. The thoracic part of the column is the first to develop a relatively fixed curvature, which is concave anteriorly. An infant can support its head at c.3 or 4 months, sit upright at c.9 months, and will commence walking between 12 and 15 months. These functional changes exert a major influence on the development of the secondary curvatures in the vertebral column and changes in the proportional size of the vertebrae, in particular in the lumbar region. The secondary lumbar curvature becomes important in maintaining the centre of gravity of the trunk over the legs when walking starts, and thus changes in body proportions exert a major influence on the subsequent shape of curvatures in the vertebral column. Adult curvatures page 736 page 737
Figure 45.7 The vertebral column: A, anterior aspect; B, lateral aspect; C, posterior aspect.
In adults, the cervical curve is a lordosis (convex forwards), and the least marked. It extends from the atlas to the second thoracic vertebra, with its apex between the fourth and fifth cervical vertebrae. Sexual dimorphism has been described in the cervical curvatures. The thoracic curve is a kyphosis (convex dorsally). It extends between the second and the eleventh and twelfth thoracic vertebrae, and its apex lies between the sixth and ninth thoracic vertebrae. This curvature is caused by the increased posterior depth of the thoracic vertebral bodies. The lumbar curve is also a lordosis. It has a greater magnitude in females and extends from the twelfth thoracic vertebra to the lumbosacral angle: there is an
increased convexity of the last three segments as a result of the greater anterior depth of the intervertebral discs and some posterior wedging of the vertebral bodies. Its apex is at the level of the third lumbar vertebra. The pelvic curve is concave anteroinferiorly and involves the sacrum and coccygeal vertebrae. It extends from the lumbosacral junction to the apex of the coccyx. The presence of these curvatures means that the cross-sectional profile of the trunk changes with spinal level. The anteroposterior diameter of the thorax is much greater than that of the lower abdomen. In the normal vertebral column there are well-marked curvatures in the sagittal plane and no lateral curvatures other than in the upper thoracic region, where there is often a slight lateral curvature, convex to the right in right-handed persons, and to the left in the lefthanded. Compensatory lateral curvature may also develop to cope with pelvic obliquity such as that imposed by inequality of leg length. The sagittal curvatures are present in the cervical, thoracic, lumbar and pelvic regions (Fig. 45.7). These curvatures have developed with rounding of the thorax and pelvis as an adaptation to bipedal gait.
VERTEBRAL COLUMN IN THE ELDERLY page 737 page 738
Figure 45.8 A, Sagittal MRI of thoracolumbosacral spine. B, Sagittal MRI of cervicothoracic spine. (By kind permission from Dr Justin Lee, Chelsea and Westminster Hospital, London.)
In older people, age-related changes in the structure of bone lead to broadening and loss of height of the vertebral bodies. These changes are more severe in females. The bony changes in the vertebral column are accompanied by changes in the collagen content of the discs and by decline in the activity of the spinal muscles. This leads to progressive decline in vertebral column mobility, particularly in the lumbar spine. The development of a 'dowager's hump' in the midthoracic region in females, caused by age-related osteoporosis, increases the thoracic kyphosis and cervical lordosis. Overall, these changes in the vertebral column lead directly to loss of total height in the individual. In mid-lumbar vertebrae the width of the body increases with age. In men there is a relative decrease of posterior to anterior body height, while in both sexes anterior height decreases relative to width. Twomey et al (1983) observed a reduction in bone density of lumbar vertebral bodies with age, principally as a result of a reduction in transverse trabeculae (more marked in females as a result of postmenopausal osteoporosis), which was associated with increased diameter and increasing concavity in their juxtadiscal surfaces (end-plates). Other changes affect the vertebral bodies. Osteophytes (bony spurs) may form from the compact cortical bone on the anterior and lateral surfaces of the bodies. Although individual variations occur, these changes appear in most individuals from c.20 years onwards. They are most common on the anterior aspect of the body, and never involve the ring epiphysis. Osteophytic spurs are frequently asymptomatic, but may result in diminished movements within the spine.
VASCULAR SUPPLY AND LYMPHATIC DRAINAGE (See also Crock 1996.) Arteries (Fig. 45.9)
Figure 45.9 Arterial supply to the vertebrae and the contents of the vertebral canal. A, Branching pattern of lumbar segmental arteries. B, Arterial anastomoses between postcentral branches of spinal arteries within the vertebral canal.
The vertebral column, its contents and its associated soft-tissues, all receive their arterial supply from derivatives of dorsal branches of the embryonic intersegmental somatic arteries (Ch. 61 and Fig. 61.24). The named artery concerned depends on the level of the column. These intersegmental vessels persist in the thoracic and lumbar regions as the posterior intercostal and lumbar arteries. In the cervical and sacral regions, longitudinal anastomoses between the intersegmental vessels persist as longitudinal vessels which themselves give spinal branches to the vertebral column. In the neck the postcostal anastomosis becomes most of the vertebral artery, while the post-transverse anastomosis forms most of the deep cervical artery. The ascending cervical artery and the lateral sacral artery are persistent parts of the precostal anastomosis. page 738 page 739
Figure 45.10 A, B Venous drainage of the vertebral column.
In the thorax and abdomen the primitive arterial pattern is retained by the paired branches of the descending aorta which supply the vertebral column (45.9A). On each side, the main trunk of the artery (posterior intercostal or lumbar (p. 1119) passes around the vertebral body, giving off primary periosteal and equatorial branches to the body, and then a major dorsal branch. The dorsal branch gives off a spinal branch which enters the intervertebral foramen, before itself supplying
the facet joints, the posterior surfaces of the laminae and the overlying muscles and skin. There is free anastomosis between these dorsal articular and soft-tissue branches, extending over several segments (Crock & Yoshizawa 1976; Boelderl et al 2002). At cervical and sacral levels the longitudinally running arteries described above have direct spinal branches. The spinal branches are the main arteries of supply to all bony elements of the vertebrae and to the dura and epidural tissues, and also contribute to the supply of the spinal cord and nerve roots via radicular branches (p. 784). As they enter the vertebral canal the spinal arteries divide into postcentral, prelaminar and radicular branches. The postcentral branches, which are the main nutrient arteries to the vertebral bodies and to the periphery of the intervertebral discs, anastomose beneath the posterior longitudinal ligament with their fellows above and below as well as across the midline (Fig. 45.9A,B). This anastomosis also supplies the anterior epidural tissues and dura. The majority of the vertebral arch, the posterior epidural tissues and dura and the ligamentum flavum are supplied by the prelaminar branches and their anastomotic plexus on the posterior wall of the vertebral canal. Veins (Fig. 45.10)
Veins of the vertebral column form intricate plexuses along the entire column, external and internal to the vertebral canal. Both groups are devoid of valves, anastomose freely with each other, and join the intervertebral veins. Interconnections are widely established between these plexuses and longitudinal veins early in fetal life. When development is complete, the plexuses drain into the caval and azygos/ascending lumbar systems via named veins which accompany the arteries described above. The veins also communicate with cranial dural venous sinuses and with the deep veins of the neck and pelvis. The venous complexes associated with the vertebral column can dilate considerably, and can form alternative routes of venous return in patients with major venous obstruction in the neck, chest or abdomen. The absence of valves allows pathways for the wide and sometimes paradoxical spread of malignant disease and sepsis. Pressure changes in the body cavities are transmitted to these venous plexuses and thus to the CSF, though the cord itself may be protected from such congestion by valves in the small veins which drain from the cord into the internal vertebral plexus. page 739 page 740
External vertebral venous plexuses
The external vertebral venous plexuses are anterior and posterior. They anastomose freely, and are most developed in the cervical region. Anterior external plexuses are anterior to the vertebral bodies, communicate with basivertebral and intervertebral veins, and receive tributaries from vertebral bodies. Posterior external plexuses lie posterior to the vertebral laminae and around spines, transverse and articular processes. They anastomose with the internal plexuses and join the vertebral, posterior intercostal and lumbar veins. Internal vertebral venous plexuses
The internal vertebral venous plexuses occur between the dura mater and vertebrae, and receive tributaries from the bones, red bone marrow and spinal cord. They form a denser network than the external plexuses and are arranged vertically as four interconnecting longitudinal vessels, two anterior and two posterior. The anterior internal plexuses are large plexiform veins on the posterior surfaces of the vertebral bodies and intervertebral discs. They flank the posterior longitudinal ligament, beneath which they are connected by transverse branches into which the large basivertebral veins open. The posterior internal plexuses, on each side in front of the vertebral arches and ligamenta flava, anastomose with the posterior external plexuses via veins which pass through and between the ligaments. The internal plexuses interconnect by venous rings near each vertebra. Around the foramen magnum they form a dense network connecting with vertebral veins, occipital and sigmoid sinuses, the basilar plexus, the venous plexus of the hypoglossal canal, and the condylar emissary veins. Basivertebral veins
The basivertebral veins emerge from the posterior foramina of the vertebral bodies. They are large and tortuous channels in bone, like those in the cranial diploë. The basivertebral veins also drain into the anterior external vertebral plexuses through small openings in the vertebral bodies. Posteriorly they form one or two short trunks which open into the transverse branches and unite anterior internal vertebral plexuses. They enlarge in advanced age. Intervertebral veins
The intervertebral veins accompany the spinal nerves through intervertebral foramina, draining the spinal cord and internal and external vertebral plexuses, and ending in the vertebral, posterior intercostal, lumbar and lateral sacral veins. Upper posterior intercostal veins may drain into the caval system via brachiocephalic veins, whereas the lower intercostals drain into the azygos system. Lumbar veins are joined longitudinally in front of the transverse processes by the ascending lumbar veins, in which they may terminate. Alternatively, they may proceed around the vertebral bodies to drain into the inferior vena cava. Whether the basivertebral or intervertebral veins contain effective valves is uncertain but experimental evidence strongly suggests that their blood flow can be reversed (Batson 1957). This may explain how pelvic neoplasms, e.g. carcinoma of the prostate, may metastasize in vertebral bodies: the cells spread into the internal vertebral plexuses via their connections with the pelvic veins when blood flow is temporarily reversed by raised intra-abdominal pressure or postural alterations. Lymphatic drainage
Little is known in detail about the lymphatic drainage of the vertebral column and its associated soft tissues. In general, deep lymphatic vessels tend to follow the arteries. The cervical vertebral column drains to deep cervical nodes, the thoracic to (posterior) intercostal nodes, and the lumbar column to lateral aortic and retroaortic nodes. The pelvic part of the column drains to lateral sacral and internal iliac nodes.
INNERVATION Innervation of the vertebral column and its associated soft tissues has been studied in greatest detail in the lumbar region. The account given here relies particularly on the work of Bogduk, to whose textbook on the lumbosacral spine (Bogduk 1997) the interested reader is referred. See also the work of Groen et al (1990). Innervation is derived from the spinal nerves where they branch, in and just beyond the intervertebral foramina. There is also an important input from the sympathetic system via grey rami communicantes or directly from thoracic sympathetic ganglia. The branches of the spinal nerve concerned are the dorsal ramus and the recurrent meningeal or sinuvertebral nerves (usually more than one at each level) (p. 782 and also Fig. 46.11). The dorsal ramus branches to supply the facet joints, periosteum of the posterior bony elements, overlying muscles and skin. The exact origin and branching pattern of the sinuvertebral nerves is controversial, but they may be best considered to be recurrent branches of the ventral rami. They receive the sympathetic input described above, then reenter the intervertebral foramina to supply the structures forming the walls of the vertebral canal as well as the dura and epidural soft tissues. Their subsequent course is described on page 782.
© 2008 Elsevier
VERTEBRAE: GENERAL FEATURES A typical vertebra has a ventral body, a dorsal vertebral (neural) arch, extended by lever-like processes, and a vertebral foramen, which is occupied in life by the spinal cord, meninges and their vessels (Fig. 45.11). Opposed surfaces of adjacent bodies are bound together by intervertebral discs of fibrocartilage. The complete column of bodies and discs forms the strong but flexible central axis of the body and supports the full weight of the head and trunk. It also transmits even greater forces generated by muscles attached to it directly or indirectly. The foramina form a vertebral canal for the spinal cord, and between adjoining neural arches, near their junctions with vertebral bodies, intervertebral foramina transmit mixed spinal nerves, smaller recurrent nerves and blood and lymphatic vessels. The cylindroid vertebral body varies in size, shape and proportions in different regions of the vertebral column. Its superior and inferior (discal) surfaces vary in shape from approximately flat (but not parallel) to sellar, with a raised peripheral smooth zone formed from an 'anular' epiphyseal disc within which the surface is rough. These differences in texture reflect variations in the early structure of intervertebral discs. In the horizontal plane the profiles of most bodies are convex anteriorly, but concave posteriorly where they complete the vertebral foramen. Most sagittal profiles are concave anteriorly but flat posteriorly. Small vascular foramina appear on the front and sides, and posteriorly there are small arterial foramina and a large irregular orifice (sometimes double) for the exit of basivertebral veins (Fig. 45.12). The adult vertebral body is not coextensive with the developmental centrum but includes, posterolaterally, parts of the neural arch.
page 740 page 741
Figure 45.11 Fourth thoracic vertebra, superior aspect. (Photograph by Sarah-Jane
Smith.)
Figure 45.12 Median sagittal section through a lumbar vertebra.
Viewed anteriorly there is a cephalocaudal increase in vertebral body width from the second cervical to the third lumbar vertebra, which is associated with an increased load-bearing function. The increase is linear in the neck but not in the thoracic and lumbar regions. There is some variation in size of the last two lumbar bodies, but thereafter width diminishes rapidly to the coccygeal apex. In the two lowest lumbar vertebrae there is an inverse relation between the areas of the upper and lower surfaces of the bodies and the size of the pedicles and transverse processes. On each side the vertebral arch has a vertically narrower ventral part, the pedicle, and a broader lamina dorsally. Paired transverse, superior and inferior articular processes project from their junctions. There is a median dorsal spinous process. Pedicles are short, thick, rounded dorsal projections from the superior part of the body at the junction of its lateral and dorsal surfaces: the concavity formed by the curved superior border of the pedicle is shallower than the inferior one (Fig. 45.12). When vertebrae articulate by the intervertebral disc and facet joints, these adjacent vertebral notches contribute to an intervertebral foramen. The complete perimeter of an intervertebral foramen consists of the notches, the dorsolateral aspects of parts of adjacent vertebral bodies and the intervening disc, and the capsule of the synovial facet joint. The laminae are directly continuous with the pedicles. They are vertically flattened and curve dorsomedially to complete, with the base of the spinous process, a vertebral foramen. Lateral to the spinous processes, vertebral grooves contain the deep dorsal muscles. At cervical and lumbar levels these grooves are shallow and mainly formed by laminae. In the thoracic region they are deeper, broader and formed by the laminae and transverse processes. The laminae are broad for the first thoracic vertebra, narrow for the second to seventh, broaden again from the
eighth to eleventh, but become narrow thereafter down to the third lumbar vertebra. The spinous process (vertebral spine) projects dorsally and often caudally from the junction of the laminae. Spines vary considerably in size, shape and direction. They lie approximately in the median plane and project posteriorly, although in some individuals a minor deflection of the processes to one side may be seen. The spines act as levers for muscles which control posture and active movements (flexion/extension, lateral flexion and rotation) of the vertebral column. The paired superior and inferior articular processes (zygapophyses) arise from the vertebral arch at the pediculolaminar junctions. The superior processes project cranially, bearing dorsal facets which may also have a lateral or medial inclination, depending on level. Inferior processes run caudally with articular facets directed ventrally, again with a medial or lateral inclination which depends on vertebral level. Articular processes of adjoining vertebrae thus contribute to the synovial zygapophyseal or facet joints (p. 757), and form part of the posterior boundaries of the intervertebral foramina. These joints permit limited movement between vertebrae: mobility varies considerably with vertebral level. Transverse processes project laterally from the pediculolaminar junctions as levers for muscles and ligaments, particularly those concerned in rotation and lateral flexion. In the cervical region, the transverse processes are anterior to the articular processes, lateral to the pedicles and between the intervertebral foramina. In the thoracic region, they are posterior to the pedicles, considerably behind those of the cervical and lumbar processes. In the lumbar region, the transverse processes are anterior to the articular processes, but posterior to the intervertebral foramina. There is considerable regional variation in the structure and length of the transverse processes. In the cervical region, the transverse process of the atlas is long and broad, which allows the rotator muscles maximum mechanical advantage. Breadth varies little from the second to the sixth cervical vertebra, but increases in the seventh. In thoracic vertebrae, the first is widest, and breadth decreases to the twelfth, where the transverse elements are usually vestigial. The transverse processes become broader in the upper three lumbar vertebrae, and diminish in the fourth and fifth. The transverse process of the fifth lumbar vertebra is the most robust. It arises directly from the body and pedicle to allow for force transmission to the pelvis through the iliolumbar ligament. The thoracic transverse processes articulate with ribs, but at other levels the mature transverse process is a composite of 'true' transverse process and an incorporated costal element. Costal elements develop as basic parts of neural arches in mammalian embryos, but become independent only as thoracic ribs. Elsewhere they remain less developed and fuse with the 'transverse process' of descriptive anatomy (Fig. 47.7). Vertebrae are internally trabecular, and have an external shell of compact bone perforated by vascular foramina (Fig. 45.12). The shell is thin on the superior and inferior body surfaces but thicker in the arch and its processes. The trabecular interior contains red bone marrow and one or two large ventrodorsal canals which contain the basivertebral veins.
Sexual dimorphism in vertebrae has received little attention, but Taylor and Twomey (1984) have described radiological differences in adolescent humans and have reported that female vertebral bodies have a lower ratio of width to depth. Vertebral body diameter has also been used as a basis for sex prediction in the analysis of skeletal material (MacLaughlin & Oldale 1992).
VERTEBRAL CANAL (Fig. 45.13) The vertebral canal extends from the foramen magnum to the sacral hiatus, and follows the vertebral curves. In the cervical and lumbar regions, which exhibit free mobility, it is large and triangular, but in the thoracic region, where movement is less, it is small and circular. These differences are matched by variations in the diameter of the spinal cord and its enlargements. In the lumbar region, the vertebral canal decreases gradually in size between L1 and L5, with a greater relative width in the female. For clinical purposes it is useful to consider the vertebral canal as having three zones. These are a central zone, between the medial margins of the facet joints, and two lateral zones, beneath the facet joints and entering the intervertebral foramina. Each lateral zone, which passes into and just beyond the intravertebral foramen, can be further subdivided into subarticular (lateral recess), foraminal and extraforaminal regions (Macnab & McCulloch 1990). The lateral zone thus described forms the canal of the spinal nerve (the radicular or 'root' canal). The central zone of the canal is a little narrower than the radiological interpedicular distance if the lateral recess is considered to be part of the radicular canal rather than part of the central zone. Spinal stenosis
Narrowing (stenosis) of the vertebral canal may occur at single or multiple spinal levels, and mainly affects the lumbar and cervical regions. Stenosis may affect the central canal and the 'root canals' either together or separately. There is a developmental form of the condition which mainly affects the central canal, but more commonly the stenosis is degenerative, and results from intervertebral disc narrowing and osteoarthritic changes in the facet joints. This latter combination is more likely to narrow the intervertebral foramen and the 'root canal', even though the sectional profile of the vertebral canal in affected lumbar vertebrae typically changes from the shape of a bell to that of a trefoil. The lumbosacral intervertebral foramen, which is normally the smallest in the region, is particularly liable to such stenosis. Severe spinal stenosis may compress the spinal cord and compromise its arterial supply. More localized 'root canal' stenosis will present with the clinical features of spinal nerve compression, but without the tension signs which characterize the stretching of nerve roots over a prolapsed disc. Ischaemia of the nerves and roots may be more responsible for the damage than is the actual physical compression of the neural tissue.
INTERVERTEBRAL FORAMINA (See also p. 735 and p. 782.) page 741 page 742
Figure 45.13 The vertebral canal in section: A, sagittal; B, transverse (axial); C, coronal.
Intervertebral foramina are the principal routes of entry and exit to and from the vertebral canal, and are closely related to the main intervertebral articulations. (Minor routes occur between the median, often partly fused, margins of the ligamenta flava.) The same general arrangement applies throughout the vertebral column, between the axis and sacrum, although there are some quantitative and structural regional variations. Because of their construction, contents and susceptibilities to multiple disorders, the intervertebral foramina are loci of great biomechanical, functional and clinical significance. The specializations cranial to the axis and at sacral levels are described with the individual bones and articulations.
Figure 45.14 The boundaries of an intervertebral foramen.
The boundaries of a generalized intervertebral foramen (Fig. 45.14) are anteriorly, from above downwards, the posterolateral aspect of the superior vertebral body, the posterolateral aspect of the intervertebral symphysis (including the disc), and a small (variable) posterolateral part of the body of the inferior vertebra; superiorly, the compact bone of the deep arched inferior vertebral notch of the vertebra above; inferiorly, the compact bone of the shallow superior vertebral notch of the vertebra below; and posteriorly a part of the ventral aspect of the fibrous capsule of the facet synovial joint. Cervical intervertebral foramina are distinct in having superior and inferior vertebral notches of almost equal depth which, in accord with the direction of the pedicles, face anterolaterally. External to them, and oriented in the same direction, is a transverse process. The thoracic and lumbar intervertebral foramina face laterally and their transverse processes
are posterior. In addition, the anteroinferior boundaries of the first to tenth thoracic foramina are formed by the articulations of the head of a rib and the capsules of double synovial joints (with the demifacets on adjacent vertebrae and the intraarticular ligament between the costocapitular ridge and the intervertebral symphysis). Lumbar foramina lie between the two principal lines of vertebral attachment of psoas major. The walls of each foramen are covered throughout by fibrous tissue which is in turn periosteal (though the presence of a true periosteum lining the vertebral canal is controversial [Newell 1999]), perichondrial, annular and capsular. The more lateral parts of the foramina may be crossed at a variable level by narrow fibrous bands, the transforaminal ligaments. The true foramen is the foraminal region of the canal of the spinal nerve (the radicular or 'root' canal). A foramen contains a segmental mixed spinal nerve and its sheaths, from two to four recurrent meningeal (sinuvertebral) nerves, variable numbers of spinal arteries, and plexiform venous connections between the internal and external vertebral venous plexuses. These structures, particularly the nerves, may be affected by trauma or one of the many disorders which may affect tissues bordering the foramen. In particular, nerve compression and irritation may be caused by intervertebral disc prolapse, or by bony entrapment as the size of the foramen decreases as a result of facet joint osteoarthritis, osteophyte formation or disc degeneration, all of which may lead to lateral or foraminal spinal stenosis.
© 2008 Elsevier
CERVICAL VERTEBRAE (Figs 45.8B, 45.15, 45.16) The cervical vertebrae are the smallest of the moveable vertebrae, and are characterized by a foramen in each transverse process. The first, second and seventh have special features and will be considered separately. The third, fourth and fifth cervical are almost identical, and the sixth, while typical in its general features, has minor distinguishing differences.
TYPICAL CERVICAL VERTEBRA (Figs 45.17, 45.18) page 742 page 743
Figure 45.15 The cervical vertebrae (anterior aspect).
Figure 45.16 Lateral radiograph of the neck. The cervical curve of the vertebral column is well shown. (Provided by Shaun Gallagher, GKT School of Medicine, London; photograph by Sarah-Jane Smith.)
Figure 45.17 Fourth cervical vertebra, superior aspect. (Photograph by Sarah-Jane Smith.)
Figure 45.18 Fourth cervical vertebra, lateral aspect. (Photograph by Sarah-Jane Smith.)
A typical cervical vertebra has a small, relatively broad vertebral body. The pedicles project posterolaterally and the longer laminae posteromedially, enclosing a large, roughly triangular vertebral foramen; the vertebral canal here accommodates the cervical enlargement of the spinal cord. The pedicles attach midway between the discal surfaces of the vertebral body, so the superior and inferior vertebral notches are of similar depth. The laminae are thin and slightly curved, with a thin superior and slightly thicker inferior border. The spinous process ('spine') is short and bifid, with two tubercles which are often unequal in size. The junction between lamina and pedicle bulges laterally between the superior and inferior articular processes to form, when articulated, an articular pillar ('lateral mass') on each side. The transverse process is morphologically
composite around the foramen transversarium. Its dorsal and ventral bars terminate laterally as corresponding tubercles. The tubercles are connected, lateral to the foramen, by the costal (or intertubercular) lamella: these three elements represent morphologically the capitellum, tubercle and neck of a cervical costal element (p. 795). The attachment of the dorsal bar to the pediculolaminar junction represents the morphological transverse process and the attachment of the ventral bar to the ventral body represents the capitellar process. page 743 page 744
The vertebral body has a convex anterior surface. The discal margin gives attachment to the anterior longitudinal ligament. The posterior surface is flat or minimally concave, and its discal margins give attachment to the posterior longitudinal ligament. The central area displays several vascular foramina, of which two are commonly relatively larger. These are the basivertebral foramina which transmit basivertebral veins to the anterior internal vertebral veins. The superior discal surface is saddle-shaped, formed by flange-like lips which arise from most of the lateral circumference of the upper margin of the vertebral body; these are sometimes referred to as uncinate or neurocentral lips or processes. The inferior discal surface is also concave: the convexity is produced mainly by a broad projection from the anterior margin which partly overlaps the anterior surface of the intervertebral disc. The discal surfaces of cervical vertebrae are so shaped in order to restrict both lateral and anteroposterior gliding movements during articulation. The paired ligamenta flava extend from the superior border of each lamina below to the roughened inferior half of the anterior surfaces of the lamina above. The superior part of the anterior surface of each lamina is smooth, like the immediately adjacent surfaces of the pedicles, which are usually in direct contact with the dura mater and cervical root sheaths to which they may become loosely attached. The spinous process of the sixth cervical vertebra is larger, and is often not bifid. The superior articular facets, flat and ovoid, are directed superoposteriorly, whereas the corresponding inferior facets are directed mainly anteriorly, and lie nearer the coronal plane than the superior facets. The dorsal rami of the cervical spinal nerves curve posteriorly, close to the anterolateral aspects of the lateral masses, and may actually lie in shallow grooves, especially on the third and fourth pair. The dorsal root ganglion of each cervical spinal nerve lies between the superior and inferior vertebral notches of adjacent vertebrae. The large anterior ramus passes posterior to the vertebral artery, which lies on the concave upper surface of the costal lamella: the concavity of the lamellae increases from the fourth to the sixth vertebra. The fourth to sixth anterior tubercles are elongated and rough for muscle attachment. The sixth is the longest, the carotid tubercle of Chassaignac. The carotid artery can be forcibly compressed in the groove formed by the vertebral bodies and the larger anterior tubercles, especially the sixth. The posterior tubercles are rounded and more laterally placed than the anterior, and all but the sixth are also more caudal; the sixth is at about the same level as the anterior. Muscle attachments The ligamentum nuchae and numerous deep extensors, including semispinalis
thoracis and cervicis, multifidus, spinales and interspinales, are all attached to the spinous processes. Tendinous slips of scalenus anterior, longus capitis and longus colli are attached to the fourth to sixth anterior tubercles. Splenius, longissimus and iliocostalis cervicis, levator scapulae and scalenus posterior and medius are all attached to the posterior tubercles. Shallow anterolateral depressions on the anterior surface of the body lodge the vertical parts of the longus colli. Ossification Cervical vertebrae ossify according to the standard vertebral pattern described on page 792. Incomplete segmentation ('block vertebra') is common in the cervical spine and most commonly involves the axis and third cervical vertebra.
C1, THE ATLAS (Fig. 45.19)
Figure 45.19 First cervical vertebra (atlas), superior aspect. (Photograph by SarahJane Smith.)
The atlas, the first cervical vertebra, supports the head. It is unique in that it fails to incorporate a centrum, whose expected position is occupied by the dens, a cranial protuberance from the axis. The atlas consists of two lateral masses connected by a short anterior and a longer posterior arch. The transverse ligament retains the dens against the anterior arch. The transverse ligament divides the vertebral canal into two compartments. The anterior third (approximately) of the canal is occupied by the dens. The posterior compartment is occupied by the spinal cord and its coverings, and the cord itself takes up about half of this space (i.e. the cord, like the dens, occupies one third of the canal). The anterior arch is slightly convex anteriorly, and carries a roughened anterior tubercle to which is attached the anterior longitudinal ligament (which is cord-like
at this level). Its upper and lower borders provide attachment for the anterior atlanto-occipital membrane and diverging lateral parts of the anterior longitudinal ligament. The posterior surface of the anterior arch carries a concave, almost circular, facet for the dens. The lateral masses are ovoid, their long axes converging anteriorly. Each bears a kidney-shaped superior articular facet for the respective occipital condyle, which is sometimes completely divided into a larger anterior and a smaller posterior part (Lang 1986). The inferior articular facet of the lateral mass is almost circular and is flat or slightly concave. It is orientated more obliquely to the transverse plane than the superior facet, and faces more medially and very slightly backwards. On the medial surface of each lateral mass is a roughened area which bears vascular foramina and a tubercle for attachment of the transverse ligament. In adults the distance between these tubercles is shorter than the transverse ligament itself, with a mean value of c.16 mm. The posterior arch forms three-fifths of the circumference of the atlantal ring. The superior surface bears a wide groove for the vertebral artery and venous plexus immediately behind, and is variably overhung by the lateral mass; the first cervical nerve intervenes. The flange-like superior border gives attachment to the posterior atlanto-axial membrane, and the flatter inferior border to the highest pair of ligamenta flava. The posterior tubercle is a rudimentary spinous process, roughened for attachment of the ligamentum nuchae. The transverse processes are longer than those of all cervical vertebrae except the seventh (Fig. 45.15). They act as strong levers for the muscles which make fine adjustments to keep the head balanced. Maximum atlantal width varies from 74-95 mm in males and 65-76 mm in females, and this affords a useful criterion for assessing sex in human remains. The apex of the transverse process, which is usually broad, flat and palpable between the mastoid process and ramus of the mandible, is homologous with the posterior tubercle of typical cervical vertebrae: the remaining part of the transverse process consists of the costal lamella. A small anterior tubercle is sometimes visible on the anterior aspect of the lateral mass. The costal lamella is sometimes deficient, which leaves the foramen transversarium open anteriorly. Muscle attachments The superior oblique parts of longus colli are attached on each side of the anterior tubercle. The anterior surface of the lateral mass gives attachment to rectus capitis anterior. Rectus capitis posterior minor is attached just lateral to the posterior tubercle. Rectus capitis lateralis is attached to the transverse process superiorly, and obliquus capitis superior is located more posteriorly. Obliquus capitis inferior is attached laterally on the apex, below which are slips of levator scapulae, splenius cervicis and scalenus medius. Ossification The atlas is commonly ossified from three centres (Fig. 45.20). One appears in each lateral mass at about the seventh week, gradually extending into the posterior arch where they unite between the third and fourth years, usually directly but occasionally through a separate centre. At birth, the anterior arch is
fibrocartilaginous, and a separate centre appears about the end of the first year. This unites with the lateral masses between the sixth and eighth year, the lines of union extending across anterior parts of the superior articular facets. Occasionally the anterior arch is formed by the extension and ultimate union of centres in the lateral masses and sometimes from two lateral centres in the arch itself. The central part of the posterior arch may be absent and replaced by fibrous tissue. Frequently bony spurs arise from the anterior and posterior margins of the groove for the vertebral artery. These are sometimes referred to as ponticles, and they occasionally convert the groove into a foramen. More often the foramen is incomplete superiorly. Rarely the atlas may be wholly or partially assimilated into (fused with) the occiput. page 744 page 745
Figure 45.20 Ossification in the vertebral column. A, Typical vertebra. B, Typical vertebra at puberty. C, Body of a typical vertebra at puberty. D, The atlas. E, The axis. F, Lumbar vertebra.
C2, THE AXIS (Figs 45.21, 45.22) The axis, the second cervical vertebra, acts as an axle for rotation of the atlas and head around the strong dens (odontoid process), which projects cranially from the superior surface of the body. The dens is conical in shape with a mean length of 15 mm in adults. It may be tilted a little, up to 14°, posteriorly, or, less often, anteriorly on the body of the axis: it may also tilt laterally up to 10°. The posterior surface bears a broad groove for the transverse ligament which is covered in cartilage. The apex is pointed, and from this point arises the apical ligament. The alar ligaments are attached to the somewhat flattened posterolateral surfaces above the groove for the transverse ligament. The anterior surface bears an ovoid articular facet for the anterior arch of the atlas, and the surface is pitted by many vascular foramina, which are most numerous near the apex.
Figure 45.21 Second cervical vertebra (axis), superior aspect. (Photograph by SarahJane Smith.)
The body consists of less compact bone than the dens. It is composite, and consists of the partly fused centra of the atlas and axis, and a rudimentary disc (synchondrosis) between them which usually remains detectable deep within the body of the axis throughout life. Large ovoid articular facets are present on either side of the dens at the junction of the body and neural arch: they are flat or slightly convex for articulation with the masses of the atlas. The facets lie in a plane anterior to the plane of the intercentral (Luschka) articulations, with which they are, in part, homologous. The somewhat triangular downward projecting anterior border gives attachment to the anterior longitudinal ligament. Posteriorly, the lower border receives the posterior longitudinal ligament and the membrana tectoria (pp. 754 and 761). The pedicles are stout, and the superior surface carries part of the superior articular facet, which also projects laterally and downwards on to the transverse process. The anterolateral surface is deeply grooved by the vertebral artery, running beneath the thin lateral part of the inferior surface of the superior articular facet, which can become quite thin. The inferior surface of each pedicle bears a deep, smooth inferior intervertebral notch, in which the large root sheath of the third cervical nerve lies. The interarticular part of the pedicle is short and lies between the relatively small posterior articular process (which is located at the pediculolaminar junction and bearing a small anteriorly facing facet) and the superior articular surface. The transverse process is pointed, projects inferiorly and laterally, and arises from
the pediculolaminar junction and the lateral aspect of the interarticular area of the pedicle. The rounded tip is homologous with the posterior tubercle of a typical cervical vertebrae. The foramen transversarium is directed laterally as the vertebral artery turns abruptly laterally under the superior articular facet. Small anterior tubercles may be present near the junction of the costal lamella with the body. The laminae are thick, and give attachment to the ligamenta flava.
page 745 page 746
Figure 45.22 Second cervical vertebra (axis), lateral aspect. (Photograph by SarahJane Smith.)
The spinous process is large, with a bifid tip and a broad base, which is concave inferiorly. The ligamentum nuchae is attached to the apical notch. Muscle attachments The anterior surface of the body carries a deep depression on each side for the attachment of the vertical part of longus colli. Levator scapulae, scalenus medius and splenius cervicis are all attached to the tips of the transverse processes and the intertransverse muscles are attached to their upper and lower surfaces (p. 768). The lateral surfaces of the spinous process give origin to obliquus capitis inferior, and rectus posterior major is attached a little more posteriorly. The inferior concavity of the process receives semispinalis and spinalis cervicis, multifidus more deeply, and the interspinales near the apex. Arterial supply Small branches arise mainly from the vertebral artery at the level of the intervertebral foramen for the third cervical nerve and form paired anterior and posterior longitudinal channels, branches of which enter the dens near the base and near the apex. The anterior channel also receives numerous twigs from nearby branches of the external carotid artery via branches to longus colli and the
ligaments of the apex, hence avascular necrosis does not occur after fracture of the base of the dens. Ossification The axis is ossified from five primary and two secondary centres (Fig. 45.20). The vertebral arch has two primary centres and the centrum one, as in a typical vertebra. The former appear about the seventh or eighth week, and that for the centrum about the fourth or fifth month. The dens is largely ossified from bilateral centres, appearing about the sixth month and joining before birth to form a conical mass, deeply cleft above by cartilage. This cuneiform cartilage forms the apex of the odontoid process. A centre appears in it which shows considerable individual variation in both time of appearance and time of fusion to the rest of the dens: it most often appears between five and eight years, but sometimes even later, fusing with the main mass about the twelfth year. The cartilage was thought to be part of the cranial sclerotomal half of the first cervical segment or pro-atlas. It has also been suggested that the apical centre for the dens is itself derived from the pro-atlas, which may also contribute to lateral atlantal masses. The dens is separated from the body by a cartilaginous disc, the circumference of which ossifies while its centre remains cartilaginous until old age; possible rudiments of adjacent epiphyses of atlas and axis may occur in the disc. A thin epiphyseal plate is formed inferior to the body around puberty. Ossification may sometimes be incomplete. Thus the apical cuneiform centre may fail to fuse with the dens, or the dens itself may fail to fuse with the body, instead forming an os odontoideum. Some believe that this results from old unrecognized trauma rather than ossification failure. Interposition of the transverse ligament may prevent union of fractures through the base of the dens. Hypoplasia of the dens is usually accompanied by atlanto-occipital assimilation and basilar invagination. Abnormalities of the dens are common, and can result in atlantoaxial subluxation. In some skeletal dysplasias there is abnormal ossification in which the dens ossifies separately and much later than the atlantal centrum. This is probably a result of abnormal mobility in the cartilaginous anlage, and normal ossification may be restored if motion is prevented by surgical fusion.
C7, THE SEVENTH CERVICAL VERTEBRA (Fig. 45.23) The seventh cervical vertebra, the vertebra prominens, has a long spinous process which is visible at the lower end of the nuchal furrow. It ends in a prominent tubercle for the attachment of the ligamentum nuchae, and the muscles detailed below. The thick and prominent transverse processes lie behind and lateral to the transverse foramina. The latter transmit vertebral veins, but not the vertebral artery, and each is often divided by a bony spicule. The costal lamella is relatively thin and may be partly deficient. It is grooved superiorly for the anterior ramus of the seventh cervical nerve, and usually carries a small and inconspicuous anterior tubercle. The posterior tubercle is prominent. The suprapleural membrane is attached to the anterior border of the transverse process. The costal lamella of the transverse process may be separate as a cervical rib. The foramina transversaria may be asymmetrical: sometimes one is absent if the
costal lamella is undeveloped.
Figure 45.23 Seventh cervical vertebra, superior aspect. (Photograph by Sarah-Jane Smith.)
Muscle attachments Trapezius, spinalis capitis, semispinalis thoracis, multifidus and interspinales all attach to the tubercle of the spinous process. The anterior border of the transverse process receives the attachment of scalenus minimus (pleuralis), when present. The first pair of levatores costarum is attached to the transverse processes. Ossification Ossific centres for the costal processes appear about the sixth month and join the body and transverse processes between the fifth and sixth years; they may remain separate and grow anterolaterally as cervical ribs. Separate ossific centres may, on occasion, also occur in the costal processes of the fourth to sixth cervical vertebrae.
© 2008 Elsevier
THORACIC VERTEBRAE THORACIC VERTEBRAE IN GENERAL AND CHANGES WITH DESCENDING LEVEL (Figs 45.11, 45.24, 45.25) All thoracic vertebral bodies display lateral costal facets and all but the lowest two or three transverse processes also have facets. The facets articulate with the head of the rib (costocapitular facet) and its tubercle (costotubercular facet) respectively. The first and ninth to twelfth vertebrae also have atypical features, but except for relatively minor details the rest are alike. The body is typically a waisted cylinder except where the vertebral foramen encroaches, and transverse and anteroposterior dimensions are almost equal. On each side there are two costal facets (which are really demifacets): the superior and usually larger pair at the upper border are anterior to the pedicles, while the inferior pair at the lower border are anterior to the vertebral notches. The vertebral foramen is small and circular, so the pedicles do not diverge as they do in cervical vertebrae: the thoracic spinal cord is smaller and more circular than the cervical cord. The laminae are short, thick and broad, and overlap from above downwards. The spinous process slants downward. The thin and almost flat superior articular processes project from the pediculolaminar junctions and face posteriorly and a little superolaterally. The inferior processes project down from the laminae and their facets are directed forwards and a little superomedially. The large, club-like transverse processes also project from the pediculolaminar junctions. Each passes posterolaterally and bears, near its tip, anterior oval facets for articulation with the tubercle of the corresponding rib. page 746 page 747
Figure 45.24 Fourth thoracic vertebra, lateral aspect. (Photograph by Sarah-Jane Smith.)
Figure 45.25 The first, ninth, tenth, eleventh and twelfth thoracic vertebrae, lateral aspect.
The bodies of upper thoracic vertebrae gradually change from cervical to thoracic in type, and the lower change from thoracic to lumbar. The body of the first is typically cervical, its transverse diameter being almost twice the anteroposterior; the second retains a cervical shape, but its two diameters differ less. The third body is the smallest, and has a convex anterior aspect unlike the flattened first and second thoracic vertebrae. The remaining bodies increase in size and, because of its increased anteroposterior diameter, the fourth is typically 'heartshaped'. The fifth to eighth increase their anteroposterior dimension but change little transversely. These four, in transverse section, are asymmetrical, their left sides being flattened by pressure of the thoracic aorta. The rest increase more rapidly in all measurements, so that the twelfth body resembles that of a typical lumbar vertebra. These modifications may contribute to the greater range of
flexion-extension seen at the cervical and lumbar ends of the thoracic vertebral column. The anterior and posterior longitudinal ligaments are attached to the borders of the bodies, and around the margins of the costal facets there are the capsular and radiate ligaments of the costovertebral joints. Thoracic pedicles show a successive caudal increase in thickness. The superior vertebral notch is recognizable only in the first thoracic vertebra, whereas the inferior notch is deep in all. Ligamenta flava are attached at the upper borders and lower anterior surfaces of the laminae. Thoracic transverse processes shorten in caudal succession. In the upper five or six vertebrae the costal facets are concave and face anterolaterally, and at lower levels the facets are flatter and face superolaterally and slightly forwards. The costotransverse ligament is attached to the anterior surface medial to the facet; the lateral costotransverse ligament is attached to its tuberculated apex and the superior costotransverse ligament is attached to its lower border. Thoracic spines overlap from the fifth to the eighth vertebra, whose spine is the longest and most oblique. Supraspinous and interspinous ligaments are attached to the spines. A change in orientation of articular processes from thoracic to lumbar type usually occurs at the eleventh thoracic vertebra, but sometimes at the twelfth or tenth. In the transitional vertebra the superior articular processes are thoracic, and face posterolaterally, while the inferior are transversely convex and face anterolaterally. The transitional vertebra marks the site of a sudden change of mobility from predominantly rotational to predominantly flexion-extension. Muscle attachments Longus colli arises from the upper three thoracic vertebral bodies, lateral to the anterior longitudinal ligament, and psoas major and minor arise from the sides of the twelfth near its lower border. Upper and lower borders of the transverse processes provide attachment for the intertransverse muscles or their fibrous vestiges. The posterior surfaces of the transverse processes provide attachment for the deep dorsal muscles, and levator costae is attached posteriorly on the apex. Trapezius, rhomboid major and minor, latissimus dorsi, serratus posterior superior and inferior and many deep dorsal muscles are attached to the spines. Rotatores attach to the posterior aspects of the laminae. Ossification Thoracic vertebrae all ossify according to the standard vertebral pattern described on page 792.
T1 (Fig. 45.25) The first thoracic vertebra resembles a cervical vertebra in its body, both in shape and in the distinctive posterolateral 'lipping' which forms the anterior border of the superior vertebral notch. There are circular superior costal facets for articulation with the whole facet on the head of the first rib. The smaller, semilunar inferior facets articulate with a demifacet on the head of the second rib. The upper costal
facet is often incomplete, in which case the first rib articulates with the seventh cervical vertebra and the intervening disc. A small, deep depression often occurs below the facet. The long, thick spine is horizontal and commonly as prominent as that of the seventh cervical vertebra.
T9 (Fig. 45.25) The ninth thoracic vertebra is otherwise typical, but it often fails to articulate with the tenth ribs, in which case the inferior demifacets are absent.
T10 (Fig. 45.25) The tenth thoracic vertebra only articulates with the tenth pair of ribs, so that superior facets only appear on the body. These are usually large and semilunar, or oval when the tenth ribs fail to articulate with the ninth vertebra and intervening disc. The transverse process may or may not bear a facet for the tenth rib tubercle.
T11 (Fig. 45.25) The eleventh thoracic vertebra articulates only with the heads of the eleventh ribs. The circular costal facets are close to the upper border of the body and extend onto the pedicles. The small transverse processes lack articular facets. The eleventh and twelfth thoracic spinous processes are triangular, with blunt apices, a horizontal lower and an oblique upper border.
T12 (Fig. 45.25) page 747 page 748
The twelfth thoracic vertebra articulates with the heads of the twelfth ribs by circular facets somewhat below the upper border, spreading on to the pedicles. The body is large and the vertebra has some lumbar features. The transverse process is replaced by three small tubercles: the superior is largest, projects upwards and corresponds to a lumbar mammillary process, though it does not lie as close to the superior articular process; the lateral tubercle is the homologue of a transverse process; the inferior is the homologue of a lumbar accessory process. The superior and inferior processes are surprisingly long in some specimens.
© 2008 Elsevier
LUMBAR VERTEBRAE (Figs 45.26, 45.27, 45.28, 45.29) LUMBAR VERTEBRAE IN GENERAL
Figure 45.26 First lumbar vertebra, superior aspect. (Photograph by Sarah-Jane Smith.)
Figure 45.27 Lumbar vertebra, lateral aspect.
Figure 45.28 Median sagittal MRI lumbar spine. (By kind permission from Dr Justin Lee, Chelsea and Westminster Hospital, London.)
page 748 page 749
Figure 45.29 Paramedian sagittal MRI lumbar spine, showing intervertebral foramina. (By kind permission from Dr Justin Lee, Chelsea and Westminster Hospital, London.)
The five lumbar vertebrae are distinguished by their large size and absence of costal facets and transverse foramina. The body is wider transversely, and normally is deeper in front. The vertebral foramen is triangular, larger than at thoracic levels but smaller than at cervical levels. The pedicles are short. The spinous process is almost horizontal, quadrangular and thickened along its posterior and inferior borders. The superior articular processes bear vertical concave articular facets facing posteromedially, with a rough mammillary process on their posterior borders. The inferior articular processes have vertical convex articular facets which face anterolaterally. The transverse processes are thin and long, except on the more substantial fifth pair. A small accessory process marks the posteroinferior aspect of the root of each transverse process. The accessory
and mammillary processes are linked by a fine ligament, the mammillo-accessory ligament, which is sometimes ossified, and beneath which runs the medial branch of the dorsal primary ramus of the spinal nerve. Strong paired pedicles arise posterolaterally from each body near its upper border. Superior vertebral notches are shallow and the inferior ones are deep. The laminae are broad and short, but do not overlap as much as those of the thoracic vertebrae. The fifth spine is the smallest, and its apex is often rounded and down-turned. Upper lumbar superior articular processes are further apart than inferior ones, but the difference is slight in the fourth and negligible in the fifth. The articular facets are reciprocally concave (superior) and convex (inferior), which allows flexion, extension, lateral bending and some degree of rotation. There are sex differences in the angle of inclination and depth of curvature of the articular facets. The facets are sometimes asymmetrical. Transverse processes, except the fifth, are anteroposteriorly compressed and project posterolaterally. The lower border of the fifth transverse process is angulated, passes laterally and then superolaterally to a blunt tip, and the whole process presents a greater upward inclination than the fourth. The angle on the inferior border may represent the tip of the costal element and the lateral end the tip of the true transverse process. The lumbar transverse processes increase in length from first to third and then shorten. The fifth pair incline both upwards and posterolaterally. The costal element is incorporated in the mature transverse process. The first lumbar vertebral foramen contains the conus medullaris of the spinal cord, while lower foramina contain the cauda equina and spinal meninges. Variation occurs in the sagittal and coronal dimensions of the lumbar vertebral canal, both within and between normal populations. Muscle and fascial attachments (See also p. 734.) Upper and lower borders of lumbar bodies give attachment to the anterior and posterior longitudinal ligaments (p. 754). The upper bodies (three on the right, two on the left) give attachments to the crura of the diaphragm lateral to the anterior longitudinal ligament. Posterolaterally, psoas major is attached to the upper and lower margins of all the lumbar bodies, and between them, tendinous arches carry its attachments across their concave sides (Fig. 2.1). The posterior lamella of the thoracolumbar fascia, erectores spinae, spinales thoracis, multifidi, interspinal muscles and ligaments, and supraspinous ligaments are all attached to spinous processes. All lumbar transverse processes present a vertical ridge on the anterior surface, nearer the tip, which marks the attachment of the anterior layer of the thoracolumbar fascia, and separates the surface into medial and lateral areas for psoas major and quadratus lumborum respectively. The middle layer of the fascia is attached to the apices of the transverse processes; the medial and lateral arcuate ligaments attach to the apices of the first pair, and the iliolumbar ligament attaches to the apices of the fifth pair. Posteriorly the transverse processes are covered by deep dorsal muscles, and fibres of longissimus thoracis are attached to them and to their accessory processes. The ventral lateral intertransverse muscles are attached to their upper and lower borders, while the dorsal attach cranially to the accessory process and caudally to the upper border of the transverse process. The mammillary process, homologous
with the superior tubercle of the twelfth thoracic vertebra, gives attachment to multifidus and the medial intertransverse muscle. The latter also attaches to the accessory process, which is sometimes difficult to identify. Ossification (Fig. 45.20) Lumbar vertebrae ossify according to the standard vertebral pattern described on page 792, but also have two additional centres for the mammillary processes. A pair of scale-like epiphyses usually appear on the tips of the costal elements of the fifth lumbar vertebra.
L5 (Fig. 45.30) The fifth lumbar vertebra has a massive transverse process which is continuous with the whole of the pedicle and encroaching on the body. The body is usually the largest and markedly deeper anteriorly, so contributing to the lumbosacral angle. Segmentation anomalies (sacralization) are considered below with the sacrum. The costal element of the first lumbar vertebra may form a short lumbar rib, which articulates with the transverse process, but not usually with the body, of the vertebra.
Figure 45.30 Fifth lumbar vertebra, superior aspect.
© 2008 Elsevier
SACRUM (Figs 45.31, 45.32, 45.33, 45.34, 45.35) The sacrum is a large, triangular fusion of five vertebrae and forms the posterosuperior wall of the pelvic cavity, wedged between the two innominate bones. Its blunted, caudal apex articulates with the coccyx and its superior, wide base with the fifth lumbar vertebra at the lumbosacral angle. It is set obliquely and curved longitudinally, its dorsal surface is convex, and the pelvic surface is concave. This ventral curvature increases pelvic capacity. Between base and apex are dorsal, pelvic and lateral surfaces and a sacral canal. In childhood, individual sacral vertebrae are connected by cartilage, and the adult bone retains many vertebral features. The sacrum consists of trabecular bone enveloped by a shell of compact bone of varying thickness. Base The base is the upper surface of the first sacral vertebra, the least modified from the typical vertebral plan. The body is large and wider transversely, and its anterior projecting edge is the sacral promontory. The vertebral foramen is triangular, its pedicles are short and diverge posterolaterally. The laminae are oblique, inclining down posteromedially to meet at a spinous tubercle. The superior articular processes project cranially, with concave articular facets directed posteromedially to articulate with the inferior articular processes of the fifth lumbar vertebra. The posterior part of each process projects backwards and its lateral aspect bears a rough area homologous with a lumbar mammillary process. The transverse process is much modified as a broad, sloping mass which projects laterally from the body, pedicle and superior articular process. It is formed by the fusion of the transverse process and the costal element to each other and to the rest of the vertebra, and forms the upper surface of the sacral lateral mass or ala. Terminal fibres of the anterior and posterior longitudinal ligaments are attached to the ventral and dorsal surfaces of the first sacral body. Its upper laminar borders receive the lowest pair of ligamenta flava. The ala is smooth superiorly, concave medially and rough laterally, and covered almost entirely by psoas major. The smooth area is grooved obliquely by the lumbosacral trunk. The rough area is for the lower band of the iliolumbar ligament, which lies lateral to the fifth lumbar spinal nerve and to the anterior sacroiliac ligament. Pelvic surface
page 749 page 750
Figure 45.31 Sacrum, anterior (pelvic) surface. (Photograph by Sarah-Jane Smith.)
The anteroinferior pelvic surface is vertically and transversely concave, but the second sacral body may produce a convexity. Four pairs of pelvic sacral foramina communicate with the sacral canal through intervertebral foramina, and transmit ventral rami of the upper four sacral spinal nerves. The large area between the right and left foramina, which is formed by the flat pelvic aspects of the sacral bodies, bears evidence of their fusion at four transverse ridges. The longitudinal bars between the foramina are costal elements, which fuse to the vertebrae. Lateral to the foramina the costal elements unite. Posteriorly they unite with the transverse processes to form the lateral part of the sacrum, which expands basally as the ala. The first three sacral ventral rami emerge from the pelvic sacral foramina and pass anterior to piriformis. The sympathetic trunks descend in contact with bone, medial to the foramina, as do the median sacral vessels in the midline. Lateral to the foramina, lateral sacral vessels are related to bone. Ventral surfaces of the first, second, and part of the third sacral bodies are covered by parietal peritoneum and crossed obliquely, left of the midline, by the attachment of the sigmoid mesocolon. The rectum is in contact with the pelvic surfaces of the third to fifth sacral vertebrae and with the bifurcation of the superior rectal artery between the rectum and third sacral vertebra. Dorsal surface
Figure 45.32 Sacrum, posterior (dorsal) surface. (Photograph by Sarah-Jane Smith.)
The posterosuperior aspect of the dorsal surface bears a raised, interrupted, median sacral crest with four (sometimes three) spinous tubercles which represent fused sacral spines. Below the fourth (or third) tubercle there is an arched sacral hiatus in the posterior wall of the sacral canal. This hiatus is produced by the failure of the laminae of the fifth sacral vertebra to meet in the median plane, and as a result the posterior surface of the body of that vertebra is exposed on the dorsal surface of the sacrum. Flanking the median crest, the posterior surface is formed by fused laminae, and lateral to this are four pairs of dorsal sacral foramina. Like the pelvic foramina, they lead into the sacral canal through intervertebral foramina, and each transmits the dorsal ramus of a sacral spinal nerve. Medial to the foramina, and vertically below each articular process of the first sacral vertebra, is a row of four small tubercles, which collectively constitute the intermediate sacral crest. These are sometimes termed articular tubercles, and represent fused contiguous articular processes. The inferior articular processes of the fifth sacral vertebra are free and project downwards at the sides of the sacral hiatus as sacral cornua, connected to coccygeal cornua by intercornual ligaments. The interrupted roughened crest to the lateral side of the dorsal sacral foramina is the lateral sacral crest which is formed by fused transverse processes, whose apices appear as a row of transverse tubercles. The upper three sacral spinal dorsal rami pierce multifidus as they emerge via dorsal foramina. Lateral surface The lateral surface is a fusion of transverse processes and costal elements. It is wide above, and rapidly narrows in its lower part. The broad upper part bears an auricular surface for articulation with the ilium, and the area posterior to this is rough and deeply pitted by the attachment of ligaments. The auricular surface, borne by costal elements, is like an inverted letter L. The shorter, cranial limb is restricted to the first sacral vertebra; the caudal limb descends to the middle of
the third. Beyond this the lateral surface is non-articular and reduced in breadth. Caudally it curves medially to the body of the fifth sacral vertebra at the inferior lateral angle, beyond which the surface becomes a thin lateral border. A variable accessory sacral articular facet sometimes occurs, posterior to the auricular surface. page 750 page 751
Figure 45.33 A, Sacrum, lateral aspect. B, Median sagittal section through the sacrum. (A, photograph by Sarah-Jane Smith.)
Figure 45.34 Sacrum, superior aspect (base). (Photograph by Sarah-Jane Smith.)
The auricular surface is covered by hyaline cartilage, and formed entirely by costal elements. It shows cranial and caudal elevations and an intermediate depression, behind which a third elevation is visible in the elderly. The surface becomes more corrugated with age. The rough area behind the auricular surface shows two or three marked depressions for the attachment of strong interosseous sacroiliac ligaments. Below the auricular surface the sacrotuberous and sacrospinous ligaments are attached between gluteus maximus dorsally and coccygeus ventrally.
Apex The apex is the inferior aspect of the fifth sacral vertebral body, and bears an oval facet for articulation with the coccyx. Sacral canal (Fig. 45.33B) The sacral canal is formed by sacral vertebral foramina, and is triangular in section. Its upper opening, seen on the basal surface, appears to be set obliquely. The inclination of the sacrum means that it is directed cranially in the standing position. Each lateral wall presents four intervertebral foramina, through which the canal is continuous with pelvic and dorsal sacral foramina. Its caudal opening is the sacral hiatus. The canal contains the cauda equina and the filum terminale, and the spinal meninges. Opposite the middle of the sacrum, the subarachnoid and subdural spaces close: the lower sacral spinal roots and filum terminale pierce the arachnoid and dura mater at that level. The filum terminale emerges below the sacral hiatus and passes downwards across the dorsal surface of the fifth sacral vertebra and sacrococcygeal joint to reach the coccyx. The fifth sacral spinal nerves also emerge through the hiatus medial to the sacral cornua, and groove the lateral aspects of the fifth sacral vertebra. Muscle attachments The pelvic surface gives attachment to piriformis in its second to fourth segments, to iliacus superolaterally, and to coccygeus inferolaterally. The dorsal surface gives attachment to the aponeurosis of erector spinae along a U-shaped area of spinous and transverse tubercles, covering multifidus which occupies the enclosed area (Fig. 45.34). On the lateral border below the auricular surface, gluteus maximus is attached dorsal and coccygeus is attached ventral to the sacrotuberous and sacrospinous ligaments. Ossification (Fig. 45.36)
page 751 page 752
page 752 page 753
Figure 45.35 A and B, Lateral radiographs of lumbosacral vertebral column in an adult male aged 26 years. C, Anteroposterior radiograph of lumbosacral vertebral column in a young adult male aged 22 years. (Provided by Shaun Gallagher, GKT School of Medicine, London; photographs by Sarah-Jane Smith.) D, High resolution computed tomogram through posterior abdominal wall at the level of the body of the fourth lumbar vertebra, showing zygapophyseal joints between fourth and fifth lumbar vertebrae.
page 753 page 754
Figure 45.36 Ossification of the sacrum and coccyx. A, At birth. B, The base of the sacrum of a child about four years old. C, At the twenty-fifth year: epiphyseal plates for each lateral surface are marked by asterisks. D, E, The epiphyses of the costal and transverse process of the sacrum at the eighteenth year.
The sacrum resembles typical vertebrae in the ossification of its segments. Primary centres for the centrum and each half vertebral arch appear between the tenth and twentieth weeks. Primary centres for the costal elements of the upper three or more segments appear superolateral to the pelvic sacral foramina, between the sixth and eighth prenatal months. Each costal element unites with its half vertebral arch between the second and fifth years, and the conjoined element so formed unites anteriorly with the centrum and posteriorly with its opposite fellow at about the eighth year. Thereafter the upper and lower surfaces of each sacral body are covered by an epiphyseal plate of hyaline cartilage which is separated from its neighbour by the fibrocartilaginous precursor of an intervertebral disc. Laterally, successive conjoined vertebral arches and costal elements are separated by hyaline cartilage; a cartilaginous epiphysis, sometimes divided into upper and lower parts, develops on each auricular and adjacent lateral surface. Soon after puberty the fused vertebral arches and costal elements of adjacent vertebrae begin to coalesce from below upwards. At the same time individual epiphyseal centres develop for the upper and lower surfaces of bodies, spinous tubercles, transverse tubercles and costal elements. The costal epiphyseal centres appear at the lateral extremities of the hyaline cartilages between adjacent costal elements; two anterior and two posterior centres appear in each of the intervals between the first, second and third sacral vertebrae. Ossification spreads from these into the auricular epiphyseal plates. One costal epiphyseal centre, placed anteriorly, occurs in each remaining interval and from them ossification spreads to the epiphyseal plate covering the lower part of the lateral surface of the sacrum. Sacral bodies unite at their adjacent margins after the twentieth year, but the central and greater part of each intervertebral disc remains unossified up to or beyond middle life. Variants The sacrum may contain six vertebrae, by development of an additional sacral element or by incorporation of the fifth lumbar or first coccygeal vertebrae. Inclusion of the fifth lumbar vertebra (sacralization) is usually incomplete and limited to one side. In the most minor degree of the abnormality a fifth lumbar transverse process is large and articulates, sometimes by a synovial joint, with the sacrum at the posterolateral angle of its base. Reduction of sacral constituents is less common but lumbarization of the first sacral vertebra does occur: it remains partially or completely separate. The bodies of the first two sacral vertebrae may remain unfused when the lateral masses are fused. The
dorsal wall of the sacral canal may be variably deficient, due to imperfect development of laminae and spines. Orientation of the superior sacral articular facets displays wide variation, as does the sagittal curvature of the sacrum. Asymmetry (facet tropism) of the superior facets alters the relation between the planes of the two lumbosacral facet joints. Sex differences in sacra Sex differences in sacra are described on page 1428.
© 2008 Elsevier
COCCYX (Fig. 45.37) The coccyx is a small triangular bone often asymmetrical in shape. It usually consists of four fused rudimentary vertebrae: the number varies from three to five, the first is sometimes separate. The bone is directed downwards and ventrally from the sacral apex: its pelvic surface is tilted upwards and forwards, its dorsum downwards and backwards. Orientation varies with mobility and between individuals. The base or upper surface of the first coccygeal vertebral body has an oval, articular facet for the sacral apex. Posterolateral to this, two coccygeal cornua project upwards to articulate with sacral cornua: they are homologues of the pedicles and superior articular processes of other vertebrae. A rudimentary transverse process projects superolaterally from each side of the first coccygeal body and may articulate or fuse with the inferolateral sacral angle, completing the fifth sacral foramina.
Figure 45.37 The coccyx. A, Anterior (pelvic) aspect. B, Posterior (dorsal) aspect.
The second to fourth coccygeal vertebrae diminish in size and are usually mere fused nodules. They represent rudimentary vertebral bodies, though the second may show traces of transverse processes and pedicles. The gap between the fifth sacral body and the articulating cornua represents, on each side, an intervertebral foramen which transmits the fifth sacral spinal nerve. The dorsal ramus descends behind the rudimentary transverse process, and the ventral ramus passes anterolaterally between the transverse process and sacrum. Muscle and ligament attachments The lateral parts of the pelvic surface, including the rudimentary transverse processes, give attachment to the levatores ani and coccygei. The anterior
sacrococcygeal ligament is attached to the front of the first and sometimes second coccygeal vertebral bodies (Fig. 111.11). The cornua give attachment to the intercornual ligaments. The lateral sacrococcygeal ligament connects the transverse process to the inferolateral sacral angle. Gluteus maximus is attached to the dorsal surface, and both levator ani and sphincter ani externus are attached to the tip of the bone. The median area gives attachment to the deep and superficial posterior sacrococcygeal ligaments, the superficial descending from the margins of the sacral hiatus and sometimes closing the sacral canal. The filum terminale, which is situated between the two ligaments, blends with them on the dorsum of the first coccygeal vertebra. Ossification Each coccygeal segment is ossified from one primary centre. A centre in the first segment appears about birth and its cornua may soon ossify from separate centres. Remaining segments ossify at wide intervals up to the twentieth year or later. Segments slowly unite: union between the first and second is frequently delayed until 30 years. The coccyx often fuses with the sacrum in later decades, especially in females.
© 2008 Elsevier
LIGAMENTS OF THE VERTEBRAL COLUMN LIGAMENTUM NUCHAE (Fig. 45.38) The ligamentum nuchae is a bilaminar fibroelastic intermuscular septum which is often considered homologous with, but structurally distinct from, the supraspinous and interspinous ligaments in the neck. Its dense bilateral fibroelastic laminae are separated by a tenuous layer of areolar tissue and the laminae are blended at its posterior free border. This border is superficial and extends from the external occipital protuberance to the spine of C7. The fibroelastic laminae are attached to the median part of the external occipital crest, the posterior tubercle of C1 and the medial aspects of the bifid spines of cervical vertebrae, as a septum for the bilateral attachment of cervical muscles and their sheaths. There is also a midline attachment to the posterior spinal dura at atlanto-occipital and atlanto-axial levels (Dean & Mitchell 2002). In bipeds the ligamentum nuchae is the reduced representative of a much thicker, complex elastic ligament which in quadrupeds aids suspension of the head and controls its flexion.
ANTERIOR LONGITUDINAL LIGAMENT (Fig. 45.41B) page 754 page 755
Figure 45.38 The ligamentum nuchae. (By permission from Kiss F, Szentagothai J 1964 Atlas of Human Anatomy. Oxford: Pergamon Press.)
The anterior longitudinal ligament is a strong band extending along the anterior surfaces of the vertebral bodies. It is broader caudally, and thicker and narrower in thoracic than in cervical and lumbar regions. It is also relatively thicker and narrower opposite vertebral bodies than at the levels of intervertebral symphyses. It extends from the basilar part of the occipital bone to the anterior tubercle of C1
and the front of the body of C2, then continues caudally to the front of the upper sacrum. Its longitudinal fibres are strongly adherent to the intervertebral discs, hyaline cartilage end-plates and margins of adjacent vertebral bodies, and are loosely attached at intermediate levels of the bodies, where the ligament fills their anterior concavities, flattening the vertebral profile. At these various levels ligamentous fibres blend with the subjacent periosteum, perichondrium and periphery of the annulus fibrosus. The anterior longitudinal ligament has several layers. The most superficial fibres are the longest and extend over three or four vertebrae, the intermediate extend between two or three, and the deepest from one body to the next. Laterally, short fibres connect adjacent vertebrae.
POSTERIOR LONGITUDINAL LIGAMENT (Fig. 45.39) The posterior longitudinal ligament lies on the posterior surfaces of the vertebral bodies in the vertebral canal, attached between the body of C2 and the sacrum, and continuous with the membrana tectoria above (p. 761). Its smooth glistening fibres, attached to intervertebral discs, hyaline cartilage end-plates and adjacent margins of vertebral bodies, are separated between attachments by basivertebral veins and the venous rami which drain them into anterior internal vertebral plexuses. At cervical and upper thoracic levels the ligament is broad and of uniform width, but in lower thoracic and lumbar regions it is denticulated, narrow over vertebral bodies and broad over discs. Its superficial fibres bridge three or four vertebrae, while deeper fibres extend between adjacent vertebrae as perivertebral ligaments, which are close to and, in adults, fused with, the annulus fibrosus of the intervertebral disc. The layers are more distinct in the immediate postnatal years.
Figure 45.39 The posterior longitudinal ligament in the lumbar region.
Figure 45.40 Ligamenta flava (anterior aspect) in the lumbar region.
LIGAMENTA FLAVA (Figs 45.39, 45.40) The ligamenta flava connect laminae of adjacent vertebrae in the vertebral canal. Their attachments extend from facet joint capsules to the point where laminae fuse to form spines. Here their posterior margins meet and are partially united; the intervals between them admit veins which connect the internal and posterior external vertebral venous plexuses. Their predominant tissue is yellow elastic tissue, whose almost perpendicular fibres descend from the lower anterior surface of one lamina to the posterior surface and upper margin of the lamina below. The anterior surface of the ligaments is covered by a fine, continuous smooth lining membrane (Newell 1999). The ligaments are thin, broad and long in the cervical region, thicker in the thoracic and thickest at lumbar levels. They arrest separation of the laminae in spinal flexion, preventing abrupt limitation, and also assist restoration to an erect posture after flexion, perhaps protecting discs from injury. page 755 page 756
INTERSPINOUS LIGAMENTS (Fig. 45.41B) Interspinous ligaments are thin, almost membranous and connect adjoining spines, their attachments extending from the root to the apex of each. They meet the ligamenta flava in front and the supraspinous ligament behind. The ligaments are narrow and elongated in the thoracic region, broader, thicker and quadrilateral at lumbar levels, and poorly developed in the neck. Some observers designate all cervical interspinous fibres as part of the ligamentum nuchae, while others regard them as distinct structures. Their fibres run obliquely posterosuperiorly from the upper border of one spine to the lower border of that immediately above.
SUPRASPINOUS LIGAMENT (Fig. 45.41B)
Figure 45.41 A, Schematic representation of the main structural features of an intervertebral disc. For clarity the number of fibrocartilaginous laminae has been greatly reduced. Note alternating obliquity of collagen fascicles in adjacent laminae (after Inoue). B, Median sagittal section through upper lumbar vertebral column showing discs and ligaments. (A, after Inoue H 1973 Three-dimensional observation of collagen framework of intervertebral discs in rats, dogs and humans. Arch Histol Jpn 36: 39-56.)
The supraspinous ligament is a strong fibrous cord which connects the tips of spinous process from C7 to the sacrum. It is thicker and broader at lumbar levels, where it is intimately blended with neighbouring fascia, though only lightly attached to the spines of L3-5. The most superficial fibres extend over three or four vertebrae, the deeper span two or three, while the deepest connect adjacent spines and are continuous with interspinous ligaments. Between the spine of C7 and the external occipital protuberance the supraspinous ligament is expanded as the ligamentum nuchae.
INTERTRANSVERSE LIGAMENTS Intertransverse ligaments run between adjacent transverse processes. At cervical levels they consist of a few, irregular fibres which are largely replaced by intertransverse muscles; in the thoracic region they are cords intimately blended
with adjacent muscles; in the lumbar region they are thin and membranous.
LIGAMENTOUS INSTABILITY Damage to the ligaments controlling stability of the column may occur in the absence of evident bony pathology. This is particularly prevalent in inflammatory disease of the upper cervical spine, where rheumatoid arthritis may weaken or destroy the ligaments on which atlanto-axial stability depends (p. 759). The transverse ligament is stronger than the dens, which usually fractures before the ligament ruptures. The alar ligaments are weaker, and combined head flexion and rotation may avulse one or both alar ligaments: rupture of one side results in an increase of about a third in the range of rotation to the opposite side. Pathological softening of the transverse and adjacent ligaments or of the lateral atlanto-axial joints results in atlanto-axial subluxation, which may cause spinal cord injury. Ligamentous damage may also occur in spinal injuries, particularly at cervical levels. Developmental laxity of ligaments may also lead to problems with instability, especially if there is an episode of trauma: this combination is probably responsible for atlanto-axial rotational instability. Laxity of cervical spinal ligaments may be a normal variant in children, and lead to diagnostic difficulties. In radiographs of the upper cervical spine in children aged less than 8, a deceptive appearance of subluxation ('pseudosubluxation') may result from a combination of ligamentous laxity and facet orientation. This usually occurs between C2 and C3, but may occasionally be seen at C3/4. Clinical and other radiological features should facilitate the correct diagnosis.
© 2008 Elsevier
JOINTS All vertebrae from C2 to S1 articulate by secondary cartilaginous joints (symphyses) between their bodies, synovial joints between their articular processes, and fibrous joints between their laminae, transverse and spinous processes. In the cervical region, from C3 to C7, there are also laterally placed articulations between the uncinate or neurocentral processes (p. 742) of the inferior vertebral body and the bevelled lateral border of the superior body at each level. These small joints, the unco-vertebral or neurocentral joints of Luschka, have a synovial element, with articular cartilage and a partial capsule.
INTERVERTEBRAL JOINTS Joints between the vertebral bodies
Joints between vertebral bodies are symphyses. Typical vertebral bodies are united by anterior and posterior longitudinal ligaments and by fibrocartilaginous intervertebral discs between sheets of hyaline cartilage (vertebral end-plates). The ligaments are described on pages 735 and 754. Articulating surfaces: intervertebral discs (Fig. 45.41A)
The intervertebral discs are the chief bonds between the adjacent surfaces of vertebral bodies from C2 to the sacrum. Except at the sites of the uncovertebral (neurocentral) joints of Luschka, disc outlines correspond with the adjacent bodies. Their thickness varies in different regions and within individual discs. Each disc consists of an outer lamellated anulus fibrosus and an inner nucleus pulposus. In cervical and lumbar regions the discs are thicker anteriorly, contributing to the anterior convexity of the vertebral column. In the thoracic region they are nearly uniform, and the anterior concavity is largely due to the vertebral bodies. Discs are thinnest in the upper thoracic region and thickest in the lumbar region. They adhere to thin layers of cartilage on the superior and inferior vertebral surfaces, the vertebral end-plates. The latter do not reach the periphery of the vertebral bodies but are encircled by ring apophyses (p. 735). The end-plates contain both hyaline cartilage and fibrocartilage. The fibrocartilaginous component lies nearer to the disc, and is sometimes considered not to be part of the endplate itself. The fibrocartilaginous components of the end-plates above and below the nucleus pulposus, together with the innermost lamellae of the anulus fibrosus, form a flattened sphere of collagen which surrounds and encloses the nucleus (Fig. 45.42). The overall proportion of fibrocartilage in the end-plate increases with age. While all discs are attached to the anterior and posterior longitudinal ligaments, discs in the thoracic region are additionally tied laterally, by intraarticular ligaments, to the heads of ribs articulating with adjacent vertebrae. Intervertebral discs form about a quarter of the length of the postaxial vertebral column: cervical and lumbar regions make a greater contribution than the thoracic and are thus more pliant. Anulus fibrosus
Figure 45.42 Structure of the vertebral end plate: the collagen fibres of the inner twothirds of the anulus fibrosus sweep around into the vertebral end plate and form its fibrocartilaginous component. The peripheral fibres of the anulus are anchored into the bone of the ring apophysis. (By permission from Bogduk N 1997 Clinical Anatomy of the Lumbar Spine and Sacrum, 3rd edn. Edinburgh: Churchill Livingstone.)
The anulus fibrosus has a narrow outer collagenous zone and a wider inner fibrocartilaginous zone. Its lamellae, which are convex peripherally when seen in vertical section, are incomplete collars. The internal vertical concavity of the lamellae conforms to the surface profile of the nucleus pulposus. In all quadrants of the anulus, about half the lamellae are incomplete; the proportion increases in the posterolateral region. The exact nature of the interlamellar substance remains in some doubt. Posteriorly, lamellae join in a complex manner. Fibres in the rest of each lamella are parallel and run obliquely between vertebrae at about 65° to the vertical. Fibres in successive lamellae cross each other obliquely in opposite directions, thus limiting rotation. The obliquity of fibres in deeper zones varies in different lamellae. Posterior fibres may sometimes be predominantly vertical, which possibly predisposes them to herniation. This standard description of the anulus may not apply at all spinal levels: a recent cadaveric study indicates that the anulus is usually incomplete posteriorly in adult cervical discs (Mercer & Bogduk 1999). Nucleus pulposus The nucleus pulposus is better developed in cervical and lumbar regions and lies between the centre of the disc and its posterior surface. At birth it is large, soft, gelatinous and composed of mucoid material. It contains a few multinucleated notochordal cells and is invaded by cells and collagen fibres from the inner zone of the adjacent anulus fibrosus. Notochordal cells disappear in the first decade, and the mucoid material is gradually replaced by fibrocartilage, derived mainly from the anulus fibrosus and the plates of hyaline cartilage adjoining the vertebral bodies. The nucleus pulposus becomes less differentiated from the remainder of the disc as age progresses, and gradually becomes less hydrated and increasingly fibrous. The type II collagen of the nucleus becomes more like the type I of the anulus as its fibril diameter increases. The quantity of aggregated proteoglycans in the nucleus decreases, while the keratan sulphate/chondroitin sulphate ratio increases. As increased cross-linking occurs between collagen and the proteoglycans the discs lose their water-binding capacity, become stiffer and more liable to injury. Contrary to what was previously thought, it has now been shown that lumbar discs do not decrease in overall height as a part of normal ageing. The anulus gradually loses height as its radial bulge increases, but the nucleus retains height and may increase in convexity as it increasingly indents the end-plate. Loss of trunk height with age results from a decrease in vertebral body depth (Bogduk 1997). When the disc is not loaded, pressure in the nucleus pulposus is low at all ages.
For a review of the structure and function of the human intervertebral disc see Adams et al (2002). Ligaments
The ligaments associated with the joints between the vertebral bodies are described on pages 735 and 754. Vascular supply (See also p. 735.)
Small offshoots of spinal branches of arteries supplying the vertebral column form an anastomosis on the outer surface of the anulus fibrosus and supply its most peripheral fibres. Normal discs are otherwise avascular and are dependent for their nutrition on diffusion from vertebral bone beneath adjacent end-plates and from the peripheral anulus. Vascular and avascular parts differ in their reaction to injury. Venous drainage is via the external and internal vertebral venous plexuses to the intervertebral veins and thence to the larger named veins which drain the vertebral column. Lymphatic drainage of the vertebral column is briefly considered above. Nothing specific is known about the lymphatic drainage of the disc. Innervation
The nerve supply of intervertebral discs has been studied in detail in the lumbar region. The outer third of the anulus is innervated by the sinuvertebral nerves: the anterior anulus is supplied by the sympathetic (grey rami) component rather than by the mixed nerve. In damaged and degenerate discs the nerves may penetrate more centrally into the disc substance. The sinuvertebral nerves are condensations within extensive nerve plexuses which lie on the posterior longitudinal ligament. Similar plexuses have been demonstrated anteriorly, covering the anterior longitudinal ligament, and laterally in the fetus. Each sinuvertebral nerve supplies both the disc at the level of its spinal nerve of origin and the disc one level above. page 757 page 758
Relations and 'at risk' structures
Posterior, lateral and anterior relations of the intervertebral disc are important in the planning of interventional investigative and therapeutic procedures ranging from discography to open disc surgery. The posterolateral surface of the disc forms the anterior boundary of the intervertebral foramen on each side (p. 757), and so is closely related to the spinal nerve and its accompanying vessels. More centrally the disc is related posteriorly to the dura mater covering the spinal cord and the cauda equina. Anterior relations of the discs vary considerably with vertebral level, but important 'at risk' structures include the pharynx and oesophagus, the descending aorta and the inferior vena cava. Laterally, relations change with level, but the parietal pleura in the thorax, and the sympathetic trunk and psoas muscles in the lumbar region, are important examples. Prolapsed intervertebral disc
A prolapsed intervertebral disc most commonly affects the 20-55 year age group, and is most often seen at the L4/5 and lumbosacral levels. It may also affect the cervical discs, particularly at C5/6 and C6/7. The thoracic discs are rarely affected. Acute tearing or chronic degeneration of the posterior lamellae of the anulus fibrosus allows deformation and herniation of the softer nucleus pulposus. The disc most often prolapses just lateral to the posterior longitudinal ligament and can compress one or two spinal nerves unilaterally (Fig. 45.43). Much less commonly, the prolapse is central, in the midline posteriorly. The compression of neural structures may then be bilateral, affecting the cord itself or the whole cauda equina. If the damaged anulus ruptures completely, some of the nuclear tissue may escape into the vertebral and 'root' canals. This sequestrated material may migrate within the canals and cause nerve compression at spinal levels distant from that of the disc rupture. The disc material itself may have an irritative effect
on the spinal nerve.
Figure 45.43 Posterolateral disc prolapse. (By permission from Moore K, Agur AMR 2002 Essential Clinical Anatomy, 2nd edn. Philadelphia: Lippincott Williams and Wilkins.)
Regarding the anatomy of the vertebral canal and intervertebral foramen in relation to disc prolapse, it is important to understand that one or both of two spinal nerves and their roots may be affected by a single prolapse, depending upon the exact site of the prolapse in the horizontal plane. At the level of each disc and foramen, there are two spinal nerves (and their roots) to consider: these are the exiting nerve and the traversing nerve (Macnab & McCulloch 1990) (Fig. 45.44). The nerve usually affected at lumbar levels is the traversing nerve, which crosses the back of the disc on its way to become the exiting nerve at the level below. Thus a lumbosacral (i.e. L5/S1) disc prolapse usually compresses the S1 nerve. However, a prolapse may affect the exiting nerve at its own level. This is especially likely if the prolapse is in the extraforaminal zone of the 'root' canal (p. 735), the so-called 'far lateral' prolapse. At cervical levels, because the roots and nerve leave the vertebral canal almost horizontally, the prolapse usually affects the exiting nerve. This nerve will still bear the number of the vertebra below the affected disc, because cervical nerves exit the canal above the pedicle of their numerically corresponding vertebra. Neurological presentation will include signs and symptoms of spinal nerve damage at the affected level. Thus pain and sensory loss will be dermatomal in distribution. Sensory changes usually precede motor loss. Internal disruption of a lumbar intervertebral disc is more common than disc prolapse, and is now an increasingly recognized cause of back pain. Typically, the nucleus is decompressed and the inner lamellae of the anulus appear to collapse into it.
For more detail on disc pathology and its consequences, see Adams et al (2002). Facet (zygapophyseal) joints
Joints between the vertebral articular processes (zygapophyses) are synovial. They are termed zygapophyseal joints, or in more common clinical usage, facet joints. For a detailed description of these joints, see Bogduk (1997). Articulating surfaces
Facet joints are of the simple (cervical and thoracic) or complex (lumbar) synovial variety: the articulating surfaces are covered in hyaline cartilage and are carried on mutually adapted articular processes. The size and shape of these processes vary with spinal level and are described with the individual vertebrae. Fibrous capsule
page 758 page 759
Figure 45.44 Exiting and traversing nerve roots. The upper root (open arrow) is the exiting root at this level: the lower (arrow) is the traversing root here, which becomes the exiting root at the level below. The dotted roots are traversing roots of the lower segment.
The fibrous capsule is thin and loose and attached peripherally to the articular facets of adjacent articular processes. The capsules are longer and looser in the cervical region. According to Bogduk, the anterior fibrous capsule is replaced entirely by the ligamentum flavum in the lumbar spine. Intracapsular structures Bogduk describes two types of intra-articular structure in lumbar facet joints, namely subcapsular fat and 'meniscoid' structures. The latter structures may be collagenous, fibroadipose or purely adipose, and project into the crevices between non-congruent articular surfaces. They resemble inclusions seen in the small joints of the hand; their function is conjectural. Ligaments
Ligaments which work in conjunction with, and modify the function of, the facet
joints throughout the vertebral column are described on page 735. Synovial membrane
The synovium attaches around the periphery of the articular cartilages and lines the fibrous capsule. In the lumbar region it is reflected over the intracapsular structures described above. Vascular supply
Arterial anastomoses around the facet joints are formed from posterior spinal branches of those arteries which supply the vertebral column. Venous drainage is via the external and internal posterior vertebral venous plexuses to the intervertebral veins and thence to the larger named veins which drain the vertebral column. Lymphatic drainage follows the principles described for the vertebral column. Innervation
The facet joints are copiously innervated by medial branches of the dorsal primary rami of the spinal nerves, which give articular branches to the joints above and below them. Relations and 'at risk' structures
Anteriorly the capsules of the facet joints form the posterior boundaries of the intervertebral foramina (p. 741). Posteriorly and laterally the joints are related to the deep muscles of the back, some of whose fibres attach to the capsules. The joints also lie in close relation to the medial branches of the dorsal rami of the spinal nerves and to their accompanying arteries and veins. Damage to the medial branches of the dorsal rami may denervate the deep back muscles. Access to the facet joints and their related nerves may be required in the diagnosis and treatment of spinal pain. Lumbar articular tropism
In the lumbar region, asymmetrical orientation of the facet joints occurs in about one fifth of the population. Such facet tropism does not predispose to degenerative disc disease.
CRANIOVERTEBRAL JOINTS (Figs 45.45, 45.46, 45.47, 45.48) The articulation between the cranium and vertebral column is specialized to provide a wider range of movement than in the rest of the axial skeleton. It consists of the occipital condyles, and the atlas and axis, and functions like a universal joint which permits horizontal and vertical scanning movements of the head and is adapted for eye-head co-ordination. Atlanto-occipital joints
The atlas articulates with the occipital bone of the skull by a pair of synovial joints. The bones are connected by articular capsules and by the anterior and posterior atlanto-occipital membranes. UPDATE Date Added: 28 February 2006 Publication Services, Inc. Abstract: Configuration of the connective tissue in the posterior atlanto-occipital interspace: A sheet plastination and confocal microscopy study Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=15959363&query_hl=6&itool=pubmed_docsum Configuration of the connective tissue in the posterior atlanto-occipital interspace: A sheet plastination and confocal microscopy study. Nash L, Nicholson H, Lee AS et al: Spine 30(12):13591366, 2005. Articulating surfaces
Each joint consists of two reciprocally curved articular surfaces, one on the
occipital condyle and the other on the lateral mass of the atlas. The atlantal facets are concave and tilted medially. Fibrous capsules
The fibrous capsules surround the occipital condyles and superior atlantal articular facets. They are thicker posteriorly and laterally, where the capsule is sometimes deficient, and may communicate with the joint cavity between the dens and the transverse ligament of the atlas.
Figure 45.45 Atlanto-occipital and atlanto-axial joints: anterior aspect. On each side a small cleft has been opened between the lateral part of the upper surface of the body of the third cervical vertebra and the bevelled, inferior surface of the body of the axis.
Figure 45.46 Atlanto-occipital and atlanto-axial joints: posterior aspect.
Ligaments
The anterior atlanto-occipital membrane The anterior atlanto-occipital membrane is a broad, dense fibrous structure which connects the anterior margin of the foramen magnum to the upper border of the anterior arch of the atlas. Laterally, it blends with the capsular ligaments, and medially it is strengthened by a median cord, which is the anterior longitudinal ligament stretching between the basilar occipital bone and anterior atlantal tubercle. The posterior atlanto-occipital membrane
Figure 45.47 Median sagittal section through the occipital bone and first to third cervical vertebrae.
Figure 45.48 Posterior aspect of the atlanto-occipital and atlanto-axial joints. The posterior part of the occipital bone and the laminae of the cervical vertebrae have been removed and the atlanto-occipital joint cavities opened.
page 759 page 760
The posterior atlanto-occipital membrane is broad, but relatively thin, and connects the posterior margin of the foramen magnum to the upper border of the posterior atlantal arch, blending laterally with the joint capsules. It arches over the grooves for the vertebral arteries, venous plexuses and first cervical nerve. The ligamentous border of the arch is sometimes ossified. The ligaments connecting the axis and the occipital bone are also functionally involved (p. 760). Synovial membrane
The synovial cavities of one or both joints may communicate with that of the posterior component of the median atlanto-axial joint. Vascular supply
The arterial supply of this region is derived from an anastomosis between branches of the deep cervical, occipital and vertebral arteries. Innervation
The joints are innervated by branches of the dorsal primary rami of the first and second cervical spinal nerves. The medial branches of these rami often communicate with each other and with the third cervical nerve to form the posterior cervical plexus. Factors maintaining stability (See also pp. 735, 760 and 768.)
Factors maintaining stability include the fibrous capsules, the atlanto-occipital membranes, the shape of the articular surfaces, the ligaments connecting the axis and the occipital bone, the ligamentum nuchae and the posterior neck muscles. The suboccipital muscles play an important proprioceptive and postural role. Movements and muscles
The long axes of the joints run anteromedially. Taken together with their articular curvatures, this means that the joints act as one around both transverse and anteroposterior axes of movement, but not about a vertical axis. The main movement is flexion (about 18° in young subjects), with a few degrees of lateral flexion and rotation. The following muscles produce these movements. For flexion: longus capitis and rectus capitis anterior; for extension: recti capitis posteriores major and minor, obliquus capitis superior, semispinalis capitis, splenius capitis and trapezius (cervical part); for lateral flexion: rectus capitis lateralis, semispinalis capitis, splenius capitis, sternocleidomastoid and trapezius (cervical part); and for rotation: obliquus capitis superior, rectus capitis posterior minor, splenius capitis and sternocleidomastoid. Relations and 'at risk' structures
Posteriorly the joints are closely related to the vertebral arteries as they pass from the foramina transversaria into the foramen magnum. The dorsal primary ramus of the first cervical nerve and rectus capitis posterior major lie posteromedially. Rectus capitis anterior lies anteriorly. Atlanto-axial joints
The atlas articulates with the axis at three synovial joints. These are a pair between the lateral masses and a median complex between the dens of the axis and the anterior arch and transverse ligament of the atlas. Articulating surfaces
The articular surfaces of the joints between the lateral masses are often classified as planar. The bony articular surfaces are more complex in shape and are usually reciprocally concave in the coronal plane; the medial parts are somewhat convex
in the sagittal plane (especially that of the axis). The cartilaginous articular surfaces are usually less concave. The median joint is a pivot between the dens and a ring formed by the anterior arch and transverse ligament of the atlas. A vertically ovoid facet on the anterior dens articulates with a facet on the posterior aspect of the anterior atlantal arch. Fibrous capsules
The fibrous capsules for the lateral joints are attached to the articular margins and are thin and loose. Each has a posteromedial accessory ligament attached below to the axial body near the base of its dens, and above to the lateral atlantal mass near the transverse ligament. The fibrous capsule for the median joint is also relatively weak and loose, especially superiorly. Ligaments
Anteriorly, the vertebral bodies are connected by the anterior longitudinal ligament: here a strong, thickened band attaches above to the lower border of the anterior tubercle of the anterior arch of the atlas and below to the front of the axial body. Posteriorly the vertebral bodies are joined by the ligamenta flava which are attached to the lower border of the atlantal arch above, and to the upper borders of the axial laminae. At this level these ligaments form a thin membrane, pierced laterally by the second cervical nerves. The transverse atlantal and cruciform ligaments
page 760 page 761
The transverse atlantal ligament is a broad, strong band which arches across the atlantal ring behind the dens: its length varies about a mean of 20 mm. It is attached laterally to a small but prominent tubercle on the medial side of each atlantal lateral mass, and broadens medially where it is covered anteriorly by a thin layer of articular cartilage. It consists almost entirely of collagen fibres, which, in the central part of the ligament, cross one another at an angle to form an interlacing mesh. From its upper margin a strong median longitudinal band arises which inserts into the basilar part of the occipital bone between the apical ligament of the dens and membrana tectoria, and from its inferior surface a weaker and less consistent longitudinal band passes to the posterior surface of the axis. These transverse and longitudinal components together constitute the cruciform ligament. The transverse ligament divides the ring of the atlas into unequal parts (Fig. 45.19). The posterior two-thirds surrounds the spinal cord and meninges, the anterior third contains the dens, which it retains in position even when all other ligaments are divided. Ligaments connecting axis and occipital bone Ligaments connecting axis and occipital bone consist of the membrana tectoria, the paired alar ligaments, the median apical ligament, and the longitudinal components of the cruciform ligament. Membrana tectoria
Inside the vertebral canal, the membrana tectoria is a broad strong band representing the upward continuation of the posterior longitudinal ligament (p. 754). Its superficial and deep laminae are both attached to the posterior surface of the axial body. The superficial lamina expands as it ascends to the upper surface of the basilar occipital bone, and attaches above the foramen magnum, where it blends with the cranial dura mater. The deep lamina consists of a strong median band which ascends to the foramen magnum, and two lateral bands which pass to, and blend with, the capsules of the atlanto-occipital joints as they reach the foramen magnum. The membrane is separated from the cruciform ligament of the atlas by a thin layer of loose areolar tissue, and sometimes by a bursa. Alar ligaments
The alar ligaments are thick cords c.11 mm long, which pass horizontally and laterally from the longitudinally ovoid flattenings on the posterolateral aspect of the apex of the dens to the roughened areas on the medial side of the occipital condyles. In most individuals there is also an anteroinferior band, c.3 mm long, which inserts into the lateral mass of the atlas in front of the transverse ligament. Fibres occasionally pass from the dens to the anterior arch of the atlas. In addition, in c.10% of cases a continuous transverse band of fibres, the transverse occipital ligament, passes between the occipital condyles immediately above the transverse ligament. The ligaments consist mainly of collagen fibres arranged in parallel. The main function of the alar ligaments is now considered to be limitation of atlanto-axial rotation, the left becoming taut on rotation to the right and vice versa. The slightly upward movement of the axis during rotation helps permit a wider range of movement by reducing tension in the alar ligaments, and in the capsules and accessory ligaments of the lateral atlanto-occipital joint. The apical ligament of the dens
The apical ligament of the dens fans out from the apex of the dens into the anterior margin of the foramen magnum between the alar ligaments, and represents the cranial continuation of the notochord and its sheath. It is separated for most of its extent from the anterior atlanto-occipital membrane and cruciform ligament by pads of fatty tissue, though it blends with their attachments at the foramen magnum, and with the alar ligaments at the apex of the dens. The ligamentum nuchae and the anterior longitudinal ligament also connect cervical vertebrae with the cranium (p. 754). Synovial membrane
The synovial membranes of the lateral joints have no special features. The median joint has two synovial cavities which sometimes communicate. The synovial cavity of the posterior component of the median joint complex is larger, lying between the horizontally orientated ovoid facet, on the posterior surface of the dens and the cartilaginous anterior surface of the transverse ligament: communication often exists with one or both of the atlanto-occipital joint cavities. Vascular supply
The arterial supply of this region is derived from an anastomosis between branches of the deep cervical, occipital and vertebral arteries. Innervation
The joints are innervated by branches of the dorsal primary rami of the first and second cervical spinal nerves. The medial branches of these rami often communicate with each other and with the third cervical nerve to form the posterior cervical plexus. Factors maintaining stability (See also pp. 735, and 768.)
The most important factors maintaining stability are the ligaments, of which the transverse atlantal ligament is the strongest. The alar ligaments are weaker. Other ligaments connecting the axis and the occipital bone, the fibrous capsules, the ligamentum nuchae and the posterior neck muscles also contribute. The suboccipital muscles play an important proprioceptive and postural role. Movements and muscles
Movement is simultaneous at all three joints, and consists almost exclusively of rotation around the axis. The shape of the articular surfaces is such that, when rotation occurs, the axis ascends slightly into the atlantal ring, limiting stretch on the lateral atlanto-axial joint capsules. Rotation is limited mainly by the alar ligaments, with a minor contribution from the accessory atlanto-axial ligament. The normal range of atlanto-axial rotation is about 40°.
The muscles producing atlanto-axial rotation act on the cranium, transverse processes of the atlas and spinous process of the axis. They are mainly obliquus capitis inferior, rectus capitis posterior major and splenius capitis of one side, and the contralateral sternocleidomastoid. Relations and 'at risk' structures
The most important 'at risk' relation is the spinal cord, lying posterior to the median atlanto-axial joint. Anteriorly the atlanto-axial articulations, capsules and ligaments are separated from the buccopharyngeal fascia and superior constrictor by the longus capitis and longus colli, the prevertebral fascia and the retropharyngeal potential space.
LUMBOSACRAL JUNCTION Articulations between the fifth lumbar and first sacral vertebrae resemble those between other vertebrae. The bodies are united by a symphysis which includes a large intervertebral disc. The latter is deeper anteriorly at the lumbosacral angle. The synovial facet joints are separated by a wider interval than those above. Segmentation (transitional) anomalies affecting the lumbosacral junction are described on page 749. Articulating surfaces
The reciprocally curved surfaces of the facet joints show considerable individual variation in alignment and shape. Asymmetry (facet tropism) is not unusual. Ligaments
Iliolumbar ligament (Figs 111.11, 45.49) The fifth lumbar vertebra is attached to the ilium and sacrum by the iliolumbar ligament, which has several distinct parts. It is attached to the tip and anteroinferior aspect of the fifth lumbar transverse process, and sometimes has a weak attachment to the fourth transverse process. It radiates laterally and is attached to the pelvis by two main bands. A lower band passes from the inferior aspect of the fifth lumbar transverse process and the body of the fifth lumbar vertebra across the anterior sacroiliac ligament to reach the posterior margin of the iliac fossa. An upper band, which is part of the attachment of quadratus lumborum, passes to the iliac crest anterior to the sacroiliac joint, and is continuous above with the anterior layer of the thoracolumbar fascia (p. 734). The lower ligament has a more vertical component which reaches the posterior iliopectineal line: this component is a lateral relation of the L5 ventral ramus. A posterior component of the iliolumbar ligament passes behind quadratus lumborum to attach to the ilium. In neonates and children the iliolumbar 'ligament' is muscular: the muscle is gradually replaced by ligament up to the fifth decade of life. Other ligaments concerned with this joint are described on page 735. Vascular supply
The vascular supply of the lumbosacral junction is derived mainly from the iliolumbar and superior lateral sacral arteries. Innervation
The lumbosacral junction is innervated by branches derived from the fourth and fifth lumbar spinal nerves: the pattern of distribution is as described on page 735. Relations and 'at risk' structures page 761 page 762
Figure 45.49 Joints and ligaments on the posterior aspect of the left half of the pelvis and the fifth lumbar vertebra.
The lumbosacral disc is related anteriorly to the common iliac veins, the median sacral vessels, and the superior hypogastric plexus of nerves. The sympathetic trunks cross it anterolaterally, while the obturator nerves and lumbosacral trunks pass close laterally. The relations of the lumbosacral facet joints are similar to those of the lumbar facet joints described above (p. 759).
SACROCOCCYGEAL JUNCTION The sacrococcygeal joint is usually a symphysis between the sacral apex and coccygeal base, united by a thin fibrocartilaginous disc which is somewhat thicker in front and behind than laterally. Its surfaces carry hyaline cartilage which varies from thin veils to small islands. Occasionally the coccyx is more mobile and the joint is synovial. Ligaments
Anterior sacrococcygeal ligament (Fig. 111.11) The anterior sacrococcygeal ligament consists of irregular fibres which descend on the pelvic surfaces of both sacrum and coccyx: it is attached like the anterior longitudinal ligament. Superior posterior sacrococcygeal ligament The superior posterior sacrococcygeal ligament is flat and passes from the margin of the sacral hiatus to the dorsal coccygeal surface (Fig. 45.49), roofing the lower sacral canal. Deep dorsal sacrococcygeal ligament The deep dorsal sacrococcygeal ligament passes from the back of the fifth sacral vertebral body to the dorsum of the coccyx and corresponds to the posterior
longitudinal ligament. Lateral sacrococcygeal ligaments The lateral sacrococcygeal ligaments are bilateral, and connect the coccygeal transverse processes to the inferolateral sacral angles, completing foramina for the fifth sacral spinal nerves. Intercornual ligaments The intercornual ligaments connect sacral and coccygeal cornua on each side. A fasciculus also connects the sacral cornua to the coccygeal transverse processes. Vascular supply
The arterial supply of the sacrococcygeal junction is derived from the inferior lateral sacral and median sacral arteries. Innervation
The innervation of the sacrococcygeal junction is derived from the lower two sacral and the coccygeal nerves.
INTERCOCCYGEAL JOINTS In the young the intercoccygeal joints are symphyses, with thin discs of fibrocartilage between coccygeal segments. Segments are also connected by extensions of the anterior and posterior sacrococcygeal ligaments. In adult males all segments unite comparatively early, but in females union is later. In advanced age the sacrococcygeal joint becomes obliterated. Occasionally the joint between the first and second segments is synovial. The apex of the terminal segment is connected to overlying skin by white fibrous tissue.
© 2008 Elsevier
MUSCLES (Figs 45.50, 45.58) The musculature of the back is arranged in a series of layers, of which only the deeper are true, intrinsic, back muscles. These true back muscles are characterized by their position and by their innervation by branches of the posterior (dorsal) rami of the spinal nerves. Those below the neck lie deep to posterior layer of the thoracolumbar fascia. In the lumbar region, where the layers of the thoracolumbar fascia are well-defined, they occupy the compartment between its posterior and middle layers (p. 734). Lying superficial to the true, intrinsic muscles are the extrinsic, 'immigrant' muscles. The most superficial of these run between the upper limb and the axial skeleton, and consist of trapezius, latissimus dorsi, levator scapulae and the rhomboid muscles. Beneath this layer lie the serratus posterior group, superior and inferior, which are variably developed but usually thin, muscles, whose function may be respiratory or possibly proprioceptive. All the extrinsic muscles are innervated by anterior (ventral) rami. Trapezius, latissimus dorsi, levator scapulae, rhomboid major and rhomboideus minor are described on pages 836-838; serratus posterior muscles are described on page 963. The muscles of the posterior abdominal wall are described on page 1115. The intrinsic muscles also have superficial and deep layers. The more superficial layers contain the splenius muscles in the neck and upper thorax, and the erector spinae group in the trunk as a whole. The deeper layers include the transversospinal group, which is itself layered into semispinalis, multifidus and the rotatores, and the suboccipital muscles. Deepest of all lie the interspinal and intertransverse muscles. The latter group are not all innervated by dorsal rami: lumbar intertransversarii mediales, thoracic intertransversarii and medial parts of cervical posterior intertransversarii are so innervated, but the others are supplied by ventral rami and are thus not true muscles of the back.
VASCULAR SUPPLY The deep muscles of the back receive their blood supply from the following arteries: vertebral artery; deep cervical artery; superficial and deep descending branches of the occipital artery; deep branch of the transverse cervical artery, when present; superior intercostal artery via dorsal branches of the upper two posterior intercostal arteries; posterior intercostal arteries of the lower nine spaces via dorsal branches; dorsal branches of the subcostal arteries; dorsal branches of the lumbar arteries; dorsal branch of arteria lumbalis ima; dorsal branches of the lateral sacral arteries. page 762 page 763
Figure 45.50 Superficial (extrinsic) muscles of the back.
The detailed pattern of the arterial supply of the deep muscles of the back has been described by Michel Salmon (Taylor & Razaboni 1994). These muscles are supplied by dorsal branches of the posterior intercostal and lumbar arteries. In the thoracic and upper lumbar regions, where the components of the erector spinae run in well-defined longitudinal columns, arterial trunks from these branches run in the sulci between the columns and between the erector spinae and multifidus, giving off branches to supply the muscles. In the lumbar region, where the erector spinae is more of a common muscle mass, this vascular pattern is less regular.
SPLENIUS CAPITIS Attachments Splenius capitis arises from the dorsal edge of the lower half of the ligamentum nuchae, the spines of the seventh cervical and upper three or four thoracic vertebrae, and their supraspinous ligaments. The muscle passes upwards and laterally to be attached to the mastoid process and the rough surface on the occipital bone just below the lateral third of the superior nuchal line.
Relations The upper part of splenius capitis lies beneath sternocleidomastoid and the remainder lies deep to serratus posterior superior, the rhomboids and trapezius. Between sternocleidomastoid and trapezius it forms part of the floor of the posterior triangle of the neck, above and behind levator scapulae. Deep to splenius lie the upper parts of the erector spinae complex and the semispinalis cervicis. Vascular supply See page 762. Innervation Splenius capitis is innervated by medial branches of the dorsal rami of the middle cervical spinal nerves. Actions The action of splenius capitis is described under splenius cervicis.
SPLENIUS CERVICIS Attachments Splenius cervicis is attached to the spines of the third to the sixth thoracic vertebrae. It ascends to the posterior tubercles of the transverse processes of the upper two or three cervical vertebrae, immediately anterior to the attachment of levator scapulae. The splenii may be absent or vary in their vertebral attachments. Accessory slips also occur. Relations Splenius cervicis lies deep to serratus posterior superior, the rhomboids and trapezius. Its deep relations include the upper parts of the erector spinae complex and the lower semispinalis muscles. Vascular supply See page 762. page 763 page 764
Innervation Splenius cervicis is innervated by the medial branches of the dorsal rami of the lower cervical and upper thoracic spinal nerves. Actions
Figure 45.51 Rhomboid muscles.
Figure 45.51 Rhomboid muscles.
Figure 45.52 Splenius cervicis and splenius capitis.
Acting together, the splenii of the two sides draw the head directly backwards. Acting separately, they draw the head to one side, and rotate it slightly, turning the face to the same side. Each is therefore synergistic with the contralateral sternocleidomastoid.
ERECTOR SPINAE (Figs 45.53, 45.54, 68.1, 68.2, 68.3) Iliocostalis Iliocostalis lumborum Iliocostalis thoracis Iliocostalis cervicis
Longissimus Longissimus thoracis Longissimus cervicis Longissimus capitis
Spinalis Spinalis thoracis Spinalis cervicis Spinalis capitis
The erector spinae (sacrospinalis) muscle complex lies on either side of the vertebral column. It forms a large musculotendinous mass, which varies in size and composition at different levels. In the sacral and lower lumbar regions it narrows and becomes increasingly strong and tendinous as it approaches its attachments. In the upper lumbar region it expands to form a thick fleshy mass which divides into three columns which are, from lateral to mid-line, iliocostalis, longissimus and spinalis. The columns may be subdivided as follows: The main muscle mass can readily be felt in the lumbar region in the living subject. Its lateral border is flanked by a visible groove (Fig. 44.3), which ascends over the back of the thorax, traversing the ribs at their angles and running first laterally, then vertically, and finally medially until it is obscured by the scapula. Attachments Erector spinae arises from the anterior surface of a broad, thick tendon or aponeurosis, which is attached in the midline to the median sacral crest, the
spines of the lumbar and the eleventh and twelfth thoracic vertebrae, their supraspinous ligaments, and laterally to the medial aspect of the posterior iliac crest and to the lateral sacral crest, where it blends with the sacrotuberous and dorsal sacroiliac ligaments. Some of its fibres are continuous with gluteus maximus and multifidus. Iliocostalis Iliocostalis lumborum is attached, by flattened tendons, to the inferior borders of the angles of the lower six or seven ribs. page 764 page 765
Figure 45.53 Erector spinae muscle group.
Iliocostalis thoracis attaches below to the upper borders of the angles of the lower six ribs medial to the tendons of insertion of iliocostalis lumborum, and above to the superior borders of the angles of the upper six ribs and the back of the transverse process of the seventh cervical vertebra. Iliocostalis cervicis attaches below to the angles of the third to the sixth ribs,
medial to the tendons of insertion of iliocostalis thoracis, and above to the posterior tubercles of the transverse processes of the fourth, fifth and sixth cervical vertebrae. Longissimus Longissimus thoracis is the largest of the continuations of the erector spinae. In the lumbar region, where it blends with iliocostalis lumborum, some of its fibres are attached to the whole length of the posterior surfaces of the transverse processes and the accessory processes of the lumbar vertebrae, and to the middle layer of the thoracolumbar fascia. In the thoracic region it is attached, by rounded tendons, to the tips of the transverse processes of all the thoracic vertebrae, and by fleshy slips to the lower nine or ten ribs between their tubercles and angles. Longissimus cervicis lies medial to longissimus thoracis. It is attached by long thin tendons to the transverse processes of the upper four or five thoracic vertebrae, and again by tendons to the posterior tubercles of the transverse processes of the second to the sixth cervical vertebrae. Longissimus capitis lies between longissimus cervicis and semispinalis capitis. It is attached below by tendons to the transverse processes of the upper four or five thoracic vertebrae and the articular processes of the lower three or four cervical vertebrae and above to the posterior margin of the mastoid process, deep to splenius capitis and sternocleidomastoid. It is usually traversed by a tendinous intersection near its upper end. Spinalis Spinalis thoracis, the medial continuation of erector spinae, is barely separable as a distinct muscle. It lies medial to longissimus thoracis, and blends intimately with it. It is attached below by three or four tendons to the eleventh and twelfth thoracic and the first and second lumbar vertebral spines: these unite in a small muscle which is attached above by separate tendons to the spines of the upper thoracic vertebrae (the number varies from four to eight). It blends closely with semispinalis thoracis, which lies anterior to it. page 765 page 766
Figure 45.54 Axial MRI of lumbar spine showing erector spinae.
Spinalis cervicis, when it is present, is attached to the lower part of the ligamentum nuchae and the spine of the seventh cervical vertebra (and sometimes to the first and second thoracic vertebrae), and to the spine of the axis. Occasionally it is also attached to the spines of the two vertebrae immediately below it. Spinalis capitis usually blends to some extent with semispinalis capitis (see
below), but can be separate. Relations Erector spinae is covered in the lumbar and thoracic regions by the thoracolumbar fascia (p. 734), and by serratus posterior inferior below and the rhomboids and splenii above. In the lumbar region it lies in the compartment between the posterior and middle layers of the thoracolumbar fascia. Vascular supply See page 762. Innervation Erector spinae is innervated by lateral and intermediate branches of the dorsal rami of the lower cervical, thoracic and lumbar spinal nerves. Actions Erector spinae as a group extends and laterally flexes the vertebral column when acting against gravity. It contracts eccentrically to control the movement as the column is flexed forwards or laterally with the aid of gravity. Contraction of the erectores spinae extends the trunk, a movement controlled largely by opposing activity of the abdominal muscle complex (rectus abdominus and the oblique abdominals). Flexion of the trunk is initiated by flexor muscles such as rectus abdominis: as the centre of gravity moves forward control is transferred to the erectores spinae which then contract eccentrically. When the trunk is fully flexed the erectores spinae are relaxed and electromyographically quiet: in this position, flexion may be limited by passive forces generated by tension in the thoracolumbar fascia and in the spinal ligaments and by resistance to deformation of the intervertebral discs. Electromyographic activity in the erector spinae group is greater when work is carried out on a low surface from a standing position. Lateral flexion is controlled by the contralateral erector spinae, with input from the oblique muscles. Longissimus capitis extends the head and turns the face to the ipsilateral side.
TRANSVERSOSPINALIS Semispinalis thoracis Semispinalis cervicis Semispinalis capitis
Multifidus
Rotatores thoracis Rotatores cervicis Rotatores lumborum
Figure 45.55 Attachments of semispinalis.
The transversospinalis muscular group consists of the following muscles: These muscles run obliquely upwards and medially from transverse processes to adjacent, and sometimes more distant, spinous processes. Bogduk and his coworkers believe that, in the lumbar region at least, multifidus should be considered to run downwards and laterally (Macintosh et al 1986). Semispinalis (Fig. 45.55) Semispinalis thoracis consists of thin, fleshy fasciculi interposed between long tendons. It is attached below by a series of tendons to the transverse processes of the sixth to the tenth thoracic vertebrae, and above, again by tendons, to the spines of the upper four thoracic and lower two cervical vertebrae. Semispinalis cervicis, a thicker muscle, is attached below by a series of tendinous and fleshy fibres to the transverse processes of the upper five or six thoracic vertebrae, and above to the spines of the second to the fifth cervical vertebrae. The fasciculus connected with the axis is the largest, and is composed chiefly of muscle. Semispinalis capitis is attached by a series of tendons to the tips of the transverse processes of the upper six or seven thoracic and seventh cervical vertebrae, to the articular processes of the fourth, fifth, and sixth cervical vertebrae and, occasionally, to the spine of the seventh cervical or first thoracic vertebra. The tendons come together in a broad muscle which attaches above to the medial part of the area between the superior and inferior nuchal lines of the occipital bone. The medial part of the muscle, which is usually more or less distinct from the rest, is sometimes called biventer cervicis, because it is traversed by an incomplete tendinous intersection. Multifidus (Fig. 45.56)
page 766 page 767
Figure 45.56 Multifidus. A, cervicothoracic. B, lumbosacral parts.
Figure 45.57 Rotatores (thoracic region). (By permission from Benninghoff, Anatomie, 15th edition © Urban and Schwarzenberg, 1994.)
Multifidus consists of a number of fleshy and tendinous fasciculi which lie deep to the foregoing muscles and fill the groove at the side of the spines of the vertebrae from the sacrum to the axis. Its fasciculi attach as follows: most caudally, to the back of the sacrum as low as the fourth sacral foramen, to the posterior superior iliac spine and dorsal sacroiliac ligaments; in the lumbar region, to all the mammillary processes; in the thoracic region, to all the transverse processes; in the cervical region, to the articular processes of the lower four vertebrae. In the lumbar region a few fibres may attach to the tendon (aponeurosis) of the erector spinae, and to the capsules of the facet joints. Each fasciculus is attached to the spinous process of one of the vertebrae above. Some fasciculi attach to the base of the spinous process, while others reach its tip. The fasciculi vary in length. Thus the most superficial connect one vertebra to the third or fourth above, those next in depth connect one vertebra to the second or third above, and the deepest connect adjacent vertebrae. For further detail of the structure of this muscle, see Kalimo et al (1989). Rotatores (Fig. 45.57)
page 767 page 768
Rotatores thoracis consists of eleven pairs of small roughly quadrilateral muscles. Each connects the upper and posterior part of the transverse process of one vertebra to the lower border and lateral surface of the lamina of the vertebra immediately above. Some fibres may extend to the base of the spinous process of the vertebra above (rotatores longi). The first is found between the first and second thoracic vertebrae, and the last between the eleventh and twelfth thoracic vertebrae. One or more may be absent from the upper or lower ends of the series. Rotatores cervicis and lumborum are represented only by irregular and variable muscle bundles, whose attachments are similar to those of rotatores thoracis. Relations The transversospinalis group lie deep to erector spinae, except in the neck where semispinalis lies mainly deep to splenius and trapezius. A small section of semispinalis capitis may lie even more superficially, forming the uppermost part of the floor of the posterior triangle of the neck. In the lumbosacral region multifidus lies immediately deep to the erector spinae tendon (aponeurosis). The components of transversospinalis themselves lie in three planes, semispinalis is the most superficial and the rotatores are the most deeply placed. Semispinalis is
absent in the lumbar and sacral regions, and the rotatores are well represented only in the thoracic region. Vascular supply See page 762. Innervation All of the transversospinalis group is innervated by dorsal rami of spinal nerves, usually by medial branches. Actions Semispinales thoracis and cervicis extend the thoracic and cervical regions of the vertebral column, and rotate them towards the opposite side. Semispinalis capitis extends the head, and turns the face slightly towards the opposite side.
INTERSPINALES Interspinales are short paired muscular fasciculi attached above and below to the apices of the spines of contiguous vertebrae, one on either side of the interspinous ligament. They are most distinct in the cervical region, where they consist of six pairs, the first between the axis and third vertebra, and the last between the seventh cervical and first thoracic vertebrae. In the thoracic region they occur between the first and second vertebrae (sometimes between the second and third), and the eleventh and twelfth vertebrae. In the lumbar region there are four pairs between the five lumbar vertebrae. A pair is occasionally found between the last thoracic and first lumbar vertebrae, and another between the fifth lumbar vertebra and the sacrum. Sometimes cervical interspinales span more than two vertebrae.
INTERTRANSVERSARII Intertransversarii are small muscles between the transverse processes of the vertebrae. They are best developed in the cervical region, where they consist of posterior and anterior sets of muscles separated by the ventral rami of spinal nerves. Posterior intertransverse muscles are divisible into medial and lateral slips, which are supplied by the dorsal and ventral rami of the spinal nerves, respectively. Each medial slip, the intertransverse muscle 'proper', is often further subdivided into medial and lateral parts by the passage through it of the dorsal ramus of a spinal nerve. Anterior intertransverse muscles and lateral parts of the posterior muscles connect the costal processes of contiguous vertebrae and medial parts of the posterior muscles connect true transverse processes. There are seven pairs of these muscles, the highest between the atlas and axis, and the lowest between the seventh cervical vertebra and the first thoracic: the anterior muscles between atlas and axis are often absent. In the thoracic region they consist of single muscles, which are present between the transverse processes of only the last three thoracic and first lumbar vertebrae. In the lumbar region they again consist of two sets of muscles. One set, intertransversarii mediales, connects the accessory process of one vertebra with the mammillary process of the next. The other set, intertransversarii laterales, can be divided into ventral and dorsal parts: the ventral parts connect the transverse processes (costal elements) of the lumbar vertebrae, and the dorsal parts connect the accessory processes to the transverse processes of succeeding vertebrae. Both ventral and dorsal lumbar intertransversarii are innervated by ventral primary rami (Bogduk 1997). Thoracic intertransverse muscles and ligaments are homologous with the medial slips of the 'proper' posterior intertransverse muscles of the cervical region, and levatores costarum (p. 962) are homologous with their lateral slips. The lateral branch of the dorsal ramus of a spinal nerve separates thoracic intertransverse from levator costae. The lumbar levatores costarum are represented by the medial intertransverse muscles; the lateral intertransverse are homologous with the intercostal muscles. For other views on the homologies and classification of transversospinal musculature consult Sato (1973). Actions of the short muscles of the back
The short muscles of the back probably function, for the most part, as postural muscles. In effect, the vertebral column consists of a series of short, jointed
levers. A mechanical arrangement of this type is unstable under compression and will tend to buckle unless movement at the individual joints is controlled. The short muscles may serve to stabilize adjoining vertebrae, controlling their movement during motion of the vertebral column as a whole, and providing for more effective action of the long erector spinae muscles. In theory, the short muscles are capable of producing extension (multifidus, interspinales), lateral flexion (multifidus, intertransversarii) and rotation (multifidus and rotatores), but their detailed patterns of activity remain unknown. The deep muscles of the back as a whole are certainly involved in the control of posture: they contract intermittently during the swaying movements that take place from an upright position.
SUBOCCIPITAL MUSCLES (Fig. 45.58) The suboccipital muscles are four small muscles which connect the occipital bone, atlas and axis posteriorly. They lie inferior to the anterior part of the occipital bone, where three of the muscles form the boundaries of the suboccipital triangle. Above and medially lie rectus capitis posterior major; above and laterally, obliquus capitis superior, and below and laterally, obliquus capitis inferior. With the head in the anatomical position the suboccipital triangle lies almost in the horizontal plane. Rectus capitis posterior major Rectus capitis posterior major is attached by a pointed tendon to the spine of the axis, becomes broader as it ascends, and is attached to the lateral part of the inferior nuchal line and the occipital bone immediately below it. As the muscles of the two sides pass upwards and laterally, they leave between them a triangular space in which parts of the recti capitis posteriores minores are visible. Rectus capitis posterior minor Rectus capitis posterior minor is attached by a narrow pointed tendon to the tubercle on the posterior arch of the atlas. As it ascends it broadens before attaching to the medial part of the inferior nuchal line and to the occipital bone between the inferior nuchal line and the foramen magnum (p. 463). Either muscle may be doubled longitudinally. There may be an attachment to the dura mater. Obliquus capitis inferior Obliquus capitis inferior, the larger of the two oblique muscles, passes laterally and slightly upwards from the lateral surface of the spine and the adjacent upper part of the lamina of the axis to the inferoposterior aspect of the transverse process of the atlas. Obliquus capitis superior Obliquus capitis superior is attached by tendinous fibres to the upper surface of the transverse process of the atlas. It expands in width as it ascends dorsally, and is attached to the occipital bone between the superior and inferior nuchal lines, lateral to semispinalis capitis and overlapping the insertion of rectus capitis posterior major. Relations of the suboccipital triangle Medially the suboccipital triangle is covered by a layer of dense adipose tissue, deep to semispinalis capitis. Laterally it lies under longissimus capitis and sometimes splenius capitis, both of which overlap obliquus capitis superior. The 'floor' of the triangle is formed by the posterior atlanto-occipital membrane and the posterior arch of the atlas. The vertebral artery and the dorsal ramus of the first cervical nerve lie in a groove on the upper surface of the posterior arch of the atlas. Vascular supply The suboccipital muscles receive their blood supply from the vertebral artery and deep descending branches of the occipital artery. Innervation All the suboccipital muscles are supplied by the dorsal ramus of the first cervical spinal nerve. page 768 page 769
Figure 45.58 Posterior view of the left suboccipital triangle.
Actions of the suboccipital triangle The suboccipital muscles are involved in extension of the head at the atlantooccipital joints and rotation of the head and atlas on the axis. Obliquus capitis superior and the two recti are probably more important as postural muscles than as prime movers, but this is difficult to confirm by direct observation. Rectus capitis posterior major extends the head and, acting with obliquus capitis inferior, rotates the face towards the ipsilateral side. Rectus capitis posterior minor extends the head. Obliquus capitis superior extends the head and laterally flexes it to the ipsilateral side.
© 2008 Elsevier
MOVEMENTS OF THE VERTEBRAL COLUMN Spinal movements between individual vertebrae cannot be measured accurately by skin-surface techniques. Only biplanar radiography is good enough for this purpose. The best values for the normal adult lumbar spine are from Pearcy (1984a,b). Movements of the entire lumbar spine can be measured using skinsurface techniques, but such measurements are of limited clinical use because of high inter- and intra- observer variation. The intervertebral discs are the principal sites of vertebral column movement. At most levels they are also the limiting factor for motion according to their available deformation. Bony deformation in the subchondral bone and articular cartilage may contribute. Regional variations in mobility of the spine depend on the geometry, orientation and properties of the facet joints and related ligamentous complexes. Physiological intervertebral movements usually combine tilting (bending) and gliding (shear), so the instantaneous centre of rotation moves continually during the movement. During flexion and extension of lumbar vertebrae, the centre of rotation usually lies near the centre of the intervertebral discs, close to the end-plate of the inferior vertebra. The oblique and ovoid articular surfaces of the facet joints ensure that spinal movements in different planes are usually 'coupled' to a certain extent. For example, lateral flexion would cause impingement of the articular surfaces on that side, leading to a posteriorly directed force on the upper vertebra which would act to rotate it about its long axis. Physiologically, coupled movements are variable, and are probably influenced by muscular control. Although movements between individual vertebrae are small, their summation gives a large total range to the vertebral column in flexion, extension, lateral flexion, and axial rotation. Each pair of vertebrae with its interposed disc and ligaments is termed a motion segment or functional spinal unit. In flexion the anterior longitudinal ligament becomes relaxed as the anterior parts of the intervertebral discs are compressed. At its limit, the posterior longitudinal ligament, ligamenta flava, interspinous and supraspinous ligaments and posterior fibres of intervertebral discs are tensed; interlaminar intervals widen, inferior articular processes glide on superior processes of subjacent vertebrae and their capsules become taut. Tension of extensor muscles is also important in limiting flexion, e.g. when carrying a load on the shoulders. Flexion is effectively absent in the thoracic region. In forward flexion of the lumbar spine, the muscles protect the osteoligamentous spine from injury but the margin of safety can be compromised during repetitive or sustained bending by a failure of the spinal reflexes. Once the muscle protection is lost, flexion injury affects first the interspinous ligaments and then the capsules of the facet joints. The ligamentum flavum has such a high content of elastin that it is always under tension, and can be stretched by 80% without damage. This ligament probably functions to provide a constant smooth lining to the vertebral canal, one which never is overstretched in flexion, and never goes slack in extension.
In extension the opposite events occur, and there is compression of posterior discal fibres. Extension is limited by tension of the anterior longitudinal ligament, anterior discal fibres and approximation of spines and facet joints. It is marked in cervical and lumbar regions, and much less at thoracic levels, partly because of the discs are thinner, but also because of the presence of the ribs and chest musculature. In full extension, the axis of movement passes posterior to the disc, moving forwards as the column straightens and passes into flexion, reaching the centre of the intervertebral disc in full flexion. In lateral flexion, which is always combined with 'coupled' axial rotation, intervertebral discs are laterally compressed and contralaterally tensed and lengthened, and motion is limited by tension of antagonist muscles and ligaments. Lateral movements occur in all parts of the column but are greatest in cervical and lumbar regions. page 769 page 770
Axial rotation involves twisting of vertebrae relative to each other with accompanying torsional deformation of intervening discs. About 70% of cervical rotation occurs at the upper two cervical levels, mainly the atlanto-axial joint. Elsewhere in the column, although movement is slight between individual vertebrae, the range summates to become large for the column as a whole. In the post-cervical column, the effective range of rotation is greatest at the thoracolumbar junction. There is very little rotation in the remainder of the lumbar region. In the cervical region the upward inclination of the superior articular facets allows free flexion and extension. The latter is usually greater, and is checked above by locking of the posterior edges of the superior facets of C1 in the occipital condylar fossae, and below by slipping of the inferior processes of C7 into grooves inferoposterior to the first thoracic superior articular processes. Flexion stops where the cervical convexity is straightened, checked by apposition of the projecting lower lips of vertebral bodies on subjacent bodies. Cervical lateral flexion and rotation are always coupled, and the superomedial inclination of the superior articular facets imparts rotation during lateral flexion. Cervical motion can be considered to involve the upper (i.e. the atlanto-occipital and atlanto-axial complexes) and the lower cervical spine (C3-7). Two physiological movements take place at the atlanto-occipital joints, those of flexionextension and lateral flexion. The atlanto-axial joints allow flexion-extension and rotation. Some studies have suggested that maximum flexion-extension occurs between the occiput and C1; however, Frobin et al (2002) noted between 12.6 and 14.5° at this level, which is less than at some of the other cervical levels. Global cervical flexion ranges from 45 to 58° depending on the method of assessment, age and sex: older subjects and females exhibit less motion (Ordway et al 1997; Trott et al 1996). At an intersegmental level, motion increases from the second cervical level, peaking at the mid cervical level, with 14-17° recorded at C4/5, before reducing at the junction of the cervical and thoracic spine (9.8-11.5°noted at C6/7) (Frobin et al 2002). Global ranges of lateral flexion range from 32 to 47°, again with a gradual reduction in range with
age and sex, whilst rotational movements range from 63 to 78°. Intersegmental ranges of motion vary from 4.7 to 6° for lateral flexion to between C2 and C7 and 2-12° for rotation (White & Panjabi 1990). In the thoracic region, especially superiorly, all movements are limited, reducing interference with respiration. Lack of upward inclination of the superior articular facets prohibits much flexion, and extension is checked by contact of the inferior articular margins with the laminae and of the spines with each other. Thoracic rotation is freer, though limited by the ribs at upper levels. Its axis is in the vertebral bodies in the midthoracic region, and in front of them elsewhere, so that rotation involves some lateral displacement. The direction of articular facets would allow free lateral flexion, but this is limited in the upper thoracic region by the resistance of the ribs and sternum. Rotation is usually combined with slight lateral flexion to the same side. Movement in the thoracic spine is frequently regionalized to upper, mid- and lower thoracic. In the upper thoracic flexion ranges from 7.8 to 9.5°, increasing to 1011.4° in the mid-thoracic and 12.5 to 12.8° in the lower thoracic (Willems et al 1996). Extension is more consistent throughout the thoracic spine ranging from 7.1 to 9.7°. Lateral flexion increases as the thoracic spine is descended ranging from 5.6 to 6.2° in the upper thoracic; 7.9 to 8.1° in the mid-thoracic and 11.9 to 13.2° in the lower thoracic. Rotation however, is greatest in the mid-thoracic region being between 11.8 to 15.9° in the upper thoracic; 21.5 to 25.3° in the midthoracic; and 8.3 to 11.8° in the lower thoracic. Lumbar flexion movements are generally greater than extension or lateral flexion. Axial rotation occurs about a centre of rotation in the posterior anulus, and is limited by bony contact in the facet joints after only 1-2° of movement. Functional transition between thoracic and lumbar regions is usually between the eleventh and twelfth thoracic vertebrae (p. 746), where the facet joints usually fit so tightly that slight compression locks them, and prevents all movements but flexion. During flexion of the lumbar spine there is an unfolding or straightening of the lumbar lordosis. Thus in full flexion the lumbar spine assumes a straight alignment or is curved slightly forwards. Normal ranges of global lumbar flexion range from 58 to 72° in under 40-year-olds and 40 to 60° in the over forties: females exhibit a reduced range compared with males (McGregor et al 1995; Dvorák et al 1995). At an intersegmental level, the L3/4 junction and L4/5 junction exhibit the greatest mobility, c.12 and 13° respectively, whilst at the lowest level L5/S1 there is only 9° and at the upper lumbar levels 8 and 10° respectively (Pearcy et al 1984). Movements into extension are the converse of those seen in flexion. Normal ranges of global extension range from 25 to 30° in under 40-yearolds and 15 to 20° in those over forty. At an intersegmental level, L5/S1 and L1/2 exhibit the greatest mobility at c.5° whilst the remaining levels exhibit less than 5° of extension. Ranges of lateral flexion and rotation in the lumbar spine are reduced compared to other regions of the spine. Global lateral flexion ranges from 20 to 35°, whilst rotation ranges from 25 to 40°, and again is reduced with age. Assessment of intersegmental rotation and lateral flexion has proven difficult because of the limitations of measurement techniques.
ROLE OF MUSCULATURE Although muscles will move the spinal column, the majority of muscular activity is involved in providing stability to maintain posture and to provide a firm platform for limb function. Hence the concept of 'core stability' in modern rehabilitation programmes, especially in sports-related problems. It is important to recognize the way in which the muscles of the back work in conjunction with those of the abdominal wall, particularly the oblique and transversus muscles, and with those of the lower limbs. The erector spinae group and the internal oblique and transversusabdominis are anatomically and functionally connected by the thoracolumbar fascia which encloses the former, and into which the latter are inserted. This fascia, together with collagenous tissue within the back muscles, plays an important role in resisting forward bending of the trunk, and in manual handling. The fascia is tensioned primarily by flexing the trunk, although this tension may be enhanced slightly by the lateral pull of the abdominal muscles. It is functionally advantageous to generate tension in the fascia and muscle sheaths, because the elastic strain energy stored in these stretched tissues can be used to help bring the trunk to an upright position and so reduce the metabolic cost of the movement (Adams et al 2002). The thoracolumbar fascia may also have an important function in transferring load between the trunk and the lower limbs: tension in the fascia can be increased by the actions of gluteus maximus and the hamstrings as well as by trunk flexion. Muscles producing vertebral movements
The spinal column is moved both directly by muscles attached to it, and indirectly by muscles attached to other bones. Gravity always plays a part. Flexion is effected by longus capitis and longus colli, scaleni, sternocleidomastoid and rectus abdominis of both sides, aided in the lumbar region by the abdominal obliques; extension is achieved by the erector spinae complex and the transversospinalis group, splenius, semispinalis capitis and trapezius of both sides, together with the sub-occipital muscles; lateral flexion by longissimus and iliocostocervicalis, oblique abdominal muscles and flexors on the side of lateral flexion, quadratus lumborum; and lastly, rotation by sternocleidomastoid, splenius cervicis, oblique abdominal muscles, rotatores and multifidus.
© 2008 Elsevier
FACTORS INVOLVED IN STABILITY The vertebral column is remarkable in that it combines mobility, stability and loadbearing capacity and also protects its contained neural structures, irrespective of its position. Much of the stability of the vertebral column depends on dynamic muscular control. However there are bony and ligamentous 'static' stabilizers. There is considerable variation between segments of the column regarding stability and mobility: the most mobile levels are the least stable. The latter are those in which the ratio of intervertebral disc height to vertebral body height is highest. Stability may be compromised by damage to any of the structures involved. Trauma may affect any vertebral region. Levels of specialized mobility (e.g. atlanto-axial joint) and junctions of mobile and relatively fixed regions (e.g. cervicothoracic, thoracolumbar) are particularly vulnerable to severe structural damage, which is often accompanied by spinal cord and nerve injury. Injuries of the vertebral column may be purely soft-tissue (ligaments, joint capsules and muscles) or may affect bony structures. Pure ligamentous/capsular injuries leading to instability may be particularly difficult to diagnose in the absence of gross radiological signs. In the cervical spine, subluxation and dislocation of the facet joints commonly occur without bony injury because of the orientation of the articular facets. Chronic infections of many types (e.g. tuberculosis) may involve the vertebrae and lead to their deformity and collapse, affect their mechanical properties and compromise their neuroprotective function. Acute infections, spreading locally or via the bloodstream, may lead to the collection of pus within the vertebral canal causing spinal cord compression (epidural abscess). page 770 page 771
The integrity of the vertebrae may also be affected by malignant disease, most commonly metastatic. Vertebrae have a copious blood supply throughout life, and many of the common cancers (e.g. breast, bronchus) spread via the arterial system. Cancers of the haemopoietic system (e.g. multiple myeloma) also commonly affect the vertebrae. Prostatic carcinoma has a predilection to metastasize to the vertebral column, often using the venous (Batson's plexus) rather than the arterial route. Metastatic deposits may occur within the epidural space, compressing the contents of the dural sac at multiple levels. Systemic inflammatory diseases may cause both deformity and instability of the vertebral column. Rheumatoid arthritis inflames facet joints and weakens ligaments, leading to instability, especially in the cervical spine. Ankylosing spondylitis and other seronegative arthritides affect joints and ligamentous attachments (entheses), leading to ectopic ossification of collagenous structures, fusion (ankylosis) of interbody and facet joints, and loss of the normal spinal curvatures. Widespread new bone formation at and around the joints of the column occurs in DISH (diffuse idiopathic skeletal hyperostosis). Such conditions would seem to increase stability of the column, at the expense of its mobility and
function, but an ankylosed spine is very liable to fracture which carries an associated risk of neural damage. Full stability and load-bearing capacity both require intact vertebral bodies and intervertebral discs. Earlier views regarding the relative importance of the discbody complex and the posterior elements have proved somewhat simplistic. Clinical observation led to the 'three-column concept' of spinal stability (Denis 1983), in which the column was divided into three longitudinal parts rather than two. The anterior column is formed by the anterior longitudinal ligament, the anterior half of the vertebral body and the anterior anulus fibrosus. The middle column is made up of the posterior longitudinal ligament, the posterior half of the vertebral body and pedicles and the posterior anulus fibrosus. The posterior column consists of the neural arch and facet joints and the posterior ligamentous complex. Failure of any column can lead to some instability: the status of the middle column is the most important. The more columns affected, the worse the instability. An injury to two columns is usually unstable. The intervertebral discs, by elastic deformability, permit tilting and axial rotation between vertebral bodies, and also help to reduce vertical accelerations of the head. The main shock-absorbing mechanism of the column stems from the spinal curves, which increase and decrease slightly during locomotion against the restraining tension of the trunk muscles. It is the elastic strain energy in the stretched tendons of the muscle which actually does the shock absorbing. Both body height and spinal stability are subject to a marked diurnal variation. Body height is affected by changes from recumbency to the upright posture. These diurnal variations appear to be due to changes which occur within the cervical, thoracic and lumbar regions of the spine. Investigations using stereophotogrammetry have demonstrated that 40% of diurnal changes occur in the thoracic spine, affecting the degree of kyphosis, and a further 40% in the lumbar spine, without affecting the lordosis. (Wing et al 1992). The greatest change in vertebral column length is found in adolescents and young adults. The height loss occurs within 3 hours of rising in the morning, with the overall loss being c.15 mm. Although the curvatures within the vertebral column contribute to the changes in height, changes within the intervertebral disc contribute both to observed height loss and to variation in stability. Magnetic resonance imaging investigations reveal a dynamic movement of fluid into and out of an intervertebral disc and adjacent vertebral body over a 24-hour period, which is related to body position. In the early morning, the discs are swollen with water, the intervertebral ligaments and the anulus fibrosus are taut, and the intrinsic bending stiffness and stability of the osteoligamentous spine are relatively high. After several hours of normal activity, the discs lose c.20% of their water and height which makes the ligaments slack and greatly reduces the bending stiffness of the spine. Relatively more of the stability of the spine must then be provided by the musculature. This diurnal expulsion of water from intervertebral discs also affects the distribution of compressive loading in the spine. As the day progresses, the hydrostatic pressure in the nucleus pulposus falls, and stress concentrations arise
in the anulus fibrosus and facet joints. All ligaments of the column, as well as the facet joint capsules, are important in the maintenance of stability. The anterior longitudinal ligament is very strong, and resists translational displacement (shear) of the vertebrae as well as extension. All the ligaments of the posterior complex resist flexion and rotation, and their integrity determines the range of movements allowed. These ligaments can support the whole column when the muscles are inactive, e.g. in quiet standing. At the limit of lumbar flexion the column is supported mainly by the thoracolumbar fascia and by collagenous tissue within the electrically silent muscles of the back. Movements are both determined and constrained by the shape and orientation of the facet joints, whose articular surfaces stabilize the column primarily by resisting horizontal gliding (shear) movements and axial rotation. In the most mobile regions the joint surfaces are flatter and more horizontally placed, as will become apparent if a typical cervical facet joint is compared with a typical lumbar joint. Certain regions of the vertebral column are further stabilized by additional, extraspinal factors. The thoracic spine is stabilized by its position as an integral part of the thoracic cage and by its strong ligamentous linkages with the ribs. The sacrum is effectively a virtually fixed integral element of the bony pelvis. The contribution to stability conferred by the musculature has in the past been grossly underrated. The whole vertebral column is stabilized by the 'guy-rope' or staying effect of the long muscles which attach it to the girdles, the head and the appendicular skeleton, especially erector spinae, which controls global posture and movement. The small and deep muscles of the back are best able to resist shear movements between vertebrae because only they have sufficient angulation to the long axis of the vertebral column to do this effectively. These deep muscles can also fine-tune intervertebral movements. For most back problems in clinical practice, especially chronic low back pain, enhancing muscle strength, stamina, and coordination with the many other muscle groups which contribute to stability, e.g. pelvic girdle muscles, is the most appropriate and effective therapeutic avenue. Only a minority of cases benefit from surgery. Furthermore, neglecting the musculature may explain the relatively high failure rates from surgery. Injury to the vertebral column may result from several different mechanisms: flexion, extension, distraction, rotation, shear and compression. Some of these mechanisms, often flexion, axial rotation and compression, commonly occur together.
© 2008 Elsevier
POSTURE AND ERGONOMICS Posture is a descriptive term for the relative position of the body segments during rest or activity. The maintenance of good posture is a compromise between minimizing the load on the spine and minimizing the muscle work required. The well-balanced erect body has a line of gravity which extends from the level of the external auditory meatus, through the dens of the axis just anterior to the body of the second thoracic vertebra, through the centre of the body of the twelfth thoracic vertebra, and through the rear of the body of the fifth lumbar vertebra to lie anterior to the sacrum. The position of the line of gravity may move anteriorly with locomotion, and may vary between individuals. The normal curvature of the cervical spine is a lordosis. However, as a result of pain, injury or poor ergonomics, this curve can become exaggerated to give a 'protruding chin' stance, i.e. hyperlordosis in the lower cervical spine with a kyphosis in the upper cervical spine. This can result in shortness and over-activity of the neck extensor muscle group, and elongation and under-activity of the neck flexor group. The thoracic spine is held convex posteriorly, and this posture primarily results from the structure of the underlying vertebrae. However, this curve or kyphosis can become exaggerated to give the impression of a rounded back. Poor posture and ergonomics can lead to this exaggerated curvature but other important causes include tuberculosis, a wedge or compression fracture of a vertebral body, Scheuermann's osteochondritis, ankylosing spondylitis, osteoporosis, and metastatic carcinoma. The lumbar spine is held in a lordosis. The degree of this lordosis is determined by the lumbosacral angle and is normally 30-45°. The muscles responsible for this posture include erector spinae, rectus abdominus, the internal and external obliques, psoas major, iliacus, the gluteal and hamstring muscles. The lordosis can be increased (as a result of weak abdominal muscles and tight hamstring muscles), decreased, flattened (common in people either with acute or chronic low back pain), or reversed. page 771 page 772
A common postural deviation seen throughout the spine is scoliosis or lateral curvature of the spine. It can be structural, compensatory or protective. In structural scoliosis, the lateral curvature is associated with vertebral rotation, and both the curve and the rotation become more accentuated on forward flexion. Such a scoliosis is common in adolescent girls and its cause is unknown. It may also be secondary to an underlying disorder, e.g. muscular dystrophy, spinal muscular atrophy or spina bifida. A compensatory scoliosis occurs when the pelvis is tilted laterally, e.g. as a result of unequal leg length or of a fixed abduction or adduction deformity at the hip joint. Usually there is no intrinsic abnormality of the spine itself and the scoliosis disappears when the pelvic tilt is corrected. A sciatic or antalgic scoliosis is a temporary deformity produced by the
protective action of muscles in certain painful conditions of the spine. Ergonomics has been defined as 'the way humans work', and it permits an appreciation of the effects of tasks and the work environment on underlying postural biomechanics. Nachemson (1975) showed that discs were loaded maximally in sitting and in lifting in a forward leaning position, so sitting posture and lifting have received considerable ergonomic attention. In sitting the goal has been to determine the seat type and reclining angle associated with lowest disc pressure and the least paraspinal muscle activity. When sitting with the hips and knees flexed to 90° the pelvis rotates posteriorly, flattening the lumbar lordosis and con sequently increasing the load on the intervertebral discs. Thus it is now advised that in sitting the angle between trunk and thigh should be between 105 and 135°, with the sacrum tilted at 16° and the fourth and fifth lumbar vertebrae supported. In lifting heavy weights there is considerable initial compression of lumbar intervertebral discs, and large increases in thoracic and intra-abdominal pressure. The compressive force acting on the spine is shared between the vertebral bodies and the neural arch. In the lumbar spine, the neural arch typically resists 20% of this force once the disc height has been reduced by diurnal fluid expulsion, and when the spine is positioned upright. However, age-related narrowing of the disc can cause more than 50% of the compressive force to be resisted by the neural arch, which may explain why osteoarthritis of the facet joints commonly follows disc degeneration. When lifting, manual handling advisors emphasize the importance of leg lifting as opposed to back lifting. Loads should also be kept close to the body to reduce the lever arm of the load. The use of deep inspiration to raise intra-abdominal pressure while lifting has also been advised, as this is believed to offer further support to the lumbar spine. The spine is at risk when lifting is combined with twisting, lateral bending, and asymmetric postures. However, heavy lifting remains one of the key work related risk factors for the spine together with whole body vibration, prolonged sitting, twisting and bending. UPDATE Date Added: 07 December 2005 Publication Services, Inc. En bloc laminoplasty without dissection of paraspinal muscles. Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=16122019&query_hl=13 En bloc laminoplasty without dissection of paraspinal muscles. Hosono N, Sakaura H, Mukai Y, Ishii T, Yoshikawa H. J Neurosurg Spine. 2005 Jul;3(1):29-33. REFERENCES Adams MA, Bogduk N, Burton K, Dolan P 2002 The Biomechanics of Back Pain. Edinburgh: Churchill Livingstone. A comprehensive and detailed source of information on the functional anatomy, tissue biology and biomechanics of the lumbar spine. Batson OV 1957 The vertebral vein system. Am J Roentgenol 78: 195-212. A pioneering study of the venous plexuses of the vertebral column which has become the standard of
reference in its field. Boelderl A, Daniaux H, Kathrein A, Maurer H 2002 Danger of damaging the medial branches of the posterior rami of spinal nerves during a dorsomedian approach to the spine. Clin Anat 15: 77-81. Detailed descriptions of the vascular supply and innervation of the posterior elements of the thoracolumbar spine and the overlying muscles. Medline Similar articles Full article Bogduk N 1997 Clinical Anatomy of the Lumbar Spine and Sacrum, 3rd edn. Edinburgh: Churchill Livingstone. The most thorough text currently available on the topographical and functional anatomy of the lumbosacral spine, with over 800 references. The book incorporates biomechanical and physiological information which is related to the clinical problem of low back pain. Cormack GC, Lamberty BGH 1994 Arterial Anatomy of Skin Flaps, 2nd edn. Edinburgh: Churchill Livingstone. A comprehensive plastic surgical textbook in which the cutaneous arterial supply is described in detail. Crock HV 1996 Atlas of Vascular Anatomy of the Skeleton and Spinal Cord. London: Martin Dunitz. Crock HV, Yoshizawa H 1976 The blood supply of the lumbar vertebral column. Clin Orthop 115: 6-21. Medline Similar articles Dean NA, Mitchell BS 2002 Anatomic relation between the nuchal ligament (ligamentum nuchae) and the spinal dura mater in the craniocervical region. Clin Anat 15: 182-5. Describes continuity in the midline between the spinal dura and the ligamentum nuchae in human cadavers. Denis F 1983 The three column spine and its significance in the classification of acute thoracolumbar spinal injuries. Spine 8: 817-31. Seminal paper for the understanding and classification of spinal instability. Medline Similar articles Full article Dvorák J, Vajda EG, Grob D, Panjabi MM 1995 Normal motion of the lumbar spine related to age and gender. Eur Spine J 4: 18-23. Medline Similar articles Frobin W, Leivseth G, Biggeman M, Brinckmann P 2002 Sagittal plane segmental motion of the cervical spine. A new precision measurement protocol and normal motion data of healthy adults. Clin Biomech 17: 21-31. Groen G, Baljet B, Drukker J 1990 The nerves and nerve plexuses of the human vertebral column. Amer J Anat 188: 282-96. An acetylcholinesterase whole-mount study of human fetal material giving detail of the perivertebral nerve plexuses and of the sinuvertebral nerves. Medline Similar articles Full article Kalimo H, Rantanen J, Viljanen T, Einola S 1989 Lumbar muscles: structure and function. Ann Med 21: 353-9. A source of detailed information, particularly on the anatomy of multifidus. Medline Similar articles Lang J 1986 Craniocervical region, osteology and articulations. Neuro Orthop 1: 67-92. Macintosh JE, Valencia F, Bogduk N, Munro RR 1986 The morphology of the lumbar multifidus muscles. Clin Biomech 1: 196-204. Macnab I, McCulloch J 1990 Backache, 2nd edn, Baltimore: Williams and Wilkins. Chapter 1. The functional anatomy of the lumbar spine, described as a basis for the clinical management of low back pain. MacLaughlin SM, Oldale KNM 1992 Vertebral body diameters and sex prediction. Ann Hum Biol 19: 28593. Describes the archaeological and forensic examination of skeletal material. Medline Similar articles Full article McGregor AH, McCarthy ID, Hughes SPF 1995 Motion characteristics of the lumbar spine in the normal population. Spine 20: 22: 2421-8. Mercer S, Bogduk N 1999 The ligaments and anulus fibrosus of human adult cervical intervertebral discs. Spine 24: 619-28. A human cadaveric microdissection study showing that the cervical anulus fibrosus is an anterior crescent rather than a uniformly circumferential structure. Medline Similar articles Full article Nachemson A 1975 Towards a better understanding of low-back pain: a review of the mechanics of the lumbar disc. Rheumatol Rehab 14: 129-43. Newell RLM 1999 The spinal epidural space. Clin Anat 12: 375-9. Review of the morphological, developmental and topographical aspects of the spinal epidural space. Medline Similar articles Full article Ordway NR, Seymour R, Donelson RG, Hojnowski L, Lee E, Edwards T 1997 Cervical sagittal range of motion using three methods. Spine 22: 501-508. Medline Similar articles Full article
Pearcy M, Protek I, Shepherd J 1984a Three-dimensional X-ray analysis of normal movement in the lumbar spine. Spine 9: 294-7. Medline Similar articles Full article Pearcy M, Tibrewal SB 1984b Axial rotation and lateral bending in the normal lumbar spine Spine 9: 5827. Sato T 1973 A new classification of the transverso-spinalis system. Proc Jap Acad 49: 51-6. An alternative view of a controversial aspect of true back muscle homology. Taylor GI, Razaboni RM (eds) 1994 Michel Salmon: Anatomic Studies, Book 1. Arteries of the Muscles of the Extremities and Trunk. St Louis: Quality Medical Publishing. A translated, updated and edited version of a classic French text first published in 1933. Now a major source-book in plastic surgery. Taylor JR, Twomey LT 1984 Sexual dimorphism in human vertebral shape. J Anat 138: 281-6. Anthropometric and radiological studies of children and adolescents. Medline Similar articles Full article page 772 page 773
Trott PH, Pearcy MJ, Ruston SA, Fulton I, Brien C 1996 Three-dimensional analysis of active cervical motion: the effect of age and gender. Clin Biomech 11: 201-206. Twomey L, Taylor J, Furniss B 1983 Age changes in the bone density and structure of the lumbar vertebral column. J Anat 136: 15-25. Medline Similar articles White AA, Panjabi MM 1990 Clinical Biomechanics of the Spine, 2nd edn. Philadelphia: JB Lippincott. Willems JM, Jull GA, Ng JK-F 1996 An in vivo study of the primary and coupled rotations of the thoracic spine. Clin Biomech 11: 311-16. Wing P, Tsang I, Gagnon F, Susak L, Gagnon R 1992 Diurnal changes in the profile shape and range of motion of the back. Spine 17: 761-5. Medline Similar articles page 773 page 774
© 2008 Elsevier
46 Macroscopic anatomy of the spinal cord and spinal nerves This chapter deals with the gross anatomy of the structures which lie within the vertebral canal and its extensions through the intervertebral foramina, the spinal nerve or radicular ('root') canals. The spinal cord, its blood vessels and nerve roots lie within a meningeal sheath, the theca, which occupies the central zone of the vertebral canal and extends from the foramen magnum, where it is in continuity with the meningeal coverings of the brain, to the level of the second sacral vertebra in the adult. Distal to this level the dura extends as a fine cord, the filum terminale externum, which fuses with the posterior periosteum of the first coccygeal segment. Tubular prolongations of the dural sheath extend around the spinal roots and nerves into the lateral zones of the vertebral canal and out into the 'root' canals, eventually fusing with the epineurium of the spinal nerves. Between the theca and the walls of the vertebral canal is the epidural (spinal extradural) space (p. 778), which is loosely filled with fat, connective tissue containing small arteries and lymphatics, and an important venous plexus. Threedimensional appreciation of the anatomy of the spinal theca and its surroundings is essential for the efficient management of spinal pain and of spinal injuries, tumours and infections. Equally significant clinically is the anatomy of the often precarious blood supply of the spinal cord and its associated structures. The increasing application and refinement of diagnostic imaging and endoscopic procedures lend a new importance to topographical detail here.
© 2008 Elsevier
SPINAL CORD (MEDULLA) (Figs 46.1, 46.2, 46.3, 46.4, 46.5, 46.6) The spinal cord is an elongated, approximately cylindrical part of the CNS, occupying the superior two-thirds of the vertebral canal. Its average length in European males is 45 cm, its weight c.30 g. (For dimensional data consult Barson & Sands 1977.) It extends from the upper border of the atlas to the junction between the first and second lumbar vertebrae: this lower level varies, and there is some correlation with the length of the trunk, especially in females. The termination may be as high as the caudal third of the twelfth thoracic vertebra or as low as the disc between the second and third lumbar vertebra, and its position rises slightly in vertebral flexion. The spinal cord is enclosed in the dura, arachnoid and pia mater, separated from each other by the subdural and subarachnoid spaces respectively. The former is a potential space, while the latter contains cerebrospinal fluid (CSF). The cord is continuous cranially with the medulla oblongata, and narrows caudally to the conus medullaris, from whose apex a connective tissue filament, the filum terminale, descends to the dorsum of the first coccygeal vertebral segment. The spinal cord varies in transverse width, gradually tapering craniocaudally, except at the levels of the enlargements. It is not cylindrical, being wider transversely at all levels, especially in the cervical segments. The cervical enlargement is the source of the large spinal nerves which supply the upper limbs. It extends from the third cervical to the second thoracic segments, its maximum circumference (c.38 mm) is in the sixth cervical segment. (A spinal cord segment provides the attachment of the rootlets of a pair of spinal nerves.) The lumbar enlargement corresponds to the innervation of the lower limbs, and extends from the first lumbar to the third sacral segments, the equivalent vertebral levels being the ninth to twelfth thoracic vertebrae. The greatest circumference (c.35 mm) is near the lower part of the body of the twelfth thoracic vertebra, below which it rapidly dwindles into the conus medullaris. Fissures and sulci extend along most of the external surface. An anterior median fissure and a posterior median sulcus and septum almost completely separate the cord into right and left halves, but they are joined by a commissural band of nervous tissue which contains a central canal. The anterior median fissure extends along the whole ventral surface with an average depth of 3 mm, although it is deeper at caudal levels. It contains a reticulum of pia mater. Dorsal to it is the anterior white commissure. Perforating branches of the spinal vessels pass from the fissure to the commissure to supply the central spinal region. The posterior median sulcus is shallower, and from it a posterior median septum of neuroglia penetrates more than halfway into the cord, almost to the central canal. The septum varies in anteroposterior extent from 4 to 6 mm, and diminishes caudally as the canal becomes more dorsally placed and the cord contracts.
page 775 page 776
Figure 46.1 Median sagittal section of the lumbosacral part of the vertebral column to show the conus medullaris and filum terminale. The section has opened up the subarachnoid space as far as the first sacral vertebra. Note the difference in levels between the inferior limits of the spinal cord and meninges. Note that there are two inaccuracies in this figure retained from an earlier edition: the epidural space is not shown; the fibres of interspinous ligaments should slope dorsocranially.
Figure 46.2 The main features of the spinal cord.
page 776 page 777
Figure 46.3 (Right) Brain and spinal cord with attached spinal nerve roots and dorsal root ganglia, photographed from the dorsal aspect. Note the fusiform cervical and lumbar enlargements of the cord, and the changing obliquity of the spinal nerve roots as the cord is descended. The cauda equina is undisturbed on the right but has been spread out on the left to show its individual components. (Dissection by MCE Hutchinson, GKT School of Medicine; photograph by Kevin Fitzpatrick on behalf of GKT School of Medicine, London.)
Figure 46.4 Lower end of spinal cord, filum terminale and cauda equina exposed from behind. The dura mater and the arachnoid have been opened and spread out.
Figure 46.5 Transverse section of the spinal cord at a mid-thoracic level.
Figure 46.6 Diagram of a spinal cord segment showing mode of formation of a typical spinal nerve and the gross relationships of the grey and white matter. Note dorsal nerve rootlets in a single linear row, ventral rootlets in three or more rows.
A posterolateral sulcus exists from 1.5 to 2.5 mm lateral to each side of the posterior median sulcus. Dorsal roots (strictly rootlets) of spinal nerves enter the cord along the sulcus. The white substance between the posterior median and posterolateral sulcus on each side is the posterior funiculus. In cervical and upper thoracic segments a longitudinal posterointermediate sulcus marks a septum dividing each posterior funiculus into two large tracts: the fasciculus gracilis
(medial) and fasciculus cuneatus (lateral). Between the posterolateral sulcus and anterior median fissure is the anterolateral funiculus. This is subdivided into anterior and lateral funiculi by ventral spinal roots which pass through its substance to issue from the surface of the cord. The anterior funiculus is medial to, and includes, the emerging ventral roots, whilst the lateral funiculus lies between the roots and the posterolateral sulcus. In upper cervical segments, nerve rootlets emerge through each lateral funiculus to form the spinal accessory nerve which ascends in the vertebral canal lateral to the spinal cord and enters the posterior cranial fossa via the foramen magnum (Fig. 46.7). The filum terminale, a filament of connective tissue c.20 cm long, descends from the apex of the conus medullaris. Its upper 15 cm, the filum terminale internum, is continued within extensions of the dural and arachnoid meninges and reaches the caudal border of the second sacral vertebra. Its final 5 cm, the filum terminale externum, fuses with the investing dura mater, and then descends to the dorsum of the first coccygeal vertebral segment. The filum is continuous above with the spinal pia mater. A few strands of nerve fibres which probably represent roots of rudimentary second and third coccygeal spinal nerves adhere to its upper part. The central canal is continued into the filum for 5-6 mm. A capacious part of the subarachnoid space surrounds the filum terminale internum, and is the site of election for access to the CSF (lumbar puncture).
© 2008 Elsevier
DORSAL AND VENTRAL ROOTS (Fig. 46.6) (See also p. 781.) The paired dorsal and ventral roots of the spinal nerves are continuous with the spinal cord. They cross the subarachnoid space and traverse the dura mater separately, uniting in or close to their intervertebral foramina to form the (mixed) spinal nerves. Since the spinal cord is shorter than the vertebral column, the more caudal spinal roots descend for varying distances around and beyond the cord to reach their corresponding foramina. In so doing they form, mostly distal to the apex of the cord, a divergent sheaf of spinal nerve roots, the cauda equina, which is gathered round the filum terminale in the spinal theca (see also p. 781). Ventral spinal roots contain efferent somatic and, at some levels, efferent sympathetic, nerve fibres which emerge from their spinal sources. There are also afferent nerve fibres in these roots. The rootlets comprising each ventral root emerge from the anterolateral sulcus over an elongated vertical elliptical area. Dorsal spinal roots bear ovoid swellings, the spinal ganglia, one on each root proximal to its junction with a corresponding ventral root in an intervertebral foramen. Each root fans out into six to eight rootlets before entering the cord in a vertical row in the posterolateral sulcus. Dorsal roots are usually said to contain only afferent axons (both somatic and visceral) from unipolar neurones in spinal root ganglia, but they may also contain a small number (3%) of efferent fibres and autonomic vasodilator fibres. page 777 page 778
Figure 46.7 Dissection showing the brain stem and upper five cervical spinal segments after removal of large portions of the occipital and parietal bones and the cerebellum together with the roof of the fourth ventricle. On the left, the foramina transversaria of the atlas and of the third, fourth and fifth cervical vertebrae have been opened to expose the vertebral artery. On the right, the posterior arch of the atlas and the laminae of the succeeding cervical vertebrae have been removed.
Each ganglionic neurone has a single short stem which divides into a medial branch which enters the spinal cord via a dorsal root, and a lateral branch which passes peripherally to a sensory end organ. The central branch is an axon while the peripheral one is an elongated dendrite (but when traversing a peripheral nerve is, in general structural terms, indistinguishable from an axon). The region of spinal cord associated with the emergence of a pair of nerves is a spinal segment, but there is no actual surface indication of segmentation. Moreover, the deep neural sources or destinations of radicular fibres may lie far beyond the confines of the 'segment' so defined.
© 2008 Elsevier
MENINGES DURA MATER The single layer of dura which lines the cranial cavity divides into two layers as it passes downwards through the foramen magnum, although it is still a single layer as it forms the anterior and posterior atlanto-occipital membranes. Within the vertebral column, it has been suggested that the outer endosteal layer becomes the periosteum of the vertebral canal, which is separated from the spinal dura mater by an extradural (epidural) space (see below). This interpretation, which would make the epidural space 'intradural', is not generally agreed (see Newell 1999). The spinal dura mater forms a tube whose upper end is attached to the edge of the foramen magnum and to the posterior surfaces of the second and third cervical vertebral bodies, and also by fibrous bands to the posterior longitudinal ligament, especially towards the caudal end of the vertebral canal. The dural tube narrows at the lower border of the second sacral vertebra. It invests the thin spinal filum terminale, descends to the back of the coccyx, and blends with the periosteum. For details of the meningeal coverings of the spinal roots and nerves see page 782. Epidural space (Fig. 46.8) page 778 page 779
The epidural space lies between the spinal dura mater and the tissues which line the vertebral canal. It is closed above by fusion of the spinal dura with the edge of the foramen magnum, and below by the posterior sacrococcygeal ligament which closes the sacral hiatus. It contains loosely packed connective tissue, fat, a venous plexus, small arterial branches, lymphatics and fine fibrous bands which connect the theca with the lining tissue of the vertebral canal. These bands, the meningovertebral ligaments, are best developed anteriorly and laterally. Similar bands tether the nerve root sheaths or 'sleeves' within their canals. There is also a midline attachment from the posterior spinal dura to the ligamentum nuchae at atlanto-occipital and atlanto-axial levels (Dean & Mitchell 2002). The venous plexus consists of longitudinally arranged chains of vessels, connected by circumdural venous 'rings'. The anteriorly placed vessels receive the basivertebral veins.
Figure 46.8 The epidural space. Adapted with permission from Rosse C & GaddumRosse P 1997. Hollinshead's Textbook of Anatomy 5th edn. Philadelphia: LippincottRaven, Fig. 13-3.
The shape of the space within each spinal segment is not uniform, though the segmental pattern is metamerically repeated. It is difficult to define the true shape of the 'space', because it changes with the introduction of fluid or as a result of preservation techniques. In the lumbar region, the dura mater is apposed to the walls of the vertebral canal anteriorly and attached by connective tissue in a
manner that permits displacement of the dural sac during movement and venous engorgement. Adipose tissue is present posteriorly in recesses between the ligamentum flavum and the dura. The connective tissue extends for a short distance through the intervertebral foramina along the sheaths of the spinal nerves. Like the main thecal sac, the root sheaths are partially tethered to the walls of the foramina by fine meningovertebral ligaments. Contrast media and other fluids injected into the epidural space at the sacral level can spread up to the cranial base. Local anaesthetics injected near the spinal nerves, just outside the intervertebral foramina, may spread up or down the epidural space to affect the adjacent spinal nerves or may pass to the opposite side. The paravertebral spaces of each side communicate via the epidural space, particularly at lumbar levels. For a review of the morphology of the epidural space, see Newell (1999). Subdural space
The subdural space is a potential space in the normal spine because the arachnoid and dura are closely apposed (Haines et al 1993). It does not connect with the subarachnoid space, but continues for a short distance along the cranial and spinal nerves. Accidental subdural catheterization may occur during extradural injections. Injection of fluid into the subdural space may either damage the cord by direct toxic effects or by compression of the vasculature.
ARACHNOID MATER (Figs 46.9, 46.10)
page 779 page 780
Figure 46.9 Part of the spinal cord exposed from the anterior aspect to show meningeal coverings.
Figure 46.10 Transverse section through the spinal cord and meninges to show the relationships between the meninges and ligaments with the spinal cord and roots: dura mater (yellow), outer layer of arachnoid mater (pale blue), intermediate layer of arachnoid mater (dark blue), pia mater (pink), subpial connective tissue (green).
The arachnoid mater which surrounds the spinal cord is continuous with the cranial arachnoid mater. It is closely applied to the deep aspect of the dura mater. At sites where vessels and nerves enter or leave the subarachnoid space, the arachnoid mater is reflected on to the surface of these structures and forms a thin coating of leptomeningeal cells over the surface of both vessels and nerves. Thus a subarachnoid angle is formed as nerves pass through the dura into the intervertebral foramina. At this point, the layers of leptomeninges fuse and become continuous with the perineurium. The epineurium is in continuity with the dura. Such an arrangement seals the subarachnoid space so that particulate matter does not pass directly from the subarachnoid space into nerves. The existence of a pathway of lymphatic drainage from the CSF is controversial.
PIA MATER (Figs 46.9, 46.10) The spinal pia mater closely invests the surface of the spinal cord and passes into the anterior median fissure. As in the cranial region, there is a subpial 'space', however over the surface of the spinal cord the subpial collagenous layer is thicker than in the cerebral region, and it is continuous with the collagenous core of the ligamentum denticulatum. The ligamentum denticulatum is a flat, fibrous sheet which lies on each side of the spinal cord between the ventral and dorsal spinal roots. Its medial border is continuous with the subpial connective tissue of the cord and its lateral border forms a series of triangular processes, the apices of which are fixed at intervals to the dura mater. There are usually 21 processes on each side. The first crosses behind the vertebral artery where it is attached to the dura mater, and is separated by the artery from the first cervical ventral root. Its site of attachment to the dura mater is above the rim of the foramen magnum, just behind the
hypoglossal nerve: the spinal accessory nerve ascends on its posterior aspect (Fig. 46.7). The last of the dentate ligaments lies between the exiting twelfth thoracic and first lumbar spinal nerves and is a narrow, oblique band which descends laterally from the conus medullaris. Changes in the form and position of the dentate ligaments during spinal movements have been demonstrated by cineradiography. Beyond the conus medullaris, the pia mater continues as a coating of the filum terminale.
INTERMEDIATE LAYER (Fig 46.10) In addition to the well-defined coats of arachnoid and pia mater, the cord is also surrounded by an extensive intermediate layer of leptomeninges. This layer is concentrated in the dorsal and ventral regions and forms a highly perforated, almost lace-like structure which is focally compacted to form the dorsal, dorsolateral and ventral ligaments of the spinal cord. Dorsally, the intermediate layer is adherent to the deep aspect of the arachnoid mater and forms a discontinuous series of dorsal ligaments which attach the spinal cord to the arachnoid. The dorsolateral ligaments are more delicate and fenestrated, and they extend from the dorsal roots to the parietal arachnoid. As the intermediate layer spreads laterally over the dorsal surface of the dorsal roots, it becomes increasingly perforated and eventually disappears. A similar arrangement is seen over the ventral aspect of the spinal cord, but the intermediate layer is less substantial. The intermediate layer is structurally similar to the trabeculae which cross the cranial subarachnoid space, in that a collagenous core is coated by leptomeningeal cells. The intermediate layers of leptomeninges around the spinal cord may act as a baffle within the subarachnoid space to dampen waves of CSF in the spinal column. Inflammation within the spinal subarachnoid space may result in extensive fibrosis within the intermediate layer and the complications of chronic arachnoiditis.
© 2008 Elsevier
CEREBROSPINAL FLUID (CSF) The cerebrospinal fluid is described in detail on page 292. Though there is free communication between the spinal and cerebral subarachnoid spaces, the mode of circulation of the spinal CSF remains uncertain. Spinal arachnoid granulations have been described.
© 2008 Elsevier
SPINAL NERVES
page 780 page 781
Figure 46.11 Formation and branching pattern of a typical spinal nerve.
In those body segments which largely retain a metameric (segmental) structure, e.g. the thoracic region, spinal nerves show a common plan (Fig. 46.11). The dorsal, epaxial, ramus passes back lateral to the articular processes and divides into medial and lateral branches which penetrate the deeper muscles of the back: both branches innervate the adjacent muscles and supply a band of skin from the posterior median line to the scapular line (p. 782). The ventral, hypaxial, ramus is connected to a corresponding sympathetic ganglion by white and grey rami communicantes. It innervates the prevertebral muscles and curves round in the body wall to supply the lateral muscles of the trunk. Near the midaxillary line it gives off a lateral branch which pierces the muscles and divides into anterior and posterior cutaneous branches. The main nerve advances in the body wall, where it supplies the ventral muscles and terminates in branches to the skin. Spinal nerves are united ventral and dorsal spinal roots, attached in series to the sides of the spinal cord. The term spinal nerve strictly applies only to the short segment after union of the roots and before branching occurs. This segment, the spinal nerve proper, lies in the intervertebral foramen: in clinical practice it is often loosely termed the 'nerve root'. There are 31 pairs of spinal nerves: 8 cervical, 12 thoracic, 5 lumbar, 5 sacral, 1 coccygeal. The abbreviations C, T, L, S and Co, with appropriate numerals, are commonly applied to individual nerves. The nerves emerge through intervertebral foramina. At thoracic, lumbar, sacral and coccygeal levels the numbered nerve exits the vertebral canal by passing below the pedicle of the corresponding vertebra, e.g. L4 nerve exits the intervertebral foramen between L4 and L5. However, in the cervical region, nerves C1-7 pass above their corresponding vertebrae. C1 leaves the vertebral canal between the occipital bone and atlas and hence is often termed the suboccipital nerve. The last pair of cervical nerves does not have a correspondingly numbered vertebra and C8 passes between the seventh cervical and first thoracic vertebrae. Each nerve is continuous with the spinal cord by ventral and dorsal roots; the latter each bears a spinal ganglion ('dorsal root ganglion').
SPINAL ROOTS AND GANGLIA Ventral (anterior) roots
Ventral roots contain axons of neurones in the anterior and lateral spinal grey columns. Each emerges as a series of rootlets in two or three irregular rows in an area c.3 mm in horizontal width.
Dorsal (posterior) roots
Dorsal roots contain centripetal processes of neurones sited in the spinal ganglia. Each consists of medial and lateral fascicles which both diverge into rootlets which enter along the posterolateral sulcus. The rootlets of adjacent dorsal roots are often connected by oblique filaments, especially in the lower cervical and lumbosacral regions. Little is known of the detail of the regions of entry and emergence of afferent and efferent rootlets in humans, but these zones of transition between the central and peripheral nervous systems have been extensively described in rodents (Fraher 2000) (see also p. 65). Appearance and orientation of roots at each spinal level
The size and direction of spinal nerve roots vary. The upper four cervical roots are small, the lower four are large. Cervical dorsal roots have a thickness ratio to the ventral roots of 3:1, which is greater than in other regions. The first dorsal root is an exception, being smaller than the ventral and it is occasionally absent. The conventional view is that the first and second cervical spinal roots are short, running almost horizontally to their exits from the vertebral canal, and that from the third to the eighth cervical levels the roots slope obliquely down. Obliquity and length increase successively, although the distance between spinal attachment and vertebral exit never exceeds the height of one vertebra. An alternative view (Kubik & Müntener 1969) states that upper cervical roots descend, the fifth is horizontal, the sixth to eighth ascend, the first two thoracic roots are horizontal, the next three ascend, the sixth is horizontal and the rest descend. This view is based on the observation that the cervicothoracic part of the spinal cord grows more in length than other parts. Thoracic roots, except the first, are small, and the dorsal root only slightly exceeds the ventral in thickness. They increase successively in length. In the lower thoracic region, the roots descend in contact with the spinal cord for at least two vertebrae before emerging from the vertebral canal. Lower lumbar and upper sacral roots are the largest, and their rootlets are the most numerous. Coccygeal roots are the smallest. Kubik & Müntener (1969) confirm that lumbar, sacral and coccygeal roots descend with increasing obliquity to their exits. The spinal cord ends near the lower border of the first lumbar vertebra, and so the lengths of successive roots rapidly increase: the consequent collection of roots is the cauda equina (Fig. 46.3). The largest roots, and hence the largest spinal nerves, are continuous with the spinal cervical and lumbar swellings and innervate the upper and lower limbs. Spinal ganglia (dorsal root ganglia)
Spinal ganglia are large groups of neurones on the dorsal spinal roots. Each is oval and reddish; its size is related to that of its root. A ganglion is bifid medially where the two fascicles of the dorsal root emerge to enter the cord. Ganglia are usually sited in the intervertebral foramina, immediately lateral to the perforation of the dura mater by the roots (Fig. 46.4). However, the first and second cervical ganglia lie on the vertebral arches of the atlas and axis, the sacral lies inside the vertebral canal, and the coccygeal ganglion usually lies within the dura mater. The first cervical ganglia may be absent. Small aberrant ganglia sometimes occur on the upper cervical dorsal roots between the spinal ganglia and the cord.
SPINAL NERVES PROPER (Figs 46.11, 46.12) Immediately distal to the spinal ganglia, ventral and dorsal roots unite to form spinal nerves. These very soon divide into dorsal and ventral rami, both of which receive fibres from both roots. At all levels above the sacral, this division occurs within the intervertebral foramen. Division of the sacral spinal nerves occurs within the sacral vertebral canal, and the dorsal and ventral rami exit separately through posterior and anterior sacral foramina at each level. Spinal nerves trifurcate at some cervical and thoracic levels, and the third branch is called a ramus intermedius. At or distal to its origin each ventral ramus gives off recurrent meningeal (sinuvertebral) branches and receives a grey ramus communicans
from the corresponding sympathetic ganglion. The thoracic and first and second lumbar ventral rami each contributes a white ramus communicans to the corresponding sympathetic ganglia. The second, third and fourth sacral nerves also supply visceral branches, unconnected with sympathetic ganglia, which carry a parasympathetic outflow direct to the pelvic plexuses.
page 781 page 782
Figure 46.12 The constitution of a typical spinal nerve. In the upper part of the diagram the spinal nerve roots show the somatic components, in the lower part the visceral components. Red: somatic efferent and preganglionic sympathetic fibres; blue: somatic and visceral afferent fibres; black: postganglionic sympathetic fibres.
Cervical spinal nerves enlarge from the first to the sixth nerve. The seventh and eighth cervical and the first thoracic nerve are similar in size to the sixth cervical nerve. The remaining thoracic nerves are relatively small. Lumbar nerves are large, increasing in size from the first to the fifth. The first sacral is the largest spinal nerve, thereafter the sacral nerves decrease in size. The coccygeal nerves are the smallest spinal nerves. Meningeal branches
The recurrent meningeal or sinuvertebral nerves number two to four filaments on each side, and occur at all vertebral levels. Each receives one or more rami from a nearby grey ramus communicans or directly from a thoracic sympathetic ganglion, and most then pursue a recurrent (often perivascular) course into the vertebral canal through the intervertebral foramen ventral to the dorsal root ganglion. Here these mixed sensory and sympathetic nerves divide into transverse, ascending and descending branches which are distributed to the dura mater, the walls of blood vessels, the periosteum, ligaments and intervertebral discs in the anterolateral region of the vertebral canal. Fine meningeal branches occasionally pass dorsal to reach the spinal ganglia to innervate the dorsal dura, periosteum and ligaments, and others pass ventrally to the posterior longitudinal
ligament. Ascending branches of the upper three cervical meningeal nerves are large and distributed to the dura mater in the posterior cranial fossa. Meningeal nerves are important in relation to the referred pain which is characteristic of many spinal disorders and in occipital headache. Coverings and relations of the spinal roots and nerves in the radicular canal (Figs 46.9, 46.10, 46.13)
Figure 46.13 A lumbar spinal nerve and its roots and meningeal coverings.
Tubular prolongations of the spinal dura mater, closely lined by the arachnoid, extend around the spinal roots and nerves as they pass through the lateral zone of the vertebral canal and through the intervertebral foramina. These prolongations, the spinal nerve sheaths ('root sheaths'), gradually lengthen as the spinal roots become increasingly oblique. Each individual dorsal and ventral root runs in the subarachnoid space with its own covering of pia mater. Each root pierces the dura separately, taking a sleeve of arachnoid with it, before joining within the dural prolongation just distal to the spinal ganglion. The dural sheaths of the spinal nerves fuse with the epineurium, within or slightly beyond the intervertebral foramina. The arachnoid prolongations within the sheaths do not extend as far distally as their dural coverings, but the subarachnoid space and its contained CSF extend sufficiently distally to form a radiologically demonstrable 'root sleeve' for each nerve. Shortening or obstruction of this sleeve seen on the radiculogram indicates compression of the spinal nerve. At the cervical level, where the nerves are short and the vertebral movement is greatest, the dural sheaths are tethered to the periosteum of the adjacent transverse processes. In the lumbosacral region there is less tethering of the dura to the periosteum, though there may be an attachment posteriorly to the facet joint capsule. In the radicular ('root') canal and intervertebral foramen, the spinal nerve is related to the spinal artery of that level and its radicular branch, and to a small plexus of veins. At the outer end of the foramen the nerve may lie above or below transforaminal ligaments. The size of the spinal nerve and its associated structures within the intervertebral
foramen is not in direct relation to the size of the foramen. At lumbar levels, though L5 is the largest nerve, its foramen is smaller than those of L1-4, which renders this nerve particularly liable to compression. Functional components of spinal nerves
Each typical spinal nerve contains somatic and visceral (autonomic) fibres. Somatic components
Somatic components are efferent and afferent. Somatic efferent fibres which innervate skeletal muscles are axons of !, " and # neurones in the spinal anterior grey column. Somatic afferent fibres convey impulses into the CNS from receptors in the skin, subcutaneous tissue, muscles, tendons, fasciae and joints: they are peripheral processes of unipolar neurones in the spinal ganglia. Visceral components
Visceral components are also afferent and efferent, and belong to the autonomic nervous system. They include sympathetic or parasympathetic fibres at different spinal levels. Preganglionic visceral efferent sympathetic fibres are axons of neurones in the spinal lateral grey column in the thoracic and upper two or three lumbar segments: they join the sympathetic trunk along corresponding white rami communicantes and synapse with postganglionic neurones distributed to nonstriated muscle or glands. The preganglionic visceral efferent parasympathetic fibres are axons of neurones in the spinal lateral grey column of the second to fourth sacral segments: they leave the ventral rami of corresponding sacral nerves and synapse in pelvic ganglia. The postganglionic axons are distributed mainly to muscle or glands in the walls of the pelvic viscera. Visceral afferent fibres have cell bodies in the spinal ganglia. Their peripheral processes pass through white rami communicantes and, without synapsing, through one or more sympathetic ganglia to end in viscera. Some visceral afferent fibres may enter the spinal cord in the ventral roots. Central processes of ganglionic unipolar neurones enter the spinal cord by posterior roots and synapse on somatic or sympathetic efferent neurones, usually through interneurones, completing reflex paths. Alternatively, they may synapse with other neurones in the spinal or brain stem grey matter which give origin to a variety of ascending tracts. Variations of spinal roots and nerves
The courses of spinal roots and nerves in relation to the thecal sac and vertebral and radicular canals may be aberrant. An individual intervertebral foramen may contain a duplicated sheath, nerve and roots, which will then be absent at an adjacent level. Abnormal communications between roots may occur within the vertebral canal. These anomalies have been described and classified for the lumbosacral spine by Neidre & Macnab (1983). Rami of the spinal nerves page 782 page 783
Ventral (anterior primary) rami supply the limbs and the anterolateral aspects of the trunk, and in general are larger than the dorsal rami. Thoracic ventral rami run independently and retain a largely segmental distribution. Cervical, lumbar and sacral ventral rami connect near their origins to form plexuses. Dorsal rami do not join these plexuses. The ventral rami are described in the appropriate regional Sections. Dorsal (posterior primary) rami of spinal nerves are usually smaller than the ventral rami and are directed posteriorly. Retaining a segmental distribution, all, except for the first cervical, fourth and fifth sacral and the coccygeal, divide into medial and lateral branches which supply the muscles and skin of the posterior regions of the neck and trunk (Fig. 46.14). Cervical dorsal spinal rami
Each cervical spinal dorsal ramus, except the first, divides into medial and lateral branches which all innervate muscles. In general only medial branches of the
second to fourth, and usually the fifth, supply the skin. Except for the first and second, each dorsal ramus passes back medial to a posterior intertransverse muscle, curving round the articular process into the interval between semispinalis capitis and semispinalis cervicis. First cervical dorsal ramus (suboccipital nerve) (Fig. 54.58)
Figure 46.14 Cutaneous distribution of the dorsal rami of the spinal nerves. The nerves are shown lying on the superficial muscles; on the left side the limit of the skin area supplied by these nerves is indicated by the dotted line. The nerves are numbered on the right side; the spines of the seventh cervical, sixth and twelfth thoracic, and first and fifth lumbar vertebrae are labelled in bold on the left side.
The first cervical dorsal ramus, the suboccipital nerve, is larger than the ventral. It emerges superior to the posterior arch of the atlas and inferior to the vertebral artery and enters the suboccipital triangle to supply rectus capitis posterior major and minor, obliquus capitis superior and inferior, and semispinalis capitis. A filament from the branch to the inferior oblique joins the second dorsal ramus. The suboccipital nerve occasionally has a cutaneous branch which accompanies the occipital artery to the scalp, and connects with the greater and lesser occipital nerves. It may also communicate with the accessory nerve. Second cervical dorsal ramus (Figs 54.58, 29.9B) The second cervical dorsal ramus is slightly larger than the ventral and all the other cervical dorsal rami. It emerges between the posterior arch of the atlas and the lamina of the axis, below inferior oblique, which it supplies. It receives a connection from the first cervical dorsal ramus and divides into a large medial and
smaller lateral branch. The medial branch, termed the greater occipital nerve, ascends between inferior oblique and semispinalis capitis, pierces the latter and trapezius near their occipital attachments, and is joined by a filament from the medial branch of the third dorsal ramus. It ascends with the occipital artery, divides into branches which connect with the lesser occipital nerve, and supplies the skin of the scalp as far forward as the vertex. It supplies semispinalis capitis and, occasionally, the back of the auricle. The lateral branch supplies splenius capitis, longissimus capitis and semispinalis capitis, and is often joined by the corresponding third cervical branch. Greater occipital neuralgia
Greater occipital neuralgia is a syndrome of pain and paraesthesiae felt in the distribution of the greater occipital nerve. It is usually due to an entrapment neuropathy of the nerve as it pierces the attachment of the neck extensors to the occiput. A similar syndrome may be caused by upper facet joint arthritis involving the second cervical root. Third cervical dorsal ramus The third cervical dorsal ramus is intermediate in size between the second and fourth. It courses back round the articular pillar of the third cervical vertebra, medial to the posterior intertransverse muscle, and divides into medial and lateral branches. Its medial branch runs between spinalis capitis and semispinalis cervicis, and pierces the splenius and trapezius to end in the skin. Deep to trapezius it gives rise to a branch, the third occipital nerve, which pierces trapezius to end in the skin of the lower occipital region, medial to the greater occipital nerve and connected to it. The lateral branch often joins a branch of the second cervical dorsal ramus. The dorsal ramus of the suboccipital nerve and medial branches of the dorsal rami of the second and third cervical nerves are sometimes joined by loops to form the posterior cervical plexus. Dorsal rami of the lower five cervical nerves The dorsal rami of the lower five cervical nerves curve back round the vertebral articular pillars and divide into medial and lateral branches. Medial branches of the fourth and fifth run between semispinalis cervicis and semispinalis capitis, reach the vertebral spines and pierce splenius and trapezius to end in the skin. The fifth medial branch may not reach the skin. The medial branches of the lowest three cervical nerves are small and end in semispinalis cervicis, semispinalis capitis, multifidus and interspinales. The lateral branches supply iliocostalis cervicis, longissimus cervicis and longissimus capitis. Thoracic dorsal spinal rami
Thoracic dorsal rami pass backwards close to the vertebral facet joints to divide into medial and lateral branches. Each medial branch emerges between a joint and the medial edges of the superior costotransverse ligament and intertransverse muscle. Each lateral branch runs in the interval between the ligament and the muscle before inclining posteriorly on the medial side of levator costae. Medial branches of the upper six thoracic dorsal rami pass between and supply the semispinalis thoracis and multifidus, then pierce the rhomboids and trapezius, and reach the skin near the vertebral spines. page 783 page 784
Medial branches of the lower six thoracic dorsal rami mainly supply multifidus and longissimus thoracis and occasionally the skin in the median region. Lateral branches increase inferiorly in size, and run through, or deep to, longissimus thoracis to the interval between it and iliocostalis cervicis, supplying these muscles and the levatores costarum. The lower five or six also have cutaneous branches, and pierce serratus posterior inferior and latissimus dorsi in line with the costal angles. Some upper thoracic lateral branches supply the skin. The twelfth thoracic lateral branch sends a filament medially along the iliac crest, then passes down to the anterior gluteal skin. Medial cutaneous branches of the thoracic dorsal rami descend close to the vertebral spines before reaching the skin; lateral branches descend across as many as four ribs before becoming
superficial. The branch of the twelfth thoracic reaches the skin a little above the iliac crest. Lumbar dorsal spinal rami
Lumbar dorsal rami pass back medial to the medial intertransverse muscles, and divide into medial and lateral branches. Medial branches run near the vertebral articular processes to end in the multifidus. They are related to the bone between the accessory and mammillary processes and may groove it, crossing a distinct notch or even a foramen. Lateral branches supply the erector spinae. In addition, the upper three rami give rise to cutaneous nerves which pierce the aponeurosis of latissimus dorsi at the lateral border of the erector spinae and cross the iliac crest posteriorly to reach the gluteal skin, some reaching as far as the level of the greater trochanter. Sacral dorsal spinal rami
Sacral dorsal rami are small, diminishing downwards, and other than the fifth, all emerge though the dorsal sacral foramina. The upper three are covered at their exit by multifidus, and divide into medial and lateral branches. Medial branches are small and end in multifidus. Lateral branches join together and with lateral branches of the last lumbar and fourth sacral dorsal rami to form loops dorsal to the sacrum. Branches from these loops run dorsal to the sacrotuberous ligament and form a second series of loops under gluteus maximus. From these, two or three gluteal branches pierce gluteus maximus (along a line from the posterior superior iliac spine to the coccygeal apex) to supply the posterior gluteal skin. The dorsal rami of the fourth and fifth sacral nerves are small and lie below multifidus. They unite with each other and with the coccygeal dorsal ramus to form loops dorsal to the sacrum: filaments from these supply the skin over the coccyx. Coccygeal dorsal spinal ramus
The coccygeal dorsal spinal ramus does not divide into medial and lateral branches. Its connections and distribution are noted above.
VASCULAR SUPPLY OF SPINAL CORD, ROOTS AND NERVES Arteries (Fig. 46.15) (See Crock 1996.)
The spinal cord, its roots and nerves are supplied with blood by both longitudinal and segmental vessels. Three major longitudinal vessels, a single anterior and two posterior spinal arteries (each of which is sometimes doubled to pass on either side of the dorsal rootlets), originate intracranially from the vertebral artery and terminate in a plexus around the conus medullaris. The anterior spinal artery forms from the fused anterior spinal branches of the vertebral artery, and descends in the anterior median fissure of the cord. Each posterior spinal artery originates either directly from the ipsilateral vertebral artery or from its posterior inferior cerebellar branch, and descends in a posterolateral sulcus of the cord. The segmental arteries are derived in craniocaudal sequence from spinal branches of the vertebral, deep cervical, intercostal and lumbar arteries. These vessels enter the vertebral canal through the intervertebral foramina and anastomose with branches of the longitudinal vessels to form a pial plexus on the surface of the cord. The segmental spinal arteries send anterior and posterior radicular branches to the spinal cord along the ventral and dorsal roots. Most anterior radicular arteries are small, and end in the ventral nerve roots or in the pial plexus of the cord. The small posterior radicular arteries also supply the dorsal root ganglia: branches enter at both ganglionic poles to be distributed around ganglion cells and nerve fibres. Segmental medullary feeder arteries
Some radicular arteries, mainly situated in the lower cervical, lower thoracic and upper lumbar regions, are large enough to reach the anterior median sulcus where they divide into slender ascending and large descending branches. These are the anterior medullary feeder arteries (Dommisse 1975). They anastomose with the anterior spinal arteries to form a single or partly double longitudinal
vessel of uneven calibre along the anterior median sulcus. The largest anterior medullary feeder, the great anterior segmental medullary artery of Adamkiewicz, varies in level, arising from a spinal branch of either one of the lower posterior intercostal arteries (T9-11), or of the subcostal artery (T12), or less frequently of the upper lumbar arteries (L1 and L2). It most often arises on the left side (Carmichael & Gloviczki 1999). Reaching the spinal cord, it sends a branch to the anterior spinal artery below and another to anastomose with the ramus of the posterior spinal artery which lies anterior to the dorsal roots. It may be the main supply to the lower two-thirds of the cord. Central branches of the anterior spinal artery enter the anterior median fissure, and then turn right or left to supply the ventral grey column, the base of the dorsal grey column, including the dorsal nucleus, and the adjacent white matter (Fig. 46.16). Each posterior spinal artery contributes to a pair of longitudinal anastomotic channels, anterior and posterior to the dorsal spinal roots. These are reinforced by posterior medullary feeders from the posterior radicular arteries. The latter are variable in number and size, but smaller, more numerous and more evenly distributed than the anterior medullary feeders. The anterior channel is joined by a ramus from the descending branch of the great anterior segmental medullary artery of Adamkiewicz. In all longitudinal spinal arteries the width of the lumen is uneven, and complete interruptions may occur. At the conus medullaris they communicate by anastomotic loops. Anastomoses other than those between the pial or peripheral spinal arterial branches may be important, e.g. a posterior spinal series of anastomoses between rami of the dorsal divisions of segmental arteries near the spinous processes. Intramedullary arteries
The central branches of the anterior spinal artery supply about two-thirds of the cross-sectional area of the cord. The rest of the dorsal grey and white columns and peripheral parts of the lateral and ventral white columns are supplied by numerous small radial vessels which branch from posterior spinal arteries and the pial plexus. In a microangiographic study of the human cervical spinal cord, up to six anterior, and eight posterior, radicular spinal arteries were described, and up to eight central branches arose from each centimetre of the anterior spinal artery (Turnbull et al 1966). Spinal cord ischaemia
The spinal cord can rely neither for its transverse nor for its longi-tudinal blood supply entirely on the longitudinal arteries. The anterior longitudinal artery and the intramedullary arteries are functional end-arteries, although overlap of territories of supply has been described. Damage to the anterior longitudinal artery can result in loss of function of the anterior two-thirds of the cord. The longitudinal arteries cannot supply the whole length of the cord, and the input of the segmental medullary feeder vessels is essential. This is especially true of the artery of Adamkiewicz (great anterior segmental medullary artery), which may effectively carry the major supply for the lower cord. The midthoracic cord, distant from the main anterior medullary feeders, is particularly liable to become ischaemic after periods of hypotension. Veins (Fig. 46.17)
The venous drainage of the spinal cord follows a similar pattern to that of its arterial supply (Gillilan 1970). Intramedullary veins within the substance of the cord drain into a plexus of surface veins, the coronal plexus. There are six tortuous longitudinal channels within this plexus, one in each of the anterior and posterior median fissures, and four others which run on either side of the ventral and dorsal nerve roots. Only the anterior median vein, which drains the central grey matter, is consistently complete. These vessels connect freely. They drain superiorly into the cerebellar veins and cranial sinuses, and segmentally mainly into medullary veins. These segmental veins drain into the intervertebral veins and thence into the external vertebral venous plexuses, the caval and azygos systems. Segmental veins
Anterior and posterior medullary veins run along some of the ventral and dorsal roots. They are larger than radicular veins, and drain the cord but not the roots themselves. Like the medullary feeder arteries, they are largest in the cervical and lumbar regions of the cord, but do not necessarily occur in the same segments as the medullary feeders. Anterior and posterior great medullary veins may arise in the lower thoracic or upper lumbar cord segments. There are 8-14 anterior medullary veins. Posterior medullary veins are more numerous. page 784 page 785
page 785 page 786
Figure 46.15 Arteries of the spinal cord. (Netter Illustrations used with permission from Icon Learning Systems, a division of MediMedia USA, Inc. All rights reserved.)
Figure 46.16 Arterial disposition within the spinal cord. (Netter Illustrations used with permission from Icon Learning Systems, a division of MediMedia USA, Inc. All rights reserved.)
Very small anterior and posterior radicular veins occur in most spinal segments, accompanying and draining the ventral and dorsal roots and some of the cord at the points of entry and exit of the rootlets. They usually drain into the intervertebral veins.
© 2008 Elsevier
SPINAL CORD INJURY AND VERTEBRAL COLUMN INJURY In the assessment of a patient with spinal injury and neurological damage, it is important to remember that the level of cord and root injury will not coincide with that of the skeletal damage to the vertebral column.
Figure 46.17 Venous disposition within the spinal cord.
In estimating the vertebral levels of cord segments in the adult, a useful approximation is that in the cervical region the tip of a vertebral spinous process corresponds to the succeeding cord segment (i.e. the sixth cervical spine is opposite the seventh spinal segment); at upper thoracic levels the tip of a vertebral spine corresponds to the cord two segments lower (i.e. the fourth spine is level with the sixth segment), and in the lower thoracic region there is a difference of three segments (i.e. the tenth thoracic spine is level with the first lumbar segment). The eleventh thoracic spine overlies the third lumbar segment and the twelfth is opposite the first sacral segment. In making this estimate by palpation of the vertebral spines, the relationship of the individual spines to their vertebral bodies should be remembered (p. 727). Complete division above the fourth cervical segment causes respiratory failure because of the loss of activity in the phrenic and intercostal nerves. Lesions between C5 and T1 paralyse all four limbs (quadriplegia), the effects in the upper limbs varying with the site of injury: at the fifth cervical segment paralysis is complete; at the sixth, each arm is positioned in abduction and lateral rotation, with the elbow flexed and the forearm supinated, due to unopposed activity in the deltoid, supraspinatus, rhomboid and brachial flexors (all supplied by the fifth cervical spinal nerves). In lower cervical lesions upper limb paralysis is less marked. Lesions of the first thoracic segment paralyse small muscles in the hand and damage the sympathetic outflow, resulting in contraction of the pupil, recession of the eyeball, narrowing of the palpebral fissure and loss of sweating in the face and neck (Horner's syndrome). However, sensation is retained in areas innervated by segments above the lesion, thus cutaneous sensation is retained in the neck and chest down to the second intercostal space, because this area is innervated by the supraclavicular nerves (C3 and C4). At thoracic levels, division
of the cord paralyses the trunk, below the segmental level of the lesion, and both lower limbs (paraplegia). The first sacral neural segment is approximately level with the thoracolumbar vertebral junction: injury, which commonly occurs here, paralyses the urinary bladder, the rectum and muscles supplied by the sacral segments, and cutaneous sensibility is lost in the perineum, buttocks, the back of the thighs and the legs and soles of the feet. The roots of lumbar nerves descending to join the cauda equina may be damaged at this level, causing complete paralysis of both lower limbs. Lesions below the first lumbar vertebra may divide or damage the cauda equina, but severe nerve damage is uncommon and is usually confined to the spinal roots at the level of the trauma. Neurological symptoms may also occur as a result of interference with the spinal blood supply, particularly in the lower thoracic and upper lumbar segments.
SPINAL CORD INJURY WITHOUT RADIOLOGICAL ABNORMALITY: 'SCIWORA' page 786 page 787
The spinal cord may be damaged without radiological evidence of skeletal injury in some injuries to the vertebral column. This is particularly liable to occur if the vertebral canal is abnormally narrowed, usually by osteoarthritic changes. In the elderly patient there may in addition be occlusive arterial disease, directly compromising an already precarious blood supply to the cord (p. 784). This type of injury not uncommonly occurs in hyperextension injuries of the cervical spine in this age group. The cause of the damage may be direct injury to neural tissue by osteophytes or by an infolded ligamentum flavum, or direct or indirect injury to the vasculature of the cord. For cervical spinal injury, several cord syndromes have been described, relating the clinical picture to the anatomy of the neurological lesion within the cord. The commonest of these is central cord syndrome, which usually results from hyperextension injury to an osteoarthritic neck, in which the major injury is to the central grey matter. This gives a greater motor loss in the upper than in the lower limbs, with variable sensory loss. In anterior cord syndrome, which may occur in flexion-compression injuries of the neck, the damage occurs in the area of supply of the anterior spinal artery, sparing the posterior columns. Here the motor loss is usually proportionately greater in the lower than in the upper limbs, while sensory loss is less of a problem.
© 2008 Elsevier
LESIONS OF THE SPINAL ROOTS, NERVES AND GANGLIA The spinal roots, nerves and ganglia may be damaged in the vertebral and 'root' canals and at the intervertebral foramina (p. 741). Neurofibromas may occur on the roots and nerves in the 'root' canals, and as they enlarge become dumb-bell in shape with both an intra- and an extraspinal component in continuity. The clinical picture may thus include paradoxical features as this asymmetrical spaceoccupying lesion grows. Root compression usually presents acutely with pain which may be severe. The pain, paraesthesiae and numbness occur in a dermatomal distribution. It may be difficult to demonstrate sensory loss on the trunk, because of the overlap of the dermatomes. Severe traction injuries of the upper limbs may cause avulsion of spinal roots from the cord in the cervical region.
THE ANATOMY OF PAIN OF SPINAL ORIGIN (See also p. 316.) In the diagnosis and description of pain of spinal origin it is particularly important to distinguish anatomically between radicular ('nerve root') pain, referred pain and radiating pain. The second and third terms are often used imprecisely and their meanings are confused. Radicular pain occurs in spinal nerve (dermatomal) distribution, is well localized and results from involvement of the spinal nerve in the pathological process, e.g. when it is compressed by a disc prolapse. Referred pain is not strictly 'of spinal origin'. The source of the pain is usually a visceral structure whose afferent innervation shares an interneuronal pool in the posterior horn of the spinal cord with the somatic structure in which the pain is felt. The pain may be felt in a dermatome; however, the pain-producing lesion is not in the spinal nerve. Radiating pain does not adopt any particular anatomical distribution. It is often vaguely localized and is described by the patient using the whole of the hand to indicate the affected area. The extent of its area of distribution often relates directly to the severity of the pain. Spinal pain of this type commonly radiates around the hip and down into the thigh.
© 2008 Elsevier
LESIONS OF THE CONUS AND CAUDA EQUINA Lesions of the conus and cauda equina, e.g. tumours, cause bilateral deficit, often with pain in the back extending into the sacral segments and to the legs. Loss of bladder and erectile function can be early features. There are lower motor neurone signs in the legs with fasciculation and muscle atrophy. Sensory loss usually involves the perineal or 'saddle area' as well as involving other lumbar and sacral dermatomes. There may be congenital abnormalities, e.g. spina bifida, lipomata or dystematomyelia, and the conus may extend below the lower border of L1, often with a tethered filum terminale. Extramedullary lesions include prolapsed intervertebral discs. A midline (central) disc protrusion in the lumbar region may present with involvement of only the sacral segments. REFERENCES Barson AJ, Sands J 1977 Regional and segmental characteristics of the human adult spinal cord. J Anat 123: 797-803. Medline Similar articles Bogduk N 1997 Clinical Anatomy of the Lumbar Spine and Sacrum, 3rd edn. Edinburgh: Churchill Livingstone. Carmichael SW, Gloviczki P 1999 Anatomy of the blood supply to the spinal cord: the artery of Adamkiewicz revisited. Perspect Vasc Surg 12: 113-22. Crock HV 1996 An Atlas of Vascular Anatomy of the Skeleton and Spinal Cord. London: Martin Dunitz. Dean NA, Mitchell BS 2002 Anatomic relation between the nuchal ligament (ligamentum nuchae) and the spinal dura mater in the craniocervical region. Clin Anat 15: 182-5. Dommisse GF 1975 The Arteries and Veins of the Human Spinal Cord From Birth. Edinburgh: Churchill Livingstone. Fraher JP 2000 The transitional zone and CNS regeneration. J Anat 196: 137-58. Medline Similar articles Gillilan LA 1970 Veins of the spinal cord. Anatomic details; suggested clinical applications. Neurology 20: 860-8. Medline Similar articles Haines DE, Harkey HL, Al-Mefty O 1993 The 'subdural' space: a new look at an outdated concept. Neurosurgery 32: 111-20. Proposes the view that the subdural 'space' is a pathological cleavage plane rather than a normal anatomical element Medline Similar articles Kubik S, Müntener M 1969 Zur Topographie der spinalen Nervenwurzeln. II Der Einfuss des Wachstums des Duralsackes, sowie der Krümmagen und der Bewegungen der spinalen Nervenwurzeln. Acta Anat 74: 149-68. An alternative view of the obliquity of the cervicothoracic spinal nerve roots based on observations of differential cord growth Medline Similar articles Neidre A, Macnab I 1983 Anomalies of the lumbosacral nerve roots. Spine 8: 294-9. Medline Similar articles Full article Newell RLM 1999 The spinal epidural space. Clin Anat 12: 375-9. Medline Full article
Similar articles
Turnbull IM, Brieg A, Hassler O 1966 Blood supply of cervical spinal cord in man. A microangiographic cadaver study. J Neurosurg 24: 951-65. Medline Similar articles page 787 page 788
© 2008 Elsevier
47 Development of the vertebral column Vertebrae and their alternating intervertebral discs are one of the-main manifestations of body segmentation or metamerism. A chain of segments arranged in sequence allows the overall structure to bend when it is moved by the associated muscles. The original body segments, the somites, provide the embryonic cell populations for bone and muscle. The vertebrae form between the early body segments by the recombination of portions of the somites on the craniocaudal axis, and the muscles attach to adjacent vertebrae. The axial skeleton, the vertebral column and associated muscles, are formed by the paraxial mesenchyme which is found lateral to the neural tube and notochord in the early embryo. Each vertebra develops from bilateral origins to form a midline centrum, two lateral arches, bearing transverse processes, which develop lateral and dorsal to the spinal cord, and a midline fused dorsal portion with a spinous process. Individual vertebrae may be distinguished by modifications of these component parts. The intervertebral discs develop from the same origins as the centra, and are composed of outer dense fibrous tissue surrounding a softer central zone.
© 2008 Elsevier
SEGMENTATION OF PARAXIAL MESENCHYME Epiblast cells which ingress through the lateral aspect of the primitive node and the rostral primitive streak (Chapter 10) become committed to a somitic lineage (Fig. 10.15). After passing through the streak the cells retain contact with both the epiblast and hypoblast basal laminae as they migrate and for some time after reaching their destination. The cells form populations of paraxial mesenchyme on each side of the notochord, termed presomitic or unsegmented mesenchyme. Somites will form from cultured presomitic mesenchyme with or without the presence of neural tube tissue or primitive node tissue. As well as specifying somitic lineage, the position of ingression of the epiblast informs the specific destination of the cells. Thus those cells which ingress through the lateral portion of Hensen's node form the medial halves of the somites, whereas those ingressing through the primitive streak c.200µm caudal to the node produce the lateral halves of the somites. The two somite halves do not appear to intermingle. The segmentation of the paraxial presomitic mesenchyme occurs as a sequential process along the craniocaudal axis. In amniote embryos-a pair of somites is formed every 90 minutes until the full number is obtained. The molecular pathway for this synchronous segmentation has been termed the segmentation clock. It has been identified as a conserved process in vertebrates from fish to mammals, and is based on the rhythmic production of mRNAs for the transcription of genes related to Notch, a large transmembrane receptor, and a number of transmembrane ligands. Intrinsically coordinated pulses of mRNA expression appear as a wave within the presomitic mesenchyme as each somite forms. As new cells enter the paraxial mesenchyme caudally they begin phases of up-regulation of the cycling genes followed by downregulation of these genes. During each cycle the most cranial presomitic mesenchyme will segment to form the next somite. Experimental evidence (from chick embryos) shows that newly formed paraxial mesenchyme cells undergo 12 such cycles before they finally form a somite (Pourquie & Kusumi 2001). Thus from ingression through the primitive streak to segmentation into a somite takes-c.18 hours. As the somite number varies between vertebrate species it is likely that the rate of somite formation also varies. Indeed those vertebrates with elongated bodies and many somites appear to form somites more rapidly than those with shorter bodies, a finding which supports the concept that somite number is controlled in part by species-specific cyclical properties of the presomitic mesenchyme (Richardson et al 1998). The final determination of somitic boundary formation has not yet been fully elucidated, but seems to require a periodic repression of the Notch pathway genes. The caudal presomitic mesenchyme cells are thought to be maintained in an immature state by their production of FGF8. The cells become competent for segmentation when FGF8 levels drop below a certain threshold. They would then be in close apposition to cells that have segmented (i.e. the next cranial somite). This area of research is moving rapidly. A detailed critique of the conceptual
models of segmentation is given by Stern & Vasiliauskas (2000). An overview of the processes involved in the development of the paraxial mesenchyme, based on the work of Christ et al (2000), is shown in Fig. 47.1.
© 2008 Elsevier
SOMITOGENESIS Following segmentation, and once the somite boundaries have been defined, the cells within the somite undergo somitogenesis, a process-in which five main stages can be identified (Figs 47.2, 47.3). These are compaction, formation of the spherical epithelial somite surrounding free somitocoele cells, epithelial/mesenchymal transition of the ventral and ventromedial walls of the somite to form the sclerotome, bilateral migration of the ventromedial mesenchyme populations towards the notochord to form the perinotochordal sheath and of similar cells around the neural tube as the sclerotome, and formation of the epithelial plate of the somite, also termed the dermomyotome, from the remaining somitic epithelium. After the onset of neurulation the paraxial mesenchyme caudal to the otic vesicle undergoes segmentation in a craniocaudal progression and forms discrete clusters of mesenchyme cells: this stage is termed compaction and denotes the start of somitogenesis. The Golgi apparatus, and actin and !-actinin are all located in the apical region of the epithelial somite cells; cilia develop on the free surface. The cells are joined by tight junctions (a variety of cell adhesion molecules has been demonstrated in epithelial somites). The basal surface rests on a basal lamina which contains collagen, laminin, fibronectin and cytotactin. Processes from the somite cells pass through this basal lamina to contact the basal laminae of the neural tube and notochord. Dorsoventral patterning of the putative vertebral column is dependent on the interaction of SHH from the notochord/floor plate with BMP4 from the roof plate and overlying ectoderm. page 789 page 790
Figure 47.1 Processes in the development of the paraxial mesenchyme in the avian embryo. Signals are indicated by open arrows; genes are printed in italics. (Redrawn with permission from Christ B, Huang R, Wilting J. The development of the avian vertebral column. Anat Embryol 202: 179-194, 2000, Springer.)
The epithelial somite undergoes rapid development. The cells of the ventromedial wall undergo an epithelial/mesenchyme transformation and break apart. The newly formed mesenchymal cells, which are collectively termed the sclerotome, proliferate and migrate medially towards the notochord. Initially they provide an axial cell population (now termed the perinotochordal sheath); they subsequently provide lateral sclerotomal populations which will give rise to the bones, joints and ligaments of the vertebral column. The remaining cells of the somite are now termed the epithelial plate of the somite (or dermomyotome). This epithelium is proliferative. It produces the cell lines which will give rise to (nearly) all the striated muscles of the body (see also p. 723). Three separate myogenic lines can be identified. First, cells produced along the craniomedial edge of the epithelial plate elongate from the cranial to the caudal edge on the underside of the basal lamina of the plate. They are collectively termed the myotome and will give rise to the skeletal muscle dorsal to the vertebrae, the epaxial musculature (p. 797). (The term 'myotome' was used previously to describe all the muscle forming cells of the somite. However, it is now usually restricted to cells which are derived from the craniomedial edge.) Second, after initiation of the myotome, cells produced from the ventrolateral edge of the epithelial plate, opposite the limb bud, migrate into the developing limb and give rise to its skeletal muscle. Cells produced from this portion of the occipital somites migrate anteriorly to give rise to the intrinsic muscles of the tongue (p. 614). The remaining epithelial plate and underlying myotome cells grow into the flank region of the body. The epithelial plate is still proliferating at the beginning of this stage. Later the epithelial plate cells revert to mesenchyme, and processes from contiguous somites fuse to form a unified premuscular mass which gives rise to the ventrolateral muscles of the body wall (p. 797).
Figure 47.2 Scanning electron micrograph of a lateral view of an embryo showing the somites. The cranial somites are at the lower border and the more caudal somites are at the upper border. A change in size of the cranially more advanced somites is apparent. (Photograph by P Collins; printed by S Cox, Electron Microscopy Unit, Southampton General Hospital.)
It was thought that the somite gave rise to segmental portions of the dermis of the skin as well as bone and muscle. However, it is now clear that the epithelial plate of the somite continues to provide a significant source of myogenic precursor cells as it elongates into the body wall. The somitic contribution to the skin from the epithelial plate is limited to the dermis over the epaxial muscles alone, which is a much smaller distribution than the segmental portion of skin usually implied by the term dermatome. The concept that an embryological dermatome, derived from the somite, produces all of the dermis of the skin is therefore outdated. The regularity of somite formation provides criteria for staging embryos. The staging scheme proposed by Ordahl (1993) will be used in the following account of relative somite development. Ordahl noted that morphogenetic events occur in successive somites at approximately the same rate. The somite most recently formed from the unsegmented mesenchyme is designated as stage I, the next most recent as stage II, etc. After the embryo forms an additional somite, the ages of the previously formed somites increase by one Roman numeral.
According to this scheme, compaction occurs at stage I; epithelialization at stages II to III; formation of mesenchymal sclerotome cells from stage V; myotome formation at stage VI; early migration of the ventrolateral lip of the epithelial plate, and production of myotome cells are still occurring at stage X.
© 2008 Elsevier
DEVELOPMENT OF SCLEROTOMES page 790 page 791
Figure 47.3 The stages of somitogenesis. Development proceeds in a craniocaudal progression. The more cranially placed somites (at the lower right of the figure) are further developed than those caudally placed (at the upper left of the figure). The stages in somitogenesis are given on the left of the figure; more detailed information is given on the right.
page 791 page 792
Figure 47.4 Formation of vertebrae and intervertebral discs from the mesenchymal sclerotomes. Each vertebra is formed from the cranial half of one bilateral pair of sclerotomes and the caudal half of the next pair of sclerotomes. The spinal nerves preferentially migrate through the cranial portion of the sclerotomes. (Redrawn from Tuchmann-Duplessis H, Haegel P 1972 Illustrated Human Embryology,-Vol 2 Organogenesis. London: Chapman and Hall.)
Sclerotomal populations form from the ventromedial border of the epithelial somite. An intrasegmental boundary (fissure or cleft, sometimes termed von Ebner's fissure) appears within the sclerotome, dividing it into loosely packed cranial and densely packed caudal halves; this boundary is initially filled with extracellular matrix and a few cells.-The epithelial plate, and later the dermomyotome, spans the two half-sclerotomes. The bilateral sclerotomal cell populations migrate towards the notochord and surround it to form the perinotochordal sheath. They undergo a matrix-mediated interaction with the notochord, differentiating chondrogenetically to form the cartilaginous precursor of the vertebral centrum. The perinotochordal sheath transiently expresses type II collagen, and this is believed to initiate type II collagen expression, and thereafter a chondrogenic fate, in those mesenchyme cells which contact it. Each vertebra is formed by the combination of much of the caudal half of one bilateral pair of sclerotomes with much of the cranial half of the next caudal pair of sclerotomes. Their fusion around the notochord produces the blastemal centrum of the vertebra (Figs 47.4, 47.5). The mesenchyme adjoining the intrasegmental sclerotomic fissure now increases greatly in density to form a well-defined perichordal disc which intervenes between the centra of two adjacent vertebrae and is the future anulus fibrosus of the intervertebral symphysis ('disc') (see below). The basic pattern of a typical vertebra is initiated by this recombination of caudal and cranial sclerotome halves (Fig. 47.6), followed by differential growth and sculpturing of the sclerotomal mesenchyme which encases the notochord and neural tube. The centrum encloses the notochord and lies ventral to the neural tube. Condensation of the sclerotomal mesenchyme around the notochord and right and left neural processes can be seen in stage 15 human embryos. The neural processes extend from the dorsolateral angles of the centrum and curve to enclose the neural tube. The neural arch consists of paired bilateral pedicles (ventrolaterally) and laminae (dorsolaterally) which coalesce in the midline dorsal to the neural tube to form the spinous process. On each side three further processes project cranially, caudally and laterally from the junction of the pedicle and laminae. The cranial and caudal projections are the blastemal articular
processes (zygapophyses) which become contiguous with reciprocal processes of adjacent vertebrae; their junctional zones mark the future zygapophyseal or facet joints. The lateral projections are the true vertebral transverse processes. Bilateral costal processes (ribs) grow anterolaterally from the ventral part of the pedicles (i.e. near the centrum), from the neighbouring perichordal disc, and, at most thoracic levels, with accessions from the next adjacent caudal pedicles. The costal processes expand to meet the tips of the transverse processes. The definitive vertebral body is compound, and is formed from a median centrum (derived from the cells of the perinotochordal sheath), and bilaterally from the expanded pedicle ends (derived from the migrating sclerotomal populations). These portions of the vertebral body fuse at the neurocentral synchondroses. The segmental nature of the vertebrae is promoted by the notochord, which induces the ventral elements of the vertebrae and represses dorsal structures, e.g. the spinous processes. Excision of the notochord in early embryos results in fusion of the centra and formation of a cartilaginous plate ventral to the neural tube. Dorsal segmentation is influenced by the spinal ganglia: experimental removal of the ganglia results in fusion of the neural arches and the formation of a uniform cartilaginous plate dorsal to the neural tube.
Intervertebral discs Vertebral centra are derived from caudal and cranial sclerotomal halves. An intervertebral disc is formed from the free somitocoele cells within the epithelial somite which migrate with the caudal sclerotomal cells. The sclerotomal mesenchyme which forms the centra of the vertebrae replaces the notochordal tissue which it surrounds. In contrast, the notochord expands between the developing vertebrae as localized aggregates of cells and matrix which form the nucleus pulposus of the intervertebral disc (Figs 47.4, 47.7). The intermediate part of each perichordal disc, which forms the anulus fibrosus, surrounds the nucleus pulposus and differentiates into an external laminated fibrous zone and an internal cuff (which lies next to the nucleus pulposus). The inner zone contributes to the growth of the outer. Near the end of the second month of embryonic life it begins to merge with the notochordal tissue, and is ultimately converted into fibrocartilage. After the sixth month of fetal life, notochordal cells in the nucleus pulposus degenerate, and are replaced by cells from the internal zone of the anulus fibrosus. This degeneration continues until the second decade of life, by which time all the notochordal cells have disappeared. In the adult, notochordal vestiges are limited, at the most, to non-cellular matrix. The original sclerotomes are coextensive with the individual metameric body segments: each sclerotomic fissure, perichordal disc, and maturing intervertebral disc lie opposite the centre of each fundamental body segment. It therefore follows that the discs correspond in level to (i.e. form the anterior boundary of) the intervertebral foramina and their associated mixed spinal nerves, ganglia, vessels and sheaths. Posteriorly, the foramina are bounded by the capsules of the synovial facet joints; the rims of the vertebral notches of adjacent vertebrae lie cranially and caudally. Thus all the structures listed (and other associated ones) are often designated segmental, whereas vertebral bodies are designated intersegmental because of their mode of development.
Figure 47.5 Contribution of the somites to the vertebrae. Each somite induces a ventral root to grow out from the spinal cord. When the sclerotomes recombine the cranial half of the first cervical sclerotome fuses with the occipital sclerotome above contributing to the occipital bone of the skull. The cervical nerves beginning with C1 exit above the corresponding vertebra. Nerve C8 exits below the last cervical vertebra (C7). After this level nerves arise below their vertebrae. (By permission from Larsen WJ 1997 Human Embryology, 2nd edn. Edinburgh: Churchill Livingstone.)
(For further discussion of resegmentation theories, the reader is directed to Muller & O'Rahilly 1986; Brand-Saberi & Christ 2000; Huang et al 2000; Stern & Vasiliauskas 2000.)
Development of vertebrae page 792 page 793
Figure 47.6 Contribution of two adjacent somites to one vertebra and rib. The intervertebral disc and the costal head are derived from somitocoele cells from one somite which migrate with the caudal half of the sclerotome. The proximal rib is formed from caudal and cranial somite halves with no mixing of cells; in the distal rib there is more mixing of cranial and caudal cells as segmentation diminishes in the ventral body wall. (Redrawn with permission from Christ B, Huang R, Wilting J. The development of the avian vertebral column. Anat Embryol 202: 179-194, 2000, Springer.)
The initial movements of sclerotomal cells round the neural tube and the expression of type II collagen signals the blastemal stage of vertebral development (Fig. 47.7). Chondrification begins at stage 17, initiating the cartilaginous stage. Each centrum chondrifies from one cartilage anlage. Each half of a neural arch is chondrified from a centre, starting in its base and extending dorsally into the laminae and ventrally into the pedicles, to meet, expand and blend with the centrum. By stage 23 there are 33 or 34 cartilaginous vertebrae (Fig. 11.6), but the spinous processes have not yet developed, so the overall appearance is of total spina bifida occulta. Fusion of the spines does not occur until the fourth month. The transverse and articular processes are chondrified in continuity with the neural arches. Intervening zones of mesenchyme which do not become cartilage mark the sites of the facet joints and the complex of costovertebral joints, within which synovial cavities later appear. A typical vertebra is ossified from three primary centres, one in each half vertebral arch and one in the centrum (Fig. 45.20). Centres in arches appear at the roots of the transverse processes, and ossification spreads backwards into laminae and spines, forwards into pedicles and posterolateral parts of the body, laterally into transverse processes and upwards and downwards into articular processes. Classically centres in vertebral arches are said to appear first in upper cervical vertebrae in the ninth to tenth week, and then in successively lower vertebrae, reaching lower lumbar levels in the twelfth week. However, in a radiographic study of unsexed human fetuses (Bagnall et al 1977), a pattern was noted which differed from such a simple craniocaudal sequence. A regular cervical progression was not observed. Centres first appeared in the lower cervical/upper thoracic region, quickly followed by others in the upper cervical region. After a short interval a third group appeared in the lower thoracolumbar region and remaining centres then appeared, spreading regularly and rapidly in craniocaudal directions. The major part of the body, the centrum, ossifies from a primary centre dorsal to the notochord. Centra are occasionally ossified from bilateral centres which may fail to unite. Suppression of one of these produces a cuneiform vertebra (hemivertebra), which is one cause of lateral spinal curvature (scoliosis). At birth (Fig. 11.6) and during early postnatal years the centrum is connected to each half neural arch by a synchondrosis or neurocentral joint. In thoracic vertebrae costal facets on the bodies are posterior to neurocentral joints. During the first year the arches unite behind, first in the lumbar region and then throughout the thoracic and cervical regions. In upper typical cervical vertebrae, centra unite with arches about the third year, but in lower lumbar vertebrae, union is not complete until the sixth. The upper and lower surfaces of bodies and apices of transverse and spinous processes are cartilaginous until puberty, at which time five secondary centres appear, one in the apex of each transverse and spinous process and two annular epiphyses ('ring apophyses') for the circumferential parts of the upper and lower surfaces of the body. Costal articular facets are extensions of these anular epiphyses. They fuse with the rest of the bone at c.25 years. There are two secondary centres in bifid cervical spinous processes. Exceptions to this pattern of ossification are described in the appropriate subsections in Chapter 45.
Vertebrae are specified as to region very early in development. If a group of thoracic somites is transplanted to the cervical region, ribs will still develop. It is the sclerotome which is restricted: the myotome will produce muscle characteristic of the new location.
OCCIPITOCERVICAL JUNCTION In humans, the junction between the head and neck (termed occipitocervical, craniovertebral or spinomedullary) corresponds to the boundary between the 4th and 5th somites (Muller & O'Rahilly 1994). It can first be determined at stage 12 by the observation of hypoglossal nerve rootlets (Fig. 47.8). At stages 14 and 15 the junction is seen between the hypoglossal rootlets and the 1st spinal ganglion. The exact position of this boundary is controversial. Wilting et al (1995) suggest that in human embryos the boundary may be between the 5th and 6th somites and that previous studies were based on older embryos in which the transitory 1st somite had already disappeared. In avian embryos, where all embryonic stages may be obtained experimentally, the occipitocervical boundary has been determined as the 5th-6th somite boundary. The first occipital somite disappears early and the caudal three fuse to form the basiocciput. Occipital sclerotomes 3 and 4 are the most distinct at stage 14, by which time the first three sclerotomes have fused. Vertebrae are formed from the 5th somite caudally: the first cervical vertebra is formed by the caudal half of occipital somite 4 and the cranial half of cervical somite 1 (Fig. 47.5). This shift of somite number and vertebral number accounts for the production of seven cervical vertebrae from eight cervical somites. The segmental pattern present in the cervical somites can be seen rostrally in the developing skull base, where mesenchymal condensations equivalent to the centra of occipital somites 2, 3 and 4 are apparent. The hypoglossal rootlets pass through the less dense portion of occipital sclerotome 4, accompanied by the hypoglossal artery. Occipital sclerotome 4 forms an incomplete centrum axially and exoccipital elements laterally, which are regarded as corresponding to neural arches. They form the rim of the foramen magnum. The occipital condyles develop from the 1st cervical somite (as also happens in the chick). In the occipitocervical junctional region the centra formed from sclerotomes 5, 6 and 7 have a different fate from those more caudally placed. The lateral portions of these sclerotomes generally develop similarly to those of lower ones. In a study of occipitocervical segmentation in human embryos, Muller and O'Rahilly designated the three complete rostral centra which develop in the atlanto-axial region X, Y and Z (Figs 47.9, 47.10; Muller & O'Rahilly 1986). They noted that the height of the XYZ complex is equal to that of three centra elsewhere. X is on the level of sclerotome 5, and Y and Z are in line with sclerotome 6 and with the less dense portion of sclerotome 7. During stage 17 a temporary intervertebral disc appears peripherally between Y and Z. It begins to disappear in stage 21, although remains may be found in the adult. No disc develops between X and Y. The origin of the anterior arch of the atlas is unclear. It is evident at stages 21-23 at the level of X or sometimes between X and Y. The posterior arch of the atlas arises from the dense area of sclerotome 5 at the level of X. The XYZ complex belongs to the axis, which means that the atlas does not incorporate a part of the central column (Muller & O'Rahilly 1994). page 793 page 794
Figure 47.7 Vertebral development through blastemal, cartilaginous and pre-and postnatal ossificatory stages. Bottom row indicates principal morphological parts of adult vertebrae. A, Vertebral development through blastemal, cartilaginous and preand postnatal ossificatory stages. B, Derivation of principal morphological parts of adult vertebrae.
The posterior arch of the axis arises from the dense area of sclerotome 6 and is at the level of Y and Z, particularly Z. XYZ correspond to the three parts of the median column of the axis, where X represents the tip of the dens, Y represents the base of the dens and Z represents the centrum of the axis. The latter differs from other cervical vertebrae in that it is thicker and square-shaped. The development of the cervical spine, particularly the upper cervical vertebrae, is closely related to the development of the basiocciput and exocciput: anomalous development will affect both regions. Anomalies of the occiput are associated with decreased skull base height; the dens lies at or above the level of the foramen magnum; a distinctive margin of the foramen magnum lies above the bottom of the posterior fossa; and the posterior arch of the atlas is at the same level as the foramen magnum. The last three structural defects are collectively termed basilar invagination. Malfusions of the caudal portion of occipital sclerotome 4 and the cranial portion of cervical sclerotome 1 may produce defects of the occipital
condyles. Disassociated occipital condyles are called-a proatlas, because in lower vertebrates the cranial half of the first cervical sclerotome forms a separate bone between the occipital bone and the atlas. page 794 page 795
Figure 47.8 Reconstructions of the occipitocervical region of human embryos. A, Stage 12: occipital somites innervated by hypoglossal fibres (small circles). Three cervical somites are shown. The crest derived ganglia of cranial nerves 5, 7, 8, 9 and 10 are shown in green. Neural crest associated with the occipital somites is hypoglossal and perhaps accessory (also shown in green). B, Stage 14: the somites have transformed into sclerotomes and moved ventrally. The less dense cranial and dense caudal parts of the sclerotomes and occipital sclerotomes 1-4 are indicated. Hypoglossal fibres and cervical ventral rami migrate through the less dense parts of the sclerotomes. The occipital neural crest is now seen to be mostly accessory. The cervical crest is subdivided into spinal ganglia. A perinotochordal sheath can be seen extending rostrally to the termination of the notochord. C, Stage 15: the dense parts of sclerotomes 1-8 are shown. Intersegmental arteries are visible in the less dense areas of the sclerotomes, as are the spinal nerve fibres. In all diagrams the occipitocervical junction is indicated by the asterisks and line. (After Muller F, O'Rahilly R 1994 Occipitocervical segmentation in staged human embryos. J Anat 185: 251-258. Permission by Blackwell Publishing.)
Decreased skull base height and diminished volume in the posterior fossa are structural anomalies associated with the Arnold-Chiari malformation. Although this is considered clinically to be a neurological defect, in that the medulla oblongata, and sometimes the cerebellar tonsils, project through the foramen magnum, there is good evidence that the underlying cause is a series of abnormalities of the occipitocervical junction. When the volume of the brain in the posterior cranial fossa and the volume of the fossa (delineated by bone and the tentorium cerebelli) were compared in controls and Arnold-Chiari cases, no significant difference was found between brain volume in the two groups, though the basiocciput, exocciput and supraocciput were significantly smaller in affected individuals (Nishikawa et al 1997). In patients with the Arnold-Chiari malformation, proatlas remnants and atlas assimilation are found. The sagittal canal diameter of the foramen magnum is critical: patients become symptomatic when this is less
than 19 mm (Menezes 1995). Secondary skull base and cervical spine deformations, e.g. foramen magnum and cervical canal enlargement, may develop in association with the Arnold-Chiari malformation. Most defects of the atlas do not contribute to abnormal occipitocervical anomalies and are not associated with basilar invagination.
Figure 47.9 The relationship between the centra and neural arches of the vertebrae and the related spinal ganglia and nerves. Scheme of the details of the early development of the occipitocervical region. A, The column of sclerotomes from occipital somite 1. B, Dorsal view of the developing vertebrae with the centra in the middle and the bilateral components of the neural processes laterally. X, Y and Z are three centra which will produce the tip of the dens of the axis (X), the base of the dens of the axis (Y) and the centrum of the axis (Z). An intervertebral disc appears temporarily between Y and Z during stage 17. No disc develops between X and Y. The occipital condyles are derived from the first cervical sclerotome. (After Muller F, O'Rahilly R 1994 Occipitocervical segmentation in staged human embryos. J Anat 185: 251-258. Permission by Blackwell Publishing.)
Abnormalities of the axis are usually concerned with fusion of the dens with the centrum of the second cervical sclerotomes. Using the classification of the three complete centra which develop in the atlantoaxial region as X, Y and Z (Muller & O'Rahilly 1986), failure of fusion of X with the YZ complex produces an ossiculum terminale, a dissociated apical odontoid epiphysis. Failure of fusion of the XY complex with Z at the dentocentral synchondrosis, or maintenance of the transitory intervertebral disc at this point, produces an os odontoideum, thought to be induced by excessive movement at the time of ossification of the dens (Crockard & Stevens 1995). Hypoplasia and aplasia of the X and Y centra, and aplasia of the Z centrum, will all lead to reduced size of the dens. There are widely differing views about whether this will lead to atlanto-axial instability.
C3-7, THIRD TO SEVENTH CERVICAL VERTEBRAE In cervical vertebrae 3-7 (Fig. 47.7) the transverse process is dorsomedial to the foramen transversarium. The costal process, corresponding to the head, neck and tubercle of a rib, limits the foramen ventrolaterally and dorsolaterally. The distal parts of these cervical costal processes do not normally develop; they do so occasionally in the case of the seventh cervical vertebra, and may even develop costovertebral joints. These cervical ribs may reach the sternum. page 795 page 796
Figure 47.10 The relationship between the centra and neural arches of the vertebrae with the related spinal ganglia and nerves. Cells of the somites from 1 to c.27 are derived from the primitive streak. Somite 31 illustrates the level of final closure of the caudal neuropore. All neural and somitic tissue caudal to that is derived from secondary neurulation. An intervertebral disc appears temporarily between Y and Z during stage 17. No disc develops between X and Y. The occipital condyles are derived from the first cervical sclerotome. (After Muller F, O'Rahilly R 1994 Occipitocervical segmentation in staged human embryos. J Anat 185: 251-258. Permission by Blackwell Publishing.)
The seventh cervical vertebra is transitional in shape between cervical and thoracic vertebrae. The laminae are longer than other cervical vertebrae in the neonate and lie almost perpendicular to the basal plane; the inferior articular facets are more upright, and resemble those of thoracic vertebrae; and, in the lateral view, the superior articular facets extend transversely to the top of the transverse processes. Anomalies of the lower cervical vertebrae are generally caused-by inappropriate cervical vertebral fusion: collectively this is termed Klippel-Feil syndrome. This term includes all congenital fusions of the cervical spine, from two segments to the entire cervical spine. Affected individuals have a low posterior hairline, short neck and limitations of head and neck movement. Scoliosis and/or kyphosis is common.
THORACIC VERTEBRAE At stage 23 the neural processes of thoracic vertebrae are short, slightly bifurcated and joined by collagenous fibres. The transverse process is prominent. The three facets for articulation with the ribs at the costovertebral and costotransverse joints are present. The thoracic neuro central and posterior synchondroses are not fused in the neonate:-the posterior synchondroses close within 2 to 3 months of postnatal development and the neurocentral synchondroses are open until 5 to -6 years of age.
In general, the thoracic spine develops ahead of the cervical and lumbar spine. However, towards the end of the second month, ossification begins in the cartilaginous vertebrae in a craniocaudal progression.
RIBS Ribs usually develop in association with the thoracic vertebrae. Occasionally they can arise from the seventh cervical and first lumbar vertebrae. The costal processes attain their maximum length as the ribs in the thoracic region. Each rib originates from the caudal half of one sclerotome and the cranial half of the next subjacent sclerotome (Fig. 47.6). The head of the rib develops from somitocoele cells from one somite, which migrate with the caudal half of the sclerotome. The proximal portion of the rib forms from both caudal and cranial sclerotomal halves; there is no mixing of cells from these origins. The distal portion of the rib forms from caudal and cranial sclerotomal halves; these cells mix as the rib extends into the ventral body wall and segmentation diminishes. The ribs arise anterolaterally from the ventral part of the pedicles, and form bilateral costal processes which expand to meet the tips of the transverse processes. As they elongate laterally and ventrally they come to lie between the myotomic muscle plates. In the thorax (Fig. 47.7) the costal processes grow laterally to form a series of precartilaginous ribs. The transverse processes grow laterally behind the vertebral ends of the costal processes, at first connected by mesenchyme which later becomes differentiated into the ligaments and other tissues of the costotransverse joints. The capitular costovertebral joints are similarly formed from mesenchyme between the proximal end of the costal processes and -the perichordal disc, and the adjacent parts of two (sometimes one) vertebral bodies, which are derived from the neural arch. Ribs 1-7 (vertebrosternal) curve round the body wall to reach the developing sternal plates. Ribs 8-10 (vertebrochondral) are progressively more oblique and shorter, only reaching the costal cartilage of the rib above, and contributing to the costal margin. Ribs 11 and 12 are free (floating), and have cone-shaped terminal cartilages to which muscles become attached (p. 955).
LUMBAR VERTEBRAE The costal processes do not develop distally in lumbar vertebrae (Fig. 47.7). Their proximal parts become the 'transverse processes', while the morphologically true transverse processes may be represented by the accessory processes of the vertebrae. Occasionally, movable ribs may develop in association with the first lumbar vertebra. Lumbar intervertebral discs are thicker than thoracic discs. By stage 23 the annulus fibrosus can be seen in the peripheral part, and internally the notochordal cells are expanding to form the nucleus pulposus.
SACRUM Sacral vertebrae have lower centra and are narrower overall from side to side than their thoracic and lumbar counterparts. Each sacral vertebra is composed of a centrum and bilateral neural processes. The contribution of the costal processes to sacral development was examined by O'Rahilly et al (1990). These authors divided the neurocentral junctional area into two parts, anterolateral or alar, and posterolateral. They found-the alar element in sacral vertebra 1 to be novel, since it was absent-in lumbar vertebra 5. There is support for this view if the course of the dorsal rami of the spinal nerves is used to distinguish the costal elements ventrally from the transverse elements dorsally. The alar elements of the sacral vertebrae are ventral to the sacral dorsal rami, and both costal and transverse portions are posterolateral. The alar element of S1 and S2 forms the auricular surface of the sacrum. At stage 23 the cartilaginous sacral vertebrae have joined and the outline of the future bone can be recognized. Individual pedicles and laminae are very small and can be detected in S3-5. Ossification of the vertebral column proceeds in a craniocaudal direction. After 16 weeks it has progressed to L5. Ossification of each additional vertebra occurs over a period of 2-3 weeks; S2 is ossified by 22 weeks. page 796 page 797
Very rarely significant malformation of the sacral or lumbosacral vertebrae may develop, often in association with a maternal history of diabetes. When there is sacral agenesis, motor paralysis is profound below the affected vertebral level, whereas the sensory disturbance does not relate to the vertebral level so clearly
and sensation may be present down to the knees. Bladder involvement is a consistent feature.
Spina bifida Spina bifida is the generic term for a range of discrete defects of neurulation and subsequent vertebral formation. The spectrum of neural tube and vertebral defects includes a range of open neural tube defects: craniorachischisis (nonfusion of the entire neural tube and no vertebral arch development); anencephaly (non-fusion of the rostal portion-of the neural tube with no calvarial or occipital development); and myelocoele (non-fusion of caudal portions of the neural tube and local failure of vertebral arch development)(see Fig. 14.8). Spina bifida cystica occurs where the meninges have developed adjacent to or over the defective neural tissue. Local accumulation of CSF may push a defective neural plaque or spinal cord superficial to the level of the vertebrae forming a meningomyelocoele. Alternatively, a meningeal sac may protrude in the midline if one or two spinous processes are absent; this condition is a meningocoele. In both cases the meningeal sac may be covered with skin or skin may be contiguous with the edges of a neural plaque. Prior to antenatal diagnosis of spina bifida by ultrasonography, most live births with spina bifida cystica had meningomyelocoele. Meningomyelocoele occurs in thoracolumbar, lumbar or lumbosacral regions. Sacral lesions are less common. The vertebral lesion usually extends cranially further than the neural lesion, showing deformities of the vertebral bodies and laminae. Vertebrae may be wedge shaped or hemivertebrae and ribs may be fused or absent. Concomitant kyphosis at birth is associated with a worse prognosis. Prenatal diagnosis of meningomyelocoele and termination of affected fetuses has led to a significant decrease in the incidence of live births with this condition. Up to 10% of spina bifida cystica cases are meningocoele. This lesion may occur in cervical and upper thoracic levels or in lower lumbar and sacral levels: the latter are most common. In its mildest form no neurological abnormality occurs with this condition. Spina bifida occulta affects about 5% of the population. Affected individuals have bifid spinous processes often at the lumbosacral junction. A very small proportion of those with this condition have a naevus, dimple or tuft of hair over the lesion.
TETHERING OF THE CORD AND DIASTEMATOMYELIA The greatest change of vertebral length occurs within the last 6 months of intrauterine life. When there is any degree of spina bifida, the upward migration of the spinal cord which normally occurs within the vertebral canal during growth is limited, and the cord is said to be tethered. Neurological deficits caused by this condition in neonates become apparent soon after birth, although some individuals show symptoms later in life, especially degrees of urinary dysfunction. Very rarely, tethering symptoms may be associated with abnormal development of the vertebral centra, e.g. when a midline cartilaginous or osseous spicule or a fibrous septum projects into the vertebral canal. These obstacles may split the spinal cord or intraspinal nerve roots into two columns, a condition termed diastematomyelia. Usually patients with this condition have a cutaneous abnormality, such as a dimple, pigmented naevi or patch of hair, along their back at the level of the tethering.
© 2008 Elsevier
DEVELOPMENT OF DERMOMYOTOMES The dorsal half of the somite, the dermomyotome, gives rise to all the skeletal muscle in the trunk. The medial dermomyotome produces-the epaxial muscles and the lateral portion produces the muscles of the tongue, limbs and of the ventrolateral body wall. Muscles in the head arise from the unsegmented paraxial mesenchyme rostral to the occipital somites (p. 453). Myogenic determination factors, MyoD, myogenin, Myf 5 and herculin/MRF 4 can first be detected in the medial half of the somite as early as stage II, several hours prior to the onset of myotome formation. Cells of the epithelial plate are mainly orientated perpendicular to the back, but they have different orientations according to their positions: they are transversely orientated along the dorsomedial edge and longitudinally orientated within the cranial edge of the epithelial plate (Figs 47.2, 47.3). Myotome cells originate from longitudinally orientated cells at the cranial edge of the epithelial plate. Individual cells, which are originally produced and anchored at the craniomedial corner of the epithelial plate, send processes to the caudal edge of the plate where they form a second anchor point. In this way, myotome formation continues caudally along the dorsomedial edge and laterally along the cranial edge, and each mononucleated myotome cell becomes very elongated perpendicular to the cells of the epithelial plate. Development of subsequent cells imparts a triangular shape to the early myotome. The growing myotome first reaches the caudal somite border on the medial side and later the ventrolateral edge. When the vertebral bodies form, the future intervertebral fissure divides the sclerotome into rostral and caudal halves. The myotome fibres span the intervertebral joints and foramina, which means that the muscles which are derived from the myotome are positioned to move adjacent vertebrae relative to each other.
EPAXIAL MUSCLE: DORSAL TRUNK MUSCLES Myotome cells are all postmitotic embryonic myoblasts: later in development they fuse to form syncytia which form the epaxial musculature. The presence of the neural tube is required for normal myoblast development. There is evidence to suggest that interactions between precursor myotome cells and the medial neural crest cells, which are commencing their migration at this time, may also be important. The epaxial muscles are innervated by the dorsal ramus of each spinal nerve (p. 782). At much later stages satellite cells enter the myotome. Interestingly the development of endo-, peri-and epimysium in relation to the epaxial muscles has not been addressed.
HYPAXIAL MUSCLE: VENTROLATERAL TRUNK MUSCLES The ventrolateral trunk muscles are formed from the epithelial plate-of the somite. After production of the myotome and the precursor myogenic cells of the limb, the remaining epithelial plate and attached myotome grows into the flank
somatopleuric mesenchyme. At this stage the epithelial plate is still proliferating and producing myogenic precursor cells. The epithelial plate has a leading edge or process from which single cells or clusters of cells migrate in a ventral direction. It may be that these epithelial plate cells, which are in a more immature state of differentiation, act as pioneer cells for further cell movement. The previously segmented processes from adjacent epithelial plates form a unified premuscular mass. Both postmitotic myoblast cells-and mitotic plate cells can be seen in the body wall; they may represent early and later forming myoblasts which will form heterokaryotic myotubes. The premuscular mass subdivides into abdominal muscle blastemata for the external and internal oblique muscles, transversus abdominis and rectus abdominis. At this time the number of somatopleural fibroblasts situated within the muscle-forming zone increases, and myotubes can be first seen. There is a subsequent ventral shift of the already separated muscle blastemata within the growing abdominal wall as they attain their definitive positions. Muscle differentiation continues and muscular connective tissue, tendons and aponeuroses develop.
OTHER STRUCTURES DERIVED FROM THE SOMITES Dermal precursors undergo an epithelial/mesenchyme transformation from the dermomyotome and migrate dorsally toward the ectoderm over the dorsal region of the embryo. Their transformation seems to be controlled by factors from the neural tube. There is a sharp boundary between dermis which is somite-derived and that which is derived from the splachnopleuric mesenchyme of the lateral plate (which covers the limbs, part of the lateral and all of the ventral body wall). Smooth muscle cells within and around the developing somites and endothelial cell precursors are now considered to arise from somitic cells. All compartments of the epithelial somite, including the somitocoele cells, give rise to angioblasts (Brand-Saberi & Christ 2000). page 797 page 798
REFERENCES Bagnall KM, Harris PF, Jones PRM 1977 A radiographic study of the human spine II. The sequence of development of ossification centres in the vertebral column. J Anat 124: 791-802. Medline Similar articles Brand-Saberi B, Christ B 2000 Evolution and development of distinct cell lineages derived from somites. Current Topics in Developmental Biology Vol 47. New York: Academic Press. Christ B, Huang R, Wilting J 2000 The development of the avian vertebral column. Anat Embryol 202: 17994. Medline Similar articles Full article Crockard HA, Stevens JM 1995 Craniovertebral junction anomalies in inherited disorders: part of the syndrome or caused by the disorder? Eur J Pediatr 154: 504-12. Medline Similar articles Full article Gumpel-Pinot M 1984 Muscle and skeleton of limbs and body wall. In: Le Douarin N, McLaren A (eds) Chimeras in Developmental Biology. London: Academic Press: 281-310. Huang R, Zhi Q, Brand-Saberi B, Christ B 2000 New experimental evidence for somite resegmentation. Anat Embryol 202: 195-200. Medline Similar articles Full article Menezes AH 1995 Primary craniovertebral anomalies and the hindbrain herniation syndrome (Chiari 1); database analysis. Pediatr Neurosurg 23: 260-9. Muller F, O'Rahilly R 1986 Somitic-vertebral correlation and vertebral levels in the human embryo. Am J
Anat 177: 3-19. Medline
Similar articles
Full article
Muller F, O'Rahilly R 1994 Occipitocervical segmentation in staged human embryos. J Anat 185: 251-8. Medline Similar articles Nishikawa M, Sakamoto H, Hakuba A, Nakanishi N, Inoue Y 1997 Pathogenesis of Chiari malformations: a morphometric study of the posterior cranial fossa. J Neurosurg 86: 40-7. Medline Similar articles O'Rahilly R, Muller F, Meyer DB 1990 The human vertebral column at the end of the embryonic period proper. 4. The sacrococcygeal region. J Anat 168: 95-111. Medline Similar articles Ordahl CP 1993 Myogenic lineages within the developing somite. In: Bernfield M (ed.) Molecular Basis of Morphogenesis. New York: Wiley Liss: 165-76. Pourquie O, Kusumi K 2001 When body segmentation goes wrong. Clin Genet 60: 409-16. Medline Similar articles Full article Richardson MK, Allen SP, Wright GM, Raynaud A, Hanken J 1998 Somite number and vertebrate evolution. Development 125: 151-60. Medline Similar articles Stern CD, Vasiliauskas D 2000 Segmentation: a view from the border. Current Topics in Developmental Biology, Vol 47. New York: Academic Press. Presents three different models of somite formation in the context of recent molecular data. Wilting J, Ebensperger C, Müller TS, Kosecki H, Wallin J, Christ B 1995 Pax-1 in the development of the cervico-occipital transitional zone. Anat Embryol 192: 221-227. Medline Similar articles
© 2008 Elsevier
SECTION 5 PECTORAL GIRDLE AND UPPER LIMB David Johnson Harold Ellis (Lead Editors) Patricia Collins (Embryology, Growth and Development) David Johnson Harold Ellis (Editors) With specialist contributions on clinical and functional anatomy by Vivien Lees (chapter 53) Critical reviewers: Paul Cartwright (chapter 48), Steve Corbett (49 & 51), David Woods (49 & 51) page page page page
798 799 799 800
© 2008 Elsevier
48 General organization and surface anatomy of the upper limb This chapter is divided into two sections. The first is an overview of the general organization of the upper limb, with particular emphasis on the distribution of the major blood vessels and lymphatic channels, and of the branches of the brachial plexus: it is intended to complement the detailed regional anatomy described in Chapters 49 to 53. The second section describes the surface anatomy of the upper limb.
© 2008 Elsevier
SKIN, FASCIA AND SOFT TISSUES The skin of the anterior aspect of the upper arm and forearm differs from that of the posterior aspect in that it is thinner and hairless. The palmar skin is thick and hairless: firm attachments to the underlying palmar fascia reflect its role in gripping and shock absorption. The dorsal skin of the hand is much thinner, lax and mobile, and this allows the extensor tendons to glide underneath the subcutaneous tissue. The direction in which skin tension is greatest varies regionally in the upper limb as in other areas of the body. Skin tension lines that follow the furrows formed when the skin is relaxed are known as 'relaxed skin tension lines' (p. 173) and can act as a guide in planning elective incisions. The superficial fascia is a layer of subcutaneous fatty tissue. Its thickness depends on the degree of obesity of the subject: measurement of the thickness of the subcutaneous tissue of the posterior upper arm is used as an indicator of obesity. There is less subcutaneous tissue in the palm of the hand than on the dorsum of the hand. The depth of deep fascia varies according to the stresses to which it is subjected in the different areas of the limb. It is a thin but quite obvious layer in the upper arm, where intermuscular septa pass to the medial and lateral sides of the humerus and separate the upper arm muscles into anterior and posterior groups within their respective compartments. In addition, each muscle also lies within its own delicate fascial sheath, an arrangement that allows individual muscles to glide upon each other. At the elbow the deep fascia condenses anteriorly as the tough bicipital aponeurosis. In the forearm it is relatively thin and is attached along the subcutaneous border of the ulna. Intermuscular septa divide the forearm into three compartments, anterior (flexors), posterior (extensors) and the mobile wad compartment for brachioradialis and extensor carpi radialis longus and brevis. At the wrist the deep fascia becomes condensed anteriorly and posteriorly as the flexor and extensor retinacula respectively. Further condensation occurs in the palm of the hand, where the palmar aponeurosis is reinforced by the insertion of the tendon of palmaris longus, and in the flexor tendon sheaths and fascial system associated with the digits.
© 2008 Elsevier
BONES AND JOINTS The bones of the upper limb are the clavicle, scapula, humerus, radius and ulna (connected for a large portion of their length by an interosseous membrane) and the bones of the hand, i.e. the carpals, metacarpals and phalanges (Figs 48.1, 48.2). UPDATE Date Added: 17 May 2005 Shanida Helena Nataraja, PhD (Dianthus Medical Limited) Update: Standard reference values for musculoskeletal ultrasonography Musculoskeletal ultrasonography is a diagnostic technique that is widely used in rheumatology, orthopedic surgery, sports medicine, physical therapy, and radiology. This recent study determines standard reference values for musculoskeletal ultrasonography in healthy adults. One hundred and two healthy Caucasian volunteers were subjected to bilateral investigations of their shoulders, elbows, hands, hips, knees, and feet using a linear probe (10-5 MHz; Esaote Technos MP). From these investigations, mean, minimum and maximum values and standard deviations were obtained for the following measurements: biceps tendon; subscapularis tendon; humeral head; infraspinatus tendon; subdeltoid bursa; axillary recess; acromioclavicular joint; and sternoclavicular joint. Standard reference values for these measurements are summarized in Table 2 of the paper, where they should be consulted. Hypoechoic rims were commonly found in joints and tendon sheaths and reflected the presence physiologic synovial fluid or cartilage. Fluid was detected around 27% (56/204) of long bicep tendons, and in 85% (173/204) of subdeltoid bursae. There was no statistical difference in measurements between the dominant and non-dominant side, between individuals 150 N in punch strength tests). The usefulness of surgical procedures to repair the transversalis fascia is questionable, given its low punch test and tensile strength. 1. Wolloscheck T, Gaumann A, Terzic A et al: Inguinal hernia: measurement of the biomechanics of the lower abdominal wall and the inguinal canal. Hernia 8(3):233-241, 2004.
UPDATE Date Added: 20 December 2005 Publication Services, Inc. Update: Abdominal wall anatomy: the key to a successful inguinal hernia repair. Fagan et al review the anatomy of the groin and explore the concept of a hernia as a pathologic hole originating from the myopectineal orifice. The latter is divided into 3 anatomic triangles. Current repair protocols focus on the medial triangle, which is the direct site of herniation within the groin. However, the lateral and femoral triangles are at risk as sites of future herniation if herniorrhaphy procedures are inadequate. The preperitoneal space, which is located in front of the bladder, is increasingly used during laparoscopic and open inguinal herniorrhaphy procedures. Surgical manipulation of this space may affect sexual or urinary function; placement of synthetic mesh over the space may adversely affect future urologic procedures. Fagan et al conclude that an effective inguinal herniorrhaphy not only repairs the pathologic hole but also covers future sites of herniation within the myopectineal orifice. Fagan SP, Awad SS: Abdominal wall anatomy: the key to a successful inguinal hernia repair. Am J Surg 188(suppl):3S-8S, 2004.
UPDATE Abstract: Surface marking of the deep inguinal ring.
Date Added: 19 July 2005
Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=15376291&query_hl=7 Surface marking of the deep inguinal ring. Boundaries
The inguinal canal is a virtual space lying between the various layers formed from the lower tissues of the anterior abdominal wall. It lies obliquely and slants downwards and medially, parallel with and a little above the inguinal ligament. It extends from the deep to the superficial inguinal rings. Its length depends on the age of the individual, but in the adult is between 3 and 5 cm long. It is bounded anteriorly by the skin, superficial fascia and aponeurosis of external oblique. In its lateral one-third, the anterior wall is reinforced by the muscular fibres of the internal oblique just above their origin from the inguinal ligament. Posterior to the canal lie the reflected inguinal ligament, the conjoint tendon and the transversalis fascia, which separate it from extraperitoneal connective tissue and peritoneum. Superiorly lie the arched fibres of internal oblique and transversus abdominis forming the conjoint tendon. Inferior to the canal is the union of the transversalis fascia with the inguinal ligament and, at the medial end, the lacunar ligament. UPDATE Abstract: 3D geometrical models of inguinal region.
Date Added: 27 July 2005
Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=14974505&query_hl=7 3D geometrical models of inguinal region. In the newborn, the deep and superficial rings are nearly superimposed and the canal is extremely short. This creates an approximately oval defect in the abdominal wall. As the child grows, the anterior abdominal wall muscles grow rapidly, causing the positions of the rings to separate and the canal to lengthen. The defect thus becomes progressively more oblique until, in adulthood, the presence of separate anterior and posterior canal walls forms a 'flap valve' effect.
Increases in intra-abdominal pressure transmitted through the posterior wall and the deep ring are supported by the presence of the thickest part of the overlying anterior wall. At the superficial ring and medial end of the anterior wall, where it is weakest, the posterior wall is strengthened by the conjoint tendon and the reflected inguinal ligament. The fibres of internal oblique and transversus abdominis, which form the conjoint tendon, are constantly active in standing; this activity increases during episodes of increased intra-abdominal pressure. Relations (Fig. 67.16)
The inferior epigastric vessels are important posterior relations of the medial end of the canal. They lie on the transversalis fascia as they ascend obliquely behind the conjoint tendon into the posterior portion of the rectus sheath. The inguinal triangle lies in the posterior wall of the canal. It is bounded inferiorly by the medial half of the inguinal ligament, medially by the lower lateral border of rectus abdominis and laterally by the inferior epigastric artery. It overlies the medial inguinal fossa and, in part, the supravesical fossa. Lacunar ligament
Figure 67.16 Deep structures of the inguinal canal. The aponeurosis of external oblique has been removed. The fibres of internal oblique and rectus abdominis have been divided for clarity. The structures passing posteroinferiorly to the inguinal ligament have also been excluded for clarity.
The lacunar ligament is a thick triangular band of tissue lying mainly posterior to the medial end of the inguinal ligament. It measures c.2 cm from base to apex and is a little larger in the male. It is formed from fibres of the medial end of the inguinal ligament and fibres from the fascia lata of the thigh, which join the medial end of the inguinal ligament from below. The inguinal fibres run posteriorly and laterally to the medial end of the pectineal line and are continuous with the pectineal fascia. They form a near horizontal, triangular sheet with a curved medial border. This edge forms the lateral border of the femoral canal. The apex of the triangle is attached to the pubic tubercle. A strong fibrous band, the pectineal ligament of Astley Cooper, extends laterally along the pectineal line from the pectineal attachment. The fibres from the fascia lata join the inferior/posterior border of the inguinal ligament, which, in combination with fibres from the transversalis fascia, fuses with the pectineal fascia as it joins the thickened periosteum of the pectineal line. This portion of the lacunar ligament forms the lower extension of the medial border of the femoral canal and femoral sheath.
© 2008 Elsevier
HERNIAS OF THE ANTERIOR ABDOMINAL WALL INGUINAL HERNIA An inguinal hernia involves the protrusion of a viscus through the tissues of the inguinal region of the abdominal wall. Although the inguinal canal is arranged such that the weaknesses in the anterior abdominal wall caused by the deep and superficial inguinal rings are supported, the region remains a potential cause of herniation. UPDATE Date Added: 09 August 2005 Abstract: Location of inguinal ring in patients with inguinal hernias. Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=14625793&query_hl=5 Location of inguinal ring in patients with inguinal hernias. Indirect inguinal hernia
An indirect hernia is defined as arising lateral to the inferior epigastric vessels. Many indirect hernias are related to the congenital abnormal persistence of the vaginal process. Other indirect hernias are acquired as a result of progressive weakening of the lateral and posterior walls of the canal. The hernia may pass through the deep ring or may expand the deep ring such that it is no longer a clear entity. Small indirect hernias tend to lie below and lateral to the fibres of the conjoint tendon, but larger hernias often distort and thin the tendon superiorly. Small hernias, which do not protrude beyond the inguinal canal, are covered by the same inner layers as the spermatic cord, including the internal spermatic fascia and cremaster. If the hernia extends through the superficial inguinal ring it is, in addition, covered by external spermatic fascia. UPDATE Date Added: 27 July 2005 Abstract: Segmental nerve damage during a McBurney's incision. Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=14625791&query_hl=5 Segmental nerve damage during a McBurney's incision. In hernias related to a persistent fully patent vaginal process, the hernia contents descend in front of the testis into the tunica vaginalis (complete congenital hernia) and the vaginal process and tunica form part of the hernial sac. Where the vaginal process is sealed off from the tunica vaginalis, the hernia contents descend to the top of the testis (incomplete congenital hernia). Although both types are related to a congenital abnormality, actual herniation into the potential sac may not occur until adult life, as a consequence of increased intra-abdominal pressure or sudden muscular strain. Direct inguinal hernia
A direct inguinal hernia is defined as arising medial to the inferior epigastric
vessels. Direct hernias are always caused by an acquired weakness of the inguinal triangle in the medial posterior wall of the canal, and frequently extend through the anterior wall of the canal or superficial ring. The hernia may protrude through the transversalis fascia, between the conjoint tendon and the inferior epigastric vessels, and enter the inguinal canal. It may closely resemble an indirect hernia, in that the coverings are similar. Its clinical presentation can also mimic an indirect inguinal hernia. Other direct hernias arise either between the fibres of the conjoint tendon or by eventration of the tendon such that it forms a thin covering to the hernia. In either case, a hernia enters the lower end of the canal, protruding through the superficial ring medial to the cord, and is covered by external spermatic fascia. Clinical features of inguinal hernias page 1111 page 1112
Indirect hernias often descend from lateral to medial in the same oblique angle as the canal, because of their origin in the lateral end of the inguinal canal. This is particularly true for congenital hernias. Direct hernias arise from the medial end of the canal and tend to protrude more directly anteriorly. With the patient in the supine position and the hernia reduced, pressure applied over the region of the deep inguinal ring may prevent appearance of an indirect hernia on standing or straining. This method of determining the type of hernia is fraught with difficulties. Indirect hernias may have a wide neck, and occlusion of the deep ring may not be possible with simple digital pressure. An acquired indirect hernia may arise medial to the deep ring. The angle of the descent of the hernia is also an unreliable guide, as it depends on the size of the defect in the posterior wall of the canal, the length of the canal and whether the hernia protrudes through the anterior wall. Direct hernias are more likely to have a wide-necked origin, making strangulation less likely. UPDATE Date Added: 09 August 2005 Update: The cutaneous nerves encountered during laparoscopic repair of inguinal hernia: new anatomical findings for the surgeon. Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=10954819&query_hl=10 The cutaneous nerves encountered during laparoscopic repair of inguinal hernia: new anatomical findings for the surgeon. UPDATE Date Added: 12 July 2005 Abstract: Retropubic vascular anatomy in patients unmdergoing endoscopic extraperitoneal inguinal hernioplasty. Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=12802654&query_hl=11 A prospective endoscopic study of retropubic vascular anatomy in 121 patients undergoing endoscopic extraperitoneal inguinal hernioplasty. Femoral hernia
A femoral hernia protrudes through the femoral ring. The femoral ring is normally closed by a femoral septum of modified extraperitoneal tissue, and is therefore a weak spot. In females, the ring is relatively large and subject to profound changes during pregnancy, explaining why femoral hernias are more common in women. When a section of intestine bulges through the ring, it pushes out a hernial sac of peritoneum. It is covered by extraperitoneal tissue (the femoral septum) and descends along the femoral canal to the saphenous opening. It is prevented from descending further along the femoral sheath by the narrow saphenous opening, by the vessels and by the close attachment of the superficial fascia and sheath to the lower part of the rim of the saphenous opening. The hernia hence turns forwards, distending the cribriform fascia and curving upwards over the inguinal ligament and the lower part of the aponeurosis of external oblique. While in the canal the hernia is usually small, because of the resistance of its surrounds, but enlarges as it escape into the inguinal loose connective tissue. Thus a femoral hernia first descends and then ascends forwards. Hence pressure to reduce it should be directed in the reverse order, with the thighs passively flexed for greatest relaxation. The coverings of a femoral hernia are, from within outwards: the peritoneum, femoral septum, femoral sheath, cribriform fascia, superficial fascia and skin. A fibrous covering, the fascia propria, may lie outside the peritoneal sac and is frequently separated from it by adipose tissue. It represents a femoral septum thickened to form a membranous sheet by hernial pressure. The fascia propria may easily be mistaken for the sac, and its contained extraperitoneal fat for omentum; the fat may resemble a lipoma, but dissection will reveal the true hernial sac in its centre. The intestine reaches only to the saphenous opening in incomplete femoral hernia, in contradistinction to complete hernia, in which it passes through the opening. The small size of an incomplete hernia renders it difficult to detect and therefore dangerous, especially in the corpulent. The site of strangulation varies: it may be at the neck of the hernial sac; more often it is at the junction of the falciform margin of the saphenous opening with the free edge of the pectineal part of the inguinal ligament; or it may be at the saphenous opening. The site of narrowing should be divided superomedially for a distance of 4-6 mm to avoid all normally positioned vessels and other important structures. However, occasionally the obturator artery is replaced by an enlarged pubic branch of the inferior epigastric artery descending almost vertically to the obturator foramen. This vessel sometimes curves along the edge of the lacunar part of the inguinal ligament, encircling the neck of a hernial sac, and may be inadvertently cut during enlargement of the femoral ring in reducing a femoral hernia. The pubic tubercle is an important landmark in distinguishing inguinal from femoral hernias; the neck of the hernia is superomedial to it in inguinal hernia, but inferolateral in the femoral form. UPDATE Date Added: 09 August 2005 Update: The iliopubic tract: an important anatomical landmark in surgery. Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?
cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=10227675&query_hl=11 The iliopubic tract: an important anatomical landmark in surgery. UPDATE Date Added: 12 July 2005 Abstract: The posterior (preperitoneal approach and iliopubic tractd repair of inguinal and femoral hernias - an update. Click on the following line to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=12820025&query_hl=10 The posterior (preperitoneal) approach and iliopubic tract repair of inguinal and femoral hernias - an update.
© 2008 Elsevier
LESIONS OF THE INTERCOSTAL NERVES Lesions of individual intercostal nerves do not produce any appreciable clinical effects. Innervation of the anterolateral muscles is from several different nerves, and it requires several lesions to produce a significant reduction in the tone of any muscle. Because of the overlap between sequential dermatomes, significant cutaneous anaesthesia is felt only after sectioning of at least two or more sequential nerves. REFERENCES Cormack GC, George B, Lamberty H 1994 The Arterial Anatomy of Skin Flaps, 2nd edn. London, Edinburgh: Churchill Livingstone: 168-72. Lytle WJ 1979 Inguinal anatomy. J Anat 128: 581-94. Medline Similar articles Rizk NN 1980 A new description of the anterior abdominal wall in man and mammals. J Anat 131: 373-85. Medline Similar articles Shafik A 1977 The cremasteric muscle. In: Johnson AD, Gomes WR (eds) The Testis. Academic Press: New York.
© 2008 Elsevier
68 Posterior abdominal wall and retroperitoneum The posterior abdominal wall consists of fasciae, muscles and their vessels and spinal nerves. The overlying skin is continuous with that of the back. It is not easily defined, and is best described as that part of the abdominal wall lying between the two mid-dorsal lines, below the posterior attachments of the diaphragm and above the pelvis. It is continuous laterally with the anterolateral abdominal wall, superiorly with the posterior wall of the thorax behind the attachments of the diaphragm and inferiorly with the structures of the pelvis. The spinal column forms part of its structure and the muscles and fasciae of the back are closely related to it, especially posterolaterally. The major vessels and lymphatic channels, in addition to the peripheral autonomic nervous systems of the abdomen, pelvis and lower limbs lie on the posterior abdominal wall. These structures, together with several viscera (including the kidneys [Ch. 91], suprarenal (adrenal) glands [Ch. 89], pancreas [Ch. 87], ureters [Ch. 92] and parts of the gut tube [Chs 73, 76-82]), lie beneath the posterior parietal peritoneum. These tissues and their surrounding connective and fascial planes are collectively referred to as the retroperitoneum. It has been suggested that the retroperitoneum can be divided into several spaces according to their relationships to the fascial layers that surround the kidneys and ureters. In this description, the layers of the perirenal fascia (p. 1114) enclose a perirenal space containing the kidney, suprarenal gland, upper ureter and their neurovascular supply. The anterior layer of the perirenal fascia is continuous across the midline anterior to the main neurovascular structures of the retroperitoneum, and the right and left perirenal spaces communicate, although this channel is limited and contains many of the midline neurovascular structures of the retroperitoneum. Behind the posterior layer of the perirenal fascia lies the posterior pararenal space. Anterior to the anterior layer of the perirenal fascia lies the anterior pararenal space, in which lie several retroperitoneal parts of the gut tube, including the duodenum and pancreas. The anterior pararenal spaces are also continuous across the midline and are limited posteriorly by the anterior communicating layer of the perirenal fascia and anteriorly by the parietal peritoneum. This description helps to explain why moderate amounts of fluid, blood or pus collecting in the retroperitoneum tend to remain constrained within the space in which they are formed although, for pathological processes such as tumour invasion, the fascial planes provide a weak barrier to local spread. Several structures, such as the pancreas, are referred to as being retroperitoneal. However although they are derived embryologically from the gut tube, they are not readily separated from the other retroperitoneal structures. Several other structures, such as the descending colon, are also referred to as being retroperitoneal, but they remain separated from the other retroperitoneal structures by a clearly defined fascial plane, which corresponds with the plane of fusion of their mesentery during development. This is of relevance during surgical exposure of the retroperitoneal organs and in some pathological processes: those defined by clear fascial planes may be mobilized with little or no risk of bleeding,
whereas mobilization of the pancreas, for example, is difficult and often very vascular.
© 2008 Elsevier
SKIN AND SOFT TISSUES The skin of the back in the region of the posterior abdominal wall is similar to that of the rest of the trunk. It is supplied by vessels from the musculocutaneous branches of the lumbar arteries and veins, and receives its innervation from the dorsal rami of the lumbar spinal and lower thoracic nerves. The soft tissues of the posterior abdominal wall and retroperitoneum (p. 743) are composed of several distinct layers of fascia, which divide them into anatomically distinct compartments.
THORACOLUMBAR FASCIA (Figs 67.10, 68.1, 68.2)
page 1113 page 1114
Figure 68.1 Fascial layers of the upper posterior abdominal wall. A, Transverse section just below the level of the hilum of the kidney. For clarity, the deep muscles of the back have not been identified separately. B, The separate layers of the thoracolumbar fascia.
Figure 68.2 Axial CT scan of the upper abdomen. A, On soft-tissue windows, to demonstrate retroperitoneal anatomy. IVC, inferior vena cava. B, On narrower window widths, to show the anterior and posterior renal fascia.
The thoracolumbar fascia in the lumbar region is in three layers. The posterior layer is attached to the spines of the lumbar and sacral vertebrae and to the supraspinous ligaments. The middle layer is attached medially to the tips of the transverse processes of the lumbar vertebrae and the intertransverse ligaments, inferiorly to the iliac crest, and superiorly to the lower border of the twelfth rib and the lumbocostal ligament. The anterior layer covers quadratus lumborum and is attached medially to the anterior surfaces of the transverse processes of the lumbar vertebrae behind the lateral part of psoas major. Inferiorly, it is attached to the iliolumbar ligament and the adjoining part of the iliac crest. Superiorly, it is attached to the apex and inferior border of the twelfth rib and then extends to the transverse process of the first lumbar vertebra, to form the lateral arcuate ligament of the diaphragm. The posterior and middle layers of the thoracolumbar fascia unite at the lateral margin of erector spinae. At the lateral border of quadratus lumborum they are joined by the anterior layer, to form the aponeurotic origin of transversus abdominis (p. 1109).
OTHER FASCIAL LAYERS (Fig. 68.3) Psoas fascia
Psoas major is enclosed within a layer of fascia over its anterior surface. The medial border is continuous with the attachments of the muscle to the transverse processes of the lumbar vertebrae, the bodies of the lumbar vertebrae and the tendinous arches. Superiorly, the fascia forms part of the medial arcuate ligament. Laterally, the fascia blends with the fascia over quadratus lumborum in the upper part of the muscle and is continuous with the iliac fascia lower down. It separates the anterior mass of psoas major from the retroperitoneal structures lying on it. The fascia extends down into the thigh. Inflammatory collections arising from the paraspinal tissues or the retroperitoneal tissues that penetrate through the fascia tend to be confined by it, and they may track down the length of the muscle, to appear in the groin where the fascia is thinnest. Iliac fascia
The iliac fascia is continuous with and indistinguishable from the psoas fascia. The fascia blends with the anterior layer of the thoracolumbar fascia over quadratus lumborum in the upper retroperitoneum. Lower down, it is attached firmly to the inner aspect of the iliac crest and medially to the periosteum of the ilium at the pelvic brim. It is also attached in the abdomen to the iliopectineal eminence. Perirenal fascia <X ref 7.20.2.1>
page 1114 page 1115
Figure 68.3 Fascial layers of the lower posterior abdominal wall. Transverse section at the level of the fifth lumbar vertebra.
The perirenal fascia (p. 1270) is a multilaminated fascial layer that surrounds the kidney, suprarenal glands, upper ureter and associated fat, which all lie in the perirenal space. Although described as having anterior and posterior layers, these are continuous with each other laterally. The posterior layer of the renal fascia is adherent to the fascia over psoas major, the iliac fascia and the anterior layers of the thoracolumbar fascia. In the obese, there may be some loose adipose tissue between these layers, but it is rarely thick. The anterior part of the renal fascia separates the kidney and the perirenal space from the overlying anterior pararenal space and its associated viscera (on the right the duodenum, ascending colon and right colonic mesentery and on the left the duodenum, descending colon and left colonic mesentery). Inferiorly, the perirenal fascia continues down and encloses the ureter. It becomes progressively thinner towards the brim of the pelvis, where it is no longer distinguishable from the loose general connective tissue of the retroperitoneum. Lateroconal fascia
The lateroconal fascia is formed from the lateral aspect of the perirenal fascia and extends anterolaterally to fuse with the fascia over transversus abdominis. It divides the anterior and posterior pararenal spaces from each other, but is
thinnest in the inferior part of the retroperitoneum.
POSTERIOR EXTRAPERITONEAL CONNECTIVE TISSUE The retroperitoneum usually contains loose connective tissue between the fascial layers. This is particularly true around the renal fascia and anterior to the psoas and iliac fascia. In all but the thinnest individuals, there is some adipose tissue present in these areas, and in the obese it may be markedly thickened. The retroperitoneal arteries and veins lie within this tissue, but the branches of the lumbar plexus of nerves lie deep to it, beneath the iliac and psoas fascia.
© 2008 Elsevier
BONES The posterior abdominal wall is supported by the bony structures of the vertebral column and bony pelvis. These include the lower two ribs (p. 957), the twelfth thoracic and five lumbar vertebrae, and the sacrum and ilium (p. 1425), in addition to their interconnecting ligaments.
© 2008 Elsevier
MUSCLES (Figs 68.4, 68.5) The majority of the muscles of the posterior abdominal wall are functionally part of the lower limb or vertebral column. They provide the surface against which the neurovascular structures of the retroperitoneum lie, and they are supported and separated from the majority of the retroperitoneal structures by fascial layers.
QUADRATUS LUMBORUM Quadratus lumborum is an irregularly shaped quadrilateral muscle, which is broader at its inferior attachment than superiorly. Attachments Quadratus lumborum is attached below by aponeurotic fibres to the iliolumbar ligament and the adjacent portion of the iliac crest for c.5 cm. The superior attachment is to the medial half of the lower border of the twelfth rib, and by four small tendons to the apices of the transverse processes of the upper four lumbar vertebrae. Sometimes it is also attached to the transverse process or body of the twelfth thoracic vertebra. Occasionally, a second layer of this muscle is found in front of the first. This duplicated layer is attached to the upper borders of the transverse processes of the lower three or four lumbar vertebrae and to the lower margin and the lower part of the anterior surface of the twelfth rib. Relations Anterior to quadratus lumborum are the colon (ascending on the right, descending on the left), kidney, psoas major and minor, and diaphragm. The subcostal, iliohypogastric and ilioinguinal nerves lie on the fascia anterior to the muscle, but are bound down to it by the medial continuation of the transversalis fascia.
page 1115 page 1116
Figure 68.4 Muscles and nerves of the posterior abdominal wall. The left psoas major has been removed to expose the origins of the lumbar plexus and quadratus lumborum.
Vascular supply Quadratus lumborum is supplied by branches of the lumbar arteries, the arteria lumbalis ima, the lumbar branch of the iliolumbar artery and branches of the subcostal artery. Innervation Quadratus lumborum is innervated by the ventral rami of the twelfth thoracic and upper three or four lumbar spinal nerves. Actions Quadratus lumborum fixes the last rib, and acts as a muscle of inspiration by helping to stabilize the lower attachments of the diaphragm. It has been suggested that this action might also provide a fixed base for controlled relaxation of the diaphragm in the precise adjustment of expiration needed for speech and singing. With the pelvis fixed, quadratus acts upon the vertebral column, flexing it to the same side. When both muscles contract, they probably help to extend the lumbar part of the vertebral column.
PSOAS MAJOR Psoas major has several sites of abdominal attachment. Posteriorly, the attachments are to the anterior surfaces and lower borders of the transverse processes of all the lumbar vertebrae. The muscle arising from these attachments is referred to as the posterior mass. The muscle also has an anterior mass. It consists of two different sets of attachments. The first part is five slips of muscle attached to the bodies of two adjoining vertebrae and their intervertebral disc (from the twelfth thoracic vertebra and the thoracolumbar disc to the lumbosacral disc and the first sacral segment). The second part is a series of tendinous arches extending across the narrow parts of the bodies of the five lumbar vertebrae between these slips. The upper four lumbar intervertebral foramina bear important relations to these attachments of the muscle. The foramina lie anterior to the transverse processes (the posterior attachments) and posterior to the vertebral bodies, discs and tendinous arches (anterior attachments). The roots of the lumbar plexus therefore enter the muscle directly between the two masses and the plexus is lodged within it. The branches then emerge from the borders and surfaces of psoas major.
PSOAS MINOR
Figure 68.5 Muscles of the posterior abdominal wall demonstrated on magnetic resonance imaging. A, Coronal T2-weighted MR image. B, Axial T1-weighted MR image. IVC, inferior vena cava.
Psoas minor (p. 1444) is often absent but, when present, lies anterior to psoas major. It arises from the sides of the bodies of the twelfth thoracic and first lumbar vertebrae and from the disc between them. It ends in a long, flat tendon, which is attached to the pectineal line and iliopectineal eminence and, laterally, to the iliac fascia.
ERECTOR SPINAE Erector spinae (p. 764) do not form part of the posterior abdominal wall itself, but are closely associated with the fascial layers of the posterior wall.
ILIACUS Iliacus (p. 1446) is a triangular sheet of muscle that arises from the superior twothirds of the concavity of the iliac fossa, the inner lip of the iliac crest, the ventral sacroiliac and iliolumbar ligaments and the upper surface of the lateral part of the sacrum. In front, it reaches as far as the anterior superior and anterior inferior iliac spines, and receives a few fibres from the upper part of the capsule of the hip joint. Most of its fibres converge into the lateral side of the strong tendon of psoas major. It lines the posterior wall of the lesser pelvis formed by the ilium.
POSTERIOR ABDOMINAL WALL HERNIAS Herniation through the posterior abdominal wall is extremely rare. The fascial layers usually provide an excellent protection against protrusion of the posterior abdominal viscera, which are relatively immobile. However, the posterior free border of external oblique and the inferior free border of latissimus dorsi do give rise to an area of potential weakness, referred to as the lumbar triangle. Spontaneous hernias through this tissue are very rare in the absence of previous surgical access such as a nephrectomy.
© 2008 Elsevier
VASCULAR SUPPLY AND LYMPHATIC DRAINGE ABDOMINAL AORTA (Figs 68.6, 68.7) page 1116 page 1117
Figure 68.6 The abdominal aorta, inferior vena cava and their branches in the male. The fascia, lymphatics and connective tissue have been removed for clarity.
page 1117 page 1118
Figure 68.7 Magnetic resonance aortoiliac angiogram. A, Coronal reformat. B, Sagittal reformat.
The abdominal aorta begins at the median, aortic hiatus of the diaphragm, anterior to the inferior border of the twelfth thoracic vertebra and the thoracolumbar intervertebral disc. It descends anterior to the lumbar vertebrae to end at the lower border of the fourth lumbar vertebra, a little to the left of the midline, by dividing into two common iliac arteries. It diminishes rapidly in calibre from above downward, because its branches are large; however, the diameter of the vessel at any given height tends to increase slightly with age. The cadaveric superior and inferior calibres are between c.9 and 14 mm and 8 and 12 mm, respectively, with little difference between the sexes. The angle of aortic bifurcation varies widely, particularly in the elderly. It has been suggested that the relationship between aortic size and shape is a possible causative factor in the development of abdominal aortic aneurysm (Newman et al 1971). This may be caused by the reflection of transmitted pressure waves, which occurs at junctions between vessels. At the aortic bifurcation, pressure oscillations and possibly turbulence may be set up as a result of differences in the luminal diameters of the common iliac arteries, and so give rise to reflected waves that may injure the intima of the distal abdominal aorta. The role of the relative calibres of the iliac arteries remains uncertain (Shah et al 1978). Relations
The upper abdominal aorta is related anteriorly to the coeliac trunk and its branches. The coeliac plexus and the lesser sac lie between it and the left lobe of the liver and lesser omentum. Below this, the superior mesenteric artery leaves the aorta, crossing anterior to the left renal vein. The body of the pancreas, with the splenic vein on its posterior surface, extends obliquely up and to the left
across the abdominal aorta, separated from it by the superior mesenteric artery and left renal vein. Below the pancreas, the proximal parts of the gonadal arteries, and the third part of the duodenum, lie anteriorly. In its lowest part it is covered by the posterior parietal peritoneum and crossed obliquely by the origin of the small intestinal mesentery. The thoracolumbar intervertebral discs, the upper four lumbar vertebrae, intervening intervertebral discs and the anterior longitudinal ligament are all posterior to the abdominal aorta. Lumbar arteries arise from its dorsal aspect and cross posterior to it. The third and fourth (and sometimes second) left lumbar veins also cross behind it to reach the inferior vena cava. The aorta may overlap the anterior border of the left psoas major. On the right, the aorta is related above to the cisterna chyli and thoracic duct, the azygos vein and the right crus of the diaphragm, which overlaps and separates it from the inferior vena cava and right coeliac ganglion. Below the second lumbar vertebra, it is closely applied to the left side of the inferior vena cava. This close relationship occasionally allows the formation of an aorto-caval fistula, particularly after aneurysmal disease surgery or trauma to the aorta. On the left, the aorta is related above to the left crus of the diaphragm and left coeliac ganglion. Level with the second lumbar vertebra, it is related to the duodenojejunal flexure and the left sympathetic trunk, the fourth part of the duodenum and the inferior mesenteric vessels. Branches (Fig. 68.8)
The branches of the aorta are described as anterior, lateral and dorsal. The anterior and lateral branches are distributed to the viscera. The dorsal branches supply the body wall, vertebral column, vertebral canal and its contents. The aorta terminates by dividing into the right and left common iliac arteries. Anterior group (Fig. 68.9)
Coeliac trunk (coeliac axis) The coeliac trunk is the first anterior branch and arises just below the aortic hiatus at the level of T12/L1 vertebral bodies. It is c.1.5-2 cm long and passes almost horizontally forwards and slightly right above the pancreas and splenic vein. It divides into the left gastric, common hepatic and splenic arteries. The coeliac trunk may also give off one or both of the inferior phrenic arteries. The superior mesenteric artery may arise with the coeliac trunk as a common origin. One or more of the superior mesenteric branches may arise from the coeliac trunk. Anterior to the coeliac trunk lies the lesser sac. The coeliac plexus surrounds the trunk, sending extensions along its branches. On the right lie the right coeliac ganglion, right crus of the diaphragm and the caudate lobe of the liver. To the left lie the left coeliac ganglion, left crus of the diaphragm and the cardiac end of the stomach. The right crus may compress the origin of the coeliac trunk, giving the appearance of a stricture. The head of the pancreas and the splenic vein are inferior to the coeliac trunk. Superior mesenteric artery
Figure 68.8 The branches of the abdominal aorta.
The superior mesenteric artery (p. 1169) originates from the aorta c.1 cm below the coeliac trunk, at the level of the L1-2 intervertebral disc. It lies posterior to the splenic vein and the body of the pancreas. The left renal vein separates it from the aorta. It runs inferiorly and anteriorly, anterior to the uncinate process of the pancreas and the third part of the duodenum. UPDATE Date Added: 20 June 2006 Abstract: A computed tomography and ultrasonography study of superior mesenteric artery syndrome Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=15957095&query_hl=1&itool=pubmed_docsum Superior mesenteric artery syndrome: CT and ultrasonography findings. Unal B, Aktas A, Kemal G et al: Diagn Interv Radiol 11:90-95, 2005. Inferior mesenteric artery The inferior mesenteric artery is usually smaller in calibre than the superior mesenteric artery. It arises from the anterior or left anterolateral aspect of the aorta at about the level of the third lumbar vertebra, 3 or 4 cm above the aortic bifurcation and posterior to the horizontal part of the duodenum. Lateral group
Suprarenal artery The middle suprarenal artery arises from the lateral aspect of the abdominal aorta, level with the superior mesenteric artery. It ascends slightly, and runs over the crura of the diaphragm to the suprarenal glands, where it anastomoses with
the suprarenal branches of the phrenic and renal arteries. The right middle suprarenal artery passes behind the inferior vena cava and near the right coeliac ganglion. The left middle suprarenal artery passes close to the left coeliac ganglion, splenic artery and the superior border of the pancreas. Renal artery The renal arteries are two of the largest branches of the abdominal aorta and arise laterally from the vessel just below the origin of the superior mesenteric artery. The right is longer and usually arises slightly higher than the left. It passes posterior to the inferior vena cava, right renal vein, head of the pancreas and second part of the duodenum. The left renal artery arises a little lower down and passes behind the left renal vein, the body of the pancreas and the splenic vein. Gonadal artery The gonadal arteries (p. 1118) are two long, slender vessels that arise from the aorta a little inferior to the renal arteries. Each passes inferolaterally under the parietal peritoneum on psoas major. page 1118 page 1119
Figure 68.9 Multislice computed tomography angiogram of the abdominal aorta. A, Single axial slice from the volume data set at the level of the coeliac axis. B, Threedimensional surface-shaded reformat of the volume dataset acquired on axial multislice CT through the entire abdomen to produce a midline sagittal view of the anterior branches of the aorta.
Dorsal group
Inferior phrenic arteries The inferior phrenic arteries usually arise from the aorta, just above the level of the coeliac trunk. Occasionally they arise from a common aortic origin with the coeliac trunk, from the coeliac trunk itself or from the renal artery. They contribute to the arterial supply of the diaphragm. Each artery ascends and runs laterally anterior to the crus of the diaphragm, near the medial border of the suprarenal gland. The left passes behind the oesophagus and forwards on the left side of its diaphragmatic opening. The right passes posterior to the inferior vena cava and then along the right of the diaphragmatic opening for the inferior vena cava. Near the posterior border of the central tendon of the diaphragm, each divides into medial and lateral branches. The medial branch curves forwards to anastomose with its fellow in front of the central tendon and with the musculophrenic and pericardiacophrenic arteries. The lateral branch approaches the thoracic wall, and anastomoses with the lower posterior intercostal and musculophrenic arteries. The lateral branch of the right artery provides the arterial supply to the wall of the inferior vena cava, whereas the left sends ascending branches to the serosal surface of the abdominal oesophagus. Each inferior phrenic artery has two or three small suprarenal branches. The capsule of the liver and spleen may also receive a small supply from the arteries. Lumbar arteries The lumbar arteries arise in series with the posterior intercostal arteries. There are usually four on each side. They arise from the posterolateral aspect of the aorta, opposite the lumbar vertebrae. A fifth, smaller, pair occasionally arise from the median sacral artery, but lumbar branches of the iliolumbar arteries usually take their place. The lumbar arteries run posterolaterally on the first to the fourth lumbar vertebral bodies, behind the sympathetic trunks, to intervals between the lumbar transverse processes. From here they continue into the muscles of the posterior abdominal wall. The right arteries pass posterior to the inferior vena cava. The upper two on the right side and first left lumbar arteries lie posterior to
the corresponding crus of the diaphragm. Arteries of both sides pass under tendinous arches, which span the lateral concavities of the vertebral bodies, which form the attachment of psoas major. They run posterior to the muscle and the lumbar plexus. They then cross the anterior surface of quadratus lumborum, the upper three posterior, and the last usually anterior to it. At the lateral border of quadratus lumborum they pierce the posterior aponeurosis of transversus abdominis, running forward between it and internal oblique. They anastomose with one another and the lower posterior intercostal, subcostal, iliolumbar, deep circumflex iliac and inferior epigastric arteries. Dorsal branche
Each lumbar artery has a dorsal branch, which passes backwards between the adjacent transverse vertebral processes to supply the dorsal muscles of the back, the joints and skin of the back. This branch also has a spinal branch which enters the vertebral canal to supply its contents and adjacent vertebra, anastomosing with the arteries above and below it and across the midline. The spinal branch of the first lumbar artery supplies the terminal spinal cord itself; the remainder supply the cauda equina, meninges and vertebral canal. Occlusion of all or most of these arteries by dissection or aneurysm of the abdominal aorta may cause ischaemia of the cauda equina, producing the so-called 'cauda equina syndrome'. This is rare, however, even after infrarenal aortic graft surgery, because of the relatively good collateral circulation of the spinal cord arteries from the descending thoracic aorta. Branches of the lumbar arteries and their dorsal branches supply the adjacent muscles, fasciae, bones, red marrow, ligaments and joints of the vertebral column. Median sacral artery
The median sacral artery is a small branch that arises from the posterior aspect of the aorta a little above its bifurcation. It descends in the midline, anterior to the fourth and fifth lumbar vertebrae, sacrum and coccyx, and ends in the coccygeal body. At the level of the fifth lumbar vertebra, it is crossed by the left common iliac vein and often gives off a small lumbar artery (arteria lumbalis ima), small branches of which reach the anorectum via the anococcygeal ligament. Anterior to the fifth lumbar vertebra the median sacral artery anastomoses with a lumbar branch of the iliolumbar artery. Anterior to the sacrum it anastomoses with the lateral sacral arteries and sends branches into the anterior sacral foramina. Aortic surgery and prostheses page 1119 page 1120
Open surgical approaches to the abdominal aorta maybe associated with several potential complications. Injury to the large lymphatic trunks in the upper abdomen may lead to chylous ascites (p. 1122). Disruption of the intermesenteric and inferior mesenteric plexuses rarely causes clinically significant disturbances of autonomic function.
INFERIOR VENA CAVA (Fig. 68.10)
Figure 68.10 Multislice computed tomography three-dimensional surface-shaded angiogram of the abdominal aorta and inferior vena cavogram.
Figure 68.11 Multislice computed tomography demonstrating a double inferior vena cava (IVC). A, Axial CT at the level of the renal hilum, showing an IVC on either side of the aorta. B, Coronal reformat showing bilateral IVC joining above the level of the renal veins.
The inferior vena cava conveys blood to the right atrium from all structures below the diaphragm. The majority of its course is within the abdomen, but a small section lies within the fibrous pericardium in the thorax. It is formed by the junction of the common iliac veins anterior to the fifth lumbar vertebral body, a little to its right. It ascends anterior to the vertebral column, to the right of the aorta. It is contained in a deep groove on the posterior surface of the liver, or sometimes in a tunnel completed by a band of liver tissue. It crosses the tendinous part of the diaphragm between its median and right 'leaves' and inclines slightly anteromedially. Passing through the fibrous pericardium and through a posterior inflexion of the serous pericardium, it opens into the inferoposterior part of the right atrium. The abdominal portion of the inferior vena cava is devoid of valves. Relations of the abdominal part of the inferior vena cava
Anteriorly, the inferior vena cava is related to the right common iliac artery at its origin. It is crossed obliquely by the root of the mesentery and its contained vessels and nerves, and by the right gonadal artery. It lies behind the peritoneum of the posterior abdominal wall and the third part of the duodenum. It ascends behind the head of the pancreas and then the first part of the duodenum, separated from them by the common bile duct and portal vein. Above the duodenum it is again covered by the peritoneum of the posterior abdominal wall, which forms the posterior wall of the epiploic foramen. This separates it from the right free border of the lesser omentum and its contents. Above this it is intimately related to the liver anteriorly. The lower three lumbar vertebral bodies, their intervertebral discs and the anterior longitudinal ligament and right psoas major, sympathetic trunk and third and fourth lumbar arteries are all posterior to the inferior vena cava. Superior to these structures, the inferior vena cava is related posteriorly to the right crus of the diaphragm, the medial part of the right suprarenal gland, the right coeliac ganglion
and the right renal, middle suprarenal and inferior phrenic arteries. The right ureter, the second part of the duodenum, medial border of the right kidney and the right lobe of the liver are all lateral to the right side of the inferior vena cava. The aorta, the right crus of the diaphragm and the caudate lobe of the liver are all lateral to the left side. Numerous anomalies may occur in the anatomy of the inferior vena cava, mostly related to its complex formation (p. 1047). It is sometimes replaced, below the level of the renal veins, by two more or less symmetric vessels (Fig. 68.11), often associated with the failure of interconnection between the common iliac veins, and as a result of persistence on the left of a longitudinal channel (usually the supracardinal or subcardinal vein) that normally disappears in early fetal life. In complete visceral transposition, the inferior vena cava lies to the left of the aorta. page 1120 page 1121
Tributaries (Fig. 68.12)
The abdominal inferior vena cava usually receives the common iliac veins at its origin and the lumbar, right gonadal, renal, right suprarenal, inferior phrenic and hepatic veins during its course. Lumbar veins
Figure 68.12 Tributaries of the inferior vena cava and lumbar veins. Only the left lumbar venous system is shown, for clarity.
Four pairs of lumbar veins collect blood by dorsal tributaries from the lumbar muscles and skin. These branches anastomose with tributaries of the lumbar origin of the azygos and hemiazygos veins (p. 987). The abdominal tributaries to the lumbar veins drain blood from the posterior, lateral and anterior abdominal walls, including the parietal peritoneum. Anteriorly, the abdominal tributaries anastomose with branches of the inferior and superior epigastric veins. These anastomoses provide routes of continued venous drainage from the pelvis and lower limb to the heart in the event of inferior vena caval obstruction. The abdominal tributaries drain into the superior epigastric veins and hence via the internal thoracic veins to the superior vena cava, whereas the dorsal tributaries carry blood into the azygos and hemiazygos system and hence into the superior vena cava. Near the vertebral column, the lumbar veins drain the vertebral plexuses and are connected by the ascending lumbar vein, which is a vessel running longitudinally anterior to the roots of the transverse processes of the lumbar vertebrae. The third and fourth lumbar veins are fairly consistent in their course and pass forward on the sides of the corresponding vertebral bodies to enter the posterior aspect of the inferior vena cava. The left lumbar veins pass behind the abdominal aorta and are therefore longer. First and second lumbar veins are much more variable and may drain into the inferior vena cava, ascending lumbar vein, or lumbar azygos veins. They are often connected to each other, and the first lumbar vein does not usually enter the inferior vena cava directly, but turns down to join the second. Alternatively, the first lumbar vein may drain directly into the ascending lumbar vein or pass forward over the first lumbar vertebral body to the lumbar azygos vein. The second lumbar vein may join the inferior vena cava at or near the level of the renal veins. Sometimes it joins the third lumbar vein, or it may drain into the ascending lumbar vein. Ascending lumbar vein The ascending lumbar vein connects the common iliac, iliolumbar and lumbar veins. It lies between psoas major and the roots of the lumbar transverse processes. There is considerable variability in the course of this vein and the related lumbar azygos and first lumbar veins. Superiorly, it commonly joins the subcostal vein to form the azygos vein on the right and the hemiazygos on the left. These veins run forward over the twelfth thoracic vertebral body, and pass deep to the crura of the diaphragm and into the thorax. The ascending lumbar vein is usually joined by a small vessel from the back of the inferior vena cava or left renal vein on the left. This little vein represents the azygos line (p. 1047) and is referred to as the lumbar azygos vein. Sometimes the ascending lumbar vein ends in the first lumbar vein, which then skirts the first lumbar vertebra with the first lumbar artery, to join the lumbar azygos vein. In this circumstance the subcostal veins join the azygos vein on the right and the hemiazygos vein on the left. Gonadal veins
Only the left gonadal vein (pp. 1323, 1306) joins the inferior vena cava directly. It opens into its right anterolateral aspect at an acute angle just inferior to the level of the left renal vein. It is often double all the way to the level of entry into the inferior vena cava. Renal veins
The renal veins (p. 1247) are large calibre vessels, which lie anterior to the renal arteries and open into the inferior vena cava almost at right angles. The left is three times longer than the right in length (7.5 cm and 2.5 cm, respectively). The left vein lies on the posterior abdominal wall posterior to the splenic vein and body of the pancreas. Close to its opening into the inferior vena cava, it lies anterior to the aorta with the superior mesenteric artery just above it. The right renal vein lies posterior to the second part of the duodenum and sometimes the lateral part of the head of the pancreas.
Suprarenal vein
The right suprarenal vein drains directly into the inferior vena cava at the level of the twelfth thoracic vertebra. Inferior phrenic veins
The inferior phrenic veins run on the inferior surface of the central tendon of the diaphragm. They drain into the posterolateral aspect of the inferior vena cava around the level of the tenth thoracic vertebra. The left vein tends to drain at a slightly higher level than the right, and runs above the level of the oesophageal opening in the diaphragm. It may be double, with a branch draining into the left renal or suprarenal vein. Collaterals in inferior vena caval occlusion
Occlusion of the inferior vena cava may follow thrombosis resulting from hypercoagulable conditions, or embolism from lower limb or pelvic thromboses. The increased pressure within the lower body circulation leads to oedema of the legs and back, without ascites. Collateral venous circulation is established through a wide range of anastomoses between branches that drain ultimately to the superior vena cava. The lumbar veins connect to branches of the superior epigastric, circumflex iliac, lateral thoracic and posterior intercostal veins. They also anastomose with tributaries of the azygos, hemiazygos and lumbar azygos veins. The interconnecting vertebral venous plexuses provide an additional route of collateral circulation between the vena cavae. Inferior vena caval filters page 1121 page 1122
Recurrent embolization of clot from the pelvic or lower limbs veins may be a serious threat to life. In an effort to prevent life-threatening pulmonary embolism, a fenestrated filter may be placed within the inferior vena cava in an effort to trap the clot. These are most commonly inserted by radiological guidance via the internal jugular vein and superior vena cava and placed at a level below the origin of the renal veins. Progressive occlusion of the filter by clot may lead to symptoms of vena caval obstruction. Rarely, damage to the medial wall of the vena cava from the retaining hooks of the filter may lead to the development of an aorto-caval fistula as a result of the close proximity of the aorta at this level.
LYMPHATIC DRAINAGE (Figs 68.13, 68.14, 68.15) The lymphatic drainage of the muscles, deep tissues and integument of the posterior abdominal wall is broadly divided into four regions. The small upper left and upper right regions drain to the lateral aortic nodes and the ipsilateral axillary lymph nodes. The larger lower left and lower right portions drain to the lateral and retro-aortic lymph nodes, although some drainage also occurs to the left and right superficial inguinal nodes.
Figure 68.13 Lymphangiogram showing the lateral aortic and proximal iliac lymphatics. The radiograph was taken approximately 3 hours after the injection of contrast medium into the lymphatics of the dorsum of the foot. (Provided by GI Verney, Addenbrooke's Hospital, Cambridge; photographs prepared by Sarah-Jane Smith and Kevin Fitzpatrick on behalf of GKT School of Medicine, London.)
The lymphatic drainage of the abdominal viscera occurs almost exclusively through the cisterna chyli and the thoracic duct. Some lymphatic drainage may occur across the diaphragm from the bare area of the liver and the uppermost retroperitoneal tissues, but this is probably of little clinical consequence other than during obstruction of the thoracic duct. The lymph nodes of the retroperitoneum lie around the abdominal aorta and form pre-aortic, lateral aortic and retro-aortic groups. Collectively, they are referred to as the para-aortic lymph nodes and clinically it is difficult to distinguish between them, either at operation or on crosssectional imaging.
Cisterna chyli and abdominal lymph trunks
page 1122 page 1123
Figure 68.14 Lymphangiogram showing the lateral aortic and proximal iliac lymph nodes. The radiograph was taken approximately 24 hours after the injection of contrast medium into the lymphatics of the dorsum of the foot. Intravenous contrast opacifies the renal collecting system. (Provided by JB Kinmonth.)
Figure 68.15 Abdominal lymph node groups. The main pre-aortic groups are shown. Only the left-sided lateral nodes are shown, for clarity.
The abdominal origin of the thoracic duct lies to the right of the midline at the level of the lower border of the twelfth thoracic vertebral body or the thoracolumbar intervertebral disc. It receives all the lymph delivered by the four main abdominal lymph trunks, which converge to an elongated arrangement of channels, referred to as the abdominal confluence of lymph trunks. This may be a simple duct-like structure or be duplicated, triplicated or plexiform. When it is wider than the thoracic duct its interior is sometimes irregular and bilocular or trilocular, and may surround intercalated lymph nodes. It is only occasionally a simple, fusiform, saccular dilatation, and the widely used name 'cisterna chyli' best describes these forms. The abdominal confluence of lymph trunks extends from the beginning of the thoracic duct, vertically downwards for 5-7 cm, and lies anterolateral to the right of the first and second lumbar vertebral bodies and their intervening discs. It lies immediately to the right of the abdominal aorta. Along its length it lies between the territories containing the upper right lateral aortic lymph nodes and right-sided members of the coeliac and superior mesenteric pre-aortic groups, branches from which may drain directly into the various trunks. The upper two right lumbar arteries and the right lumbar azygos vein are between the confluence and the vertebral column. The medial edge of the right crus of the diaphragm lies anterior to the abdominal confluence of lymph trunks. The confluence receives the right and left lumbar and intestinal lymph trunks, although rarely these may drain directly into the thoracic duct.
The lumbar lymph trunks are formed by vessels draining from the lateral aortic nodes. Thus, either directly or after traversing intermediary groups, they carry lymph from: the lower limbs, the full thickness of the pelvic, perineal and infraumbilical abdominal walls, the deep tissues of most of the supra-umbilical abdominal walls, most of the pelvic viscera, gonads, kidneys and suprarenal glands. The intestinal lymph trunks receive vessels draining from coeliac nodes and, via these nodes, the superior and inferior mesenteric nodes, which are collectively the pre-aortic nodes. Either directly or via intermediary groups, they drain the entire abdominal gastrointestinal tract down to the anus. The intimate relationship of the cisterna chyli and abdominal lymph trunks to the abdominal aorta may lead to problems after aortic surgery, particularly dissections carried out around the aorta above the level of the coeliac axis. The large calibre of the trunks, coupled with the volume of lymph flowing through them, means that they do not readily self-seal after injury, and this lead to problematic recurrent chylous (lymphatic) ascites. The thoracic duct leaves the superior end of the cisterna chyli or the abdominal confluence and immediately passes through the aortic aperture of the diaphragm posterolateral to the aorta. Pre-aortic group
The pre-aortic groups tend to lie around the origins of the anterior (visceral) arteries and receive lymph from the gastrointestinal tract and its accessory structures (liver, spleen and pancreas) from the abdominal oesophagus to the level of the anus. They give rise to lymphatic vessels, which drain upwards to form the intestinal trunks that enter the abdominal confluence of lymph trunks. They are divisible into coeliac, superior mesenteric and inferior mesenteric groups, being near the origins of these arteries. Coeliac nodes
The coeliac nodes lie anterior to the abdominal aorta around the origin of the coeliac artery. They are a terminal group and receive lymph draining from the regional lymph nodes around the branches of the coeliac artery (left gastric, hepatic and pancreaticosplenic nodes). They also receive lymph from the lower pre-aortic groups (the superior mesenteric and inferior mesenteric). The coeliac nodes give rise to right and left intestinal lymph trunks. Gastric There are a great number of gastric lymph node groups. They drain the stomach, upper duodenum, abdominal oesophagus and the greater omentum. They drain to the coeliac group. Hepatic The hepatic nodes extend in the lesser omentum along the hepatic arteries and bile duct. They vary in number and site, but almost always occur at the junction of the cystic and common hepatic ducts (the cystic node), alongside the upper common bile duct and in the anterior border of the epiploic foramen. Hepatic nodes drain the majority of the liver, gallbladder and bile ducts, but also receive drainage from some parts of the stomach, duodenum and pancreas. They drain to the coeliac nodes and thence to the intestinal trunks. Pancreaticosplenic The pancreaticosplenic nodes drain the spleen, pancreas and part of the stomach. Their afferents join the coeliac nodes. Superior mesenteric and inferior mesenteric nodes
The superior and inferior mesenteric nodes lie anterior to the aorta near the origins of their respective arteries. The superior and inferior mesenteric nodes are
preterminal groups for the alimentary canal from the duodenojejunal flexure to the upper anal canal. They collect from outlying groups, including the mesenteric, ileocolic, colonic and pararectal nodes and drain into the coeliac nodes. Lateral aortic group
The lateral aortic nodes lie on either side of the abdominal aorta anterior to the medial margins of psoas major, diaphragmatic crura and sympathetic trunks. On the right, some nodes lie lateral and anterior to the inferior vena cava near the end of the right renal vein. Nodes rarely lie between the aorta and inferior vena cava where they are closely related. The lateral aortic nodes drain the viscera and other structures supplied by the lateral and dorsal aortic branches. The upper lateral groups receive the lymph drainage directly from the suprarenal glands, kidneys, ureters, gonads, uterine tubes and upper uterus. They also receive lymph directly from the deeper tissues of the posterior abdominal wall. Lymphatics from the pelvis, most of the pelvic viscera, the perineum and the anterolateral abdominal wall pass first to regional nodes largely related to the iliac arteries and their branches. These include the common iliac, external iliac, internal iliac and circumflex iliac nodes, in addition to the inferior epigastric and sacral nodes. Lymph from the lower limbs passes through the pelvic lymph nodes via the iliac groups. The lateral aortic group drains into the two lumbar lymph trunks, one on each side, which terminate in the confluence of lymph trunks. A few vessels may pass to the pre-aortic and retro-aortic nodes and others cross the midline to flow into the contralateral nodes, forming a loose plexus. page 1123 page 1124
Retro-aortic group
The retro-aortic group is the smallest of all the para-aortic lymph nodes. They have no particular areas of drainage, although they may receive some lymph directly from the paraspinal posterior abdominal wall. They effectively provide peripheral nodes of the lateral aortic groups and interconnect between surrounding groups.
© 2008 Elsevier
INNERVATION The posterior abdominal wall contains the origin of the lumbar plexus and numerous autonomic plexuses and ganglia, which lie close to the abdominal aorta and its branches. Lumbar ventral rami increase in size from first to last and are joined, near their origins, by grey rami communicantes from the four lumbar sympathetic ganglia. These rami, long and slender, accompany the lumbar arteries round the sides of the vertebral bodies, behind psoas major. Their arrangement is irregular: one ganglion may give rami to two lumbar nerves, one lumbar nerve may receive rami from two ganglia; rami often leave the sympathetic trunk between ganglia. The first and second, and sometimes the third, lumbar ventral rami are each connected with the lumbar sympathetic trunk by a white ramus communicans. The lumbar ventral rami descend laterally into psoas major. The first three and most of the fourth form the lumbar plexus; the smaller moiety of the fourth joins the fifth as a lumbosacral trunk, which joins the sacral plexus. The fourth is often termed the nervus furcalis, being divided between the two plexuses; but the third is occasionally the nervus furcalis; or both third and fourth may be furcal nerves, in which case the plexus is termed prefixed. More frequently, the fifth nerve is furcal, the plexus then being termed postfixed. These variations modify the sacral plexus.
LUMBAR PLEXUS (Fig. 68.16) The lumbar plexus lies within the substance of the posterior part of psoas major, anterior to the transverse processes of the lumbar vertebrae. It is formed by the first three, and most of the fourth, lumbar ventral rami. The first lumbar ramus receives a branch from the last thoracic ventral ramus. The paravertebral part of psoas major consists of posterior and anterior masses, which arise from different attachments. The lumbar plexus lies between these masses and hence is in 'line' with the intervertebral foramina. Although there may be minor variations, the most common arrangement of the plexus is described here. The first lumbar ventral ramus, joined by a branch from the twelfth thoracic ventral ramus, bifurcates, and the upper and larger part divides again into the iliohypogastric and ilioinguinal nerves. The smaller lower part unites with a branch from the second lumbar ventral ramus to form the genitofemoral nerve. The remainder of the second, third, and part of the fourth lumbar ventral rami join the plexus and divide into ventral and dorsal branches. Ventral branches of the second to fourth rami join to form the obturator nerve. The main dorsal branches of the second to fourth rami join to form the femoral nerve. Small branches from the dorsal branches of the second and third rami join to form the lateral femoral cutaneous nerve. The accessory obturator nerve, when it exists, arises from the third and fourth ventral branches. The lumbar plexus is supplied by branches from the lumbar vessels which supply psoas major. UPDATE Date Added: 27 July 2005 Abstract: Anatomy of ilioinguinal and iliohypogastric nerves in relation to trocar placement and low transverse incisions. Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=14710069&query_hl=9 Anatomy of ilioinguinal and iliohypogastric nerves in relation to trocar placement and low transverse incisions. Muscular Iliohypogastric Ilioinguinal Genitofemoral Lateral femoral cutaneous
T12, L1-4 L1 L1 L1, L2 L2, L3
Femoral Obturator Accessory obturator
L2-4 dorsal divisions L2-4 ventral divisions L2, L3
The branches of the lumbar plexus are: Division of constituent ventral rami into ventral and dorsal branches is not as clear in the lumbar and lumbosacral plexuses as it is in the brachial plexus. Anatomically, the obturator and tibial nerves (via the sciatic) arise from ventral divisions, and the femoral and peroneal nerves (via the sciatic) from dorsal divisions. Lateral branches of the twelfth thoracic and first lumbar ventral rami are drawn into the gluteal skin, but otherwise these nerves are typical. The second lumbar ramus is difficult to interpret. It not only contributes substantially to the femoral and obturator nerves, but also has an anterior terminal branch (the genital branch of the genitofemoral) and a lateral cutaneous branch (lateral femoral cutaneous nerve and the femoral branch of the genitofemoral). Anterior terminal branches of the third to fifth lumbar and first sacral rami are suppressed, but the corresponding parts of the second and third sacral rami supply the skin, etc. of the perineum. Inflammatory processes may occur in the posterior abdominal wall in the tissues anterior to psoas major, such as retrocaecal appendicitis on the right and diverticular abscess on the left. This may cause irritation of one or more of the branches of the lumbar plexus and lead to presenting symptoms of pain or dysaesthesia in the distribution of the affected nerves such as in the thigh, hip or buttock skin. Muscular branches
Small branches from all five lumbar roots. Iliohypogastric nerve
Distribution The iliohypogastric nerve originates from the L1 ventral ramus. It emerges from the upper lateral border of psoas major, crosses obliquely behind the lower renal pole and in front of quadratus lumborum. Above the iliac crest, it enters the posterior part of transversus abdominis. Between transversus abdominis and internal oblique, it divides into lateral and anterior cutaneous branches, and also supplies both muscles. The lateral cutaneous branch runs through internal and external oblique above the iliac crest, a little behind the iliac branch of the twelfth thoracic nerve, and is distributed to the posterolateral gluteal skin. The anterior cutaneous branch runs between and supplies internal oblique and transversus abdominis. It runs through internal oblique c.2 cm medial to the anterior superior iliac spine, and through the external oblique aponeurosis c.3 cm above the superficial inguinal ring, and is then distributed to the suprapubic skin. The iliohypogastric nerve connects with the subcostal and ilioinguinal nerves (Fig. 67.3). It is occasionally injured during an oblique surgical approach to the appendix. However, because the suprapubic skin is innervated from several sources, there is rarely any detectable sensory loss. Division of the iliohypogastric nerve above the anterior superior iliac spine may weaken the posterior wall of the inguinal canal and predispose to direct formation of a direct hernia. Motor The iliohypogastric nerve supplies a small motor contribution to transversus abdominis and internal oblique, including the conjoint tendon. Sensory The iliohypogastric nerve supplies sensory fibres to transversus abdominis, internal oblique and external oblique, and innervates the posterolateral gluteal and suprapubic skin. Ilioinguinal nerve
Distribution The ilioinguinal nerve originates from the L1 ventral ramus. It is smaller than the iliohypogastric nerve and arises with it from the first lumbar ventral ramus, to emerge from the lateral border of psoas major, with or just inferior to the iliohypogastric nerve. It passes obliquely across quadratus lumborum and the upper part of iliacus and enters transversus abdominis near the anterior end of the iliac crest. It sometimes connects with the iliohypogastric nerve at this point. It pierces internal oblique and supplies it and then traverses the inguinal canal below the spermatic cord. It emerges with the cord from the superficial inguinal ring to supply the proximal medial skin of the thigh and the skin over the root of the penis and upper part of the scrotum in males, or the skin covering the mons pubis and the adjoining labium majus in females. The ilioinguinal and iliohypogastric nerves are reciprocal in size. The ilioinguinal is occasionally very small and ends by joining the iliohypogastric, a branch of which then takes its place. Occasionally, the ilioinguinal nerve is completely absent when the iliohypogastric nerve supplies its territory. The nerve may be injured during inguinal surgery, particularly for hernia, which produces paraesthesia over the skin of the genitalia. Entrapment of the nerve during surgery may cause troublesome recurrent pain in this distribution. Motor The ilioinguinal nerve supplies motor nerves to transversus abdominis and internal oblique. page 1124 page 1125
Figure 68.16 The lumbar plexus, its branches and the muscles which they supply. The ventral branches of the ventral rami are coloured yellow and the dorsal branches orange.
Sensory The ilioinguinal nerve supplies sensory fibres to transversus abdominis and internal oblique. It innervates the medial skin of the thigh and the skin over the root of the penis and upper part of the scrotum in males or the skin covering the mons pubis and the adjoining labium majus in females.
Genitofemoral nerve
Distribution
page 1125 page 1126
The genitofemoral nerve originates from the L1 and L2 ventral rami. It is formed within the substance of psoas major and descends obliquely forwards through the muscle to emerge on its abdominal surface near the medial border, opposite the third or fourth lumbar vertebra. It descends beneath the peritoneum on psoas major, crosses obliquely behind the ureter and divides above the inguinal ligament into genital and femoral branches. It often divides close to its origin; its branches then emerge separately from psoas major. The genital branch crosses the lower part of the external iliac artery, enters the inguinal canal by the deep ring and supplies cremaster and the skin of the scrotum in males. In females, it accompanies the round ligament and ends in the skin of the mons pubis and labium majus. The femoral branch descends lateral to the external iliac artery, and sends a few filaments round it. It then crosses the deep circumflex iliac artery, passes behind the inguinal ligament and enters the femoral sheath lateral to the femoral artery. It pierces the anterior layer of the femoral sheath and fascia lata and supplies the skin anterior to the upper part of the femoral triangle. It connects with the femoral intermediate cutaneous nerve and supplies the femoral artery. The genital branch may be injured during inguinal surgery, in the same way as the ilioinguinal nerve. Motor The genitofemoral nerve innervates cremaster via the genital branch. Cutaneous The genitofemoral nerve innervates the skin of the scrotum in males or mons pubis and labium majus in females via the genital branch, and the anteromedial skin of the thigh via the femoral branch. Femoral nerve
The femoral nerve (p. 1455) descends through psoas major and emerges low on its lateral border. It passes between psoas major and iliacus deep to the iliac fascia and runs posterior to the inguinal ligament into the thigh. It gives off branches, which supply iliacus and pectineus and sensory fibres to the femoral artery. Posterior to the inguinal ligament, it lies lateral to the femoral artery and is separated from it by a part of psoas major. Lateral femoral cutaneous nerve of the thigh
The lateral femoral cutaneous nerve of the thigh (p. 1454) emerges from the lateral border of psoas major and crosses iliacus obliquely towards the anterior superior iliac spine. It supplies sensory fibres to the parietal peritoneum in the iliac fossa. The right nerve passes posterolateral to the caecum, separated from it by the iliac fascia and peritoneum. The left nerve passes behind the lower part of the descending colon. Both pass behind or through the inguinal ligament c.1 cm medial to the anterior superior iliac spine and anterior to, or through, sartorius into the thigh. UPDATE Date Added: 14 August 2006 Abstract: Anatomy of the lateral femoral cutaneous nerve and clinical relevance to iliac crest bone grafts Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? db=pubmed&cmd=Retrieve&dopt=AbstractPlus&list_uids=16414244&query_hl=20&itool=pubmed_docsum Lateral femoral cutaneous nerve and iliac crest bone grafts-anatomical and clinical considerations. Mischkowski RA, Selbach I, Neugebauer J, et al: Int J Oral Maxillofac Surg 35:366-372, 2006. UPDATE Date Added: 12 July 2005 Abstract: Lateral femoral cutaneous neuralgia: an anatomical insight. Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=12794914&query_hl=12 Lateral femoral cutaneous neuralgia: an anatomical insight.
Obturator nerve
The obturator nerve (p. 1455) descends within the substance of psoas major to emerge from its medial border at the level of the pelvic brim. It passes posterior to the common iliac vessels and lateral to the internal iliac vessels. It then descends on the lateral wall of the pelvis attached to the fascia over obturator internus. Here it lies anterosuperior to the obturator vessels before running into the obturator foramen to enter the thigh. It gives no branches in the abdomen or pelvis. Accessory obturator nerve
When present, the accessory obturator nerve (p. 1456) emerges from the medial border of psoas major and runs along this border over the posterior surface of the superior pubic ramus posterior to pectineus. It gives off branches here to supply pectineus and the hip joint, and it may join with the main obturator nerve.
LUMBOSACRAL PLEXUS The lumbosacral plexus (p. 1456) provides the nerve supply to the pelvis and lower limb, in addition to part of the autonomic supply to the pelvic viscera. It gives origin to the sciatic, inferior gluteal, superior gluteal and pudendal nerves (p. 1364), in addition to the nerves to quadratus femoris, obturator internus and the posterior cutaneous nerve of the thigh.
LUMBAR SYMPATHETIC SYSTEM The lumbar part of each sympathetic trunk usually contains four interconnected ganglia. It runs in the extraperitoneal connective tissue anterior to the vertebral column and along the medial margin of psoas major. Superiorly, it is continuous with the thoracic trunk posterior to the medial arcuate ligament. Inferiorly, it passes posterior to the common iliac artery and is continuous with the pelvic sympathetic trunk. On the right side, it lies posterior to inferior vena cava, and on the left it is posterior to the lateral aortic lymph nodes. It is anterior to most of the lumbar vessels, but may pass behind some lumbar veins. The first, second and sometimes third lumbar ventral spinal rami send white rami communicantes to the corresponding ganglia. Grey rami communicantes pass from all four lumbar ganglia to the lumbar spinal nerves. They are long, and accompany the lumbar arteries round the sides of the vertebral bodies, medial to the fibrous arches to which psoas major is attached. Four lumbar splanchnic nerves pass from the ganglia to join the coeliac, inferior mesenteric (or occasionally abdominal aortic) and superior hypogastric plexuses. The first lumbar splanchnic nerve, from the first ganglion, gives branches to the coeliac, renal and inferior mesenteric plexuses. The second nerve joins the inferior part of the intermesenteric or inferior mesenteric plexus. The third nerve arises from the third or fourth ganglion and passes anterior to the common iliac vessels to join the superior hypogastric plexus. The fourth lumbar splanchnic nerve from the lowest ganglion passes above the common iliac vessels to join the lower part of the superior hypogastric plexus, or the inferior hypogastric 'nerve'. Vascular branches from all lumbar ganglia join the abdominal aortic plexus. Fibres of the lower lumbar splanchnic nerves pass to the common iliac arteries and form a plexus, which continues along the internal and external iliac arteries as far as the proximal part of the femoral artery. Many postganglionic fibres travel in the muscular, cutaneous and saphenous branches of the femoral nerve, supplying vasoconstrictor nerves to the femoral artery and its branches in the thigh. Other postganglionic fibres travel via the obturator nerve to the obturator artery. Considerable uncertainty persists regarding the exact path of the sympathetic nerve supply to the lower limb (Pick 1970). Segmental sympathetic supply to abdominal viscera
For information on the plexuses, please see the appropriate chapters: coeliac plexus, superior mesenteric plexus, abdominal aortic plexus (intermesenteric plexus), inferior mesenteric plexus, superior hypogastric plexus, inferior hypogastric plexuses.
LUMBAR PARASYMPATHETIC SYSTEM The parasympathetic supply to the abdominal viscera is provided by the vagus nerve to the coeliac and superior mesenteric plexuses, and from the pelvic splanchnic nerves to the inferior mesenteric, superior hypogastric and inferior hypogastric plexuses.
PARA-AORTIC BODIES The para-aortic bodies are condensations of chromaffin tissue that are found in close relation to the aortic autonomic plexuses and lumbar sympathetic chains. They are largest in the fetus, become relatively smaller in childhood and have largely disappeared by adulthood. They are most commonly found as a pair of bodies lying anterolateral to the aorta in the region of the intermesenteric, inferior mesenteric and superior hypogastric plexuses. They may lie as high as the coeliac plexus, as low as the inferior hypogastric plexus in the pelvis, or may be closely applied to the sympathetic ganglia of the lumbar chain. Scattered cells that persist into adulthood may, rarely, be the sites of development of tumours of the chromaffin tissue (phaeochromocytoma), although these are much more commonly found arising from the cells of the suprarenal medulla. The wide variation in site of persistent para-aortic body tissue accounts for the range of locations of such tumours. REFERENCES Burkhill GJC, Healy JC 2000 Anatomy of the retroperitoneum. Imaging 12: 10-20. Newman DL, Gosling RG, Bowden R 1971 Changes in aortic distensibility and area ratio with the development of atherosclerosis. Atherosclerosis 14: 231-40. Medline Similar articles Full article Pick J 1970 The Autonomic Nervous System. Philadelphia: Lippincott. Shah PM, Scarton HA, Tsapogas MJ 1978 Geometric anatomy of the aorto-common iliac bifurcation. J Anat 126: 451-8. Both this and the Newman reference describe the anatomical factors that may affect the development of aortoiliac atherosclerosis. Medline Similar articles Full article
© 2008 Elsevier
69 Peritoneum and peritoneal cavity PERITONEUM AND PERITONEAL REFLECTIONS STRUCTURE OF THE PERITONEUM The peritoneum is the largest serous membrane in the body, and its arrangements are often complex. In males it forms a closed sac, but in females it is open at the lateral ends of the uterine tubes. It consists of a single layer of flat mesothelial cells lying on a layer of loose connective tissue. The mesothelium usually forms a continuous surface, but in some areas may be fenestrated. Neighbouring cells are joined by junctional complexes, but probably permit the passage of macrophages. The submesothelial connective tissue may also contain macrophages, lymphocytes and adipocytes (in some regions). Mesothelial cells may transform into fibroblasts, which may play an important role in the formation of peritoneal adhesions after surgery or inflammation of the peritoneum. The peritoneal cavity is a potential space between the parietal peritoneum, which lines the abdominal wall, and infoldings of visceral peritoneum, which suspend the abdominal viscera within the cavity. It contains a small amount of serous fluid, but is otherwise empty. The fluid lubricates the visceral peritoneum and allows the mobile viscera to glide freely on the abdominal wall and each other within the limits dictated by their attachments. It contains water, proteins, electrolytes and solutes derived from interstitial fluid in the adjacent tissues and from the plasma in the local blood vessels. It normally contains a few cells, including desquamated mesothelium, nomadic peritoneal macrophages, mast cells, fibroblasts, lymphocytes and other leukocytes. Some cells, particularly macrophages, migrate freely between the peritoneal cavity and the surrounding connective tissue. Lymphocytes provide both cellular and humoral immunological defence mechanisms within the peritoneal cavity. The intraperitoneal fluid is directed by gravity to dependent sites within the peritoneal cavity, and also flows in a cephalad direction as a consequence of the negative upper intra-abdominal pressures which are generated by respiration. The peritoneal cavity never contains gas in normal circumstances, although the amount of fluid may be increased in inflammatory conditions of the viscera. In females blood or fluid may occasionally escape from the uterine tubes into the pelvic peritoneal cavity during menstruation. Extraperitoneal connective tissue separates the parietal peritoneum from the muscular layers of the abdominal walls. The parietal peritoneum covering the anterior abdominal wall and pelvic walls is generally attached only loosely by this tissue, an arrangement which allows for considerable alteration in the size of the bladder and rectum. The extraperitoneal tissue on the inferior surface of the diaphragm and behind the linea alba is denser and more firmly adherent. The extraperitoneal tissue frequently contains large amounts of fat over the posterior abdominal wall, especially in obese males. The visceral peritoneum is firmly adherent to the underlying tissues and cannot be easily detached. Its connective tissue layer is often continuous with the fibrous matrix of the wall of the underlying viscera and rarely contains much loose connective or adipose tissue. The visceral peritoneum is often considered as part of the underlying viscus for clinical and pathological purposes such as the staging of carcinoma.
GENERAL ARRANGEMENT OF THE PERITONEUM (Fig. 69.1) In utero, the alimentary tract develops as a single tube suspended in the coelomic cavity by ventral and dorsal mesenteries (p. 1254). Ultimately, the ventral mesentery is largely resorbed, although some parts persist in the upper abdomen and form structures such as the falciform ligament. The mesenteries of the intestines in the adult are the remnants of the dorsal mesentery. The migration
and subsequent fixation of parts of the gastrointestinal tract produce the so-called 'retroperitoneal' segments of bowel (duodenum, ascending colon, descending colon, and rectum), and four separate intraperitoneal bowel loops suspended by mesenteries of variable lengths. These are all covered by visceral peritoneum which is continuous with the parietal peritoneum covering the posterior abdominal wall. The first intraperitoneal loop is formed by the intraperitoneal oesophagus, the stomach and first part of the duodenum. The second loop is made up of the duodenojejunal junction, jejunum, ileum and occasionally the caecum and proximal ascending colon. The third loop contains the transverse colon and the final loop contains the sigmoid colon and occasionally the distal descending colon. Where the visceral peritoneum encloses or suspends organs within the peritoneal cavity, the peritoneum and related connective tissues are known as the peritoneal ligaments, omenta or mesenteries. All but the greater omentum are composed of two layers of visceral peritoneum separated by variable amounts of connective tissue. The greater omentum is folded back on itself and is therefore made up of four layers of closely applied visceral peritoneum, which are separated by variable amounts of adipose tissue. The mesenteries attach their respective viscera to the posterior abdominal wall: the attachment is referred to as the mesenteric root, and the peritoneum of the mesentery is continuous with that of the posterior abdominal wall in this area. Although they are described as intraperitoneal, strictly speaking the suspended organs do not lie within the peritoneal cavity, because they are covered by visceral peritoneum. They are continuous with the extraperitoneal tissues, including the retroperitoneum, via subperitoneal tissue lying between the folds of visceral peritoneum. The loose areolar connective tissues of the extraperitoneal and subperitoneal tissues are sometimes conceptualized as 'spaces' because fluid or blood collects relatively easily within them. The subperitoneal tissues contain the neurovascular bundles and lymphatic channels which supply the suspended organs. In obese individuals, extensive adipose tissue within the mesenteries and omenta may obscure the neurovascular bundles. In contrast, in the very young, the elderly or the malnourished, the mesentery may contain very little adipose tissue and the neurovascular bundles are usually obvious. Peritoneum of the upper abdomen (Figs 69.2, 69.3, 69.4)
The abdominal oesophagus, stomach, liver and spleen all lie within a double fold of visceral peritoneum which runs from the posterior to the anterior abdominal wall. This fold has no recognized name, but has been referred to as the mesogastrium by Coakley and Hricak 1999 because it is derived from the fetal mesogastrium (p. 1254). It has a complex attachment to the wall of the abdominal cavity and gives rise to the falciform ligament, coronary ligaments, lesser omentum (gastrohepatic and hepatoduodenal ligaments), greater omentum (including gastrocolic ligament), gastrosplenic ligament, splenorenal ligament, and phrenicocolic ligament. The falciform ligament page 1127 page 1128
Figure 69.1 The posterior abdominal wall, showing the lines of peritoneal reflexion, after removal of the liver, spleen, stomach, jejunum, ileum, caecum, transverse colon and sigmoid colon. The various sessile (retroperitoneal) organs are seen shining through the posterior parietal peritoneum. Note: the ascending and descending colon, duodenum, kidneys, suprarenals, pancreas and inferior vena cava. Line WW represents the plane of Fig. 69.2. Line YY represents the plane of Fig. 69.4. Line XX represents the plane of Fig. 69.3. Line AA represents the plane of Fig. 69.5. Line BB represents the plane of Fig. 69.6. Line CC represents the plane of Fig. 69.7.
The falciform ligament is a thin anteroposterior peritoneal fold which connects the liver to the posterior aspect of the anterior abdominal wall just to the right of the midline. It extends inferiorly to the level of the umbilicus, and is widest between the liver and umbilicus. The ligament narrows superiorly as the distance between the liver and anterior abdominal wall reduces and narrows to just a centimetre or so in height over the superior surface of the liver. Its two peritoneal layers divide to enclose the liver and are continuous with the visceral peritoneum adherent to the surface of the liver. Superiorly, they are reflected onto the inferior surface of the diaphragm and are continuous with the parietal peritoneum over the right dome. At the posterior limit, or apex, of the falciform ligament, the two layers are also reflected vertically left and right, and are continuous with the anterior layers of the left triangular ligament and the superior layer of the coronary ligament of
the liver. The inferior aspect of the falciform ligament forms a free border where the two peritoneal layers become continuous with each other as they fold over to enclose the ligamentum teres (p. 1213). Because the peritoneum of the falciform ligament is continuous with that covering the posterior abdominal wall and the periumbilical anterior abdominal wall, blood arising from retroperitoneal haemorrhage (commonly acute haemorrhagic pancreatitis) may track between the folds of peritoneum and appear as haemorrhagic discolouration around the umbilicus (Cullen's sign). Spread of inflammatory change from the pancreas runs through the gastrohepatic ligament (lesser omentum) and then via the falciform ligament to the umbilicus. The peritoneal connections of the liver (Figs 69.5, 69.6, 69.7)
The liver is almost completely covered in visceral peritoneum, only the 'bare area' is in direct contact with the right dome of the diaphragm. Peritoneal folds, the ligaments of the liver, run from the liver to the surrounding viscera and to the abdominal wall and diaphragm. They are described in detail in Chapter 85 (p. 1213). The coronary ligament is formed by the reflection of the peritoneum from the diaphragm onto the posterior surfaces of the right lobe of the liver. Between the two layers of this ligament, a large area of liver, the bare area, is devoid of peritoneal covering. At this point the liver is attached to the diaphragm by areolar tissue and is in continuity inferiorly with the uppermost part of the anterior pararenal space. The layers of the coronary ligament are continuous on the right with the right triangular ligament. page 1128 page 1129
Figure 69.2 Sagittal section through the abdomen to the right of the epiploic foramen along one line of WW in Fig. 69.1.
The upper layer of the coronary ligament is continuous superiorly with the peritoneum over the inferior surface of the diaphragm and inferiorly with the peritoneum over the right and superior surfaces of the liver. At the lower margin of the bare area, the lower layer of the coronary ligament is continuous inferiorly with the peritoneum of the posterior abdominal wall over the right suprarenal gland and upper pole of the right kidney, and superiorly with the peritoneum over the inferior surface of the liver. The left triangular ligament is a double layer of peritoneum which extends over the superior border of the left lobe of the liver to a variable length. Medially, its anterior leaf is continuous with the left layer of the falciform ligament. The posterior layer is continuous with the left layer of the lesser omentum. The left triangular ligament lies in front of the abdominal part of the oesophagus, the upper end of the lesser omentum and part of the fundus of the stomach. Intraoperative division of the left triangular ligament permits mobilization of the left lobe of the liver in order to expose the abdominal oesophagus and crura of the diaphragm. The right triangular ligament is a short V-shaped fold formed by the approximation of the two layers of the coronary ligament at its right lateral end, and is continuous with the peritoneum of the right posterolateral abdominal wall. The coronary ligament is reflected inferiorly and is directly continuous with the peritoneum over the upper pole of the right kidney. This fold is sometimes referred to as the hepatorenal ligament. The recess formed between the peritoneum of the inferior surface of the liver, the hepatorenal ligament and the peritoneum over the right kidney is known as the hepatorenal pouch (of Morison). In the supine position this is the most dependent part of the peritoneal cavity in the upper abdomen, and is a common site of pathological fluid accumulation. The peritoneum is reflected inferolaterally from the posterior layer of the left triangular ligament onto the posterior abdominal wall above the oesophageal opening of the diaphragm. It lines the inferior surface of the left dome of the diaphragm and continues backwards onto the posterior abdominal wall. Inferiorly, it is reflected behind the spleen onto the most lateral part of the mesentery of the transverse colon and the splenic flexure. It continues down lateral to the descending colon into the pelvis, and forms the left paracolic 'gutter' (p. 1136). Medially, the peritoneum covering the left upper posterior abdominal wall is reflected anteriorly to form the left layer of the upper end of the lesser omentum, the peritoneum over the left aspect of the abdominal oesophagus and the left layer of the splenorenal ligament.
page 1129 page 1130
Figure 69.3 Section through the upper part of the abdominal cavity, along the line XX in Fig. 69.1. The boundaries of the epiploic foramen are shown and a small recess of the lesser sac is displayed in front of the head of the pancreas. Note that the transverse colon and its mesocolon are adherent to the posterior two layers of the greater omentum.
Figure 69.4 Sagittal section through the abdomen, approximately in the median plane. Compare with Fig. 69.1. The section cuts the posterior abdominal wall along the line YY in Fig. 69.1. The peritoneum is shown in blue except along its cut edges, which are left white.
page 1130 page 1131
Figure 69.5 Transverse section through the abdomen, at the level of line AA in Fig. 69.1, viewed from above. The peritoneal cavity is shown in dark blue; the peritoneum and its cut edges in lighter blue.
Figure 69.6 Transverse section through the abdomen at the level of the line BB in Fig. 69.1, viewed from above. Colours as in Fig. 69.5.
Figure 69.7 Transverse section through the abdomen at the level of the line CC in Fig. 69.1. The line passes through the bare area of the liver at the superior end of the lesser omentum. The parts of the left subphrenic space are clearly seen although they are continuous with each other.
From the inferior layer of the coronary ligament the peritoneum descends over the anterior surface of the right kidney to the front of the first part of the duodenum and hepatic flexure of the colon. Medially it passes in front of a short segment of the inferior vena cava between the duodenum and liver. At this point the peritoneum forms the posterior wall of the epiploic foramen. It forms a narrow strip, which broadens out as it continues across the midline onto the posterior wall of the lesser sac. It lines the posterior abdominal wall over the diaphragmatic crura, the upper abdominal aorta, the coeliac axis, nodes and plexus and the upper border of the pancreas. Inferiorly, below the liver, the peritoneum continues down on the posterior abdominal wall to the right of the ascending colon, forming the right paracolic 'gutter' between the anterolateral abdominal wall and colon. The lesser omentum page 1131 page 1132
The lesser omentum is formed of two layers of peritoneum separated by a variable amount of connective tissue and is derived from the ventral mesogastrium. It runs from the inferior visceral surface of the liver to the abdominal oesophagus, stomach, pylorus and first part of the duodenum. Superiorly, its attachment to the inferior surface of the liver forms an L-shape. The vertical component of the L is formed by the fissure for the ligamentum venosum. Inferiorly, the attachment turns and runs horizontally to complete the L in the portal fissure. The vertical and horizontal components of the lesser omentum run between the liver and the stomach and duodenum and are known as the gastrohepatic and hepatoduodenal ligaments, respectively. At the lesser curvature of the stomach, the layers of the lesser omentum split to enclose the stomach and are continuous with the visceral peritoneum covering the anterior and posterior surfaces of the stomach. The anterior layer of the lesser omentum descends from the fissure for the ligamentum venosum onto the anterior surface of the abdominal oesophagus, stomach and duodenum. The posterior layer descends from the posterior part of the fissure for the ligamentum venosum and runs onto the posterior surface of the stomach and pylorus. The lesser omentum forms the anterior surface of the lesser sac. The gastrohepatic ligament contains
the right and left gastric vessels, branches of the vagus nerves, and gastrohepatic lymph nodes between its two layers near their attachment to the stomach. The right lateral border of the lesser omentum is thickened and extends from the junction between the first and second parts of the duodenum to the porta hepatis. This border is free and forms the anterior wall of the epiploic foramen. It contains the portal vein, common bile duct, hepatic artery, portocaval lymph nodes and lymphatics and the hepatic plexus of nerves ensheathed in a perivascular fibrous capsule. Occasionally the free margin extends to the right of the epiploic foramen, runs to the gallbladder and is referred to as the cystoduodenal ligament. The left border of the lesser omentum is short and runs over the inferior surface of the diaphragm between the liver and medial aspect of the abdominal oesophagus. The lesser omentum is thinner on the left and may be fenestrated or incomplete. The variations in thickness are dependent upon the amount of connective tissue, especially fat. The greater omentum
The greater omentum is the largest peritoneal fold and hangs inferiorly from the greater curvature of the stomach. It is a double sheet: each sheet consists of two layers of peritoneum separated by a scant amount of connective tissue. The two sheets are folded back on themselves and are firmly adherent to each other. The anterior sheet descends from the greater curvature of the stomach and first part of the duodenum. The most anterior layer is continuous with the visceral peritoneum over the anterior surface of the stomach and duodenum and the posterior layer is continuous with the peritoneum over the posterior wall of the stomach and pylorus. The anterior sheet descends a variable distance into the peritoneal cavity and then turns sharply on itself to ascend as the posterior sheet. The posterior sheet passes anterior to the transverse colon and transverse mesocolon. It is attached to the posterior abdominal wall above the origin of the small intestinal mesentery and anterior to the head and body of the pancreas. The anterior layer of the posterior sheet is continuous with the peritoneum of the posterior wall of the lesser sac. The posterior layer is reflected sharply inferiorly and is continuous with the anterior layer of the transverse mesocolon. The posterior sheet is adherent to the transverse mesocolon at its root and is often known as the gastrocolic ligament, which is the supracolic part of the greater omentum. In early foetal life the greater omentum and transverse mesocolon are separate structures, and this arrangement sometimes persists. During surgical mobilization of the transverse colon, the plane between the transverse mesocolon and greater omentum can be entered opposite the taenia omentalis, and the greater omentum can be separated entirely from the transverse colon and mesocolon if required. Access into the lesser sac can be obtained via this approach if the upper part of the posterior sheet of the greater omentum is then divided. This gives a relatively bloodless plane of entry for surgical access to the posterior wall of the stomach and to the anterior surface of the pancreas. The greater omentum is continuous with the gastrosplenic ligament on the left, and on the right it extends to the start of the duodenum. A fold of peritoneum, the hepatocolic ligament, may run from either the inferior surface of the right lobe of the liver or the first part of the duodenum to the right side of the greater omentum or hepatic flexure of the colon. The right border of the greater omentum is occasionally adherent to the anterior surface of the ascending colon down as far as the caecum: its peritoneal layers are not continuous with the peritoneum over this part of the colon. A thin sheet of peritoneum referred to as Jackson's membrane may run from the front of the ascending colon and caecum to the posterolateral abdominal wall and may merge with the greater omentum. It often contains several small blood vessels. Occasionally, a band passes from the right side of the ascending colon to the lateral abdominal wall near the level of the iliac crest. It has been called the 'sustentaculum hepatis' but plays no role in the support of the liver. Other folds between the ascending colon and posterolateral abdominal wall may divide the right lateral paracolic gutter into several small recesses. Less commonly the
greater omentum is adherent to the anterior surface of the left colon; very occasionally it extends to the level of the sigmoid colon. When the undisturbed abdomen is opened, the greater omentum is frequently wrapped around the upper abdominal organs. Only rarely is it evenly dependent anterior to the coils of the small intestine, although this is the disposition which is frequently illustrated. It is usually thin and cribriform, but it always contains some adipose tissue and is a common site for storage of fat in obese individuals, particularly males. Between the two layers of the anterior fold of the greater omentum, close to the greater curvature of the stomach, the right and left gastroepiploic vessels form a wide anastomotic arc. Numerous vessels are given off from the arc and extend the full length of the omentum. This supply appears to exceed the metabolic requirements of the omentum, and perhaps reflect the role the greater omentum may play in peritoneal disease processes. The greater omentum is highly mobile and frequently becomes adherent to inflamed viscera within the abdominal cavity. This action may help to limit the spread of infection and the omentum may provide a source of well-vascularized tissue to take part in the early reparative process. It contains numerous fixed macrophages, which are easily mobilized. These may accumulate into dense, oval or round visible 'milky-spots'. The peritoneal connections of the spleen
The peritoneal connections of the spleen include the gastrosplenic, splenorenal and phrenicocolic ligaments, which suspend the spleen in the left upper quadrant of the abdomen. The gastrosplenic ligament runs between the greater curvature of the stomach and the hilum of the spleen and is in continuity with the left side of the greater omentum. The layers of the gastrosplenic ligament separate to enclose the spleen and then rejoin to form the splenorenal ligament and phrenicocolic ligaments. The splenorenal ligament extends from the spleen to the posterior abdominal wall and the phrenicocolic ligament extends to the anterolateral abdominal wall. The splenorenal ligament is formed from two layers of peritoneum. The anterior layer is continuous medially with the peritoneum of the posterior wall of the lesser sac over the left kidney and runs up to the splenic hilum where it is continuous with the posterior layer of the gastrosplenic ligament. The posterior layer of the splenorenal ligament is continuous laterally with the peritoneum over the inferior surface of the diaphragm and runs onto the splenic surface over the renal impression. The splenic vessels lie between the layers of the splenorenal ligament: the tail of the pancreas is usually present in its lower portion. The gastrosplenic ligament also has two layers. The posterior layer is continuous with the peritoneum of the splenic hilum and the peritoneum over the posterior surface of the stomach. The anterior layer is formed from the peritoneum reflected off the gastric impression of the spleen and is continuous with the peritoneum over the anterior surface of the stomach. The short gastric and left gastroepiploic branches of the splenic artery pass between the layers of the gastrosplenic ligament. The phrenicocolic ligament extends from the splenic flexure of the colon to the diaphragm at the level of the eleventh rib. It extends inferiorly and laterally and is continuous with the peritoneum of the lateral end of the transverse mesocolon at the lateral margin of the pancreatic tail, and the splenorenal ligament at the hilum of the spleen. A fan-shaped presplenic fold frequently extends from the anterior aspect of the gastrosplenic ligament near the greater curvature of the stomach below the inferolateral pole of the spleen. It blends with the phrenicocolic ligament. If the peritoneal attachments of the spleen are not recognized during surgery, the splenic capsule is at risk of injury and there may be subsequent serious bleeding. Downward traction on the phrenicocolic ligament during handling of the descending colon, especially during mobilization of the splenic flexure, may cause rupture of the splenic capsule. This is less likely if traction on the phrenicocolic
ligament is made laterally or medially. The superior border and anterior diaphragmatic surface of the splenic capsule are often adherent to the peritoneum of the greater omentum. Medial traction on the omentum during surgery may cause splenic capsular injury: such injury is less likely, if any limited traction required is applied inferiorly. Peritoneum of the lower abdomen page 1132 page 1133
The posterior surface of the lower anterior abdominal wall is lined by parietal peritoneum which extends from the linea alba centrally to the lateral border of quadratus lumborum. Here it is continuous with the peritoneum of the lateral paracolic gutter and is reflected over the sides and front of the ascending colon on the right and the descending colon on the left. Occasionally the ascending and descending colon are suspended by a short mesentery from the posterior abdominal wall. Between the ascending and descending colon, the peritoneum lines the posterior abdominal wall other than the oblique area, where it is reflected anteriorly to form the right and left layers of the small intestinal mesentery. Over the posterior abdominal wall it covers the left and right psoas major, inferior vena cava, duodenum, vertebral column and right and left ureters. At the upper extent of the posterior abdominal wall the peritoneum is reflected anteriorly and is continuous with the peritoneum of the posterior layer of the transverse mesocolon. Transverse mesocolon
The mesentery of the transverse colon is a broad fold of visceral peritoneum reflected anteriorly from the posterior abdominal wall and suspends the transverse colon in the peritoneal cavity. The root of the transverse mesocolon lies along an oblique line passing from the anterior aspect of the second part of the duodenum, over the head and neck of the pancreas, above the duodenojejunal junction and over the upper pole of the left kidney to the splenic flexure. It varies considerably in length but is shortest at either end. It contains the middle colic vessels and their branches together with branches of the superior mesenteric plexus, lymphatics and regional lymph nodes. Its two layers pass to the posterior surface of the transverse colon where they separate to cover the colon. The upper layer of peritoneum is reflected from the posterior abdominal wall immediately anteriorly and inferiorly and becomes continuous with the posterior layer of the greater omentum to which it is adherent. The lower layer of peritoneum of the transverse mesocolon is continuous with the peritoneum of the posterior abdominal wall. Lateral extensions of the transverse mesocolon produce two shelf-like folds on the right and left sides of the abdominal cavity. On the right the duodenocolic ligament extends from the transverse mesocolon at the hepatic flexure to the second part of the duodenum. On the left the phrenicocolic ligament extends from the transverse mesocolon at the splenic flexure to the diaphragm at the level of the eleventh rib. Near the uncinate process of the pancreas, the root of the transverse mesocolon is closely related to the upper limit of the root of the small intestinal mesentery. Mesentery of the small intestine
The mesentery of the small intestine is arranged as a complex fan formed from two layers of peritoneum (anterosuperior and posteroinferior) separated by connective tissue and vessels. The root of the mesentery lies along a line running diagonally from the duodenojejunal flexure on the left side of the second lumbar vertebral body to the right sacroiliac joint. The root crosses over the third part of the duodenum, aorta, inferior vena cava, right ureter and right psoas major. The length of the root of the mesentery is c.15 cm long in adults while the mesentery along its intestinal attachment is the same length as the small intestine (c.5 m), and consequently the mesentery is usually thrown into multiple folds along its intestinal border. The average depth of the mesentery from the root to the intestinal border is c.20 cm, but this varies along the length of the small intestine: it is shortest at the jejunum and terminal ileum and longest in the region of the
mid ileum. Its two peritoneal layers contain the jejunum, ileum, jejunal and ileal branches of the superior mesenteric vessels, branches of the superior mesenteric plexus, lacteals and regional lymph nodes. Because of the length and mobility of the mesentery, identification of the proximal and distal ends of a loop of small intestine may be difficult through small surgical incisions. Tracing the continuity of the right peritoneal layer of the mesentery onto the posterior abdominal wall above the root towards the ascending colon, and the continuity of the left layer towards the descending and sigmoid colon, may be useful in helping to orientate an individual loop of ileum. The mesentery of the small intestine is sometimes joined to the transverse mesocolon at the duodenojejunal junction by a peritoneal band. Occasionally the fourth part of the duodenum possesses a very short mesentery which is continuous with the upper end of the root of the small bowel mesentery. Pronounced bands of peritoneum may extend to the posterior abdominal wall at the terminal ileum. The root of the mesentery of the small intestine is continuous with the peritoneum surrounding the appendix and caecum in the right iliac fossa. Mesoappendix
The mesentery of the appendix is a triangular fold of peritoneum around the vermiform appendix. It is attached to the posterior surface of the lower end of the mesentery of the small intestine close to the ileocaecal junction. It usually reaches the tip of the appendix but sometimes fails to reach the distal third, in which case a vestigial low peritoneal ridge containing fat is present over the distal third. It encloses the blood vessels, nerves and lymph vessels of the vermiform appendix, and usually contains a lymph node. Sigmoid mesocolon
The sigmoid mesocolon shows individual variation in length and depth. The root of the sigmoid colon forms a shallow inverted V with an apex near the division of the left common iliac artery but may vary from a very short straight line at the pelvic brim to a long curved attachment. The upper, left end of the attachment runs medially over the left psoas major. The lower, right end passes into the pelvis towards the midline at the level of the third sacral vertebra. The root extends for a variable distance over the brim of the pelvis and the lower posterior abdominal wall. The anteromedial peritoneal layer of the mesentery of the sigmoid colon is continuous with the peritoneum of the lower left posterior abdominal wall and its posterolateral layer is continuous with the peritoneum of the pelvis and lateral abdominal wall. The proximal and distal ends of the sigmoid colon are occasionally joined together by a fibrous band which is usually associated with a narrow based sigmoid mesentery and may predispose the sigmoid colon to volvulus. Pronounced bands of peritoneum may also be found running from the proximal sigmoid colon to the posterior abdominal wall. The sigmoid and superior rectal vessels run between its layers and the left ureter descends into the pelvis behind its apex. Peritoneum of the lower anterior abdominal wall (Fig. 69.8)
The peritoneum of the lower anterior abdominal wall is raised into five ridges which diverge as they descend from the umbilicus. These are the median and right and left lateral and medial umbilical folds. The median umbilical fold extends from the umbilicus to the apex of the bladder and contains the urachus or its remnant (p. 1259). The obliterated umbilical artery lies under the medial umbilical fold which ascends from the pelvis to the umbilicus. The supravesical fossa lies between the medial and median umbilical folds on either side of the midline. The lateral umbilical fold covers the inferior epigastric artery below its entry into the rectus sheath, and is separated from the medial umbilical fold by the medial inguinal fossa. The lateral inguinal fossa lies lateral to the lateral umbilical fold, and covers the deep inguinal ring. The femoral fossa lies inferomedial to the lateral inguinal fossa, from which it is separated by the medial end of the inguinal ligament. It overlies the femoral ring (Chapter 67). Peritoneum of the pelvis
The parietal peritoneum of the posterior surface of the anterior abdominal wall and that lining the posterior abdominal wall continue into the pelvis as the pelvic peritoneum. The pelvic peritoneum then follows the surfaces of the true pelvic viscera and pelvic side walls although there are important differences between the sexes. Peritoneum of the male pelvis (Fig. 69.9) page 1133 page 1134
Figure 69.8 The infra-umbilical part of the anterior abdominal wall of a male subject: posterior surface, with the peritoneum in situ. Note the pelvic bones flanking the wide greater pelvis (middle) and narrower lesser pelvis (below) containing the bladder.
In males, the peritoneum of the left lower abdominal wall is reflected from the junction of the sigmoid colon and anterolateral surface of the rectum to line the brim and upper inner surface of the true pelvis. The peritoneum passes down into the true pelvis, lying over the anterior surface of the rectum, which then becomes an extraperitoneal organ. Laterally, the peritoneum is reflected to the pelvic side walls to form the right and left pararectal fossae: these vary in size according to the degree of distension of the rectum. The peritoneum is reflected anteriorly from the anterior surface of the rectum over the upper poles of the seminal vesicles and onto the posterior surface of the bladder, producing the rectovesical pouch. Anteriorly the rectovesical pouch is limited laterally by peritoneal folds, the sacrogenital folds, which extend from the sides of the bladder posteriorly to the anterior aspect of the sacrum. The peritoneum covers the superior surface of the bladder, and forms a paravesical fossa on each side limited laterally by a ridge of peritoneum which contains the ductus deferens. The size of the paravesical fossae depends on the volume of urine in the bladder. When the bladder is empty, a variable transverse vesical fold divides each fossa into two. The anterior ends of the sacrogenital folds may sometimes be joined by a ridge separating a middle fossa from the main rectovesical pouch. Between the paravesical and pararectal fossae the ureters and internal iliac vessels may cause slight elevations in the peritoneum. From the apex of the bladder the peritoneum extends superiorly along the posterior surface of the lower anterior abdominal wall to the umbilicus (p. 1132). When the bladder distends, the peritoneum is lifted from the lower anterior abdominal wall so that part of the anterior surface of the bladder is in direct contact with the posterior surface of the lower median area of the anterior abdominal wall. This relationship means that a well-distended bladder can be entered by direct puncture through the lower anterior abdominal wall without entering the peritoneal cavity (suprapubic puncture).
Peritoneum of the female pelvis
In females the peritoneum covers the upper rectum as it does in the male, but it descends further over the anterior surface of the rectum. The lateral limit of the pararectal and paravesical fossae is the peritoneum covering the round ligament of the uterus (p. 1133). The rectovesical pouch is occupied by the uterus and vagina. The peritoneum from the rectum is thus reflected anteriorly onto the posterior surface of the posterior fornix of the vagina and the uterus producing the recto-uterine pouch (of Douglas). The peritoneum covers the fundus of the uterus to its anterior (vesical) surface as far as the junction of the body and cervix, from which it is reflected forwards to the upper surface of the bladder, forming a shallow vesico-uterine pouch. Peritoneum is reflected from the bladder to the posterior surface of the anterior abdominal wall as it is in males. Marginal rectouterine folds correspond to the sacrogenital folds in males and pass back to the sacrum from the sides of the cervix, lateral to the rectum. Peritoneum is reflected from the anterior and posterior uterine surfaces to the lateral pelvic walls as the broad ligament of the uterus. This consists of anteroinferior and posterosuperior layers which are continuous at the upper border of the ligament (p. 1332). The broad ligament extends from the sides of the uterus to the lateral pelvic walls, and contains the uterine tubes in its free superior margins and the ovaries attached to its posterior layer. Below, it is continuous with the lateral pelvic parietal peritoneum. Between the ridges formed by the obliterated umbilical arteries and the ureter, the peritoneum forms a shallow depression on the lateral pelvic wall, the ovarian fossa, which lies behind the lateral attachment of the broad ligament. The ovary usually rests in the fossa in nulliparous females.
Figure 69.9 The peritoneum of the male pelvis: anterosuperior view. The median umbilical fold contains both the unpaired median and the paired medial umbilical ligaments in the plane of section in this subject.
© 2008 Elsevier
PERITONEUM AND PERITONEAL REFLECTIONS STRUCTURE OF THE PERITONEUM The peritoneum is the largest serous membrane in the body, and its arrangements are often complex. In males it forms a closed sac, but in females it is open at the lateral ends of the uterine tubes. It consists of a single layer of flat mesothelial cells lying on a layer of loose connective tissue. The mesothelium usually forms a continuous surface, but in some areas may be fenestrated. Neighbouring cells are joined by junctional complexes, but probably permit the passage of macrophages. The submesothelial connective tissue may also contain macrophages, lymphocytes and adipocytes (in some regions). Mesothelial cells may transform into fibroblasts, which may play an important role in the formation of peritoneal adhesions after surgery or inflammation of the peritoneum. The peritoneal cavity is a potential space between the parietal peritoneum, which lines the abdominal wall, and infoldings of visceral peritoneum, which suspend the abdominal viscera within the cavity. It contains a small amount of serous fluid, but is otherwise empty. The fluid lubricates the visceral peritoneum and allows the mobile viscera to glide freely on the abdominal wall and each other within the limits dictated by their attachments. It contains water, proteins, electrolytes and solutes derived from interstitial fluid in the adjacent tissues and from the plasma in the local blood vessels. It normally contains a few cells, including desquamated mesothelium, nomadic peritoneal macrophages, mast cells, fibroblasts, lymphocytes and other leukocytes. Some cells, particularly macrophages, migrate freely between the peritoneal cavity and the surrounding connective tissue. Lymphocytes provide both cellular and humoral immunological defence mechanisms within the peritoneal cavity. The intraperitoneal fluid is directed by gravity to dependent sites within the peritoneal cavity, and also flows in a cephalad direction as a consequence of the negative upper intra-abdominal pressures which are generated by respiration. The peritoneal cavity never contains gas in normal circumstances, although the amount of fluid may be increased in inflammatory conditions of the viscera. In females blood or fluid may occasionally escape from the uterine tubes into the pelvic peritoneal cavity during menstruation. Extraperitoneal connective tissue separates the parietal peritoneum from the muscular layers of the abdominal walls. The parietal peritoneum covering the anterior abdominal wall and pelvic walls is generally attached only loosely by this tissue, an arrangement which allows for considerable alteration in the size of the bladder and rectum. The extraperitoneal tissue on the inferior surface of the diaphragm and behind the linea alba is denser and more firmly adherent. The extraperitoneal tissue frequently contains large amounts of fat over the posterior abdominal wall, especially in obese males. The visceral peritoneum is firmly adherent to the underlying tissues and cannot be easily detached. Its connective tissue layer is often continuous with the fibrous matrix of the wall of the underlying viscera and rarely contains much loose connective or adipose tissue. The visceral peritoneum is often considered as part of the underlying viscus for clinical and pathological purposes such as the staging of carcinoma.
GENERAL ARRANGEMENT OF THE PERITONEUM (Fig. 69.1) In utero, the alimentary tract develops as a single tube suspended in the coelomic cavity by ventral and dorsal mesenteries (p. 1254). Ultimately, the ventral mesentery is largely resorbed, although some parts persist in the upper abdomen and form structures such as the falciform ligament. The mesenteries of the intestines in the adult are the remnants of the dorsal mesentery. The migration and subsequent fixation of parts of the gastrointestinal tract produce the so-called 'retroperitoneal' segments of bowel (duodenum, ascending colon, descending
colon, and rectum), and four separate intraperitoneal bowel loops suspended by mesenteries of variable lengths. These are all covered by visceral peritoneum which is continuous with the parietal peritoneum covering the posterior abdominal wall. The first intraperitoneal loop is formed by the intraperitoneal oesophagus, the stomach and first part of the duodenum. The second loop is made up of the duodenojejunal junction, jejunum, ileum and occasionally the caecum and proximal ascending colon. The third loop contains the transverse colon and the final loop contains the sigmoid colon and occasionally the distal descending colon. Where the visceral peritoneum encloses or suspends organs within the peritoneal cavity, the peritoneum and related connective tissues are known as the peritoneal ligaments, omenta or mesenteries. All but the greater omentum are composed of two layers of visceral peritoneum separated by variable amounts of connective tissue. The greater omentum is folded back on itself and is therefore made up of four layers of closely applied visceral peritoneum, which are separated by variable amounts of adipose tissue. The mesenteries attach their respective viscera to the posterior abdominal wall: the attachment is referred to as the mesenteric root, and the peritoneum of the mesentery is continuous with that of the posterior abdominal wall in this area. Although they are described as intraperitoneal, strictly speaking the suspended organs do not lie within the peritoneal cavity, because they are covered by visceral peritoneum. They are continuous with the extraperitoneal tissues, including the retroperitoneum, via subperitoneal tissue lying between the folds of visceral peritoneum. The loose areolar connective tissues of the extraperitoneal and subperitoneal tissues are sometimes conceptualized as 'spaces' because fluid or blood collects relatively easily within them. The subperitoneal tissues contain the neurovascular bundles and lymphatic channels which supply the suspended organs. In obese individuals, extensive adipose tissue within the mesenteries and omenta may obscure the neurovascular bundles. In contrast, in the very young, the elderly or the malnourished, the mesentery may contain very little adipose tissue and the neurovascular bundles are usually obvious. Peritoneum of the upper abdomen (Figs 69.2, 69.3, 69.4)
The abdominal oesophagus, stomach, liver and spleen all lie within a double fold of visceral peritoneum which runs from the posterior to the anterior abdominal wall. This fold has no recognized name, but has been referred to as the mesogastrium by Coakley and Hricak 1999 because it is derived from the fetal mesogastrium (p. 1254). It has a complex attachment to the wall of the abdominal cavity and gives rise to the falciform ligament, coronary ligaments, lesser omentum (gastrohepatic and hepatoduodenal ligaments), greater omentum (including gastrocolic ligament), gastrosplenic ligament, splenorenal ligament, and phrenicocolic ligament. The falciform ligament page 1127 page 1128
Figure 69.1 The posterior abdominal wall, showing the lines of peritoneal reflexion, after removal of the liver, spleen, stomach, jejunum, ileum, caecum, transverse colon and sigmoid colon. The various sessile (retroperitoneal) organs are seen shining through the posterior parietal peritoneum. Note: the ascending and descending colon, duodenum, kidneys, suprarenals, pancreas and inferior vena cava. Line WW represents the plane of Fig. 69.2. Line YY represents the plane of Fig. 69.4. Line XX represents the plane of Fig. 69.3. Line AA represents the plane of Fig. 69.5. Line BB represents the plane of Fig. 69.6. Line CC represents the plane of Fig. 69.7.
The falciform ligament is a thin anteroposterior peritoneal fold which connects the liver to the posterior aspect of the anterior abdominal wall just to the right of the midline. It extends inferiorly to the level of the umbilicus, and is widest between the liver and umbilicus. The ligament narrows superiorly as the distance between the liver and anterior abdominal wall reduces and narrows to just a centimetre or so in height over the superior surface of the liver. Its two peritoneal layers divide to enclose the liver and are continuous with the visceral peritoneum adherent to the surface of the liver. Superiorly, they are reflected onto the inferior surface of the diaphragm and are continuous with the parietal peritoneum over the right dome. At the posterior limit, or apex, of the falciform ligament, the two layers are also reflected vertically left and right, and are continuous with the anterior layers of the left triangular ligament and the superior layer of the coronary ligament of
the liver. The inferior aspect of the falciform ligament forms a free border where the two peritoneal layers become continuous with each other as they fold over to enclose the ligamentum teres (p. 1213). Because the peritoneum of the falciform ligament is continuous with that covering the posterior abdominal wall and the periumbilical anterior abdominal wall, blood arising from retroperitoneal haemorrhage (commonly acute haemorrhagic pancreatitis) may track between the folds of peritoneum and appear as haemorrhagic discolouration around the umbilicus (Cullen's sign). Spread of inflammatory change from the pancreas runs through the gastrohepatic ligament (lesser omentum) and then via the falciform ligament to the umbilicus. The peritoneal connections of the liver (Figs 69.5, 69.6, 69.7)
The liver is almost completely covered in visceral peritoneum, only the 'bare area' is in direct contact with the right dome of the diaphragm. Peritoneal folds, the ligaments of the liver, run from the liver to the surrounding viscera and to the abdominal wall and diaphragm. They are described in detail in Chapter 85 (p. 1213). The coronary ligament is formed by the reflection of the peritoneum from the diaphragm onto the posterior surfaces of the right lobe of the liver. Between the two layers of this ligament, a large area of liver, the bare area, is devoid of peritoneal covering. At this point the liver is attached to the diaphragm by areolar tissue and is in continuity inferiorly with the uppermost part of the anterior pararenal space. The layers of the coronary ligament are continuous on the right with the right triangular ligament. page 1128 page 1129
Figure 69.2 Sagittal section through the abdomen to the right of the epiploic foramen along one line of WW in Fig. 69.1.
The upper layer of the coronary ligament is continuous superiorly with the peritoneum over the inferior surface of the diaphragm and inferiorly with the peritoneum over the right and superior surfaces of the liver. At the lower margin of the bare area, the lower layer of the coronary ligament is continuous inferiorly with the peritoneum of the posterior abdominal wall over the right suprarenal gland and upper pole of the right kidney, and superiorly with the peritoneum over the inferior surface of the liver. The left triangular ligament is a double layer of peritoneum which extends over the superior border of the left lobe of the liver to a variable length. Medially, its anterior leaf is continuous with the left layer of the falciform ligament. The posterior layer is continuous with the left layer of the lesser omentum. The left triangular ligament lies in front of the abdominal part of the oesophagus, the upper end of the lesser omentum and part of the fundus of the stomach. Intraoperative division of the left triangular ligament permits mobilization of the left lobe of the liver in order to expose the abdominal oesophagus and crura of the diaphragm. The right triangular ligament is a short V-shaped fold formed by the approximation of the two layers of the coronary ligament at its right lateral end, and is continuous with the peritoneum of the right posterolateral abdominal wall. The coronary ligament is reflected inferiorly and is directly continuous with the peritoneum over the upper pole of the right kidney. This fold is sometimes referred to as the hepatorenal ligament. The recess formed between the peritoneum of the inferior surface of the liver, the hepatorenal ligament and the peritoneum over the right kidney is known as the hepatorenal pouch (of Morison). In the supine position this is the most dependent part of the peritoneal cavity in the upper abdomen, and is a common site of pathological fluid accumulation. The peritoneum is reflected inferolaterally from the posterior layer of the left triangular ligament onto the posterior abdominal wall above the oesophageal opening of the diaphragm. It lines the inferior surface of the left dome of the diaphragm and continues backwards onto the posterior abdominal wall. Inferiorly, it is reflected behind the spleen onto the most lateral part of the mesentery of the transverse colon and the splenic flexure. It continues down lateral to the descending colon into the pelvis, and forms the left paracolic 'gutter' (p. 1136). Medially, the peritoneum covering the left upper posterior abdominal wall is reflected anteriorly to form the left layer of the upper end of the lesser omentum, the peritoneum over the left aspect of the abdominal oesophagus and the left layer of the splenorenal ligament.
page 1129 page 1130
Figure 69.3 Section through the upper part of the abdominal cavity, along the line XX in Fig. 69.1. The boundaries of the epiploic foramen are shown and a small recess of the lesser sac is displayed in front of the head of the pancreas. Note that the transverse colon and its mesocolon are adherent to the posterior two layers of the greater omentum.
Figure 69.4 Sagittal section through the abdomen, approximately in the median plane. Compare with Fig. 69.1. The section cuts the posterior abdominal wall along the line YY in Fig. 69.1. The peritoneum is shown in blue except along its cut edges, which are left white.
page 1130 page 1131
Figure 69.5 Transverse section through the abdomen, at the level of line AA in Fig. 69.1, viewed from above. The peritoneal cavity is shown in dark blue; the peritoneum and its cut edges in lighter blue.
Figure 69.6 Transverse section through the abdomen at the level of the line BB in Fig. 69.1, viewed from above. Colours as in Fig. 69.5.
Figure 69.7 Transverse section through the abdomen at the level of the line CC in Fig. 69.1. The line passes through the bare area of the liver at the superior end of the lesser omentum. The parts of the left subphrenic space are clearly seen although they are continuous with each other.
From the inferior layer of the coronary ligament the peritoneum descends over the anterior surface of the right kidney to the front of the first part of the duodenum and hepatic flexure of the colon. Medially it passes in front of a short segment of the inferior vena cava between the duodenum and liver. At this point the peritoneum forms the posterior wall of the epiploic foramen. It forms a narrow strip, which broadens out as it continues across the midline onto the posterior wall of the lesser sac. It lines the posterior abdominal wall over the diaphragmatic crura, the upper abdominal aorta, the coeliac axis, nodes and plexus and the upper border of the pancreas. Inferiorly, below the liver, the peritoneum continues down on the posterior abdominal wall to the right of the ascending colon, forming the right paracolic 'gutter' between the anterolateral abdominal wall and colon. The lesser omentum page 1131 page 1132
The lesser omentum is formed of two layers of peritoneum separated by a variable amount of connective tissue and is derived from the ventral mesogastrium. It runs from the inferior visceral surface of the liver to the abdominal oesophagus, stomach, pylorus and first part of the duodenum. Superiorly, its attachment to the inferior surface of the liver forms an L-shape. The vertical component of the L is formed by the fissure for the ligamentum venosum. Inferiorly, the attachment turns and runs horizontally to complete the L in the portal fissure. The vertical and horizontal components of the lesser omentum run between the liver and the stomach and duodenum and are known as the gastrohepatic and hepatoduodenal ligaments, respectively. At the lesser curvature of the stomach, the layers of the lesser omentum split to enclose the stomach and are continuous with the visceral peritoneum covering the anterior and posterior surfaces of the stomach. The anterior layer of the lesser omentum descends from the fissure for the ligamentum venosum onto the anterior surface of the abdominal oesophagus, stomach and duodenum. The posterior layer descends from the posterior part of the fissure for the ligamentum venosum and runs onto the posterior surface of the stomach and pylorus. The lesser omentum forms the anterior surface of the lesser sac. The gastrohepatic ligament contains
the right and left gastric vessels, branches of the vagus nerves, and gastrohepatic lymph nodes between its two layers near their attachment to the stomach. The right lateral border of the lesser omentum is thickened and extends from the junction between the first and second parts of the duodenum to the porta hepatis. This border is free and forms the anterior wall of the epiploic foramen. It contains the portal vein, common bile duct, hepatic artery, portocaval lymph nodes and lymphatics and the hepatic plexus of nerves ensheathed in a perivascular fibrous capsule. Occasionally the free margin extends to the right of the epiploic foramen, runs to the gallbladder and is referred to as the cystoduodenal ligament. The left border of the lesser omentum is short and runs over the inferior surface of the diaphragm between the liver and medial aspect of the abdominal oesophagus. The lesser omentum is thinner on the left and may be fenestrated or incomplete. The variations in thickness are dependent upon the amount of connective tissue, especially fat. The greater omentum
The greater omentum is the largest peritoneal fold and hangs inferiorly from the greater curvature of the stomach. It is a double sheet: each sheet consists of two layers of peritoneum separated by a scant amount of connective tissue. The two sheets are folded back on themselves and are firmly adherent to each other. The anterior sheet descends from the greater curvature of the stomach and first part of the duodenum. The most anterior layer is continuous with the visceral peritoneum over the anterior surface of the stomach and duodenum and the posterior layer is continuous with the peritoneum over the posterior wall of the stomach and pylorus. The anterior sheet descends a variable distance into the peritoneal cavity and then turns sharply on itself to ascend as the posterior sheet. The posterior sheet passes anterior to the transverse colon and transverse mesocolon. It is attached to the posterior abdominal wall above the origin of the small intestinal mesentery and anterior to the head and body of the pancreas. The anterior layer of the posterior sheet is continuous with the peritoneum of the posterior wall of the lesser sac. The posterior layer is reflected sharply inferiorly and is continuous with the anterior layer of the transverse mesocolon. The posterior sheet is adherent to the transverse mesocolon at its root and is often known as the gastrocolic ligament, which is the supracolic part of the greater omentum. In early foetal life the greater omentum and transverse mesocolon are separate structures, and this arrangement sometimes persists. During surgical mobilization of the transverse colon, the plane between the transverse mesocolon and greater omentum can be entered opposite the taenia omentalis, and the greater omentum can be separated entirely from the transverse colon and mesocolon if required. Access into the lesser sac can be obtained via this approach if the upper part of the posterior sheet of the greater omentum is then divided. This gives a relatively bloodless plane of entry for surgical access to the posterior wall of the stomach and to the anterior surface of the pancreas. The greater omentum is continuous with the gastrosplenic ligament on the left, and on the right it extends to the start of the duodenum. A fold of peritoneum, the hepatocolic ligament, may run from either the inferior surface of the right lobe of the liver or the first part of the duodenum to the right side of the greater omentum or hepatic flexure of the colon. The right border of the greater omentum is occasionally adherent to the anterior surface of the ascending colon down as far as the caecum: its peritoneal layers are not continuous with the peritoneum over this part of the colon. A thin sheet of peritoneum referred to as Jackson's membrane may run from the front of the ascending colon and caecum to the posterolateral abdominal wall and may merge with the greater omentum. It often contains several small blood vessels. Occasionally, a band passes from the right side of the ascending colon to the lateral abdominal wall near the level of the iliac crest. It has been called the 'sustentaculum hepatis' but plays no role in the support of the liver. Other folds between the ascending colon and posterolateral abdominal wall may divide the right lateral paracolic gutter into several small recesses. Less commonly the
greater omentum is adherent to the anterior surface of the left colon; very occasionally it extends to the level of the sigmoid colon. When the undisturbed abdomen is opened, the greater omentum is frequently wrapped around the upper abdominal organs. Only rarely is it evenly dependent anterior to the coils of the small intestine, although this is the disposition which is frequently illustrated. It is usually thin and cribriform, but it always contains some adipose tissue and is a common site for storage of fat in obese individuals, particularly males. Between the two layers of the anterior fold of the greater omentum, close to the greater curvature of the stomach, the right and left gastroepiploic vessels form a wide anastomotic arc. Numerous vessels are given off from the arc and extend the full length of the omentum. This supply appears to exceed the metabolic requirements of the omentum, and perhaps reflect the role the greater omentum may play in peritoneal disease processes. The greater omentum is highly mobile and frequently becomes adherent to inflamed viscera within the abdominal cavity. This action may help to limit the spread of infection and the omentum may provide a source of well-vascularized tissue to take part in the early reparative process. It contains numerous fixed macrophages, which are easily mobilized. These may accumulate into dense, oval or round visible 'milky-spots'. The peritoneal connections of the spleen
The peritoneal connections of the spleen include the gastrosplenic, splenorenal and phrenicocolic ligaments, which suspend the spleen in the left upper quadrant of the abdomen. The gastrosplenic ligament runs between the greater curvature of the stomach and the hilum of the spleen and is in continuity with the left side of the greater omentum. The layers of the gastrosplenic ligament separate to enclose the spleen and then rejoin to form the splenorenal ligament and phrenicocolic ligaments. The splenorenal ligament extends from the spleen to the posterior abdominal wall and the phrenicocolic ligament extends to the anterolateral abdominal wall. The splenorenal ligament is formed from two layers of peritoneum. The anterior layer is continuous medially with the peritoneum of the posterior wall of the lesser sac over the left kidney and runs up to the splenic hilum where it is continuous with the posterior layer of the gastrosplenic ligament. The posterior layer of the splenorenal ligament is continuous laterally with the peritoneum over the inferior surface of the diaphragm and runs onto the splenic surface over the renal impression. The splenic vessels lie between the layers of the splenorenal ligament: the tail of the pancreas is usually present in its lower portion. The gastrosplenic ligament also has two layers. The posterior layer is continuous with the peritoneum of the splenic hilum and the peritoneum over the posterior surface of the stomach. The anterior layer is formed from the peritoneum reflected off the gastric impression of the spleen and is continuous with the peritoneum over the anterior surface of the stomach. The short gastric and left gastroepiploic branches of the splenic artery pass between the layers of the gastrosplenic ligament. The phrenicocolic ligament extends from the splenic flexure of the colon to the diaphragm at the level of the eleventh rib. It extends inferiorly and laterally and is continuous with the peritoneum of the lateral end of the transverse mesocolon at the lateral margin of the pancreatic tail, and the splenorenal ligament at the hilum of the spleen. A fan-shaped presplenic fold frequently extends from the anterior aspect of the gastrosplenic ligament near the greater curvature of the stomach below the inferolateral pole of the spleen. It blends with the phrenicocolic ligament. If the peritoneal attachments of the spleen are not recognized during surgery, the splenic capsule is at risk of injury and there may be subsequent serious bleeding. Downward traction on the phrenicocolic ligament during handling of the descending colon, especially during mobilization of the splenic flexure, may cause rupture of the splenic capsule. This is less likely if traction on the phrenicocolic
ligament is made laterally or medially. The superior border and anterior diaphragmatic surface of the splenic capsule are often adherent to the peritoneum of the greater omentum. Medial traction on the omentum during surgery may cause splenic capsular injury: such injury is less likely, if any limited traction required is applied inferiorly. Peritoneum of the lower abdomen page 1132 page 1133
The posterior surface of the lower anterior abdominal wall is lined by parietal peritoneum which extends from the linea alba centrally to the lateral border of quadratus lumborum. Here it is continuous with the peritoneum of the lateral paracolic gutter and is reflected over the sides and front of the ascending colon on the right and the descending colon on the left. Occasionally the ascending and descending colon are suspended by a short mesentery from the posterior abdominal wall. Between the ascending and descending colon, the peritoneum lines the posterior abdominal wall other than the oblique area, where it is reflected anteriorly to form the right and left layers of the small intestinal mesentery. Over the posterior abdominal wall it covers the left and right psoas major, inferior vena cava, duodenum, vertebral column and right and left ureters. At the upper extent of the posterior abdominal wall the peritoneum is reflected anteriorly and is continuous with the peritoneum of the posterior layer of the transverse mesocolon. Transverse mesocolon
The mesentery of the transverse colon is a broad fold of visceral peritoneum reflected anteriorly from the posterior abdominal wall and suspends the transverse colon in the peritoneal cavity. The root of the transverse mesocolon lies along an oblique line passing from the anterior aspect of the second part of the duodenum, over the head and neck of the pancreas, above the duodenojejunal junction and over the upper pole of the left kidney to the splenic flexure. It varies considerably in length but is shortest at either end. It contains the middle colic vessels and their branches together with branches of the superior mesenteric plexus, lymphatics and regional lymph nodes. Its two layers pass to the posterior surface of the transverse colon where they separate to cover the colon. The upper layer of peritoneum is reflected from the posterior abdominal wall immediately anteriorly and inferiorly and becomes continuous with the posterior layer of the greater omentum to which it is adherent. The lower layer of peritoneum of the transverse mesocolon is continuous with the peritoneum of the posterior abdominal wall. Lateral extensions of the transverse mesocolon produce two shelf-like folds on the right and left sides of the abdominal cavity. On the right the duodenocolic ligament extends from the transverse mesocolon at the hepatic flexure to the second part of the duodenum. On the left the phrenicocolic ligament extends from the transverse mesocolon at the splenic flexure to the diaphragm at the level of the eleventh rib. Near the uncinate process of the pancreas, the root of the transverse mesocolon is closely related to the upper limit of the root of the small intestinal mesentery. Mesentery of the small intestine
The mesentery of the small intestine is arranged as a complex fan formed from two layers of peritoneum (anterosuperior and posteroinferior) separated by connective tissue and vessels. The root of the mesentery lies along a line running diagonally from the duodenojejunal flexure on the left side of the second lumbar vertebral body to the right sacroiliac joint. The root crosses over the third part of the duodenum, aorta, inferior vena cava, right ureter and right psoas major. The length of the root of the mesentery is c.15 cm long in adults while the mesentery along its intestinal attachment is the same length as the small intestine (c.5 m), and consequently the mesentery is usually thrown into multiple folds along its intestinal border. The average depth of the mesentery from the root to the intestinal border is c.20 cm, but this varies along the length of the small intestine: it is shortest at the jejunum and terminal ileum and longest in the region of the
mid ileum. Its two peritoneal layers contain the jejunum, ileum, jejunal and ileal branches of the superior mesenteric vessels, branches of the superior mesenteric plexus, lacteals and regional lymph nodes. Because of the length and mobility of the mesentery, identification of the proximal and distal ends of a loop of small intestine may be difficult through small surgical incisions. Tracing the continuity of the right peritoneal layer of the mesentery onto the posterior abdominal wall above the root towards the ascending colon, and the continuity of the left layer towards the descending and sigmoid colon, may be useful in helping to orientate an individual loop of ileum. The mesentery of the small intestine is sometimes joined to the transverse mesocolon at the duodenojejunal junction by a peritoneal band. Occasionally the fourth part of the duodenum possesses a very short mesentery which is continuous with the upper end of the root of the small bowel mesentery. Pronounced bands of peritoneum may extend to the posterior abdominal wall at the terminal ileum. The root of the mesentery of the small intestine is continuous with the peritoneum surrounding the appendix and caecum in the right iliac fossa. Mesoappendix
The mesentery of the appendix is a triangular fold of peritoneum around the vermiform appendix. It is attached to the posterior surface of the lower end of the mesentery of the small intestine close to the ileocaecal junction. It usually reaches the tip of the appendix but sometimes fails to reach the distal third, in which case a vestigial low peritoneal ridge containing fat is present over the distal third. It encloses the blood vessels, nerves and lymph vessels of the vermiform appendix, and usually contains a lymph node. Sigmoid mesocolon
The sigmoid mesocolon shows individual variation in length and depth. The root of the sigmoid colon forms a shallow inverted V with an apex near the division of the left common iliac artery but may vary from a very short straight line at the pelvic brim to a long curved attachment. The upper, left end of the attachment runs medially over the left psoas major. The lower, right end passes into the pelvis towards the midline at the level of the third sacral vertebra. The root extends for a variable distance over the brim of the pelvis and the lower posterior abdominal wall. The anteromedial peritoneal layer of the mesentery of the sigmoid colon is continuous with the peritoneum of the lower left posterior abdominal wall and its posterolateral layer is continuous with the peritoneum of the pelvis and lateral abdominal wall. The proximal and distal ends of the sigmoid colon are occasionally joined together by a fibrous band which is usually associated with a narrow based sigmoid mesentery and may predispose the sigmoid colon to volvulus. Pronounced bands of peritoneum may also be found running from the proximal sigmoid colon to the posterior abdominal wall. The sigmoid and superior rectal vessels run between its layers and the left ureter descends into the pelvis behind its apex. Peritoneum of the lower anterior abdominal wall (Fig. 69.8)
The peritoneum of the lower anterior abdominal wall is raised into five ridges which diverge as they descend from the umbilicus. These are the median and right and left lateral and medial umbilical folds. The median umbilical fold extends from the umbilicus to the apex of the bladder and contains the urachus or its remnant (p. 1259). The obliterated umbilical artery lies under the medial umbilical fold which ascends from the pelvis to the umbilicus. The supravesical fossa lies between the medial and median umbilical folds on either side of the midline. The lateral umbilical fold covers the inferior epigastric artery below its entry into the rectus sheath, and is separated from the medial umbilical fold by the medial inguinal fossa. The lateral inguinal fossa lies lateral to the lateral umbilical fold, and covers the deep inguinal ring. The femoral fossa lies inferomedial to the lateral inguinal fossa, from which it is separated by the medial end of the inguinal ligament. It overlies the femoral ring (Chapter 67). Peritoneum of the pelvis
The parietal peritoneum of the posterior surface of the anterior abdominal wall and that lining the posterior abdominal wall continue into the pelvis as the pelvic peritoneum. The pelvic peritoneum then follows the surfaces of the true pelvic viscera and pelvic side walls although there are important differences between the sexes. Peritoneum of the male pelvis (Fig. 69.9) page 1133 page 1134
Figure 69.8 The infra-umbilical part of the anterior abdominal wall of a male subject: posterior surface, with the peritoneum in situ. Note the pelvic bones flanking the wide greater pelvis (middle) and narrower lesser pelvis (below) containing the bladder.
In males, the peritoneum of the left lower abdominal wall is reflected from the junction of the sigmoid colon and anterolateral surface of the rectum to line the brim and upper inner surface of the true pelvis. The peritoneum passes down into the true pelvis, lying over the anterior surface of the rectum, which then becomes an extraperitoneal organ. Laterally, the peritoneum is reflected to the pelvic side walls to form the right and left pararectal fossae: these vary in size according to the degree of distension of the rectum. The peritoneum is reflected anteriorly from the anterior surface of the rectum over the upper poles of the seminal vesicles and onto the posterior surface of the bladder, producing the rectovesical pouch. Anteriorly the rectovesical pouch is limited laterally by peritoneal folds, the sacrogenital folds, which extend from the sides of the bladder posteriorly to the anterior aspect of the sacrum. The peritoneum covers the superior surface of the bladder, and forms a paravesical fossa on each side limited laterally by a ridge of peritoneum which contains the ductus deferens. The size of the paravesical fossae depends on the volume of urine in the bladder. When the bladder is empty, a variable transverse vesical fold divides each fossa into two. The anterior ends of the sacrogenital folds may sometimes be joined by a ridge separating a middle fossa from the main rectovesical pouch. Between the paravesical and pararectal fossae the ureters and internal iliac vessels may cause slight elevations in the peritoneum. From the apex of the bladder the peritoneum extends superiorly along the posterior surface of the lower anterior abdominal wall to the umbilicus (p. 1132). When the bladder distends, the peritoneum is lifted from the lower anterior abdominal wall so that part of the anterior surface of the bladder is in direct contact with the posterior surface of the lower median area of the anterior abdominal wall. This relationship means that a well-distended bladder can be entered by direct puncture through the lower anterior abdominal wall without entering the peritoneal cavity (suprapubic puncture).
Peritoneum of the female pelvis
In females the peritoneum covers the upper rectum as it does in the male, but it descends further over the anterior surface of the rectum. The lateral limit of the pararectal and paravesical fossae is the peritoneum covering the round ligament of the uterus (p. 1133). The rectovesical pouch is occupied by the uterus and vagina. The peritoneum from the rectum is thus reflected anteriorly onto the posterior surface of the posterior fornix of the vagina and the uterus producing the recto-uterine pouch (of Douglas). The peritoneum covers the fundus of the uterus to its anterior (vesical) surface as far as the junction of the body and cervix, from which it is reflected forwards to the upper surface of the bladder, forming a shallow vesico-uterine pouch. Peritoneum is reflected from the bladder to the posterior surface of the anterior abdominal wall as it is in males. Marginal rectouterine folds correspond to the sacrogenital folds in males and pass back to the sacrum from the sides of the cervix, lateral to the rectum. Peritoneum is reflected from the anterior and posterior uterine surfaces to the lateral pelvic walls as the broad ligament of the uterus. This consists of anteroinferior and posterosuperior layers which are continuous at the upper border of the ligament (p. 1332). The broad ligament extends from the sides of the uterus to the lateral pelvic walls, and contains the uterine tubes in its free superior margins and the ovaries attached to its posterior layer. Below, it is continuous with the lateral pelvic parietal peritoneum. Between the ridges formed by the obliterated umbilical arteries and the ureter, the peritoneum forms a shallow depression on the lateral pelvic wall, the ovarian fossa, which lies behind the lateral attachment of the broad ligament. The ovary usually rests in the fossa in nulliparous females.
Figure 69.9 The peritoneum of the male pelvis: anterosuperior view. The median umbilical fold contains both the unpaired median and the paired medial umbilical ligaments in the plane of section in this subject.
© 2008 Elsevier
PERITONEAL CAVITY GENERAL ARRANGEMENT OF THE PERITONEAL CAVITY The peritoneal cavity is a single continuous space between the parietal peritoneum lining the abdominal wall and the visceral peritoneum enveloping the abdominal organs. It consists of a main region, termed the greater sac, which is equivalent to the main abdominal cavity surrounding the majority of the abdominal and pelvic viscera. The lesser sac, or omental bursa, is a small diverticulum lined with peritoneum, which is situated behind the stomach and lesser omentum and in front of the pancreas and retroperitoneum (p. 1135). These two areas communicate via the epiploic foramen. For clinical purposes the peritoneal cavity can be divided into several spaces because pathological processes are often contained within these spaces and their anatomy may influence diagnosis and treatment. It is useful to divide the peritoneal cavity into two main compartments, supramesocolic and inframesocolic, which are partially separated by the transverse colon and its mesentery (the latter connects the transverse colon to the posterior abdominal wall). The pelvic peritoneal spaces are described above (p. 1133). Supramesocolic compartment
The supramesocolic space lies above the transverse mesocolon between the diaphragm and the transverse colon. It can be arbitrarily divided into right and left supramesocolic spaces. These regions can be further subdivided into a number of subspaces, which are normally in communication, but are frequently subdivided by inflammatory adhesions in disease. The right supramesocolic space can be divided into three subspaces; the right subphrenic space, the right subhepatic space, and the lesser sac. The left supramesocolic space can be divided into two subspaces; the left subphrenic space and the left perihepatic space. Right subphrenic space
The right subphrenic space lies between the diaphragm and the anterior, superior and right lateral surfaces of the right lobe of the liver. It is bounded on the left side by the falciform ligament and behind by the upper layer of the coronary ligament. It is a relatively common site for collections of fluid after right sided abdominal inflammation. Right subhepatic space (hepatorenal recess)
The right subhepatic space lies between the right lobe of the liver and the right kidney. It is bounded superiorly by the inferior layer of the coronary ligament, laterally by the right lateral abdominal wall, posteriorly by the anterior surface of the upper pole of the right kidney and medially by the second part of the duodenum, hepatic flexure, transverse mesocolon and part of the head of the pancreas. In the supine position the posterior right subhepatic space is more dependent than the right paracolic gutter: postoperative infected fluid collections are common in this location. Lesser sac (omental bursa)
The lesser sac is a cavity lined with peritoneum and connected to the larger general peritoneal cavity (greater sac) by the epiploic foramen. It is considered part of the right supramesocolic space because embryologically the liver grows into the right peritoneal space and stretches the dorsal mesentery to form the lesser sac behind the stomach. The sac varies in size according to the size of the viscera making up its walls. It has posterior and anterior walls as well as superior, inferior, right and left borders. The anterior wall is made up of the posterior peritoneal layer of the lesser omentum, the peritoneum over the posterior wall of the stomach and first part of the duodenum, and the uppermost part of the anterior layer of the greater omentum. At its right border, the anterior wall is mostly formed by the lesser omentum but moving towards the left, the lesser omentum becomes progressively shorter and more of the anterior wall is formed by the posterior aspect of the stomach and greater omentum. The posterior wall is formed mainly by the peritoneum covering the posterior abdominal wall in this area. In the lower part, the posterior wall is made up of the anterior layer of the posterior sheet of the greater omentum as it lies on the transverse mesocolon. The posterior wall covers, from below upwards, a small part of the head and the whole neck and body of the pancreas, the medial part of the anterior aspect of the left kidney, most of the left suprarenal (adrenal) gland, the commencement of the abdominal aorta and coeliac artery and part of the diaphragm. The inferior phrenic, splenic, left gastric and hepatic arteries lie partly behind the bursa. Many of these structures form the 'bed' of the stomach and are separated from it only by the linings of the lesser sac. The superior border of the lesser sac is narrow and lies between the right side of the oesophagus and the upper end of the fissure for the ligamentum venosum. Here peritoneum of the posterior wall of the lesser sac is reflected anteriorly from the diaphragm to join the posterior layer of the lesser omentum. The inferior border of the lesser sac runs along the line of the fusion of the layers of the greater omentum. This runs from the gastrosplenic ligament to the peritoneal fold behind the first part of the duodenum. In cases where the layers are not completely adherent to each other, the lesser sac may extend as far as the bottom of the two sheets of the greater omentum. In adults, even in these circumstances of separation of the layers, the lowest extent of the inferior border is rarely below the level of the transverse colon. The right border of the lesser sac is formed by the reflection of the peritoneum from the pancreatic neck and head onto the inferior aspect of the first part of the duodenum. The line of this reflection ascends to the left, along the medial side of the gastroduodenal artery. Near the upper duodenal margin the right border joins the floor of the epiploic foramen round the hepatic artery proper. The epiploic foramen thus forms a break in the right border. Above the epiploic foramen the right border is formed by the reflection of peritoneum from the diaphragm to the right margin of the caudate lobe of the liver and along the left side of the inferior vena cava, enclosing the hepatic recess.
The left border of the lesser sac runs from the left end of the root of the transverse mesocolon and is mostly formed by the inner layer of peritoneum of the splenorenal and gastrosplenic ligaments. The part of the lesser sac lying between the splenorenal and gastrosplenic ligaments is referred to as the splenic recess. Above the level of the spleen, the two ligaments are merged as the short gastrophrenic ligament, which passes forwards from the diaphragm to the posterior aspect of the fundus of the stomach and forms part of the upper left border of the lesser sac. The two layers of the gastrophrenic ligament diverge near the abdominal oesophagus, leaving part of the posterior gastric surface devoid of peritoneum. The left gastric artery runs forwards here into the lesser omentum. The lesser sac is narrowed by two crescentic peritoneal folds produced by the hepatic and left gastric arteries. The left gastropancreatic fold overlies the left gastric artery as it runs from the posterior abdominal wall to the lesser curvature of the stomach. The right gastropancreatic fold overlies the hepatic artery as it runs from the posterior abdominal wall to the lesser omentum. The folds vary in size. When prominent, they divide the lesser sac into a smaller superior and a larger inferior recess. The superior recess lies posterior to the lesser omentum and liver, and encloses the caudate lobe of the liver, which is covered by peritoneum on both its anterior and posterior surfaces. It extends superiorly into the fissure for the ligamentum venosum and lies adjacent to the right crus of the diaphragm posteriorly. The inferior recess of the lesser sac lies between the stomach and pancreas and is contained in the double sheet of the greater omentum. Epiploic foramen (of Winslow)
page 1135 page 1136
The epiploic foramen (foramen of Winslow, aditus to the lesser sac), is a short, vertical slit, c.3 cm height in adults, in the upper part of the right border of the lesser sac. It leads into the greater sac. The hepatoduodenal ligament, which is formed by the thickened right edge of the lesser omentum extending from the flexure between the first and second parts of the duodenum, forms the anterior margin of the foramen. The anterior border contains the common bile duct (on the right), portal vein (posteriorly) and hepatic artery (on the left) between its two layers. Superiorly the peritoneum of the posterior layer of the hepatoduodenal ligament runs over the caudate lobe of the liver which forms the roof of the epiploic foramen. This layer of peritoneum is then reflected onto the inferior vena cava which forms the posterior margin of the epiploic foramen. At the upper border of the first part of the duodenum the peritoneum runs forwards from the inferior vena cava, above the head of the pancreas, and is continuous with the posterior layer of the lesser omentum, forming the floor of the epiploic foramen. A narrow passage, the vestibule of the lesser sac, may be found to the left of the foramen between the caudate process and the first part of the duodenum. To the right, the rim of the foramen is continuous with the peritoneum of the greater sac. The roof is continuous with the peritoneum on the inferior surface of the right hepatic lobe. The anterior and posterior walls of the foramen are normally apposed.
Left subphrenic space
The left subphrenic space lies between the diaphragm, the anterior and superior surfaces of the left lobe of the liver, the anterosuperior surface of the stomach and the diaphragmatic surface of the spleen. It is bounded to the right by the falciform ligament and behind by the anterior layer of the left triangular ligament. It is much enlarged in the absence of the spleen and is a common site for fluid collection particularly after splenectomy. The left subphrenic space is substantially larger than the right and is sometimes described as being divided into anterior and posterior parts, although no obvious demarcation exists in the absence of disease. The left posterior subphrenic space is small and lies between the fundus of the stomach and the diaphragm above the origin of the splenorenal ligament. The left anterior subphrenic space is large and lies between the superior and anterolateral surfaces of the spleen and the left dome of the diaphragm. Inferiorly and medially, this space is bounded by the splenorenal, gastrosplenic, and phrenicocolic ligaments which produces a partial barrier to the left paracolic gutter. This may explain why left subphrenic collections are less frequent than right subphrenic collections following lower abdominal and pelvic surgery, but the left subphrenic space is the commonest site of fluid collection after upper abdominal, particularly splenic, surgery. Left perihepatic space
The left perihepatic space is sometimes subdivided into anterior and posterior spaces. The posterior perihepatic space is also known as the left subhepatic space or gastrohepatic recess. The left anterior perihepatic space lies between the anterosuperior surface of the left lobe of the liver and diaphragm. The left posterior perihepatic space lies inferior to the left lobe of the liver, and extends into the fissure for the ligamentum venosum on the right, anterior to the main portal vein. Posteriorly, the lesser omentum separates this space from the superior recess of the lesser sac. On the left, the space is bounded by the lesser curvature of the stomach. Inframesocolic compartment
The inframesocolic compartment lies below the transverse mesocolon and transverse colon are far as the true pelvis. It is divided in two unequal spaces by the root of the mesentery of the small intestine. It contains the right and left paracolic gutters lateral to the ascending and descending colon. As a consequence of the mobility of the transverse mesocolon and mesentery of the small intestine, disease processes are rarely well contained within these spaces, and fluid within the infracolic space tends to descend into the pelvis or the paracolic gutters. Right infracolic space
The right infracolic space is a triangular space. It is smaller than its counterpart on the left, and lies posterior and inferior to the transverse colon and mesocolon and to the right of the small intestinal mesentery. The space is narrowest inferiorly because the attachment of the root of the mesentery of the small intestine lies well to the right of the midline. The vermiform appendix often lies in the lower part of the right infracolic space.
Left infracolic space
The left infracolic space is larger than its counterpart on the right and is in free communication with the pelvis to the right of the midline. It lies posterior and inferior to the transverse colon and mesocolon and to the left of the mesentery of the small intestine. The sigmoid colon and its mesentery may partially restrict the flow of fluid or blood into the pelvis to the left of the midline. Paracolic gutters
The right and left paracolic gutters are peritoneal recesses on the posterior abdominal wall lying alongside the ascending and descending colon. The main paracolic gutter lies lateral to the colon on each side. A less obvious medial paracolic gutter may be formed, especially on the right side, if the colon possesses a short mesentery for part of its length. The right (lateral) paracolic gutter runs from the superolateral aspect of the hepatic flexure of the colon, down the lateral aspect of the ascending colon, and around the caecum. It is continuous with the peritoneum as it descends into the pelvis over the pelvic brim. Superiorly, it is continuous with the peritoneum which lines the hepatorenal pouch and, through the epiploic foramen, the lesser sac. Bile, pus or blood released from viscera anywhere along its length may run along the gutter and collect in sites quite remote from the organ of origin. In supine patients, infected fluid from the right iliac fossa may ascend in the gutter to enter the lesser sac. In patients nursed in a sitting position, fluid from the stomach, duodenum or gallbladder may run down the gutter to collect in the right iliac fossa or pelvis and may mimic acute appendicitis or form a pelvic abscess. The right paracolic gutter is larger than the left, which together with the partial barrier provided by the phrenicocolic ligament, may explain why right subphrenic collections are more common than left subphrenic collections. Extraperitoneal subphrenic spaces
There are two potential 'spaces' which actually lie outside the peritoneal coverings of the abdomen but are of clinical relevance because of the possibility that fluid collections will accumulate in them. The right extraperitoneal space is bounded by the two layers of the coronary ligament, the bare area of the liver and the inferior surface of the right dome of the diaphragm. The left extraperitoneal space lies anterior to the left suprarenal gland and upper pole of the left kidney. It contains extraperitoneal connective tissue. Clinical management of fluid collections in the peritoneal cavity
Fluid collections frequently occur within the peritoneal cavity as a result of a wide range of pathological processes. In the absence of any inflammation, peritoneal adhesions or previous surgery, serous fluid is almost always distributed freely between the peritoneal spaces and is not confined to any particular area. Simple ascites, for example, can therefore be drained freely from any convenient dependent part of the peritoneal cavity. This is most commonly performed by blind or ultrasound guided insertion of a catheter into the lower left or right paracolic gutters. These spaces usually readily fill with fluid and although the colon and some loops of small bowel may be present, their relatively mobility results in very little risk of injury to them.
Fluid collections caused by inflammatory processes are often much more viscid because they contain pus, fibrin or blood and are usually associated with peritoneal inflammation which results in, at least transient, peritoneal adhesions. These factors mean that collections may become localized if the flow of fluid is restricted by the, partial compartmentalization of the peritoneum. Once collected in one 'space', this fluid often becomes further confined by ongoing inflammation and may even form a truly walled-off cavity over time. Any of the spaces of the peritoneum may develop a collection but the subphrenic, subhepatic and pelvic spaces are the commonest since they are most well defined by the fixed peritoneal folds and organs forming their boundaries. These spaces are also the most dependent spaces within the peritoneum in the supine position and consequently any initially free fluid tends to gravitate to them. Surgical access to the peritoneal spaces is rarely necessary today because of the great advances which have been made in radiologically guided drainage. When necessary, lateral subcostal or intercostal incisions may give adequate access to the subphrenic spaces and the anterior wall of the rectum is also a useful route to access the rectouterine or rectovesical space. Computerized tomography or ultrasound guided drainage offers a much more reliable and versatile method of accessing even difficult spaces such as subhepatic, perihepatic, paracolic or even intermesenteric collections. Posterolateral translumbar or trans-sciatic approaches can be used to access these more difficult areas. Peritoneal dialysis
The mesothelium resembles vascular endothelium in being a dialysing membrane which fluids and small molecules may traverse. Numerous endocytic vesicles occur near the cell surfaces, the remaining cytoplasm being poor in organelles, indicating low metabolic activity. Normally the volumes of fluid transmitted by peritoneal surfaces are small, but large volumes may be administered via the intraperitoneal route. Conversely, substances such as urea can be dialysed from blood into fluid circulated through the peritoneal cavity. Ventriculoperitoneal shunts page 1136 page 1137
The absorptive capabilities of the peritoneum can be used to absorb excess transitional fluids from several sites in the body. The commonest of these is the absorption of cerebrospinal fluid drained from the intracerebral ventricles or the intrathecal space via a fine calibre catheter. The catheter can be placed within the peritoneum with a one way valve preventing reflux of peritoneal fluid into the cerebrospinal fluid. The fluid is then continuously absorbed maintaining a low pressure within the intrathecal or intraventricular space.
RECESSES OF THE PERITONEAL CAVITY Peritoneal folds may create fossae or recesses within the peritoneal cavity. These are of clinical interest because a length of intestine may enter one and be constricted by the fold at the entrance to the recess: it may subsequently become a site of internal herniation. The contents of the peritoneal fold may be important if surgical incision is required to reduce such a hernia. Although internal herniation
may occur into the lesser sac via the epiploic foramen, the sac is not usually considered to be a peritoneal recess. Duodenal recesses (Fig. 69.10)
Several folds of peritoneum may exist around the fourth part of the duodenum and the duodenojejunal junction forming several recesses. Superior duodenal recess
The superior duodenal recess is occasionally present, usually in association with an inferior duodenal recess. It lies to the left of the end of the fourth part of the duodenum, opposite the second lumbar vertebra, and behind a crescentic superior duodenal fold (duodenojejunal fold). The fold has a semilunar free lower edge which merges to the left with the peritoneum anterior to the left kidney. The inferior mesenteric vein is directly behind the junction of the left (lateral) end of this fold and the posterior parietal peritoneum. The recess varies in size but is commonly is c.2 cm deep, admitting a fingertip. It opens downwards, its orifice being in the angle formed by the left renal vein as it passes across the abdominal aorta. Inferior duodenal recess
The inferior duodenal recess is usually present often associated with a superior recess with which it may share an orifice. It lies to the left of the fourth part of the duodenum, opposite the third lumbar vertebra. It sits behind a non-vascular, triangular inferior duodenal fold (duodenomesocolic fold), which has a sharp upper edge. It is usually c.3 cm deep, admits one or two fingers and opens upwards towards the superior duodenal recess. It sometimes extends behind the fourth part of the duodenum and to the left, in front of the ascending branch of the left colic artery and the inferior mesenteric vein. Paraduodenal recess (Fig. 69.11)
The paraduodenal recess may occur in conjunction with superior and inferior duodenal recesses. It is rare in adults but is more commonly seen in newborn children. It lies a little to the left and slightly behind of the fourth part of the duodenum, behind a falciform paraduodenal fold. The free right edge of the fold contains the inferior mesenteric vein and ascending branch of the left colic artery, and represents part of the upper left colic mesentery. Its free edge lies in front of the wide orifice of the recess, which faces right. Retroduodenal recess
The retroduodenal recess is the largest of the duodenal recesses, but is rarely present. It lies behind the third and fourth parts of the duodenum in front of the abdominal aorta. It ascends nearly to the duodenojejunal junction, is 8-10 cm deep, and bounded on both sides by duodenoparietal folds. It has a wide orifice which faces down and to the left. Duodenojejunal recess
The duodenojejunal or mesocolica recess occurs in c.20% of adults. When present, it is almost never associated with any other duodenal recesses. It is c.3 cm deep and lies to the left of the abdominal aorta, between the duodenojejunal
junction and the root of the transverse mesocolon. It is bounded above by the pancreas, on the left by the kidney, and below by the left renal vein. It has a circular opening between two peritoneal folds, and faces down and to the right. Mesentericoparietal recess
Figure 69.10 The superior and inferior duodenal recesses. The transverse colon and jejunum have been displaced. (After Jonnesco, from Poirier P, Charpy A 1901 Traite d'Anatomie Humaine. Paris: Masson et Cie.)
Figure 69.11 The paraduodenal recess.
Figure 69.11 The paraduodenal recess.
The mesentericoparietal recess is only rarely present in adults. It lies just below the third part of the duodenum and invaginates into the upper part of the mesentery towards the right. Its orifice is large and faces left behind a fold of mesentery raised by the superior mesenteric artery. Caecal recesses (Fig. 69.12)
Several folds of peritoneum may exist around the caecum and form recesses. Paracaecal recesses are common sites for abscess formation following acute appendicitis. Superior ileocaecal recess
The superior ileocaecal recess is usually present and best developed in children. It is often reduced and absent in the aged, especially the obese. It is formed by the vascular fold of the caecum, which arches over the anterior caecal artery, supplying the anterior part of the ileocaecal junction, and its accompanying vein. It is a narrow slit bounded in front by the vascular fold, behind by the ileal mesentery, below by the terminal ileum and on the right by the ileocaecal junction. Its orifice opens downwards to the left. Inferior ileocaecal recess page 1137 page 1138
Figure 69.12 The peritoneal folds and recesses in the caecal region.
The inferior ileocaecal recess is well marked in youth but frequently obliterated by fat in adults. It is formed by the ileocaecal fold, which extends from the anteroinferior aspect of the terminal ileum to the front of the mesoappendix (or to the appendix or caecum). It is also known as the 'bloodless fold of Treves', although it sometimes contains blood vessels and will often bleed if divided during surgery. If inflamed, especially when the appendix and its mesentery are retrocaecal, it may be mistaken for the mesoappendix. The recess is bounded in front by the ileocaecal fold, above by the posterior ileal surface and its mesentery, to the right by the caecum, and behind by the upper mesoappendix. Its orifice opens downwards to the left. Retrocaecal recess
The retrocaecal recess lies behind the caecum. It varies in size and extent and ascends behind the ascending colon, often being large enough to admit an entire finger. It is bounded in front by the caecum (and sometimes the lower ascending colon), behind by the parietal peritoneum and on each side by caecal folds (parietocolic folds) passing from the caecum to the posterior abdominal wall. The vermiform appendix frequently occupies this recess when in the retrocaecal position. Intersigmoid recess
The intersigmoid recess is constant in fetal life and infancy, but may disappear during later development. It lies behind the apex of the V-shaped parietal attachment of the sigmoid mesocolon and is funnel shaped. It is directed upwards and opens downwards. It varies in size from a slight depression to a shallow fossa. Its posterior wall is formed by the parietal peritoneum of the posterior abdominal wall which covers the left ureter as it crosses the bifurcation of the left common iliac artery. Occasionally the recess is within the layers of the sigmoid mesocolon, and is nearer the bowel wall than the mesenteric root. It is probably produced by an imperfect blending of the mesocolon with the posterior parietal peritoneum.
© 2008 Elsevier
VASCULAR SUPPLY AND LYMPHATIC DRAINAGE Parietal and visceral peritoneum develop from the somatopleural and splanchnopleural layers respectively of lateral plate mesoderm (Chapter 108). Parietal peritoneum is therefore supplied by somatic blood vessels of the abdominal and pelvic walls. Its lymphatics join those in the body wall and drain to parietal lymph nodes. Visceral peritoneum is best considered as an integral part of the viscera which it overlies. It derives its blood supply from the viscera, and its lymphatics join the visceral vessels to drain to the regional lymph nodes.
© 2008 Elsevier
INNERVATION The parietal peritoneum is innervated by branches from nerves which supply the muscles and skin of the overlying body wall and thus has a similar spinal level of origin. The visceral peritoneum is innervated by branches of visceral afferent nerves which travel with the autonomic supply to the underlying viscera. The different sensations arising from pathologies which affect either the parietal or visceral peritoneum reflect these differences in patterns of innervation. Welllocalized pain is elicited by mechanical, thermal or chemical stimulation of the nocioceptors of the parietal peritoneum. The sensation is usually confined to one or two dermatomes for each area of peritoneum stimulated and is both lateralized and well-localized. Somatic nerves of the parietal peritoneum also supply the corresponding segmental areas of skin and muscles and, when the parietal peritoneum is irritated, muscles tend to contract by reflex, causing localized hypercontractility (guarding) or even rigidity of the abdominal wall. The parietal peritoneum on the underside of the diaphragm is supplied with afferent fibres from the phrenic nerves and peripherally by the lower six intercostal nerves and subcostal nerve. Peripheral irritation of the diaphragm may result in pain, tenderness and muscular rigidity in the distribution of the lower thoracic spinal nerves, while central irritation may result in pain in the cutaneous distribution of cervical spinal nerves III-V, i.e. the shoulder region. The visceral peritoneum and viscera are not affected by these stimuli since the visceral afferent innervation provides a much more limited sensation of discomfort. When stimulated, the sensation of pain is of a less severe nature and referred to the area of abdominal wall according to the region of the intestinal tract affected. Discomfort from foregut structures is felt in the region of the epigastrium, midgut structures in the region of the umbilicus, and hindgut structures in the suprapubic region: none of these sensations shows significant lateralization. However, stretch of the visceral peritoneum is a potent cause of certain sensations and responses. Various neural elements in the visceral walls, mesenteries and overlying peritoneum mediate poorly localized sensations of discomfort when stimulated by stretch, and may also elicit profound reflex autonomic reactions involving vasomotor and cardiac changes. This is of considerable clinical relevance. The effects of division of the parietal peritoneum may be rendered painless by local or regional local anaesthesia. In marked contrast, the direct central connections of the visceral afferents, particularly via the vagus, mean that stretching the visceral peritoneum may have profound effects, and may produce acute haemodynamic instability despite high spinal anaesthesia. Ischaemia of the underlying viscera causes poorly localized abdominal pain, probably due to the spasms of visceral smooth muscle. REFERENCES Healy JC, Reznek RH 1999 The anterior abdominal wall and peritoneum. In Butler P, Mitchell A, Ellis H (eds) Applied Radiological Anatomy. Cambridge: Cambridge University Press: 189-200. Demonstrates the imaging anatomy of the peritoneal spaces and reflections using cross-sectional imaging. Meyers M 1994 Dynamic Radiology of the Abdomen. Normal and Pathologic Anatomy. New York: Springer. Provides a systematic application of anatomic and dynamic principles to the understanding and diagnosis
of intraabdominal disease. Coakley FV, Hricak H 1999 Imaging of peritoneal and mesenteric disease: key concepts for the clinical radiologist. Clin Radiol 54: 563-574 Explains the complex anatomy of the upper abdominal peritoneal fold suspending the stomach, liver and spleen. Medline Similar articles Full article
© 2008 Elsevier
70 GASTROINTESTINAL TRACT General microstructure of the gut wall The gut wall displays a common structural plan which is modified regionally (p. 41) to take account of local functional differences. The general microstructure is best appreciated by reference to the development of the gut (Chapter 90). Much of the alimentary canal originates as a tube of endoderm enclosed in splanchnopleuric mesoderm. Its external surface faces the embryonic coelom, and the endodermal lining forms the epithelium of the canal and also the secretory and ductal cells of various glands which secrete into the lumen, including the pancreas and liver. The splanchnopleuric mesoderm forms the connective tissue, muscle layers, blood vessels and lymphatics of the wall, and its external surface becomes the visceral mesothelium or serosa (p. 41). There is no serosa surrounding the cervical and thoracic portions of the gut, or where the hindgut traverses the pelvic floor: in these sites the gut tube is surrounded by a connective tissue adventitia. Neural elements invade the gut from neural crest tissue (Chapter 14). The smooth muscle of the muscularis externa layers of the alimentary canal is supplemented with striated muscle both cranially (from the branchial arches) and caudally. An outline of the general microstructural organization of the gut wall is shown in Fig. 70.1.
© 2008 Elsevier
MATURE GUT WALL The gut wall has four main layers, namely mucosa, submucosa, muscularis externa and serosa (p. 41). The mucosa (mucous membrane) is the innermost layer and is subdivided into a lining epithelium, an underlying lamina propria (a layer of loose connective tissue, where many of the glands are also found) and a thin layer of smooth muscle, the muscularis mucosae. The submucosa is a strong and highly vascularized layer of connective tissue. The muscularis externa consists of inner circular and outer longitudinal layers which are present throughout the gut wall: a partial oblique layer is present only in the stomach. The external surface is bounded by a serosa or adventitia, depending on its position within the body.
MUCOSA Epithelium
The epithelium is a protective barrier and the site of secretion and absorption. Its protective function (against mechanical, thermal and chemical injury) is particularly evident in the oesophagus and in the terminal part of the rectum, where it is thick, stratified, and covered in mucus, which serves as a protective lubricant. Other than in these sites, the epithelium lining the gut wall is singlelayered, and either cuboidal (in glands) or columnar. It contains cells modified for absorption as well as various types of secretory cell. The barrier function and selectivity of absorption depend on tight junctions (p. 7) over the entire epithelium. The surface area of the lumen available for secretion or absorption is increased by the presence of mucosal folds, pits, crypts, villi and glands. Microvilli on the surfaces of individual absorptive cells amplify the area of apical plasma membrane in contact with the contents of the gut. Some glands lie in the lamina propria, some in the submucosa, and others (the liver and pancreas) are totally external to the wall of the gut. All of these glands drain into the lumen of the gut through individual ducts. The epithelium also contains scattered neuroendocrine (enteroendocrine) cells. Lamina propria
The lamina propria consists of compact connective tissue, often rich in elastin fibres, which supports the surface epithelium and provides nutrient vessels and lymphatics. Lymphoid follicles are present in many regions of the gut, most notably in Peyer's patches. Cells within the lamina propria are the source of growth factors which regulate cell turnover, differentiation and repair in the overlying epithelium. Muscularis mucosae
The muscularis mucosae is well developed in the oesophagus and in the large intestine, especially in the terminal part of the rectum. In addition, single muscle cells originating from the muscularis mucosae are found inside the villi or between the tubular glands of the stomach and large intestine. By its contraction, the muscularis mucosae can alter the surface configuration of the mucosa locally, allowing it to adapt to the shapes and mechanical forces imposed by the contents of the lumen, and in the intestinal villi, promoting vascular exchange and lymphatic drainage.
SUBMUCOSA The submucosa contains large bundles of collagen and is the strongest layer of the gut wall. However, it is also pliable and deformable and can therefore adjust to changes in the length and diameter of the gut. It contains the largest arterial network of the wall, which supplies both the mucosa and the muscle coat. The submucosa invades the folds which project into the lumen of the oesophagus and rectum, the rugae of the gastric wall, and the plicae circulares of the small intestine, but does not enter the villi.
MUSCULARIS EXTERNA The muscularis externa usually consists of distinct inner circular and outer longitudinal layers whose antagonistic activities create waves of peristalsis responsible for the movement of ingested material through the lumen of the gut.
In the stomach, where movements are more complex, there is a partial oblique layer, internal to the other two layers. The layer of circular muscle is invariably thicker than the longitudinal muscle, except in the colon, where the longitudinal muscle is gathered into three cords (taenia coli). The muscularis externa is composed almost exclusively of smooth muscle, except in the upper oesophagus, where smooth muscle blends with striated muscle. Although the oesophageal musculature resembles that of the pharynx, it is entirely under involuntary control. For most of its length, the smooth muscle of the gut wall consists of ill-defined bundles of cells, typically visceral in type, and somewhat larger than vascular smooth muscle cells. They are c.500 µm long, regardless of body size, and are electrically and mechanically coupled. Their fasciculi lack a perimysium, but have sharp boundaries. The arrangement of the musculature means that a segment of gut can change extensively in diameter (to virtual occlusion of the lumen) and also in length, although elongation is limited by the presence of mesenteries. The co-ordinated activity of the two muscle layers produces a characteristic motor behaviour which is mainly propulsive and directed anally (peristalsis), combined with a nonpropulsive motor activity which either mixes the luminal contents, as occurs in the stomach, or partitions them, as occurs at the pyloric sphincter. The muscle maintains constant volume, so that shortening of a segment of the gut wall is accompanied by an increase in muscle layer circumference. page 1139 page 1140
Figure 70.1 The general arrangement of the alimentary canal to show the layers of the gut wall at the levels indicated (highly diagrammatic). The transverse colon (above right) has been displaced downwards to reveal the duodenum.
Intestinal smooth muscle exhibits variable and changing degrees of contraction on which rhythmic (or phasic) contractions are superimposed. Slow waves of rhythmic electrical activity, driven by changes in membrane potential in pacemaker cells (interstitial cells), spread throughout the thickness of the circular and longitudinal smooth muscle coats. After spreading circumferentially, slow waves can move in either oral or anal directions, causing segmental contraction. The distances of propagation and the patterns of this spontaneous activity vary
between areas of the intestine. Neural regulation of slow and phasic contractions involves excitatory and inhibitory transmitters which are released from the myenteric plexus. This motor control is closely co-ordinated with mucosal absorption and secretion and is mediated via intrinsic nerves in the submucous plexus. The peristaltic reflex occurs during passage of luminal contents down the intestine. It involves ascending contraction and descending relaxation: the sensory limb is mediated by sensory neurones that respond to either mucosal stimulation (intrinsic primary afferents) or muscle stretch (extrinsic afferents). Interstitial cells page 1140 page 1141
Interstitial cells (of Cajal) are thought to act as pacemakers for the myogenic contraction of muscularis externa, establishing the rhythm of bowel contractions through their influence on electrical slow wave activity. They receive modulatory inputs from the enteric nervous system and from the extrinsic innervation of the gut. Interstitial cells are thin, flat, and branched. They resemble smooth muscle cells because they contain actin and myosin filaments, and are linked by gap junctions to typical smooth muscle cells. However, they are phenotypically distinct from muscle cells because their intermediate filaments are vimentin rather than desmin, (which is typical of muscle cells). They lie in close apposition to varicose nerve endings of at least two types; one contains small, round clear vesicles (50 nm diameter), the other contains flat, discoidal vesicles (70 nm diameter). The positions of the interstitial cell layers vary regionally. In general, they lie in the same layers as the enteric plexuses. They are scattered among the cells of the circular muscle layer in the oesophagus and stomach, lie between the inner and outer layers of circular smooth muscle in the small intestine, and colocalize with the myenteric plexus and the single layer of the submucosal plexus on the luminal side of the circular component of the muscularis externa, in the large intestine.
SEROSA AND ADVENTITIA There is a layer of connective tissue external to the muscularis externa. It is of variable thickness and in many places contains adipose tissue. Where the gut is covered by visceral peritoneum, the external layer is a serosa, which consists of a thin connective tissue layer and an external coat of mesothelium. Elsewhere the connective tissue blends with that of the surrounding fasciae and is referred to as an adventitia. Where the alimentary tract is retroperitoneal, the surface facing the abdominal cavity is covered by serosa, and the other parts are covered by adventitia.
VASCULAR PLEXUSES Vascular plexuses are present at various levels of the wall, especially in the submucosa and mucosa: they connect with plexuses of vessels which supply the surrounding tissues or those entering through the mesentery, and accompany the ducts of outlying glands.
INNERVATION The gut is densely innervated by the autonomic and enteric nervous systems, and is under extrinsic and intrinsic neuronal control. Neuronal cell bodies of the enteric nervous system lie between the circular and longitudinal components of the muscularis externa (myenteric plexus) and within the submucosa (submucosal plexus). They provide the intrinsic sensory and motor supply of the gut wall and connect with extrinsic sensory, motor and sensorimotor nerves of cranial or spinal origin. Extrinsic innervation
The extrinsic innervation is derived from neurones outside the gut, and contains functional components from the sympathetic, parasympathetic and visceral sensory divisions of the peripheral nervous system. Visceral sensory endings (p. 59) respond to excessive muscular contraction or distension: their cell bodies are situated in the nodose ganglion of the vagus nerve and in thoracic and lumbar spinal or dorsal root ganglia. The cell bodies of parasympathetic efferent axons lie in the vagal dorsal motor nucleus in the medulla oblongata. Sympathetic efferent fibres arise from the thoracic and lumbar spinal cord and relay in prevertebral
sympathetic ganglia (coeliac, mesenteric and pelvic). A subserosal plexus, which sometimes contains neuronal cell bodies, connects the extrinsic nerve fibres with the myenteric plexus and is particularly prominent near the mesentery. Fibres from this plexus run through the longitudinal muscle layer to reach the myenteric plexus. Intrinsic innervation (Fig. 70.2)
The intrinsic innervation of the gut wall is derived from neurones which are located entirely within the wall in intramural ganglionated plexuses (for more details see pp. 59, 238). The myenteric (Auerbach's) plexus is a network of fine bundles of axons and small ganglia which lies within the muscularis externa, between the circular and longitudinal layers. It is often associated with secondary and tertiary plexuses of nerve fibres which sometimes contain isolated neuronal cell bodies. There are two or more submucosal plexuses, the most superficial of which is Meissner's plexus. Non-ganglionated nerve plexuses lie at various levels in the wall, notably in the lamina propria (mucosal plexus); at the interface between the submucosa and muscularis externa; between the circular and longitudinal muscles (the nonganglionated part of the myenteric plexus); and within the serosa. An additional non-ganglionated plexus lies between the internal and external components of the circular muscle of the small intestine. All parts of the myenteric plexus are continuous not only with each other, but also with the nerve fibre bundles in the circular muscle. The latter are connected to the ganglionated and nonganglionated plexuses of the submucosa, and these in turn are connected with the mucosal plexus by fibres which pass through the muscularis mucosa. UPDATE Date Added: 14 August 2006 Abstract: Morphology and motor function of the gastrointestinal tract examined with endosonography Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? db=pubmed&cmd=Retrieve&dopt=AbstractPlus&list_uids=16718809&query_hl=24&itool=pubmed_docsum Morphology and motor function of the gastrointestinal tract examined with endosonography. Odegaard S, Nesje LB, Hoff DA, et al: World J Gastroenterol 12:2858-2863, 2006.
page 1141 page 1142
Figure 70.2 Wall of the small intestine with the mesh-like appearance of the myenteric plexus highlighed by selective neuronal staining. The ganglia are the elongated structures running vertically, comprised of ganglion neurones, joined by connecting strands of nerve fibres. The circular musculature, virtually unstained, runs vertically and the unstained longitudinal musculature runs transversely. (Courtesy of Professor G Gabella, University College, London.) (By kind permission from G Gabella, Department of Anatomy and Embryology, University College, London.)
© 2008 Elsevier
71 GASTROINTESTINAL TRACT Stomach and abdominal oesophagus The stomach is the widest part of the alimentary tract and lies between the oesophagus and the duodenum. It is situated in the upper abdomen, extending from the left upper quadrant downwards, forwards and to the right, lying in the left hypochondriac, epigastric and umbilical areas. It occupies a recess beneath the diaphragm and anterior abdominal wall that is bounded by the upper abdominal viscera on either side. Its mean capacity increases from c.30 ml at birth, to 1000 ml at puberty, to c.1500 ml in adults. The peritoneal surface of the stomach is interrupted by the attachments of the greater and lesser omenta, which define the greater and lesser curvatures separating two surfaces (Figs 71.1, 71.6).
© 2008 Elsevier
PARTS OF THE STOMACH The stomach is divided for descriptive purposes into the fundus, body, pyloric antrum and pylorus, by arbitrary lines drawn on its external surface. The internal appearance and microstructure of these regions varies to some degree. The fundus is dome shaped and projects above and to the left of the cardiac orifice to lie in contact with the left dome of the diaphragm. It lies above a line drawn horizontally from the incisura cardiaca to the greater curvature. The body extends from the fundus to the incisura angularis, which is a constant external notch at the lower end of the lesser curvature. A line drawn from the incisura angularis to an indentation on the greater curvature defines the lower boundary of the body. The pyloric antrum extends from this line to the sulcus intermedius. At this point, the stomach narrows to become the pyloric canal, which is usually only 1-2 cm in length and terminates at the pyloric orifice.
Figure 71.1 The parts of the stomach.
© 2008 Elsevier
GASTRIC RELATIONS GASTRIC CURVATURES Lesser curvature
The lesser curvature extends between the cardiac and pyloric orifices and forms the medial (posterior and superior) border of the stomach. It descends from the medial side of the oesophagus in front of the decussating fibres of the right crus of the diaphragm. It curves downwards and to the right and lies anterior to the superior border of the pancreas. It ends at the pylorus just to the right of the midline. In the most dependent part there is typically a notch, the incisura angularis, whose position and appearance vary with gastric distension. The lesser omentum is attached to the lesser curvature and contains the right and left gastric vessels. Greater curvature
The greater curvature is four or five times longer than the lesser. It starts from the incisura cardiaca formed between the lateral border of the abdominal oesophagus and the fundus of the stomach. It arches upwards, posterolaterally and to the left. Its highest convexity, the apex of the fundus, is approximately level with the left fifth intercostal space just below the left nipple in males, but varies with respiration. From this level it sweeps inferiorly and anteriorly, slightly convex to the left, almost as far as the tenth costal cartilage in the supine position, where it turns medially to end at the pylorus. There is frequently a groove, termed the sulcus intermedius, in the curvature close to the pyloric constriction. The start of the greater curvature is covered by peritoneum, which continues over the anterior surface of the stomach. Laterally the greater curvature gives attachment to the gastrosplenic ligament and beyond this to the greater omentum, which contains the gastroepiploic vessels. The gastrosplenic ligament and the greater omentum, together with the gastrophrenic and splenorenal ligaments, are continuous parts of the original dorsal mesogastrium. The names merely indicate regions of the same continuous sheet of peritoneum and associated connective tissue. Gastric volvulus page 1143 page 1144
Volvulus of the stomach is much less common than volvulus of either the sigmoid colon or caecum. Two types of gastric volvulus may occur. The first, organoaxial volvulus, occurs about a line of rotation running from below the cardiac orifice to the pylorus. The antrum, body and fundus rotate upwards, with the greater curvature coming to lie above the lesser curvature as the volvulus progresses. The second, mesenteroaxial volvulus, occurs about a line drawn 'across' the body of the stomach, usually just above the incisura angularis. This type of volvulus is perpendicular to the line of organoaxial volvulus. The distal body and antrum rotate anteriorly, superiorly and laterally whilst the upper body and fundus rotate posteriorly, medially and inferiorly. Although relatively mobile within the upper abdomen, the stomach is normally tethered to the oesophagus at the gastrooesophageal junction, to the duodenum at the pylorus, to the spleen by the gastrosplenic omentum, and to the liver by the lesser omentum. The attachment to the transverse colon via the gastrocolic omentum also restrains the stomach but is the most mobile of all. For either type of gastric volvulus to occur, it is necessary for some or all of these points of tethering to be loosened either by previous surgical division or by chronic lengthening and loosening of their connective tissue. Organoaxial volvulus is most common because the lesser omentum, gastrosplenic ligament and gastrocolic omentum are more likely to undergo chronic lengthening by traction than the other attachments of the stomach. Mesenteroaxial volvulus requires the gastro-oesophageal junction and
pylorus to be sufficiently mobile as to come into close approximation. These structures are firmly tethered and consequently this form of gastric volvulus is much less common. Despite the profuse gastric arterial supply, either type of volvulus may compromise the vascularity of the stomach.
GASTRIC SURFACES When the stomach is empty and contracted, the two surfaces tend to lie facing almost superiorly and inferiorly, but with increasing degrees of distension they come to face progressively more anteriorly and posteriorly. Anterior (superior) surface
The lateral part of the anterior surface is posterior to the left costal margin and in contact with the diaphragm, which separates it from the left pleura, the base of the left lung, the pericardium and the left sixth to ninth ribs (Fig. 71.2). It lies posterior to the costal attachments of the upper fibres of transversus abdominis, which separate it from the seventh to ninth costal cartilages. The upper and left part of this surface curves posterolaterally and is in contact with the gastric surface of the spleen. The right half of the anterior surface is related to the left and quadrate lobes of the liver and the anterior abdominal wall. When the stomach is empty, the transverse colon may lie adjacent to the anterior surface. The entire anterior (superior) surface is covered by peritoneum. Posterior (inferior) surface
Figure 71.2 Anterior relations of the stomach, viewed from behind.
The posterior surface lies anterior to the left crus and lower fibres of the diaphragm, the left inferior phrenic vessels, the left suprarenal gland, the superior pole of the left kidney, the splenic artery, the anterior pancreatic surface, the
splenic flexure of the colon and the upper layer of the transverse mesocolon (Fig. 71.3). Together these form the shallow stomach bed: they are separated from the stomach by the lesser sac (over which the stomach slides as it distends). The upper left part of the surface curves anterolaterally and lies in contact with the gastric surface of the spleen. The greater omentum and the transverse mesocolon separate the stomach from the duodenojejunal flexure and ileum. The posterior surface is covered by peritoneum, except near the cardiac orifice, where a small, triangular area contacts the left diaphragmaticcrus and sometimes the left suprarenal gland. The left gastric vessels reach the lesser curvature at the right extremity of this bare area in the left gastropancreatic fold. The gastrophrenic ligament passes from the lateral aspect of this bare area to the inferior surface of the diaphragm.
© 2008 Elsevier
GASTRIC ORIFICES CARDIAC ORIFICE AND GASTRO-OESOPHAGEAL JUNCTION The opening from the oesophagus into the stomach is the cardiac orifice (Fig. 71.4). It is typically situated to the left of the midline behind the seventh costal cartilage at the level of the eleventh thoracic vertebra. It is c.10 cm from the anterior abdominal wall and 40 cm from the incisor teeth. The short abdominal part of the oesophagus curves sharply to the left as it descends and is continuous with the cardiac orifice. The right side of the oesophagus is continuous with the lesser curvature, the left side with the greater curvature. There is no specific anatomical cardiac sphincter related to the orifice. Internally, the transition between oesophagus and stomach is difficult to define because mucosa of gastric fundal pattern extends a variable distance up into the abdominal oesophagus. It usually forms a 'zig-zag' squamo-columnar epithelial junction with the oesophageal epithelium above this Z line (p. 1152). This is often referred to as the gastro-oesophageal junction, for histological and endoscopic purposes. A sling of longitudinal gastric muscle forms a loop on the superior, left, side of the gastro-oesophageal junction between the oesophagus and the lesser curvature, and this is taken as the external boundary of this junction.
GASTRO-OESOPHAGEAL REFLUX page 1144 page 1145
Figure 71.3 Posterior relations of the stomach.
Figure 71.4 The valve-like structure formed by the angle of the wall at the cardiac orifice. (Provided by Donald E Low, Department of Surgery, Virginia Mason, Seattle, USA.)
Reflux of gastric contents into the abdominal and lower thoracic oesophagus as a result of transient relaxation of the lower oesophageal sphincter occurs as a normal event in most individuals for a small percentage of their daily life. It also occurs as a result of a weak lower oesophageal sphincter, or of hiatus hernia which disrupts the normal anatomical barriers (p. 1083). Several anatomical and physiological factors normally prevent gastro-oesophageal reflux. The folds of gastric mucosa present in the gastro-oesophageal junction, the mucosal rosette, contribute to the formation of a fluid-and gas-tight seal. They also help to ensure that even low levels of tone within the lower oesophageal wall muscles may occlude the lumen of the junction against low pressures of gastric gas. The angle of the cardiac orifice may help to form a type of 'flap valve' and the length of abdominal oesophagus is buttressed externally by pads of adipose connective tissue at and below the level of the diaphragmatic hiatus. However, the major anti-reflux mechanism is the tonic contractions of the lower oesophageal musculature, which forms an effective high pressure zone (HPZ) (p. 986). The specialized smooth muscle of the wall of the lower oesophagus and the encircling fibres of the crural diaphragm exert a radial pressure that can be measured by a sensing device as it is withdrawn from the stomach into the oesophagus (Paterson 2001). If reflux is to be prevented, this pressure must always exceed the difference between the pressures on either side of the junction, i.e. the difference between intra-abdominal pressure (transferred to the stomach, and augmented by any contraction of the stomach wall itself), and intrathoracic pressure (transferred to the oesophagus).
During expiration, pressure exerted by tonic contraction of the smooth muscle of the lower oesophagus is normally sufficient to oppose the gastro-oesophageal pressure gradient. During inspiration, intra-abdominal pressure rises and intrathoracic pressure becomes more negative, increasing the risk of reflux. This tendency is opposed by additional pressure exerted by contraction of the crural fibres of the diaphragm. (Activation of the crural diaphragm slightly before the costal diaphragm would ensure that contraction of peri-oesophageal fibres preceded the increase in gastro-oesophageal pressure gradient.) The anti-reflux barrier must of course be lowered for swallowing and vomiting. Swallowing is followed immediately by expiration, which relaxes the crural fibres and allows the oesophageal contents to be transferred to the stomach by peristaltic movement. Vomiting is produced by bursts of activity involving co-contraction of the diaphragm, intercostal and abdominal muscles in a pattern distinct from that of respiration: this activity is coordinated with relaxation of the crural fibres around the oesophagus (Miller, 1990). Barrett's oesophagus
The squamous epithelium lining the lower oesophagus may be pathologically replaced by a columnar, gastric type epithelium. This may occur as islands, strips, or circumferentially, and may extend for a variable length up the lower oesophagus. This process is most likely to be the result of the chronic reflux of gastric contents, acid or alkali, into the oesophagus with a resultant change in mucosal cell type. The abnormal columnar type epithelium present in the anatomical oesophagus is referred to as Barrett's epithelium.
PYLORIC ORIFICE page 1145 page 1146
The pyloric orifice is the opening into the duodenum. The circular pyloric constriction on the surface of the stomach usually indicates the location of the pyloric sphincter and is often marked by a prepyloric vein crossing the anterior surface vertically downwards. The pyloric orifice typically lies 1-2 cm to the right of the midline in the transpyloric plane with the body supine and the stomach empty. The pyloric sphincter is a muscular ring formed by a marked thickening of the circular gastric muscle interlaced with some longitudinal fibres.
© 2008 Elsevier
GASTRIC FORM AND INTERNAL APPEARANCES It is clear from contrast radiographic studies that the form and position of the stomach are extremely variable depending on posture, the volume of its contents, and the surrounding viscera. They are also influenced by the tone of the abdominal wall and gastric musculature and by the build of the individual. The empty stomach is most commonly J-shaped and, in the erect posture, the pylorus descends to the level of the second or the third lumbar vertebra. The lowest part of the antrum often lies below the level of the umbilicus. The fundus usually contains gas. The overall axis of the organ is, therefore, slightly inclined from the vertical (Figs 71.5, 71.6). In short, obese individuals the axis of the stomach lies more towards the horizontal as a 'steer-horn' shape. Variation caused by the contents of the stomach mainly affects the body because the pyloric part usually remains contracted during digestion. As the stomach fills, it expands forwards and downwards but, when the colon or small bowel is distended, the fundus enlarges towards the liver and diaphragm. As stomach capacity increases, the pylorus is displaced to the right and the axis of the whole organ lies in a more oblique direction (Figs 71.5, 71.6). In this position the anterior and posterior surfaces tend to face forwards and backwards and the lowest part is the pyloric antrum, which extends below the umbilicus. When intestinal distension interferes with downward expansion of the body, the stomach retains a horizontal position.
INTERNAL APPEARANCES
Figure 71.5 Axes of the empty and full stomach. As the stomach distends, the greater curvature 'rolls' downwards and the anterosuperior surface comes to lie almost completely vertical as the anterior surface.
During endoscopic examination (Fig. 71.7), the stomach is typically at least
partially distended by air. The cardiac orifice and the lowest portion of the abdominal oesophagus viewed from above are typically closed at rest by tonic contraction of the lower oesophageal musculature. The gastric mucosa lining the orifice is puckered into ridges. It is present for a short but variable distance into the abdominal oesophagus and the transition between columnar and squamous epithelium is usually clearly visible. The presence of abnormal columnar epithelium within the anatomical oesophagus is referred to as Barrett's oesophagus but the precise definition of this condition is difficult. From within the distended stomach, the cardiac orifice appears in the medial wall of the fundus and is asymmetrical. The medial edge of the cardiac orifice is continuous with the medial wall of the body of the stomach. The mucosa is slightly thickened at this point with a raised profile, forming part of the 'mucosal rosette' that lines the orifice. The 'rosette' aids closure of the cardiac orifice and helps prevent reflux of stomach contents into the oesophagus. The medial edge of the orifice is more clearly visible than the lateral edge as it forms a more acute angle with the mucosal lining of the abdominal oesophagus. In the partly distended stomach, the mucosa of the fundus is thrown into gentle folds with no particular pattern. As the stomach fills towards capacity, however, these folds rapidly become less pronounced, and the wall is nearly smooth when the stomach is over-inflated. The body of the stomach has the most pronounced mucosal folds. Even in moderate distension, they appear as long, broad mucosal ridges running in sinuous strips from fundus to pyloric antrum (Fig. 71.6). They are seen on all mucosal surfaces of the body but are most obvious on the anterolateral, lateral and posterolateral parts (which correspond to the inner surface of the anterior and posterior external surfaces and to the greater curvature). Here they are occasionally called the magenstrasse, a reference to their possible role in directing liquid entering the stomach immediately down into the pyloric antrum. These folds are least prominent on the medial surface (corresponding to the inner surface of the lesser curvature), which is much smoother, particularly when the stomach distends. The areae gastricae within the antrum are small nodular elevations of the mucosal surface that are readily seen on double contrast barium meal (Fig. 71.6). The few folds present in the antrum when the stomach is relaxed disappear with distension. The antrum adjacent to the pyloric canal, the prepyloric antrum, has a smooth mucosal surface culminating in a slight puckering of the mucosa at the pyloric orifice caused by the contraction of the pyloric sphincter.
GASTROSTOMY Since the lower body and antrum of the stomach is related to the posterior aspect of the left anterior abdominal wall, it may usefully be accessed to form a gastrostomy. Its mobility enables the anterior surface of the stomach to be readily approximated to the parietal peritoneum on the posterior surface of the abdominal wall and a communication to be established between the lumen of the stomach and the surface of the skin. Although this may be performed as a direct open surgical procedure under general anaesthetic it is much more commonly performed using a percutaneous puncture guided by either endoscopic visualization of the stomach or radiological imaging. The procedure is made easier
by the fact that the anterior surface of the stomach lies most nearly in the vertical plane when the stomach is distended. One of the main hazards of the procedure results from the occasional interposition of the transverse colon between the stomach and anterior abdominal wall. This may lead to inadvertent transfixion of the colon by the needle puncture system. The variable length of the transverse colonic mesentery means that it may sometimes lie adjacent to the anterior gastric surface when a subject is recumbent. These risks may be reduced by radiological guidance.
© 2008 Elsevier
VASCULAR SUPPLY AND LYMPHATIC DRAINAGE ARTERIES The arterial supply to the stomach comes predominantly from the coeliac axis although intramural anastomoses exist with vessels of other origins at the two ends of the stomach (Figs 71.8, 71.9). The left gastric artery arises directly from the coeliac axis. The splenic artery gives origin to the short gastric arteries as well as the left gastroepiploic artery and may occasionally give origin to a posterior gastric artery. The hepatic artery gives origin to the right gastric artery and the gastroduodenal artery, which in turn gives origin to the right gastroepiploic artery. UPDATE Date Added: 21 June 2005 Shanida Helena Nataraja, PhD (Dianthus Medical Limited) Abstract: Assessment of vascular anatomy around stomach before laparoscopyassisted gastrectomy. Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=15208129&query_hl=12 Assessment of vascular anatomy around stomach before laparoscopy-assisted gastrectomy. Left gastric artery
The left gastric artery is the smallest branch of the coeliac axis. It ascends to the left of the midline and crosses the left crus of the diaphragm beneath the peritoneum of the upper posterior wall of the lesser sac. Here it lies adjacent to the left inferior phrenic artery and medial or anterior to the left suprarenal gland. It runs forwards into the superior portion of the lesser omentum adjacent to the superior end of the lesser curvature. It turns anteroinferiorly to run along the lesser curvature between the two peritoneal leaves of the lesser omentum. At the highest point of its course, it gives off an oesophageal branch. In its course along the lesser curvature, it gives off multiple branches that run onto the anterior and posterior surfaces of the stomach and anastomose with the right gastric artery in the region of the incisura angularis. page 1146 page 1147
Figure 71.6 Double contrast barium meal. A, Initial stomach filling demonstrates a horizontally lying stomach with prominent gastric rugal folds. B, The area gastricae within the antrum are clearly identified on distension of the stomach. C, In the erect position the stomach has a more 'J'-shaped configuration.
The left gastric artery may arise from the common hepatic artery or its branches. The most common variant is an origin from the left hepatic artery, when the left gastric artery passes between the peritoneal layers of the superior lesser omentum to reach the lesser curvature of the stomach. Other variants include a common origin with the common hepatic artery. An aberrant left hepatic artery can occasionally arise from the left gastric artery: identification of an aberrant origin may be of importance during surgical mobilization of the upper stomach. UPDATE Date Added: 13 December 2005 Publication Services, Inc. Abstract: An anomalous case of the left gastric artery, the splenic artery, and
hepato-mesenteric trunk independently arising from the abdominal aorta. Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract_uids=16119612&query_hl=3 An anomalous case of the left gastric artery, the splenic artery, and hepatomesenteric trunk independently arising from the abdominal aorta. Saga T, Hirao T, Kitashima S et al. Kurume Med J 52(1-2):49-52, 2005.
Short gastric arteries
The short gastric arteries are variable in number, commonly between five and seven, and arise from the splenic artery, its divisions, or from the proximal left gastroepiploic artery. They pass between layers of the gastrosplenic ligament to supply the cardiac orifice and gastric fundus, and anastomose with branches of the left gastric and left gastroepiploic arteries. An accessory left gastric artery may arise with these vessels from the distal splenic artery. Left gastroepiploic artery
The left gastroepiploic artery arises from the splenic artery as its largest branch near the splenic hilum. It runs anteroinferiorly between the layers of the gastrosplenic ligament and into the upper gastrocolic omentum. It lies between the layers of peritoneum close to the greater curvature, running inferiorly to anastomose with the right gastro epiploic artery. It gives off gastric branches to the fundus of the stomach through the gastrosplenic ligament and to the body of the stomach through the gastrocolic omentum. These are necessarily longer than the gastric branches of the right gastroepiploic artery and may be 8-10 cm long. Epiploic (omental) branches arise along the course of the vessel and descend between the layers of the gastrocolic omentum into the greater omentum. A particularly large epiploic branch commonly originates close to the origin of the left gastroepiploic artery, descends in the lateral portion of the greater omentum and provides a large arterial supply to the lateral half of the omentum. Posterior gastric artery
Variant: A distinct posterior gastric artery may occur. When present, it arises from the splenic artery in its middle section posterior to the body of the stomach. It ascends behind the peritoneum of the lesser sac towards the fundus. It reaches the posterior surface of the stomach in the gastrophrenic fold. Right gastric artery page 1147 page 1148
Figure 71.7 Endoscopic appearance of the stomach: A, cardiac orifice from below; B, body greater curvature; C, body lesser curvature; D, pylorus.
The right gastric artery arises from the hepatic artery as it passes forwards from the posterior wall of the lesser sac into the lower border of the lesser omentum above the first part of the duodenum. The right gastric artery then runs between the peritoneal layers of the lesser omentum just above the medial end of the lesser curvature. It passes superiorly along the lesser curvature, giving off multiple branches onto the anterior and posterior surfaces of the stomach, and anastomoses with the left gastric artery. The origin of the right gastric artery is often variant. The most common alternative origins are from the common hepatic, left hepatic, gastroduodenal or supraduodenal arteries. Gastroduodenal artery
The gastroduodenal artery arises from the common hepatic artery posterior and superior to the first part of the duodenum. It gives origin to the right gastroepiploic and superior pancreaticoduodenal arteries at the lower border of the first part of the duodenum.
Figure 71.8 Arterial supply of the stomach.
Right gastroepiploic artery
The right gastroepiploic artery originates from the gastroduodenal artery behind the first part of the duodenum, anterior to the head of the pancreas. It passes inferiorly towards the midline between the layers of the gastrocolic omentum. It lies inferior to the pylorus and then runs laterally along the greater curvature. It ends by anastomosing with the left gastroepiploic artery. It is adjacent to the pylorus but, more distally, lies c.2 cm from the greater curvature of the stomach. Gastric branches ascend onto the anterior and posterior surfaces of the antrum and lower body of the stomach while epiploic branches descend into the greater omentum. It also contributes to the supply of the inferior aspect of the first part of the duodenum. Arterial anastomoses of the stomach
There is an anastomosis between the oesophageal arteries originating from the thoracic aorta and the vessels supplying the fundus in the region of the cardiac orifice. At the pyloric orifice the extensive network of vessels supplying the duodenum allows for some anastomosis between vessels of superior mesenteric artery origin and the pyloric vessels. The major named vessels supplying the stomach form extensive arterial anastomoses both on the serosal surface and around the curvatures. The right and left gastroepiploic arteries and the left and right gastric arteries anastomose freely with each other along the greater and lesser curvatures respectively. Anastomoses also form between the short gastric and left gastric arteries in the region of the fundus, and between the right gastric
and right gastroepiploic arteries in the region of the antrum. In addition to the extensive serosal anastomoses, networks form within the stomach wall at intramuscular, submucosal and mucosal levels. A true plexus of small arteries and arterioles is present within the submucosa: it supplies the mucosa and shows considerable regional variation both in the gastric wall and in the proximal duodenum. The rich arterial supply to the stomach ensures that the high mucosal blood flow required for physiological functioning is maintained even if one or more vessels become occluded. As a consequence, the stomach exhibits considerable resistance to ischaemia even when multiple arterial supplies are lost. page 1148 page 1149
Figure 71.9 The coeliac axis and its branches demonstrated on: A, digital subtraction angiogram demonstrating a replaced right hepatic artery arising from the origin of the superior mesenteric artery and being filled by a collateral from the left gastric artery. (A, by kind permission from Dr Adam Mitchell, Charing Cross Hospital London; B and C, by kind permission from GE Worldwide Medical Systems.)
The pyloric arteries are rami of the right gastric and right gastroepiploic arteries and pierce the duodenum distal to the sphincter around its entire circumference. They pass through the muscular layer to the submucosa where they divide into two or three rami, which turn back into the pyloric canal beneath the mucosa and run to the end of the pyloric antrum (Fig. 71.10). They supply the entire mucosa of the pyloric canal. Branches of these pyloric submucosal arteries may anastomose close to their origin with the duodenal submucosal arteries. Their terminal rami also anastomose with gastric arteries from the prepyloric antrum. The pyloric sphincter is supplied by the gastric and pyloric arteries via rami that leave their parent vessels in the subserosal and submucosal levels to penetrate the sphincter. Dieu la Foy lesions
Abnormalities of the intramural vascularity of the stomach are a rare cause of bleeding from the upper gastrointestinal tract. So-called 'Dieu la Foy' lesions commonly occur in the proximal body or fundus. When not actively bleeding, they
appear as small, raised, red dots marking the mucosal surface of the proximal body or fundus. They were originally thought to be small arteriovenous malformations of the submucosal plexus. It is now considered that such lesions are caused by a larger than normal penetrating arterial vessel running through the muscular coat of the stomach into the submucosa before branching into the submucosal plexus. Although not a pathological abnormality, the vessel has a greater than normal calibre for arteries at this level. The pulsatile flow, combined with its proximity to the overlying mucosa, may then lead to focal ulceration and rupture of the vessel following minor trauma, leading to profuse intraluminal bleeding.
VEINS The stomach veins drain ultimately into the portal vein. A rich submucosal and intramural network of veins gives rise to veins that usually accompany the corresponding named arteries. They drain either into the splenic or superior mesenteric veins although some pass directly into the portal vein. Short gastric veins
Four or five short gastric veins drain the gastric fundus and the upper part of the greater curvature. They drain into the splenic vein or one of its large tributaries. Left gastroepiploic vein
Figure 71.10 Blood supply of the stomach and the proximal duodenum. A scheme of arterial arrangements at the gastroduodenal junction. Dotted lines indicate sites where the submucous plexus may be non-continuous. Shaded areas represent the muscular layer of the visceral wall. (Redrawn courtesy of C Piasecki, Department of Anatomy, Royal Free Hospital School of Medicine, London and the Journal of Anatomy.)
The left gastroepiploic vein drains both anterior and posterior gastric surfaces and
the adjacent greater omentum. It runs superolaterally along the greater curvature, between the layers of the gastrocolic omentum. It receives multiple tributaries from the anterior and posterior surfaces of the body of the stomach and the greater omentum, and drains into the splenic vein within the gastrosplenic ligament. Right gastroepiploic vein page 1149 page 1150
The right gastroepiploic vein drains the greater omentum, distal body and antrum of the stomach. It passes medially, inferior to the greater curvature, in the upper portion of the gastrocolic omentum. Just proximal to the pyloric constriction it passes posteriorly to drain into the superior mesenteric vein below the neck of the pancreas. It may receive the superior pancreaticoduodenal vein close to its entry into the superior mesenteric vein. Left gastric vein
The left gastric vein drains the upper body and fundus of the stomach. It ascends along the lesser curvature to the oesophageal opening where it receives several lower oesophageal veins. It then curves posteriorly and medially behind the posterior peritoneal surface of the lesser sac. It drains into the portal vein directly at the level of the upper border of the first part of the duodenum. Right gastric vein
The right gastric vein is typically small and runs along the medial end of the lesser curvature. It passes under the peritoneum as it is reflected from the posterior aspect of the pylorus and first part of the duodenum onto the posterior wall of the lesser sac. It drains directly into the portal vein at the level of the first part of the duodenum. It receives the prepyloric vein as it ascends anterior to the pylorus at the level of the pyloric opening. Posterior gastric veins
Distinct posterior gastric veins may occur. When present, they accompany the posterior gastric artery from the middle of the posterior surface of the stomach. They drain into the splenic vein and may occur as multiple small vessels. Gastric varices
Variceal dilatation of the submucosal veins of the stomach may occur in the presence of portal hypertension. The anastomosis between portal and systemic venous circulations occurs around the lower oesophagus and upper stomach. Submucosal veins close to the cardiac orifice may become involved in the pathological flow of blood from the stomach and other upper abdominal viscera into the oesophageal veins. Gastric varices present less commonly in clinical practice than oesophageal varices. Occasionally gastric varices exist without the presence of oesophageal varices. In these circumstances, it may be that the effective 'point of meeting' between portal and systemic venous systems is lower than usual and occurs in the upper stomach rather than the lower oesophagus.
LYMPHATIC DRAINAGE
The stomach has a rich network of lymphatics that connect with lymphatics draining the other visceral organs of the upper abdomen. At the gastrooesophageal junction the lymphatics are continuous with those draining the lower oesophagus. In the region of the pylorus they are continuous with those draining the duodenum. In the main, they follow the course of the arteries supplying the stomach, however many separate node groups are now recognized (Fig. 71.11). The relationship of separate node groups to the regions of the stomach and the vascular territories supplied is of great importance during resection of the stomach, particularly for malignancy. Pancreatic and hepatic lymphatics play a considerable role in draining areas of the stomach during disease.
© 2008 Elsevier
INNERVATION The stomach is innervated by sympathetic and parasympathetic fibres. The sympathetic supply originates from the fifth to twelfth thoracic spinal segments and is mainly distributed to the stomach via the greater and lesser splanchnic nerves and the coeliac plexus. Periarterial plexuses form along the arteries and supply the stomach from the coeliac axis. Additional innervation comes from fibres of the hepatic plexus, which pass to the upper body and fundus via the upper limit of the lesser omentum. Some innervation is also provided via direct branches from the greater splanchnic nerves. The parasympathetic supply is from the vagus nerves (Fig. 71.12). Usually one or two rami branch on the anterior and posterior aspects of the gastro-oesophageal junction. The anterior nerves are mostly from the left vagus and the posterior from the right vagus, both emerging from the oesophageal plexus. The anterior nerves supply filaments to the cardiac orifice and divide near the oesophageal end of the lesser curvature into gastric, pyloric and hepatic branches. Gastric branches (between four and ten) radiate on the anterior surface of the body and fundus. The greater anterior gastric nerve is the major gastric branch and lies in the lesser omentum near the lesser curvature. Pyloric branches (generally two) originate below the cardiac orifice. The smaller of the two nerves runs between the peritoneal layers of the lesser omentum almost horizontally towards its free edge and turns down on the left side of the hepatic artery to reach the pylorus. The larger nerve usually arises from the greater anterior gastric nerve during its course over the anterior surface of the stomach and runs inferomedially to the pyloric antrum. Hepatic branches (one or two) originate from the pyloric branches and run superiorly to contribute to the hepatic plexus. The posterior nerves produce two main groups of branches, gastric and coeliac. Gastric branches originate behind the cardiac orifice and upper body of the stomach. They radiate over the posterior surface of the body and fundus and extend to the antrum but do not reach the pyloric sphincter. The largest is termed the greater posterior gastric nerve and runs posteriorly along the lesser curvature, giving branches to the coeliac plexus. Coeliac branches are often larger than the gastric branches. They run beneath the peritoneum, deep to the posterior wall of the lesser sac, at the upper limit of the lesser omentum to reach the coeliac plexus. Hepatic branches (one or two) are often small and originate from the coeliac branches. No true plexus occurs on either the anterior or posterior gastric surfaces, but plexuses are present in the submucosa and between the layers of the muscularis externa. The gastric sympathetic nerves are vasoconstrictor to the gastric vasculature and inhibitory to gastric musculature. The sympathetic supply to the pylorus is motor, and brings about pyloric constriction. The sympathetic supply also conducts afferent impulses that mediate sensations, including pain. The parasympathetic gastric supply is secretomotor to the gastric mucosa and motor to the gastric musculature. It is also responsible for coordinated relaxation of the pyloric sphincter during gastric emptying. Coeliac plexus
The coeliac plexus is the largest major autonomic plexus, sited at the level of the twelfth thoracic and first lumbar vertebrae. It is a dense network uniting two large coeliac ganglia and surrounds the coeliac artery and the root of the superior mesenteric artery (Fig. 71.13). It is posterior to the stomach and lesser sac, anterior to the crura of the diaphragm and the commencement of the abdominal aorta, and lies between the suprarenal glands. The plexus and ganglia are joined by greater and lesser splanchnic nerves and branches from the vagus and phrenic nerves. The plexus extends as numerous secondary plexuses along adjacent arteries. The coeliac ganglia are irregular masses on each side of the coeliac trunk adjacent to the suprarenal glands. They lie anterior to the crura of the diaphragm. The right ganglion is posterior to the inferior vena cava, the left ganglion posterior to the origin of the splenic artery. The ipsilateral greater splanchnic nerve joins the upper part of each ganglion. The lower part of each ganglion forms a distinct subdivision usually termed the aorticorenal ganglion. This receives the ipsilateral lesser splanchnic nerve and gives origin to the majority of the renal plexus. It most commonly lies anterior to the origin of the renal artery. The coeliac plexus is connected to or gives rise to the phrenic, splenic, hepatic, superior mesenteric, suprarenal, renal and gonadal plexuses.
Phrenic plexus
The phrenic plexus lies around the inferior phrenic arteries on the crura of the diaphragm. It arises as a superior extension of the coeliac ganglion and often receives one or two sensory branches from the phrenic nerve. The left phrenic plexus is usually larger than the right. On the left it supplies branches to the left suprarenal gland and the cardiac orifice of the stomach. The right phrenic plexus joins the phrenic nerve, forming a small phrenic ganglion. This distributes branches to the inferior vena cava, suprarenal gland and hepatic plexus.
REFERRED PAIN page 1150 page 1151
page 1151 page 1152
Figure 71.11 Lymph node stations of A, the stomach and B, upper abdominal viscera.
Figure 71.12 Distribution of the vagal nerves to the stomach.
The majority of the sensation of pain arising from the stomach is poorly localized. In common with other structures of foregut origin, it is referred to the central epigastrium. Pain arising from the region of the gastro-oesophageal junction may involve innervation from the oesophagus and is commonly referred to the lower retrosternal and subxiphoid areas.
© 2008 Elsevier
MICROSTRUCTURE The gastric wall consists of the major layers found elsewhere in the gut, i.e. mucosa, submucosa, muscularis externa and serosa, together with gastric vessels and nerves (Figs 71.14, 71.15). The microstructure reflects the functions of the stomach as an expandable muscular sac lined by secretory epithelium, although there are local structural and functional variations in this pattern.
Mucosa
Figure 71.13 Distribution of the upper abdominal autonomic plexuses.
The mucosa is a thick layer with a soft, smooth surface that is mostly reddish brown in life but pink in the pyloric region. In the contracted stomach the mucosa is folded into numerous folds or rugae, most of which are longitudinal. They are most marked towards the pyloric end and along the greater curvature (Fig. 71.7). The rugae represent large folds in the submucosal connective tissue (see below) rather than variations in the thickness of the mucosa covering them, and they are obliterated when the stomach is distended. As elsewhere in the gut, the mucosa is composed of a surface epithelium, lamina propria and muscularis mucosae.
EPITHELIUM When viewed microscopically at low magnification, the internal surface of the stomach wall (Fig. 71.15) appears honeycombed by small, irregular gastric pits c.0.2 mm in diameter. The base of each gastric pit receives several long, tubular gastric glands that extend deep into the lamina propria as far as the muscularis mucosae. Simple columnar mucus-secreting epithelium covers the entire luminal surface including the gastric pits, and is composed of a continuous layer of surface mucous cells which release gastric mucus from their apical surfaces to form a thick protective, lubricant layer over the gastric lining. This epithelium commences abruptly at the cardiac orifice, where there is a sudden transition from oesophageal stratified epithelium. Gastric glands
Although all gastric glands are tubular, they vary in form and cellular composition in different parts of the stomach. They can be divided into three groups-the cardiac, principal (in the body and fundus) and pyloric glands. The most highly specialized are the principal glands. Principal gastric glands
The principal glands are found in the body and fundus, three to seven opening into each gastric pit (Figs 71.14, 71.15, 71.16). Their junction with the base of the pit is termed the isthmus of the gland and immediately basal to this is the neck, the remainder being the base. In the walls of the gland are at least five distinct cell types: chief, parietal, mucous neck, stem and neuroendocrine. page 1152 page 1153
Figure 71.14 Diagram showing the principal regions of the interior of the stomach and the microstructure of tissues and cells within its wall. Undifferentiated, dividing cells are shown in white.
The chief (peptic) cells (Figs 71.14, 71.16) are the source of the digestive enzymes pepsin and lipase. They are usually basal in position, their shape is cuboidal and their nuclei rounded and euchromatic. They contain secretory zymogen granules and because of the abundant cytoplasmic RNA they are strongly basophilic. The parietal (oxyntic) cells are the source of gastric acid and of intrinsic factor, a glycoprotein necessary for the absorption of vitamin B 12 . They are large, oval and strongly eosinophilic, with centrally placed nuclei. They are mainly situated in the more apical half of the gland, reaching as far as the isthmus. They occur only at intervals along the walls, and bulge laterally into the surrounding connective tissue. Parietal cells have a unique ultrastructure related to their ability to secrete hydrochloric acid. The luminal side of the cell is deeply invaginated to form a series of blind-ended channels (canaliculi) bearing numerous irregular microvilli. Within the cytoplasm facing these channels are numerous fine membranous tubules (the tubulo-vesicular system) directed towards the canalicular surface. Abundant mitochondria are interspersed among these organelles. The plasma membrane covering the microvilli has a high concentration of H+ /K + ATPase antiporter channels that actively secrete hydrogen ions into the lumen, chloride ions following along the electrochemical gradient. The precise structure of the cell varies with its secretory phase: when stimulated, the numbers and surface areas of the microvilli increase up to fivefold, thought to be by the rapid fusion of the tubulo-vesicular system with the plasma membrane. At the end of stimulated secretion, this process is reversed, the excess membrane retreats back into the tubulo-alveolar system and microvilli are lost. Mucous neck cells are numerous at the necks of the glands and are scattered along the walls of the more basal regions. They are typical mucus-secreting cells, with apical secretory vesicles containing mucins, and basally displaced nuclei. However, their products are distinct histochemically from those of the superficial mucous cells. page 1153 page 1154
Figure 71.15 Low power micrograph showing the stomach wall, in the region of a longitudinal fold or ruga, visible macroscopically. The surface epithelium is infolded microscopically to form gastric pits, into the bases of which open gastric glands extending through the thickness of the mucosal lamina propria. A muscularis mucosae layer and submucosa follow the contours of the ruga and part of the external muscularis layers is seen below. (By permission from Kierszenbaum AL 2002 Histology and Cell Biology. St Louis: Mosby.)
Stem cells are relatively undifferentiated mitotic cells from which the other types of gland cell are derived. They are relatively few in number, and are situated in the isthmus region of the gland and bases of the gastric pits. These cells are columnar, with a few short apical microvilli. They periodically undergo mitosis, the cells they produce migrating apically to differentiate into new surface mucous cells, or basally to form mucous neck, parietal and chief cells, and also the
neuroendocrine cells. All of these cells have a limited lifespan, especially the mucus-secreting types, and are constantly replaced. The replacement period for surface mucous cells is c.3 days; mucous neck cells are replaced after c.1 week. Other cell types appear to live much longer. Neuroendocrine (enteroendocrine) cells occur in all types of gastric gland but more frequently in the body and fundus. They are situated mainly in the deeper parts of the glands, among the chief cells. They are basally situated, pleomorphic cells with irregular nuclei surrounded by granular cytoplasm containing clusters of large (0.3µm) secretory granules. These cells synthesize a number of biogenic amines and polypeptides important in the control of motility and glandular secretion. In the stomach they include cells designated as G cells secreting gastrin, D cells (somatostatin), and ECL (enterochromaffin-like) cells (histamine). They form part of the system of dispersed neuroendocrine cells. Cardiac glands
Cardiac glands are confined to a small area near the cardiac orifice (Fig. 71.17); some are simple tubular glands, others are compound branched tubular. Mucussecreting cells predominate and parietal and chief cells, although present, are few. Pyloric glands
page 1154 page 1155
Figure 71.16 A, Micrograph showing gastric glands in the fundic region, opening into the bases of gastric pits lined by mucous cells similar to those covering the epithelial surface. Eosinophilic parietal cells and basophilic chief cells line the glands, shown at higher magnification in B. B, Higher power micrograph showing eosinophilic parietal cells (short arrows) and basophilic chief cells (long arrows) lining the gastric glands (GG); these open into gastric pits (GP), which are invaginations of the mucussecreting surface epithelium. (Photograph by Sarah-Jane Smith.)
Figure 71.17 Micrograph showing the junction between the stratified squamous nonkeratinizing epithelium of the oesophagus (left) and the stomach, with cardiac glands. A lymphoid follicle (an example of MALT) is seen in the submucosa of the junction zone (bottom left). (Photograph by Sarah-Jane Smith.)
Pyloric glands enter as groups of two or three short convoluted tubes into the bases of the deep gastric pits of the pyloric antrum: the pits occupy about twothirds of the mucosal depth (Fig. 71.18). Pyloric glands are mostly populated with mucus-secreting cells, parietal cells are few and chief cells scarce. In contrast, neuroendocrine cells are numerous, especially G cells, which secrete gastrin when activated by appropriate mechanical stimulation (causing increased gastric motility and secretion of gastric juices). Although parietal cells are infrequent in pyloric glands, they are always present in fetal and postnatal tissue. In adults they may appear in the duodenal mucosa proximally near the pylorus.
LAMINA PROPRIA The lamina propria forms a connective tissue framework between the glands and contains lymphoid tissue that collects in small masses, gastric lymphatic follicles, which resemble solitary intestinal follicles (especially in early life). The lamina propria also contains a complex periglandular vascular plexus, which is thought to be important in the maintenance of the mucosal environment, including the removal of bicarbonate produced in the tissues as a counterpart to acid secretion. Neural plexuses are present and contain sensory and motor terminals.
MUSCULARIS MUCOSAE The muscularis mucosae is a thin layer of smooth muscle fibres lying external to the layer of glands. Its fibres are arranged as inner circular and outer longitudinal layers, and there is also a discontinuous external circular layer. The inner layer sends strands of smooth muscle cells between the glands, and their contraction probably assists in emptying into the gastric pits.
Submucosa The submucosa is a variable layer of loose connective tissue containing thick collagen bundles, numerous elastin fibres, blood vessels and nervous plexuses, including the ganglionated submucosal (Meissner's) plexus of the stomach.
Figure 71.18 Micrograph showing the pyloric region of the stomach with pyloric glands, stained with the periodic acid-Schiff (PAS) technique to show mucin (magenta) in the gastric pits and glands. Pale staining cells are the larger parietal cells (P) and smaller enteroendocrine cells (E). (By permission from Dr JB Kerr,
Monash University, from Kerr JB 1999 Atlas of Functional Histology. London: Mosby.)
Muscularis externa The muscularis externa is a thick muscle coat immediately under the serosa, with which it is closely connected by subserous loose connective tissue. From innermost outwards it has oblique, circular and longitudinal layers of smooth muscle fibres, although the separation between layers may be indistinct in places. The circular layer is poorly developed in the oesophageal region but is thickened at the distal pyloric antrum to form the annular pyloric sphincter. The outer longitudinal layer is most pronounced in the upper two-thirds of the stomach and the inner oblique layer in the lower half. The actions of the muscularis externa of the stomach produce a churning movement that mixes food with the gastric secretions. When the muscles contract, they reduce the volume of the stomach and throw the mucosa into longitudinal folds or rugae (see above). These flatten as the stomach distends with food and the musculature relaxes and thins. Muscle activity is controlled by a network of unmyelinated autonomic nerve fibres and their ganglia, lying between the muscle layers in the myenteric (Auerbach's) plexus.
Serosa or visceral peritoneum The serosa is an extension of the visceral peritoneum and covers the entire surface except along the greater and lesser curvatures at the attachment of the greater and lesser omenta, where the peritoneal layers leave space for vessels and nerves. It is also absent from a small posteroinferior area near the cardiac orifice where the stomach contacts the diaphragm at the reflections of the gastrophrenic and left gastropancreatic folds. REFERENCES DiDio LJ, Anderson MC 1968 The 'Sphincters' of the Digestive System. Baltimore: Williams and Wilkins. Japanese Research Society for Gastric Cancer 1998 Japanese Classification of Gastric Carcinoma. Tokyo: Kanehara & Co Ltd and Gastric Cancer 1: 10-24, 25-30. A detailed, widely accepted, description of the lymph node fields of the upper abdominal viscera, particularly in relation to malignancy. Silverstein FE, Tytgat GNJ 1991 Atlas of Gastrointestinal Endoscopy. 2nd edition. New York: Gower Medical Publishing. page 1155 page 1156
© 2008 Elsevier
72 GASTROINTESTINAL TRACT - SMALL INTESTINE Microstructure of the small intestine The intestinal wall is composed of mucosa, submucosa, muscularis externa and serosa or adventitia (Figs 70.1, 72.1 to 72.7). The mucosa is thick and very vascular in the proximal small intestine, but thinner and less vascular distally. In places it is ridged by the underlying submucosa to form circular folds, and mucosal finger-or leaf-like intestinal villi cover the whole surface. There are numerous simple tubular intestinal glands or crypts between the bases of the villi, and additional submucosal glands in the duodenum.
© 2008 Elsevier
CIRCULAR FOLDS Large, crescentic folds of mucosa (Figs 72.2, 72.3) project into the intestinal lumen transversely or slightly obliquely to the long axis. Unlike gastric folds they are not obliterated by distension of the intestine. Most extend round half or twothirds of the luminal circumference: some are complete circles; some bifurcate and join adjacent folds; some are spiral but extend only once (occasionally two or three times) round the lumen. Larger folds are c.8 mm deep at their broadest, but most are smaller than this, and larger folds often alternate with smaller ones. Folds begin to appear c.2.5-5 cm beyond the pylorus. Distal to the major duodenal papilla they are large and close together, as they also are in the proximal half of the jejunum. From here to midway along the ileum they diminish, and they disappear almost completely in the distal ileum, which accounts for the thinness of this part of the intestinal wall. The circular folds slow the passage of the intestinal contents and increase the absorptive surface. They are visible in radiographs after a barium meal.
Figure 72.1 Low-power micrograph showing the wall of the duodenum, with villi projecting into the lumen; intestinal crypts (of Lieberkühn) in the mucosa, seen mainly in transverse section; well-defined muscularis mucosae; submucosal seromucous
(Brunner's) glands and muscularis externa. (Photograph by Sarah-Jane Smith.)
© 2008 Elsevier
INTESTINAL VILLI (Figs 70.1, 72.4, 72.5)
Figure 72.2 Internal aspect of a representative sample of the proximal jejunum, showing circular folds.
page 1157 page 1158
Figure 72.3 Low-power micrograph showing several circular folds in the wall of the ileum. The folds are covered with villi projecting into the lumen and the submucosa extends into the core of each fold. Circular (innermost) and longitudinal smooth muscle layers form the underlying muscularis externa. Large masses of lymphoid tissue (Peyer's patches) occupy the mucosa at the left of the field. (By permission from Young B, Heath JW 2000 Wheater's Functional Histology. Edinburgh: Churchill Livingstone.)
Figure 72.4 A three-dimensional reconstruction of the architecture of an intestinal villus, indicating the underlying layers of the intestinal wall. Also shown are arteries and arterioles (red), veins and venules (blue), central lacteals and other lymphatic channels (yellow), lymphoid follicles (purple), neural elements (green), smooth muscle fibres (pink), and fibroblasts (white). Note the orifices of the intestinal crypts (of Lieberkühn). Types of cells in the epithelium include absorptive cells, goblet cells and neuroendocrine cells. Arrows indicate the direction of cell migration. The various layers are not drawn to scale.
Intestinal villi are highly vascular projections of the mucosal surface, just visible to the naked eye. They cover the entire intestinal mucosa, increase the surface area of the lumen about eight-fold, and give it a velvety texture. Villi are large and numerous in the duodenum and jejunum, and smaller and fewer in the ileum. In the first part of the duodenum they appear as broad ridges, become tall and foliate in the distal duodenum and proximal jejunum, and then gradually shorten to a finger-like form in the distal jejunum and ileum. Villi vary in density from 10 to 40 per square millimetre and from c.0.5 to 1.0 mm in height. UPDATE Date Added: 28 June 2005 Abstract: Direct visualization of intestinal villi by high-resolution magnifying upper endoscopy. Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=15557949&query_hl=6 Direct visualization of intestinal villi by high-resolution magnifying upper endoscopy.
© 2008 Elsevier
MUCOSA The mucosa (Figs 70.1, 72.5) consists of epithelium, lamina propria and muscularis mucosae.
EPITHELIUM (Fig. 72.5) A single-layered epithelium covers the intestinal villi (Fig. 72.5), and also lines the intestinal glands (crypts) that open between the bases of villi. Two types of cell, enterocytes and goblet cells, cover the surfaces of the villi. Microfold cells (M cells) are restricted to the dome epithelium covering localized accumulations of lymphoid tissues. These cell types are all in contact basally with a basal lamina to which they adhere, and all derive from a common stem cell in the intestinal crypts. Enterocytes page 1158 page 1159
Figure 72.5 High-power micrograph of an intestinal villus (toluidine blue-stained resin section) showing absorptive enterocytes (A), bearing microvilli (MV) covering its surface, interspersed with goblet cells (G) filled with pale mucinogen granules. Enteroendocrine cells (E) and intraepithelial lymphocytes (L) are also seen. The
Enteroendocrine cells (E) and intraepithelial lymphocytes (L) are also seen. The central core is lamina propria containing blood vessels (BV), a large lacteal lymphatic vessel (LV) and other connective tissue elements; smooth muscle cells are also present. (By permission from Dr JB Kerr, Monash University, from Kerr JB 1999 Atlas of Functional Histology. London: Mosby.)
Enterocytes are columnar absorptive cells, c.20 µm tall (Fig. 72.7). They are the most numerous type of cell in the intestinal lining, and are responsible for nutrient absorption. Their surfaces bear up to 3000 microvilli, which greatly increase the surface area for absorption. Collectively, microvilli are visible by light microscall rav1as a brush (striated) border c.1 µm thick: individual microvilli can be resolved only by electron microscopy. Enterocyte nuclei are elongated vertically, mainly euchromatic and located just below the centre of the cell. The cells have a lifespan after differentiation of about 5 days, and their position on the villus wall reflects their stage in the life cycle: at the tips of intestinal villi they undergo programmed, apoptotic, cell death and are shed from the epithelium. They are replaced at the base of the villus by stem cell mitosis. The apical cell surface is resistant to protease attack because microvilli possess a specialized glycoprotein-rich surface coat (glycocalyx) which, with an overlying layer of mucus, protects the epithelium against pancreatic enzymes in the intestinal lumen. The cell coat also contains a number of digestive enzymes as integral membrane proteins. These include enzymes that degrade disaccharides and oligopeptides prior to absorption. Further details of the structure of microvilli are given on page 20. The luminal surface is an important barrier to diffusion. Nutrients generally have to pass through enterocytes (transcellular absorption) before they reach the underlying lamina propria and its blood vessels and lymphatics (lacteals). Classical epithelial junctional complexes (p. 7) encircle the apical plasma membranes of adjacent enterocytes, and their tight junctions form an effective barrier to non-selective diffusion between the gut lumen and the body as a whole. The lateral plasma membranes of enterocytes are highly folded, interdigitating with each other to form complex intercellular boundaries, anchored periodically by desmosomes, and making contact at gap junctions. The lateral intercellular space expands during active absorption and is an additional conduit (supplementing transport across the basal cell surface) for the passage of fluids, nutrients and other solutes to the vessels of the lamina propria.
Figure 72.6 Part of a transverse section of the ileum showing Paneth cells containing orange-stained zymogen granules (containing defensins, including lysozyme) at the base of an intestinal gland. Undifferentiated epithelial cells are also present.
Goblet cells
Goblet cells are most numerous in the distal small intestine, increasing in number from the duodenum to their highest density in the terminal ileum. They have elongated, basal nuclei, an apical region containing many membrane-bound mucinogen granules (Fig. 72.7), and apical surfaces that bear a few short microvilli. Goblet cell mucins contribute to protection against microorganisms and toxins in the gut lumen, and also provide lubrication and mechanical protection from the intestinal contents. Microfold (M) cells
Microfold cells are present where the epithelium covers lymphoid aggregates (MALT; p. 77) in the intestinal wall. They are cuboidal or flattened in shape and have long, widely spaced microfolds rather than microvilli on their apical surfaces. They sample luminal antigens by endocytosis and transport antigen to lymphocytes that occupy intercellular pockets formed by deep invaginations of the M-cell basolateral plasma membranes. See page 80 for details of antigen processing and presentation. Lymphocytes
Intraepithelial lymphocytes are found in close association with M cells and also between the basolateral regions of enterocytes and goblet cells. They are
migratory cells derived from the underlying lymphoid tissue and constitute an important means of immune defence. Intestinal glands or crypts
Intestinal glands or crypts (of Lieberkühn) are tubular pits that open into the lumen throughout the intestinal mucosa via small circular apertures between the bases of the villi (Figs 70.1, 72.1, 72.4). Their thin walls are composed of columnar enterocytes supplemented by mucous cells, Paneth cells, stem cells and neuroendocrine cells. They are separated by a basement membrane from a rich capillary plexus within the lamina propria. page 1159 page 1160
Figure 72.7 Electron micrograph of columnar enterocytes covering a villus in the small intestine, each with an apical brush border of microvilli. A single goblet cell containing mucinogen granules is also present. The small cell towards the left between two enterocytes is an intraepithelial lymphocyte. A capillary and other connective tissue elements of the lamina propria lie beneath the basal lamina.
Enterocytes Enterocytes in the crypts secrete ions and alkaline fluid to dilute chyme and aid
absorption by structurally similar cells covering the villi. Mucous cells The mucous cells in the crypts are similar to the goblet cells of the villi. Paneth cells Paneth cells are numerous in the deeper parts of the intestinal crypts, particularly in the duodenum. They are rich in zinc and contain large acidophilic granules (Fig. 72.6) that stain strongly with eosin or phosphotungstic haematoxylin. Paneth cells secrete lysozyme, a highly specific antibacterial enzyme, and other defensive proteins (defensins) such as tumour necrosis factor alpha (TNF-!), which protect the intestinal luminal surface. UPDATE Abstract: Defensins and mucosal protection.
Date Added: 28 June 2005
Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=15669635&query_hl=10 Defensins and mucosal protection. Stem cells Stem cells occur in a zone just above the basal region of the crypts and are the source of most of the cell types of the intestinal epithelium. Their progeny, transit (transient) amplifying cells, have one of the most rapid proliferation rates in the body. They migrate out of the intestinal crypts along the sides of the villi, where they differentiate mainly into the short-lived columnar enterocytes or goblet cells (which are thus continually replaced). When not dividing, their apical surfaces have fewer and more irregular microvilli than the differentiated enterocytes. Neuroendocrine cells Several types of neuroendocrine cell are scattered among the walls of the intestinal crypts, and less commonly over the villi. They secrete bioactive peptides, such as gastrin, cholecystokinin and secretin , basally into the surrounding lamina propria. Crypt neuroendocrine cells are derived from stem cells, which also give rise to enterocytes and other epithelial elements. For further details of the dispersed neuroendocrine system see page 180.
LAMINA PROPRIA The lamina propria is composed of connective tissue and provides mechanical support for the epithelium. It has a rich vascular plexus, receives nutrients absorbed by the enterocytes, and forms the cores of the villi. It also contains lymphoid tissue, fibroblasts and connective tissue extracellular matrix fibres, smooth muscle cells, eosinophils, macrophages, mast cells, capillaries, lymphatic vessels and unmyelinated nerve fibres. Plasma cells are numerous and lymphocytes in many regions are clustered in solitary and aggregated lymphatic follicles, Peyer's patches, some of which extend through the muscularis mucosae into the submucosa. Core of the villus Each villus has a core of delicate connective tissue that contains a large blind-
ending lymphatic vessel or lacteal (so called because of its content of suspended chylomicrons, the droplets of apoprotein-lipid complex elaborated by enterocytes from absorbed dietary fats). The core also contains blood vessels, nerves and smooth muscle cells derived from fine extensions of the muscularis mucosae. Each lacteal, usually single but occasionally double, starts in a closed dilated extremity near the tip of a villus, and extends through the core to the base of the villus, where it joins a narrower lymphatic plexus in the deeper lamina propria. Its wall is a single layer of endothelial cells. Smooth muscle cells surround the lacteal throughout the villus and their contraction propels its contents into the underlying lymphatic plexus. Capillaries within the core are lined by fenestrated endothelium to facilitate the rapid intake of nutrients diffusing from the covering absorptive epithelium. Mucosa-associated lymphoid tissue Mucosa-associated lymphoid tissue (MALT) is found mainly in the lamina propria, but sometimes expands into the submucosa. It is the source of B and T lymphocytes and other related cells for the immune defence of the gut wall (p. 77). MALT consists of lymphoid follicles covered by intestinal epithelium that includes a few M cells. Solitary lymphoid follicles are scattered along the length of the intestinal mucosa, and are most numerous in the distal ileum. Aggregated follicles, Peyer's patches, are largest and most numerous in the ileum, whilst in the distal jejunum they are small, circular and few; they are only occasionally found in the duodenum. They are usually situated in the intestinal wall opposite the mesenteric attachment. Aggregated lymphoid follicles are circular or oval masses containing 10-260 follicles, varying in length from 2 to 10 cm and visible macroscopically as domelike elevations. Villi are small or absent over the larger follicular groups. Like other masses of MALT (except lymph nodes), solitary and aggregated lymphoid follicles are most prominent around the age of puberty, after which they diminish in number and size, although many persist into old age. For further details of intestinal MALT, including Peyer's patches, see the review by Kraehenbuhl and Neutra (1992).
MUSCULARIS MUCOSAE The muscularis mucosae forms the base of the mucosa, and has external longitudinal and internal circular layers of smooth muscle cells. It follows the surface profiles of the circular folds and sends fine fascicles of smooth muscle cells into the cores of the villi.
© 2008 Elsevier
SUBMUCOSA page 1160 page 1161
The submucosa is composed of loose connective tissue carrying blood vessels, lymphatics and nerves. Its ridged elevations form the cores of the circular folds. The geometry of its collagen and elastin fibres permits the considerable changes in transverse and longitudinal dimensions that accompany peristalsis, whilst still providing adequate support, elasticity and strength.
SUBMUCOSAL GLANDS Submucosal glands are limited to the submucosa of the duodenum (Figs 70.1, 72.1). Their ducts traverse the muscularis mucosae to enter the bases of the mucosal crypts. They are largest and most numerous near the pylorus, and form an almost complete layer in the proximal half of the descending duodenum. Thereafter they gradually diminish in number and disappear at the duodenojejunal junction. They are small, branched tubuloacinar glands (p. 34): each has several secretory acini lined by low columnar epithelial cells that produce an alkaline (c.pH 9) mucoid secretion which effectively neutralizes acidic chyme from the stomach. Many neuroendocrine cells are present among the acinar cells.
© 2008 Elsevier
MUSCULARIS EXTERNA The muscularis externa consists of a thin external longitudinal layer and a thick internal circular layer of smooth muscle cells. It is thicker in the proximal small intestine. For details, see Gabella (1988).
© 2008 Elsevier
SEROSA Serosa is visceral peritoneum. It consists of a subserous stratum of loose connective tissue covered by mesothelium. The retroperitoneal portion of the duodenum is mainly covered by a connective tissue adventitia rather than by serosa. REFERENCES Gabella G 1988 Structure of intestinal musculature. In: Handbook of Physiology; The Gastrointestinal System I. New York: American Physiological Society and Oxford University Press: 103-39. Kraehenbuhl JP, Neutra MR 1992 Molecular and cellular basis of immune protection of mucosal surfaces. Physiol Rev 72: 853-79. Medline Similar articles page 1161 page 1162
© 2008 Elsevier
73 GASTROINTESTINAL TRACT - SMALL INTESTINE: Duodenum The adult duodenum is c.20-25 cm long and is the shortest, widest and most predictably placed part of the small intestine. It is only partially covered by peritoneum although the extent of the peritoneal covering varies along its length: the proximal 2.5 cm is intraperitoneal; the remainder is retroperitoneal. The duodenum forms an elongated 'C' that lies between the level of the first and third lumbar vertebrae in the supine position. The lower 'limb' of the C extends further to the left of the midline than the upper limb. The head and uncinate process of the pancreas lie within the concavity of the C. The duodenum lies entirely above the level of the umbilicus and is described as having four parts (Figs 73.1, 73.2). UPDATE Date Added: 02 February 2005 Abstract: Vascular anatomy of the pancreaticoduodenal region: A review. Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=10436238 Vascular anatomy of the pancreaticoduodenal region: A review.
FIRST (SUPERIOR) PART The first part of the duodenum is c.5 cm long and starts as a continuation of the duodenal end of the pylorus. It is the most mobile portion of the duodenum. Close to the pylorus, peritoneum covers the anterior, superior and upper part of the posterior aspect where the duodenum forms part of the anterior wall of the epiploic foramen. Here the lesser omentum is attached to its upper border and the greater omentum to its lower border. The first 2 or 3 cm have a bland internal mucosal appearance and readily distend on insufflation during endoscopy. This part is frequently referred to as the duodenal 'cap': it has a triangular, homogeneous appearance during contrast radiology and shows the same pattern of rugae as the pylorus. It is often visible on plain radiographs of the abdomen as an isolated triangular gas shadow to the right of the first or second lumbar vertebra. The first part then passes superiorly, posteriorly and laterally for c.5 cm before curving sharply inferiorly into the superior duodenal flexure, which marks the end of the first part. Through this course it rapidly becomes more retroperitoneal and is covered by peritoneum only on its anterior aspect. From the end of the duodenal cap, the internal appearance is characterized by extensive, deep mucosal folds that involve up to half of the circumference of the lumen. Even during endoscopic insufflation, these folds are pronounced (Fig. 73.3) and they are readily seen on contrast radiographs (Fig. 73.2). The section from the duodenal cap to the superior duodenal flexure lies posterior and inferior to the quadrate lobe of the liver. At the junction with the second part of the duodenum it lies posterior to the neck of the gallbladder. The first part of the duodenum lies anterior to the gastroduodenal artery, common bile duct and portal vein and anterosuperior to the head and neck of the pancreas. The gastroduodenal artery lies immediately posterior to the outer muscular layers of the posterior wall of the first part. Peptic ulceration is commonly found on the posterior wall in this region and penetration of the wall with erosion of the gastroduodenal artery may lead to dramatic haemorrhage. The common hepatic and hepatoduodenal lymph nodes also lie close to the first part of the duodenum and can be visualized using endoscopic ultrasound. This may be important in the staging of gastric, pancreatic or bile duct tumours. The proximity of the common bile duct to the first part of the
duodenum allows endoscopic ultrasound examination of the distal common bile duct and the formation of a surgical anastomosis between bile duct and duodenum (choledochoduodenostomy) when required. Penetrating peptic ulceration in the anterior wall may lead to free perforation into the peritoneal cavity because the anterior surface of the first part is covered only by peritoneum.
SECOND (DESCENDING) PART The second part of the duodenum is c.8-10 cm long. It starts at the superior duodenal flexure and runs inferiorly in a gentle curve, which is convex to the right side of the vertebral column, extending to the lower border of the third lumbar vertebral body. It then turns sharply medially into the inferior duodenal flexure which marks its junction with the third part. It is covered by peritoneum only on its upper anterior surface, lies posterior to the neck of the gallbladder and the right lobe of the liver at its start, and is crossed anteriorly by the transverse colon. The origin of the transverse mesocolon is attached to the anterior surface of the duodenum by loose connective tissue. Below the attachment of the transverse mesocolon, the connective tissue and vessels forming the mesentery of the upper ascending colon and hepatic flexure are loosely attached to the anterior surface of the duodenum. This section of duodenum is at risk of injury during surgical mobilization of the ascending colon. The second part lies anterior to the hilum of the right kidney, the right renal vessels, the edge of the inferior vena cava and the right psoas major. The head of the pancreas and the common bile duct are medial and the hepatic flexure is above and lateral. A small part of the pancreatic head is sometimes embedded in the medial duodenal wall, and pancreatic rests in the duodenal wall produce small filling defects on double contrast barium meal. The internal appearance is similar to that of the distal portion of the first part of the duodenum, with pronounced mucosal folds (Fig. 73.3). The common bile duct and pancreatic duct enter the medial wall obliquely and unite to form the hepatopancreatic ampulla (p. 1228). The narrow, distal, end opens on the summit of the major duodenal papilla (ampulla of Vater), situated on the posteromedial wall of the second part c.8-10 cm distal to the pylorus (Fig. 73.3). An accessory pancreatic duct may open c.2 cm above the major papilla on a minor duodenal papilla (p. 1233). Peptic ulceration of the second part is less common than that of the first part, but tends to occur on the anterior or lateral wall. Duodenal diverticula
The duodenum is the most common site for a diverticulum in the small intestine. Diverticula are congenital and usually solitary. They almost always arise in the medial wall of the second part of the duodenum, where they are intimately related to the head of the pancreas. This relationship means that diverticula are frequently related to the major duodenal papilla (ampulla of Vater), and the latter may be found either on the mucosal fold at the mouth of a diverticulum or arising from the mucosa within the body of a diverticulum, particularly on the superomedial wall. This may complicate interpretation of contrast radiographs of the duodenum or biliary system and may cause difficulties during attempted endoscopic cannulation of the papilla because the diverticulum may restrict access. If the papilla is located on or close to the anterior wall of a diverticulum, the procedure of opening the ampulla with an electrocautery current during attempted cannulation (sphincterotomy) is made more hazardous because the wall is thin at this point and so there is a risk of free perforation into the peritoneal cavity.
THIRD (HORIZONTAL) PART page 1163
page 1164
Figure 73.1 A, The four parts of the duodenum. B, C The relations of the duodenum: B, anterior surface; C, posterior surface.
Figure 73.2 Contrast radiographic appearance of the duodenum showing a distended duodenal cap and the remainder of the duodenum up to the duodenojejunal flexure.
page 1164 page 1165
Figure 73.3 The endoscopic appearances of the duodenum: A, duodenal cap; B, distal first part; C, second part showing the major duodenal papilla; D, third part.
The third part of the duodenum starts at the inferior duodenal flexure and is c.10 cm long. It runs from the right side of the lower border of the third lumbar vertebra, angled slightly superiorly, across to the left, anterior to the inferior vena cava, and ends in continuity with the fourth part in front of the abdominal aorta. It lies posterior to the transverse mesocolon, the origin of the small bowel mesentery and the superior mesenteric vessels. Peritoneum covers the lower portion of the anterior aspect and is reflected anteriorly to form the posterior layer of the origin of the small bowel mesentery. The anterior surface of the left lateral end, close to the junction with the fourth part, is also covered with peritoneum. The third part is anterior to the right ureter, right psoas major, right gonadal
vessels, inferior vena cava and abdominal aorta (at the origin of the inferior mesenteric artery), and inferior to the head of the pancreas. Anteroinferiorly, loops of jejunum lie in the right and left infracolic compartments. In the mid portion, the third part is potentially 'pinched' between the superior mesenteric vessels just below their origin anteriorly, and the abdominal aorta posteriorly: this arrangement very occasionally gives rise to intermittent obstruction of the duodenum at this point.
FOURTH (ASCENDING) PART The fourth part of the duodenum is c.2.5 cm long. It starts just to the left of the aorta and runs superiorly and laterally to the level of the upper border of the second lumbar vertebra. It then turns anteroinferiorly at the duodenojejunal flexure and is continuous with the jejunum. The aorta, left sympathetic trunk, left psoas major, left renal and left gonadal vessels are all posterior, and the left kidney and left ureter are posterolateral. The main trunk of the inferior mesenteric vein lies either posterior to the duodenojejunal flexure or beneath the adjacent peritoneal fold. (The duodenojejunal flexure is a useful landmark to locate the vein radiologically or surgically.) Anteriorly are the upper part of the root of the small bowel mesentery, the left lateral transverse mesocolon and transverse colon, which separate it from the stomach. The peritoneum of the root of the small bowel mesentery continues over the anterior surface. The lower border of the body of the pancreas is superior. At its left lateral end, the fourth part becomes progressively covered in peritoneum on its superior and inferior surfaces, such that it is suspended from the retroperitoneum by a double fold of peritoneum, the ligament of Treitz, at the start of the duodenojejunal flexure. The ligament of Treitz is not a mesentery because the vascular supply to the fourth part of the duodenum continues to enter its wall from the posteromedial aspect. It may contain a small slip of muscle called the suspensory muscle of the duodenum. When present, the suspensory muscle contains skeletal muscle fibres that run from the left crus of the diaphragm to connective tissue around the coeliac axis, and smooth muscle fibres that run from the coeliac axis: its function is unknown. The ligament of Treitz is an important landmark in the radiological diagnosis of incomplete rotation and malrotation of the small intestine (p. 1257). UPDATE Date Added: 28 June 2005 Abstract: Development and distribution of mast cells and neuropeptides in human fetus duodenum. Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=15484316&query_hl=8 Development and distribution of mast cells and neuropeptides in human fetus duodenum.
© 2008 Elsevier
VASCULAR SUPPLY AND LYMPHATIC DRAINAGE ARTERIES The main vessels supplying the duodenum are the superior and inferior pancreaticoduodenal arteries. The first and second parts also receive contributions from several sources including the right gastric artery, the supraduodenal artery, the right gastroepiploic artery, the hepatic artery and the gastroduodenal artery. Branches of the superior pancreaticoduodenal artery may contribute to the supply of the pyloric canal, with some anastomosis in the muscular layer across the pyloroduodenal junction. UPDATE Date Added: 02 February 2005 Abstract: The pancreaticoduodenal arteries in human foetal development. Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=15478102 The pancreaticoduodenal arteries in human foetal development. Krakowiak-Sarnowska E, Flisinski P, Szpinda M, Flisinski M, Sarnowski J. The pancreaticoduodenal arteries in human foetal development. Folia Morphol (Warsz). 2004 Aug; 63(3): 281-4.
UPDATE Date Added: 02 February 2005 Abstract: Blood supply to the duodenal papilla and the communicating artery between the anterior and posterior pancreaticoduodenal arterial arcades. Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=11455486 Blood supply to the duodenal papilla and the communicating artery between the anterior and posterior pancreaticoduodenal arterial arcades. Gastroduodenal artery
The gastroduodenal artery arises from the common hepatic artery behind, or sometimes above, the first part of the duodenum. It is of moderately large calibre and descends between the first part of the duodenum and the neck of the pancreas, immediately to the right of the peritoneal reflection from the posterior surface of the first part. It usually lies to the left of the common bile duct but is occasionally anterior. At the lower border of the first part of the duodenum it divides into the right gastroepiploic and superior pancreaticoduodenal arteries. Before its division the lowest part of the artery gives rise to small branches that supply the pyloric end of the stomach and the pancreas, and retroduodenal branches that supply the first part and the proximal portion of the second part of the duodenum directly. The supraduodenal artery often arises from the gastroduodenal artery behind the upper border of the first part of the duodenum and supplies the superior aspect of the first part. Although the gastroduodenal artery usually branches from the common hepatic artery it may also arise as a trifurcation with the right and left hepatic arteries. Occasionally the origin is from the superior mesenteric artery or the left hepatic artery, and rarely it may arise from the coeliac axis and right hepatic artery. The supraduodenal artery occasionally arises from the common hepatic artery or right gastric artery.
Superior pancreaticoduodenal arteries
The superior pancreaticoduodenal artery is usually double. The anterior artery is a terminal branch of the gastroduodenal artery and descends in the anterior groove between the second part of the duodenum and the head of the pancreas. It supplies branches to the first and second parts of the duodenum and to the head of the pancreas, and anastomoses with the anterior division of the inferior pancreaticoduodenal artery. The posterior artery is usually a separate branch of the gastroduodenal artery and is given off at the upper border of the first part of the duodenum. It descends to the right, anterior to the portal vein and common bile duct as the latter lies behind the first part of the duodenum. It then runs behind the head of the pancreas, crosses posterior to the common bile duct (which is embedded in the head of the pancreas), enters the duodenal wall and anastomoses with the posterior division of the inferior pancreaticoduodenal artery. The posterior artery supplies branches to the head of the pancreas, the first and second parts of the duodenum, and several branches to the lowest part of the common bile duct. UPDATE Date Added: 02 February 2005 Abstract: The arterial blood supply of the pancreas: A review. I. The superior pancreaticoduodenal and the anterior superior pancreaticoduodenal arteries. An anatomical and radiological study. Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=7482159 The arterial blood supply of the pancreas: A review. I. The superior pancreaticoduodenal and the anterior superior pancreaticoduodenal arteries. An anatomical and radiological study. Inferior pancreaticoduodenal artery
The inferior pancreaticoduodenal artery arises from the superior mesenteric artery or its first jejunal branch, near the superior border of the third part of the duodenum. It usually divides directly into anterior and posterior branches. The anterior branch passes to the right, anterior to the lower border of the head of the pancreas, and runs superiorly to anastomose with the anterior superior pancreaticoduodenal artery. The posterior branch runs posteriorly and superiorly to the right, lying posterior to the lower border of the head of the pancreas, and anastomoses with the posterior superior pancreaticoduodenal artery. Both branches supply the pancreatic head, its uncinate process and the second and third parts of the duodenum. UPDATE Date Added: 02 February 2005 Abstract: The arterial blood supply of the pancreas: A review. IV. The anterior inferior and posterior pancreaticoduodenal aa., and minor sources of blood supply for the head of the pancreas. An anatomical review and radiologic study. Click on the following link to view the abstract: The arterial blood supply of the pancreas: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=9381324 A review. The anterior inferior and posterior pancreaticoduodenal aa., and minor sources of
blood supply for the head of the pancreas. An anatomical review and radiologic study. UPDATE Date Added: 02 February 2005 Abstract: The arterial blood supply of the pancreas: A review. III. The inferior pancreaticoduodenal artery. An anatomical review and a radiological study. Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=8782310 The arterial blood supply of the pancreas: A review. III. The inferior pancreaticoduodenal artery. An anatomical review and a radiological study. Jejunal artery branches
The first jejunal branch of the superior mesenteric artery has branches that supply the fourth part of the duodenum. They frequently form an anastomosis with the terminal branch of the anterior inferior pancreaticoduodenal artery, which means that the fourth part of the duodenum has a potential collateral supply from the coeliac axis and superior mesenteric artery and so is not commonly affected by ischaemia.
VEINS The duodenal veins drain ultimately into the portal vein. Submucosal and intramural veins give rise to pancreaticoduodenal veins that usually accompany the corresponding named arteries. The superior pancreaticoduodenal vein is formed medial to the mid point of the second part of the duodenum. It runs superomedially on the posterior surface of the head of the pancreas, passes posterior to the distal common bile duct and drains into the portal vein behind the neck of the pancreas. The inferior pancreaticoduodenal vein runs from the anteromedial aspect of the second part of the duodenum inferiorly in the groove between the second and third parts and the head of the pancreas. It usually drains into the superior mesenteric vein but may drain into the right gastroepiploic vein. Small veins from the first and upper second part of the duodenum may drain into the prepyloric vein whilst veins from the third and fourth parts may drain directly into the superior mesenteric vein.
LYMPHATICS Duodenal lymphatics run to anterior and posterior pancreatic nodes that lie in the anterior and posterior grooves between the pancreatic head and the duodenum. These drain widely into the suprapyloric, infrapyloric, hepatoduodenal, common hepatic and superior mesenteric nodes.
© 2008 Elsevier
INNERVATION page 1165 page 1166
The duodenum is innervated by the parasympathetic and sympathetic systems. Preganglionic sympathetic fibres originate from neurones in the fifth to the twelfth thoracic spinal segments and travel via the greater and lesser splanchnic nerves to the coeliac plexus where they synapse on neurones in the coeliac ganglion: postganglionic axons are distributed via periarterial plexuses on the branches of the coeliac axis and superior mesenteric artery. The parasympathetic supply is from the vagus nerve via branches from the coeliac plexus.
REFERRED PAIN In common with other structures derived from the foregut, the visceral sensation of pain arising from the duodenum is poorly localized and referred to the central epigastrium. REFERENCE Jackson JE 1999 Vascular anatomy of the gastrointestinal tract. In: Butler P, Mitchell AWM, Ellis H (eds) Applied Radiological Anatomy. Cambridge, UK: Cambridge University Press. Provides details and illustrations of common vascular variants.
© 2008 Elsevier
74 GASTROINTESTINAL TRACT - SMALL INTESTINE: Jejunum and ileum The small intestine consists of the duodenum, jejunum and ileum and extends from the pylorus to the ileocaecal valve. In the living adult it has a total length of c.5 metres, but this can range widely from less than 3 metres to more than 7 metres. The proximal two-fifths is referred to as the jejunum and the distal threefifths as the ileum, although there is no clear distinction between the two parts. There is, however, a gradual change in morphology from the proximal to distal ends of the small bowel. The distal 30 cm or so of the ileum is often referred to as the terminal ileum, which has some specialized physiological functions. The jejunum and ileum occupy the central and lower parts of the abdominal cavity and usually lie within the boundary formed by the abdominal colon. They are attached to the posterior abdominal wall by a mesentery and this allows considerable mobility of the loops of small bowel. In the supine position, loops of jejunum may be found anterior to the transverse colon, stomach and even lesser omentum. When upright, loops of ileum may descend into the pelvis anterior to the rectum and, in women, may fill the rectouterine pouch. The majority of the jejunum and ileum is covered anteriorly by the greater omentum (p. 1132). The jejunum and ileum are covered by peritoneum on all but their mesenteric borders where the adipose connective tissue of the mesentery abuts the muscular wall. The peritoneum continues over these tissues to enclose the mesentery (p. 1133). Mesenteric fat covers c.20% of the circumference wall of the ileum and somewhat less of the jejunum. The mucosa of the jejunum and ileum is thrown into numerous circular folds, plicae circulares, which protrude into the lumen. The submucosa of the small bowel contains aggregates of lymphoid tissue, which are more numerous in the ileum.
© 2008 Elsevier
JEJUNUM The jejunum has a median external diameter of c.4 cm and an internal diameter of 2.5 cm. It has a thicker wall than the ileum, and possesses a profuse arterial blood supply so that it appears redder than the ileum. The plicae circulares are most pronounced in the proximal jejunum, where they are more numerous and deeper than elsewhere in the small bowel. They frequently 'branch' around the lumen and may appear to be ranged one on top of another, giving the jejunum a characteristic appearance during single contrast radiography (Fig. 74.1). Lymphoid aggregates are almost absent from the proximal jejunum; they are present distally but are still fewer in number and smaller than in the ileum. They are usually discoid in shape and impalpable. In the supine position the jejunum usually occupies the upper left infracolic compartment extending down to the umbilical region. The first loop or two often occupies a recess between the left part of the transverse mesocolon and the left kidney. On supine radiological examination, the jejunal loops are characteristically situated in the upper abdomen, to the left of the midline, whereas the ileal loops tend to lie in the lower right part of the abdomen and pelvis. This distribution can be reversed during ileus or small bowel obstruction due to rotation around the mesenteric attachment following bowel distension.
JEJUNAL FEEDING In situations where the stomach and duodenum are either unsuitable or unavailable for receiving oral nutrition, delivery of prepared feed to the jejunum is possible. This can be performed either using a surgically created jejunostomy or by insertion of a feeding tube. Because the jejunum is highly mobile, it is possible to bring the first or second loop of jejunum into contact with the abdominal wall to create a surgical jejunostomy. Insertion of a fine-bore feeding tube via the nose as far as the jejunum is also possible. The end of the feeding tube must lie beyond the duodenojejunal flexure to prevent reflux of the feed into the duodenum and stomach; this is usually confirmed by radiological monitoring of the progress of the tube through the duodenum.
© 2008 Elsevier
ILEUM The ileum has a median external diameter of 3.5 cm and an internal diameter of 2 cm; it tends to have a thinner wall than the jejunum. The plicae circulares become progressively less obvious in the distal mucosa in the ileum: they tend to be single and flatter with less pronounced crests (Figs 74.1, 74.2). The mucosa of the terminal ileum immediately proximal to the ileocaecal valve may appear almost flat. Lymphoid aggregates are larger and more numerous than in the jejunum and may be easily palpable in the terminal ileum. They are most prominent in early childhood, become less so prior to puberty, and are of the adult type by late teenage years. In the supine position, the ileum lies mainly in the hypogastric region and right iliac fossa. The terminal ileum frequently lies in the pelvis, from which it ascends over the right psoas major and right iliac vessels to end in the right iliac fossa, where it opens into the ileocaecal valve.
MECKEL'S DIVERTICULUM The ileal diverticulum (of Meckel) exists in c.3% of adults: it represents the remnant of the proximal part of the intestino-vitelline duct. It projects from the antimesenteric border of the terminal ileum and is commonly located between 50 and 100 cm from the ileocaecal valve. It has a median length of c.5 cm and often possesses a short 'mesentery' of adipose tissue that extends from the ileal mesentery up to the base. The lumen of the diverticulum is usually wide, with a calibre similar to that of the ileum. The tip is usually free but occasionally may be connected to the anterior abdominal wall near the umbilicus by a fibrous band. The mucosa is ileal in type but small areas may be lined by gastric body type epithelium, which occasionally gives rise to bleeding in the adjacent normal ileal mucosa. Heterotopic areas of pancreatic, colonic or other tissues may occur in its wall. Inflammation may mimic acute appendicitis: Meckel's diverticulum is derived from midgut structures, and so pain is referred to the periumbilical region as it is in early appendicitis.
© 2008 Elsevier
VASCULAR SUPPLY AND LYMPHATIC DRAINAGE ARTERIES The arterial supply to the jejunum and ileum arises from the superior mesenteric artery. Branches divide as they approach the mesenteric border and extend between the serosal and muscular layers. From these, numerous branches traverse the muscle, supplying it and forming an intricate submucosal plexus from which minute vessels pass to the glands and villi. Although there is a profuse anastomotic network of arteries within the mesentery, anastomoses between the terminal branches close to the intestinal wall are few, and alternate vessels are often distributed to opposite sides of the jejunum/ileum (Fig. 74.3). page 1167 page 1168
page 1168 page 1169
Figure 74.1 Barium studies of the jejunum and ileum. A, Barium follow-through. The feathery appearance of the small intestine is due to the plicae circulares, this is most prominent in the jejunum. The constrictions are the result of peristalsis. B, Small bowel enema (enteroclysis). The plicae circulares are clearly demonstrated by this technique. C, caecum; I, ileum; J, jejunum; PC, plicae circulares; TI, terminal part of ileum.
Figure 74.2 Endoscopic appearance of the terminal ileum.
SUPERIOR MESENTERIC ARTERY The superior mesenteric artery (Figs 74.4, 74.5) originates from the aorta c.1 cm below the coeliac trunk, at the level of the intervertebral disk between the first and
second lumbar vertebrae. It runs inferiorly and anteriorly, anterior to the uncinate process of the pancreas and the third part of the duodenum, and posterior to the splenic vein and the body of the pancreas. The left renal vein lies behind it and separates it from the aorta (Fig. 74.6). As it descends in the root of the small bowel mesentery, the artery crosses anterior to the inferior vena cava, right ureter and right psoas major. The calibre of the vessel steadily decreases as successive branches are given off to the loops of jejunum and ileum: it ends in a terminal branch which anastomoses with the ileocolic artery. The superior mesenteric artery gives off the middle colic (p. 1193), right colic (p. 1203), ileocolic, jejunal and ileal branches. A fibrous strand from the region of the last ileal branch may be present in the mesentery and represents the vestige of the embryonic artery that originally connected it to the yolk sac. The superior mesenteric artery may be the source of the common hepatic, gastroduodenal, accessory right hepatic, accessory pancreatic or splenic arteries. It may arise from a common coeliac-mesenteric trunk. Jejunal branches Jejunal branches arise from the left side of the upper portion of the superior mesenteric artery (Figs 74.4, 74.5). They are usually five to ten in number and are distributed to the jejunum as a series of short arcades. These form a single or occasionally double tier of anastomotic arcs before giving rise to multiple straight vessels that run directly towards the jejunal wall (Fig. 74.3). These vessels run almost parallel in the mesentery and are distributed alternately to opposite aspects of its wall. Small twigs supply regional lymph nodes and other structures in the mesentery. Ileal branches Ileal branches arise from the left and anterior aspects of the superior mesenteric artery. They are more numerous than the jejunal branches but smaller in calibre. The length of the mesentery is greater in the ileum and the branches form three, four or sometimes five tiers of arcs within the mesentery before giving rise to multiple straight vessels that run directly towards the ileal wall. As with the jejunal branches, these run parallel in the mesentery and are distributed to alternate aspects of the ileum. They are longer and smaller than similar jejunal vessels, particularly in the distal ileum. In the terminal ileum, the arcades receive a contribution from the ileal branch of the ileocolic artery and are often larger in calibre than the mid-ileal vessels.
VEINS Superior mesenteric vein
The superior mesenteric vein drains the small intestine, caecum, ascending and transverse parts of the colon (Figs 74.5B, 74.7). It is formed in the right lower mesentery of the small bowel by the union of tributaries from the terminal ileum, caecum and vermiform appendix. It ascends in the mesentery to the right of the superior mesenteric artery. It passes anterior to the right ureter, inferior vena cava, third part of the duodenum and uncinate process of the pancreas, and joins the splenic vein behind the neck of the pancreas to form the portal vein.
page 1169 page 1170
Figure 74.3 Specimens of the jejunum (A) and ileum (B) (overleaf) from a subject in whom the superior mesenteric artery was injected with a red coloured mass of gelatin before fixation. Subsequently the specimens were dehydrated and then cleared in benzene followed by methyl salicylate. The largest vessels present are the jejunal and ileal branches of the superior mesenteric artery and these are succeeded by anastomotic arterial arcades, which are relatively few in number (1-3) in the jejunum, becoming more numerous (5-6) in the ileum. From the arcades, straight arteries pass towards the gut wall; frequently, successive straight arteries are distributed to opposite sides of the gut. Note the denser vascularity of the jejunal wall. (Specimens prepared by MCE Hutchinson; photographs by Kevin Fitzpatrick on behalf of GKT School of Medicine, London.)
page 1170 page 1171
Figure 74.4 The superior mesenteric artery and its branches. The first loop of the jejunum and the terminal loop of the ileum have been spread out to show the arrangement of their arteries.
Figure 74.5 The superior mesenteric artery and its branches: digital subtraction angiogram of the superior mesenteric artery (A) and vein (B); C, sagittal reformat of multislice CT superior mesenteric angiogram. (A, by kind permission from Dr Adam Mitchell, Charing Cross Hospital, London.) (B, by kind permission from Dr Adam Mitchell, Charing Cross Hospital, London; C, by kind permission from GE Worldwide Medical Systems.)
Tributaries The superior mesenteric vein receives jejunal, ileal, ileocolic, right colic (p. 1191), middle colic (p. 1193), right gastroepiploic (p. 1149) and pancreaticoduodenal (p. 1165) veins in a similar distribution to the corresponding arteries.
LYMPHATICS Lymph vessels, called lacteals, are arranged at two levels within the wall of the small bowel. The first is mucosal and the second in the muscular coat. Lymph vessels from the villi arise from an intricate plexus in the mucosa and submucosa and are joined by vessels from lymph spaces at the bases of solitary lymphoid follicles. They drain to larger vessels at the mesenteric aspect of the gut. The lymph vessels of the muscular tunic form a close plexus that runs mostly between the two muscle layers. They communicate freely with mucosal vessels and also open into vessels at the mesenteric border. Mesenteric lacteals pass between the layers of the mesentery. They drain into a series of mesenteric lymph nodes arranged in tiers within the mesentery which follow the same distribution as the regional arterial supply and which may form a 'chain' along the major arteries. Elsewhere in the ileal and jejunal mesentery they form an extensive network that affords a relatively wide field of lymph node drainage. This arrangement makes radical surgical resection of lymph nodes difficult if the vessels to the remaining unaffected small bowel are to be preserved. The mesenteric nodes drain into superior mesenteric nodes around the root of the superior mesenteric artery.
© 2008 Elsevier
INNERVATION SUPERIOR MESENTERIC PLEXUS The superior mesenteric plexus, an inferior continuation of the coeliac plexus, lies in the preaortic connective tissue around the origin of the superior mesenteric artery, posterior to the pancreas. It receives preganglionic parasympathetic elements via the right vagus nerve. Preganglionic sympathetic fibres originate from neurones in the mid-thoracic spinal segments and travel in the greater and lesser splanchnic nerves to the coeliac and superior mesenteric ganglia where they synapse. The superior mesenteric ganglion lies superiorly in the plexus, usually above the origin of the superior mesenteric artery. Postganglionic axons accompany the superior mesenteric artery into the mesentery and are distributed along branches of the artery.
REFERRED PAIN In common with other structures derived from the midgut, the visceral sensation of pain arising from the jejunum or ileum is poorly localized. It is commonly referred to the periumbilical region or central epigastrium. page 1171 page 1172
Figure 74.6 Ultrasound images taken through the origin of the superior mesenteric artery demonstrated in the axial (A) and sagittal (B) planes.
Figure 74.7 The portal vein and its tributaries (semi-diagrammatic). Portions of the stomach, pancreas and left lobe of the liver and the transverse colon have been removed.
REFERENCES Kadir S 1991 Atlas of Normal and Variant Angiographic Anatomy. Philadelphia: WB Saunders. Underhill BML 1955 Intestinal length in man. Br Med J 2: 1243-6. Medline Similar articles
© 2008 Elsevier
75 GASTROINTESTINAL TRACT - LARGE INTESTINE: Microstructure of the large intestine The layers of tissue in the large intestinal wall (Figs 70.1, 75.1) resemble those in the small intestine (Ch. 72), except that villi and circular folds are absent and the glands (crypts) are longer.
© 2008 Elsevier
MUCOSA The mucosa is pale, smooth, and, in the colon, raised into numerous crescentshaped folds between the sacculi. In the rectum it is thicker, darker, more vascular, and more loosely attached to the submucosa.
EPITHELIUM OF THE CAECUM, APPENDIX, COLON AND UPPER RECTUM (Figs 75.1, 75.2, 75.3) The luminal surface is lined by columnar cells, mucous (goblet) cells, and occasional microfold (M) cells (p. 1158) that are restricted to the epithelium overlying lymphoid follicles. Columnar and mucous cells are also present in the intestinal glands (crypts) which additionally contain stem cells and neuroendocrine cells. The glands generally lack Paneth cells, but these may be present in the caecum. Columnar (absorptive) cells Columnar (absorptive) cells are the most numerous of the epithelial cell types. They are responsible for ion exchange and other transepithelial transport functions including water resorption, particularly in the colon. Although there is some variation in their structure, they all bear apical microvilli, which are shorter and less regular than those on enterocytes in the small intestine. All cells have typical junctional complexes around their apices, and these limit extracellular diffusion from the lumen across the intestine wall. Mucous (goblet) cells Mucous cells have a similar structure to those of the small intestine, but are more numerous. They are outnumbered by absorptive cells for most of the length of the colon, but they are equally frequent towards the rectum, where their numbers increase further. Microfold (M) cells Microfold cells are similar to those of the small intestine: they are flattened or cuboidal cells with long, blunt microfolds rather than typical microvilli, and they are restricted to epithelium overlying lymphoid follicles. Stem cells Stem cells are the source of the other epithelial cell types in the large intestine. They are located at or near the bases of the intestinal glands, where they divide by mitosis. They provide cells that migrate towards the luminal surface of the intestine: their progeny differentiate, undergo apoptosis and are shed after approximately 5 days. Neuroendocrine cells Neuroendocrine cells are situated mainly at the bases of the glands, and secrete basally into the lamina propria.
INTESTINAL GLANDS (CRYPTS) OF THE LARGE INTESTINE The crypts are narrow perpendicular tubular glands which are longer, more
numerous and closer together than those of the small intestine. Their openings give a cribriform appearance to the mucosa in surface view (Fig. 75.2). The glands are lined by low columnar epithelial cells, mainly goblet cells (Figs 75.1, 75.3), between which are columnar absorptive cells and neuroendocrine cells. Epithelial stem cells at their bases give rise to all three cell types.
LAMINA PROPRIA The lamina propria is composed of connective tissue that supports the epithelium. It forms a specialized pericryptal fibroblast sheath around each intestinal gland. Solitary lymphoid follicles within the lamina propria are most abundant in the caecum, appendix and rectum, but are also present scattered along the rest of the large intestine. They are similar to those of the small intestine (Ch. 72) and efferent lymphatic vessels originate within them. Lymphatic vessels are absent from the lamina propria core between crypts.
MUSCULAR MUCOSAE The muscularis mucosae of the large intestine is essentially similar to that of the small intestine: it has prominent longitudinal and circular layers.
© 2008 Elsevier
SUBMUCOSA The submucosa of the large intestine is similar to that of the small intestine.
© 2008 Elsevier
MUSCULARIS EXTERNA The muscularis externa has outer longitudinal and inner circular layers of smooth muscle. The longitudinal fibres form a continuous layer but, with the exception of the uniform outer muscle layer of most of the appendix, macroscopically these are aggregated as longitudinal bands or taeniae coli (Fig. 76.10). Between the taeniae coli the longitudinal layer is much thinner, less than half the circular layer in thickness. The circular fibres form a thin layer over the caecum and colon, and are aggregated particularly in the intervals between the sacculi. In the rectum they form a thick layer and in the anal canal they form the internal anal sphincter. There is an interchange of fascicles between circular and longitudinal layers, especially near the taeniae coli. Deviation of longitudinal fibres from the taeniae to the circular layer may, in some instances, explain the haustration of the colon.
© 2008 Elsevier
SEROSA The serosa or visceral peritoneum is variable in extent. Along the colon the peritoneum forms small fat-filled appendices epiploicae (Fig. 76.10) which are most numerous on the sigmoid and transverse colon but generally absent from the rectum. Subserous loose connective tissue attaches the peritoneum to the muscularis externa.
© 2008 Elsevier
MICROSTRUCTURE OF THE APPENDIX The layers of the wall of the appendix are essentially those of the rest of the large intestine but some features are notably different and are described here. The serosa forms a complete covering, except along the mesenteric attachment. The longitudinal muscular fibres form a complete layer of uniform thickness, except over a few small areas where both muscular layers are deficient, leaving the serosa and submucosa in contact. At the base of the appendix, the longitudinal muscle thickens to form rudimentary taeniae that are continuous with those of the caecum and colon. page 1173 page 1174
page 1174 page 1175
Figure 75.1 Diagrams showing the major regions of the large intestine, the microstructure of the colonic wall and its epithelial cells. Note the aggregations of lymphocytes (yellow) and undifferentiated epithelial cells (white).
The submucosa typically contains many large lymphoid aggregates that extend from the mucosa and obscure the muscularis mucosae layer: consequently this
becomes discontinuous. These aggregates also cause the mucosa to bulge into the lumen of the appendix, so that it narrows irregularly (Fig. 75.4). They are absent at birth but accumulate over the first 10 years of life to become a prominent feature. The mucosa is covered by a columnar epithelium as it is elsewhere in the large intestine, and M cells are present in the epithelium that overlies the mucosal lymphoid tissue. Glands (crypts) are similar to those of the colon but are fewer in number and thus less densely packed. They penetrate deep into the lymphoid tissue of the mucosal lamina propria (Fig. 75.4). The submucosal lymphoid tissue frequently exhibits germinal centres within its follicles (p. 77), indicative of B-cell activation, as it is in secondary lymphoid tissue elsewhere (p. 74). In adults, the normal layered structure of the appendix is lost and the lymphoid follicles atrophy and are replaced by collagenous tissue. In the elderly, the appendix may be filled with fibrous scar tissue.
Figure 75.2 Scanning electron micrograph of the luminal surface of human rectal mucosa. The outlines of absorptive epithelial cells bearing microvilli and the openings of rectal crypts can be seen. (Material provided by DS Rampton; prepared and photographed by Michael Crowder.)
Figure 75.3 Micrograph showing the colonic mucosa with mucus-secreting surface epithelium and glands (crypts). The smooth muscle of the muscularis mucosae is seen at the bottom of the field. (Photograph by Sarah-Jane Smith.)
page 1175 page 1176
Figure 75.4 Low-power micrograph of the appendix in transverse section, showing part of its circumference and a faecal pellet lodged in its lumen. Lymphoid tissue (basophilic staining) occupies much of the mucosa between crypts, and part of the submucosa. The muscularis externa and outermost serosa layer are seen at the bottom of the field. (Photograph by Sarah-Jane Smith.)
© 2008 Elsevier
76 GASTROINTESTINAL TRACT - LARGE INTESTINE Overview of the large intestine The large intestine extends from the distal end of the ileum to the anus, and is c.1.5 m long, although there is considerable variation in its length. Its calibre is greatest near the caecum and gradually diminishes to the level of mid rectum. It enlarges in the lower third of the rectum to form the rectal ampulla above the anal canal. The large intestine differs from the small intestine in that it has a greater calibre; it is for the most part more fixed in position; its longitudinal muscle, though a complete layer, is concentrated into three longitudinal bands, taeniae coli, in all but the distal sigmoid colon and rectum; small adipose projections, appendices epiploicae, are scattered over the free surface of the whole colon (they tend to be absent from the caecum, vermiform appendix and rectum). Moreover, the colonic wall is puckered into sacculations (haustrations), which may, in part, be due to the presence of the taeniae coli, and which may be demonstrated on plain radiographs as incomplete septations arising from the bowel wall. The function of the large intestine is chiefly absorption of fluid and solutes. Broadly speaking, the large intestine lies in a curve which extends from the right iliac fossa, ascends in the right flank, crosses the mid upper abdomen in a variable course, descends in the left flank, passes through the left iliac fossa and thence posteroinferiorly into the pelvis. (Fig. 76.1). It tends to form a border to the loops of the small intestine which are located centrally within the abdomen. The colon commences in the right iliac fossa as the caecum, from which the vermiform appendix arises. The caecum proceeds directly into the ascending colon which ascends in the right lumbar and hypochondriac regions to the inferior aspect of the liver where it bends to the left forming the hepatic flexure (right colic flexure). The large intestine then loops across the abdomen with an anteroinferior convexity as the transverse colon, and on reaching the left hypochondriac region it curves inferiorly to form the splenic flexure (left colic flexure). The descending colon proceeds through the left lumbar and iliac regions and becomes the sigmoid colon in the left iliac fossa. The sigmoid colon descends deep into the pelvis as the rectum and ends in the anal canal below the level of the pelvic floor. The large intestine develops as a fully mesenteric organ (p. 1259). However, after the rotation of the gut tube in utero, large portions of it come to lie adherent to the retroperitoneum, which means that some parts of the colon are fixed within the retroperitoneum, and other parts are suspended by a mesentery within the peritoneal cavity. Those portions of the colon within the retroperitoneum are separated from other retroperitoneal structures by a thin layer of connective tissue which forms an avascular field during surgical dissection, but which offers little or no barrier to the spread of disease within the retroperitoneum.
page 1177 page 1178
Figure 76.1 Overview of the abdominal colon and its relations.
The caecum may be within the retroperitoneum, but more frequently is suspended by a short mesentery. The ascending colon is usually a retroperitoneal structure although the hepatic flexure may be suspended by a mesentery. The transverse colon emerges from the retroperitoneum on a rapidly elongating mesentery and lies, often freely mobile, in the upper abdomen. The transverse mesocolon shortens to the left of the upper abdomen and may become retroperitoneal at the splenic flexure. Occasionally the splenic flexure is suspended by a short mesentery. The descending colon is retroperitoneal usually to the level of the left iliac crest. As the colon enters the pelvis it becomes increasingly more mesenteric again at the origin of the sigmoid colon, although the overall length of the sigmoid mesentery is highly variable. The distal sigmoid colon lies on a rapidly shortening mesentery as it approaches the pelvis and by the level of the rectosigmoid junction the mesentery has all but disappeared, so that the rectum enters the pelvis as a retroperitoneal structure. In the neonate and infant, the caecum and proximal ascending colon are often more mobile on a longer mesentery than in the adult. The mesenteries of the colon consist of visceral peritoneum enclosing connective and adipose tissues which envelop the vessels, nerves and lymphatics as they course from the retroperitoneum. There is a direct communication from the retroperitoneum to the suspended colon within the mesenteries via the so-called subperitoneal space.
EXTERNAL APPEARANCE
The haustrations of the colon are often absent in the caecum proximal to the origin of the ascending colon and are often relatively sparse in the ascending and proximal transverse colon. In these regions the taeniae coli are usually thin and occupy only a small percentage of the circumference of the colon. There are few if any appendices epiploicae on the serosal surface of the caecum, and only a limited number on the surface of the ascending colon. The haustrations become more pronounced from the middle of the transverse colon to the distal portion of the descending colon: the sigmoid colon is often characterized by marked sacculation. The width of the taeniae coli remains fairly constant throughout the length of colon but the number of appendices usually increases, becoming most numerous in the sigmoid colon where they can be fairly large in the obese individual. The taeniae are located in fairly constant positions beneath the serosal surface of the colon except in the transverse colon. They are oriented anteriorly, opposite the midline of the mesenteric attachment on the anti-mesenteric aspect of the colon (taenia libera), posterolaterally (taenia omentalis) and posteromedially (taenia mesocolica) midway between the taenia libera and the mesentery. In the caecum and descending colon, which are partly retroperitoneal structures, the posterolateral taenia is often obscured from view by the peritoneal reflection onto the colonic wall. In the transverse colon, the taeniae are rotated through 90° anterior being inferior, posteromedial being posterior and posterolateral being superior - as a consequence of the mobility and dependent position of this part of the colon. The taeniae coli broaden to occupy more of the circumference of the sigmoid colon in its distal portion and by the level of the rectosigmoid junction have widened to form distinct anterior and posterior bands, which unite to form a complete longitudinal muscle covering for the rectum. The rectum therefore has no external sacculation and no serosal appendices epiploicae.
INTERNAL APPEARANCE
Figure 76.2 Endoscopic appearance of the caecum. The characteristic trefoil appearance of the confluence of the three taeniae is usually obvious.
Figure 76.3 Endoscopic appearance of the ascending colon.
Figure 76.4 Endoscopic appearance of the transverse colon. The characteristic triangular appearance of the haustrations when viewed collectively is obvious.
page 1178 page 1179
Figure 76.5 Endoscopic appearance of the descending colon.
Figure 76.6 Endoscopic appearance of the sigmoid colon.
Figure 76.7 Appearance of the abdominal colon on double contrast barium enema examination demonstrating the transverse colon, hepatic and splenic flexures.
Figure 76.8 Endoscopic appearance of the rectum. Large transverse folds with little else in the way of mucosal folds characterizes the rectum. Prominent veins are often seen particularly in the lower third.
Throughout its length, the internal aspect of the colon is characterized by the presence of haustrations. These infoldings of the wall consist of mucosa and submucosa, and may partially span the lumen, but they never form a complete, circumferential ring. The pattern of the haustrations and appearance of the colonic mucosa help the clinician appreciate the level reached during flexible endoscopic examinations of the colon. In the portion of the caecum where haustrations occur, the three longitudinal taeniae coli converge to form a characteristic 'trefoil' pattern on the caecal wall (Fig. 76.2). Elsewhere, the wall of the lower pole of the caecum is usually devoid of haustrations, although a spiral mucosal pattern is often seen in the region of the appendix orifice (Fig. 78.3). The upper caecum and ascending colon possess shallow but long haustrations which may extend across one-third of the lumen (Fig. 76.3). This pattern is most pronounced in the transverse colon where the long haustrations often confer a triangular cross section on the lumen (Fig. 76.4). The wall of the colon is thinnest in this region and is most at risk of perforation during therapeutic endoscopic procedures. The haustrations of the descending and sigmoid colon tend to be thicker and shorter than those of the transverse colon, which gives a more circular cross-section to the lumen. The overall luminal diameter is often smallest in the descending colon (Fig. 76.5). During endoscopy the pattern of the submucosal vessels becomes more conspicuous in the sigmoid colon (Fig. 76.6). (The mobility of the sigmoid colon on its mesentery means that shorter lengths of colon tend to be visible during endoscopy than anywhere else in the colon.) The haustrations of the rectum tend to form consistent and recognizable folds: the pattern of the submucosal vessels is more pronounced than anywhere else in the colon (Fig. 76.8). Distinct veins are usually visible during endoscopy, and they are most marked above the anorectal junction.
CROSS-SECTIONAL APPEARANCE Cross-sectional imaging of the colon can be performed with computerized tomography (CT) and magnetic resonance imaging (MRI) permitting visualization of the bowel wall. On axial imaging the colon may be filled with particulate faeces and air (Fig. 76.9). The wall in normal individuals is imperceptibly thin. Diverticular disease is very common in adults, and air-filled diverticula are frequently identified as out-pouchings of the colonic wall, especially in the sigmoid colon. The air content and position of the colon facilitate its identification throughout the retroperitoneum and peritoneum suspended by its mesenteries. The caecum and ascending colon often contain faecal residue and are easily identified in the right retroperitoneum. The transverse colon may contain faeces or gas, but lies in a variable position suspended by its mesentery. The descending colon in the left retroperitoneum is frequently collapsed and contains little faecal residue. The volume data sets produced by modern multislice CT can now produce virtual colonoscopic mucosal images in the distended and cleaned colon, and surfacerendered images of the external surface of the bowel. In addition post-processing allows interrogation of the bowel mucosa in stretched out segments of bowel that can be opened up like a surgical specimen (Fig. 76.9). In addition, both CT and MR abdominal angiography are possible following an intravenous injection of contrast: these techniques are less invasive than conventional angiography, which is increasingly being reserved for patients who need interventional procedures performed under imaging guidance.
© 2008 Elsevier
VASCULAR SUPPLY AND LYMPHATIC DRAINAGE ARTERIES The arterial supply of the large intestine is derived from both the superior and inferior mesenteric arteries. Those parts derived from the midgut (caecum, appendix, ascending colon and right two-thirds of the transverse colon) are supplied from colic branches of the superior mesenteric artery; whilst hindgut derivatives (left part of the transverse, descending and sigmoid colon, rectum and upper anal canal) are supplied predominantly from the inferior mesenteric artery, with small contributions from branches of the internal iliac artery. The larger unnamed branches of these vessels ramify between the muscular layers of the colon which they supply. They subdivide into smaller submucosal rami and enter the mucosa. The terminal branches divide into vasa brevia and vasa longa which either enter the colonic wall directly or run through the subserosa for a short distance before crossing the circular smooth muscle to give off branches to the appendices epiploicae (Fig. 76.10). Superior mesenteric artery
The superior mesenteric artery supplies the caecum, appendix, ascending colon and right two-thirds of the transverse colon via the ileocolic, right colic and middle colic branches (Figs 76.11, 76.12). The ileocolic artery is formed as the distal continuation of the superior mesenteric artery in the root of the small bowel mesentery after the origin of the last ileal artery. Although it has many variations in its terminal distribution, it usually divides into a superior branch, which anastomoses with the right colic artery, and an inferior branch, which anastomoses with the distal superior mesenteric artery. The right colic artery usually arises as a common trunk with the middle colic artery. Occasionally it arises separately from the right side of the superior mesenteric artery and is absent rarely. Sometimes it arises from the ileocolic when it is named the accessory right colic artery. The middle colic artery is one of the first branches of the superior mesenteric artery and usually originates on its anterolateral aspect as a common trunk with the right colic artery. It arises just inferior to the uncinate process of the pancreas, anterior to the third part of the duodenum and ascends in the root of the transverse colon mesentery, just to the right of the midline dividing into terminal branches. Occasionally the middle colic artery arises separately from the right colic artery. It may arise from the dorsal pancreatic artery and rarely may arise from an accessory or replaced hepatic artery arising from the superior mesenteric artery. The artery may end in left and right main branches but frequently divides into three or more main branches within the mesentery. A large branch may be present, which runs parallel and posterior to the middle colic artery in the transverse mesocolon. This provides a direct communication between the superior and inferior mesenteric arteries and is known as the arc of Riolan. Inferior mesenteric artery page 1179 page 1180
The inferior mesenteric artery is usually smaller in calibre than the superior mesenteric artery, and arises from the anterior or left anterolateral aspect of the aorta at about the level of the third lumbar vertebra, 3 or 4 cm above the aortic bifurcation and posterior to the horizontal part of the duodenum. It descends deep to the peritoneum, initially anterior and then to the left of the aorta. It crosses the origin of the left common iliac artery medial to the left ureter and then enters, and continues in, the root of the sigmoid mesocolon as the superior rectal artery. Distally the inferior mesenteric vein is lateral to it. The principal branches are the left colic, sigmoid (of which there may be several) and superior rectal arteries (Figs 76.11, 76.13).
MARGINAL ARTERY OF THE COLON (Fig. 76.12)
page 1180 page 1181
Figure 76.9 Appearance of the colon on multislice computerized tomographic examination. The data acquired in the axial plane can be presented in a number of ways using multiplanar and volume rendered reformatting as demonstrated below. A, Axial CT showing air within the ascending, transverse and descending colon. B, Coronal reformat from axial data set showing the caecum, ascending and descending colon. C, Volume-rendering of the colonic wall using the axial data set to produce virtual colonoscopic views to show the triangular lumen of the transverse colon. D, Volume-rendering of the air-filled colon using the axial data set to give an image similar to a double contrast barium enema. (Images provided by kind permission from GE Worldwide Medical Systems.) E, Volume-rendering of the axial data set to produce a virtual dissection of the transverse colon.
Figure 76.10 Typical pericolic arrangement of arterial vasculature.
The marginal artery (marginal artery of Drummond) of the colon is the vessel which lies closest to and parallels the bowel wall. It is formed by the main trunks, and the arcades arising from, the ileocolic and right, middle and left colic arteries. Anastomoses form between the main terminal branches which run parallel to the
colon within the mesentery and give rise to vasa recta and vasa brevia to supply the colon. In the region of the splenic flexure the marginal artery receives contributions from the left branch of the middle colic artery - a branch of the superior mesenteric artery - which ramifies and anastomoses with an ascending branch of the left colic artery to supply the upper descending colon. The descending branch of the left colic artery ramifies and anastomoses with upper branches of the highest sigmoid artery to supply the descending colon. The origin of the primary arterial supply for the splenic flexure and distal third of the transverse colon is usually via the left colic artery but may be from the left branch of the middle colic artery. The marginal artery in the region of the splenic flexure may be absent or of such small calibre as to be of little clinical relevance. It may hypertrophy significantly when one of the main visceral arteries is compromised, e.g. following stenosis or occlusion of the inferior mesenteric artery, and it then provides a vessel of collateral supply. Colonic vascular occlusion
The marginal artery of the colon may become massively dilated when there is chronic, progressive occlusion of the superior mesenteric artery, because under these conditions it is required to supply the majority of the midgut (except the proximal portion which is supplied by collateral vessels from the coeliac artery). Occlusion of the aorta or common iliac arteries may also result in dilatation of the marginal and inferior mesenteric arteries, which become an important collateral supply to the legs via dilated middle rectal vessels arising from the internal iliac artery. Occlusion of the inferior mesenteric artery does not always result in irreversible ischaemia of the descending and sigmoid colon, because the marginal artery of the colon usually receives an adequate supply from the left branch of the middle colic artery. Moreover, the sigmoid arteries may be supplied by the superior rectal artery, which anastomoses with the middle and inferior rectal arteries. When ischaemia does occur, it is usually maximal in the proximal descending colon because this region is furthest from the collateral arterial supplies. Vascular ligation in colonic resections
During resection of the descending and sigmoid colon, ligation of the inferior mesenteric artery close to its origin preserves the bifurcation of the left colic artery. This allows continued flow in the left colic artery to the proximal descending colon supplied by flow from the middle colic artery via the marginal artery. Less radical resection, involving ligation of the left colic artery close to its bifurcation, may interfere with or obliterate this supply and render the descending colon more likely to become ischaemic. The same is true for ligation of the left colic vein. If the inferior mesenteric vein is ligated, then the bifurcation of the vein forms the route of venous drainage for the descending colon to the middle colic vein territory. Ligation of the branches separately will impair the venous drainage.
VEINS The venous drainage of the large intestine is primarily into the hepatic portal vein via the superior mesenteric and inferior mesenteric veins, although a small amount of drainage from the rectum occurs via middle rectal veins into the internal iliac vein and via inferior rectal veins into the pudendal vein. Those parts of the colon derived from the midgut (caecum, appendix, ascending colon and right two-thirds of the transverse colon) drain into colic branches of the superior mesenteric vein, whilst hindgut derivatives (left part of the transverse, descending and sigmoid colon, rectum and upper anal canal) drain into the inferior mesenteric vein. page 1181 page 1182
Figure 76.11 Relations and main branches of the superior and inferior mesenteric arteries.
Figure 76.12 Digital subtraction arteriogram of the marginal artery running parallel to the colon and anastomosing with the branches of the superior mesenteric artery supplying the right side of the colon. (By kind permission from Dr J Jackson, Hammersmith Hospital, London.)
Superior mesenteric vein
The superior mesenteric vein receives middle colic, right colic and ileocolic veins.
Venous blood from the wall of the caecum, appendix, ascending colon and right two-thirds of the transverse colon drains into mesenteric arcades and subsequently into segmental veins, which accompany their respective arteries. The segmental veins drain into the superior mesenteric vein, which lies to the right of the mesenteric artery. The veins tend to follow variations in arterial drainage. Inferior mesenteric vein
The inferior mesenteric vein drains the rectum, sigmoid, descending and distal transverse colon (Figs 76.13, 76.14). It begins as the superior rectal vein, from the rectal plexus, through which it connects with middle and inferior rectal veins. The superior rectal vein leaves the pelvis and crosses the left common iliac vessels medial to the left ureter with the superior rectal artery, and continues upwards as the inferior mesenteric vein. The inferior mesenteric vein lies to left of the inferior mesenteric artery, ascending deep to the peritoneum and anterior to the left psoas major. It may cross the testicular or ovarian vessels or ascend medial to them, and then passes above, or behind, the duodenojejunal flexure. It usually drains into the splenic vein, but occasionally drains into the confluence of the splenic and superior mesenteric veins or directly into the superior mesenteric vein. If a duodenal or paraduodenal fossa exists, the vein is usually in its anterior wall. The inferior mesenteric vein receives tributaries from several sigmoid veins, the middle and the left colic veins.
LYMPHATICS Lymphatic vessels of the caecum, ascending and proximal transverse colon drain ultimately into lymph nodes related to the superior mesenteric artery, while those of the distal transverse colon, descending colon, sigmoid colon and rectum drain into nodes following the course of the inferior mesenteric artery (Fig. 76.15). In cases where the distal transverse colon or splenic flexure is predominantly supplied by vessels from the middle colic artery, the lymphatic drainage of this area may be predominantly to superior mesenteric nodes. Colic nodes page 1182 page 1183
Figure 76.13 Digital subtraction arteriogram showing A, the inferior mesenteric artery and its branches and B, the inferior mesenteric vein and its tributaries. (By kind permission from Dr Adam Mitchell, Charing Cross Hospital, London.)
Figure 76.14 Relations and main branches of the superior and inferior mesenteric veins.
Lymph nodes related to the colon form four groups, namely epicolic, paracolic, intermediate colic and preterminal colic nodes. Epicolic nodes are minute nodules on the serosal surface of the colon, sometimes in the appendices epiploicae. Paracolic nodes lie along the medial borders of the ascending and descending colon and along the mesenteric borders of the transverse and sigmoid colon. Intermediate colic nodes lie along the named colic vessels (the ileocolic, right colic, middle colic, left colic, sigmoid and superior rectal arteries). Preterminal colic nodes lie along the main trunks of the superior and inferior mesenteric arteries and drain into para-aortic nodes at the origin of these vessels. These are
commonly referred to as the highest nodes of the territory which they drain. Lymph node clearance in colorectal cancer resections page 1183 page 1184
Figure 76.15 The lymph vessels and nodes of the transverse, descending and sigmoid colon. (After Jamieson JK, Dobson JF 1908 The lymphatics of the colon. Proc R Soc Med 2: 149-174, by permission from the Royal Society of Medicine.)
Radical lymphadenectomy during resection for colorectal cancer requires removal of the highest possible lymph node draining the area of colon in which the tumour is located. In cases of cancer involving the rectum and sigmoid colon, this usually involves resection of the preterminal colic nodes of the inferior mesenteric artery and thus ligation of the inferior mesenteric artery at its root or just below the origin of the left colic artery. A detailed description of the classification of the lymph nodes with regard to the site of the primary tumour within the colon has been suggested by the Japanese Society for the Cancer of the Colon and Rectum.
© 2008 Elsevier
INNERVATION The colon and rectum are innervated by the sympathetic and parasympathetic systems (Figs 76.16, 76.17). The sympathetic supply to the caecum, appendix, ascending colon and right twothirds of the transverse colon (derivatives of the midgut) originates in the fifth to the twelfth thoracic spinal segments. The preganglionic sympathetic axons which arise from these segments, and which are destined to influence the gut wall, do not synapse locally in the sympathetic chain, but instead are conveyed to the coeliac plexus via the greater and lesser splanchnic nerves, and synapse on ganglionic neurones in the coeliac and superior mesenteric plexuses. Postganglionic axons derived from neurones in these plexuses travel along branches of the superior mesenteric artery, and are distributed to the walls of the colon from periarterial plexuses. The parasympathetic supply is derived from the vagus nerve via the coeliac and superior mesenteric plexuses. The sympathetic supply of the left third of the transverse colon, the descending and sigmoid colon, rectum and upper anal canal - derivatives of the hindgut originates in the lumbar and upper sacral spinal segments. The fibres are distributed via the lumbar splanchnic nerves through the abdominal aortic and the inferior mesenteric plexuses and via the sacral splanchnic nerves through the superior and inferior hypogastric plexuses. Postganglionic axons reach the gut wall via periarterial plexuses on branches of the inferior mesenteric artery. The parasympathetic supply travels via the pelvic splanchnic nerves (nervi erigentes) from cell bodies in the second to the fourth sacral spinal segments. Those distributed to the rectum and upper anal canal run through the inferior and superior hypogastric plexuses to branches of the inferior mesenteric artery. Some of those distributed to the descending and sigmoid colon run via these plexuses; however, a large number of fibres pass directly through the retroperitoneal tissues to the splenic flexure, descending and sigmoid colon independently of the inferior mesenteric artery. The ultimate distribution within the wall of the large intestine is similar to the small intestine. The colic sympathetic nerves are motor to the ileocaecal valve musculature and inhibitory to mural muscle in the colon and rectum. Some fibres are vasoconstrictor to the colic vasculature. Parasympathetic nerves are motor to the colic and rectal musculature and inhibitory to the internal anal sphincter. Afferent impulses mediating sensations of distension are carried by visceral afferent fibres which travel with the parasympathetic nerves; pain impulses pass in visceral afferents travelling with the sympathetic and parasympathetic nerves supplying the rectum and the upper part of the anal canal.
SUPERIOR MESENTERIC PLEXUS The superior mesenteric plexus lies in the preaortic connective tissue around the origin of the superior mesenteric artery posterior to the pancreas. It is an inferior continuation of the coeliac plexus, and includes branches from the right vagus nerve and the coeliac plexus. Its branches accompany the superior mesenteric artery into the mesentery, and it divides into secondary plexuses which are distributed along the branches of the artery. The superior mesenteric ganglion lies superiorly in the plexus, usually above the origin of the superior mesenteric artery.
ABDOMINAL AORTIC PLEXUS (INTERMESENTERIC PLEXUS) page 1184 page 1185
Figure 76.16 Schematic diagram of the autonomic plexuses of the colon and rectum.
The abdominal aortic plexus lies on the sides and front of the aorta, between the origins of the superior and inferior mesenteric arteries. It consists of 4-12 intermesenteric nerves, which are connected by oblique branches. It is continuous above with the coeliac plexus and below with the superior hypogastric plexus. It is formed by parasympathetic and sympathetic branches from the coeliac plexus and receives rami from the first and second lumbar splanchnic nerves which contain sympathetic fibres. It is connected with the testicular, inferior mesenteric, iliac and superior hypogastric plexuses.
INFERIOR MESENTERIC PLEXUS The inferior mesenteric plexus lies around the origin of the inferior mesenteric artery and is distributed along its branches. It is formed predominantly from the aortic plexus and the first and second lumbar splanchnic nerves (sympathetic fibres), but it also receives connections from the superior hypogastric plexus (sympathetic and parasympathetic fibres).
SUPERIOR HYPOGASTRIC PLEXUS The superior hypogastric plexus lies anterior to the aortic bifurcation, the left common iliac vein, medial sacral vessels, fifth lumbar vertebral body and sacral promontory and between the common iliac arteries. It is often termed the presacral nerve, but is seldom a single nerve and it is prelumbar rather than presacral. Most frequently found to the left side of the midline, it lies in extraperitoneal connective tissue from which the parietal peritoneum can easily be stripped. The breadth and condensation of its constituent nerves vary. The attachment of the sigmoid mesocolon, containing the superior rectal vessels, is anterior and to the left of the lower part of the plexus. The superior hypogastric plexus is formed by branches from the aortic plexus and the third and fourth lumbar splanchnic nerves (which are mainly sympathetic). It may also contain parasympathetic fibres from the pelvic splanchnic nerves - which ascend from the two inferior hypogastric plexuses - via a series of filaments sometimes identified as the right and left hypogastric 'nerves'. The latter lie in loose connective tissue just posterolateral to the start of the mesorectum and pass over the pelvic brim
medial to the internal iliac vessels. The superior hypogastric plexus supplies fibres to the inferior mesenteric plexus and to the ureteric, testicular, ovarian and common iliac plexuses. Preganglionic sympathetic axons originate in the lower three thoracic and upper two lumbar spinal segments; they synapse in the ganglia associated with the lumbar and sacral sympathetic trunk, or in the lower part of the aortic, superior or inferior hypogastric plexuses. Preganglionic parasympathetic axons originate from neurones in the second to fourth sacral spinal segments.
INFERIOR HYPOGASTRIC (PELVIC) PLEXUSES The inferior hypogastric plexus lies in the thin extraperitoneal connective tissue lateral to the mesorectum. Laterally lie the internal iliac vessels, and attachments of levator ani, coccygeus and obturator internus; superiorly lie the superior vesical and obliterated umbilical arteries; posteriorly lie the sacral and coccygeal plexuses. In males the inferior hypogastric plexus lies posterolaterally on either side of the seminal vesicles, prostate and the posterior part of the urinary bladder. In females each plexus lies lateral to the uterine cervix, vaginal fornix and the posterior part of the urinary bladder, and often extends into the broad ligaments of the uterus. page 1185 page 1186
Figure 76.17 Anatomic illustration of the autonomic plexuses of the colon and rectum.
The inferior hypogastric plexus is formed from the right and left hypogastric 'nerves' - which are mainly sympathetic - and long branches of the parasympathetic pelvic splanchnic nerves (S2, 3, 4). Branches from the lowest lumbar splanchnic nerve and sacral splanchnic nerves may also join the plexus via the hypogastric 'nerves'. In males the plexus supplies the vas deferens, seminal vesicles, prostate, accessory glands and penis. In females it supplies the ovary, fallopian tubes, uterus, uterine cervix and vagina. It supplies the urinary
bladder and distal ureter in both sexes. Preganglionic sympathetic axons originate in the lower thoracic and upper two lumbar spinal segments; they synapse in the ganglia associated with the lumbar and sacral sympathetic trunk, or in the lower part of the aortic, superior or inferior hypogastric plexuses. Preganglionic parasympathetic axons originate from neurones in the second to fourth sacral spinal segments. REFERENCES Fenlon HM 2002 Virtual colonoscopy. Br J Surg 89(1): 1-3. Describes and reviews the use of multislice CT to image the colon. Fisher DF Jr, Fry WJ 1987 Collateral mesenteric circulation. Surg Gynecol Obstet 164(5): 487-92. Reviews collateral mesenteric circulations that develop during disease processes. Medline Similar articles Jackson JE 1999 Vascular anatomy of the gastrointestinal tract. In: Butler P, Mitchell AWM, Ellis H (eds) Applied Radiological Anatomy. Cambridge: Cambridge University Press. Japanese Society for the Cancer of the Colon and Rectum 1997. Japanese Classification of Colorectal Carcinoma. Tokyo: Kanehara & Co Ltd. Describes the topography of colonic mesenteric lymph nodes with particular reference to radical excision of carcinoma. Oliphant M, Berne AS, Meyers MA 1996 The subperitoneal space of the abdomen and pelvis: planes of continuity. Am J Roentgenol 167(6): 1433-9. Silverstein FE, Tytgat GNJ 1991 Atlas of Gastrointestinal Endoscopy, 2nd edn. London: Gower Medical Publishing.
© 2008 Elsevier
77 GASTROINTESTINAL TRACT - LARGE INTESTINE Caecum The caecum is a large blind pouch of large intestine lying in the right iliac fossa below the ileocaecal valve and continuing distally as the ascending colon. The blind-ending vermiform appendix usually arises on its medial side at the level of the ileal opening. Its average axial length is c.6 cm and its breadth c.7.5 cm. It rests posteriorly on the right iliacus and psoas major, with the lateral cutaneous nerve of the thigh interposed. Posteriorly lies the retrocaecal recess which frequently contains the vermiform appendix. The anterior abdominal wall is immediately anterior to the caecum except when it is empty, when the greater omentum and some loops of the small intestine may be interposed. Usually the caecum is entirely covered by peritoneum, but occasionally this is incomplete posterosuperiorly where it lies attached to the iliac fascia by loose connective tissue. In early fetal life the caecum is usually short, conical and broad at the base, with an apex turned superomedially towards the ileocaecal junction. As the fetus grows, the caecum increases in length more than in breadth, to form a longer tube with a narrower base but retaining the same inclination. Distal growth later ceases, but the proximal part continues to grow in breadth, so that at birth a narrow vermiform appendix extends from the apex of a conical caecum. This infantile form persists throughout life in only a very small percentage of individuals. Occasionally the conical caecum takes on a quadrate shape as a result of the outgrowth of a saccule on each side of the anterior taenia: the saccules are of equal size and the appendix arises from the depression between them instead of from the apex of a cone. In the normal adult form, the right saccule grows more rapidly than the left, forming a new 'apex'. The original apex, with the appendix attached, is pushed towards the ileocaecal junction. The caecum commences the process of fluid and electrolyte reabsorption, which occurs to a large extent in the ascending and transverse colon. The distensible nature and 'sac-like' morphology of the caecum are adaptations for the storage of larger volumes of semi-liquid chyme entering from the small bowel via the ileocaecal valve.
© 2008 Elsevier
ILEOCAECAL VALVE The ileum opens into the posteromedial aspect of the large intestine at the junction of the caecum and colon. The ileocaecal orifice has a so-called 'valve', consisting of two flaps which project into the lumen of the large intestine (Figs 77.1, 77.2). The precise shape and form of the valve varies but in the distended, fixed, caecum the flaps are often semilunar. The upper flap, approximately horizontal, is attached to the junction of the ileum and colon; the lower flap is longer and more concave, and is attached to the junction of the ileum and caecum. At their ends the flaps fuse, continuing as narrow membranous ridges, the frenula of the valve. The orifice may appear in many different shapes depending on the state of contraction or distension of the caecum: it is commonly either a slit or an oval. The margin of the ileocaecal valve is a reduplication of the intestinal mucosa and circular muscle. Longitudinal muscle fibres are partly reduplicated as they enter the valve, but the more superficial fibres and the peritoneum continue from the small to the large intestine without interruption. The valve may prevent reflux of chyme from the caecum to the ileum, and may slow the passage of ileal contents into the caecum when the circular muscle of the valve is contracted by sympathetic stimulation. Although circular and longitudinal muscle layers of the terminal ileum continue into the valve, there is little evidence that it constitutes a true functional sphincter. The ileal valvular surfaces are covered with villi and have the structure of the small intestinal mucosa, whereas their caecal surfaces have no villi but display numerous orifices of tubular glands peculiar to the colonic mucosa.
Figure 77.1 Double contrast barium enema appearance of the caecum and ileocaecal valve.
Figure 77.2 Endoscopic appearance of the ileocaecal valve.
© 2008 Elsevier
VASCULAR SUPPLY AND LYMPHATIC DRAINAGE ILEOCOLIC ARTERY page 1187 page 1188
Figure 77.3 The arteries of the caecum and vermiform appendix.
The caecum is supplied principally from the ileocolic artery, which is the last branch from the right side of the superior mesenteric artery (Fig. 77.3). It descends to the right beneath the parietal peritoneum into the right iliac fossa, where it divides into two branches: the superior branch anastomoses with the right colic artery, and the inferior branch anastomoses with the end of the superior mesenteric artery. The ileocolic artery crosses anterior to the right ureter, testicular or ovarian vessels, and psoas major. The inferior branch approaches the superior border of the ileocolic junction. It has the following named branches: ascending (colic) artery (which passes up on the ascending colon); anterior and posterior caecal arteries; appendicular artery (which descends behind the terminal ileum to enter the mesoappendix and gives off a recurrent branch which anastomoses with a branch of the posterior caecal artery); ileal artery (which ascends to the left on the lower ileum, supplies it and anastomoses with the terminal ileal arcade arteries).
ILEAL ARTERIES The terminal arcades of the ileal arteries provide a collateral supply to the caecum via anastomoses with the ileal branch of the ileocolic artery.
ILEOCOLIC VEIN The ileocolic vein is formed from superior and inferior tributaries and ascends alongside the ileocolic artery beneath the peritoneum of the ileocaecal mesentery to drain into the superior mesenteric vein. The inferior tributary receives appendicular, anterior and posterior caecal and ileal veins and the superior tributary drains the ascending colic veins.
LYMPHATIC DRAINAGE
Figure 77.4 The lymph vessels and nodes of the caecum and vermiform appendix: anterior aspect. (Modified with permission from Elsevier: The Lancet, 1907, 1, 10611066.)
Anterior lymphatic vessels pass in front of the caecum and drain to the anterior ileocolic nodes and nodes of the ileocolic chain; posterior vessels ascend behind the caecum to the posterior and inferior ileocolic nodes. Lymph drains from the nodes in the ileocolic chain into the superior mesenteric nodes in the root of the small bowel mesentery (Fig. 77.4).
© 2008 Elsevier
INNERVATION The caecum is innervated by sympathetic and parasympathetic nerves via the superior mesenteric plexus.
© 2008 Elsevier
CAECAL VOLVULUS Because the caecum and ascending colon may possess a mesentery to a variable degree, it is possible for the caecum (and lower portion of the ascending colon) to rotate about their mesenteric attachment (mesentero-axial volvulus) such that the lower pole of the caecum comes to lie in the left or right upper quadrant of the abdomen. The fixed apex of this rotation is formed by the attachment of the caecal mesentery to the retroperitoneum. Volvulus is extremely unlikely in individuals where the caecum and ascending colon possess a short broadly attached mesentery. Volvulus occurs most commonly in those individuals where a large portion of the ascending colon lies on a mesentery and the common origin of the caecal and ascending colic mesentery is narrow. True isolated volvulus of the caecum is extremely rare.
© 2008 Elsevier
78 GASTROINTESTINAL TRACT - LARGE INTESTINE Vermiform appendix The vermiform appendix is a narrow, vermian (worm-shaped) tube which arises from the posteromedial caecal wall, c.2 cm below the end of the ileum. It may occupy one of several positions (Fig. 78.1). Thus it may be retrocaecal, retrocolic (behind the caecum or lower ascending colon respectively), pelvic or descending (when it hangs dependently over the pelvic brim, in close relation to the right uterine tube and ovary in females). These are the commonest positions seen in clinical practice. Other positions are occasionally seen especially when there is a long appendix mesentery allowing greater mobility. These include subcaecal (below the caecum); preilial (anterior to the terminal ileum); postileal (behind the terminal ileum). The three taeniae coli on the ascending colon and caecum converge on the base of the appendix, and merge into its longitudinal muscle. The anterior caecal taenia is usually distinct and can be traced to the appendix, which affords a guide to its location in clinical practice. The appendix varies from 2 to 20 cm in length: it is often relatively longer in children and may atrophy and shorten after mid-adult life. It is connected by a short mesoappendix to the lower part of the ileal mesentery. This fold is usually triangular, extending almost to the appendicular tip along the whole viscus. The lumen of the appendix is small and opens into the caecum by an orifice lying below and slightly posterior to the ileocaecal opening. The orifice is sometimes guarded by a semilunar mucosal fold forming a valve (Fig. 78.2). The lumen may be widely patent in early childhood and is often partially or wholly obliterated in the later decades of life. The appendix usually contains numerous patches of lymphoid tissue although these tend to decrease in size from early adulthood.
Figure 78.1 Diagram illustrating the major positions of the appendix encountered at surgery or postmortem.
Figure 78.2 Endoscopic appearance of the appendix orifice. The orifice varies from a small depression to an obvious lumenal structure.
© 2008 Elsevier
VASCULAR SUPPLY AND LYMPHATIC DRAINAGE APPENDICULAR ARTERY The main appendicular artery, a branch from the lower division of the ileocolic artery, runs behind the terminal ileum and enters the mesoappendix a short distance from the appendicular base. Here it gives off a recurrent branch, which anastomoses at the base of the appendix with a branch of the posterior caecal artery: the anastomosis is sometimes extensive. The main appendicular artery approaches the tip of the organ, at first near to, and then in the edge of, the mesoappendix. The terminal part of the artery lies on the wall of the appendix and may be thrombosed in appendicitis, which results in distal gangrene or necrosis. Accessory arteries are common, and many individuals possess two or more arteries of supply.
APPENDICULAR VEINS The appendix is drained via one or more appendicular veins into the posterior caecal or ileocolic vein and thence into the superior mesenteric vein.
LYMPHATICS Lymphatic vessels in the appendix are numerous: there is abundant lymphoid tissue in its walls. From the body and apex of the appendix 8-15 vessels ascend in the mesoappendix, and are occasionally interrupted by one or more nodes. They unite to form three or four larger vessels which run into the lymphatic vessels draining the ascending colon, and end in the inferior and superior nodes of the ileocolic chain.
© 2008 Elsevier
INNERVATION page 1189 page 1190
The appendix and overlying visceral peritoneum are innervated by sympathetic and parasympathetic nerves from the superior mesenteric plexus. Visceral afferent fibres carrying sensations of distension and pressure mediate the symptoms of 'pain' felt during the initial stages of appendicular inflammation. In keeping with other structures derived from the midgut, these sensations are poorly localized initially, and referred to the central (periumbilical) region of the abdomen. It is not until parietal tissues adjacent to the appendix become involved in any inflammatory process that somatic nociceptors are stimulated, and there is an associated change in the nature and localization of pain.
© 2008 Elsevier
ACUTE APPENDICITIS The genesis of acute appendicitis varies between individuals. It may follow obstruction of the lumen, when it is a consequence of increased intraluminal pressure and retention of (infected) contents which allows acute suppuration to occur. The size of the orifice of the appendix in some individuals may contribute to the risk that this will happen. Appendicoliths may be visible on plain radiographs in 7-15% of the normal population: in those patients with acute abdominal pain, their presence indicates a high risk of appendicitis being present. The increased size of the orifice and lumen at the extremes of life may be a reason why acute appendicitis is relatively uncommon in these age groups. Acute appendicitis may also present as a primary suppuration of the tissues of the appendix itself: the reduction in appendicular lymphoid tissue that occurs in later life may be another reason why the disease is infrequent in the elderly. Although the appendix is well supplied by arterial anastomoses at its base, the appendicular artery is an end artery from the midpoint upwards and its close proximity to the wall makes it susceptible to thrombosis during episodes of acute inflammation. This may render the distal appendix ischaemic and explains the frequency of gangrenous perforation seen in the disease. REFERENCE Buschard K, Kjaeldgaard A 1973 Investigations and analysis of the positions, fixation, length and embryology of the vermiform appendix. Acta Chir Scand 139: 293-8. Medline Similar articles
© 2008 Elsevier
79 GASTROINTESTINAL TRACT - LARGE INTESTINE Ascending colon The ascending colon is c.15 cm long and narrower than the caecum. It ascends to the inferior surface of the right lobe of the liver, on which it makes a shallow depression, and then turns abruptly forwards and to the left, at the hepatic flexure. It is a retroperitoneal structure covered anteriorly and on both sides by peritoneum. Its posterior surface is connected by loose connective tissue to the iliac fascia, the iliolumbar ligament, the quadratus lumborum muscles, the aponeurosis of transversus abdominis, and the anterior peri-renal fascia inferolateral to the right kidney. The lateral femoral cutaneous nerve, usually the fourth lumbar artery, and sometimes the ilioinguinal and iliohypogastric nerves, lie posteriorly as they cross the quadratus lumborum muscles. Laterally the peritoneum forms the paracolic gutter. The ascending colon possesses a narrow mesocolon for part of its course in up to one-third of cases. Anteriorly it is in contact with loops of ileum, the greater omentum and the anterior abdominal wall.
© 2008 Elsevier
HEPATIC FLEXURE The hepatic flexure forms the junction of the ascending and transverse colon as the latter turns down, forwards and to the left. It is variable in position and usually has a less acute angle than the splenic flexure. The anterior surface of the lower pole of the right kidney is posterior and the right lobe of the liver is superior and anterolateral. The descending (second) part of the duodenum is medial and the fundus of the gallbladder is anteromedial. The posterior aspect of the hepatic flexure is not covered by peritoneum and is in direct contact with renal fascia.
© 2008 Elsevier
VASCULAR SUPPLY AND LYMPHATIC DRAINAGE ARTERIES Ileocolic artery - ascending branch
The ascending branch of the ileocolic artery supplies the lower half of the ascending colon and anastomoses freely with the right colic artery. Right colic artery
The right colic artery is a small vessel and may be absent. When present, it arises near the middle of the superior mesenteric artery, or in common with the middle colic artery. It passes towards the ascending colon, deep to the parietal peritoneum and anterior to the right gonadal vessels, right ureter and psoas major. Sometimes it arises more superiorly and crosses the descending duodenum and inferior pole of the right kidney. Near the colon it divides into a descending branch, which anastomoses with the ileocolic artery, and an ascending branch which anastomoses with the middle colic arteries. These form arches from which vessels are distributed to the upper two-thirds of the ascending colon, and to the hepatic flexure.
VEINS
page 1191 page 1192
Figure 79.1 The relationship of the ascending colon and hepatic flexure to right lobe of liver, kidneys, second part of duodenum, and gallbladder in A axial and B coronal reformat CT images. (By kind permission from Dr Louise Moore, Chelsea and Westminster Hospital, London.)
Ascending tributaries of the ileocolic and right colic veins accompany their respective arteries into the root of the mesentery and drain via the ileocolic vein into the superior mesenteric vein. Although the right colic vein usually drains into the superior mesenteric vein, it is occasionally absent or may join the right gastroepiploic or inferior pancreaticoduodenal vein to form a 'gastrocolic trunk' which drains into the superior mesenteric vein.
LYMPHATICS Lymphatic vessels originate from both anterior and posterior aspects of the colon and drain into nodes located along the ascending branch of the ileocolic and the right colic arteries. Lymphatic drainage from the distal ascending colon and hepatic flexure may be predominantly to the nodes of the right colic artery. The lymphatic anastomoses are rich and the preterminal nodes for both routes of drainage are the ileocolic nodes which are located close to the superior mesenteric artery.
© 2008 Elsevier
INNERVATION The ascending colon is innervated by sympathetic and parasympathetic nerves via the superior mesenteric plexus. REFERENCE Yamaguchi S, Kuroyanagi H, Milson JW, Sim R, Shimada H 2002 Venous anatomy of the right colon: precise structure of the major veins and gastrocolic trunk in 58 cadavers. Dis Colon Rectum 45: 1337-40. Medline Similar articles Full article
© 2008 Elsevier
80 GASTROINTESTINAL TRACT - LARGE INTESTINE Transverse colon The transverse colon is c.50 cm long, and extends from the hepatic flexure in the right lumbar region across into the left hypochondriac region, where it curves posteroinferiorly below the spleen as the splenic flexure. It is highly variable in length and position, as may be confirmed by radiological assessment, but it often describes an inverted arch, with its concavity directed posteriorly and superiorly. Near the splenic flexure an abrupt U-shaped curve may descend lower than the main arch. The posterior surface at the hepatic flexure is devoid of peritoneum and is attached by loose connective tissue to the front of the descending part of the duodenum and the head of the pancreas. The transverse colon from here to the splenic flexure is almost completely invested by peritoneum, and is suspended from the anterior border of the body of the pancreas by the transverse mesocolon. The latter is attached from the inferior part of the right kidney, across the second part of the duodenum and pancreas, to the inferior pole of the left kidney. The transverse colon hangs down between the flexures to a variable extent, and sometimes reaches the pelvis. Above it are the liver and gallbladder, the greater curvature of the stomach and the body of the spleen. The transverse colon is usually attached to the greater curvature of the stomach by the gastrocolic ligament, which is in continuity with the greater omentum, lying anteriorly and extending inferiorly (p. 1127). Behind and below the transverse colon lie the descending part of the duodenum, the head of the pancreas, the upper end of the small bowel mesentery, the duodenojejunal flexure and loops of the jejunum and ileum. The transverse mesocolon permits considerable mobility of the transverse colon: occasionally the colon may be interposed between the liver and the diaphragm (Chilaiditi syndrome), and may be mistaken for free intraperitoneal gas.
© 2008 Elsevier
SPLENIC FLEXURE The splenic flexure forms the junction of the transverse and descending colon, and lies in the left hypochondriac region anteroinferior to the lower part of the spleen and anterior to the pancreatic tail. The left kidney lies behind and lateral to it (Fig. 80.1). The splenic flexure often adopts a very acute angle such that the end of the transverse colon overlaps the beginning of the descending colon. It lies more superiorly and posteriorly than the right hepatic flexure, and is attached to the diaphragm at the level of the tenth and eleventh ribs by the phrenicocolic ligament (p. 1127) which lies below the anterolateral pole of the spleen.
© 2008 Elsevier
VASCULAR SUPPLY AND LYMPHATIC DRAINAGE ARTERIES The proximal two-thirds of the transverse colon is supplied by the superior mesenteric artery via the middle colic artery. The distal third is usually supplied by the ascending branch of the left colic artery via the marginal artery of the colon, although this is somewhat variable. Middle colic artery
Figure 80.1 Relations of the splenic flexure.
The middle colic artery leaves the superior mesenteric artery just inferior to the pancreas and anterior to the third part of the duodenum. Initially it passes inferiorly, then turns to run anteriorly and superiorly within the transverse mesocolon, where it usually divides into a right and left branch. The right branch anastomoses with the right colic artery, and the left branch anastomoses with the
left colic artery. The arterial arches thus formed lie 3 or 4 cm from the transverse colon, which they supply. Sometimes the middle colic artery divides into three or more branches within the transverse mesocolon, in which case the most lateral branches form the arterial anastomoses.
VEINS Several tributaries drain into one or more middle colic veins. The middle colic veins drain either into the superior mesenteric vein, just before its junction with the splenic vein, or directly into the hepatic portal vein.
LYMPHATICS Lymph vessels drain into nodes along the middle colic arteries and then into the superior mesenteric nodes. The predominant lymphatic drainage of the splenic flexure is usually via nodes along the left colic artery which drain into the inferior mesenteric nodes: this arrangement is dependent on the arterial supply to the distal third of the transverse colon.
© 2008 Elsevier
INNERVATION page 1193 page 1194
The proximal two-thirds of the transverse colon is innervated by sympathetic and parasympathetic nerves via the superior mesenteric plexus. The distal third usually receives a sympathetic supply from the inferior mesenteric plexus and a parasympathetic supply from this plexus and from retroperitoneal fibres which travel in the pelvic splanchnic nerves from neurones in the second, third and fourth sacral segments (p. 1178). REFERENCE Murphy JM, Maibaum A, Alexander G, Dixon AK 2000 Chilaiditi's syndrome and obesity. Clin Anat 13(3): 181-4. Medline Similar articles Full article
© 2008 Elsevier
81 GASTROINTESTINAL TRACT - LARGE INTESTINE Descending colon The descending colon is c.25 cm long. It descends through the left hypochondriac and lumbar regions, initially following the lateral border of the lower pole of the left kidney, and then descending in the angle between psoas major and quadratus lumborum to the iliac crest. It then curves inferomedially, lying anterior to iliacus and psoas major, to become the sigmoid colon at the inlet of the lesser pelvis. It is a retroperitoneal structure covered anteriorly and on both sides by peritoneum. Its posterior surface is separated by loose connective tissue from the anterior perirenal fascia inferolateral to the left kidney, the aponeurosis of transversus abdominis, quadratus lumborum, iliacus and psoas major. The subcostal vessels and nerves, iliohypogastric and ilioinguinal nerves, fourth lumbar artery (usually), the lateral femoral cutaneous, femoral and genitofemoral nerves, the gonadal vessels and the external iliac artery all pass behind the descending colon. Loops of jejunum lie anteriorly: if the anterior abdominal walls are relaxed, the most inferior part of the descending colon may be directly palpated transabdominally. The descending colon is smaller in calibre, more deeply placed, and more frequently covered posteriorly by peritoneum, than the ascending colon.
© 2008 Elsevier
VASCULAR SUPPLY AND LYMPHATIC DRAINAGE ARTERIES The arterial supply of the descending colon is from the inferior mesenteric artery via its left colic branch, which also anastomoses with the marginal artery of the colon (in the region of the splenic flexure), and the sigmoid arteries (in the region of the junction with the sigmoid colon). Left colic artery
The left colic artery ascends anterior to the left psoas major, and divides into ascending and descending branches: this division can occur soon after its origin. The trunk and its branches cross the left ureter and gonadal vessels. The ascending branch passes anterior to the left kidney in the upper left retroperitoneum and anastomoses with the left branch of the middle colic artery in the subperitoneal space within the transverse mesocolon. The descending branch passes laterally in the retroperitoneum approaching the descending colon where it forms part of the marginal artery and anastomoses with the highest sigmoid artery. The arterial arches thus formed supply the distal third of the transverse and the descending colon.
VEINS The left colic vein is formed from several tributaries including ascending and descending branches which correspond to the equivalent arteries. These tributaries may not form a discrete vein until they drain into the inferior mesenteric vein, and occasionally there may be two distinct veins which both run into the inferior mesenteric vein. The left colic vein usually lies superior to its equivalent artery and has a shorter course because it ascends more steeply to drain into the inferior mesenteric vein.
LYMPHATICS Lymph vessels drain into nodes along the left colic artery and subsequently into the inferior mesenteric nodes.
© 2008 Elsevier
INNERVATION The descending colon receives a sympathetic supply from the inferior mesenteric plexus and a parasympathetic supply from this plexus and from retroperitoneal fibres which travel in the pelvic splanchnic nerves from neurones in the second, third and fourth sacral segments (p. 1178).
page 1195 page 1196
Figure 81.1 The vascular supply of descending colon from the inferior mesenteric artery via left colic artery with ascending and descending branches A axial and B coronal. CT reformat. (By kind permission from Dr Louise Moore, Chelsea and Westminster Hospital, London.)
© 2008 Elsevier
82 GASTROINTESTINAL TRACT - LARGE INTESTINE Sigmoid colon
Figure 82.1 Relations of the sigmoid colon.
The sigmoid colon begins at the pelvic inlet and ends at the rectum. Characteristically it forms a mobile loop which normally lies in the lesser pelvis. It is completely invested in peritoneum and is attached to the posterior pelvic wall and lower posterior abdominal walls by the fan-shaped sigmoid mesocolon. The root of the sigmoid mesocolon has an inverted 'V' attachment to the posterior abdominal wall. The sigmoid colon initially descends adjacent the left pelvic wall, but then comes to lie in an extremely variable position. It may remain folded principally in contact with the peritoneum overlying iliacus, or it may cross the pelvic cavity between the rectum and bladder in males, or the rectum and uterus in females, and it may even reach the right pelvic wall. If long, the sigmoid loop may rise out of the pelvis into the abdominal cavity and lie in contact with loops of ileum. The sigmoid loop ends in a relatively constant position lying just to the left of the midline at the level of the third sacral body, where it bends inferiorly and is continuous with the rectum. The sigmoid loop is fixed at its junctions with the descending colon and rectum but quite mobile between them. Its relations are therefore variable (Fig. 82.1). Laterally it is related to the left external iliac
vessels, the obturator nerve, ovary or vas deferens and the lateral pelvic wall. Posteriorly lie the left internal iliac vessels, gonadal vessels, ureter, piriformis and sacral plexus. Inferiorly are the bladder in males, or uterus and bladder in females, and superiorly and to the right the sigmoid colon is in contact with loops of the ileum. The gonadal vessels and ureter lie in a distinct fascial plane which is separate from the thin fascial coverings of the mesosigmoid. The fascial plane surrounding the sigmoid can be recognized during dissection of the mesentery of the sigmoid colon because it does not contain the numerous small vessels which are often present in the loose connective tissue surrounding the ureter and gonadal vessels. The position and shape of the sigmoid colon vary greatly, depending on its length; the length and mobility of its mesocolon; the degree of distension (when distended it rises into the abdominal cavity, and sinks again into the lesser pelvis when empty); and the condition of the rectum, bladder and uterus (the sigmoid colon tends to rise when these are distended, and to fall when they are empty). The length and diameter of the sigmoid colon also vary in different races.
SIGMOID VOLVULUS Rotation or volvulus around the mesenteric attachment of the sigmoid colon, similar to the rotation of the caecum around its mesentery, may occur. Only the sigmoid colon is affected, because the descending colon rarely possesses sufficient mesentery to become involved in the rotation. Rotation usually involves at least 270°. Volvulus is more common as the length of the sigmoid colon increases, producing a 'fan-like' mesentery with a short retroperitoneal attachment. Volvulus does not occur in individuals where the sigmoid colon is short or where the mesentery is short and its attachment runs over the brim of the pelvis. The combination of anatomical features predisposing to sigmoid volvulus is most commonly found in sub-Saharan Africans and chronically institutionalized patients.
DIVERTICULAR DISEASE The development of acquired diverticula occurs commonly in the sigmoid colon, particularly in white Caucasian populations. The aetiopathogenesis is probably multifactorial and includes dietary factors; however, the location of the diverticula would seem to be related to the underlying anatomy of the wall of the colon. Diverticula commonly occur midway between the antimesenteric and lateral taeniae, i.e. where the wall is potentially weak, not only because the circular muscle lacks the support of the longitudinal muscle, but also because it is traversed here by arteries as they access the submucosal vascular plexus. The predilection of the diverticula for the sigmoid colon probably relates to causative factors rather than intrinsic differences in the structure of its walls. In marked contrast, congenital diverticula may occur anywhere throughout the colon, and are often found adjacent to the mesenteric attachment of the colon.
© 2008 Elsevier
VASCULAR SUPPLY AND LYMPHATIC DRAINAGE ARTERIES Sigmoid (inferior left colic) arteries
There are between two and five sigmoid arteries, which are branches of the inferior mesenteric artery. They descend from the retroperitoneum obliquely within the subperitoneal space anterior to the left psoas major, ureter and gonadal vessels. Branches supply the lower descending colon and sigmoid colon, and anastomose superiorly with the left colic artery and inferiorly with the superior rectal artery (Fig. 82.2). Unlike the small intestine, arterial arcades do not form until the arteries are close to the colon wall. At this point small branches arise and anastomose with each other to form a marginal artery along the mesenteric border of the sigmoid colon. A significant space often exists in the mesentery between the highest sigmoid artery and the descending branch of the left colic artery. page 1197 page 1198
Figure 82.2 Details of sigmoid colon arterial supply.
VEINS Several sigmoid veins drain the sigmoid colon and run superiorly alongside their respective arteries to drain into the inferior mesenteric vein.
LYMPHATICS Lymphatic vessels drain into sigmoid nodes in the sigmoid mesocolon and join with superior rectal and left colic vessels to drain into the inferior mesenteric nodes.
© 2008 Elsevier
INNERVATION The sigmoid colon receives a sympathetic supply from the inferior mesenteric plexus and a parasympathetic supply from this plexus and from retroperitoneal fibres which travel in the pelvic splanchnic nerves from neurones in the second, third and fourth sacral segments (p. 1178).
© 2008 Elsevier
83 GASTROINTESTINAL TRACT - LARGE INTESTINE Rectum Although the rectum is continuous with the sigmoid colon, it has several features which distinguish it functionally from the rest of the colon. These features suit its specialized role in defaecation and continence in combination with the anal canal. The rectum is continuous with the sigmoid colon at the level of the third sacral vertebra and terminates at the upper end of the anal canal. It descends along the sacrococcygeal concavity as the sacral flexure of the rectum, initially inferoposteriorly and then inferoanteriorly to join the anal canal by passing through the pelvic diaphragm. The anorectal junction is 2-3 cm in front of and slightly below the tip of the coccyx, which is opposite the apex of the prostate in males. From this level the anal canal passes inferiorly and posteriorly from the lower end of the rectum. The posterior bend is termed the perineal flexure of the rectum and the angle it forms with the upper anal canal is termed the anorectal angle. The rectum also deviates in three lateral curves. The upper is convex to the right, the middle (the most prominent) bulges to the left, and the lower is convex to the right. Both ends of the rectum are in the median plane (Fig. 83.1). Although variable in absolute length, a common landmark used in clinical practice to define the rectum is a length of 15 cm above the external anal margin. It commences with a similar diameter to the sigmoid colon, more inferiorly it is dilated as the rectal ampulla. The rectum differs from the sigmoid colon in having no sacculations, appendices epiploicae, or mesentery. The taeniae blend c.5 cm above the rectosigmoid junction, forming two wide muscular bands which descend anteriorly and posteriorly in the rectal wall. These then fuse to form an encircling layer of longitudinal muscle, which invests the entire length of the rectum. At the rectal ampulla a few strands of the anterior longitudinal fibres pass forwards to the perineal body, as the musculus recto-urethralis. In addition, two fasciculi of smooth muscle may pass anteroinferiorly from the front of the second and third coccygeal vertebrae to blend with the longitudinal muscle fibres on the posterior wall of the anal canal, forming the rectococcygeal muscles. The upper third of the rectum is covered by peritoneum on its anterior and lateral aspects. It is related anteriorly to the sigmoid colon or loops of ileum if these lie in the pelvis, otherwise it is related to the urinary bladder in males or cervix and body of the uterus in females. The middle third of the rectum is covered by peritoneum only on the anterior aspect. The peritoneum is reflected superiorly onto the urinary bladder in males, to form the rectovesical pouch, or onto the posterior vaginal wall in females to form the recto-uterine pouch (pouch of Douglas). The level of this reflection is higher in males with the rectovesical pouch is c.7.5 cm (about the length of the index finger) from the anus. In females the recto-uterine pouch is c.5.5 cm from the anus. In the male neonate, peritoneum extends on to the front of the rectum as far as the lower limit of the prostate. Superiorly the peritoneum is firmly attached to the muscle layer of the sigmoid colon by fibrous connective tissue, but as it descends onto the rectum it is more loosely attached by fatty connective tissue, allowing for considerable expansion of
the upper half of the rectum (Figs 83.2, 83.3). There are no haustra in the rectum. When empty the mucosa forms a number of longitudinal folds in its lower part which become effaced during distension. In addition the rectum commonly has three (although the number can vary) permanent semilunar transverse or horizontal folds, most marked in rectal distension (Fig. 76.8). Two forms of horizontal fold have been recognized. One consists of the mucosa, a circular muscle layer and part of the longitudinal muscle, and is marked externally by an indentation. The other is devoid of longitudinal muscle and has no external marking. The most superior fold at the beginning of the rectum may be either on the left or right and occasionally encircles the rectal lumen. The middle fold is largest and most constant. It lies immediately above the rectal ampulla, projecting from the anterior and right wall just below the level of the anterior peritoneal reflection. The circular muscle is more marked in this fold than in the others. The most inferior and variable fold is found on the left c.2.5 cm below the middle fold. Sometimes a fourth fold is found on the left c.2.5 cm above the middle fold.
page 1199 page 1200
Figure 83.1 Coronal T2-weighted MRI of the rectum.
Figure 83.2 Sagittal T2-weighted MRI of the rectum in the male.
page 1200 page 1201
Figure 83.3 Sagittal T2-weighted MRI of the rectum in the female.
MESORECTUM, RECTAL FASCIAE AND 'SPACES' The mesorectum (mesentery of the rectum) and its contents are intimately related to the rectum down to the level of levator ani (Figs 83.4, 83.5, 83.6).
Figure 83.4 Axial T2-weighted MRI of the upper rectum in the female.
Figure 83.5 Axial T2-weighted MRI of the mid rectum below the peritoneal reflection in
Figure 83.5 Axial T2-weighted MRI of the mid rectum below the peritoneal reflection in a male.
Figure 83.6 Axial T2-weighted MRI of the low rectum below the peritoneal reflection in a female.
The mesorectum is a distinct compartment that derives from the embryological hindgut. It contains the superior rectal artery and its branches, the superior rectal vein and tributaries, the lymphatic vessels and nodes along the superior rectal artery, and branches from the inferior mesenteric plexus which descend to innervate the rectum and loose adipose connective tissue.
page 1201 page 1202
The mesorectum is enclosed by the mesorectal fascia which is a distinct covering derived from the visceral peritoneum. It is also known as the visceral fascia of the mesorectum, fascia propria of the rectum or the presacral wing of the hypogastric sheath. The fascia bounds the mesorectum posteriorly and thus lies anterior to the retrorectal space and the presacral fascia. The mesorectum and its fascia are surrounded by loose areolar tissue which separates them from the posterior and lateral walls of the true pelvis. Superiorly, the mesorectal fascia blends with the connective tissue bounding the sigmoid mesentery. Laterally, the mesorectal fascia extends around the rectum and becomes continuous with a denser condensation of fascia anteriorly. In males this anterior fascia is known as the rectovesical fascia of Denonvilliers. In females it forms the fascia of the rectovaginal septum. On MRI scanning, the mesorectum is seen as a fat-containing envelope in which vessels are depicted as low signal due to signal void produced by blood flow. Lymph nodes appear as high signal ovoid structures. Small nerves within the mesorectum are not visualized, but interlacing connective tissue within the mesorectum can be seen as low signal intensity strands. The mesorectal fascia is demonstrated on axial views as a low signal layer surrounding the mesorectum, which corresponds to the distinct condensation of fascia seen on histological sections containing the mesorectum (Brown et al 1999) (Fig. 83.4). Identification of the involvement of this layer by malignant tumours of the rectum on MRI scanning may help plan preoperative radiotherapy, and predict the chance of successful surgical resection. Anterolaterally, branches of the inferior hypogastric plexus and branches of the middle rectal artery and veins run into the mesorectum. They are ensheathed by fascia and together are referred to as the 'lateral rectal ligaments'. The number and calibre of the middle rectal vessels are highly variable. They may be very small or even absent (Sato & Sato 1991). The 'lateral ligaments' are not seen on MRI or CT scanning, and only appear as an identifiable structure with surgical traction on the rectum. The fascia of the 'ligaments' is flimsy and they probably play very little role in support of the rectum. The parietal fascia that covers the levator and pelvic side-wall muscles forms a denser condensation of fascial tissue overlying the sacrum (Fig. 83.7).
RELATIONS OF THE RECTUM
Figure 83.7 Wholemount cadaveric specimen of the mid rectum in a male.
Posterior to the rectum and mesorectum, and separated from them by the presacral fascia in the median plane are the lower three sacral vertebrae, coccyx, median sacral vessels, and the lowest portion of the sacral sympathetic chain. Laterally, the upper part of the rectum is related to the pararectal fossa and its contents (sigmoid colon or terminal ileum), while below the peritoneal reflection lie piriformis, the anterior rami of the lower three sacral and coccygeal nerves,
sympathetic trunk, lower lateral sacral vessels, the coccygei and levatores ani muscles. Anteriorly above the level of the peritoneal reflection lie loops of sigmoid colon or terminal ileum - if these lie in the pelvis - otherwise the rectum is related to the upper parts of the base of the bladder in males, or cervix/body of the uterus and upper vagina in females. The lower parts of the base of the bladder, the seminal vesicles, vas deferens, terminal parts of the ureters, and the prostate in males (Figs 83.5, 83.6), or the lower part of the vagina in females lie below the reflection. In females, the rectovaginal septum is composed of a condensation of connective tissue which is continuous with the connective tissue of the outer layers of the rectal and vaginal walls. In postmenopausal females, and following childbirth, the connective tissue of the rectovaginal septum may atrophy or be thinned, reducing the support for the anterior rectal and posterior vaginal wall.
RECTAL PROLAPSE Prolapse of the rectum involving all layers of the rectal wall may occur. The precise aetiology is not known. During prolapse, the mid and upper rectum descends towards the pelvic floor on straining. This descent tends to occur within the lumen of the rectum as a form of intussusception, and may be a consequence of the large diameter of the lower rectal ampulla and relative fixation of the anorectal junction and anal canal. The upper mesorectum is formed by relatively loose adipose connective tissue, and so relatively little tissue actually fixes the mid and upper rectum in position within the pelvis. The 'lateral ligaments' are composed mainly of vascular structures and offer only limited support against rectal descent. Chronic enlargement of the anorectal space bound by levatores ani commonly occurs in patients suffering rectal prolapse, but it is thought to be a consequence rather than a cause of the prolapse. During prolapse, the rectouterine pouch in females also descends with the anterior rectal wall, and in extreme cases may become everted through the anus between the layers of rectal wall. This rarely happens to the rectovesical pouch in males.
RECTOCOELE When the rectovaginal septum is grossly effaced, especially in postmenopausal females, the pressure of defaecation transmitted forward along the sacral flexure of the rectum can eventually cause bulging of the rectal wall into the posterior vagina, and, in extreme cases, through the vaginal introitus. Failure of support of the rectum and perineum by the puborectalis and pubovaginalis muscles contributes to this prolapse by allowing descent of the posterior perineum during straining.
MESORECTAL EXCISION IN RECTAL CANCER page 1202 page 1203
An important concept in the oncological treatment of adenocarcinoma of the rectum is the integrity of the rectum and its mesorectal tissue. The epirectal and pararectal lymphatic tissues are located throughout the mesorectum. If the mesorectum is divided or disrupted during surgical excision of rectal carcinoma, involved nodes may be left in situ, which may predispose to local recurrence of tumour (Heald et al 1982). Oncological excision of the rectum involves mobilization of the rectum in the plane formed by the mesorectal fascia
posterolaterally around the mesorectum. This plane allows the rectum and mesorectum to be removed as a whole without disruption of the presacral fascia and its underlying venous plexus. It is continuous with the plane between the mesentery surrounding the origin of the superior rectal artery and the presacral fascia over the sacral promontory, which is opened during mobilization of the inferior mesenteric artery origin prior to ligation. The mesorectal plane extends laterally and anteriorly, disrupted only by small branches of the middle rectal vessels, and is continuous with the rectovesical or rectovaginal fascia, which provides a similar oncological plane of dissection. Since the inferior hypogastric nerves are closely related to the plane of the mesorectal fascia in its upper third, dissection must spare the nerves if bladder dysfunction (and erectile dysfunction in males) is to be avoided following surgery. Complete excision of the rectum total mesorectal excision - involves mobilization of this plane to the level of levatores ani. During the excision it is important to follow the sacral curve of the rectum to prevent entry into the presacral fascia or anococcygeal ligament.
LOCAL EXCISION OF THE RECTUM Full thickness excision of lesions of the rectal wall is possible in those parts of the rectum lying below the peritoneal reflections in the extraperitoneal compartment. The mesorectal adipose tissue contains any leakage of contents and provides support to the closure of the defect in the rectal wall. Since only the lower third of the rectum is wholly extraperitoneal, full thickness local excisions of the rectum should be avoided on the anterior wall of the middle third and anterolateral walls of the upper third of the rectum. The height to which such excision can be performed anteriorly in females is lower than in males, because of the lower level of the peritoneal reflection which makes entry into the peritoneal cavity more likely.
© 2008 Elsevier
VASCULAR SUPPLY AND LYMPHATIC DRAINAGE ARTERIES The principal arterial supply to the upper two-thirds of the rectum is the superior rectal artery. Branches of the middle rectal artery provide some additional supply to the middle third, and ascending branches of the inferior rectal artery supply the distal third. A small contribution also comes from the median sacral artery, which is the terminal midline branch of the aorta and enters the posterior wall of the anorectal junction on the sacrorectal fascia (Fig. 83.8). Superior rectal artery The superior rectal artery is the principal continuation of the inferior mesenteric artery. It descends into the pelvis in the sigmoid mesocolon, crosses the left common iliac vessels and passes over the sacral promontory usually just to the left of the midline. It is straddled by the inferior hypogastric nerves on either side. It lies anterior to the upper sacral vertebrae and passes into the upper mesorectum. It descends, initially in the midline, and divides into two branches at the level of the third sacral vertebra. These lie initially posterolateral, then lateral, to the rectal wall as they descend one on each side of the rectum. Terminal branches pierce the muscle wall from the level of the upper mesorectum to enter the rectal submucosa, where they anastomose with ascending branches of the inferior rectal arteries. Middle rectal arteries The middle rectal arteries arise from the anterior division of the internal iliac artery and enter the mesorectum anterolaterally in the 'lateral rectal ligaments'. They are frequently absent (Sato & Sato 1991). When present they provide an arterial supply to the muscle of the mid and lower rectum, but form only poor anastomoses with the superior and inferior rectal arteries. Inferior rectal arteries
Figure 83.8 Details of the arterial supply of the rectum (viewed from behind).
The inferior rectal arteries are terminal branches of the internal pudendal arteries. They enter the upper anal canal laterally and supply the internal and external sphincters, the anal canal below its valves, and the perianal skin. They also provide ascending branches in the submucosa, which anastomose with the terminal branches of the superior rectal artery.
VEINS Rectal venous plexus The rectal venous plexus surrounds the rectum, and connects anteriorly with the vesical plexus in males or the uterovaginal plexus in females. It consists of an internal part, beneath the rectal and anal epithelium, and an external part outside the muscular wall. In the anal canal the internal plexus displays longitudinal dilatations, connected by transverse branches in circles immediately above the anal valves. The dilatations are most prominent in the left lateral, right anterolateral and right posterolateral sectors. The internal plexus drains mainly to the superior rectal vein but connects widely with the external plexus. The inferior portion of the external plexus is drained by the inferior rectal vein into the internal pudendal vein, the middle portion by a middle rectal vein into the internal iliac vein, and its superior part by the superior rectal vein, which is the start of the inferior mesenteric vein. Communication between portal and systemic venous systems is thus established in the rectal plexus. Superior rectal veins The superior rectal veins are formed from the internal rectal plexus. The tributaries of the superior rectal vein ascend in the rectal submucosa as about six
vessels of considerable size which pierce the rectal wall c.7.5 cm above the anus. The branches unite to form the superior rectal vein, which runs along the superior rectal artery in the root of the mesorectum and mesosigmoid, passes to the left of the midline and continues as the inferior mesenteric vein. Middle rectal veins The middle rectal veins pass alongside the middle rectal arteries to drain into the anterior division of the internal iliac vein on the lateral wall of the pelvis.
LYMPHATICS Lymphatics draining the rectum and the upper anal canal above the level of the dentate line pass superiorly, initially through the rectal wall, and then as a fine network over the surface of the rectum, before draining into epirectal nodes in the mesorectum. They usually lie very close to the outer fibres of the rectal longitudinal muscle. The pararectal nodes lie within the mesorectum, a variable distance from the rectal wall. The overall direction of drainage is upwards along the branches of the superior rectal artery. The nodes lie within the loose adipose connective tissue of the mesorectum. Drainage to intermediate groups occurs up to the nodes near the origin of the inferior mesenteric artery (Fig. 83.9). page 1203 page 1204
Figure 83.9 Lymph nodes of the rectum and upper anal canal (viewed from behind).
© 2008 Elsevier
INNERVATION The rectum is innervated primarily via the inferior mesenteric plexus. Both sympathetic and parasympathetic fibres (p. 1185) form a plexus along branches of the superior rectal artery. A small contribution is also made by fibres of the middle rectal plexus along the branches of the middle rectal artery. These are derived from the inferior hypogastric plexus (p. 1185). The role of rectal innervation in continence and its relation to anal innervation is dealt with on page 1210. REFERENCES Broden B, Snellman B 1968 Procidentia of the rectum studied with cineradiography. A contribution to the discussion of causative mechanisms. Dis Colon Rectum 11: 330-47. Medline Similar articles Full article Brown G, Richards CJ, Newcombe RG et al 1999 Rectal carcinoma: thin-section MR imaging for staging in 28 patients. Radiology 211(1): 215-22. Medline Similar articles Heald RJ, Husband EM, Ryall RDH 1982 The mesorectum in rectal cancer surgery: the clue to pelvic recurrence? Br J Surg 69: 613-16. Original description of the mesorectal plane and its relevance to the surgical excision of rectal tumours. Medline Similar articles Full article Sato K, Sato T 1991 The vascular and neuronal composition of the lateral ligament of the rectum and the rectosacral fascia. Surg Radiol Anat 13(1): 17-22. Medline Similar articles Full article
© 2008 Elsevier
84 GASTROINTESTINAL TRACT - LARGE INTESTINE Anal canal The anal canal begins at the anorectal junction and ends at the anal verge (Figs 84.1, 84.2, 84.3, 84.4). It is angulated in relation to the rectum because the pull of the sling-like puborectalis produces the anorectal angle. It lies 2-3 cm in front of and slightly below the tip of the coccyx, which is opposite the apex of the prostate in males. The anal verge is marked by a sharp turn where the squamous epithelium which lines the lower anal canal becomes continuous with the skin of the perineum. The pigmentation of skin around the anal verge demarcates the extent of the external sphincter. Identification of the anal verge may be difficult, particularly in males in whom the perineum may 'funnel' upwards into the lower anal canal. However, the characteristic puckering of the external epithelium caused by the penetrating fibres of the conjoint longitudinal layer (p. 1208) makes a useful landmark. The functional anal canal is represented by a zone of high pressure which roughly equates to the anatomical canal. The anal canal consists of an inner epithelial lining, a vascular subepithelium, the internal and external anal sphincters and fibromuscular supporting tissue. It is between 2.5 and 5 cm long in adults although the anterior wall is slightly shorter than the posterior. It is usually shorter in females. At rest it forms an oval slit in the anteroposterior plane rather than a circular canal due to the arrangement of the external anal sphincter. The anal canal is attached posteriorly to the coccyx by the anococcygeal ligament, a midline fibroelastic structure which may possess some skeletal muscle elements, and which runs between the posterior aspect of the external sphincter and the coccyx. Just above this is the raphe of the levator plate, the fusion of the two halves of the iliococcygeus, which merges anteriorly with puborectalis. Between these two structures is a potential 'postanal' space. The anus is surrounded laterally and posteriorly by loose adipose tissue within the ischioanal fossae, a potential pathway for the spread of perianal sepsis from one side to the other (p. 1366). The ischial spines may be palpated laterally. The pudendal nerves pass over the ischial spines at this point (p. 1126) and pudendal nerve motor terminal latency may be measured digitally using a modified electrode worn on the examining glove. Anteriorly the perineal body separates the anal canal from the membranous urethra and penile bulb in males or from the lower vagina in females.
LINING OF THE ANAL CANAL The upper portion of the anal canal is lined by columnar epithelium similar to that of the rectum. It contains secretory and absorptive cells with numerous tubular glands or crypts. The subepithelial tissues are mobile and relatively distensible and possess profuse submucosal arterial and venous plexuses. Terminal branches of the superior rectal vessels pass downwards towards the anal columns. The submucosal veins drain into the submucosal rectal venous plexus and also through the fibres of the upper internal anal sphincter into an intermuscular venous plexus.
page 1205 page 1206
Figure 84.1 Coronal section through the anal canal. Glandular and vascular structures are only shown unilaterally for clarity.
Figure 84.2 A, Mid-coronal MRI endocoil image of the anal canal. B, Anterior coronal MRI endocoil section in a woman showing the transverse perineii (TP) joining the external anal sphincter anteriorly (between arrows). EAS, external anal sphincter; IAS, internal anal sphincter; PR, puborectalis.
Figure 84.3 Sagittal section through the anal canal. The anal canal is angled posteriorly with the anterior sphincteric structures appearing to be lower than posterior structures. The glandular, vascular and fibromuscular structures and the conjoint longitudinal coat have been omitted for clarity. EAS, external anal sphincter.
There are 6-10 vertical folds, the anal columns, in the mid anal canal. They tend to be obvious in children but less well-defined in adults. Each column contains a terminal radicle of the superior rectal artery and vein: the vessels are largest in the left-lateral, right-posterior and right-anterior quadrants of the wall of the canal where the subepithelial tissues expand into three 'anal cushions'. The cushions
seal the anal canal, helping to maintain continence to flatus and fluid. They are also important in the pathogenesis of haemorrhoids. The lower ends of the columns may form small crescentic folds called the anal valves, between which lie small recesses referred to as anal sinuses. The anal valves and sinuses together form the dentate (or pectinate) line. About six anal glands open into small depressions in the anal valves, the anal crypts. The glands, which are branched, are lined by stratified columnar epithelium. Cystic dilatations may be seen in the glands, which may extend through the internal sphincter and even into the external sphincter (Seow-Cheon Ho 1994). page 1206 page 1207
Figure 84.4 MRI endocoil mid sagittal view of the anal canal in a man. Cs, corpus spongiosus; Tp, transverse perineii; Eas, external anal sphincter; Lm, longitudinal muscle; Ias, internal anal sphincter; PR, puborectalis; Bs, bulbospongiosus.
The mucosa below the dentate line is smooth, and termed the pectin. It is nonkeratinized stratified squamous epithelium which lacks sweat and sebaceous glands and hair follicles, but contains numerous somatic nerve endings. It extends down to the intersphincteric groove, a depression at the lower border of the internal sphincter. The canal below the intersphincteric groove is lined by hairbearing, keratinizing, stratified epithelium which is continuous with the perianal skin. The submucosa in this region contains profuse arterial and venous plexuses and has more connective tissue than the upper canal. It may be tethered to fibres of the conjoint longitudinal layer in the region of the intersphincteric groove. The junction between the columnar and squamous epithelia is referred to as the anal transition zone (ATZ), which is variable in height and position, and often contains islands of squamous epithelium extending up into the columnar mucosa. Nerve endings including thermoreceptors exist in the submucosa around the upper ATZ. They probably play a role in continence by providing a highly specialized 'sampling' mechanism by which the contents of the lower rectum are identified during periods when the upper anal canal relaxes to allow rectal contents to come into contact with the upper anal canal epithelium (Duthie &
Gairns 1960). The well-defined muscularis mucosae of the rectum continues into the upper canal. Fibres from the longitudinal muscle pass through the internal sphincter and surround the submucosal venous plexus, before turning upwards to merge with the muscularis mucosae to form the musculus submucosae ani. Referred pain
Pain in the anus is usually felt with a high degree of acuity and is well localized to the perineum and anal canal itself. Haemorrhoids
Haemorrhoids represent abnormal enlargement of the anal cushions. The partial drainage of the venous plexus into the intermuscular plane may play a role in this chronic engorgement as a result of obstruction of the venous flow during prolonged straining and defaecation. Since the subepithelial vascular cushions of the upper, mid and lower anal canal form a continuous plexus, the differentiation of haemorrhoids into internal and external is somewhat arbitrary. The laxity of the upper anal canal submucosa is probably responsible for the fact that internal haemorrhoids may form more easily, although engorgement of the plexus deep to the lower anal epithelium may also occur. Since the epithelium in the lower canal is well supplied with sensory nerve endings, acute distension or invasive treatment to haemorrhoids in this area causes profound discomfort, whereas invasive or destructive therapy with relatively few symptoms is possible in the upper canal because the latter is lined with insensate columnar mucosa. Anal fissures
Vertical breaks in the anal epithelium are not uncommon in human beings. The development of chronic, non-healing fissures is less common. They are more likely to develop in the anterior, and particularly the posterior, midline of the anus, possibly because stress within the anal canal is concentrated in these regions by the arrangement of the external sphincter fibres. The relative paucity of the arterial supply in these regions probably contributes to poor healing. Persistent hypertonicity of the anal sphincter is a primary pathogenic factor, possibly developing as a consequence of disordered local reflex mechanisms between the lower rectum and internal anal sphincter. Cryptoglandular sepsis and fistula in ano
Infection of the anal glands is the main cause of anal sepsis and cryptogenic fistula formation (Parks et al 1976). Because many of these glands extend into the intersphincteric plane, so-called cryptoglandular sepsis may readily spread into this tissue plane. The commonest route of drainage of this sepsis is downwards between the internal and external anal sphincters to appear beneath the skin at the anal verge. Rupture or drainage of the sepsis at this point will form a fistula in ano in the intersphincteric plane. Less commonly, the infection drains outwards from the intersphincteric plane. It passes through the fibres of the external anal sphincter, possibly along the spreading septa of the conjoint longitudinal coat to appear in the perianal skin outside the external anal sphincter as a transphincteric fistula. If the sepsis does not originate in the intersphincteric plane, drainage is likely to be beneath the anal mucosa alone with the formation of a submucous fistula. Sepsis within the ischioanal fossa surrounding the lower anal canal can spread with relative ease since the connective tissue is mostly loose adipose tissue. Pus may spread posteriorly, behind the lower anal canal, into the contralateral ischioanal fossa due to the absence of any septa or ligamentous attachments of the lower anal canal (Parkes 1961).
MUSCLES OF THE ANAL CANAL The anal canal is encircled by the internal and external anal sphincters, separated
by the longitudinal layer, and has connections superiorly to puborectalis and the transverse perineii (p. 1358) (Figs 84.5, 84.6). Internal anal sphincter
Attachments The internal anal sphincter is a well-defined ring of obliquely orientated smooth muscle fibres continuous with the circular muscle of the rectum, terminating at the junction of the superficial and subcutaneous components of the external sphincter. Its thickness varies between 1.5 and 3.5 mm, depending upon the height within the anal canal and whether the canal is distended. It is usually thinner in females and becomes thicker with age. It may also be thickened in disease processes such as rectal prolapse and chronic constipation. The lower portion of the sphincter is crossed by fibres from the conjoint longitudinal coat which pass into the submucosa of the lower canal. Vascular supply The internal anal sphincter is supplied from the terminal branches of the superior rectal vessels and branches of the inferior rectal vessels. Innervation The internal anal sphincter is innervated by the sympathetic and parasympathetic systems from fibres extending down from the lower rectum. Sympathetic fibres originate in the lower two lumbar segments, are distributed to the sphincter via the inferior hypogastric plexus, and cause contraction of the sphincter. Parasympathetic fibres originate in the second to fourth sacral spinal segments, are distributed through the inferior hypogastric plexus, and cause relaxation of the sphincter. The internal anal sphincter relaxes following stimulation of the rectum suggesting that local reflex pathways exist between the lower rectal sensory fibres and sphincteric motor fibres. This relaxation also occurs on stimulation of somatic sensory nerves present in the pelvic floor indicating that an additional reflex pathway exists via the sacral spinal segments.
EXTERNAL ANAL SPHINCTER The external anal sphincter is an oval tube-shaped complex of striated muscle, composed mainly of type 1 (slow twitch) skeletal muscle fibres, which are well suited to prolonged contraction. page 1207 page 1208
Figure 84.5 A-C, Axial views of the anal canal at three levels on endoanal ultrasound in a woman. The endoanal ultrasound probe is the black structure centrally. A, Upper anal canal. The 'U' shape of the puborectalis (PR) is visible. Ias, internal anal sphincter. B, Middle anal canal. The external anal sphincter (Eas) is now a complete ring anteriorly (arrowhead). Lm, longitudinal muscle; S, subepithelial tissues C, Lower anal canal. Below the termination of the internal anal sphincter, the longitudinal layer extends through the subcutaneous external anal sphincter (between arrowheads). D, Key for levels of the anal canal.
Although previously described as consisting of deep, superficial and subcutaneous parts, the external anal sphincter forms a single functional and anatomical entity. Endoanal ultrasound and magnetic resonance imaging reveal that the uppermost fibres blend with the lowest fibres of puborectalis. Anteriorly some of these upper fibres decussate into the superficial transverse perineal muscles and posteriorly, some fibres are attached to the anococcygeal raphe. The majority of the middle fibres of the external anal sphincter surround the lower part of the internal sphincter. This portion is attached anteriorly to the perineal body and posteriorly to the coccyx via the anococcygeal ligament. Some fibres from each side of the sphincter decussate in these areas to form a commissure in the anterior and posterior midline. The lower fibres lie below the level of the internal anal sphincter and are separated from the lowest anal epithelium by submucosa. The length and thickness of the external anal sphincter varies between the sexes: in females, the anterior portion tends to be shorter, the wall may be slightly thinner, and the tube may take the form of an asymmetrical cone (Rociu et al 2000). In women the transverse perineii and bulbospongiosus fuse with the external sphincter in the lower part of the perineum. In men the annular external sphincter is separate from the central point of the perineum into which the transverse perineii and bulbospongiosus fuse, so that there is a surgical plane of cleavage between the external sphincter and perineum (Fig. 84.7). Vascular supply The external anal sphincter is supplied from the terminal branches of the inferior rectal vessels with a small contribution from the median sacral artery. Innervation The external anal sphincter is innervated by the inferior rectal branch of the pudendal nerve originating in the anterior divisions of the second to fourth sacral nerve roots.
FIBROMUSCULAR STRUCTURES OF THE ANAL CANAL The longitudinal layer and conjoint longitudinal coat
The longitudinal layer is situated between the internal and external sphincters and contains a fibromuscular layer, the conjoint longitudinal coat, and the intersphincteric space with its connective tissue components (Lunniss & Phillips 1992). The longitudinal layer has a muscular and fibroelastic component. The muscle part is formed by fusion of striated muscle fibres from puboanalis, the innermost part of puborectalis, with smooth muscle from the longitudinal muscle of the rectum. Endoanal ultrasound and magnetic resonance imaging demonstrate either muscle bundles or incomplete sheets of muscle extending down between the sphincters in the upper canal in both sexes: in men these often end just above the lower border of the internal sphincter. The layer then becomes completely fibroelastic, and splits into septa running between bundles of the subcutaneous external sphincter to terminate in the perianal skin. The area bounded by these septa is generally referred to as the perianal space. The most peripheral of the septa extend between the fibres of the external sphincter into the ischioanal fat. The most central septa pass through the fibres of the internal sphincter to reach the anal lining and may help to form the intersphincteric groove. The conjoint longitudinal coat is innervated by autonomic fibres from the same
origin as those supplying the internal sphincter. page 1208 page 1209
page 1209 page 1210
Figure 84.6 MRI scan images of the anal canal. A, Key for levels of the anal canal. B, Upper anal canal. High in the canal with the sling of puborectalis (PR) extending anteriorly to the pubic bones. Vag, vagina; Ur, urethra. C, Mid anal canal. Mid canal level shows the transverse perineii (Tp) fusing into the external anal sphincter anteriorly. The superficial (middle) external anal sphincter (SpEas) is attached either side of the anococcygeal ligament (Acl). Ias, internal anal sphincter. D, Low anal canal. Low canal level, below the internal anal sphincter. ScEas, subcutaneous (lower) part of the external anal sphincter.
Figure 84.7 Typical male and female external anal sphincter anatomy. The puborectalis is shown in isolation with the external anal sphincter viewed from superolateral. The anterior portion of the external anal sphincter is typically shorter and thinner in females.
Other fibromuscular structures
A layer of smooth muscle, yellow elastic fibres, and collagenous connective tissue is found in the anal submucosa, inferior to the anal sinuses. It is derived mainly from strands of the conjoint longitudinal coat, which descend inwards between the fibres of the internal sphincter. Some of the strands end by turning outwards around the lower edge of the internal sphincter to rejoin the main longitudinal layer. Most continue obliquely downwards and insert into the dermis below the intersphincteric groove. These attachments may help to form the corrugations seen in the perianal skin. The septa end in a honeycomb-like arrangement of fibres, which prevents easy distension of the lowest anal lining. This may explain the severe pain produced by pus or blood which collects here, since small volumes of fluid rapidly produce high pressure within the subepithelium.
© 2008 Elsevier
VASCULAR SUPPLY AND LYMPHATIC DRAINAGE ARTERIES The arterial supply to the anal canal is derived from terminal branches of the superior rectal artery, the inferior rectal branch of the pudendal artery and branches of the median sacral artery. The supply to the anal canal lining is not distributed uniformly. The anterior and, more particularly, the posterior, midline epithelia have a relatively poorer arterial supply than that lining the lateral portions of the canal (Klosterhalfen et al 1989). This is of relevance in the perpetuation of chronic anal fissures (p. 1207).
VEINS The venous drainage of the upper anal canal mucosa, internal anal sphincter and conjoint longitudinal coat passes via the terminal branches of the superior rectal veins into the inferior mesenteric vein. The lower anal canal and external sphincter drain via the inferior rectal branch of the pudenal vein into the internal iliac vein.
LYMPHATICS Lymphatics from the upper anal mucosa, internal anal sphincter and conjoint longitudinal coat drain upwards into the submucosal and intramural lymphatics of the rectum. The lower anal canal epithelium and external anal sphincter lymphatics drain downwards via perianal plexuses into vessels which drain into the external inguinal lymph nodes. The lymphatics of puborectalis drain into the internal iliac lymph nodes. This arrangement has considerable importance for tumours of the lower rectum and upper anal canal. If the tumour is confined to the tissue of origin, malignant spread is confined to the mesorectal lymph nodes. However, if the tumour involves puborectalis or tissues associated with the external anal sphincter, the possibility of lymph node spread to these other groups may require radical excision and much more extensive surgery.
© 2008 Elsevier
ANORECTAL CONTINENCE To keep the anal canal closed the pressure within the anal canal has to be higher than in the rectum. The resting pressure in the canal is maintained mainly by tonic activity of the internal sphincter, with sudden increases of intrarectal pressure, as with coughing or exertion, compensated for by rapid contraction of the external sphincter and puborectalis. Angulation of the anorectal junction and the 'flap valve' theory is no longer considered important in continence. The internal sphincter does not close the canal completely, and the (c.7 mm) gap is sealed by the vascular subepithelial tissues.
DEFECATION Defecation is a conscious physiological act in response to feeling the need to pass stool in the rectum, requiring coordinated relaxation of the pelvic floor muscles and anal sphincter. The dynamics have been studied using pressure measurements in the colon, rectum and anus (Herbst et al 1997) and by various imaging techniques including fluoroscopy, ultrasonography and magnetic resonance imaging (Kruyt et al 1991). The process is initiated by mass colonic contractions driving faeces into the rectum. Rectal distension lowers internal sphincter tone in preparation for defecation. Defecation may be deferred by conscious contraction of the external anal sphincter until contractions cease and retrograde rectal peristalsis moves stool out of the distal rectum, so that the sensation to defecate passes off. Initiation of defecation involves relaxation of the pelvic floor muscles and external sphincter, so that the pelvic floor descends and the anal canal opens. Abdominal contraction will aid expulsion from the rectum, but continuing mass colonic contractions push more faeces down into the rectum, so that the entire left colon may be emptied. Integrating the sensory input from the anal canal in order to control the activity of the anal musculature occurs at many levels in the nervous system, including the spinal cord, brain stem, thalamus and cortex. Neural activity monitors and regulates defecation, and other more subtle behaviours within the rectum and anal canal, such as the separation of faeces from rectal gas; local adjustments to faecal consistency and quantities; self-cleansing movements in the rectum and anal canal; and coordination with other actions of the perineal and abdominal muscles. UPDATE Date Added: 13 December 2005 Abstract: Mechanisms controlling normal defecation and the potential effects of spinal cord injury. Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=16198712&query_hl=5 Mechanisms controlling normal defecation and the potential effects of spinal cord injury. Brading AF, Ramalingam T. Prog Brain Res. 152:345-58, 2005.
ANAL INCONTINENCE Anal incontinence is common and may occur for a variety of reasons. Abnormal high rectal pressures, as in severe diarrhoea, may overcome a normal sphincter, and autonomic disorders may produce abnormal motor activity or loss of normal sensation, so that there is no awareness of stool. There may also be damage to the anal sphincter, mostly from vaginal delivery, or atrophy of the external sphincter from pudendal damage, also of obstetric origin, except in the elderly. A damaged sphincter will not be able to overcome normal fluctuations of intrarectal pressures, or there may be a combination of damage and rectal dysfunction resulting in anal incontinence. page 1210 page 1211
REFERENCES Duthie HL, Gairns FW 1960 Sensory nerve-endings and sensation in the anal region of man. Br J Surg 47: 585-95. Medline Similar articles Full article Herbst F, Kamm MA, Morris GP, Britton K, Woloszko J, Nicholls RJ 1997 Gastrointestinal transit and prolonged ambulatory colonic motility in health and faecal incontinence. Gut 41: 381-9. Medline Similar articles Klosterhalfen B, Vogel P, Rixen H, Mittermayer C 1989 Topography of the inferior rectal artery: a possible cause of chronic primary anal fissure. Dis Colon Rectum 32: 43-52. A detailed postmortem angiographic study demonstrating the arrangement of anal arterial supply Medline Similar articles Full article Kruyt RH, Delemarre JB, Doornbos J, Vogel HJ 1991 Normal anorectum: dynamic MR imaging anatomy. Radiology 179(1): 159-63. Describes the MR anatomy of the anorectum in relation to surrounding structures and the anorectal angle at rest, during perineal contraction, and during straining, in asymptomatic subjects Medline Similar articles Lunniss PJ, Phillips RK 1992 Anatomy and function of the anal longitudinal muscle. Br J Surg 79: 882-4. Medline Similar articles Full article Parkes AG 1961 The pathogenesis and treatment of fistula in ano. Br Med J 1: 463-9. An early, full description of the relationship between the anatomy of anal glands and cryptoglandular sepsis Medline Similar articles Parks AG, Gordon PH, Hardcastle JD 1976 A classification of fistula-in-ano. Br J Surg 63(1): 1-12. A classification of anal fistulas based on the pathogenesis of the disease and the normal muscular anatomy of the pelvic floor Medline Similar articles Full article Rociu E, Stoker J, Eijkemans MJ, Lameris JS 2000 Normal anal sphincter anatomy and age-and sexrelated variations at high-spatial-resolution endoanal MR imaging. Radiology 217: 395-401. Medline Similar articles Seow-Choen F, Ho JM 1994 Histoanatomy of anal glands. Dis Colon Rectum 37: 1215-18. Medline Similar articles page 1211 page 1212
© 2008 Elsevier
85 HEPATOBILIARY SYSTEM: Liver The liver is the largest of the abdominal viscera, occupying a substantial portion of the upper abdominal cavity. It performs a wide range of metabolic activities necessary for homeostasis, nutrition and immune defence. It is composed largely of epithelial cells (hepatocytes), which are bathed in blood derived from the hepatic portal veins and hepatic arteries. There is continuous chemical exchange between the cells and the blood. Hepatocytes are also associated with an extensive system of minute canals, which form the biliary system into which products are secreted. The liver is important in the removal and breakdown of toxic, or potentially toxic, materials from the blood. It regulates blood glucose and lipids, and plays a role in the storage of certain vitamins, iron, and other micronutrients as well as breaking down or modifying amino acids . It is involved in a plethora of other biochemical reactions. Since the majority of these processes are exothermic, a substantial part of the thermal energy production of the body, especially at rest, is provided by the liver. The liver is populated by phagocytic macrophages, which form part of the mononuclear phagocyte system of the body, and are important in the removal of particulates from the blood stream. In fetal life the liver is an important site of haemopoiesis. The liver lies in the upper right part of the abdominal cavity. It occupies most of the right hypochondrium and epigastrium, although it frequently extends into the left hypochondrium as far as the left lateral line. In adults the liver weighs c.2% of body mass. The liver has an overall wedge shape, which is in part determined by the form of the upper abdominal cavity into which it grows. The narrow end of the wedge lies towards the left hypochondrium, with the anterior edge pointing anteriorly and inferiorly. The superior and right lateral aspects are shaped by the anterolateral abdominal and chest wall as well as the diaphragm. The inferior aspect is shaped by the adjacent viscera. In life it is reddish brown in colour and although firm and pliant its weight and texture depend in part on the volume of venous blood it contains. The liver capsule plays an important part in maintaining the integrity of its shape. Once the capsule is lacerated, the liver tissue is easily parted and provides only limited support for surgical sutures. These features, in combination with its exceptional vascular supply, make the liver prone to potentially lethal injuries if it is split open. The liver is probably supported in its position in the upper abdomen by several factors. Tone in the anterolateral abdominal muscles is important in holding many viscera, including the liver, in place. Ligamentous attachments of the liver capsule to the diaphragm and anterior abdominal wall (p. 1213) provide some support, and prevent rotation of the liver about its vascular attachments at the porta hepatis (p. 1215) and hepatic veins (p. 1220). The liver is also attached to the relatively fixed retroperitoneal inferior vena cava by the hepatic and caudate veins. The relative importance of these attachments can be seen after orthotopic liver transplantation where, despite the absence of ligamentous structures, the liver remains within the right upper quadrant although it is more prone to torsion and rotation.
EXTERNAL FEATURES Hepatic attachments
The liver is attached to the anterior abdominal wall, diaphragm and other viscera by several ligaments, which are formed from condensations of the peritoneum as described on page 1127. Falciform ligament
The liver is attached in front to the anterior abdominal wall by the falciform ligament. The two layers of this ligament descend from the posterior surface of the anterior abdominal wall and diaphragm and turn onto the anterior and superior surfaces of the liver. On the dome of the superior surface, the right leaf runs laterally and is continuous with the upper layer of the coronary ligament. The left layer of the falciform ligament turns medially and is continuous with the anterior layer of the left triangular ligament. The ligamentum teres - which represents the obliterated left umbilical vein - (p. 1215) runs in the lower free border of the falciform ligament and continues into a fissure on the inferior surface of the liver. Coronary ligament
The coronary ligament is formed by the reflection of the peritoneum from the
diaphragm onto the posterior surfaces of the right lobe of the liver. Between the two layers of this ligament there is a large triangular area of liver devoid of peritoneal covering called the 'bare area' of the liver. Here the liver is attached to the diaphragm by areolar tissue and is in continuity inferiorly with the anterior pararenal space. The coronary ligament is continuous on the right with the right triangular ligament. On the left, it becomes closely applied, and forms the left triangular ligament. The upper layer of the coronary ligament is reflected superiorly onto the inferior surface of the diaphragm and inferiorly onto the right and superior surface of the liver. The lower layer of the coronary ligament reflects inferiorly over the right suprarenal gland and right kidney, and superiorly onto the inferior surface of the liver. Surgical division of the right triangular and coronary ligaments allows the right lobe of the liver to be brought forward, and exposes the lateral aspect of the inferior vena cava behind the liver. Triangular ligaments
The left triangular ligament is a double layer of peritoneum which extends to a variable length over the superior border of the left lobe of the liver. Medially the anterior leaf is continuous with the left layer of the falciform ligament. The posterior layer is continuous with the left layer of the lesser omentum. The left triangular ligament lies in front of the abdominal part of the oesophagus, the upper end of the lesser omentum and part of the fundus of the stomach. Division of the left triangular ligament allows the left lobe of the liver to be mobilized for exposure of the abdominal oesophagus and crura of the diaphragm. The right triangular ligament is a short structure which lies at the apex of the 'bare area' of the liver and is continuous with the layers of the coronary ligament. Lesser omentum
The lesser omentum is a fold of peritoneum which extends from the lesser curve of the stomach and proximal duodenum to the inferior surface of the liver. The attachment to the inferior surface of the liver is L-shaped. The vertical component follows the line of the fissure for the ligamentum venosum - the fibrous remnant of the ductus venosus. More inferiorly the attachment runs horizontally to complete the L in the porta hepatis. At its upper end, the superior layer of lesser omentum is continuous on the left with the posterior layer of the left triangular ligament, and the inferior layer is continuous on the right with the coronary ligament as it encloses the inferior vena cava. At its lower end, the two layers diverge to surround the structures of the porta hepatis. A thin fibrous condensation of fascia usually runs from the medial end of the porta hepatis into the fissure in the inferior surface which contains the ligamentum teres. This fascia is continuous with the lower border of the falciform ligament when the ligamentum teres re-emerges at the inferior border of the liver. Care should be taken when dividing the lesser omentum, since an aberrant left hepatic artery may run in the medial end. page 1213 page 1214
Hepatic surfaces (Fig. 85.1)
The liver is usually described as having superior, anterior, right, posterior and inferior surfaces, and has a distinct inferior border. However, superior, anterior and right surfaces are continuous and no definable borders separate them. It would be more appropriate to group them as the diaphragmatic surface, which is mostly separated from the inferior, or visceral surface, by a narrow inferior border. The border is rounded between the right and inferior surfaces, but becomes angled much more sharply between the anterior and inferior surfaces. This part of the inferior border is notched by the ligamentum teres, just to the right of the midline. The inferior border follows the right costal margin lateral to the fundus of the gallbladder, which usually corresponds to a second notch, 4-5 cm to the right of the midline. To the left of the ligamentum teres, the inferior border ascends below the medial end of the right costal margin. It crosses the infrasternal angle to pass behind the medial end of the left costal margin near the tip of the eighth costal cartilage. At the infrasternal angle the inferior border is related to the anterior abdominal wall and is accessible to examination by percussion, but is not usually palpable. In the midline, the inferior border of the liver is near the transpyloric plane, about a hand's breadth below the xiphisternal joint. In women and children the border often projects a little below the right costal margin. Superior surface
The superior surface is the largest surface and lies immediately below the diaphragm, separated from it by peritoneum except for a small triangular area
where the two layers of the falciform ligament diverge. The majority of the superior surface lies beneath the right dome; however, centrally there is a shallow cardiac impression corresponding to the position of the heart above the central tendon of the diaphragm. The left side of the superior surface lies beneath part of the left dome of the diaphragm. The superior surface blends imperceptibly with the anterior, right and posterior surfaces over the 'dome' of the liver. It is related to the right diaphragmatic pleura and base of the right lung, to the pericardium and ventricular part of the heart, and to part of the left diaphragmatic pleura and base of the left lung.
Figure 85.1 The surfaces and external features of the liver. Top left, superior view; top right, posterior view; bottom left, anterior view; bottom right, inferior view.
Anterior surface
The anterior surface, which is approximately triangular and convex, is covered by peritoneum except at the attachment of the falciform ligament. Much of it is in contact with the anterior attachment of the diaphragm. On the right the diaphragm separates it from the pleura and sixth to tenth ribs and cartilages, and on the left from the seventh and eighth costal cartilages. The thin margins of the base of the lungs are thus quite close to the upper part of this surface, more extensively so on the right. The median area of the anterior surface lies behind the xiphoid process and the anterior abdominal wall in the infracostal angle. Right surface
The right surface is covered by peritoneum and lies adjacent to the right dome of the diaphragm which separates it from the right lung and pleura and the seventh to eleventh ribs. Above and lateral to its upper third, lie both the right lung and basal pleura between the diaphragm and the seventh and eighth ribs. The diaphragm, the costodiaphragmatic recess lined by pleura, and the ninth and tenth ribs all lie lateral to the middle third of the right surface. Lateral to the lower third, the diaphragm and thoracic wall are in direct contact. Rarely, the hepatic flexure and proximal transverse colon may lie on a long mesentery over the right and superior surfaces of the liver, referred to as Chilaiditi syndrome. Liver biopsy
page 1214 page 1215
The liver lies conveniently close to the abdominal wall on its right and lateral anterior surfaces. Under normal circumstances, no other structures lie between the liver parenchyma and the diaphragm overlying the intercostal spaces of the ninth and tenth ribs. During deep inspiration the lung may descend to fill the pleura-lined costodiaphragmatic recess as far down as the tenth rib, but on forced
expiration it also shrinks, and the eight or even seventh intercostal space may come into direct contact with the diaphragm overlying the liver. Blind percutaneous needle biopsy of the liver can be achieved through these intercostal spaces provided the patient can 'fix' the position of the liver by holding a forced expiration, and so ensure that the lung is not interposed between the body wall and the diaphragm. More commonly, biopsy is performed under ultrasound guidance. This not only ensures the lung is not liable to injury, but that the biopsy is directed to specific lesions within the parenchyma. Posterior surface
The posterior surface is convex, wide on the right, but narrow on the left. A deep median concavity corresponds to the forward convexity of the vertebral column close to the attachment of the ligamentum venosum. Much of the posterior surface is attached to the diaphragm by loose connective tissue, which forms the so-called 'bare area'. The 'bare area' is triangular in shape. It is bounded above and below by the layers of the coronary ligament and its apex is directed down to the right, running into the right triangular ligament. Lateral to its lower end, the 'bare area' is an anterior relation of the upper pole of the left suprarenal gland. The inferior vena cava lies in a groove or tunnel in the medial end of the 'bare area'. To the left of the caval groove the posterior surface of the liver is formed by the caudate lobe, which is covered by a layer of peritoneum. This peritoneum is continuous with that of the inferior layer of the coronary ligament and the layers of the lesser omentum. The potential space between these three peritoneal layers is often referred to as the superior omental recess. The caudate lobe is related to the diaphragmatic crura above the aortic opening and the right inferior phrenic artery. It is separated by these structures from the descending thoracic aorta. The fissure for the ligamentum venosum separates the posterior aspect of the caudate from the main part of the left lobe. The fissure cuts deeply in front of the caudate lobe and contains the two layers of the lesser omentum. Below, it curves laterally to the left end of the porta hepatis. The ligamentum venosum is attached below to the posterior aspect of the left branch of the portal vein. It ascends in the floor of the fissure and passes laterally. At the upper end of the caudate lobe it joins the left hepatic vein near its entry into the inferior vena cava, or sometimes the vena cava itself. The posterior surface over the left lobe bears a shallow oesophageal impression near the upper end of the fissure for the ligamentum venosum caused by the abdominal part of the oesophagus. The posterior surface of the left lobe to the left of this impression is related to part of the fundus of the stomach (Fig. 85.2)
Figure 85.2 Relations of the liver. Top left, superior view; top right, posterior view; bottom left, anterior view; bottom right, inferior view.
Inferior surface
The inferior surface is bounded by the inferior edge of the liver. It blends with the posterior surface in the region of the origin of the lesser omentum, the porta hepatis and the lower layer of the coronary ligament. It is marked near the midline by a sharp fissure which contains the ligamentum teres - the obliterated fetal left umbilical vein. Occasionally this fissure is bridged by liver tissue to form a tunnel. Posteriorly the inferior surface is related to the ligamentum venosum and the gallbladder. The latter usually lies in a shallow fossa but may vary from having a short mesentery, to being completely intrahepatic when it lies within a cleft in the liver parenchyma (p. 1227). Between the fissure for the ligamentum teres and the gallbladder lies the quadrate lobe. The inferior surface of the left lobe is related inferiorly to the fundus of the stomach and the upper lesser omentum. The quadrate lobe lies adjacent to the pylorus, first part of the duodenum and the lower part of the lesser omentum. Occasionally the transverse colon lies between the duodenum and the quadrate lobe. To the right of the gallbladder, the inferior surface is related to the hepatic flexure of the colon, the right suprarenal gland and right kidney, and the first part of the duodenum. The inferior surface may demonstrate a pronounced 'mound' of liver tissue close to the left border of the fissure for the ligamentum teres. This is referred to as the tuber omentale. The porta hepatis (Figs 85.3, 85.4)
Figure 85.3 Cross-section of the structures at the porta hepatis.
Figure 85.4 Axial CT of the porta hepatis. The hepatic ducts lie anteriorly, the portal vein posteriorly, and hepatic artery between the two.
page 1215 page 1216
The porta hepatis is the area of the inferior surface through which all the neurovascular and biliary structures, except the hepatic veins, enter and leave the liver. It is situated between the quadrate lobe in front and the caudate process behind. The porta hepatis is actually a deep fissure into which the portal vein, hepatic artery and hepatic nervous plexus ascend into the parenchyma of the liver. The right and left hepatic bile ducts and some lymph vessels emerge from it. At the porta hepatis, the hepatic ducts lie anterior to the portal vein and its branches, and the hepatic artery with its branches lies between the two. All these structures are enveloped in the perivascular fibrous capsule - hepatobiliary capsule of Glisson - a sheath of loose connective tissue which surrounds the vessels as they course through the portal canals in the liver. It is also continuous with the fibrous hepatic capsule. The dense aggregation of vessels, supporting connective tissue, and liver parenchyma just above the porta hepatis is often referred to as the 'hilar plate' of the liver. It may be dissected surgically to gain access to the intrahepatic branches of the bile ducts and vessels. The left hepatic duct remains extrahepatic as it runs down to the bifurcation along the base of segment IV - the quadrate lobe. This extrahepatic length of duct is particularly useful when performing high biliary duct reconstructions where a length of jejunum is anastomosed to form a biliary enteric bypass, for strictures of the common hepatic duct.
LOBATION AND SEGMENTATION The liver has four lobes or eight segments, depending on whether it is defined by its gross anatomical appearance or by its internal architecture. Classification of the liver by internal architecture divides it into lobes, segments or sectors. The biliary, hepatic arterial and portal venous supply of the liver tend to follow very similar distributions used to define the hepatic segments. The hepatic venous anatomy follows a markedly different pattern. At the edge of segments, there is considerable overlap between vascular and biliary structures and segments are not identifiable by either gross external or internal examination of the liver. The value of the segmental classification, according to vascular and biliary supply, is that surgical resection of a segment, multiple segments or a whole lobe, may be planned and performed to encounter the fewest possible major vascular structures. The surface of the liver is often marked by indentations or accessory fissures which do not relate directly to the lobes or segments. They occur most often over the superior, right, and anterior surfaces. UPDATE Date Added: 19 April 2005 Abstract: Insights into liver anatomy: the Canals of Hering Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=15085485 Insights into liver anatomy: the Canals of Hering. Gross anatomical lobes
Historically the gross anatomical appearance of the liver has been divided into right, left, caudate and quadrate lobes by the surface peritoneal and ligamentous attachments. The falciform ligament superiorly and the ligamentum venosum (p. 1219) inferiorly, mark the division between right and left lobes. On the inferior surface, to the right of the groove formed by the ligamentum venosum, there are two prominences separated by the porta hepatis. The quadrate lobe lies anteriorly, the caudate lobe posteriorly. The gallbladder usually lies in a shallow fossa to the right of the quadrate lobe. Right lobe
The right lobe is the largest in volume and contributes to all surfaces. It is demarcated by the line of attachment of the falciform ligament superiorly. Inferiorly the fissure for the ligamentum teres, the groove for the ligamentum venosum, and the attachment of the lesser omentum, mark its border. The inferior border of the right lobe, to the right of the gallbladder, often demonstrates a bulge of tissue, which when pronounced, is referred to as Riedel's lobe. Although the right inferior border of the liver is not usually palpable, the presence of a Riedel's lobe may be clinically detectable and may give rise to confusion as an apparent pathological right upper quadrant mass. Quadrate lobe
The quadrate lobe is only visible from the inferior surface. It is bounded by the
gallbladder fossa to the right, a short portion of the inferior border anteriorly, the fissure for the ligamentum teres to the left, and the porta hepatis posteriorly. In gross anatomical descriptions it is said to be a lobe arising from the right lobe; however, it is functionally related to the left lobe. Caudate lobe
The caudate lobe is visible on the posterior surface. It is bounded on the left by the fissure for the ligamentum venosum, below by the porta hepatis, and on the right by the groove for the inferior vena cava. Above, it continues into the superior surface on the right of the upper end of the fissure for the ligamentum venosum. Below and to the right, it is connected to the right lobe by a narrow caudate process, which is immediately behind the porta hepatis and above the epiploic foramen. Depending on the depth of the fissure for the ligamentum venosum, the caudate lobe often has an anterior surface, which forms the posterior wall of the fissure and is in contact with the hepatic part of the lesser omentum. In gross anatomical descriptions this lobe is said to arise from the right lobe, but it is functionally separate. Left lobe
The left lobe is the smaller of the two 'main' lobes. It lies to the left of the falciform ligament, has no subdivisions, and ends in a thin apex pointing into the left upper quadrant. Since it is substantially thinner than the right lobe it is more flexible. It is nearly as large as the right lobe in young children, possibly due to a more even distribution in portal and hepatic arterial supply, which may progressively come to favour growth of the right lobe during development of the body cavity. A fibrous band may be present at the left end of the adult left lobe: it represents an atrophied remnant of the more extensive left lobe found in children. If present, it contains atrophied bile ducts called the hepatic vasa aberrantia. Similar remnants may occur in the inferior border of the lobe and near the inferior vena cava. Couinaud segments (Figs 85.5, 85.6) page 1216 page 1217
Figure 85.5 Segmentation of the liver - Couinaud. Top left, superior view; top right, posterior view; bottom left, anterior view; bottom right, inferior view. The segments are sometimes referred to by name - I, caudate (sometimes subdivided into left and right parts); II, lateral superior; III, lateral inferior; IV, medial (sometimes subdivided into superior and inferior parts); V, anterior inferior; VI, posterior inferior; VII, posterior superior; VIII, anterior superior.
Although a variety of definitions have been used to describe the anatomy of the liver segments, the most widely accepted clinical nomenclature is that described by Couinaud (1957), and Healey and Schroy (1953). The internal architecture of the liver is divided into segments, commonly referred to as Couinaud's segments. Couinaud based his work on the distribution of the portal and hepatic veins whilst
Healey and Schroy studied the arterial and biliary anatomy. The liver is divided by the 'principal plane' into two halves of approximately equal size. The principal plane is defined by an imaginary parasagittal line from the gallbladder anteriorly to the inferior vena cava posteriorly. The usual functional division of the liver into right and left lobes lies along this plane. The liver is further subdivided into segments, each supplied by a principal branch of the hepatic artery, portal vein and bile duct. Segments I, II, III and IV make up the functional left lobe, and segments V, VI, VII and VIII make up the functional right lobe. The right lobe can be further divided into a posterior and anterior section or sector. The right posterior section is made up of segments VI and VII, and the right anterior section is made up of segments V and VIII. The left lobe can also be divided into sections: segment IV is referred to as the left medial section, and segments II and III as the left lateral section. The hepatic veins lie in liver parenchyma between the sections. Segment I corresponds to the gross anatomical caudate lobe and segment IV to the quadrate lobe. The delineation of the internal liver architecture has been confirmed by crosssectional imaging techniques as well as hepatic portography and arteriography. On axial CT and MRI, segment I is situated posterior and to the right of the inferior vena cava; segments VII, VIII, IV and II run in a clockwise fashion above the portal vein; and segments VI, V, IV and III are situated in a similar manner below the portal vein. The value of the identification of the liver segments and sections according to vascular and biliary supply is that surgical resection of a segment, section, multiple segments, a lobe or greater volume of tissue may be performed whilst encountering the least number of possible major vascular structures. Liver resection
Surgical resection of the liver for primary and secondary neoplasia is now routine, and there is very low morbidity and mortality. Knowledge of the internal anatomy of the liver is essential. As much as 80% of the liver mass can be removed safely. The liver has the unique capacity of regeneration, and will regrow to its original size some 6-12 months after resection. The identification of the hepatic arterial and portal venous segmentation means that lesions seen on cross-sectional imaging can be placed within segments, so that the feasibility of resection can be assessed. Although detailed arterial and portal venous imaging is usually required to allow definitive surgical planning, understanding the segmental anatomy of the liver has allowed considerable advances to be made. Since the hepatic venous anatomy differs widely from the portal and arterial anatomy, resections rarely lie within wholly convenient vascular planes. Currently, the main limitation to liver resection is not the difficulty of segmental anatomy but the involvement of vital structures such as the inferior vena cava, or the need to preserve an adequate volume of functioning liver tissue after resection. Resection of the liver can be described as either non-anatomical or anatomical. Non-anatomical resections are usually minor and the lines of resection are not related to Couinaud's segments. Major resections usually follow the planes between the segments and are anatomical. The nomenclature of liver resections has been historically confusing. Recently, Strasberg (1997) has attempted to simplify this by describing the removal of segments V to VIII as a right hemihepatectomy. The resection of liver has otherwise been described with respect to sections and segments. For example, removal of segments IV to VIII is a right trisectionectomy, involving the removal of the right posterior, right anterior and left medial sections. Major anatomical liver resections involve control of the appropriate inflow and outflow to the part of liver to be resected. Once the inflow has been stopped a line of ischaemic demarcation appears which then guides the surgeon along a plane of resection. Liver transplantation
Liver transplantation is an established form of treatment for patients with end stage liver disease. Orthotopic transplantation involves a standard hepatectomy including the hepatic inferior vena cava. The implant of the graft then requires a superior and inferior caval anastomosis, followed by anastomosis of the portal vein. In adults, most surgeons use venovenous bypass during the anhepatic phase of the transplant. This allows splanchnic and systemic venous blood to return to the heart via the internal jugular vein. The transplant is completed by
performing the arterial and biliary anastomoses. UPDATE Date Added: 26 April 2005 Shanida Helena Nataraja, PhD (Dianthus Medical Limited) Update: Retrospective analysis reveals that anatomic variations are common in right liver grafts. Right lobe liver living donor transplantation is a promising surgical alternative for adult patients with end-stage liver disease. A precise knowledge of liver anatomy is essential for this surgical intervention. Between June 1999 and January 2002, 96 right lobe living donor transplantations were performed at the Recanati/Miller Transplantation Institute, Mount Sinai Hospital, New York. The mean (±SD) age of the 65 male and 31 female patients was 38.4 (±10) years. The radiologic and histologic findings from these 96 patients were reviewed retrospectively to assess surgical liver anatomy. Portal vein anatomy: type 1, 86.4%; type 2, 6.3%; type 3, 7.3%. Biliary tree anatomy: type 1, 55.8%; type 2, 14.3%; type 3a, 5.2%; type 3b, 5.6; type 4a, 2.6%; type 4b, 6.5%. Twenty-seven donors (28%) had a classic pattern. Hepatic artery anatomy: type 1, 70.8%; type 2a, 6.25%; type 2b, 6.25%; type 3a, 3.1%; type 3b,10.4%; type 4b, 2.1%; and type 5, 1.1%. Only 28% of patients displayed a "classic" anatomy, i.e., type 1 for hepatic artery, portal vein, and bile ducts. Of the patients with type 3b biliary anatomy, 60% were found to have type 3 portal anatomy compared with 12.5% of donors with type 1 and 2 portal anatomy. Anatomic variations in right liver grafts are therefore very common, and classical anatomy is only present in a minority of patients. Although these anatomic variations should not be viewed as a contraindication for transplantation, clinicians need to be aware of them. Varotti G, Gondolesi GE, Goldman J, Wayne M, Florman SS, Schwartz ME, Miller CM, Sukru E. Anatomic variations in right liver living donors. J Am Coll Surg. 2004;198(4):577-82. Medline Similar articles page 1217 page 1218
For children, a cadaveric liver may be split into two along the lines of segmentation such that a single liver can be used for two transplants, usually an adult and a child or two small adults. Live donor liver transplants may be performed since the left lateral section and sometimes a right hemiliver (segments V to VIII) can be removed from a healthy donor and then transplanted.
Figure 85.6 Couinaud segments of the liver seen on axial CT scan. A, Contrast enhanced CT shows the left (L), middle (M), and right (R) hepatic veins at the superior aspect of the liver. B, Inferior to this the caudate lobe (segment I) lies between the inferior vena cava (IVC) and the main portal vein (PV). The left portal vein (LPV) separates segment II superiorly from segment III inferiorly. C, The right portal vein (RPV) divides segments V and VI inferiorly (C) from segments VII and VIII superiorly (B).
© 2008 Elsevier
VASCULAR SUPPLY AND LYMPHATIC DRAINAGE The vessels connected with the liver are the portal vein, hepatic artery and hepatic veins. The portal vein and hepatic artery ascend in the lesser omentum to the porta hepatis, where each bifurcates. The hepatic bile duct and lymphatic vessels descend from the porta hepatis in the same omentum (Figs 85.3, 85.4). The hepatic veins leave the liver via the posterior surface and run directly into the inferior vena cava. UPDATE Date Added: 12 July 2005 Shanida Helena Nataraja, PhD (Dianthus Medical Limited) Abstract: Evaluation of hepatic vascular anatomy in potential right lobe donors. Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=15795845&query_hl=13 Evaluation of hepatic vascular anatomy in potential right lobe donors. UPDATE Date Added: 21 June 2005 Shanida Helena Nataraja, PhD (Dianthus Medical Limited) Abstract: Relevant vascular anatomy of the surgical plane in split-liver transplantation. Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=14595144&query_hl=10 Relevant vascular anatomy of the surgical plane in split-liver transplantation. UPDATE Date Added: 14 June 2005 Shanida Helena Nataraja, PhD (Dianthus Medical Limited) Update: Importance of anatomic assessment of liver allografts before split-liver transplantation. Successful split-liver transplantation (SLT) and avoidance of post-operative complications requires both an intact arterial and portal blood supply and sufficient biliary and venous drainage from the remaining hepatic segments. The importance of pre-operative assessment of hepatic anatomic structures before SLT has been demonstrated in a study of human cadaveric livers. Sixty livers were obtained during routine autopsy, and resections were done en bloc with liver, celiac trunk, left gastric artery, lesser omentum, superior mesenteric artery and head of the pancreas. The caudate lobe was excised in order to divide the liver into its right and left lobes. The hepatic artery, portal vein, biliary tree and hepatic veins were dissected and correlated with the hepatic segments. The right, the median and the left hepatic veins were unique in 59 (98.3%), 53 (88.3%) and 46 (76.3%) cases, respectively. The portal vein trunk divided into right and left branches in 59 (98.3%) cases, a median branch appeared in nine (15.2%) cases and no portal vein bifurcation occurred in one (1.6%) case. There were multiple right and left hepatic ducts in 47 (78.3%) and 57 (95%) cases, respectively. Analysis of the intra-hepatic distribution of the right hepatic duct revealed that there were four branches in 28 (59%) cases (segments V, VI, VII and VIII), two branches in 11 (23%) cases, (segments V and VI) and two branches in eight (17%) cases (segments VII and VIII). Analysis of the intra-hepatic distribution of the left hepatic duct, on the other hand, showed that the major branches were distributed toward hepatic segments II and III. Three separate branches of the left hepatic duct were found in 11 (19%) cases (segments II, III and IV). The arterial supply of the liver was predominantly provided by the right and left hepatic arteries; in nine (15%) cases this role was fulfilled by the median hepatic artery. The right hepatic artery, coming from the superior mesenteric artery, was present in 15 (25%) cases, whereas the left hepatic artery, originating from the left gastric
artery, was observed in only two (3.3%) cases. The right and left hepatic artery was accessory in 11 (18.3%) and 2 (3.3%) cases, respectively. The right hepatic artery was dominant in four (6.6%) cases. Bifurcation of the portal vein occurred in the majority of livers assessed in this study. Chaib E, Ribeiro MA Jr, Saad WA et al: The main hepatic anatomic variations for the purpose of split-liver transplantation. Transplant Proc 37(2):1063-1066, 2005.
HEPATIC ARTERY (Figs 85.7, 85.8) In adults the hepatic artery is intermediate in size between the left gastric and splenic arteries. In fetal and early postnatal life it is the largest branch of the coeliac axis. After its origin from the coeliac axis, it passes anteriorly and laterally below the epiploic foramen to the upper aspect of the superior part of the duodenum. The artery may be subdivided into the common hepatic artery - from the coeliac trunk to the origin of the gastroduodenal artery - and the hepatic artery 'proper' - from that point to its bifurcation. It passes anterior to the portal vein and ascends between the layers of the lesser omentum. It lies anterior to the epiploic foramen and passes in the free border of the lesser omentum medial to the common bile duct and anterior to the portal vein. At the porta hepatis it divides into right and left branches before these run into the parenchyma of the liver. The right hepatic artery usually crosses posterior (occasionally anterior) to the common hepatic duct. It almost always divides into an anterior branch supplying segments V and VIII, and a posterior branch supplying segments VI and VII. The anterior division often supplies a branch to segment I and the gallbladder. The segmental arteries are macroscopically end arteries although some collateral circulation occurs between segments via fine terminal branches. UPDATE Date Added: 08 June 2005 Shanida Helena Nataraja, PhD (Dianthus Medical Limited) Abstract: Vascular variations in living related liver transplant donor candidates. Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=15848625&query_hl=8 Vascular variations in living related liver transplant donor candidates. Bob Browne
UPDATE Date Added: 08 June 2005 Shanida Helena Nataraja, PhD (Dianthus Medical Limited) Abstract: Anatomic variation of right hepatic artery and its reconstruction in living donor liver transplantation. Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=15848624&query_hl=7 Anatomic variation of right hepatic artery and its reconstruction in living donor liver transplantation. Bob Browne Medline
Similar articles
page 1218 page 1219
Figure 85.7 Dissection to show the relations of the hepatic artery, bile duct and portal vein to each other in the lesser omentum: anterior aspect.
Figure 85.8 Hepatic arteriogram. A, A selective hepatic arteriogram shows normal left hepatic artery branches and small right hepatic artery branches. B, The right hepatic artery is arising from the origin of the superior mesenteric artery.
There are a small number of normal variants, which are important to demonstrate angiographically because they may influence surgical and interventional radiological procedures. A vessel which supplies a lobe in addition to its normal vessel is defined as an accessory artery. A replaced hepatic artery is a vessel that does not originate from an orthodox position and is the sole supply to that lobe. Rarely a replaced common hepatic artery arises from the superior mesenteric artery. More commonly a replaced right hepatic artery or an accessory right hepatic artery arises from the superior mesenteric artery (Fig. 85.8B). In this case they run behind the portal vein and bile duct in the lesser omentum. Occasionally, a replaced left hepatic artery or an accessory branch arises from the left gastric artery. This provides a source of collateral arterial circulation in cases of occlusion of the vessels in the porta hepatis and may be injured during mobilization of the stomach as it lies in the upper portion of the lesser omentum. Rarely, accessory left or right hepatic arteries may also arise from the gastroduodenal artery or aorta. UPDATE Date Added: 21 June 2005 Shanida Helena Nataraja, PhD (Dianthus Medical Limited) Abstract: Comparison of reliability of hepatic artery configuration in 3D computerized tomography (CT) angiography compared with conventional angiography. Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=15819796&query_hl=13 Comparison of reliability of hepatic artery configuration in 3D computerized tomography (CT) angiography compared with conventional angiography. UPDATE Date Added: 15 March 2005 Helen Elizabeth Wiggett, PhD (Dianthus Medical Limited) Update: Anatomical variations of the arterial pattern in the right hemiliver
This study has analyzed the arterial supply to the right hemiliver and attempted to classify variations in the right hepatic artery (RHA), particularly at the segmental and sectional level, using corrosion casts prepared from 80 undamaged livers removed during autopsy. The RHA (supplies segments 5, 8, 6, and 7 and is a branch of the proper hepatic artery [PrHA]) was classified according to its origin and branching pattern. In addition, the origin and number of sectional and segmental arteries for the right hemiliver were determined for each liver. The right hemiliver was supplied by one artery in 96% (77 cases) of cases, originating from either the PrHA (73/77) or the superior mesenteric artery (SMA; 4/77) with anastomosis to the left hepatic artery. In the remaining cases, two arteries supplied the right hemiliver, one originating from the PrHA and the other from the SMA, and there was always an anastomosis between the two arteries. The replacing RHA (originating from the SMA) supplied the whole right hemiliver in 5% (4 cases) of cases and the incomplete (supplies < 4 but > 2 right liver segments) replacing artery supplied segments 5, 6, 7, and a part of 8 in 3% of cases (2 cases) and only segments 5, 6, and a part of 8 in 1 case. In the majority of cases (61%), the anterior section (segments 5 and 6) was supplied by one artery originating from either the RHA or the replacing RHA. Of the remaining cases, 30% (24 cases) were supplied by two arteries which either both originated from the RHA (14/24) or one originated from the RHA and one from the posterior sectional artery (PSA; supplies sections 6 and 7; 10/24); 9% (7 cases) were supplied by three arteries, either two originating from the RHA and one from the PSA (4/7), or from the incomplete RHA, the incomplete replacing RHA, and the PSA, respectively (3/7). The posterior section was supplied by one artery in the majority of cases (66%), and this originated from either the RHA or the replacing RHA. In 31% of cases (25 cases), the posterior section was supplied by two arteries, either both originating from the RHA (16/25); or from the RHA and the anterior sectional artery (ASA) respectively (8/25); or from the incomplete RHA and incomplete RHA respectively (1/25). The remaining two cases were supplied by three arteries, either all originating from the RHA, or one from the RHA and two from the ASA. Segments 5 and 7 were predominantly supplied by one artery (49% and 51% of cases, respectively) and segments 6 and 8 by two arteries (59% and 55% of cases, respectively). In summary, the study has classified the arteries of the hemiliver into 10 groups that differ according to their origin and branching pattern. Mlakar B, Gadzijev EM, Ravnik D, Hribernik M. Anatomical variations of the arterial pattern in the right hemiliver. Eur J Morphol. 2002 Dec;40(5):267-73.
The hepatic artery gives off right gastric, gastroduodenal and cystic branches as well as direct branches to the bile duct from the right hepatic and sometimes the supraduodenal artery.
VEINS The liver has two venous systems. The portal system conveys venous blood from the majority of the gastrointestinal tract and its associated organs to the liver. The hepatic venous system drains blood from the liver parenchyma into the inferior vena cava. Portal venous system
The portal system includes all the veins draining the abdominal part of the digestive tube with the exception of the lower anal canal, but including the abdominal part of the oesophagus. It also drains the spleen, pancreas and gallbladder. The portal vein conveys the blood from these viscera to the liver, where it ramifies like an artery, and ends in the sinusoids from which vessels again converge to reach the inferior vena cava via the hepatic veins (p. 1222). The blood running through the portal system therefore passes through two sets of 'exchange' vessels, namely the capillaries of the gut, spleen, pancreas or gallbladder, and the hepatic sinusoids.
In adults, the portal vein and its tributaries have no valves. In fetal life and for a short postnatal period valves are demonstrable in its tributaries, but they usually atrophy. Rarely some persist in an atrophic form into adulthood. Portal vein (Figs 88.3, 85.9, 85.10)
The portal vein begins at the level of the second lumbar vertebra and is formed from the convergence of the superior mesenteric and splenic veins. It is c.8 cm long and lies anterior to the inferior vena cava and posterior to the neck of the pancreas. It lies obliquely to the right and ascends behind the first part of the duodenum, the common bile duct and gastroduodenal artery. At this point it is directly anterior to the inferior vena cava. It enters the right border of the lesser omentum, and ascends anterior to the epiploic foramen to reach the right end of the porta hepatis. It then divides into right and left main branches which accompany the corresponding branches of the hepatic artery into the liver. In the lesser omentum it lies posterior to both the common bile duct and hepatic artery. It is surrounded by the hepatic nerve plexus and accompanied by many lymph vessels and some lymph nodes. The right branch usually receives the cystic vein and then enters the right lobe. In common with the hepatic artery, it usually forms an anterior division supplying segments V and VIII and a posterior division supplying segments VI and VII. The anterior division may give a branch to segment I. The left branch has a longer extraparenchymal course and tends to lie slightly more horizontal than the right branch but is often of smaller calibre. It gives off branches to segments I (caudate), II, III and IV (quadrate). As it enters the left lobe it is joined by para-umbilical veins and the ligamentum teres, which contains the functionless and partly obliterated left umbilical vein. It is connected to the inferior vena cava by the ligamentum venosum, a vestige of the obliterated ductus venosus. The small extrahepatic section of the left branch, from which the branches to segments II, III and IV arise, is a persistent part of the left umbilical vein. The portal vein receives many branches including the splenic, superior mesenteric, left gastric, right gastric, para-umbilical and cystic veins. Portal venous blood is one route through which hepatic metastases from gastrointestinal primary malignancies may spread. Blood within the portal vein flows at such a rate that streaming may occur so that the blood from the splenic vein tends to remain on the left side of the portal blood stream and drain preferentially to the left main branch. The clinical evidence to support this is very limited since colorectal cancer metastases commonly occur in the right lobe. The portal vein supplies the liver with c.5% of its resting oxygen consumption but significantly more of its metabolic nutrition. Progressive occlusion of the hepatic artery rarely results in complete necrosis of the liver, which is due principally to the blood supply derived from the portal vein. Porto-systemic shunts page 1219 page 1220
Figure 85.9 The portal vein and its tributaries (semi-diagrammatic). Portions of the stomach, pancreas and left lobe of the liver and the transverse colon have been removed.
Figure 85.10 Coronal CT of the portal vein and superior mesenteric vein.
Increase in the pressure within the portal venous system may occur for a wide range of reasons. Chronic hypertension of the portal system results in dilatation of the portal vein and its tributaries in response to the raised postcapillary pressure. Tiny venous channels not normally visible may dilate and become engorged. In those areas where these veins form anastomoses with veins draining into the systemic venous circulation, a reversal of flow may occur due to the pressure difference within the two systems. Portal venous blood then flows into the systemic circulation without having been processed by the liver tissue. The following are common sites of porto-systemic shunts. Between the left gastric and lower oesophageal veins (portal) and the lower branches of the oesophageal veins draining into the azygos and accessory hemiazygos veins (systemic). Enlargement of these anastomoses may result in the formation of varices, either oesophageal or gastric. These may give rise to potentially fatal torrential bleeding. Between the superior rectal veins (portal) and the middle and inferior rectal veins draining into the internal iliac and pudendal veins (systemic). The dilated veins may be seen on the rectal wall, but rarely give rise to troublesome bleeding and are not a cause for internal haemorrhoids. Between persistent tributaries of the left branch of the portal vein running in the ligamentum teres and the peri-umbilical branches of the superior and inferior epigastric veins (systemic), forming the so-called 'caput medusae'. Between intraparenchymal branches of the right branch of the portal vein lying in liver tissue exposed in the 'bare area' and retroperitoneal veins draining into the lumbar, azygos and hemiazygos veins. Between omental and colonic veins (portal) and retroperitoneal veins (systemic) in the region of the hepatic and splenic flexure. Rarely, between a patent ductus venosus connected to the left branch of the portal vein and the inferior vena cava. Hepatic veins (Figs 85.6A, 85.11, 85.12) page 1220 page 1221
Figure 85.11 Sagittal ultrasound of the middle hepatic vein. Middle hepatic vein (MHV) draining into the inferior vena cava (IVC).
Figure 85.12 Arrangement of the hepatic venous territories. Multiple lower group veins may be present. Individual segments may drain into more than one hepatic venous territory.
The hepatic veins convey blood from the liver to the inferior vena cava. The tributaries arise within the parenchyma of the liver and have a thin tunica adventitia which binds them to the walls of their canals within the liver. They are a major source of bleeding following open liver injury since they are less able to collapse sufficiently to allow haemostasis to occur. The veins commence as intralobular veins, which drain the sinusoids and lead to sublobular veins, which eventually unite into hepatic veins. These emerge from the posterior hepatic surface to open directly into the inferior vena cava in its groove on the posterior hepatic surface. Hepatic veins are arranged in upper and lower groups. The upper group are usually large veins and commonly referred to as the right, middle and left hepatic veins. The right hepatic vein drains segments V, VI, VII and VIII.
The middle hepatic vein lies between segments IV and VIII and drains both these segments and segment V. The left hepatic vein drains segments II and III with some drainage from segment IV. The lower group vary in number and extent of distribution. They are small veins draining directly into the inferior vena cava from segment I and occasionally from segments VII and VIII. The hepatic veins have no valves. The caudate lobe often has small veins draining directly into the inferior vena cava and therefore may hypertrophy in conditions involving thrombosis of the large hepatic veins, e.g. Budd-Chiari syndrome. UPDATE Date Added: 21 June 2005 Shanida Helena Nataraja, PhD (Dianthus Medical Limited) Abstract: Hepatic vein anatomy of the medial segment for living donor liver transplantation. Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=15776411&query_hl=11 Hepatic vein anatomy of the medial segment for living donor liver transplantation. Transjugular intraparenchymal porto-systemic shunt (TIPS) procedure for portal hypertension
In situations where severe, possibly life-threatening, complications due to chronic portal hypertension exist, the formation of a large calibre anastomosis between portal and systemic circulations may be used to reduce the intraportal pressure. The safest way to perform this is to use the proximity of the hepatic veins and dilated portal branches within the liver parenchyma. Balloon catheters introduced via the internal jugular vein can be passed down the superior and inferior vena cava into the hepatic veins. Under radiological guidance, puncture across an appropriate strip of liver tissue can be achieved into a dilated portal branch. Balloon rupture and dilatation of this tissue 'window' can then be performed with relative safety, since the surrounding liver tissue provides some degree of support to the damaged tissue.
LYMPHATICS Lymph from the liver has abundant protein content. Lymphatic drainage from the liver is wide and may pass to nodes both above and below the diaphragm. This can be seen since obstruction of the hepatic venous drainage increases the flow of lymph in the thoracic duct. Hepatic collecting vessels are divided into superficial and deep systems. Superficial hepatic vessels
The superficial vessels run in subserosal areolar tissue over the whole surface of the liver and drain in four directions. Lymph vessels from the majority of the posterior surface, the surface of the caudate lobe, and the posterior part of the inferior surface of the right lobe, accompany the inferior vena cava and drain into pericaval nodes. Vessels in the coronary and right triangular ligaments may directly enter the thoracic duct without any intervening node. Vessels from the majority of the inferior surface, anterior surface and most of the superior surface all converge on the porta hepatis to drain into the hepatic nodes. A few lymph vessels from the posterior surface of the lateral end of the left lobe pass towards the oesophageal opening to drain into the paracardiac nodes. One or two lymph trunks from the right surface and right end of the superior surface accompany the inferior phrenic artery across the right crus to drain into the coeliac nodes. page 1221 page 1222
Deep hepatic lymphatics
The great majority of the liver parenchyma is drained by lymphatic vessels within the substance of the liver. The fine lymphatic vessels merge to form larger vessels. Some run superiorly through the parenchyma to form the ascending trunks. They accompany the hepatic veins and pass through the caval opening in the diaphragm to drain into nodes round the end of the inferior vena cava. Vessels from the lower portion of the liver form descending trunks which emerge from the porta hepatis to drain into the hepatic nodes.
© 2008 Elsevier
INNERVATION The liver has a dual innervation. The parenchyma is supplied by hepatic nerves, which arise from the hepatic plexus and contain sympathetic and parasympathetic (vagal) fibres (p. 1150). They enter the liver at the porta hepatis and largely accompany the hepatic arteries and bile ducts. A very few may run directly within the liver parenchyma. The capsule is supplied by some fine branches of the lower intercostal nerves, which also supply the parietal peritoneum, particularly in the area of the 'bare area' and superior surface. This is seen clinically when distension or disruption of the liver capsule causes quite well localized sharp pain.
HEPATIC PLEXUS The hepatic plexus is the largest derivative of the coeliac plexus. It also receives branches from the anterior and posterior vagi. It accompanies the hepatic artery and portal vein and their branches into the liver, where its fibres run close to the branches of the vessels. These branches not only supply vasomotor fibres to the hepatic vessels and biliary tree, but also innervate the hepatocytes directly and are involved in the control of some homeostatic mechanisms. Branches to the gallbladder form a delicate cystic plexus. Multiple fine branches from the plexus supply the common and hepatic bile ducts directly. The vagal fibres are motor to the musculature of the gallbladder and bile ducts and inhibitory to the sphincter of the bile duct. Nerves run from the hepatic plexus with the branches of the common hepatic artery to supply, or contribute to the supply of, foregut derivatives. Branches may run inferiorly from the plexus to accompany the right gastric artery and contribute to the supply of the pylorus; with the gastroduodenal artery and branches to reach the pylorus and the first part of the duodenum; with the right gastroepiploic artery to provide a small contribution to the supply the right side of the stomach and the greater curvature. The superior pancreaticoduodenal extension supplies the descending part of the duodenum, the pancreatic head, and the intrapancreatic part of the common bile duct.
REFERRED PAIN Pain arising from the parenchyma of the liver is poorly localized. In common with other structures of foregut origin, pain is referred to the central epigastrium. Stretch of or involvement of the liver capsule by inflammatory or neoplastic processes rapidly produces well-localized pain of a 'somatic' nature.
© 2008 Elsevier
MICROSTRUCTURE The liver is essentially an epithelial-mesenchymal outgrowth of the caudal part of the foregut, with which it retains its connection via the biliary tree. The surface of the liver facing the peritoneal cavity is covered by a typical serosa, the visceral peritoneum. Beneath this, and enclosing the whole structure, is a thin (50-100 µm) layer of connective tissue from which extensions pass into the liver as connective tissue septa and trabeculae. Branches of the hepatic artery and hepatic portal vein, together with bile ductules and ducts, run within these connective tissue trabeculae which are termed portal tracts (portal canals). The combination of the two types of vessel and a bile duct is termed a portal triad; these structures are usually accompanied by one or more lymphatic vessels. The liver parenchyma consists of a complex network of epithelial cells, supported by connective tissue, and perfused by a rich blood supply from the hepatic portal vein and hepatic artery. The epithelial cells, hepatocytes, carry out the major metabolic activities of this organ, but additional cell types possess storage, phagocytic and mechanically supportive functions. In the mature liver, hepatocytes are arranged mainly in plates - or cords, as seen in two-dimensional sections - usually only one cell thick. Until about seven years of age, plates are normally two cells thick (p. 1255). Between the plates are venous sinusoids, which anastomose with each other via gaps in the hepatocyte plates. Bile secreted by the hepatocytes is collected in a network of minute tubes (canaliculi). The hepatocytes can therefore be regarded as exocrine cells, secreting bile to the alimentary tract ultimately via the hepatic ducts and bile duct. However, their other metabolic orientation is towards the blood, with which hepatocytes carry out complex biochemical exchanges. The fetal liver is a major haemopoietic organ; erythrocytes, leukocytes and platelets develop from the mesenchyme covering the sinusoidal endothelium.
LOBULATION OF THE LIVER (Figs 85.13, 85.14)
page 1222 page 1223
Figure 85.13 The histological organization of the liver, showing the principal types of subdivisions which have been proposed. For purposes of clarity, the territories of the classic hepatic lobules are shown as regular hexagons, unlike their real appearance which is highly variable. The portal lobule, centred on the portal triad and biliary drainage is also shown. Liver function is emphasized by the territory of the liver acinus, which centres on blood flow to sectors of adjacent hepatic lobules (see text) and reflects gradients of metabolism across its zones.
Figure 85.14 The structural organization of human liver tissue into lobules, bordered by delicate connective tissue septa (arrows) in which run branches of the hepatic portal vein, hepatic artery and bile duct, grouped as portal triads. A central vein drains each lobule. (By permission from Dr JB Kerr, Monash University, from Kerr JB 1999 Atlas of Functional Histology. London: Mosby.)
In humans, the arrangement of hepatic plates into discrete lobules is less clear than in some species, where the classic lobular units of structure are delimited microscopically by distinct connective tissue septa. These lobules are comprised of polygonal (often hexagonal) clusters of hepatocytes, about 1 mm in diameter, bounded by loose connective tissue which in humans is scant. Within each lobular unit, hepatic plates with intervening sinusoids radiate around a central vein, a tributary of the hepatic vein draining the tissue. These plates do not pass straight to the periphery of a lobule like the spokes of a wheel but run irregularly, as they anastomose and branch. Detailed studies of human liver, using threedimensional reconstruction and morphometric analysis, combined with histopathological observations, have revealed a highly orderly arrangement of the human liver into functional units, the liver (portal) acini. They are approximately oval masses of tissue, centred on a terminal branch of a hepatic arteriole and portal venule, and with their long axes defined by the territory between two adjacent central veins. Each acinus includes the hepatic tissue served by these afferent vessels and is bounded by the territories of other acini. The acinar definition of hepatic micro-organization has clarified important problems of liver histopathology, especially the development of zones of anoxic damage, glycogen deposition and removal, and of toxic trauma, which are all
related to the direction of blood flow and thus tend to follow the acinar pattern. There are also real metabolic differences between hepatocytes within the acini and so they are divided into three zones: zone 1 (periportal) nearest to the terminal branches of afferent vessels; zone 2 intermediate zone; and zone 3 around the central venous drainage.
BLOOD SUPPLY Preterminal hepatic arterioles in the portal canals branch to convey arterial blood to the sinusoids by several routes, mainly via a fine capillary plexus which drains to branches of the portal veins. Some arterial blood passes directly to the hepatic sinusoids, bypassing these capillary plexuses; but they represent only a small part of the total flow. Sinusoids thus contain mixed venous and arterial blood. Central veins from adjacent lobules form interlobular veins, which unite as hepatic veins, draining blood to the inferior vena cava. Hepatic veins draining the tissue run quite separately with respect to the portal triad system, freely crossing the boundaries of triad territories.
Figure 85.15 A central vein draining a hepatic lobule. The surrounding liver parenchyma consists of plates (cords) of hepatocytes radiating from the central vein and interposed hepatic sinusoids through which blood flows towards the central vein. (By permission from Dr JB Kerr, Monash University, from Kerr JB 1999 Atlas of Functional Histology. London: Mosby.)
Figure 85.16 A portal triad in the liver, toluidine blue stained. It contains branches of the hepatic portal vein (centre; generally the largest profile), the hepatic artery (here, a small arteriole, top left profile) and the elongated profile of a bile duct, with typical round epithelial cell nuclei (top right). Other small bile ductule and arteriolar branches are also visible. (Photograph by Sarah-Jane Smith.)
CELLS OF THE LIVER (Figs 85.15, 85.16) Cells of the liver include hepatocytes, hepatic stellate cells - also known as perisinusoidal lipocytes, or Ito cells - sinusoidal endothelial cells, macrophages (Kupffer cells), the cells of the biliary tree - cuboidal to columnar epithelium - and connective tissue cells of the capsule and portal tracts. Hepatocytes (Figs 85.17, 85.18, 85.19, 85.20) page 1223 page 1224
Figure 85.17 The chief cellular features of a hepatic cord, showing hepatocytes, grooved by bile canaliculi. A discontinuous fenestrated endothelium lines the sinusoids, shown containing erythrocytes. Also shown are a Kupffer cell and a hepatic stellate cell. Fine collagen fibres occupy the space of Disse.
Figure 85.18 High power electron micrograph of the border of a hepatic sinusoid showing two hepatocytes, with their plasma membranes facing a sinusoid containing erythrocytes. Their lateral membranes enclose a small bile canaliculus (top left). The sinusoid is lined by fenestrated endothelial cells; the nucleus of an endothelial cell is seen in the centre of the field. The endothelium lacks a basal lamina and is separated from hepatocytes by the space of Disse, into which hepatocyte microvilli project. (By permission from Young B, Heath JW 2000 Wheater's Functional Histology. Edinburgh: Churchill Livingstone.)
page 1224 page 1225
Figure 85.19 Diagram of hepatic microstructure. Perisinusoidal endothelial cells and macrophages (Kupffer cells) are not shown. (Modified after H Elias, Department of Anatomy, Chicago Medical School.)
Figure 85.20 Electron micrograph showing portions of three adjacent hepatocytes and the intervening bile canaliculi, bounded by tight junctions.
About 80% of the liver volume and 60% of its cell number are formed by hepatocytes (parenchymal cells). They are polyhedral, with 5-12 sides and are from 20 to 30µm across. Their nuclei are round, euchromatic and often tetraploid, polyploid or multiple - two or more in each cell. Their cytoplasm typically contains much rough and smooth endoplasmic reticulum, many mitochondria, lysosomes and well-developed Golgi apparatus, features indicating a high metabolic activity. Glycogen granules and lipid vacuoles are usually prominent. Numerous, large peroxisomes and vacuoles containing enzymes, e.g. urease in distinctive crystalline forms, indicate the complex metabolism of these cells. Their role in iron metabolism is shown by the presence of storage vacuoles containing crystals of ferritin and haemosiderin. The surfaces of hepatocytes facing the sinusoids exhibit numerous microvilli, c.0.5 µm long, which create a large area of membrane - 70% of the hepatocyte surface - exposed to blood plasma. Elsewhere, hepatocytes are linked by numerous gap junctions and desmosomes. Lateral plasma membranes of adjacent hepatocytes form microscopic channels, the bile canaliculi, which are specialized regions of intercellular space formed by apposing grooves in hepatocyte plasma membranes, sealed from extraneous interstitial space by tight junctions. Numerous membrane-bound exocytotic vesicles cluster near the lumen of the canaliculi, since the secretion of bile components is targeted to the canalicular plasma membrane. These canaliculi form the origins of the biliary tree and their tight junctions prevent bile from entering interstitial fluid or blood plasma: this is the blood-bile barrier. Hepatic stellate cells
Hepatic stellate cells are also known as perisinusoidal lipocytes or Ito cells and are much less numerous than hepatocytes. They are irregular in outline and lie within the hepatic plates, between the bases of hepatocytes. They are thought to be mesenchymal in origin and are characterized by numerous cytoplasmic lipid droplets. These cells secrete most of the intralobular matrix components, including collagen type III (reticular) fibres. They store the fat-soluble vitamin A in their lipid droplets and are a significant source of growth factors active in liver homeostasis and regeneration. Hepatic stellate cells also play a major role in pathological processes. In response to liver damage, they become activated and predominantly myofibroblast-like. They are responsible for the replacement of toxically damaged hepatocytes with collagenous scar tissue - hepatic fibrosis, seen initially in zone 3, around central veins. This can progress to cirrhosis, where the parenchymal architecture and pattern of blood flow are destroyed, with major systemic consequences. Sinusoidal endothelial cells
Hepatic venous sinusoids are generally wider than blood capillaries and are lined by a thin but highly fenestrated endothelium which lacks a basal lamina (Fig. 85.18). The endothelial cells are typically flattened, each with a central nucleus and joined to each other by junctional complexes. The fenestrae are grouped in clusters with a mean diameter of 100nm, allowing plasma direct access to the basal plasma membranes of hepatocytes. Their cytoplasm contains numerous typical transcytotic vesicles. Kupffer cells
Kupffer cells are hepatic macrophages derived from circulating blood monocytes. They are long-term hepatic residents, lying within the sinusoidal lumen (Fig. 85.17), attached to the endothelial surface. They originate in the bone marrow, and form a major part of the mononuclear phagocyte system (p. 81), responsible for removing cellular and microbial debris from the circulation, and secreting cytokines involved in defence. Kupffer cells remove aged and damaged red cells from the hepatic circulation, a function normally shared with the spleen, but fulfilled entirely by the liver after splenectomy. Kupffer cells are irregular in shape, with long processes extending into the sinusoidal lumen.
HEPATIC PLATES (CORDS) The endothelial linings of the sinusoids are separated from hepatocytes of the hepatic plates by a narrow gap, the perisinusoidal space of Disse which is normally about 0.2-0.5 µm wide, but distends in anoxic conditions. It contains fine collagen fibres - chiefly type III, with some types I and IV - the microvilli of adjacent hepatocytes, and occasional non-myelinated nerve terminals. There is no basal lamina within the space of Disse. Minute bile canaliculi form nets with polygonal meshes in the hepatic plates. Each polygonal hepatocyte is surrounded by canaliculi except on the surfaces - at least two - facing sinusoids. Hepatic plates thus enclose a network of canaliculi which pass to the lobular periphery, where they join to form narrow intralobular ductules (terminal ductules or the canals of Hering) lined by squamous or cuboidal epithelium. These enter bile ductules in the portal canals, lined by cuboidal or columnar cells. The flow of bile is thus towards the periphery of lobules, in the opposite direction to the blood flow, which is centripetal. REFERENCES Couinaud C 1957 Le foie: etudes anatomique et chirurgicules. Masson: Paris. The original description of
hepatic segmentation by Couinaud Healey JE, Schroy PC 1953 Anatomy of the biliary ducts within the human liver; analysis of the prevailing pattern of branchings and the major variations of the biliary ducts. Arch Surg 66: 599-616. Mitchell AWM, Dick R 1999 Liver, gall-bladder, pancreas and spleen. In: Butler P, Mitchell AWM, Ellis H (eds) Applied Radiological Anatomy. Cambridge: Cambridge University Press: 239-58. Strasberg SM 1997 Terminology of liver anatomy and liver resections: coming to grips with hepatic babel. J Am Coll Surg 184: 413-34. A review and suggested system for the clinical nomenclature of liver surgery according to segmental anatomy Medline Similar articles page 1225 page 1226
© 2008 Elsevier
86 HEPATOBILIARY SYSTEM: Gallbladder and biliary tree The biliary tree consists of the system of vessels and ducts which collect and deliver bile from the liver parenchyma to the second part of the duodenum. It is conventionally divided into intrahepatic and extrahepatic biliary ducts. The intrahepatic ducts are formed from the larger bile canaliculi (p. 1222) which come together to form segmental ducts. These fuse close to the porta hepatis into right and left hepatic ducts. The extrahepatic biliary tree consists of the right and left hepatic ducts, the common hepatic duct, the cystic duct and gallbladder and the common bile duct (Fig. 86.1).
© 2008 Elsevier
GALLBLADDER (Fig. 86.2)
Figure 86.1 Overall arrangement of the intrahepatic and extrahepatic biliary tree. The biliary tree to the level of the segmental ducts shown in relation to the conventional arterial anatomy. The segmental ducts often branch just before, or are multiple, as they enter the main lobar ducts, but are shown as single ducts for clarity here. Note that segment I usually has drainage to both right and left hepatic ducts. The level of the liver parenchyma at the porta hepatis is shown by the dashed black line. Duodenum, brown; portal veins, blue; hepatic arteries, red.
The gallbladder is a flask-shaped, blind-ending diverticulum attached to the common bile duct by the cystic duct. In life, it is grey-blue in colour and usually lies attached to the inferior surface of the right lobe of the liver by connective tissue. In the adult the gallbladder is between 7 and 10 cm long with a capacity of up to 50 ml. It usually lies in a shallow fossa in the liver parenchyma covered by peritoneum continued from the liver surface. This attachment can vary widely. At one extreme the gallbladder may be almost completely buried within the liver surface, having no peritoneal covering (intraparenchymal pattern); at the other extreme it may hang from a short mesentery formed by the two layers of peritoneum separated only by connective tissue and a few small vessels (mesenteric pattern). The gallbladder is described as having a fundus, body and neck. The neck lies at the medial end close to the porta hepatis, and almost always has a short peritoneal covered attachment to the liver (mesentery); this mesentery usually contains the cystic artery. The mucosa at the medial end of the neck is obliquely ridged, forming a spiral groove continuous with the spiral valve of the cystic duct. At its lateral end the neck widens out to form the body of the gallbladder and this widening is often referred to in clinical practice as 'Hartmann's pouch'. The neck lies anterior to the second part of the duodenum. The body of the gallbladder normally lies in contact with the liver surface. When the neck possesses a mesentery, this rapidly shortens along the length of the body as it comes to lie in the gallbladder fossa. It lies anterior to the second part
of the duodenum and the right end of the transverse colon. The fundus lies at the lateral end of the body and usually projects past the inferior border of the liver to a variable length. It often lies in contact with the anterior abdominal wall behind the ninth costal cartilage where the lateral edge of the right rectus abdominis crosses the costal margin. This is the location where enlargement of the gallbladder is best sought on clinical examination. The fundus commonly lies adjacent to the transverse colon. The gallbladder varies in size and shape. The fundus may be elongated and highly mobile. Rarely the fundus of the gallbladder is folded back upon the body of the gallbladder, the so-called Phrygian cap. On ultrasound this may be wrongly interpreted as an apparent septum within an otherwise normal gallbladder. Rarely, the gallbladder may be bifid or completely duplicated, usually with a duplicated cystic duct.
Figure 86.2 Interior of the gallbladder and bile ducts.
© 2008 Elsevier
EXTRAHEPATIC BILIARY TREE UPDATE Date Added: 12 April 2005 Abstract: New algorithm for reconstructing magnetic resonance cholangiography (MRC) biliary structure Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=15127745 New algorithm for reconstructing magnetic resonance cholangiography (MRC) biliary structure.
CYSTIC DUCT The cystic duct drains the gallbladder into the common bile duct. It is between 3 and 4 cm long, passes posteriorly to the left from the neck of the gallbladder, and joins the common hepatic duct to form the common bile duct. It almost always runs parallel to, and is adherent to, the common hepatic duct for a short distance before joining it. The junction usually occurs near the porta hepatis but may be lower down in the free edge of the lesser omentum. The cystic duct may have several important variations in its anatomy. Rarely, the cystic duct lies along the right edge of the lesser omentum all the way down to the level of the duodenum before the junction is formed, but in these cases the cystic and common bile ducts are usually closely adherent. The cystic duct occasionally drains into the right hepatic duct in which case it may be elongated, lying anterior or posterior to the common hepatic duct, and joins the right hepatic duct on its left border. Rarely, the duct is double or even absent in which case the gallbladder drains directly into the common bile duct. One or more accessory hepatic ducts occasionally emerge from segment V of the liver and join either the right hepatic duct, the common hepatic duct, the common bile duct, the cystic duct, or the gallbladder directly. These variations in cystic duct anatomy are of considerable importance during surgical excision of the gallbladder. Ligation or clip occlusion of the cystic duct must be performed at an adequate distance from the common bile duct to prevent angulation or damage to it. Accessory ducts must not be confused with the right hepatic or common hepatic ducts. The mucosa of the cystic duct bears 5-12 crescentic folds, continuous with those in the neck of the gallbladder. They project obliquely in regular succession, appearing to form a spiral valve when the duct is cut in longitudinal section. When the duct is distended, the spaces between the folds dilate and externally it appears twisted like the neck of the gallbladder.
HEPATIC BILE DUCTS The main right and left hepatic ducts emerge from the liver and unite near the right end of the porta hepatis as the common hepatic duct. This descends about 3 cm before being joined on its right at an acute angle by the cystic duct to form the common bile duct. The common hepatic duct lies to the right of the hepatic artery and anterior to the portal vein in the free edge of the lesser omentum. UPDATE Date Added: 22 February 2005 Helen Elizabeth Wiggett, PhD (Dianthus Medical Limited) Update: Magnetic resonance cholangiopancreatography of anatomic variants of the biliary tree in Taiwanese patients Magnetic resonance cholangiopancreatography (MRCP), a new non-invasive method for evaluating the hepatobiliary and pancreatic ductal systems, was used to investigate the anatomic variants of the biliary tree in Taiwanese patients. Over a 29-month period, 170 patients with abdominal or liver diseases underwent diagnostic MRCP using a 1.5-T scanner (Signa, GE, Milwaukee, USA). Classification of enrolled patients was based on seven anatomic biliary tree variants (types I-VII) previously described by Koenraad and Pablo (2001). These variants differ with respect to the positions of the left hepatic duct (LHD), the right anterior segment duct (RASD), and the right posterior segment duct (RPSD). Among the 170 study patients, common diagnoses included gallstones (44%), common bile duct stenosis (20%), and common bile stones (15%). The most
common anatomic biliary tree variants were type I (30%) and type IIA (40%). In type I variants, the anatomy is characterized by simultaneous emptying of the RPSD, RASD, and LHD into the common hepatic duct. Type IIA is characterized by drainage of the RPSD into the RASD to form the right hepatic duct, and then confluence with the LHD to form the common hepatic duct. Of the remaining patients, 11.6% were type III, 1.5% were type IV, and 6.9% were type V. In addition, six patients had a biliary tree anatomy that was similar but not identical to type IIA and these patients were classified as either type IIB (five patients [3.8%]) or type IIC (one patient [0.8%]), and seven patients had a biliary tree anatomy that did not match any of the variants described by Koenraad and Pablo and therefore represented new variants. These new variants were called type VI (3.8%), type VII (0.8%), and type VIII (0.8%). None of the patients had variants that matched Koenraad and Pablo's type VI or VII. The results of this analysis show that the most common anatomic biliary tree variants in Taiwanese patients are type I and type IIA, which together describe 70% of the total number of patients. Since a balanced distribution of type I-type VII variants has previously been reported in the literature, the differences seen in this patient population may represent racial and ethnic variation. Evaluations of biliary anatomy are extremely important clinically, particularly before hepatic surgery, for staging and localization of intrahepatic liver neoplasms or bile duct tumors, and before complex biliary interventional procedures. These findings highlight racial and ethnic variations in biliary tree anatomy that must be borne in mind when considering these procedures in Taiwanese patients. Koenraad JM, Pablo RR. Anatomic variants of the biliary tree: MR cholangiographic findings and clinical applications. AJR Am J Roentgenol. 2001;177:389-94. Lee CM, Chen HC, Leung TK, Chen YY. Magnetic resonance cholangiopancreatography of anatomic variants of the biliary tree in Taiwanese., J Formos Med Assoc. 2004;103(2):155-9. Medline Similar articles
COMMON BILE DUCT (Figs 86.3, 86.4) The common bile duct is formed near the porta hepatis, by the junction of the cystic and common hepatic ducts. It is usually between 6 and 8 cm long. Its diameter tends to increase somewhat with age but is usually around 6 mm in adults. It descends posteriorly and slightly to the left, anterior to the epiploic foramen, in the right border of the lesser omentum. It lies anterior and to the right of the portal vein and to the right of the hepatic artery. It passes behind the first part of the duodenum with the gastroduodenal artery on its left, and then runs in a groove on the superolateral part of the posterior surface of the head of the pancreas (Fig. 87.1). It lies anterior to the inferior vena cava and is sometimes embedded in the pancreatic tissue. The duct may lie close to the medial wall of the second part of the duodenum or as much as 2 cm from it. Even when it is embedded in the pancreas, a groove in the gland marking its position can be palpated behind the second part of the duodenum. Hepatopancreatic ampulla (of Vater)
As it lies medial to the second part of the duodenum, the common bile duct approaches the right end of the pancreatic duct. The ducts enter the duodenal wall together, and usually unite to form the hepatopancreatic ampulla. Rarely the common bile duct and pancreatic duct drain into the duodenum separately. Circular muscle usually surrounds the lower part of the common bile duct (bile duct sphincter) and frequently also surrounds the terminal part of the main pancreatic duct (pancreatic duct sphincter) and the hepatopancreatic ampulla (sphincter of Oddi). When all elements are present, this arrangement may allow for separate control of pancreatic and common bile duct emptying. Division of the upper part of the ampulla and ampullary sphincter (sphincterotomy) may be required to allow access to the common bile duct during endoscopic retrograde cholangiography).
Figure 86.3 Endoscopic retrograde cholangiopancreatogram.
Figure 86.4 Magnetic resonance cholangiopancreatogram.
Calot's triangle page 1228 page 1229
The near triangular space formed between the cystic duct, the common hepatic duct and the inferior surface of segment V of the liver (Suzuki et al 2000), is commonly referred to as Calot's triangle. It is enclosed by the double layer of peritoneum which forms the short mesentery of the cystic duct. Since the two layers are not closely opposed, there is an appreciable amount of loose
connective tissue within the triangle. It is perhaps better described as a pyramidal 'space' with one apex lying at the junction of the cystic duct and fundus of the gallbladder, one at the porta hepatis, and two closer apices at the attachments of the gallbladder to the liver bed. The base of the triangle thus lies on the inferior surface of the liver. This space usually contains the cystic artery as it approaches the gallbladder, the cystic lymph node and lymphatics from the gallbladder, one or two small cystic veins, the autonomic nerves running to the gallbladder and some loose adipose tissue. It may contain any accessory ducts which drain into the gallbladder from the liver. Appreciation of the variations in ductal and arterial anatomy as they relate to the triangle are of considerable importance during excision of the gallbladder in order to avoid mistakenly ligating the common hepatic or common bile duct.
BILIARY STONES Gallstones usually form in the gallbladder. As the gallbladder empties, gallstones move towards the cystic duct. When small stones enter the cystic duct they may irritate the columnar mucosa which leads to spasm of the smooth muscle in the cystic duct wall. This spasm generates pain known as biliary colic, which is often very severe. The mucosal folds in the neck of the gallbladder and the cystic duct provide a common site of entrapment of gallstones. Stones occluding the neck of the gallbladder may cause a sterile distension of the gallbladder; providing the gallbladder has not undergone acute inflammation previously it remains nonfibrotic and readily distensible, and an enlarged fundus often becomes palpable below the costal margin. Stones lodged in the distal cystic duct may cause swelling in the tissues around the duct. Due to the close relationship of the distal cystic duct to the common hepatic duct, this swelling may give rise to secondary compression of the common hepatic duct with resultant partial obstruction to the flow of bile and the appearance of mild jaundice; so called 'Mirizzi syndrome'. Once stones have passed through the cystic duct they often become impacted at the junction of the common bile duct and pancreatic duct just proximal to the hepatopancreatic ampulla, producing obstructive jaundice.
ENDOSCOPIC CHOLANGIOGRAPHY The common bile duct may be accessed endoscopically from the duodenum for diagnostic cholangiography and therapeutic interventions. Due to the angled relationship of the distal common bile duct and pancreatic duct, direct cannulation of the bile duct may be difficult. When the duct lies in a more vertical position, embedded within the wall of the duodenum, a direct incision in the base of the hepatopancreatic ampulla and the adjacent wall of the duodenum (pre-cut sphincterotomy) may expose the duct to allow cannulation. This is occasionally associated with haemorrhage due to the duodenal wall vessels. Division of the smooth muscle fibres of the common bile duct sphincter may be necessary for endoscopic access to the bile duct but it often results in uncontrolled reflux of duodenal contents into the distal common bile duct, and this may result in recurrent biliary infections.
BILIARY DRAINAGE The proximity of the fundus of the gallbladder to the anterior abdominal wall provides a useful route of access for percutaneous drainage of a distended, obstructed gallbladder. It is rarely obscured by liver tissue, can often be accessed below the costal margin, and is most often performed under ultrasound guidance. Because of the nature of the cystic duct, drainage of the gallbladder is rarely adequate to decompress the biliary tree if it is blocked, and this must be achieved endoscopically, surgically or by a percutaneous, transhepatic approach. This latter technique requires the identification of dilated intrahepatic bile ducts, usually by ultrasound imaging with direct guided puncture via a right subcostal approach. Percutaneous access is obtained via segment III in the left lobe of the liver and via segments V and VI in the right lobe of the liver. UPDATE Date Added: 14 August 2006 Abstract: A study of the subvesical bile duct (duct of luschka) in resected liver specimens
Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? db=pubmed&cmd=Retrieve&dopt=AbstractPlus&list_uids=16830216&query_hl=18&itool=pubmed_docsum A study of the subvesical bile duct (duct of luschka) in resected liver specimens. Ko K, Kamiya J, Nagino M, et al: World J Surg 30:1316-1320, 2006.
© 2008 Elsevier
VASCULAR SUPPLY AND LYMPHATIC DRAINAGE CYSTIC ARTERY The cystic artery usually arises from the right hepatic artery. It usually passes posterior to the common hepatic duct and anterior to the cystic duct to reach the superior aspect of the neck of the gallbladder. It divides into superficial and deep branches. The superficial branch ramifies on the inferior aspect of the gallbladder body, the deep branch on the superior aspect. These arteries anastomose over the surface of the body and fundus. The cystic artery may arise from the common hepatic artery, sometimes from the left hepatic artery, and rarely from the gastroduodenal or superior mesenteric arteries. In these cases it crosses anterior (or less commonly posterior) to the common bile duct or common hepatic duct to reach the gallbladder. An accessory cystic artery may arise from the common hepatic artery or one of its branches and the cystic artery often bifurcates close to its origin, giving rise to two vessels which approach the gallbladder. Multiple fine arterial branches may arise from the parenchyma of segments IV or V of the liver and contribute to the supply of the body, particularly when the gallbladder is substantially intrahepatic. This makes the gallbladder relatively resistant to necrosis during inflammation which otherwise occludes the cystic artery. The cystic artery gives rise to multiple fine branches which supply the common and lobar hepatic ducts and upper part of the common bile duct. These fine branches form a network which anastomoses with the vessels ascending around the common bile duct and with the vessels from the liver parenchyma which descend with the right and left hepatic ducts. UPDATE Date Added: 22 March 2005 Helen Elizabeth Wiggett, PhD (Dianthus Medical Limited) Update: Anatomical variations of the cystic artery Identifying the origin of the cystic artery is important during surgery, particularly in controlling intraoperative or postoperative bleeding in the gallbladder fossa. This study has analyzed variations in the cystic artery using 60 corrosion casts taken from undamaged livers removed at autopsy and 21 livers from embalmed cadavers. In the majority of cases (70; 86%), the gallbladder was supplied by one cystic artery which most commonly arose from either the right hepatic artery (RHA; 52%) or the replacing RHA (originating from the superior mesenteric artery; 7%). In 21% of cases, the cystic artery arose from a sectional artery, either the anterior sectional artery (ASA), which supplies segments 5 and 6 (16%) or the posterior sectional artery (PSA), which supplies segments 6 and 7 (5%). In 6% of cases, the cystic artery arose from segmental arteries coming from either segment 4 (2.5%), 5 (1.2%), 6 (1.2%), or 8 (1.2%). The gallbladder was supplied by two cystic arteries in 14% of cases, both of which arose from either the RHA (7.4%) or the ASA (2.5%), or from the RHA and either the ASA (2.5%) or the left hepatic artery (1.2%), respectively. In two cases, a branch of the cystic artery (subsegmental artery) was identified that lead to either segment 5 or segment 6. The authors identified 12 cystic artery variants: in the majority of cases, one cystic artery was present, and this arose most commonly from the RHA. Mlakar B, Gadzijev EM, Ravnik D, Hribernik M. Anatomical variations of the cystic artery. Eur J Morphol. 2003 Feb;41(1):31-4.
DUCTAL ARTERIES The common bile duct and hepatic ducts are supplied by a fine network of vessels, which lie in close proximity to the ducts themselves. This network usually
vessels, which lie in close proximity to the ducts themselves. This network usually has contributions from several sources. Disruption of the network during surgical exposure of the bile ducts over a long length frequently causes chronic ischaemia and a resultant stenosis of the duct. Approaches which spare the network are necessary to avoid this complication. Anterior to the common bile duct, two to four ascending vessels arise from the retroduodenal branch of the gastroduodenal artery as it crosses the anterior surface of the duct at the upper border of the duodenum. Three or four descending branches of the right hepatic and cystic arteries arise as these vessels pass close to the lower common hepatic duct. These ascending and descending arteries form long narrow anastomotic channels along the length of the duct, which are approximately disposed into medial and lateral 'trunks' although they may lie more anterolateral and posteromedial. Posteriorly, a retroportal artery often arises from the coeliac axis, superior mesenteric artery or one of their major branches close to the origin from the aorta. It runs upwards on the posterior surface of the portal vein. It usually ends by joining the retroduodenal artery close to the lower end of the supraduodenal bile duct, but occasionally it passes up behind the bile duct to join the right hepatic artery. When present, the retroportal artery contributes to the arterial network supplying the supraduodenal bile duct system.
CYSTIC VEINS The venous drainage of the gallbladder is rarely by a single cystic vein. There are usually multiple small veins. Those arising from the superior surface of the body and neck lie in areolar tissue between the gallbladder and liver and enter the liver parenchyma to drain into the segmental portal veins. The remainder form one or two small cystic veins, which enter the liver either directly or after joining the veins draining the hepatic ducts and upper bile duct. Only rarely does a single or double cystic vein drain into the right portal branch.
LYMPHATICS Numerous lymphatic vessels run from the submucosal and subserosal plexuses on all aspects of the gallbladder and cystic duct. Those on the hepatic aspect of the gallbladder connect with the intrahepatic lymph vessels. The remainder drain into the cystic node, which usually lies above the cystic duct in the tissue of Calot's triangle. This node, and some lymphatic channels which bypass the cystic node, drain into a node lying in the anterior border of the free edge of the lesser omentum. Hepatic nodes lying in the porta hepatis collect lymph from vessels accompanying the hepatic ducts and the upper part of the bile duct. Lymphatics from the lower part of the common bile duct drain into the inferior hepatic and upper pancreaticosplenic nodes.
© 2008 Elsevier
INNERVATION page 1229 page 1230
The gallbladder and the extrahepatic biliary tree are innervated by branches from the hepatic plexus (p. 1222). The retroduodenal part of the common bile duct also has contributions from the pyloric branches of the vagi, which also innervate the smooth muscle of the hepatopancreatic ampulla.
REFERRED PAIN In common with other structures of foregut origin, pain from stretch of the common bile duct or gallbladder is referred to the central epigastrium. Involvement of the overlying somatic peritoneum produces pain which is more localized to the right upper quadrant.
© 2008 Elsevier
MICROSTRUCTURE GALLBLADDER (Fig. 86.5) The fundus of the gallbladder is covered by a serosa (p. 41), but this only covers the inferior surfaces and sides of the body and neck of the gallbladder unless the gallbladder is mesenteric. Beneath it is subserous loose connective and adipose peritoneal tissue. The gallbladder wall microstructure generally resembles that of the small intestine. The mucosa is yellowish-brown and elevated into minute rugae with a honeycomb appearance (Fig. 86.2). In section, projections of the mucosa into the gallbladder lumen resemble intestinal villi, but these are not fixed structures and the surface flattens as the gallbladder fills with bile. Its epithelium is a single layer of columnar absorptive cells bearing apical microvilli. Goblet cells are absent. Basally, the spaces between epithelial cells are dilated. Many capillaries lie beneath the basement membrane. The epithelial cells actively absorb water and solutes from the bile to concentrate it, up to ten-fold. The thin fibromuscular layer is composed of fibrous tissue mixed with smooth muscle cells which are arranged loosely in longitudinal, circular and oblique bundles.
BILE DUCTS
Figure 86.5 Low power micrograph showing the gallbladder wall, with a mucosal projection which flattens in the full gallbladder, and the thin muscular layer. (Photograph by Sarah-Jane Smith.)
The large biliary ducts have external fibrous and internal mucosal layers. The former is fibrous connective tissue which contains a variable amount of
longitudinal, oblique and circular smooth muscle cells. The mucosa is continuous with that of the hepatic ducts, gallbladder and duodenum. The epithelium is columnar. Many tubuloalveolar mucous glands occur in the walls of these ducts. Expulsion of gallbladder contents is under neuroendocrine control. Fat in the duodenum causes the release of cholecystokinin (CCK), stimulating the gallbladder to contract because muscle cells in its walls have surface receptors for CCK. When the pressure exceeds 100 mm of bile, the sphincter of Oddi relaxes and bile enters the duodenum. The termination of the united bile and pancreatic ducts is packed with villous, valvular folds of mucosa with muscle cells in their connective tissue cores. Contraction is thought to result in retraction and clumping of the folds, preventing reflux of duodenal contents and controlling the exit of bile. REFERENCES Suzuki M, Akaishi S, Rikiyama T, Naitoh T, Rahman MM, Matsuno S. 2000. Laparoscopic cholecystectomy, calot's triangle, and variations in cystic arterial supply. Surgical Endoscopy 14: 141-4. Medline Similar articles IHPBA Brisbane 2000 Terminology of Liver Anatomy and Restrictions.
© 2008 Elsevier
87 PANCREAS, SPLEEN AND SUPRARENAL GLAND Pancreas UPDATE Date Added: 02 February 2005 Abstract: Surgical anatomy of the pancreas for limited resection. Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=11180873 Surgical anatomy of the pancreas for limited resection. Kimura W. Surgical anatomy of the pancreas for limited resection. J Hepatobiliary Pancreat Surg. 2000; 7(5): 473-9. Medline Similar articles
The pancreas is the largest of the digestive glands and performs a range of both endocrine and exocrine functions. The major part of the gland is exocrine, secreting a range of enzymes which are involved in the digestion of lipids, carbohydrates and proteins. The endocrine function of the pancreas is derived from cells scattered throughout the substance of the gland. They take part in glucose homeostasis as well as being involved in the control of upper gastrointestinal motility and function. The pancreas is salmon pink in colour with a firm, lobulated smooth surface. The main portion of the pancreas is divided into four parts - head, neck, body and tail and it possesses one accessory lobe (the uncinate process) (Fig. 87.1). The division into the parts is purely on the basis of anatomical relations and there are only very minor functional or anatomical differences between them. The uncinate process is an anatomically and embryologically distinct portion of the pancreas. In adults the pancreas measures between 12 and 15 cm long and is shaped as a flattened 'tongue' of tissue, thicker at its medial end (head) and thinner towards the lateral end (tail). With age, the amount of exocrine tissue tends to decline, as does the amount of fatty connective tissue within the substance of the gland, and this leads to a progressive thinning atrophy which is particularly noticeable on CT scanning. The pancreas lies within the curve of the first, second and third parts of the duodenum, and extends transversely and slightly upwards across the posterior abdominal wall to the hilum of the spleen, behind the stomach. It does not lie in one plane. It is effectively 'draped' over the other structures in the retroperitoneum and the vertebral column and so forms a distinct shallow curve, the neck and medial body being the most anterior parts. Because of its flattened shape, the parts of the pancreas, particularly the body, are often referred to as having surfaces and borders. UPDATE Date Added: 05 April 2005 Shanida Helena Nataraja, PhD (Dianthus Medical Limited) Update: Diagnostic accuracy of image analysis in patients with anomalous pancreaticobiliary junction (APBJ) Anomalous pancreaticobiliary junction (APBJ) is frequently associated with biliary tract carcinoma and with pancreatic disorders such as acute pancreatitis. This paper is a retrospective evaluation of image analysis in the diagnosis of 64 patients with APBJ who were treated at the Beijing Tongren Hospital from February 1979 through to August 2001. All patients were subjected to the following assessments before surgery: ultrasound imaging, endoscopic retrograde cholangiopancreatography (ERCP), and magnetic resonance cholangiopancreatography (MRCP). The diagnostic accuracy of these assessments was reviewed retrospectively, as were the outcomes of surgical intervention. Of the enrolled patients, 22 were men and 42 women, and their mean age was 46.8 years (range: 15 to 70 years). Thirty-two (50%) of patients had congenital choledochal cyst, 40 (62.5%) patients were associated with cholelithiasis, and 21 (32.8%) patients were associated with biliary tract neoplasm. Both ERCP and MRCP revealed that the mean length of the common
channel was 19 mm (range: 17 to 49 mm), and that the junction was greater than 75° in 49 (76.6%) patients. Of the enrolled patients, 29 were defined as being of bile duct type (C-P), 4 were defined as common channel type, and 28 were defined as pancreatic duct type (P-C). Patients with congenital choledochal cyst underwent excision of the dilated extrahepatic bile duct and Roux-en-Y hepaticojejunostomy. Patients associated with cholelithiasis were treated by cholecystectomy. The findings of this study therefore indicate that both ERCP and MRCP can accurately diagnose APBJ. Once a diagnosis has been established, patients with congenital choledochal cyst should be treated with cyst excision and hepaticojejunostomy, whereas cholecystectomy alone may be sufficient in other patients, both with and without cholecystolithiasis. Yu ZL, Zhang LJ, Fu JZ, Li J, Zhang QY, Chen FL. Anomalous pancreaticobiliary junction: image analysis and treatment principles. Hepatobiliary Pancreat Dis Int. 2004;3(1):136-9. Medline Similar articles
Figure 87.1 CT scan of pancreas.
HEAD The head of the pancreas lies to the right of the midline, anterior and to the right side of the vertebral column. It is the thickest and broadest part of the pancreas but is still flattened in the anteroposterior plane. It lies within the curve of the duodenum. Superiorly it lies adjacent to the first part of the duodenum but close to the pylorus the duodenum is on a short mesentery, and here the duodenum lies anterior to the upper part of the head (p. 1163). The duodenal border of the head is flattened and slightly concave, and is firmly adherent to the second part of the duodenum. Occasionally a small part of the head is actually embedded in the wall of the second part of the duodenum. The superior and inferior pancreaticoduodenal arteries lie between the head and the duodenum in this area. The inferior border lies superior to the third part of the duodenum and is continuous with the uncinate process (p. 1233). Close to the midline, the head is continuous with the neck. The boundary between head and neck is often marked anteriorly by a groove for the gastroduodenal artery and posteriorly by a similar but deeper deep groove containing the union of the superior mesenteric and splenic veins to form the portal vein. UPDATE Date Added: 02 February 2005 Abstract: Study of surgical anatomy for duodenum-preserving resection of the
head of the pancreas. Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=7726671 Study of surgical anatomy for duodenum-preserving resection of the head of the pancreas. Anterior surface (Fig. 87.2) The anterior surface of the head is covered in peritoneum and is related to the origin of the transverse mesocolon. Posterior surface The posterior surface of the head is related to the inferior vena cava, which ascends behind it and covers almost all of this aspect. It is also related to the right renal vein and the right crus of the diaphragm.
NECK The neck of the pancreas is only c.2 cm wide and links the head and body. It is often the most anterior portion of the gland. It is defined as that portion of the pancreas which lies anterior to the portal vein, and is closely related to the upper posterior surface. The lower part of the neck lies anterior to the superior mesenteric vein just before the formation of the portal vein. This is important during surgery for pancreatic cancer since malignant involvement of these vessels may make resection impossible. The anterior surface of the neck is covered with peritoneum. It lies adjacent to the pylorus just inferior to the epiploic foramen. The gastroduodenal and anterior superior pancreaticoduodenal arteries descend in front of the gland in the region of the junction of the neck and head.
BODY The body of the pancreas runs from the left side of the neck to the tail. It is the longest portion of the gland and becomes progressively thinner and less broad towards the tail. It is slightly triangular in cross-section and is described as having three surfaces: anterosuperior, posterior and anteroinferior. Anterosuperior surface The anterosuperior surface of the pancreas makes up most of the anterior aspect of the gland close to the neck. Laterally, it narrows and lies slightly more superiorly to share the anterior aspect with the anteroinferior surface. It is covered by peritoneum, which runs anteroinferiorly from the surface of the gland to be continuous with the anterior, ascending layer of the greater omentum (p. 1132), (Fig. 69.4). It is separated from the stomach by the lesser sac. Posterior surface (Fig. 87.3)
page 1231 page 1232
Figure 87.2 Anterior relations of the pancreas. The extent of the lesser sac varies slightly between individuals. The splenic flexure may lie below or anterior to part of the tail.
Figure 87.3 Posterior relations of the pancreas. The posterior surfaces of the pancreas with their relations (viewed from behind).
The posterior surface of the pancreas is devoid of peritoneum. It lies anterior to the aorta and the origin of the superior mesenteric artery, the left crus of the diaphragm, left suprarenal gland and the left kidney and renal vessels, particularly the left renal vein. It is closely related to the splenic vein which runs from left to right forming a shallow groove in the gland. The splenic vein lies between the posterior surface and the other posterior relations. The left kidney is also separated from the posterior surface by perirenal fascia and fat. Anteroinferior surface The anteroinferior surface of the pancreas begins as a narrow strip just to the left of the neck. As the body runs laterally, it broadens out to form more of the anterior aspect of the body. It is covered by peritoneum which is continuous with that of the posteroinferior layer of the transverse mesocolon. The fourth part of the duodenum, the duodenojejunal flexure and coils of jejunum lie inferiorly. The lateral end of the inferior border often lies superior and posterior to the splenic flexure. The peritoneum of the anterosuperior layer of the transverse mesocolon is reflected onto the upper part of the anteroinferior surface.
Superior border On the right side the superior border of the pancreas is initially blunt and somewhat flat. As the gland is followed to the left, the surface changes to become narrower and sharper. An omental tuberosity usually projects from the right end of the superior border above the level of the lesser curvature of the stomach, in contact with the posterior surface of the lesser omentum. The superior border is related to the coeliac artery. The common hepatic artery runs to the right just above the gland, the splenic artery runs to the left along the superior border. The course of the artery is often highly tortuous and it tends to rise above the level of the superior border at several points along its course. Anterior border The anterior border of the pancreas separates the anterosuperior from the anteroinferior surfaces. The two layers of the transverse mesocolon diverge along this border. One passes up over the anterosuperior surface whilst the other runs downwards and backwards over the anteroinferior surface. Inferior border The inferior border of the pancreas separates the posterior from the anteroinferior surfaces. At the medial end of the inferior border, adjacent to the neck of the pancreas, the superior mesenteric vessels emerge from behind the gland. More laterally, the inferior mesenteric vein runs under the border to join the splenic vein on the posterior surface. This is a useful site of identification of the inferior mesenteric vein during left-sided colonic resections and on CT imaging.
TAIL The tail of the pancreas is the narrowest, most lateral portion of the gland and lies between the layers of the splenorenal ligament (p. 1132). It is continuous medially with the body and is between 1.5 and 3.5 cm long in adults. It may finish at the base of the splenorenal ligament or extend up nearly as far as the splenic hilum, in which case it is prone to injury at splenectomy during ligation of the splenic vessels. Posteriorly it is related to the splenic branches of the splenic artery and the splenic vein and its tributaries. The tip of the tail may lie in contact with the splenic hilum. page 1232 page 1233
UNCINATE PROCESS The uncinate process of the pancreas extends from the inferior lateral end of the head of the gland. It is embryologically separate from the rest of the gland, and as a consequence of its development it lies posterior to the superior mesenteric vessels. These lie in close contact to its anterior surface as they descend and run forward into the root of the ileal mesentery. Posteriorly it lies in front of the aorta, and inferiorly it lies on the upper surface of the third part of the duodenum. Tumours of the uncinate process do not cause obstruction to the common bile duct but frequently compress the third part of the duodenum as a result of this close relationship.
© 2008 Elsevier
PANCREATIC DUCTS The exocrine pancreatic tissue drains into multiple small lobular ducts, which drain into a single main, and usually, a single accessory duct (Fig. 86.3). The main pancreatic duct runs within the substances of the gland from left to right. It tends to lie more towards the posterior than anterior surface. It is formed by the junction of several lobular ducts in the tail. As it runs within the body it increases in calibre as it receives further lobular ducts, which join it almost at right angles to the axis of the main duct to form a 'herringbone pattern'. On ultrasound the duct can often be demonstrated, measuring c.3 mm in diameter in the head, 2 mm in the body, and 1 mm in the tail in adults. As it reaches the neck of the gland it usually turns inferiorly and posteriorly towards the bile duct, which lies on its right side. The two ducts enter the wall of the descending part of the duodenum obliquely and unite in a short dilated hepatopancreatic ampulla (p. 1163). A separate accessory pancreatic duct usually drains the lower part of the head and uncinate process. It is much smaller in calibre than the main duct and forms within the substance of the head from several lobular ducts. It ascends anterior to the main duct and usually communicates with it through several small branches. The accessory duct occasionally opens onto a small rounded minor duodenal papilla, which lies about 2 cm anterosuperior to the major papilla. If the duodenal end of the accessory duct fails to develop, the duct drains along the connecting channels into the main duct. The main and accessory pancreatic ducts demonstrate some variability in their anatomy (Fig. 87.4). Occasionally the accessory duct is absent and the main duct drains the uncinate process directly. The main duct may drain directly into the duodenum and the uncinate process drains via an accessory duct. Rarely the two ducts are conjoined.
© 2008 Elsevier
VASCULAR SUPPLY AND LYMPHATIC DRAINAGE ARTERIES (Fig. 87.5) The pancreas has a rich arterial supply derived from the coeliac axis and superior mesenteric arteries via both named vessels and multiple small un-named vessels. UPDATE Abstract: Pancreas head carcinoma
Date Added: 05 June 2006
Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=16006800&query_hl=8&itool=pubmed_docsum Pancreas head carcinoma: Frequency of invasion to soft tissue adherent to the superior mesenteric artery. Noto M, Miwa K, Kitagawa H et al: Am J Surg Pathol 29:1056-1061, 2005. Inferior pancreaticoduodenal artery
The inferior pancreaticoduodenal artery arises from the superior mesenteric artery or its first jejunal branch, near the superior border of the third part of the duodenum. It usually divides directly into anterior and posterior branches. The anterior branch passes to the right, anterior to the lower border of the head of the pancreas, and runs superiorly to anastomose with the anterior superior pancreaticoduodenal artery. The posterior branch runs posteriorly and superiorly to the right, lying posterior to the lower border of the head of the pancreas and anastomoses with the posterior superior pancreaticoduodenal artery. Both branches supply the pancreatic head, its uncinate process and the second and third parts of the duodenum. Superior pancreaticoduodenal artery
Figure 87.4 Variations in the ductal anatomy of the pancreas.
The superior pancreaticoduodenal artery is usually double. The anterior artery is a terminal branch of the gastroduodenal artery and descends in the anterior groove between the second part of the duodenum and head of the pancreas. It supplies branches to the head of the pancreas. It anastomoses with the anterior division of the inferior pancreaticoduodenal artery. The posterior artery is usually a separate branch of the gastroduodenal artery arising at the upper border of the first part of the duodenum. It descends to the right, anterior to the portal vein and common bile duct, where the duct passes behind the first part of the duodenum. The artery runs posterior to the head of the pancreas and then crosses posterior to the common bile duct embedded in the head of the pancreas. It enters the duodenal wall and anastomoses with the posterior division of the inferior
pancreaticoduodenal artery. The posterior superior artery supplies branches to the head of the pancreas and the first and second parts of the duodenum. Pancreatic branches page 1233 page 1234
Figure 87.5 Arterial supply of the pancreas.
The pancreas is supplied by numerous small arterial branches which usually run into the gland directly from their arteries of origin. These are particularly numerous in the region of the neck, body and tail. Most originate from the splenic artery as it runs along the superior border of the gland and supply the left part of the body and tail. A dorsal branch descends posterior to the pancreas, dividing into right and left branches. It sometimes arises from the superior mesenteric, middle colic, hepatic or rarely, the coeliac artery. The right branch is often double and runs between the neck and uncinate process to form a prepancreatic arterial arch as it anastomoses with a branch from the anterior superior pancreaticoduodenal artery. The left branch runs along the inferior border to the pancreatic tail where it anastomoses with the greater pancreatic artery (arteria pancreatica magna) and the artery to tail of the pancreas (arteria caudae pancreatis). Small un-named branches also arise from the first jejunal arcade of the superior mesenteric artery and the arterial branches of the retroperitoneal vessels. Small arteries characteristically run along the inferior and superior borders of the gland, either lying in a deep groove or within the tissue of the gland. They supply branches, which penetrate the substance of the gland at right angles to the vessel and receive contributions from the arteries supplying the gland but mainly from the inferior and superior pancreaticoduodenal arteries. They may bleed profusely on cutting the parenchyma of the gland during resection and usually require ligation.
VEINS The venous drainage of the pancreas is primarily into the portal system. The head and neck drain primarily via superior and inferior pancreaticoduodenal veins (p. 1165). The body and tail drain mostly via small veins running directly into the splenic vein along the posterior aspect of the gland or occasionally directly into the portal vein. Small venous channels exist between the gland and the retroperitoneal veins, draining into the lumbar veins and these may hypertrophy and become clinically significant in cases of portal hypertension.
LYMPHATICS Lymph capillaries commence around the pancreatic acini. The larger lymph vessels follow the arterial supply and drain into the lymph nodes around the pancreas and adjacent node groups. The tail and body lymphatics drain mostly into the pancreaticosplenic nodes although some drain directly to pre-aortic nodes. Lymphatics from the neck and head drain more widely into nodes along the pancreaticoduodenal, superior mesenteric and hepatic arteries. Drainage also occurs to the pre-aortic nodes and coeliac axis nodes. There are no lymphatics in the pancreatic islets.
© 2008 Elsevier
INNERVATION OF THE EXOCRINE PANCREAS The exocrine lobules of the pancreas are innervated by a fine network of sympathetic and parasympathetic fibres. The sympathetic supply originates from the sixth to tenth thoracic spinal segments and is mainly distributed to the pancreas via the sympathetic contribution to the coeliac ganglia. The postganglionic fibres are distributed to the gland via the arterial supply as periarterial plexuses. The parasympathetic supply is from the posterior vagus nerve and the parasympathetic component of the coeliac plexus. The supply to the gland is both vasomotor (sympathetic) and parenchymal (sympathetic and parasympathetic) in distribution. The exocrine lobules are innervated by a fine network of parasympathetic and sympathetic fibres. Sensory fibres running from the gland run in both the sympathetic and parasympathetic systems. These mediate the sensation of pain arising from the gland and may also carry other sensory information. In chronic inflammation or inoperable tumours of the gland, thermal or chemical ablation of the coeliac plexus may be required to control chronic pain mediated by these fibres.
REFERRED PAIN Pain arising in the pancreas is poorly localized. In common with other foregut structures, the majority of pain arising from the pancreas is referred to the epigastrium. Inflammatory or infiltrative processes arising from the gland rapidly involve the tissues of the retroperitoneum and their supply from somatic nerves, and this is referred to the posterior paravertebral region around the lower thoracic spine.
© 2008 Elsevier
PANCREATITIS AND PSEUDOCYST Pancreatitis is one of the major pathological processes affecting the pancreas. Gallstones lying within the common bile duct are associated with pancreatitis. The presence of a common drainage for the common bile duct and the pancreatic duct may allow reflux of bile or pancreatic enzymes into the pancreatic duct during the passage of a gallstone through the ampulla. This may also occur due to oedema of the common bile duct wall even when the gallstones have not entered the common ampulla. Inflammation in the pancreas may cause a range of secondary pathologies. The course of the superior mesenteric artery and vein behind the neck and between the inferior border and uncinate process makes these vessels vulnerable to compression and secondary inflammation, which may result in an inflammatory aneurysm of the superior mesenteric artery or thrombosis of the superior mesenteric vein. Inflammatory aneurysms may rupture producing major haemorrhage. Thrombosis of the superior mesenteric vein may cause potentially lethal venous ischaemia of the small intestine. Thrombosis of smaller arterial branches such as the origin of the middle colic artery may also occur, causing ischaemia of individual organs such as the transverse colon. page 1234 page 1235
page 1235 page 1236
Figure 87.6 Microstructure of the exocrine pancreas and the mechanisms by which its secretion is controlled.
The profuse arterial and venous supply to the gland makes it particularly prone to haemorrhage. The extravasated blood collects in the retroperitoneal tissues as the neck and body of the gland lie largely in the loose connective tissue of the retroperitoneum. The pancreas lies anterior to the thoracolumbar and perirenal fasciae and the blood can track freely in the retroperitoneal tissues to appear either in the flanks - so-called Grey-Turner's sign - in the groins, or above the iliac crest where the iliac fascia is attached. Blood tracking laterally from the head of the pancreas may enter the lesser omentum and 'bare area' of the liver, from where it may run forward into the falciform ligament and appear around the skin of the umbilicus - so-called Cullen's sign. During acute episodes of inflammation, the close anterior relationship of the stomach may contribute to gastric stasis and vomiting. The origin of the superior mesenteric plexus also lies close to the pancreas and secondary inflammation in the tissues around the pancreas may affect the autonomic supply to the midgut and contribute to the paralytic ileus which frequently develops. In severe cases, pancreatic inflammation may cause the collection of fluid within and around the pancreatic tissue. Intrapancreatic collections frequently resolve spontaneously over time but coalescence of the fluid may occur anterior to the pancreas beneath the layer of peritoneum covering its anterior surfaces, although this actually lies beneath the posterior wall of the lesser sac. If this collection persists and grows, the peritoneum anterior and superior to the pancreas is stretched and comes to lie in contact with the anterior wall of the lesser sac. This collection is referred to as a pseudocyst. The anterior wall of the pseudocyst is formed of the twin layers of peritoneum lying adjacent to the posterior wall of the stomach, the lesser omentum and, occasionally, the gastrosplenic ligament. The lateral wall of the pseudocyst includes the splenorenal ligament. The posterior wall is a mixture of fibrous tissue resulting from previous inflammation, the anterior surface of the pancreas and the retroperitoneal tissues. Treatment of the pseudocyst usually involves drainage of the contents into the lumen of the stomach. This may be established endoscopically by placement of a drain through the posterior stomach wall and the two thickened layers of peritoneum into the cyst. Alternatively, the drain may be placed using radiological guidance, via the anterior abdominal wall and the anterior stomach wall.
© 2008 Elsevier
PANCREATIC RESECTION Resection of the pancreas is complicated by several factors. The extensive vascular supply requires careful haemostasis. Spleen-preserving resections may be undertaken if the underlying pathology does not involve the splenic vein as it lies in the groove on the posterior surface of the gland, although the multiple small pancreatic veins draining into it may cause troublesome bleeding. Resection of the head and neck is possible provided the plane between the neck and the portal vein has not been involved by disease. Occasional small venous branches may enter the portal vein directly and may also cause bleeding during this mobilization. Resection of the head and neck are always accompanied by resection of the distal first and second parts of the duodenum because of the dense adherence between the two and the common arterial supply. Resection without removal of the proximal part of the first part of the duodenum and pylorus - pylorus preserving pancreatectomy - may be possible provided the arterial supply to the pylorus from the stomach, and directly from the pre-pyloric vessels, is adequate.
© 2008 Elsevier
MICROSTRUCTURE (Figs 87.6, 87.7, 87.8) The pancreas is composed of two different types of glandular tissue. The main tissue mass is exocrine (p. 34), in which are embedded pancreatic islets of endocrine cells (p. 34).
EXOCRINE PANCREAS
Figure 87.7 Pancreatic tissue. Exocrine acinar cells (A) are deeply stained basally, indicating the high ribosomal concentration. A small duct (D) is shown. An endocrine islet (of Langerhans) (I) is shown centrally, with pale-staining cells surrounded by a network of capillaries, seen as clear spaces. Connective tissue septa (C) separate lobules. (By permission from Dr JB Kerr, Monash University, from Kerr JB 1999 Atlas of Functional Histology. London: Mosby.)
The exocrine pancreas is a branched acinar gland, surrounded and incompletely lobulated by delicate loose connective tissue. It is formed of pyramidal, secretory cells arranged mainly as spherical clusters, or acini. A narrow intercalated, intralobular duct originates within each secretory acinus, lined initially by flattened or cuboidal centro-acinar cells. These small ductules form branching links which run within and between adjacent acini, explaining why structurally distinct intralobular pancreatic ducts are infrequent (see Kerr 1999, for details). More distally, these are replaced by taller cuboidal and eventually columnar epithelium in the larger interlobular ducts. The latter are surrounded by loose connective tissue of the septa, containing smooth muscle and autonomic nerve fibres. Neuroendocrine cells are present amongst the columnar ductal cells and mast cells are numerous in the surrounding connective tissue. Acinar cells
Acinar cells of the exocrine pancreas have a basal nucleus and, in their basal cytoplasmic domain, abundant rough endoplasmic reticulum which results in their basophilic staining characteristics. Dense secretory zymogen granules stain deeply with eosin in the apical region. A prominent supranuclear Golgi complex is surrounded by large, membrane-bound granules containing the proteinaceous constituents of pancreatic secretion, including enzymes which are only active after release. Ganglionic neurones and cords of undifferentiated epithelial cells are also found within the acini. The structure of the exocrine pancreas and its functional regulation are summarized in Fig. 87.6.
ENDOCRINE PANCREAS (Fig. 87.8) The endocrine pancreas consists of pancreatic islets of Langerhans, composed of spherical or ellipsoid clusters of cells embedded in the exocrine tissue. The human pancreas may contain more than a million islets, usually most numerous in the tail. An islet is a mass of polyhedral cells, each in close proximity to fenestrated capillaries and a rich autonomic innervation. Specialized staining procedures or immunohistochemical techniques are necessary to distinguish the three major types of cell, designated alpha, beta and delta. Their general organization is shown in Fig. 87.8. page 1236 page 1237
Figure 87.8 Microstructure and control of function of the endocrine pancreas.
The most numerous cells, types alpha and beta, secrete glucagon and insulin respectively. Alpha cells tend to be concentrated at the periphery of islets, and beta cells more centrally. A third type, the delta cell, secretes somatostatin and gastrin, and like alpha cells, is peripherally placed within the islets. A minor cell type, the F cell, secretes pancreatic polypeptide (PP), which is stored in smaller secretory granules. The autonomic transmitters acetylcholine (ACh) and noradrenalin affect islet cell secretion. ACh augments insulin and glucagon release, noradrenalin inhibits glucose-induced insulin release and they may also affect somatostatin and PP secretion. Innervation of endocrine pancreas
The innervation of the endocrine islets is almost exclusively from the parasympathetic system. Fine branches ramify among the cells and form plexuses around the islets. Fibres frequently synapse with acinar cells before innervating the islets, suggesting a close linkage between neural control of exocrine and endocrine components. Many fibres enter the islets with the arterioles. Parasympathetic ganglia lie in the connective tissue within and between lobules, and in the former case are frequently associated with islet cells, forming neuroinsular complexes. Both alpha and beta cells are involved in these neuroinsular complexes. Three types of nerve terminal are seen in islets.
Cholinergic terminals have agranular vesicles with a diameter of 30-50 nm, adrenergic terminals have dense-cored vesicles with a diameter of 30-50 nm and a third, uncharacterized, type have dense-cored vesicles with a diameter of 60200 nm (Smith & Porte 1976). No selective link with any one type of insular cell has been found. Sometimes more than one type of terminal contacts a single cell and some of the chemical synapses between axon terminals and islet cells show narrow areas in the synaptic clefts suggesting an electrical synapse or gap junction. Such junctions also occur between islet cells, and electrical coupling of nerve supply to a functional network of islet cells may occur (Orci 1974). REFERENCES Kerr JB 1999 Atlas of Functional Histology, Chapter 14. London: Mosby. Orci L 1974 A portrait of a pancreatic B-cell. Diabetologia 10: 163-87. Medline Full article
Similar articles
Smith PH, Porte D Jr 1976 Neuropharmacology of the pancreatic islets. Annu Rev Pharmacol Toxicol 16: 269-85. Medline Similar articles page 1237 page 1238
© 2008 Elsevier
88 PANCREAS, SPLEEN AND SUPRARENAL GLAND Spleen (Fig. 88.1) The spleen consists of a large encapsulated mass of vascular and lymphoid tissue situated in the upper left quadrant of the abdominal cavity between the fundus of the stomach and the diaphragm. Its shape varies from a slightly curved wedge to a 'domed' tetrahedron. The shape is mostly determined by its relations to neighbouring structures during development. The superolateral aspect is shaped by the left dome of the diaphragm with the inferomedial aspect being influenced mostly by the neighbouring splenic flexure of the colon, the right kidney and stomach. Its long axis lies approximately in the plane of the tenth rib. Its posterior border is c.4 cm from the mid-dorsal line at the level of the tenth thoracic vertebral spine. Its anterior border usually reaches the mid-axillary line. The size and weight of the spleen vary with age and between the sexes. It can also vary slightly in the same individual under different conditions. In the adult it is usually c.12 cm long, 7 cm broad and between 3 and 4 cm wide. It is comparatively largest in the young child, and although its weight increases during puberty, by adulthood it is relatively smaller in comparison to the neighbouring organs. It tends to diminish in size and weight in senescence. Its average adult weight is about 150 g although the normal range is wide, between 80 g and 300 g, in part reflecting the amount of blood it contains. Additional collections of fully functional splenic tissue may exist near the spleen, especially within the gastrosplenic ligament and greater omentum. These accessory spleens, or spleniculi, are usually isolated but can be connected to the spleen by thin bands of similar tissue. They may be numerous and widely scattered in the abdomen. The spleen may retain its fetal lobulated form or show deep notches on its diaphragmatic surface and inferior border in addition to those usually present on the superior border.
Figure 88.1 The visceral surface of the spleen.
RELATIONS The spleen has a superolateral diaphragmatic and an inferomedial visceral surface. There are superior and inferior borders and anterior and posterior extremities or poles. The diaphragmatic surface is convex and smooth and faces mostly superiorly and laterally although the posterior part may face posteriorly and almost medially as it approaches the inferior border. The diaphragmatic surface is related to the abdominal surface of the left dome of the diaphragm which separates it from the basal pleura, the lower lobe of the left lung and the ninth to eleventh left ribs. The pleural costodiaphragmatic recess extends down as far as its inferior border. The visceral surface faces inferomedially towards the
abdominal cavity and is irregular. It is marked by gastric, renal, pancreatic and colic impressions. The gastric impression faces anteromedially and is broad and concave where the spleen lies adjacent to the posterior aspect of the fundus, upper body and upper greater curvature of the stomach. It is separated from the stomach by a peritoneal recess, which is limited by the gastrosplenic ligament. The renal impression is slightly concave and lies on the lowest part of the visceral surface. It is separated from the gastric impression above by a raised strip of splenic tissue and the splenic hilum. It faces inferomedially and slightly backwards, being related to the upper and lateral area of the anterior surface of the left kidney and sometimes to the superior pole of the left suprarenal gland. The colic impression lies at the inferior pole of the spleen and is usually flat. It is related to the splenic flexure of the colon and the phrenicocolic ligament. The pancreatic impression is often small when present and lies between the colic impression and the lateral part of the hilum. It is related to the tail of the pancreas which lies in the splenorenal ligament. The hilum of the spleen lies in the visceral surface closer to the inferior border and anterior extremity. It is a long fissure pierced by several irregular apertures through which the branches of the splenic artery and vein as well as nerves and lymphatics enter and leave the spleen. The superior border separates the diaphragmatic surface from the gastric impression and is usually convex. Near the anterior extremity there may be one or two notches persisting from the lobulated form of the spleen in early fetal life. These notches are often absent and are not a reliable guide to the identification of the spleen during clinical examination. The inferior border separates the renal impression from the diaphragmatic surface and lies between the diaphragm and the upper part of the lateral border of the left kidney. It is more blunt and rounded than the superior border and corresponds in position to the eleventh rib's lower margin. The posterior extremity, or superior pole, usually faces the rounded vertebral column. The anterior extremity, or inferior pole, is larger and less angulated than the posterior extremity and connects the lateral ends of the superior and inferior borders. It is related to the colic impression and may lie adjacent to the splenic flexure and the phrenicocolic ligament.
PERITONEAL CONNECTIONS OF THE SPLEEN (Fig. 88.2) page 1239 page 1240
Figure 88.2 Peritoneal connections of the spleen seen on CT scan. The peritoneal ligaments contain low attenuation fat on CT. The posterior connection is the splenorenal ligament (E) the anterolateral connection is the phrenicocolic ligament (AB), and the anterior connection is the gastrosplenic ligament (A-D).
The spleen is almost entirely covered by peritoneum, which is firmly adherent to its capsule. Recesses of the greater sac separate it from the stomach and left kidney. It develops in the upper dorsal mesogastrium (p. 1254), (Figs 90.6, 90.7, 90.8) and remains connected to the posterior abdominal wall, anterolateral abdominal wall and stomach by three folds of peritoneum. The posterior connection is the splenorenal ligament, the anterolateral connection is the phrenicocolic ligament, and the anterior connection is the gastrosplenic ligament. The splenorenal ligament is formed from two layers of peritoneum. The anterior layer is continuous with the peritoneum of the posterior wall of the lesser sac over the left kidney and is continuous with peritoneum of the splenic hilum where it runs into the posterior layer of the gastrosplenic ligament. The posterior layer of the splenorenal ligament is continuous with the peritoneum over the inferior surface of the diaphragm and runs onto the splenic surface over the renal impression. The splenic vessels lie between the layers of the splenorenal ligament and the tail of the pancreas is usually present in its lower portion (Fig. 69.5). The length of the splenorenal ligament may vary. Longer ligaments tend to make the spleen more mobile and may predispose the spleen to injury due to rotational shear forces during trauma but also make the mobilization of the spleen easier during surgery. The presence of the pancreatic tail within the splenorenal ligament must be remembered as it can be injured during ligation of the splenic vessels causing pancreatitis or a pancreatic duct fistula to form. The gastrosplenic ligament also has two layers. The posterior is continuous with the peritoneum of the splenic hilum and that over the posterior aspect of the stomach. The anterior layer is formed from the peritoneum reflected off the gastric impression and reaches the greater curvature of the stomach anteriorly. The short gastric and left gastroepiploic branches of the splenic artery pass between its layers. Division of the gastrosplenic ligament during surgery may be hazardous if the ligament is short since ligation of the short gastric vessels may risk injury to the greater curvature of the stomach. The phrenicocolic ligament extends from the splenic flexure of the colon to the diaphragm at the level of the eleventh rib. It extends inferiorly and laterally and is continuous with the peritoneum of the lateral end of the transverse mesocolon at the lateral margin of the pancreatic tail, and the splenorenal ligament at the hilum of the spleen. If the peritoneal attachments of the spleen are not recognized surgery may risk injury to the splenic capsule and subsequent serious bleeding. Downward traction on the phrenicocolic ligament during handling of the descending colon, especially during mobilization of the splenic flexure, may cause rupture of the splenic capsule. This is less likely if traction on the phrenicocolic ligament is made laterally or medially. The superior border and anterior diaphragmatic surface are often adherent to the peritoneum of the greater omentum. Medial traction on the omentum during surgery may cause capsular injury which is less likely if any limited traction required is applied inferiorly. The diaphragmatic surface of the spleen is occasionally adherent to the peritoneum over the inferior surface of the diaphragm. These adhesions often occur after inflammation in the spleen but may also be present congenitally.
SPLENOMEGALY page 1240 page 1241
Any massive immune response may be accompanied by splenic enlargement. This also occurs in many other systemic inflammatory and degenerative conditions. In splenomegaly, the anterior border, anterior diaphragmatic surface and notched superior border may become clearly palpable below the left costal margin; the notches are often exaggerated and may be clearly palpable. The transverse colon and splenic flexure are displaced downward.
SPLENIC TRAUMA Because of its relatively mobile peritoneal connections, the spleen is particularly prone to rotational injury during rapid deceleration or compressions. This may lead to tearing injuries to the splenic vessels at the hilum or burst injuries of the splenic pulp. Fractures of the overlying lower left ribs may cause sharp penetrating injuries to the splenic capsule and pulp. The spleen may also be injured during surgical procedures by tearing of its capsule through peritoneal adhesions and connections. Minor capsular tears may be treated by application of various haemostatic substances to the exposed pulp. Direct sutured repair of more extensive tears of the spleen is rarely successful because of the fragile nature of the splenic pulp. Moderately severe injuries may be treated by compression of the splenic tissue until haemostasis occurs but extensive burst injuries or major injuries to the hilar vessels usually require splenectomy.
SPLENECTOMY Partial splenectomy is followed by rapid regeneration of lost tissue and there is no significant loss in any of the functions of the spleen. Total splenectomy has few haematological consequences since the functions of the spleen are largely assumed by the liver. There is, however, a loss in immune function, particularly in the antibody response to systemic infections with encapsulated bacteria. This is referred to as 'overwhelming post-splenectomy sepsis syndrome'. It is particularly a problem when the spleen is lost in early childhood, but is still a significant risk even if the spleen is lost in late adult life. Splenectomy in adults is usually followed by an increased white blood cell count with increased lymphocytic, neutrophil, eosinophil and platelet counts in the peripheral blood. These effects fade and disappear within a few weeks.
© 2008 Elsevier
VASCULAR SUPPLY AND LYMPHATIC DRAINAGE SPLENIC ARTERY The spleen is supplied exclusively from the splenic artery. This is the largest branch of the coeliac axis and its course is among the most tortuous in the body (Fig. 87.5). From its origin the artery runs a little way inferiorly, then turns rapidly to the left to run initially horizontally above the level of the neck of the pancreas, before ascending as it passes more laterally. It is less steeply inclined than the body and tail of the pancreas and so comes to lie posterior to the superior border of the gland. It lies in multiple loops or even coils which appear above the superior border of the pancreas and descend to lie behind the gland. The splenic artery lies anterior to the left kidney and left suprarenal gland. It runs in the splenorenal ligament posterior to the tail of the pancreas and divides into two or three main branches before entering the hilum of the spleen. As these branches enter the hilum they divide further into four or five segmental arteries. These vessels each supply a segment of the splenic tissue. There is relatively little arterial collateral circulation between the segments which means that occlusion of a segmental vessel often leads to infarction of part of the spleen. There is, however, considerable venous collateral circulation between the segments, which makes segmental resection of the spleen practically impossible. The splenic artery gives off various branches to the pancreas in its course (p. 1233) and gives off short gastric arteries to the stomach just prior to dividing or from its terminal branches (p. 1147).
SPLENIC VEIN (Fig. 88.3)
Figure 88.3 Axial oblique CT slice of the portal vein and splenic vein.
The splenic vein is formed by five or six tributaries emerging from the hilum of the spleen. It is actually formed within the splenorenal ligament close to the tip of the tail of the pancreas. The splenic vein tributaries are thin walled and often spread over several centimetres as the hilum is long and thin (Fig. 88.1). This is important during surgical removal of the spleen since the venous tributaries must be divided close to the hilum to avoid injury to the pancreatic tail. They should be ligated in several groups to prevent the risk of avulsion of the veins from the splenic hilum and profuse bleeding, before the resection is complete. The splenic vein runs in the splenorenal ligament below the splenic artery and posterior to the tail of the pancreas. It descends to the right, and crosses the posterior abdominal wall inferior to the splenic artery and posterior to the body of the pancreas. It receives numerous short tributaries from the gland. It crosses anterior to the left kidney and renal hilum. It is separated from the left sympathetic trunk and left crus of the diaphragm by the left renal vessels, and from the abdominal aorta by the superior mesenteric artery and left renal vein. It ends behind the neck of the pancreas, where it joins the superior mesenteric vein to form the portal vein. The short gastric and left gastro-epiploic veins drain into the splenic vein through the folds of the gastrosplenic ligament near its origin (p. 1149).
LYMPHATICS Lymphatic vessels drain along the splenic trabeculae to pass out of the hilum into
the lymphatic vessels accompanying the splenic artery and vein. The vessels run posterior to pancreas close to the splenic artery and drain into nodes at the hilum, along the splenic artery and into the coeliac nodes.
© 2008 Elsevier
INNERVATION The spleen is innervated by the splenic plexus. This is formed by branches of the coeliac plexus, left coeliac ganglion, and right vagus. It accompanies the splenic artery. The fibres are mainly sympathetic and terminate in blood vessels and nonstriated muscle of the splenic capsule and trabeculae. These fibres appear to be mainly noradrenergic vasomotor, concerned with the regulation of blood flow through the spleen. Adrenergic agonists inhibit the concentration of red cells in the splenic blood (so called 'plasma skimming') indicating that sympathetic activity causes an increase in the 'fast' circulation of the spleen as opposed to slow filtration (p. 1244) (Reilly 1985).
REFERRED PAIN The majority of the sensation of pain arising from the pulp of the spleen is poorly localized. In common with other structures of foregut origin, it is referred to the central epigastrium. Distension of the splenic capsule stretches the parietal layers of the peritoneum and produces pain localized to the posterior left upper quadrant. page 1241 page 1242
Figure 88.4 The main features of splenic structure, not to scale. Shown are the capsule, trabeculae, reticular fibres and cells, the perivascular lymphoid sheaths and follicles (white pulp), and the cellular cords and venous sinusoids of the red pulp. The 'open' and 'closed' theories of splenic circulation are illustrated, although it is likely that most of the circulation is of the open form. The venous sinusoids are lined by
specialized 'stave' cells (blue) with their intercellular gaps over-emphasized for clarity.
© 2008 Elsevier
MICROSTRUCTURE (Figs 88.4, 88.5, 88.6, 88.7) The spleen is essentially concerned with phagocytosis and immune responses. In the fetus it is also an important site of haemopoiesis. Postnatally it may become haemopoietic in certain pathological conditions. Although important to the defence of the body it is not absolutely essential since many of its functions can be assumed by the liver and by other lymphoid tissues if the spleen is removed.
Figure 88.5 A section through the spleen. White pulp is present as ovoid areas of basophilic tissue, many with germinal centres (GC) due to the high density of lymphocytes surrounding splenic arterioles (periarteriolar lymphoid sheaths, PALS). Arterioles derive from trabecular arteries (TA). Paler areas within the white pulp are germinal centres. Red pulp (RP) lies between white pulp tissue and consists of splenic sinusoids and intervening cellular cords. Part of the capsule is seen top right. (By permission from Dr JB Kerr, Monash University, from Kerr JB 1999 Atlas of Functional Histology. London: Mosby.)
page 1242 page 1243
Figure 88.6 Medium power view of the junctional zone between the densely packed lymphocytes of white pulp (top), surrounding a small eccentrically placed penicillar arteriole, and the open sinusoids and cellular cords of red pulp, below. A section of fibromuscular trabecular tissue (pink) is seen at the bottom left of the field. (Photograph by Sarah-Jane Smith.)
Figure 88.7 High power micrograph of spleen, trichrome-stained, showing splenic sinusoids (open spaces) in red pulp and cellular cords in between. One sinusoid is sectioned tangentially through its wall to show, centre field, the fine parallel strap-like endothelial stave cells. The nuclei of stave cells lining transversely sectioned sinusoids bulge into the lumen, which also contains migratory lymphocytes and erythrocytes re-entering the circulation. Numerous orange erythrocytes are also seen outside the circulation, within the splenic cords, which are populated by macrophages. (Photograph by Sarah-Jane Smith.)
Microscopically, the parenchymal tissue of the spleen consists of two major components, known as white pulp and red pulp, from their appearance when a fresh spleen is transected. The white pulp is composed of lymphoid tissue (p. 74) in which B and T lymphocytes mature and proliferate under antigenic stimulation. The red pulp is a unique filtration device with which the spleen clears particulate material from the blood as it perfuses the spleen. Red pulp is composed of a complex system of interconnected spaces populated by large numbers of phagocytic macrophages (p. 80). These cells remove effete red blood cells, microorganisms, cellular debris and other particulate matter from the circulation.
FIBROUS FRAMEWORK OF THE SPLEEN The serosa of the peritoneum covers the entire organ except at its hilum and along the lines of reflexion of the splenorenal and gastrosplenic ligaments. Deep to this layer is the connective tissue capsule, a continuous layer c.1.5 mm thick, containing abundant collagen (type I) but also some elastin fibres. The capsule has an outer and an inner lamina in which the directions of collagen fibres differ, increasing its strength. Numerous trabeculae extend from the capsule into the substance of the spleen, branching within it to form a supportive framework. The
largest trabeculae enter at the hilum and provide a conduit for the splenic vessels and nerves, dividing into branches in the splenic pulp (Figs 88.4, 88.8). Within the spleen, branching trabeculae are continuous with a delicate network of fine collagen (type III, reticular) fibres pervading both the white and red pulp, which is laid down by numerous fibroblasts present in its meshes.
WHITE PULP (Fig. 88.6) In the spleen parenchyma, branches of the splenic artery radiate out from the hilum within trabeculae, ramifying and narrowing to arterioles. In their terminal few millimetres, their connective tissue adventitia is replaced by a sheath of Tlymphocytes, the periarteriolar lymphatic sheath (PALS). This sheath is enlarged in places by lymphoid follicles (p. 74), which are aggregations of B lymphocytes visible to the naked eye on the freshly cut spleen surface as white semi-opaque dots, 0.25-1 mm in diameter, which contrast with the surrounding deep reddishpurple of the red pulp. Follicles are usually situated near the terminal branches of arterioles and typically protrude to one side of the vessel, which therefore appears eccentrically placed within the follicle. Arterioles branch laterally within follicles to form a series of parallel terminal arterioles - called penicilli, or penicillar arterioles, from their resemblance to the penicillium mould. Like the periarteriolar sheaths, follicles are centres of lymphocyte proliferation as well as aggregation, and when antigenically stimulated, the white pulp increases in size as lymphocytes proliferate. The primary follicles become intensely active in B-cell proliferation, and they develop germinal centres, as are found in lymph nodes (p. 75). The presentation of antigen by follicular dendritic cells (p. 81) is involved in this process. The germinal centres regress when the infection subsides. Follicles generally atrophy with increasing age and may be absent in the very elderly.
Figure 88.8 Transverse section through the spleen, showing the trabecular tissue and the splenic vein and its tributaries (from the first edition of Gray's Anatomy, 1858). (From the first edition of Gray's Anatomy, 1858.)
RED PULP (Figs 88.4, 88.7) The red pulp constitutes the majority (75%) of the total splenic volume. Within it lie large numbers of venous sinusoids which drain into tributaries of the major splenic veins. The sinusoids are separated from each other by a fibrocellular network, the reticulum, formed by numerous reticular fibroblasts, and small bundles of delicate collagen type III fibres, in the meshes of which lie splenic macrophages. Seen in two-dimensional sections, these intersinusoidal regions appear as strips of tissue, the splenic cords, which alternate with splenic sinuses (Figs 88.4, 88.7). In reality they form a three-dimensional continuum around the venous spaces. Venous sinusoids
Venous sinusoids are elongated ovoid vessels c.50 µm in diameter, lined by a characteristic, 'incomplete' endothelium unique to the spleen. The endothelial cells are long and narrow, aligned with the long axis of the sinusoid - for this reason they are often called stave cells, reminiscent of planks in a barrel (Figs 88.4, 88.7). They are attached at intervals along their length to their neighbours by short stretches of intercellular junctions which alternate with intercellular slits through which blood can pass. A perforated, discontinuous basal lamina is present on the aspect of the sinus facing away from the lumen. The presence of slits between the endothelial cells allows blood cells to squeeze into the lumen of the sinusoid from the surrounding splenic cords. The sinusoids are supported externally by circumferential and longitudinal reticular fibres which are connected to the fibrous reticulum around them. Reticular tissue of the splenic cords
A population of large, stellate fibroblasts, the reticular cells, lie around the sinusoids, amongst the network of collagen fibres. The flattened extensions of these cells help to divide the reticular space into a series of defined compartments containing macrophages. Reticular cells synthesize the matrix components of the reticulum, including collagen and proteoglycans. Blood is released into the reticular space from the ends of capillaries which originate from penicillar arterioles. As it percolates through the spaces within cords, macrophages are able to remove particulate material, including ageing and damaged erythrocytes, from the blood. Under conditions where there are many damaged erythrocytes in the circulation to be removed by splenic macrophages, the reticular cells proliferate and increase the size of the red pulp considerably, thus causing enlargement of the whole spleen, and in extreme cases, splenomegaly.
MARGINAL ZONE page 1243 page 1244
The marginal zone lies at the interface between the white and red pulp. It is a
region of great importance to the function of the spleen. Here the lymphocytes are more loosely arranged than in the white pulp, and are held in a dense network of reticular fibres and cells. The arterioles leaving the white pulp are surrounded by a small aggregation of macrophages, the periarteriolar macrophage sheath. The marginal zone is a region where blood is delivered into the red pulp, and also where many lymphocytes leave the circulation to migrate into their respective Tand B-lymphocyte areas of the white pulp.
SPLENIC MICROCIRCULATION (Fig. 88.4) The segmental splenic arteries enter the hilum and ramify in the trabeculae throughout the organ. The splenic vein forms in the ligament from an equal number of tributaries emerging from the hilum. Small arteries tapering to arterioles pass through the white pulp then turn abruptly to form penicillar branches which, after a course of c.0.5 mm, pass out of the white pulp into the marginal zone and red pulp. The passage of blood through the vascular compartments between the arterioles and splenic veins is referred to collectively as the intermediate circulation of the spleen. Ultimately, blood is passed to the venous sinusoids from which it enters venules leading to small veins - running within trabeculae - and thence into larger veins draining the spleen at its hilum. Open and closed splenic circulations
Views on the intermediate circulation of the spleen differ on whether blood passes from the arterioles (or their terminal capillaries) directly into the venous sinuses (a closed circulation), or is instead discharged into a network of spaces in the splenic cords before entering the sinuses through the minute slits in their walls (an open circulation). In humans, evidence favours the presence of an anatomically and physiologically open circulation, in which blood percolates slowly through the reticular tissue of the splenic cords and filters through slits in the sinus walls before joining the majority of the blood flow. There is thought to be an additional closed vascular route but this is likely to provide only a minor contribution to splenic circulation; however, this has not been determined conclusively. REFERENCE Reilly FD 1985 Innervation and vascular pharmacodynamics of the mammalian spleen. Experientia 41: 187-92. Medline Similar articles Full article
© 2008 Elsevier
89 PANCREAS, SPLEEN AND SUPRARENAL GLAND Suprarenal (adrenal) gland (Figs 89.1, 89.2) The suprarenal (adrenal) glands lie immediately superior and slightly anterior to the upper pole of either kidney. Golden yellow in colour, each gland possesses two functionally and structurally distinct areas: an outer cortex and an inner medulla. The glands are surrounded by connective tissue containing perinephric fat and they are enclosed within the renal fascia. They are separated from the kidneys by a small amount of fibrous tissue. In the adult the glands measure c.50 mm vertically, 30 mm transversely and 10 mm anteroposteriorly. They each weigh c.5 g (the medulla being about one-tenth of the total weight). The dimensions of the suprarenal glands in vivo have been defined by Vincent and colleagues (1994) using computed tomography (CT). The mean dimensions of the body of the suprarenal gland are 0.61 cm (right) and 0.79 cm (left). The mean dimensions of suprarenal limbs are 0.28 cm (right) and 0.33 cm (left). No individual suprarenal limb should measure more than 6.5 mm across.
Figure 89.1 Suprarenal glands: anterior (A) and posterior (B) aspects.
The glands are macroscopically slightly different in external appearance (Fig. 89.1). The right gland is pyramidal in shape and has two well-developed lower projections (limbs) giving a cross-sectional appearance similar to a broad-headed arrow. The left gland has a more semilunar form and is flattened in the anteroposterior plane. The left gland is marginally larger than the right. The bulk of the right suprarenal sits on the apex of the right kidney and usually lies slightly higher than the left gland, which sits on the anteromedial aspect of the upper pole of the left kidney. At birth the glands are comparatively larger and are approximately one-third the size of the ipsilateral kidney. The cortex of each gland reduces in size immediately after birth and the medulla grows comparatively little. By the end of the second month the weight of the suprarenal has reduced by 50%. The glands begin to grow by the end of the second year and regain their weight at birth by puberty. There is little further weight increase in adult life. Small accessory suprarenal glands composed mainly of cortical tissue may occur in the areolar tissue near the main suprarenal glands. Accessory glands, cortical bodies, may also occur in the spermatic cord, epididymis and broad ligament of the uterus.
Figure 89.2 Vertical section through a whole adult human suprarenal gland.
© 2008 Elsevier
RIGHT SUPRARENAL GLAND (Fig. 89.3)
Figure 89.3 A, Axial and B, coronal multislice CT scans of the right and left suprarenal glands.
The right suprarenal gland lies posterior to the inferior vena cava from which it is only separated by a thin layer of fascia and connective tissue. It is posterior to the right lobe of the liver and lies anterior to the right crus of the diaphragm and
superior pole of the right kidney (Fig. 89.3). Its inferior surface is referred to as the base and adjoins the anterosuperior aspect of the superior pole of the right kidney. It often overlaps the apex of the upper pole of the right kidney as the two lower projections (limbs) straddle the renal tissue. The anterior surface faces slightly laterally and possesses two distinct facets. The medial facet is somewhat narrow, runs vertically and lies posterior to the inferior vena cava. The lateral facet is triangular and lies in contact with the bare area of the liver. The lowest part of the anterior surface may be covered by peritoneum, reflected onto it from the inferior layer of the coronary ligament. At this point it may lie posterior to the lateral border of the second part of the duodenum. Below the apex, near the anterior border of the gland, the hilum lies in a short sulcus from which the right suprarenal vein emerges to join the inferior vena cava. This vein is particularly short and makes surgical resection of the gland potentially hazardous, because ligation may be difficult. The vein may be avulsed from the inferior vena cava during surgery or occasionally by high-energy deceleration injuries. The posterior surface is divided into upper and lower areas by a curved transverse ridge. The large upper area is slightly convex and rests on the diaphragm. The small lower area is concave and lies in contact with the superior aspect of the upper pole of the right kidney. The medial border of the gland is thin and lies lateral to the right coeliac ganglion and the right inferior phrenic artery as the artery runs over the right crus of the diaphragm.
© 2008 Elsevier
LEFT SUPRARENAL GLAND The left suprarenal gland lies closely applied to the left crus of the diaphragm and is separated from it only by a thin layer of fascia and connective tissue (Fig. 89.3). The medial aspect is convex whilst the lateral aspect is concave since it is shaped by the medial side of the superior pole of the left kidney. The superior border is sharply defined while the inferior surface is more rounded. The anterior surface has a large superior area covered by peritoneum on the posterior wall of the lesser sac, which separates it from the cardia of the stomach and sometimes from the posterior aspect of the spleen. The smaller inferior area is not covered by peritoneum and lies in contact with the pancreas and splenic artery. The hilum faces inferiorly from the medial aspect and is near the lower part of the anterior surface. The left suprarenal vein emerges from the hilum and runs inferomedially to join the left renal vein. The posterior surface is divided by a ridge into a lateral area adjoining the kidney and a smaller medial area which lies in contact with the left crus of the diaphragm. The convex medial border lies lateral to the left coeliac ganglion and the left inferior phrenic and left gastric arteries, which ascend on the left crus of the diaphragm.
© 2008 Elsevier
SUPRARENAL GLAND EXCISION Removal of one or both suprarenal glands may be performed either through 'open surgery' or via a laparoscopic approach. There are essentially three 'open surgical' methods for removing the suprarenal glands. The posterior and lateral approaches involve removal of either the eleventh or twelfth ribs. The surgeon remains in the retroperitoneal plane and accesses the gland via the perirenal fat. This approach can be performed laparoscopically, but is difficult: landmarks are not easily identified because of the surrounding fat and the 'workspace' is frequently small. Not surprisingly, few surgeons employ this approach. During laparoscopic surgery the anterior, transperitoneal approach is most commonly used. The patient is placed in the lateral decubitus position. On the left side, the splenic flexure is mobilized inferiorly revealing the kidney. The lateral phrenicocolic ligament is then fully divided allowing the spleen to be mobilized medially and exposing the splenorenal ligament. Division of the splenorenal ligament with the patient in the lateral decubitus position allows the spleen to 'drop' medially exposing the plane between the splenic hilum and medial border of the left kidney. Here the pancreatic tail is identified as it runs to the hilum of the spleen. The renal fascia is divided to expose the suprarenal gland lying over the superomedial aspect of the upper pole of the kidney. On the right side, the right triangular ligament of the liver is divided. This allows the liver to be retracted caudally. In most individuals the inferior vena cava is easily identified behind the peritoneum lateral to the second part of the duodenum. The peritoneal reflection on the inferior border of the liver is divided from the inferior vena cava to the lateral border of the liver. This exposes the suprarenal gland in the angle between the inferior vena cava and liver. The peritoneal reflection along the lateral aspect of the inferior vena cava is divided and the gland is then mobilized laterally exposing the suprarenal vein, which often emerges from the posterior aspect of the inferior vena cava. Once the suprarenal vein is divided the small middle suprarenal arteries running from the aorta behind the cava to the gland are easily identified. The gland is then usually mobilized very easily in a lateral direction. However, the gland can lie either behind the inferior vena cava or under the liver, in which case dissection at the junction of the liver and inferior vena cava often becomes very difficult. In obese (or Cushingoid) patients, identification of a relatively normal size suprarenal gland can be extremely difficult and time consuming. The suprarenal vein emerges from the lower medial border of the gland and it is often a very substantial structure. In contrast, the supplying arteries tend to be rather smaller in size and individual arteries are often difficult to identify during surgery. The anterior open approach follows the procedure as described above through either a subcostal or a midline incision.
© 2008 Elsevier
VASCULAR SUPPLY AND LYMPHATIC DRAINAGE ARTERIES The suprarenal glands are very vascular. Each gland is supplied by superior, middle and inferior suprarenal arteries, whose main branches may be duplicated or even multiple. They ramify over the capsule before entering the gland to form a subcapsular plexus, from which fenestrated sinusoids pass around clustered glomerulosal cells and between columns in the zona fasciculata to a deep plexus in the zona reticularis. From this plexus venules pass between medullary chromaffin cells to medullary veins, which they enter between prominent bundles of smooth muscle fibres. Some relatively large arteries bypass this indirect route and pass directly to the medulla (see Fig. 89.5). Superior suprarenal arteries
The superior suprarenal artery arises from the inferior phrenic artery, which is a branch of the abdominal aorta (p. 1119). It is often small and may be absent. Middle suprarenal arteries (See also p. 1247)
The middle suprarenal artery arises from the lateral aspect of the abdominal aorta, at the level of the superior mesenteric artery. It ascends slightly and runs over the crura of the diaphragm to the suprarenal glands, where it anastomoses with the suprarenal branches of the inferior phrenic and renal arteries. The right middle suprarenal artery passes behind the inferior vena cava and near the right coeliac ganglion. It is frequently multiple. The left middle suprarenal artery passes close to the left coeliac ganglion, splenic artery and the superior border of the pancreas. Inferior suprarenal arteries
The inferior suprarenal arteries arise from the renal arteries (p. 1118), usually from the main renal artery but occasionally from its upper pole branches.
VEINS Medullary veins emerge from the hilum to form a suprarenal vein, which is usually single. The right vein is very short, passing directly and horizontally into the posterior aspect of the inferior vena cava. An accessory vein is occasionally present and runs from the hilum superomedially to join the inferior vena cava above the right suprarenal vein. The left suprarenal vein descends medially, anterior and lateral to the left coeliac ganglion. It passes posterior to the pancreatic body and drains into the left renal vein. Since the venous drainage from each gland is usually via a single vein, damage to a suprarenal vein is more likely to cause infarction of that gland than damage to one of the suprarenal arteries.
LYMPHATIC DRAINAGE Small lymphatic channels from both cortex and medulla drain to the hilum where larger calibre lymphatics emerge to drain directly into the lateral groups of paraaortic nodes (p. 1123).
© 2008 Elsevier
INNERVATION SUPRARENAL PLEXUS The suprarenal gland, relative to its size, has a larger autonomic supply than any other organ. The suprarenal plexus on each side lies between the medial aspect of the glands and the coeliac and aorticorenal ganglia. It contains mostly preganglionic sympathetic fibres which originate in the lower thoracic spinal segments and which reach the plexus via branches from the coeliac ganglion and plexus, and via the greater splanchnic nerve. These fibres synapse, often in deep invaginations, with large medullary chromaffin cells, which may thus be considered as homologous with postganglionic sympathetic neurones. A preponderance of non-myelinated axons has been described in the human suprarenal plexus. Both cortex and medulla also contain acetylcholinesterase (AChE)-positive fibres which presumably reach the gland from the coeliac plexus: some synapse with ganglion cells in the zonae fasciculata and reticularis.
© 2008 Elsevier
MICROSTRUCTURE In section, the suprarenal gland has an outer cortex, which is yellowish in colour and forms the main mass, and a thin medulla, forming about one-tenth of the gland, which is dark red or greyish, depending on its content of blood (Fig. 89.2). The medulla is completely enclosed by cortex, except at the hilum. The gland has a thick collagenous capsule, which extends deep trabeculae into the cortex. The capsule contains a rich arterial plexus which supplies branches to the gland.
SUPRARENAL CORTEX (Figs 89.4, 89.5) The suprarenal cortex consists of the zona glomerulosa, zona fasciculata and zona reticularis (Figs 89.4, 89.5). The outer subcapsular zona glomerulosa consists of a narrow region of small polyhedral cells in rounded clusters. The cells have deeply staining nuclei and a basophilic cytoplasm containing a few lipid droplets. Ultrastructurally, the cytoplasm displays abundant smooth endoplasmic reticulum, which is typical of cells which synthesize steroids. Deep to this, the broader zona fasciculata consists of large polyhedral basophilic cells arranged in straight columns, two cells wide, with parallel fenestrated venous sinusoids between them. The cells contain many lipid droplets and large amounts of smooth endoplasmic reticulum. The innermost part of the cortex, the zona reticularis, consists of branching interconnected columns of rounded cells whose cytoplasm also contains smooth endoplasmic reticulum, many lysosomes and aggregates of brown lipofuscin pigment which accumulate with age.
page 1247 page 1248
Figure 89.4 Cortex of the suprarenal gland. Beneath the capsule (top, absent from this specimen) is the thin zona glomerulosa (ZG), then the zona fasciculata (ZF) which occupies much of the cortex, and innermost, the zona reticularis (ZR). Sinusoids (clear spaces) and a delicate connective tissue (green) surround groups of endocrine cells, defining the organization of cells in the three functional zones. (Photograph by Sarah-Jane Smith.)
page 1248 page 1249
Figure 89.5 The gross sectional appearance, microstructure, vasculature and ultrastructure of the suprarenal gland. Brief functional summaries are appended.
Cortical cells produce several hormones and the cells of the zonae fasciculata and reticularis are also rich in ascorbic acid . Cells in the zona glomerulosa produce mineralocorticoids, e.g. aldosterone, which regulates electrolyte and
water balance; cells in the zona fasciculata produce hormones maintaining carbohydrate balance (glucocorticoids) e.g. cortisol (hydrocortisone ); cells in the zona reticularis produce sex hormones (progesterone , oestrogens and androgens). The cortex is essential to life; complete removal is lethal without replacement therapy. It exerts considerable control over lymphocytes and lymphoid tissue: increase in secretion of corticosteroids can result in a marked reduction in lymphocyte numbers. The deeper part of the zona fasciculata widens in pregnancy and in women of childbearing age in summer. Cortical atrophy in elderly males is greatest in the same region.
SUPRARENAL MEDULLA (Fig. 89.6) The suprarenal medulla is composed of groups and columns of chromaffin cells (phaeochromocytes) separated by wide venous sinusoids and supported by a network of reticular fibres. Chromaffin cells, so-called from their colour reaction to dichromate fixatives, form part of the neuroendocrine system (p. 180) and are functionally equivalent to postganglionic sympathetic neurones. They are neural crest derivatives (p. 244) and synthesize, store (as granules), and release the catecholamines noradrenaline (norepinephrine) and adrenaline (epinephrine ) into the venous sinusoids. Release is under preganglionic sympathetic control; single or small groups of sympathetic neurons are found in the medulla.
Figure 89.6 Section through the medulla of the suprarenal gland (trichrome-stained), showing large deeply staining chromaffin cells and a nerve fibre bundle (stained green, below). (Photograph by Sarah-Jane Smith.)
The majority of chromaffin cells synthesize adrenaline and store it in small granules with a dense core. Less numerous noradrenaline-secreting cells have larger granules with a dense eccentric core. Some cells synthesize both hormones. Chromogranin proteins package catecholamines within the granules, which also contain enkephalins, opiate-like proteins that may have endogenous analgesic effects in some circumstances. All of the cells are large, with large nuclei and basophilic, faintly granular cytoplasm. They form single rows along the venous sinusoids. Sympathetic axon terminals synapse with the chromaffin cells on their surfaces which face away from the sinusoids (Fig. 89.5). The sinusoids, which are lined by fenestrated endothelium, drain to the central medullary vein and hilar suprarenal vein. Normally, little adrenaline or noradrenaline is released but in response to fear, anger or stress, secretion is increased. Unlike the cortex, the suprarenal medulla is not essential to life. REFERENCE Vincent JM, Morrison ID, Armstrong P, Reznek RH 1994 The size of normal adrenal glands on computed tomography. Clin Radiol 49: 453-55. Medline Similar articles Full article page 1249 page 1250
© 2008 Elsevier
90 DEVELOPMENT OF THE PERITONEAL CAVITY, GASTROINTESTINAL TRACT AND ITS ADNEXAE Development of the peritoneal cavity, gastrointestinal tract and its adnexae POSTPHARYNGEAL FOREGUT The postdiaphragmatic gut is subdivided into three embryological portions: foremid- and hindgut, but there are no corresponding fundamental morphological and cytological distinctions between the three parts (Fig. 90.1). Thus the foregut produces a portion of the duodenum as does the midgut, and the midgut similarly produces large intestine, as does the hindgut. The differences between portions of the gut develop as a result of interactions between the three embryonic tissue layers which give rise to the gut, namely the endodermal inner epithelium, the thick layer of splanchnopleuric mesenchyme, and the outer layer of proliferating splanchnopleuric coelomic epithelium. The epithelial layer of the mucosa and connected ducts and glands are derived from the endodermal epithelium. The lamina propria and muscularis mucosa, the connective tissue of the submucosa, the muscularis externa and the external connective tissue are all derived from the splanchnopleuric mesenchyme. The outer peritoneal epithelium is derived from the splanchnopleuric coelomic epithelium.
Figure 90.1 Major epithelial populations within the early embryo. The early gut tube is close to the notochord and neural tube dorsally. The splanchnopleuric layer of the intraembryonic coelomic epithelium is in contact with the foregut ventrally and laterally, with the midgut laterally and with the hindgut ventrally and laterally.
Throughout the gut, blood vessels, lymphatics and lymph nodes develop from local populations of angiogenic mesenchyme. The nerves, which are distributed within the enteric and autonomic systems, are derived from the neural crest (p. 254). There is a craniocaudal developmental gradient along the gut in that the stomach and small intestine develop in advance of the colon. Fig. 90.2A,B shows the gut in a stage 12 embryo in relation to the other developing viscera, especially the heart and liver. Fig. 90.3 shows the overall development of the gut from stages 13-17. These diagrams should be compared. Fig. 90.1 shows the fundamental relationship of the intraembryonic coelom to the developing gut.
All regions of the gut develop from epithelial/mesenchymal interactions which are dependent on the sequential expression of a range of basic and specific genes; on the regulation of the developmental clock, seen in all areas of development; on endogenous regulatory mechanisms and local environmental influences (Lebenthal 1989). Although all these factors pertain to the whole range of developing tissues, local differences in any one of these factors along the length of the developing gut promotes the differentiation of, for example, the gastric mucosa and hepatocytes; the rotation of the midgut; and the final disposition of the sessile portions of the fully formed gastrointestinal tract. The gut is functional prior to birth and able to interact with the extrauterine environment in preterm infants.
OESOPHAGUS The oesophagus can be distinguished from the stomach at stage 13 (embryo 5 mm). It elongates during successive stages and its absolute length increases more rapidly than the embryo as a whole. Cranially it is invested by splanchnopleuric mesenchyme posterior to the developing trachea, and more caudally between the developing lungs and pericardio-peritoneal canals posterior to the pericardium (for details of tracheo-oesophageal fistulae see p. 1091). Caudal to the pericardium, the terminal, pregastric segment of the oesophagus has a short thick dorsal meso-oesophagus (from splanchnopleuric mesenchyme), while ventrally it is enclosed in the cranial stratum of the septum transversum mesenchyme. Each of the above are continuous caudally with their respective primitive dorsal and ventral mesogastria (p. 1254). Thus the oesophagus has only limited areas related to a primary coelomic epithelium. However, note the subsequent development of the para-oesophageal right and left pneumatoenteric recesses (see Fig. 90.7), the relation of the ventral aspect of the middle third of the oesophagus to the oblique sinus of the pericardium, and the relation of its lateral walls in the lower thorax to the mediastinal pleura. All the foregoing are secondary extensions from the primary coelom. The oesophageal mucosa consists of two layers of cells by stage 15 (week 5), but the proliferation of the mucosa does not occlude the lumen at any time. The mucosa becomes ciliated at 10 weeks, and stratified squamous epithelium is present at the end of the 5th month: occasionally patches of ciliated epithelium may be present at birth. Circular muscle can be seen at stage 15 but longitudinal muscle has not been identified until stage 21. Neuroblasts can be demonstrated in the early stages; the myenteric plexuses have cholinesterase activity by 9.5 weeks and ganglion cells are differentiated by 13 weeks. It has been suggested that the oesophagus is capable of peristalsis in the first trimester. Oesophageal atresia is one of the more common obstructive conditions of the alimentary tract: fetuses swallow amniotic fluid, and so the condition may be indicated by polyhydramnios. Oesophagus at birth page 1251 page 1252
Figure 90.2 A, The digestive tube of a human embryo at stage 12, with 29 paired somites, a CR length of 3.4 mm and an estimated age of 27 days. Note pharyngeal development. B, Reconstruction of a human embryo at the end of the fourth week. The alimentary canal and its outgrowths are shown in median section. The brain is shown in outline, but the spinal cord is omitted. The heart is shown in perspective, the left horn of the sinus venosus having been divided. The somites are indicated in outline. (Modified with permission from Streeter GL 1942 Developmental horizons in human embryos. Contrib Embryol Carnegie Inst Washington 30: 211-245.)
At birth the oesophagus extends 8-10 cm from the cricoid cartilage to the gastric cardiac orifice. It starts and ends 1-2 vertebrae higher than in the adult, extending from between the fourth to the sixth cervical vertebra to the level of the ninth thoracic vertebra (Fig. 11.5). Its average diameter is 5 mm and it possesses the constrictions seen in the adult. The narrowest constriction is at its junction with the pharynx, where the inferior pharyngeal constrictor muscle functions to constrict the lumen: this region may be easily traumatized with instruments or catheters. In the neonate the mucosa may contain scattered areas of ciliated columnar epithelium, but these disappear soon after birth. Peristalsis along the oesophagus and at the lower oesophageal sphincter is immature at birth and results in frequent regurgitation of food in the newborn period. The pressure at the lower oesophageal sphincter approaches that of the adult at 3-6 weeks of age.
STOMACH At the end of the fourth and beginning of the fifth week the stomach can be recognized as a fusiform dilation cranial to the wide opening of the midgut into the yolk sac (Figs 90.2, 90.3). By the fifth week this opening has narrowed into a tubular vitelline intestinal duct, which soon loses its connection with the digestive tube. At this stage the stomach is median in position and separated cranially from the pericardium by the septum transversum (see Fig. 90.5), which extends caudally on to the cranial side of the vitelline intestinal duct and ventrally to the somatopleure. Dorsally, the stomach is related to the aorta and, reflecting the presence of the pleuroperitoneal canals on each side, is connected to the body wall by a short dorsal mesentery, the dorsal mesogastrium (see Fig. 90.7). The latter is directly continuous with the dorsal mesentery (mesenteron) of almost all of the remainder of the intestine (except its caudal short segment). In human embryos of 10 mm (stage 15-16), the characteristic gastric curvatures are already recognizable. Growth is more active along the dorsal border of the viscus: its convexity markedly increases and the rudimentary fundus appears. Because of more rapid growth along the dorsal border, the pyloric end of the stomach turns ventrally and the concave lesser curvature becomes apparent (Figs 90.3, 90.6). The stomach is now displaced to the left of the median plane and apparently becomes physically rotated, which means that its original right surface becomes dorsal and its left surface becomes ventral. Accordingly the right vagus is distributed mainly to the dorsal, and the left vagus mainly to the ventral, surfaces of the stomach. The dorsal mesogastrium increases in depth and becomes folded on itself. The ventral mesogastrium becomes more coronal than sagittal. The pancreaticoenteric recess (see Fig. 90.7), hitherto usually described as a simple depression on the right side of the dorsal mesogastrium, becomes dorsal to the stomach and excavates downwards and to the left between the folded layers. It may now be termed the inferior recess of the bursa omentalis. Put simply, the stomach has undergone two 'rotations'. The first is 90° clockwise, viewed from the cranial end, the second is 90° clockwise, about an anteroposterior axis. The displacement, morphological changes and apparent 'rotation' of the stomach have been attributed variously to its own and surrounding differential growth changes, extension of the pancreaticoenteric recess with changes in its mesenchymal walls, and pressure, particularly that exerted by the rapidly growing liver. Mucosa
Mucosal and submucosal development can be seen in the 8th to 9th weeks. No villi form in the stomach, unlike other regions of the gut; instead glandular pits can be seen in the body and fundus. These develop in the pylorus and cardia by weeks 10 and 11 when parietal cells can be demonstrated. Although acid secretion has not been demonstrated in the fetal stomach before 32 weeks' gestation, preterm infants from 26 weeks' gestation onwards are able to secrete acid soon after birth. Intrinsic factor has been detected after 11 weeks. This increases from the 14th to 25th week, at which time the pylorus, which contains more parietal cells than it does in the adult, also contains a relatively larger quantity of intrinsic factor. The significance of the early production of intrinsic factor and the late production of acid by the parietal cells is not known. Chief cells can be identified after weeks 12-13, although they cannot be demonstrated to contain pepsinogen until term. Mucous neck cells actively produce mucus from week 16. Gastrin-producing cells have been demonstrated in the antrum between 19 and 20 weeks and gastrin levels have been measured in cord blood and in the plasma at term. Cord serum contains gastrin levels 2-3 times higher than those in maternal serum. Muscularis
The stomach muscularis externa develops its circular layer at 8-9 weeks, when neural plexuses are developing in the body and fundus. The longitudinal muscle develops slightly later. The pyloric musculature is thicker than the rest of the stomach: in general, the thickness of the total musculature of the stomach at term is reduced compared to the adult. page 1252 page 1253
page 1253 page 1254
Figure 90.3 The shape of the endodermal epithelium of the gut at succeeding stages. The scale is constant illustrating the enormous growth of the gut over a 13-day period. A, stage 13; B, stage 14; C, stage 15; D, stage 16 and E, stage 17. Note the separation of the respiratory diverticulum, the elongation of the foregut and expansion of the stomach, the formation of the hepatic and pancreatic diverticula, the lengthening of the midgut loop which protrudes into the umbilical cord, and the separation of the cloaca into enteric and allantoic portions. (Modified from O'Rahilly and Muller. Developmental Stages in Human Embryos 1987 Carnegie Institution of Washington. Pub 637.)
Serosa
The serosa of the stomach is derived from the splanchnopleuric coelomic epithelium. No part of this serosa undergoes absorption. The original left side of the gastric serosa faces the greater sac, the right side faces the lesser sac. Stomach at birth
The stomach exhibits fetal characteristics until just after birth when the initiation of pulmonary ventilation, the reflexes of coughing and swallowing, and crying, cause the ingestion of large amounts of air and liquid. Once postnatal swallowing has started the stomach distends to four or five times its contracted state, and shifts its position in relation to the state of expansion and contraction of the other abdominal viscera, and to the position of the body. In the neonate, the anterior surface of the stomach is generally covered by the left lobe of the liver, which extends nearly as far as the spleen (Figs 11.4, 65.5). Only a small portion of the greater curvature of the stomach is visible anteriorly. The capacity of the stomach is 30-35 ml in the full-term neonate, rising to 75 ml in the second week and 100 ml by the fourth week (adult capacity is on average 1000 ml). The mucosa and submucosa are relatively thicker than in the adult, however, the muscularis is only moderately developed and peristalsis is not coordinated. At birth gastric acid secretion is low, which means that gastric pH is high for the first 12 postnatal hours. It falls rapidly with the onset of gastric acid secretion, usually after the first feed. Acid secretion usually remains low for the first 10 days postnatally. Gastric emptying and transit times are delayed in the neonate.
DUODENUM The duodenum develops from the caudal part of the foregut and the cranial part of the midgut. A ventral mesoduodenum, which is continuous cranially with the ventral mesogastrium, is attached only to the foregut portion. Posteriorly the duodenum has a thick dorsal mesoduodenum which is continuous with the dorsal mesogastrium cranially and the dorsal mesentery of the midgut caudally. Anteriorly the extreme caudal edge of the ventral mesentery of the foregut extends onto the short initial segment of the duodenum. The liver arises as a diverticulum from the ventral surface of the duodenum at the foregut-midgut junction, i.e. where the midgut is continuous with the yolk sac wall (the cranial intestinal portal). The ventral pancreatic bud also arises from this diverticulum. The dorsal pancreatic bud evaginates posteriorly into the dorsal mesoduodenum slightly more cranially than the hepatic diverticulum. The rotation, differential growth, and cavitations related to the developing stomach and omenta cause corresponding movements in the duodenum, which forms a loop directed to the right, with its original right side now adjacent to the posterior abdominal wall. This shift is compounded by the migration of the bile duct and ventral pancreatic duct around the duodenal wall. Their origin shifts until it reaches the medial wall of the second part of the fully formed duodenum: the bile duct passes posteriorly to the duodenum and travels in the free edge of the ventral duodenum and ventral mesogastrium. Local adherence and subsequent absorption of part of the duodenal serosa and the parietal peritoneum results in almost the whole of the duodenum, other than a short initial segment, becoming retroperitoneal (sessile). Duodenal atresia is a developmental defect found in 1 in 5000 live births (Whittle 1999). It may be associated with an annular pancreas which may compress the duodenum externally (20% of duodenal atresia), or with abnormalities of the bile duct. In 40-60% of cases the atresia is complete and pancreatic tissue fills the lumen. The condition can be diagnosed on ultrasound examination, which reveals a typical double bubble appearance caused by fluid enlarging the stomach and the proximal duodenum. Polyhydramnios is invariably present and often the indication for the scan. Duodenal atresia commonly occurs with other developmental defects, e.g. cardiac and skeletal anomalies and in Down's syndrome.
DORSAL AND VENTRAL MESENTERIES OF THE FOREGUT The epithelium of the stomach and duodenum does not rotate relative to its investing mesenchyme. The rotation includes the coelomic epithelial walls of the pericardioperitoneal canals, which are on each side of the stomach and duodenum and form its serosa, and the elongating dorsal mesogastrium or the much shorter dorsal mesoduodenum. A ventral mesogastrium can be seen when the distance between the stomach and liver increases. Whereas the dorsal mesogastrium takes origin from the posterior body wall in the midline, its connection to the greater curvature of the stomach, which lengthens as the stomach grows, becomes directed to the left as the stomach undergoes its first rotation. With the second rotation a portion of the dorsal mesogastrium now faces
caudally. The ventral mesogastrium remains as a double layer of coelomic epithelium which encloses mesenchyme and forms the lesser omentum (see Figs 90.7, 90.8B). Movement of the stomach is associated with an extensive lengthening of the dorsal mesogastrium, which becomes the greater omentum, and which now, from its posterior origin, droops caudally over the small intestine, then folds back anteriorly and ascends to the greater curvature of the stomach. The greater omentum is therefore composed of a fold containing, technically, four layers of peritoneum. The dorsal mesoduodenum, or suspensory ligament of the duodenum, is a much thicker structure, and it fixes the position of the duodenum when the rest of the midgut and its dorsal mesentery elongate and pass into the umbilical cord. For a more detailed account of this process see page 1256.
SPECIAL GLANDS OF THE POSTPHARYNGEAL FOREGUT Pancreas
page 1254 page 1255
Figure 90.4 Development of the pancreas in a human embryo. A, An early stage, 7.5 mm embryo; lateral view. B, A later stage, 14.5 mm embryo; ventral view. (Modified with permission from Streeter GL 1942 Developmental horizons in human embryos. Contrib Embryol Carnegie Inst Washington 30: 211-245.)
The pancreas develops from two evaginations of the foregut which fuse to form a single organ. A dorsal pancreatic bud can be seen in stage 13 embryos as a thickening of the endodermal tube which proliferates into the dorsal mesogastrium (Figs 90.3, 90.4). A ventral pancreatic bud evaginates in close proximity to the liver primordium but cannot be clearly identified until stage 14 when it appears as an evagination of the bile duct itself. At stage 16 (5 weeks) differential growth of the wall of the duodenum results in movement of the ventral pancreatic bud and the bile duct to the right side and ultimately to a dorsal position. It is not clear whether there is a corresponding shift of mesenchyme during this rotation.
However, the ventral pancreatic bud and the bile duct rotate from a position within the ventral mesogastrium (ventral mesoduodenum) to one in the dorsal mesogastrium (dorsal mesoduodenum) which is destined to become fixed onto the posterior abdominal wall. By stage 17 the ventral and dorsal pancreatic buds have fused, although the origin of the ventral bud from the bile duct is still obvious. Three-dimensional reconstruction of the ventral and dorsal pancreatic buds have confirmed that the dorsal pancreatic bud forms the anterior part of the head, the body and the tail of the pancreas and the ventral pancreatic bud forms the posterior part of the head and the posterior part of the uncinate process. The ventral pancreatic bud does not form all of the uncinate process (Collins 2002). The developing pancreatic ducts usually fuse in such a way that most of the dorsal duct drains into the proximal part of the ventral duct (Figs 90.3, 90.4). The proximal portion of the dorsal duct usually persists as an accessory duct. The fusion of the ducts takes place late in development or in the postnatal period: 85% of infants have patent accessory ducts as compared to 40% of adults. Fusion may not occur in 10% of individuals, in which case separate drainage into the duodenum is maintained, so-called pancreatic divisum. Failure of the ventral pancreatic diverticulum to migrate will result in an annular pancreas which may constrict the duodenum locally. UPDATE Date Added: 06 July 2005 Abstract: Case report of annular pancreas in male adult. Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=15779573&query_hl=6 Case report of annular pancreas in male adult. UPDATE Date Added: 28 June 2005 Abstract: Annular pancreas: magnetic resonance cholangiopancreatography findings. Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=15232386&query_hl=9 Annular pancreas: magnetic resonance cholangiopancreatography findings. UPDATE Date Added: 28 June 2005 Abstract: Annular pancreas: magnetic resonance cholangiopancreatography findings. Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=15232386&query_hl=9 Annular pancreas: magnetic resonance cholangiopancreatography findings. The ventral pancreas does not always extend anterior to the superior mesenteric vein but remains related to its right lateral surface. Initially the body of the pancreas extends into the dorsal mesoduodenum and then cranially into the dorsal mesogastrium. As the stomach rotates, this portion of the dorsal mesogastrium is directed to the left forming the posterior wall of the lesser sac. The posterior layer of this portion of dorsal mesogastrium fuses with the parietal layer of the coelom wall (peritoneum) and the pancreas becomes mainly retroperitoneal. The region of fusion of the dorsal mesogastrium does not extend so far left as to include the tail of the pancreas which passes into the splenorenal (lienorenal) ligament. The anterior border of the pancreas later provides the main line of attachment for the posterior leaves of the greater omentum. UPDATE Date Added: 30 May 2006 Abstract: Fetal development of pancreaticoduodenal arteries Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=15478102&query_hl=10&itool=pubmed_docsum The pancreaticoduodenal arteries in human foetal development. Krakowiak-Sarnowska E, Flisinski P, Szpinda M et al: Folia Morphol (Warsz) 63:281-284, 2004. Cellular development of the pancreas
The early specification of pancreatic endoderm involves the proximity of the notochord to the dorsal endoderm, which locally represses the expression of Shh transcription factor. Endoderm caudal to the pancreatic region does not respond to notochordal signals. The ventral pancreatic endoderm does not seem to undergo the same induction. Pancreatic mesenchyme is derived from two regions. The mesenchyme which surrounds the dorsal pancreatic bud proliferates from the splanchnopleuric coelomic epithelium of the medial walls of the
pericardioperitoneal canals, whereas the ventral pancreatic bud is invested by septum transversum mesenchyme and by mesenchyme derived from the lower ventral walls of the pericardioperitoneal canals. The primitive endodermal ductal epithelium provides the stem cell population for all the secretory cells of the pancreas. Initially these endocrine cells are located in the duct walls or in buds developing from them, and later they accumulate in pancreatic islets. The remaining primitive duct cells will differentiate into definitive ductal cells. In the fetus they develop microvilli and cilia but lack the lateral interdigitations seen in the adult. Branches of the main duct become interlobular ductules which terminate as blind ending acini or as tubular, acinar elements. The ductal branching pattern and acinar structure of the pancreas is determined by the pancreatic mesenchyme. This mesenchyme gives rise to connective tissue between the ducts which, in the fetus, appears to be important in stimulating pancreatic proliferation and maintaining the relative proportions of acinar, ! and " cells during development. It also provides cell lines for smooth muscle within the pancreas. Angiogenic mesenchyme invades the developing gland to produce blood and lymphatic vessels. The process of islet differentiation is divided into two phases (Collins 2002). Phase I, characterized by proliferation of polyhormonal cells, occurs from weeks 9-15. Phase II, characterized by differentiation of monohormonal cells, is seen from week 16 onwards. The " cells, producing insulin and amylin, differentiate first, followed by !-cells which produce glucagon. The # cells which produce somatostatin are seen after 30 weeks. The dorsal bud gives rise mostly to ! cells, and the ventral bud to most of the pancreatic polypeptide producing cells. The " cells develop from the duct epithelium throughout development and into the neonatal period. Later, in weeks 10-15, some of the primitive ducts differentiate into acinar cells in which zymogen granules or acinar cell markers can be detected at 12-16 weeks. The pancreas in the neonate has all of the normal subdivisions of the adult. The head is proportionately large in the newborn and there is a smooth continuation between the body and the tail. The inferior border of the head of the pancreas is found at the level of the second lumbar vertebra. The body and tail pass cranially and to the left, and the tail is in contact with the spleen (Fig. 11.4). Liver
The liver is one of the most precocious embryonic organs and is the main centre for haemopoiesis in the fetus. It develops from an endodermal evagination of the foregut and from septum transversum mesenchyme which is derived from the proliferating coelomic epithelium in the protocardiac region. The development of the liver is intimately related to the development of the heart. The vitelline veins, succeeded by the umbilical veins passing to the sinus venosus are disrupted by the enlarging septum transversum to form a hepatic plexus, the forerunner of the hepatic sinusoids (p. 1046). (See Collins 2002 for a detailed account of hepatic development.) Early liver development
As the head fold and early intraembryonic coelom form, the ventral parietal wall of the pericardial cavity gives rise to populations of cells termed precardiac or cardiac mesenchyme. Hepatic endoderm is induced to proliferate by this mesenchyme, although all portions of the early heart tube, truncus arteriosus, atria, ventricle, both endocardium and myocardium, have hepatic induction potency which is tissue-specific, but not species-specific. As the heart and foregut become separated by the accumulation of the cardiac mesenchyme, the mesenchyme itself is renamed septum transversum. It is seen as a ventral mass, caudal to the heart which passes dorsally on each side of the developing gut to join the mesenchyme proliferating from the walls of the pericardioperitoneal canals. The majority of the cells within the septum transversum are destined to become hepatic mesenchyme. In the stage 11 embryo the location of the hepatic endoderm has been identified at the superior boundary of the rostral intestinal portal. By stage 12, the hepatic endodermal primordium is directed ventrally and begins to proliferate as a diverticulum. There are two parts: a caudal part, which will produce the cystic duct and gallbladder, and a cranial part which forms the liver biliary system (Figs 90.3, 90.5). The cells start to express liver-specific molecular markers and glycogen storage. Around the cranial portion of the hepatic diverticulum the basal lamina is
progressively disrupted and individual epithelial cells migrate into the surrounding septum transversum mesenchyme. The previously smooth contour of the diverticulum merges into columnar extensions of endoderm, the epithelial trabeculae, which stimulate the hepatic mesenchymal cells to form blood islands and endothelium. The advance of the endodermal epithelial cells promotes the conversion of progressively more hepatic mesenchyme into endothelium and blood cells, and only a little remains to form the scanty liver capsule and interlobular connective tissue. This invasion by the hepatic epithelium is completed in stage 13, when it approaches the caudal surface of the pericardial cavity, and is separated from it only by a thin lamina of mesenchyme which will give rise to part of the diaphragm. During this early phase of development the liver is far more highly vascularized than the rest of the gut. The hepatic capillary plexus is connected bilaterally with the right and left vitelline veins. Dorsolaterally they empty by multiple channels into enlarged hepatocardiac channels, which lead to the right and left horns of the sinus venosus; usually the channel on the right side is most developed. Both left and right channels bulge into the pericardioperitoneal canals, forming sites for the exchange of fluid from the coelom into the vascular channels. The growth of the hepatic tissue in these regions is sometimes referred to as the left and right horns of the liver. page 1255 page 1256
Figure 90.5 The hepatic endodermal primordium proliferates ventrally into the septum transversum mesenchyme. The endodermal cells forming the hepatic trabeculae will become hepatocytes; the septum transversum mesenchymal cells will become the endothelium of the liver sinusoids and early blood cells. (Modified with permission from Collins P 2002 Embryology of the liver and bile ducts. In: Howard, ER, Stringer MD, Clombani PM (eds) Surgery of the Liver, Bile Ducts and Pancreas in Children, Chapter 7, Part 3. London: Arnold.)
The liver remains proportionately large during its development and constitutes a sizeable organ dorsal to the heart at stage 14 (Fig. 90.8B,C), then more caudally placed by stage 16. By this stage hepatic ducts can be seen separating the hepatic epithelium from the extrahepatic biliary system, but even at stage 17 the ducts do not penetrate far into the liver. Maturation of the liver
At 3 months' gestation, the liver almost fills the abdominal cavity and its left lobe is nearly as large as its right. When the haematopoietic activity of the liver is assumed by the spleen and bone marrow the left lobe undergoes some degeneration and becomes smaller than the right. The liver remains relatively larger than in the adult throughout the remainder of gestation. In the neonate it constitutes 4% of the body weight, compared to 2.5-3.5% in adults. It is in contact with the greater part of the diaphragm and extends below the costal margin anteriorly, and in some cases to within 1 cm of the iliac crest posteriorly. The left lobe covers much of the anterior surface of the stomach and constitutes nearly one-third of the liver (Figs 11.4, 65.5). Although its haemopoietic functions cease before birth its enzymatic and synthetic functions are not completely mature at
birth. Development of intrahepatic biliary ducts
The development of the intrahepatic biliary ducts follows the branching pattern of the portal vein radicles (Collins 2002). The cranial hepatic diverticulum gives rise to the liver hepatocytes, the intrahepatic large bile ducts (right and left hepatic ducts, segmental ducts, area ducts and their first branches) and the small bile ducts (septal bile ducts, interlobular ducts and bile ductules). The portal and hepatic veins arise together from the vitelline veins. Early in development the accumulation of mesenchyme around these veins is similar, whereas later mesenchyme increases around the portal veins. This is a prerequisite for bile duct development. Primitive hepatocytes surround the portal vein branches and associated mesenchyme and form a sleeve of cells termed the ductal plate. Portions of the ductal plate divide to produce lines of epithelial cells which migrate close to a portal vein branch where they differentiate into bile ducts. As the bile ducts develop, angiogenic mesenchymal cells form blood vessels which connect to the hepatic artery from 10 weeks. Thus the portal triads are patterned by the portal vein radicles which initially induce bile duct formation and then artery formation. The development of the biliary system extends from the hilum to the periphery. Abnormalities of the biliary tree are associated with abnormalities of the branching pattern of the portal vein. The developing bile ducts remain patent throughout development; the solid stage of ductal development previously promulgated has been refuted. Atresia of the extrahepatic bile ducts has been noted, often in assocition with extrahepatic atresia. The cause of this condition is not clear; inflammatory process may be involved, although some cases have features of ductal plate malformation (Howard 2002). Development of extrahepatic biliary ducts
The caudal part of the hepatic endodermal diverticulum forms the extrahepatic biliary system, the common hepatic duct, gallbladder, cystic duct and common bile duct. The bile duct, which originated from the ventral wall of the foregut (now duodenum), migrates with the ventral pancreatic bud first to the right and then dorsomedially into the dorsal mesoduodenum. The right and left hepatic ducts arise from the cranial end of the common hepatic duct from 12 weeks' gestation. Atresia of the extrahepatic bile ducts in neonates occurs alone or in conjunction with a range of other anomalies, including situs inversus, malrotation, polysplenia and cardiac defects. In such cases the intrahepatic bile ducts have a mature tubular shape but also show features of ductal plate malformation. In the neonate the gallbladder has a smaller peritoneal surface than in the adult, and its fundus often does not extend to the liver margin. It is generally embedded in the liver and in some cases may be covered by bands of liver. After the second year the gallbladder assumes the relative size it has in the adult.
© 2008 Elsevier
POSTPHARYNGEAL FOREGUT The postdiaphragmatic gut is subdivided into three embryological portions: foremid- and hindgut, but there are no corresponding fundamental morphological and cytological distinctions between the three parts (Fig. 90.1). Thus the foregut produces a portion of the duodenum as does the midgut, and the midgut similarly produces large intestine, as does the hindgut. The differences between portions of the gut develop as a result of interactions between the three embryonic tissue layers which give rise to the gut, namely the endodermal inner epithelium, the thick layer of splanchnopleuric mesenchyme, and the outer layer of proliferating splanchnopleuric coelomic epithelium. The epithelial layer of the mucosa and connected ducts and glands are derived from the endodermal epithelium. The lamina propria and muscularis mucosa, the connective tissue of the submucosa, the muscularis externa and the external connective tissue are all derived from the splanchnopleuric mesenchyme. The outer peritoneal epithelium is derived from the splanchnopleuric coelomic epithelium.
Figure 90.1 Major epithelial populations within the early embryo. The early gut tube is close to the notochord and neural tube dorsally. The splanchnopleuric layer of the intraembryonic coelomic epithelium is in contact with the foregut ventrally and laterally, with the midgut laterally and with the hindgut ventrally and laterally.
Throughout the gut, blood vessels, lymphatics and lymph nodes develop from local populations of angiogenic mesenchyme. The nerves, which are distributed within the enteric and autonomic systems, are derived from the neural crest (p. 254). There is a craniocaudal developmental gradient along the gut in that the stomach and small intestine develop in advance of the colon. Fig. 90.2A,B shows the gut in a stage 12 embryo in relation to the other developing viscera, especially the heart and liver. Fig. 90.3 shows the overall development of the gut from stages 13-17. These diagrams should be compared. Fig. 90.1 shows the fundamental relationship of the intraembryonic coelom to the developing gut. All regions of the gut develop from epithelial/mesenchymal interactions which are dependent on the sequential expression of a range of basic and specific genes; on the regulation of the developmental clock, seen in all areas of development; on endogenous regulatory mechanisms and local environmental influences (Lebenthal 1989). Although all these factors pertain to the whole range of developing tissues, local differences in any one of these factors along the length of the developing gut promotes the differentiation of, for example, the gastric
mucosa and hepatocytes; the rotation of the midgut; and the final disposition of the sessile portions of the fully formed gastrointestinal tract. The gut is functional prior to birth and able to interact with the extrauterine environment in preterm infants.
OESOPHAGUS The oesophagus can be distinguished from the stomach at stage 13 (embryo 5 mm). It elongates during successive stages and its absolute length increases more rapidly than the embryo as a whole. Cranially it is invested by splanchnopleuric mesenchyme posterior to the developing trachea, and more caudally between the developing lungs and pericardio-peritoneal canals posterior to the pericardium (for details of tracheo-oesophageal fistulae see p. 1091). Caudal to the pericardium, the terminal, pregastric segment of the oesophagus has a short thick dorsal meso-oesophagus (from splanchnopleuric mesenchyme), while ventrally it is enclosed in the cranial stratum of the septum transversum mesenchyme. Each of the above are continuous caudally with their respective primitive dorsal and ventral mesogastria (p. 1254). Thus the oesophagus has only limited areas related to a primary coelomic epithelium. However, note the subsequent development of the para-oesophageal right and left pneumatoenteric recesses (see Fig. 90.7), the relation of the ventral aspect of the middle third of the oesophagus to the oblique sinus of the pericardium, and the relation of its lateral walls in the lower thorax to the mediastinal pleura. All the foregoing are secondary extensions from the primary coelom. The oesophageal mucosa consists of two layers of cells by stage 15 (week 5), but the proliferation of the mucosa does not occlude the lumen at any time. The mucosa becomes ciliated at 10 weeks, and stratified squamous epithelium is present at the end of the 5th month: occasionally patches of ciliated epithelium may be present at birth. Circular muscle can be seen at stage 15 but longitudinal muscle has not been identified until stage 21. Neuroblasts can be demonstrated in the early stages; the myenteric plexuses have cholinesterase activity by 9.5 weeks and ganglion cells are differentiated by 13 weeks. It has been suggested that the oesophagus is capable of peristalsis in the first trimester. Oesophageal atresia is one of the more common obstructive conditions of the alimentary tract: fetuses swallow amniotic fluid, and so the condition may be indicated by polyhydramnios. Oesophagus at birth page 1251 page 1252
Figure 90.2 A, The digestive tube of a human embryo at stage 12, with 29 paired somites, a CR length of 3.4 mm and an estimated age of 27 days. Note pharyngeal development. B, Reconstruction of a human embryo at the end of the fourth week. The alimentary canal and its outgrowths are shown in median section. The brain is shown in outline, but the spinal cord is omitted. The heart is shown in perspective, the left horn of the sinus venosus having been divided. The somites are indicated in outline. (Modified with permission from Streeter GL 1942 Developmental horizons in human embryos. Contrib Embryol Carnegie Inst Washington 30: 211-245.)
At birth the oesophagus extends 8-10 cm from the cricoid cartilage to the gastric cardiac orifice. It starts and ends 1-2 vertebrae higher than in the adult, extending from between the fourth to the sixth cervical vertebra to the level of the ninth thoracic vertebra (Fig. 11.5). Its average diameter is 5 mm and it possesses the constrictions seen in the adult. The narrowest constriction is at its junction with the pharynx, where the inferior pharyngeal constrictor muscle functions to constrict the lumen: this region may be easily traumatized with instruments or catheters. In the neonate the mucosa may contain scattered areas of ciliated columnar epithelium, but these disappear soon after birth. Peristalsis along the oesophagus and at the lower oesophageal sphincter is immature at birth and results in frequent regurgitation of food in the newborn period. The pressure at the lower oesophageal sphincter approaches that of the adult at 3-6 weeks of age.
STOMACH At the end of the fourth and beginning of the fifth week the stomach can be recognized as a fusiform dilation cranial to the wide opening of the midgut into the yolk sac (Figs 90.2, 90.3). By the fifth week this opening has narrowed into a tubular vitelline intestinal duct, which soon loses its connection with the digestive tube. At this stage the stomach is median in position and separated cranially from the pericardium by the septum transversum (see Fig. 90.5), which extends caudally on to the cranial side of the vitelline intestinal duct and ventrally to the somatopleure. Dorsally, the stomach is related to the aorta and, reflecting the presence of the pleuroperitoneal canals on each side, is connected to the body wall by a short dorsal mesentery, the dorsal mesogastrium (see Fig. 90.7). The latter is directly continuous with the dorsal mesentery (mesenteron) of almost all of the remainder of the intestine (except its caudal short segment). In human embryos of 10 mm (stage 15-16), the characteristic gastric curvatures are already recognizable. Growth is more active along the dorsal border of the viscus: its convexity markedly increases and the rudimentary fundus appears. Because of more rapid growth along the dorsal border, the pyloric end of the stomach turns ventrally and the concave lesser curvature becomes apparent (Figs 90.3, 90.6). The stomach is now displaced to the left of the median plane and apparently becomes physically rotated, which means that its original right surface becomes dorsal and its left surface becomes ventral. Accordingly the right vagus is distributed mainly to the dorsal, and the left vagus mainly to the ventral, surfaces of the stomach. The dorsal mesogastrium increases in depth and becomes folded on itself. The ventral mesogastrium becomes more coronal than
sagittal. The pancreaticoenteric recess (see Fig. 90.7), hitherto usually described as a simple depression on the right side of the dorsal mesogastrium, becomes dorsal to the stomach and excavates downwards and to the left between the folded layers. It may now be termed the inferior recess of the bursa omentalis. Put simply, the stomach has undergone two 'rotations'. The first is 90° clockwise, viewed from the cranial end, the second is 90° clockwise, about an anteroposterior axis. The displacement, morphological changes and apparent 'rotation' of the stomach have been attributed variously to its own and surrounding differential growth changes, extension of the pancreaticoenteric recess with changes in its mesenchymal walls, and pressure, particularly that exerted by the rapidly growing liver. Mucosa
Mucosal and submucosal development can be seen in the 8th to 9th weeks. No villi form in the stomach, unlike other regions of the gut; instead glandular pits can be seen in the body and fundus. These develop in the pylorus and cardia by weeks 10 and 11 when parietal cells can be demonstrated. Although acid secretion has not been demonstrated in the fetal stomach before 32 weeks' gestation, preterm infants from 26 weeks' gestation onwards are able to secrete acid soon after birth. Intrinsic factor has been detected after 11 weeks. This increases from the 14th to 25th week, at which time the pylorus, which contains more parietal cells than it does in the adult, also contains a relatively larger quantity of intrinsic factor. The significance of the early production of intrinsic factor and the late production of acid by the parietal cells is not known. Chief cells can be identified after weeks 12-13, although they cannot be demonstrated to contain pepsinogen until term. Mucous neck cells actively produce mucus from week 16. Gastrin-producing cells have been demonstrated in the antrum between 19 and 20 weeks and gastrin levels have been measured in cord blood and in the plasma at term. Cord serum contains gastrin levels 2-3 times higher than those in maternal serum. Muscularis
The stomach muscularis externa develops its circular layer at 8-9 weeks, when neural plexuses are developing in the body and fundus. The longitudinal muscle develops slightly later. The pyloric musculature is thicker than the rest of the stomach: in general, the thickness of the total musculature of the stomach at term is reduced compared to the adult. page 1252 page 1253
page 1253 page 1254
Figure 90.3 The shape of the endodermal epithelium of the gut at succeeding stages. The scale is constant illustrating the enormous growth of the gut over a 13-day period. A, stage 13; B, stage 14; C, stage 15; D, stage 16 and E, stage 17. Note the separation of the respiratory diverticulum, the elongation of the foregut and expansion of the stomach, the formation of the hepatic and pancreatic diverticula, the lengthening of the midgut loop which protrudes into the umbilical cord, and the separation of the cloaca into enteric and allantoic portions. (Modified from O'Rahilly and Muller. Developmental Stages in Human Embryos 1987 Carnegie Institution of Washington. Pub 637.)
Serosa
The serosa of the stomach is derived from the splanchnopleuric coelomic epithelium. No part of this serosa undergoes absorption. The original left side of the gastric serosa faces the greater sac, the right side faces the lesser sac. Stomach at birth
The stomach exhibits fetal characteristics until just after birth when the initiation of pulmonary ventilation, the reflexes of coughing and swallowing, and crying, cause the ingestion of large amounts of air and liquid. Once postnatal swallowing has started the stomach distends to four or five times its contracted state, and shifts its position in relation to the state of expansion and contraction of the other abdominal viscera, and to the position of the body. In the neonate, the anterior surface of the stomach is generally covered by the left lobe of the liver, which extends nearly as far as the spleen (Figs 11.4, 65.5). Only a small portion of the greater curvature of the stomach is visible anteriorly. The capacity of the stomach is 30-35 ml in the full-term neonate, rising to 75 ml in the second week and 100 ml by the fourth week (adult capacity is on average 1000 ml). The mucosa and submucosa are relatively thicker than in the adult, however, the muscularis is only moderately developed and peristalsis is not coordinated. At birth gastric acid secretion is low, which means that gastric pH is high for the first 12 postnatal hours. It falls rapidly with the onset of gastric acid secretion, usually after the first feed. Acid secretion usually remains low for the first 10 days postnatally. Gastric emptying and transit times are delayed in the neonate.
DUODENUM The duodenum develops from the caudal part of the foregut and the cranial part of the midgut. A ventral mesoduodenum, which is continuous cranially with the ventral mesogastrium, is attached only to the foregut portion. Posteriorly the duodenum has a thick dorsal mesoduodenum which is continuous with the dorsal mesogastrium cranially and the dorsal mesentery of the midgut caudally. Anteriorly the extreme caudal edge of the ventral mesentery of the foregut extends onto the short initial segment of the duodenum. The liver arises as a diverticulum from the ventral surface of the duodenum at the foregut-midgut junction, i.e. where the midgut is continuous with the yolk sac wall (the cranial intestinal portal). The ventral pancreatic bud also arises from this diverticulum. The dorsal pancreatic bud evaginates posteriorly into the dorsal mesoduodenum slightly more cranially than the hepatic diverticulum. The rotation, differential growth, and cavitations related to the developing stomach and omenta cause corresponding movements in the duodenum, which forms a loop directed to the right, with its original right side now adjacent to the posterior abdominal wall. This shift is compounded by the migration of the bile duct and ventral pancreatic duct around the duodenal wall. Their origin shifts until it reaches the medial wall of the second part of the fully formed duodenum: the bile duct passes posteriorly to the duodenum and travels in the free edge of the ventral duodenum and ventral mesogastrium. Local adherence and subsequent absorption of part of the duodenal serosa and the parietal peritoneum results in almost the whole of the duodenum, other than a short initial segment, becoming retroperitoneal (sessile). Duodenal atresia is a developmental defect found in 1 in 5000 live births (Whittle 1999). It may be associated with an annular pancreas which may compress the duodenum externally (20% of duodenal atresia), or with abnormalities of the bile duct. In 40-60% of cases the atresia is complete and pancreatic tissue fills the lumen. The condition can be diagnosed on ultrasound examination, which reveals a typical double bubble appearance caused by fluid enlarging the stomach and the proximal duodenum. Polyhydramnios is invariably present and often the indication for the scan. Duodenal atresia commonly occurs with other developmental defects, e.g. cardiac and skeletal anomalies and in Down's syndrome.
DORSAL AND VENTRAL MESENTERIES OF THE FOREGUT The epithelium of the stomach and duodenum does not rotate relative to its investing mesenchyme. The rotation includes the coelomic epithelial walls of the pericardioperitoneal canals, which are on each side of the stomach and duodenum and form its serosa, and the elongating dorsal mesogastrium or the much shorter dorsal mesoduodenum. A ventral mesogastrium can be seen when the distance between the stomach and liver increases. Whereas the dorsal mesogastrium takes origin from the posterior body wall in the midline, its connection to the greater curvature of the stomach, which lengthens as the stomach grows, becomes directed to the left as the stomach undergoes its first rotation. With the second rotation a portion of the dorsal mesogastrium now faces
caudally. The ventral mesogastrium remains as a double layer of coelomic epithelium which encloses mesenchyme and forms the lesser omentum (see Figs 90.7, 90.8B). Movement of the stomach is associated with an extensive lengthening of the dorsal mesogastrium, which becomes the greater omentum, and which now, from its posterior origin, droops caudally over the small intestine, then folds back anteriorly and ascends to the greater curvature of the stomach. The greater omentum is therefore composed of a fold containing, technically, four layers of peritoneum. The dorsal mesoduodenum, or suspensory ligament of the duodenum, is a much thicker structure, and it fixes the position of the duodenum when the rest of the midgut and its dorsal mesentery elongate and pass into the umbilical cord. For a more detailed account of this process see page 1256.
SPECIAL GLANDS OF THE POSTPHARYNGEAL FOREGUT Pancreas
page 1254 page 1255
Figure 90.4 Development of the pancreas in a human embryo. A, An early stage, 7.5 mm embryo; lateral view. B, A later stage, 14.5 mm embryo; ventral view. (Modified with permission from Streeter GL 1942 Developmental horizons in human embryos. Contrib Embryol Carnegie Inst Washington 30: 211-245.)
The pancreas develops from two evaginations of the foregut which fuse to form a single organ. A dorsal pancreatic bud can be seen in stage 13 embryos as a thickening of the endodermal tube which proliferates into the dorsal mesogastrium (Figs 90.3, 90.4). A ventral pancreatic bud evaginates in close proximity to the liver primordium but cannot be clearly identified until stage 14 when it appears as an evagination of the bile duct itself. At stage 16 (5 weeks) differential growth of the wall of the duodenum results in movement of the ventral pancreatic bud and the bile duct to the right side and ultimately to a dorsal position. It is not clear whether there is a corresponding shift of mesenchyme during this rotation.
However, the ventral pancreatic bud and the bile duct rotate from a position within the ventral mesogastrium (ventral mesoduodenum) to one in the dorsal mesogastrium (dorsal mesoduodenum) which is destined to become fixed onto the posterior abdominal wall. By stage 17 the ventral and dorsal pancreatic buds have fused, although the origin of the ventral bud from the bile duct is still obvious. Three-dimensional reconstruction of the ventral and dorsal pancreatic buds have confirmed that the dorsal pancreatic bud forms the anterior part of the head, the body and the tail of the pancreas and the ventral pancreatic bud forms the posterior part of the head and the posterior part of the uncinate process. The ventral pancreatic bud does not form all of the uncinate process (Collins 2002). The developing pancreatic ducts usually fuse in such a way that most of the dorsal duct drains into the proximal part of the ventral duct (Figs 90.3, 90.4). The proximal portion of the dorsal duct usually persists as an accessory duct. The fusion of the ducts takes place late in development or in the postnatal period: 85% of infants have patent accessory ducts as compared to 40% of adults. Fusion may not occur in 10% of individuals, in which case separate drainage into the duodenum is maintained, so-called pancreatic divisum. Failure of the ventral pancreatic diverticulum to migrate will result in an annular pancreas which may constrict the duodenum locally. UPDATE Date Added: 06 July 2005 Abstract: Case report of annular pancreas in male adult. Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=15779573&query_hl=6 Case report of annular pancreas in male adult. UPDATE Date Added: 28 June 2005 Abstract: Annular pancreas: magnetic resonance cholangiopancreatography findings. Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=15232386&query_hl=9 Annular pancreas: magnetic resonance cholangiopancreatography findings. UPDATE Date Added: 28 June 2005 Abstract: Annular pancreas: magnetic resonance cholangiopancreatography findings. Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=15232386&query_hl=9 Annular pancreas: magnetic resonance cholangiopancreatography findings. The ventral pancreas does not always extend anterior to the superior mesenteric vein but remains related to its right lateral surface. Initially the body of the pancreas extends into the dorsal mesoduodenum and then cranially into the dorsal mesogastrium. As the stomach rotates, this portion of the dorsal mesogastrium is directed to the left forming the posterior wall of the lesser sac. The posterior layer of this portion of dorsal mesogastrium fuses with the parietal layer of the coelom wall (peritoneum) and the pancreas becomes mainly retroperitoneal. The region of fusion of the dorsal mesogastrium does not extend so far left as to include the tail of the pancreas which passes into the splenorenal (lienorenal) ligament. The anterior border of the pancreas later provides the main line of attachment for the posterior leaves of the greater omentum. UPDATE Date Added: 30 May 2006 Abstract: Fetal development of pancreaticoduodenal arteries Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=15478102&query_hl=10&itool=pubmed_docsum The pancreaticoduodenal arteries in human foetal development. Krakowiak-Sarnowska E, Flisinski P, Szpinda M et al: Folia Morphol (Warsz) 63:281-284, 2004. Cellular development of the pancreas
The early specification of pancreatic endoderm involves the proximity of the notochord to the dorsal endoderm, which locally represses the expression of Shh transcription factor. Endoderm caudal to the pancreatic region does not respond to notochordal signals. The ventral pancreatic endoderm does not seem to undergo the same induction. Pancreatic mesenchyme is derived from two regions. The mesenchyme which surrounds the dorsal pancreatic bud proliferates from the splanchnopleuric coelomic epithelium of the medial walls of the
pericardioperitoneal canals, whereas the ventral pancreatic bud is invested by septum transversum mesenchyme and by mesenchyme derived from the lower ventral walls of the pericardioperitoneal canals. The primitive endodermal ductal epithelium provides the stem cell population for all the secretory cells of the pancreas. Initially these endocrine cells are located in the duct walls or in buds developing from them, and later they accumulate in pancreatic islets. The remaining primitive duct cells will differentiate into definitive ductal cells. In the fetus they develop microvilli and cilia but lack the lateral interdigitations seen in the adult. Branches of the main duct become interlobular ductules which terminate as blind ending acini or as tubular, acinar elements. The ductal branching pattern and acinar structure of the pancreas is determined by the pancreatic mesenchyme. This mesenchyme gives rise to connective tissue between the ducts which, in the fetus, appears to be important in stimulating pancreatic proliferation and maintaining the relative proportions of acinar, ! and " cells during development. It also provides cell lines for smooth muscle within the pancreas. Angiogenic mesenchyme invades the developing gland to produce blood and lymphatic vessels. The process of islet differentiation is divided into two phases (Collins 2002). Phase I, characterized by proliferation of polyhormonal cells, occurs from weeks 9-15. Phase II, characterized by differentiation of monohormonal cells, is seen from week 16 onwards. The " cells, producing insulin and amylin, differentiate first, followed by !-cells which produce glucagon. The # cells which produce somatostatin are seen after 30 weeks. The dorsal bud gives rise mostly to ! cells, and the ventral bud to most of the pancreatic polypeptide producing cells. The " cells develop from the duct epithelium throughout development and into the neonatal period. Later, in weeks 10-15, some of the primitive ducts differentiate into acinar cells in which zymogen granules or acinar cell markers can be detected at 12-16 weeks. The pancreas in the neonate has all of the normal subdivisions of the adult. The head is proportionately large in the newborn and there is a smooth continuation between the body and the tail. The inferior border of the head of the pancreas is found at the level of the second lumbar vertebra. The body and tail pass cranially and to the left, and the tail is in contact with the spleen (Fig. 11.4). Liver
The liver is one of the most precocious embryonic organs and is the main centre for haemopoiesis in the fetus. It develops from an endodermal evagination of the foregut and from septum transversum mesenchyme which is derived from the proliferating coelomic epithelium in the protocardiac region. The development of the liver is intimately related to the development of the heart. The vitelline veins, succeeded by the umbilical veins passing to the sinus venosus are disrupted by the enlarging septum transversum to form a hepatic plexus, the forerunner of the hepatic sinusoids (p. 1046). (See Collins 2002 for a detailed account of hepatic development.) Early liver development
As the head fold and early intraembryonic coelom form, the ventral parietal wall of the pericardial cavity gives rise to populations of cells termed precardiac or cardiac mesenchyme. Hepatic endoderm is induced to proliferate by this mesenchyme, although all portions of the early heart tube, truncus arteriosus, atria, ventricle, both endocardium and myocardium, have hepatic induction potency which is tissue-specific, but not species-specific. As the heart and foregut become separated by the accumulation of the cardiac mesenchyme, the mesenchyme itself is renamed septum transversum. It is seen as a ventral mass, caudal to the heart which passes dorsally on each side of the developing gut to join the mesenchyme proliferating from the walls of the pericardioperitoneal canals. The majority of the cells within the septum transversum are destined to become hepatic mesenchyme. In the stage 11 embryo the location of the hepatic endoderm has been identified at the superior boundary of the rostral intestinal portal. By stage 12, the hepatic endodermal primordium is directed ventrally and begins to proliferate as a diverticulum. There are two parts: a caudal part, which will produce the cystic duct and gallbladder, and a cranial part which forms the liver biliary system (Figs 90.3, 90.5). The cells start to express liver-specific molecular markers and glycogen storage. Around the cranial portion of the hepatic diverticulum the basal lamina is
progressively disrupted and individual epithelial cells migrate into the surrounding septum transversum mesenchyme. The previously smooth contour of the diverticulum merges into columnar extensions of endoderm, the epithelial trabeculae, which stimulate the hepatic mesenchymal cells to form blood islands and endothelium. The advance of the endodermal epithelial cells promotes the conversion of progressively more hepatic mesenchyme into endothelium and blood cells, and only a little remains to form the scanty liver capsule and interlobular connective tissue. This invasion by the hepatic epithelium is completed in stage 13, when it approaches the caudal surface of the pericardial cavity, and is separated from it only by a thin lamina of mesenchyme which will give rise to part of the diaphragm. During this early phase of development the liver is far more highly vascularized than the rest of the gut. The hepatic capillary plexus is connected bilaterally with the right and left vitelline veins. Dorsolaterally they empty by multiple channels into enlarged hepatocardiac channels, which lead to the right and left horns of the sinus venosus; usually the channel on the right side is most developed. Both left and right channels bulge into the pericardioperitoneal canals, forming sites for the exchange of fluid from the coelom into the vascular channels. The growth of the hepatic tissue in these regions is sometimes referred to as the left and right horns of the liver. page 1255 page 1256
Figure 90.5 The hepatic endodermal primordium proliferates ventrally into the septum transversum mesenchyme. The endodermal cells forming the hepatic trabeculae will become hepatocytes; the septum transversum mesenchymal cells will become the endothelium of the liver sinusoids and early blood cells. (Modified with permission from Collins P 2002 Embryology of the liver and bile ducts. In: Howard, ER, Stringer MD, Clombani PM (eds) Surgery of the Liver, Bile Ducts and Pancreas in Children, Chapter 7, Part 3. London: Arnold.)
The liver remains proportionately large during its development and constitutes a sizeable organ dorsal to the heart at stage 14 (Fig. 90.8B,C), then more caudally placed by stage 16. By this stage hepatic ducts can be seen separating the hepatic epithelium from the extrahepatic biliary system, but even at stage 17 the ducts do not penetrate far into the liver. Maturation of the liver
At 3 months' gestation, the liver almost fills the abdominal cavity and its left lobe is nearly as large as its right. When the haematopoietic activity of the liver is assumed by the spleen and bone marrow the left lobe undergoes some degeneration and becomes smaller than the right. The liver remains relatively larger than in the adult throughout the remainder of gestation. In the neonate it constitutes 4% of the body weight, compared to 2.5-3.5% in adults. It is in contact with the greater part of the diaphragm and extends below the costal margin anteriorly, and in some cases to within 1 cm of the iliac crest posteriorly. The left lobe covers much of the anterior surface of the stomach and constitutes nearly one-third of the liver (Figs 11.4, 65.5). Although its haemopoietic functions cease before birth its enzymatic and synthetic functions are not completely mature at
birth. Development of intrahepatic biliary ducts
The development of the intrahepatic biliary ducts follows the branching pattern of the portal vein radicles (Collins 2002). The cranial hepatic diverticulum gives rise to the liver hepatocytes, the intrahepatic large bile ducts (right and left hepatic ducts, segmental ducts, area ducts and their first branches) and the small bile ducts (septal bile ducts, interlobular ducts and bile ductules). The portal and hepatic veins arise together from the vitelline veins. Early in development the accumulation of mesenchyme around these veins is similar, whereas later mesenchyme increases around the portal veins. This is a prerequisite for bile duct development. Primitive hepatocytes surround the portal vein branches and associated mesenchyme and form a sleeve of cells termed the ductal plate. Portions of the ductal plate divide to produce lines of epithelial cells which migrate close to a portal vein branch where they differentiate into bile ducts. As the bile ducts develop, angiogenic mesenchymal cells form blood vessels which connect to the hepatic artery from 10 weeks. Thus the portal triads are patterned by the portal vein radicles which initially induce bile duct formation and then artery formation. The development of the biliary system extends from the hilum to the periphery. Abnormalities of the biliary tree are associated with abnormalities of the branching pattern of the portal vein. The developing bile ducts remain patent throughout development; the solid stage of ductal development previously promulgated has been refuted. Atresia of the extrahepatic bile ducts has been noted, often in assocition with extrahepatic atresia. The cause of this condition is not clear; inflammatory process may be involved, although some cases have features of ductal plate malformation (Howard 2002). Development of extrahepatic biliary ducts
The caudal part of the hepatic endodermal diverticulum forms the extrahepatic biliary system, the common hepatic duct, gallbladder, cystic duct and common bile duct. The bile duct, which originated from the ventral wall of the foregut (now duodenum), migrates with the ventral pancreatic bud first to the right and then dorsomedially into the dorsal mesoduodenum. The right and left hepatic ducts arise from the cranial end of the common hepatic duct from 12 weeks' gestation. Atresia of the extrahepatic bile ducts in neonates occurs alone or in conjunction with a range of other anomalies, including situs inversus, malrotation, polysplenia and cardiac defects. In such cases the intrahepatic bile ducts have a mature tubular shape but also show features of ductal plate malformation. In the neonate the gallbladder has a smaller peritoneal surface than in the adult, and its fundus often does not extend to the liver margin. It is generally embedded in the liver and in some cases may be covered by bands of liver. After the second year the gallbladder assumes the relative size it has in the adult.
© 2008 Elsevier
MIDGUT The midgut forms the third and fourth parts of the duodenum, jejunum, ileum and two-thirds of the way along the transverse colon: its development produces most of the small and a portion of the large intestine. In embryos of stages 10 and 11 it extends from the cranial to the caudal intestinal portals and communicates directly with the yolk sac over its entire length. Although it has a dorsal wall, at these stages the lateral walls have not yet formed. By stage 12 the connection with the yolk sac has narrowed such that the midgut has ventral walls cranially and caudally. This connection is reduced to a yolk stalk containing the vitellointestinal duct during stage 13, at which time the yolk sac appears as a sphere in front of the embryo. Posterior to the midgut the splanchnopleuric coelomic epithelia converge forming the dorsal mesentery. Ventrolaterally the intraembryonic coelom is in wide communication with the extraembryonic coelom. At stage 14 the midgut has increased in length more than the axial length of the embryonic body and, with elongation of the dorsal mesentery, it bulges ventrally, deviating from the median plane. For all these stages consult Fig. 90.3.
PRIMARY INTESTINAL (OR MIDGUT) LOOP page 1256 page 1257
Figure 90.6 Three-dimensional schematization of the major developmental sequences of the subdiaphragmatic embryonic and fetal gut, together with its associated major glands, peritoneum and mesenteries: left anterolateral aspect. The development sequence A-F spans 1.5 months to the perinatal period. Three-dimensional schematization of the major developmental sequences of the subdiaphragmatic embryonic and fetal gut, together with its associated major glands, peritoneum and mesenteries: left anterolateral aspect. The development sequence A-F spans 1.5 months to the perinatal period. H denotes the general disposition of the remaining viscera, mesenteric roots with their lines of attachment, and principal contained vessels, which approximate to the adult state for comparison.
The midgut loop can first be seen at stage 15 when a bulge, the caecal bud, can be discerned on the lower limb of the loop, caudal to the yolk stalk (which arises from the apex or summit of the loop) (Figs 90.3, 90.6). Later, the original proximal limb of the loop moves to the right and the distal limb to the left. The longest portion of the dorsal mesentery is at the level of the yolk stalk: there is less relative lengthening near the caudal end of the duodenum or the cranial half of the colon. The midgut extends into the umbilical coelom having already rotated through an angle of 90° (anticlockwise viewed from the ventral aspect). This relative position is approximately maintained so long as the protrusion persists, during which time the proximal limb which forms the small intestine elongates
greatly. It becomes coiled, and its adjacent mesentery adopts a pleated appearance. The origin of the root of the mesentery is initially both median and vertical, while at its intestinal attachment it is elongated like a ruffle and folded along a horizontal zone. The mesenteric sheet and its contained vessels has spiralled through 90°. The distal, colic, part of the loop elongates less rapidly and has no tendency to become coiled. By the time the fetus has attained a length of 40 mm (10 weeks), the peritoneal cavity has enlarged and the relative size of the liver and mesonephros is much less. The re-entry of the gut occurs rapidly and in a particular sequence during which it continues the process of rotation. The proximal loop returns first, with the jejunum mainly on the left and the ileum mainly on the right of the subhepatic abdominal cavity. As they re-enter the abdominal cavity the coils of jejunum and ileum slide inwards over the right aspect of the descending mesocolon, and so displace the descending colon to the left. The transverse colon passes superiorly to the origin of the root of the mesentery. The caecum is the last to re-enter and at first lies on coils of ileum on the right. Later development of the colon leads to its elongation and to the establishment of the hepatic and splenic flexures. Anomalies of midgut rotation
If the midgut loop fails to return to the abdominal cavity at the appropriate time a range of ventral defects can result. Failure of obliteration of the vitelline-intestinal duct connecting the midgut to the yolk sac results in Meckel's diverticulum. This may present as a short segment of vitelline duct attached to the original ventral side of the ileum; it may remain attached to the umbilicus as a fistula; or it may remain as a ligamentous attachment to the umbilicus. page 1257 page 1258
An umbilical hernia occurs when loops of gut protrude into a widened umbilical cord at term. The degree of protuberance may increase when the infant cries, which raises the intra-abdominal pressure: these hernias usually resolve without treatment. Exomphalos is a ventral wall defect with midline herniation of the intraabdominal contents into the base of the umbilical cord. Herniated viscera are covered by the peritoneum internally and amnion externally. The omphalocoele so formed ranges in size from a large umbilical hernia to a very large mass containing most of the visceral organs. Even after the exomphalos has been repaired these babies will still have a deficient anterior abdominal wall. page 1258 page 1259
Gastroschisis is a para-umbilical defect of the anterior abdominal wall associated with evisceration of the abdominal organs. The organs are not enclosed in membranes, thus gastroschisis can be detected by prenatal ultrasonography and differentiated from exomphalos. Gastroschisis is thought to result from periumbilical ischaemia caused by vascular compromise of either the umbilical vein or arteries. The incidence of this condition appears to be increasing, especially in babies born to young women less than 20 years old (Whittle 1999). Congenital volvulus arises if the midgut loop does not rotate appropriately. A number of types of this condition are identified. Left-sided colon occurs if the midgut loop has not rotated at all; mixed rotation results in the caecum lying
inferior to the pylorus; failure of attachment of the peritoneum appropriately may result in the small intestine being attached at only two points on the posterior abdominal wall. All of these arrangements lead to a risk of volvulus which may result in necrosis of the gut. The position and configuration of the duodenal loop are of particular importance in children. The normal duodenal loop has a U-shaped configuration. The suspensory ligament of the duodenum (ligament of Treitz) is usually found to the left of the body of the first or second lumbar vertebral body after normal gut rotation: any other position of this ligament may indicate some degree of gut malrotation. On barium studies the duodenojejunal flexure should thus lie to the left of the upper lumbar spine at the level of the pylorus. If the caecum has remained in the right upper quadrant it may become fixed in that position by peritoneal attachments passing to the right, so called Ladd's bands. These may compress the underlying duodenum and give rise to duodenal stenosis. The high positioning of the caecum close to the duodenal jejunal flexure, in some cases in the midline, is associated with later development of volvulus. The identification of intestinal malrotation can be made by X-ray investigation, however, ultrasonography has the advantage of showing the position of the superior mesenteric vein and artery. The vein should lie to the right of the artery. Most cases of volvulus will show inversion of this normal relationship, but malrotation can occur with apparently normally related vessels, particularly in malrotation with bowel obstruction due to Ladd's bands and not volvulus. UPDATE Date Added: 12 July 2005 Shanida Helena Nataraja, PhD (Dianthus Medical Limited) Update: Disorders of intestinal rotation and fixation. Diagnostic imaging of malrotation and malrotation with volvulus is challenging. Prompt diagnosis and treatment of this acute life-threatening condition is crucial. Strouse provides a brief overview of the embryology of the intestine, followed by detailed information on cases of malrotation (i.e. failure during development of normal rotation of any part of the intestinal tract). Malrotation occurs in approximately 1 in 500 live births and is most commonly noted on prenatal sonography. The classical presentation of malrotation with volvulus is bilious vomiting. However, this symptom is not exclusive to neonates with malrotation, and 62% of infants will prove not to have an anatomic obstruction. The majority of children presenting with malrotation do not have any predisposing syndrome or genetic susceptibility, but malrotation is invariably present in children with congenital diaphragmatic hernia, gastroschisis and omphalocele, because these conditions all interfere with the normal spatial development of the gut. The imaging workup of a child with suspected malrotation should start with radiography (anteroposterior supine view and either an anteroposterior upright view or cross-table lateral view). These investigations rarely suggest the diagnosis of malrotation but are useful in terms of their ability to exclude other etiologies and guide further imaging studies. The preferred modality for the radiologic diagnosis of malrotation and malrotation with volvulus is an upper GI using barium: signs of malrotation include abnormal position of the
duodenojejunal junction; spiral, corkscrew, or Z-shaped course of the distal duodenum and proximal jejunum; and location of the proximal jejunum in the right abdomen. In most patients, upper GI will allow a clear distinction between normal or abnormal anatomy, but this distinction cannot always be made definitively because malrotation represents a spectrum of abnormality. Other imaging modalities, e.g. ultrasound, computerized tomography and magnetic resonance imaging, are not the preferred diagnostic modality for malrotation, but they can be used to exclude other disease processes that may present in a similar manner to malrotation and malrotation with volvulus. The standard approach to treatment is surgical, and the Ladd's procedure, which can be performed laparoscopically, is associated with high survival rates. Strouse PJ: Disorders of intestinal rotation and fixation ("malrotation"). Pediatr Radiol 34(11):837-851, 2004. Medline Similar articles
UMBILICAL CORD (See also p. 1341) During the period when the midgut loop protrudes into the umbilical coelom, the edges of the ventral body wall are becoming relatively closer, forming a more discrete root for the umbilical cord. Somatic mesenchyme, which will form the ventral body wall musculature, migrates into the somatopleuric mesenchyme and passes ventrally toward the midline. The umbilical cord forms all of the ventral body wall between the pericardial bulge and the developing external genitalia. It encloses a portion of the extraembryonic coelom, the umbilical coelom, into which midgut loop protrudes. When the midgut loop is abruptly returned to the abdominal cavity the more recognizable umbilical cord forms. The vitellointestinal duct and vessels involute. The cranial end of the allantois becomes thinned and its lumen partially obliterated, and it forms the urachus. The mesenchymal core of the umbilical cord is derived by coalescence from somatopleuric amniotic mesenchyme, splanchnopleuric vitellointestinal (yolk sac) mesenchyme, and splanchnopleuric allantoic (connecting stalk) mesenchyme. These various layers become fused and are gradually transformed into the viscid, mucoid connective tissue (Wharton's jelly) which characterizes the more mature cord. The changes in the circulatory system result in a large cranially oriented left umbilical vein (the right umbilical vein regresses), and two spirally disposed umbilical arteries.
MATURATION OF THE SMALL INTESTINE Mucosa
The exact timing of the cellular morphogenesis of the gut is difficult to establish, especially as it undergoes a proximodistal gradient in maturation. Developmental differences between parts of the small intestine or colon have not yet been correlated with age. The endodermal cells of the small intestine proliferate and form a layer some three to four cells thick with mitotic figures throughout. From 7 weeks, blunt projections of the endoderm have begun to form in the duodenum and proximal jejunum; these are the developing villi which increase in length until in the duodenum the lumen becomes difficult to discern. The concept of occlusion of the lumen and recanalization which is described in many accounts of development does not match the cytodifferentiation which occurs in the gut epithelia. Thus it is no longer thought that there is secondary recanalization of the gut lumen. By 9 weeks the duodenum, jejunum and proximal ileum have villi and
the remaining distal portion of ileum develops villi by 11 weeks. The villi are covered by a simple epithelium. Primitive crypts, epithelial downgrowths into the mesenchyme between the villi, appear between 10 and 12 weeks similarly along a craniocaudal progression. Brunner's glands are present in the duodenum from 15 weeks and the muscularis mucosa can be seen in the small intestine from 18 weeks. Whereas mitotic figures are initially seen throughout the endodermal layer of the small intestine prior to villus formation, by 10-12 weeks they are limited to the intervillous regions and the developing crypts. It is believed that an 'adult' turnover of cells may exist when rounded-up cells can be observed at the villus tips, in position for exfoliation. The absorptive enterocytes have microvilli at their apical borders before 9 weeks. An apical tubular system appears at this time composed of deep invaginations of the apical plasma membrane and membrane-bound vesicles and tubules; many lysosomal elements (meconium corpuscles) appear in the apical cytoplasm. These latter features are more developed in the ileum than jejunum, are most prominent at 16 weeks, and diminish by 21 weeks. There are abundant deposits of glycogen in the fetal epithelial cells, and it has been suggested that prior to the appearance of hepatic glycogen the intestinal epithelium serves as a major glycogen store. Goblet cells are present in small numbers by 8 weeks, Paneth's cells differentiate at the base of the crypts in weeks 11 and 12, and enteroendocrine cells appear between weeks 9 and 11. M cells (membrane or microfold cells) are present from 14 weeks. Meconium can be detected in the lumen of the intestine by the 16th week. It is derived from swallowed amniotic fluid which contains vernix and cellular debris, salivary, biliary, pancreatic and intestinal secretions, and sloughed enterocytes. As the mixture passes along the gut, water and solutes are removed and cellular debris and proteins concentrated. Meconium contains enzymes from the pancreas and proximal intestine in higher concentrations in preterm than full-term babies. Muscularis layer
The muscularis layer is derived from the splanchnopleuric mesenchyme as it is in other parts of the gut. Longitudinal muscle can be seen from 12 weeks. At 26-30 weeks the gut shows contractions without regular periodicity; from 30-33 weeks repetitive groups of regular contractions have been seen in preterm neonates. Serosa
The small intestine possesses only a dorsal mesentery. The movement of the root of this dorsal mesentery, and the massive lengthening of its enteric border in order to match the longitudinal growth of the gut tube, reflect the spiralizing of the midgut loop in the umbilical coelom. The specific regions of adherence of the serosa and parietal peritoneum of the small intestine in the peritoneal cavity are given on page 1261. Small intestine at birth
In the neonate the small intestine forms an oval-shaped mass with its greater diameter transversely orientated in the abdomen, rather than vertically as in the adult (Fig. 65.5). The mass of the small intestine inferior to the umbilicus is
compressed by the urinary bladder which is anterior at this point. The small intestine is 300-350 cm long at birth and its width when empty is 1-1.5 cm. The ratio between the length of the small and large intestine at birth is similar to the adult ratio. The mucosa and submucosa are fairly well developed and villi are present throughout the small intestine, however, some epithelial differentiation is incomplete. The muscularis is very thin, particularly the longitudinal layer, and there is little elastic tissue in the wall. There are few or no circular folds in the small intestine, and the jejunum and ileum have little fat in their mesentery.
© 2008 Elsevier
PRIMITIVE HINDGUT page 1259 page 1260
Just as the foregut has an extensive, ventral endodermal diverticulum which contributes to a system separate from the gut, so too the hindgut has a ventral diverticulum, the allantois, destined for a different system. However, unlike the respiratory diverticulum of the foregut the allantois is formed very early in development, prior even to formation of the embryonic endoderm and tail folding. With the reorganization of the caudal region of the embryo at stage 10, part of the allantois is drawn into the body cavity. The early embryonic hindgut thus consists of a dorsal tubular region extending from the caudal intestinal portal to the cloacal membrane, and a ventral blind-ending allantois extending from the cloacal region into the connecting stalk. The slightly dilated cavity, lined by endoderm, that cranially receives the enteric hindgut proper and the root of the allantoenteric diverticulum is termed the endodermal cloaca. It is closed ventrally by the cloacal membrane (endoderm opposed to proctodeal ectoderm), and it also has, transiently, a small recess of endoderm in the root of the tail, the postanal gut. As elsewhere, the hindgut, allantois and endodermal cloaca are encased in splanchnopleuric mesenchyme. Proliferation of the mesenchyme and endoderm in the angle of the junction of hindgut and allantois produces a urorectal septum (Fig. 109.4). Continued proliferation of the urorectal septum and elongation of the endodermal structures thrusts the endodermal epithelium towards the cloacal membrane with which it fuses centrally, separating the presumptive rectum and upper anal canal (dorsally) from the presumptive urinary bladder and urogenital sinus (ventrally) (Figs 109.4, 109.7). The cloacal membrane is thus divided into anal (dorsal) and urogenital (ventral) membranes. The nodal centre of division is the site of the future perineal body, the functional centre of the perineum. Details of the development of the allantoic hindgut are given on page 1391.
ENTERIC HINDGUT The development of the large intestine, whether derived from mid- or hindgut, seems to be similar. The proximal end of the colon can be first identified at stage 15 when an enlargement of a local portion of gut on the caudal limb of the midgut loop defines the developing caecum. An evagination of the distal portion of the caecum forms the vermiform appendix at stage 17 (Fig. 90.3). Apart from the embryonic studies of Streeter (1942) there is little information about the development of the large intestine in humans. The early endodermal lining of the colon appears stratified, and mitoses occur throughout the layers. A series of longitudinal folds arise initially at the rectum and caecum and later in the regions of colon between these two points. The folds segment into villi with new villi forming between. The developing mucosa invaginates into the underlying mesenchyme between the villi to form glands which increase in number by splitting longitudinally from the base upwards. The villi gradually diminish in size and are absent by the time of birth.
MATURATION OF THE LARGE INTESTINE The similarity of development of the small and large intestines is further mirrored
in their cytological differentiation. Fetal gut from 11 weeks shows dipeptidase activity in the colon as well as in the small intestine. Throughout preterm development meconium corpuscles are seen in the colon and in the small intestine: they are believed to be the phagocytosed remains of neighbouring cells which have died as a result of programmed cell death. There is little direct evidence of colonic function in the human fetus and neonate. However, the specific results of mammalian studies are being correlated to human studies where possible. A number of distinct and important differences between the function of adult and fetal colon have been reported. Mucosa
The absorption of glucose and amino acids does not take place through the colonic mucosa in adult life, but there is evidence of direct absorption of these nutrients during development. At birth the normal cycle of bile acids is not mature. In the adult, bile is secreted by the liver, stored in the gallbladder and then secreted into the intestine where it is absorbed by the jejunum and ileum. In the fetus and neonate, the transport of bile acids by an active process from the ileum does not occur, and so bile salts pass on into the colon. In the adult the presence of bile salts in the colon stimulates the secretion of water and electrolytes which results in diarrhoeal syndrome; however, the fetal and neonatal colon seems protected from this effect. The colon is not considered a site of significant nutrient absorption in the adult, and yet neonates are unable to assimilate the full lactose load of a normal breast feed from the small intestine and a large proportion of it may be absorbed from the colon. Thus it appears that the colon fulfills a slightly different role in the preterm and neonatal period, conserving nutrient absorption and minimizing fluid loss until the neonate has adjusted to extrauterine life, oral feeding, and the establishment of the symbiotic bacterial flora. Muscularis
The muscularis is present and functioning by the 8th week, when peristaltic waves have been observed. The specific orientation of the longitudinal muscle layer into taeniae coli occurs in the 11th to 12th weeks when haustra appear. The enteric nerves are present in Meissner's and Auerbach's plexuses at 8 and 12 weeks respectively: there is a craniocaudal migration of neurones into the gut wall. A normal distribution of ganglion cells has been noted in preterm babies of 24 weeks, although there is a region devoid of ganglia just above the anal valves. Abnormal migration of neural crest cells to the gut may give rise to Hirschsprung's disease (p. 1261). Puborectalis appears in 20-30 mm embryos, following opening of the anal membrane. Serosa
The development of the serosa of the intestine is considered with the development of the peritoneal cavity (p. 1261). Colon at birth
In the neonate the colon is c.66 cm long and averages 1 cm in width. The caecum is relatively smaller than in the adult; it tapers into the vermiform appendix. The ascending colon is shorter in the neonate, reflecting the shorter lumbar region.
The transverse colon is relatively long, whereas the descending colon is short, but twice the length of the ascending colon (Figs 11.4, 65.5). The sigmoid colon may be as long as the transverse colon; it often touches the inferior part of the anterior body wall on the left and, in c.50% of neonates, part of the sigmoid colon lies in the right iliac fossa. The muscularis, including the taeniae coli, is poorly developed in the colon as it is in the small intestine. Appendices, epiploicae and haustra are not present, which gives a smooth external appearance to the colon. Haustra appear within the first 6 months. The rectum is relatively long; its junction with the anal canal forms at nearly a right angle.
ANAL CANAL Mesenchymal proliferation occurs around the rim of the ectodermal aspect of the anal membrane which thus comes to lie at the bottom of a depression, the proctodeum (Fig. 109.4). With the absorption and disappearance of the anal membrane the anorectum communicates with the exterior. The lower part of the anal canal is formed from the proctodeal ectoderm and underlying mesenchyme, but its upper part is lined by endoderm. The line of union corresponds with the edges of the anal valves in the adult. The dual origin of the anal canal is reflected in its innervation: the endodermal portion is innervated by autonomic nerves, and the ectodermal proctodeum is innervated by spinal nerves. In the fourth and fifth weeks a small part of the hindgut, the postanal gut, projects caudally beyond the anal membrane towards the region of the tail; it usually disappears before the end of the fifth week. Imperforate anus is a term used to describe many different anorectal malformations. The most common is anal agenesis which is found in c.45% of all cases of imperforate anus. The condition is usually associated with a fistula which opens into the vulva (females) or into the urethra (males). It is more rare for the anal membrane to fail to perforate. The condition cannot reliably be diagnosed prenatally by ultrasound diagnosis, and it may be confused with Hirschsprung's disease and colonic atresia. The prognosis is good for low lesions of the anal canal. The principal concern in all cases is the degree of bowel control, urinary control and in some cases sexual function, which is compromised by the condition. Anorectal malformations may be indicators of other abnormalities, for example those forming the 'VATER' syndrome (Vertebral, Anal, TracheoOesophageal and Renal abnormalities).
© 2008 Elsevier
DEVELOPMENT OF GUT-ASSOCIATED LYMPHOID TISSUE AND ENTERIC NERVOUS SYSTEM DEVELOPMENT OF GUT-ASSOCIATED LYMPHOID TISSUE page 1260 page 1261
The neonatal gut becomes colonized by a range of bacterial flora, some of which exist in a symbiotic relationship with their host, some of which may be considered pathogenic. The gut plays a significant role in the defence of the body. Individual lymphocytes appear in the lamina propria of the gut from approximately week 12 of development, and lymphoid aggregates, Peyer's patches, have been noted between 15 and 20 weeks: it is not clear whether these cells migrate in from distant sources or differentiate from the investing mesenchyme. The endodermal epithelium overlying the lymphoid aggregates is often distorted into a dome shape. The cells within the dome are a mixed population of enterocytes, endocrine cells, goblet cells and M cells. M cells are specialized to provide a mechanism for the transport of micro-organisms and intact macromolecules across the epithelium to the intraepithelial space and lamina propria. They have been observed in the fetus by 17 weeks; it is believed that they are formed by a specialized epithelial/mesenchymal interaction of the endoderm and underlying lymphoid type mesenchyme. There are similarly specialized epithelial cells between the enterocytes. Intraepithelial leukocytes account for c.15% of the epithelial cells of the gut in the adult. They have been observed at 11 weeks, with a distribution of c.3 intraepithelial leukocytes/100 gut epithelial cells. Both T and B lymphocytes have been described in the developing gut wall. For an account of the development of the immune cells of the gut consult Butzner & Befus (1989).
DEVELOPMENT OF THE ENTERIC NERVOUS SYSTEM Enteric neurones are derived from trunk neural crest cells at somite levels 1-7 and from 28 onwards (Fig. 14.11 and p. 254). After neurulation the crest cells begin their ventral migration and invade the gut via the dorsal mesentery. Glial cells associated with the gut have been identified as arising from similar levels. The local splanchnopleuric mesenchyme patterns the crest cells such that those which enter the gut layers attain an enteric fate, whereas those that remain outside the gut become committed as parasympathetic postganglionic neurones. The enteric neurones also migrate to the glands of the gut, e.g. the pancreas. Hirschsprung's disease
Hirschsprung's disease is usually characterized by an aganglionic portion of gut which does not display peristalsis, and a dilated segment of colon proximal to this site. Histologically there is either an absence or a reduction in the number of ganglia and postganglionic neurones in the myenteric plexus; postganglionic innervation of the muscle layers is also often defective. It is believed that the condition is caused by a failure of neural crest cells to colonize the gut wall appropriately. An overabundance of basal laminal components, perhaps at the
mesothelial/mesenchyme interface, may prevent the early migrating neural crest cells from penetrating the gut wall; their new position outside the gut does not confer on them the environmental stimuli for enteric nerve differentiation and so non-enteric development occurs in local ganglia adjacent to the gut (Gershon 1987). A variable length of large intestine may be affected: the lower and midrectum are the most common sites, but in severe cases the rectum, sigmoid, descending and even proximal colon can be aganglionic. The chronic dilatation of the colon or rectum proximal to the affected segment gives rise to the common name, megacolon. It occurs as a consequence of functional obstruction due to the failure of peristalsis within the affected segment, and the dilated colon is structurally normal. Occasionally aganglionosis affects only a very short length of rectum proximal to the anorectal junction and the degree of functional obstruction is minimal: in these cases of 'ultra-short segment Hirschsprung's disease', clinical abnormalities arise later in life. Infants with Hirschsprung's disease show delay in the passage of meconium, constipation, vomiting and abdominal distension.
© 2008 Elsevier
FUNCTIONAL MATURITY OF THE GUT AT BIRTH After birth the first passage of stool of the newborn is termed meconium. This is a dark, sticky, viscid substance formed from the passage of amniotic fluid, sloughed cells, digestive enzymes and bile salts along the fetal gut. Meconium becomes increasingly solid as gestation advances but does not usually pass out of the fetal body while in utero. Fetal distress produced by anoxia may induce the premature defecation of meconium into the amniotic fluid, with the risk of its inhalation. At birth the colon contains 60-200 g of meconium. The majority of neonates defaecate within the first 24 hours after birth. Delayed passage of stool beyond this time is associated with Hirschsprung's disease (p. 1261) or imperforate anus (p. 1261). The normal passage of meconium continues for the first 2 or 3 days after birth, and is followed by a transition to faecal stools by day 7.
© 2008 Elsevier
DEVELOPMENT OF THE PERITONEAL CAVITY The early development of the intraembryonic coelom which gives rise to the peritoneal cavity is described on pages 198 and 207. Fig. 10.23 show the shape of the early peritoneal cavity and indicate the mesenchymal populations derived from its epithelial walls. Initially the peritoneal cavity associated with the lower end of the foregut has separate right and left components, the pleuroperioneal canals (Fig. 90.1). At the level of the midgut, the pleuroperitoneal canals join a confluent cavity surrounding the developing gut, which transitorily is in communication with the extraembryonic coelom. The description of the development of the peritoneal cavity which follows pertains to changes which occur as a consequence of the differential growth of the gut.
PERITONEUM Peritoneum develops from a specific portion of the intraembryonic coelomic walls (pp. 198, 207). Initially the intraembryonic coelomic epithelium is a pseudostratified germinal layer from which cellular progeny with different fates arise in specific sites and at specific developmental times. The portion which will give rise to the peritoneum is derived from the lower portion of the pericardioperitoneal canals and the somatopleure and splanchnopleure associated with the lower foregut, midgut and upper portions of the hind gut (Figs 10.24, 90.1). The proliferative splanchnopleuric epithelium produces cell populations for the mucosa and muscularis of the gut and also the lamina propria and epithelium of the visceral peritoneum, (the serosa of the gut wall). The somatopleuric epithelium gives rise to the lamina propria and epithelium of the parietal peritoneum. The visceral and parietal peritoneal layers constitute a mesothelium, which denotes their origin from the intraembryonic mesoderm of the coelomic wall. As the gut grows, splanchnic mesenchyme accumulates around the endodermal epithelium and the whole unit generally moves ventrally. There is a concomitant enlargement of the caudal ends of the developing pericardioperitoneal (pleuroperitoneal) canals and developing peritoneal cavity. The medial walls of the intraembryonic coelom move closer and there is a relative decrease in the mesenchyme between them. The regions where the medial portions of the intraembryonic coelom come together are termed mesenteries. They are composed of two layers of peritoneum with intervening mesenchyme and contain the neurovascular structures which pass to and from the gut. At the caudal ends of the pleuroperitoneal canals the gut has both ventral and dorsal mesenteries, whereas caudal to this there is only a dorsal mesentery. The mesenteries attached to the gut lengthen to permit large movements or rotations of the gut tube. Later, when part or the whole of the mesentery lies against the parietal peritoneum, their apposed surfaces fuse and are absorbed, i.e. they become sessile. Only those viscera developed in direct apposition to one of the primary coelomic regions, or a secondary extension of the latter, retain a partial or almost complete visceral serous cover. Thus the original line of reflexion of mesenteries becomes altered, or in some cases the organ may become retroperitoneal. These mechanisms are significant throughout the subdiaphragmatic gut, but are predominant in the small and large intestine. All serous membranes may vary their thickness, lines of reflexion, disposition, 'space' enclosed and their channels of communication, by a process of areal and thickness growth on one aspect combined with cavitation (leading to expanding embryonic recess formation) on the other (Fig. 90.7A). Although all of the gut tube and its derived glands are associated with mesenteries formed as described above, the nomenclature for some portions of the gut and glands is different. Thus the mesenteries of the stomach are called omenta and the reflections of peritoneum around the liver, which develop from a
confluence of splanchnopleuric, somatopleuric and septum transversum-derived portions, are called ligaments. The movements of the developing viscera within the peritoneal cavity occur with associated movements of the mesothelia which surround them. The descriptions of peritoneal cavity development which follow are thus describing a sequence of changes which affect a complex space and its boundaries. Mesenteries of the developing gut page 1261 page 1262
page 1262 page 1263
Figure 90.7 Development of the subdiaphragmatic foregut and the right and left pericardioperitoneal/pleuroperitoneal canals, with particular reference to the terminal oesophagus, stomach, duodenum, spleen, the lesser sac of peritoneum and omenta: seen in semicoronal section (left column) and transverse section at the levels indicated (right column).
The cervicothoracic oesophagus develops between the pericardioperitoneal canals (Fig. 90.1). It is encased in prevertebral, retrotracheal and retrocardiac mesenchyme. As the pericardioperitoneal canals expand with the developing lung buds, and the diaphragm forms immediately below them, the oesophagus at this level has no true dorsal or ventral mesentery. At superior and intermediate thoracic levels parts of the lateral aspects of the oesophagus are closely related to the secondary, mediastinal, parietal pleura. In the lower thorax the oesophagus inclines ventrally anterior to the descending thoracic aorta. The dorsocaudally sloping midline diaphragm between oesophageal and aortic orifices may be homologized with part of a dorsal meso-oesophagus, and is used in that context in descriptions of diaphragmatic development. A ventral midline diaphragmatic strip may also be considered to be a derivative of a ventral meso-oesophagus, however, this region is more usually thought of as septum transversum. The alimentary tube, from the diaphragm to the start of the rectum, initially possesses a sagittal dorsal mesentery. Its line of continuity with the dorsal parietal peritoneum (i.e. its 'root' or 'line of reflexion') is initially also midline. The abdominal foregut, from the diaphragm to the future hepatopancreatic duodenal papilla, also has a ventral mesentery. This extends from the ventrolateral margins of the abdominal oesophagus and as yet 'unrotated' primitive stomach and proximal duodenum, cranially to the future diaphragm and anteriorly to the ventral abdominal wall (to the level of the cranial rim of the umbilicus). Caudally, between umbilicus and duodenum, it presents a crescentic free border. The midgut and hindgut have no ventral mesentery; thus the pleural and supraumbilical peritoneal cavities are initially, and transiently, bilaterally symmetrical above the umbilicus. Below the umbilicus, the peritoneal cavity is freely continuous across the midline ventral to the gut (Fig. 90.5). Foregut mesenteries
The ventral and dorsal foregut mesenteries are relatively large compared with the slender endodermal tubes they encase: they are composed of mesenchyme sandwiched between two layers of splanchnopleuric coelomic epithelium. (Compare the endodermal profile seen in Fig. 90.3 with the endoderm and surrounding splanchnopleure in Fig. 90.8.) A complex series of recesses develop in the splanchnopleuric mesenchyme and become confluent. As a result of foregut rotation, differential growth of stomach, liver, pancreas and spleen, and completion of the diaphragm, the territories of the greater sac and lesser sac (omental bursa) are delimited, and the mesenteric complexes of these organs (omenta and 'ligaments') are defined (Figs 90.6, 90.7, 90.8).
Consequences of rotation of the stomach
A number of processes occur concurrently which conceptually can be visualized as the movement of the right pleuroperitoneal canal to a position posterior to the stomach such that its communication with the remainder of the peritoneal cavity is reduced (p. 1252). These processes include the differential growth of the walls of the stomach, the formation and specific local extension of the omenta (dorsal and ventral mesogastria), the growth of the liver and particularly of the vessels and ducts which enter and leave the liver. These developments permit stomach expansion both anteriorly and posteriorly when food is ingested and free movement of peristalsis. The right pleuroperitoneal canal forms a discrete region of the peritoneal cavity, the lesser sac, and the remaining left pleuroperitoneal canal and the remainder of the peritoneal cavity form the greater sac. The entrance to the original right pleuroperitoneal canal (lesser sac) becomes reduced in size. It is called the epiploic foramen, foramen of Winslow, or the aditus of the omental bursa (bursa omentalis) (Fig. 90.8C). Early stages of lesser sac development
page 1263 page 1264
Figure 90.8 A, Transverse section of a human embryo, 8 mm long, showing the right pneumatoenteric recess. B, Transverse section through the same embryo as A but 530 mm more caudally. Note that rotation of the stomach has taken place and that the sinusoidal spaces in the liver communicate freely with one another. C, Transverse section through the same embryo as B, but 150 mm more caudally. Compare with the preceding figure and observe that the omental bursa (pancreaticoenteric recess) communicates with the general peritoneal cavity at this level.
The lesser sac is first indicated by the appearance of multiple clefts in the paraoesophageal mesenchyme on both left and right aspects of the oesophagus. Although they may become confluent, the left clefts are transitory and soon atrophy. The right clefts merge to form the right pneumatoenteric recess that extends from the oesophageal end of the lesser curvature of the stomach as far as the caudal aspect of the right lung bud. At its gastric end it communicates with the general peritoneal cavity and lies ventrolateral to the gut; more rostrally it lies directly lateral to the oesophagus. It is not, as commonly stated, a simple progressive excavation of the right side of the dorsal mesogastrium. The right pneumatoenteric recess undergoes further extension, subdivision and modification (Fig. 90.8A). From its caudal end a second process of cleft and cavity formation occurs which produces the hepatoenteric recess. This thins and expands the splanchnopleure between the liver and the stomach and proximal duodenum, and reaches the diaphragm (Figs 90.7C, 90.8B,C). The resulting, structurally bilaminar, mesenteric sheet is the lesser omentum. It is derived, cranially to caudally, from the small meso-oesophagus; the much larger ventral mesogastrium and the most caudal free border, is from the ventral mesoduodenum. As differential growth of the duodenum occurs, the biliary duct is repositioned and most of the duodenum becomes sessile. The duodenal attachment of the free border and a continuous neighbouring strip of the lesser omentum become confined to the upper border of a short segment of its superior part. The contrasting growth and positioning of its attached viscera cause the free border to change gradually from the horizontal to the vertical. It carries the bile duct, portal vein and hepatic artery, and its hepatic end is reflected around the porta hepatis. An alternative name for this part of the lesser omentum is the hepatoduodenal ligament: it forms the anterior wall of the epiploic foramen. The floor of the foramen is the initial segment of the superior part of the duodenum, its posterior wall is the peritoneum covering the immediately subhepatic part of the inferior vena cava, and its roof the peritonealized caudate process of the liver. The major part of the lesser omentum from the lesser gastric curvature passes in an approximately coronal plane to reach the floor of the increasingly deep groove for the ductus venosus on the hepatic dorsum: this part is sometimes called the hepatogastric ligament. Ligaments of the liver
The liver is precociously large during development because of its early role in haematopoiesis. Thus the liver mass projects into the abdominal cavity with equal growth on the two sides of the peritoneal cavity. The ligaments associated with the liver develop from the ventral mesogastrium - which passes from the foregut to the ventral abdominal wall down to the cranial rim of the intestinal portal - and from the reflections of peritoneum from the liver to the diaphragm. The medial portions of the germinative coelomic epithelial walls containing splanchnopleuric mesenchyme, septum transversum mesenchyme and developing liver constitute the early ventral mesogastrium (Fig. 90.5). The mesenchyme between these layers is continuous superiorly with the septum transversum mesenchyme of the diaphragm. The coelomic epithelial layers of the ventral mesogastrium almost touch anterior and posterior to the liver, and are separated by a slender lamina of mesenchyme. They form the falciform ligament and the lesser omentum respectively, and where they are in contact with the liver directly they form visceral peritoneum. When the diaphragm is formed above the liver (p. 1093), local cavities coalesce and open into the general coelomic cavity as extensions of the greater (and lesser) sacs. In this way almost all the ventrosuperior, visceral and some of the posterior aspects of the liver become peritonealized. The process of extending the greater sac continues over the right lobe and stops when the future superior and inferior layers of the coronary ligament and the right triangular ligament are defined. Those, plus a medial boundary provided by an extension of the lesser sac, enclose the 'bare area' of the liver where loose areolar tissue of septum transversum origin persists. Later in development, when the haematopoietic function of the liver declines, the left lobe becomes relatively smaller than the right, which presumably accounts for the smaller size of the left triangular ligament. Where the superior layers of the coronary and left triangular ligaments meet they continue as a (bilaminar) ventral mesentery attached to the ventrosuperior aspects of the liver. Its somewhat arched umbilicohepatic free caudal border carries the left umbilical vein. As the ventral body wall develops this falciform ligament, which initially attaches to the early cranial intestinal portal, is drawn to the diminishing cranial rim of the umbilicus. It may be considered the final ventral part of the ventral mesogastrium, although its free border has a ventral mesoduodenal origin. Its passage to the ventral body wall becomes increasingly oblique, curved and falciform (sickle-shaped) as the umbilicus becomes more defined. In the early embryo the connection between one pericardioperitoneal canal and the other was directly across the ventral surface of the cranial midgut, immediately caudal to the developing primitive ventral mesogastrium. By stage 14 the passage from one side of the falciform ligament to the other necessitates passing below the greatly enlarged liver, or the curved lower edge of the falciform ligament, or the lesser omentum. The position of the falciform ligament is of clinical interest in the neonate in diagnosing pneumoperitoneum because it is silhouetted by air on abdominal X-rays. Caval fold
The caval fold is a linear eminence, with divergent cranial and caudal ends, which passes from the upper abdominal to the lower thoracic region and protrudes from the dorsal wall of the pleuroperitoneal canal. Cranially it becomes continuous, lateromedially, with the root of the pulmonary anlage and pleural coelom, the uppermost portion of the septum transversum mesenchyme, and the retrocardiac mediastinal mesenchyme. Caudally it forms an arch with dorsal and ventral horns. The dorsal horn merges with the primitive dorsal mesentery and the mesonephric ridge and associated gonad and suprarenal (adrenal) gland. The ventral horn is confluent with the dorsal surface of the septal mesenchyme. Thus the caval fold is a zone where intestinal, mesenteric, intermediate, hepatic,
pericardial, pulmonary and mediastinal mesenchymes meet and blend. It provides a mesenchymal route for the upper abdominal, transdiaphragmatic and transpericardial parts of the inferior vena cava, and it is also prominent in the development of parts of the liver, lesser sac of peritoneum, and certain mesenteries. The left fold regresses whereas the right fold enlarges rapidly (Fig. 90.7). The pneumatoenteric recess continues to expand to the right into the substance of the caval fold. It stops near the left margin of the hepatic part of the inferior vena cava, which remains extraperitoneal and crosses the base of the now roughly triangular bare area of the liver and this new expanded line of reflexion. With closure of the pleuroperitoneal canals the rostral part of the right pneumatoenteric recess is sequestered by the diaphragm but often persists as a small serous sac in the right pulmonary ligament. The remaining caval fold mesenchyme to the left of the inferior vena cava - which forms the right wall of the upper part of the lesser sac - becomes completely invaded by embryonic hepatic tissue and is transformed into the caudate lobe of the liver. This smooth, vertically elongated mass projects into the cavity of the lesser sac: both its posterior, and much of its anterior, surfaces become peritonealized as a result of the increasing depth of the groove for the ductus venosus and the attachment of the lesser omentum to its floor. Later stages of lesser sac development
The lower (inferior) part of the lesser sac starts development at c.8-9 mm CR length. At this stage, the early pneumatoenteric and hepatoenteric recesses are well established. Progressive differential gastric growth produces an elliptical transverse sectional profile, with a right-sided lesser curvature, which corresponds to the original ventral border of the gastric tube. The lesser omental gastric part of the ventral mesogastrium remains attached to this border. The greater curvature of the stomach is a new, rapidly expanding, region: its convex profile projects mainly to the left, but also cranially and caudally (Fig. 90.7). The original dorsal border of the gastric tube now traverses the dorsal aspect of the expanding rudiment, curving along a line near the lesser curvature. The primitive dorsal mesogastrium is transiently attached to it, and blends with the thick layer of compound gastric mesenchyme clothing the posterior aspect and greater curvature of the miniature stomach. Because of its thickness, the mesenchyme projects cranially, caudally, and particularly to the left, beyond the 'new' greater curvature of the endodermal lining of the stomach. page 1264 page 1265
The processes already described in relation to the ventral mesenteries now supervene. Multiple clefts appear at various loci in the mesenchyme, and there are local mesenchyme to epithelial transitions. The groups of clefts rapidly coalesce to form transiently isolated closed spaces which soon join with each other and with the preformed upper part of the lesser sac; the newly formed epithelia join the coelomic epithelium. In sequence, the initial loci occur in the compound posterior gastric mesenchyme nearer the lesser curvature and along its zone of blending with the primitive dorsal mesogastrium; in the dorsal mesoduodenum; and independently in the caudal rim, where greater curvature mesenchyme and dorsal mesogastrium blend. As these cavities become confluent and their 'reniform' expansion follows, matches and then exceeds that of the gastric greater curvature, there are several major sequelae. The primitive dorsal mesogastrium increases in area by intrinsic growth, and, as cavitation proceeds, by incorporating substantial contributions from the dorsal lamella separated by cleavage of the posterior gastric mesenchyme. It is now called the secondary dorsal mesogastrium (Fig. 90.7C). The gastric attachment of the secondary dorsal mesogastrium changes progressively: it may be regarded as a set of somewhat spiral lines, longitudinally disposed, that move with time to the left, from near the lesser curvature, towards and finally reaching, the definitive greater curvature. The parietal mesogastrial and (cleaving) mesoduodenal attachment remains, for a time, in the dorsal midline, but subsequently undergoes
profound changes. With the confluence of the cavities that collectively form the lower part of the lesser sac, its communication with the upper part - which corresponds to the lesser gastric curvature and right and left gastropancreatic folds - becomes better defined. Ventral to the lower part of the cavity postcleavage splanchnopleure covers the posteroinferior surface of the stomach and a short proximal segment of the duodenum. This ventral wall is continued beyond the greater curvature and duodenum as the splanchnopleuric strip of visceral attachment of the secondary dorsal mesogastrium and mesoduodenum. The radial width of the strip is relatively short cranially (gastric fundus) and gradually increases along the descending left part of the greater curvature. It is longest throughout the remaining perimeter of the greater curvature as far as the duodenum: this prominent part shows continued marginal (caudoventral and lateral) growth with extended internal cavitation (its walls constitute the expanding greater omentum, Fig. 90.7E). The margins of the cavity of the inferior part of the lesser sac are limited by the reflexed edges of the ventrally placed strata derived from the secondary dorsal mesogastrium just described. These converge to form the splanchnopleuric dorsal wall, which is initially 'free' throughout except at its midline dorsal root. At roughly midgastric levels, the pancreatic rudiment grows obliquely encased in this dorsal wall; its tail ultimately reaches the left limit of the lesser sac at the level of the junction between gastric fundus and body. Greater omentum
The greater omentum continues to grow both laterally, and particularly caudoventrally. It covers and is closely applied to the transverse mesocolon, transverse colon and inframesocolic and infracolic coils of small intestine (Fig. 90.6D-G). At this stage the quadrilaminar nature of the dependent part of the greater omentum is most easily appreciated. 'Simple' mesenteries are bilaminar: they possess two mesothelial surfaces derived from splanchnopleuric coelomic epithelium, which enclose a connective tissue core derived from splanchnopleuric mesenchyme. In the greater omentum, the gastric serosa covering its posteroinferior surface (single mesothelium) and the anterosuperior serosa (single mesothelium) converge to meet at the greater curvature and initial segment of the duodenum. The resulting bilaminar mesentery continues caudoventrally as the 'descending' stratum of the omentum. This, on reaching the omental margins, is reflexed and now passes craniodorsally to its parietal root as the 'ascending' posterior bilaminar stratum. The two bilaminar strata are initially in fairly close contact caudally, but separated by a fine, fluid-containing, cleft-like extension of the lower part of the lesser sac. The posterior mesothelium of the posterior stratum makes equally close contact with the anterosuperior surface of the transverse colon, starting at the taenia omentalis, and with its transverse mesocolon. Maturation of the lesser sac
At this stage, and subsequently, it is convenient to designate the lower part of the lesser sac as consisting of three subregions: retrogastric, perigastric and greater omental (Fig. 90.7E). The names are self-explanatory but their confines are all modified by various factors. Two phenomena are particularly prominent: gastric 'descent' relative to the liver, and fusion of peritoneal layers with altered lines of reflexion, adhesion of surfaces and loss of parts of cavities.
Figure 90.9 Fusion of the proximal part of the dorsal mesogastrium with the peritoneum on the posterior abdominal wall. Note also the conversion of the dorsal mesogastrium into the gastrosplenic and lienorenal ligaments. 1, Transverse section of an embryo in which the dorsal mesogastrium is still at the stage shown in Fig. 90.8A. 2 and 3, Transverse sections of older embryos made at the same level, simplified by retaining the shape and size of the stomach and spleen.
After the third month hepatic growth diminishes, particularly of the left lobe, and the whole organ recedes into the upper abdomen. Meanwhile the stomach elongates and some descent occurs, despite its relatively fixed cranial and caudal ends. This produces the angular flexure of the stomach which persists postnatally. The concavity of the lesser curvature is now directed more precisely to the right, the lesser omentum is more exactly coronal and its free border vertical. Ventral to the liver the free border of the falciform ligament passes steeply craniodorsally from umbilicus to liver (see disposition in neonate in Figs 65.5, 11.4 and 11.5). The mesenchymal dorsal wall of the lower part of the lesser sac, which is crossed obliquely by the growing pancreas, has hitherto remained free and retained its original dorsal midline root. Substantial areas now fuse with adjacent peritonealized surfaces of retroperitoneal viscera, the parietes, or another mesenteric sheet or fold. Where sheets fuse there is a variable loss of apposed mesothelia and some continuity of their mesenchymal cores, but they remain surgically separable and no vascular anastomosis develops across the interzone. Above the pancreas the posterior secondary dorsomesogastrial wall of the sac becomes closely applied to the peritoneum covering the posterior abdominal wall and its sessile organs, the diaphragm, much of the left suprarenal gland, the ventromedial part of the upper pole of the left kidney, the initial part of the abdominal aorta, the coeliac trunk and its branches, and other vessels, nerves, and lymphatics (Figs 90.7D, E, 11.5). Their peritoneal surfaces fuse. However, albeit with some tissue loss, a single mesothelium remains covering these structures, intercalated as a new secondary dorsal wall for this part of the lesser sac (Fig. 90.9). The pancreas grows from the duodenal loop, penetrating the substance of the dorsal mesoduodenum and secondary dorsal mesogastrium, their mesenchymes and mesothelia initially clothing its whole surface, except where there exist peritoneal lines of reflexion. Its posterior peritoneum becomes closely applied to that covering all the posterior abdominal wall structures it crosses (the inferior vena cava, abdominal aorta, splenic vein, superior mesenteric vessels, inferior mesenteric vein, portal vein, left renal vessels, the caudal pole of the left suprarenal, a broad ventral band on the left kidney and various muscles (Fig. 11.4). The intervening peritoneal mesothelia fuse and atrophy, and the mesenchymal cores form fascial sheaths and septa. The pancreas is now sessile.
The peritoneum covering the upper left part of its head, neck and the anterosuperior part of its body forms the central part of the dorsal wall of the lesser sac. The pancreatic tail remains peritonealized by a persisting part of the secondary dorsal mesogastrium as it curves from the ventral aspect of the left kidney towards the hilum of the spleen. The infracolic parts of the pancreas are covered with greater sac peritoneum. In the greater omental subregion of the lower part of the lesser sac two contrasting forms of mesenteric adhesion occur. The posterior 'returning' bilaminar stratum of the omentum undergoes partial fusion with the peritoneum of the transverse colon at the taenia omentalis and with its mesocolon. The layers remain surgically separable: no anastomosis occurs between omental and colic vessels. UPDATE Date Added: 22 August 2006 Abstract: Three-dimensional computed tomography pancreatography of an annular pancreas to elucidate embryogenesis Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? db=pubmed&cmd=Retrieve&dopt=AbstractPlus&list_uids=16670627&query_hl=7&itool=pubmed_docsum Three-dimensional computed tomography pancreatography of an annular pancreas to elucidate embryogenesis. Ueki T, Yao T, Beppu T, et al: Pancreas 32:426-429, 2006. page 1265 page 1266
The original dorsal midline attachment to the parietes of the foregut dorsal mesentery is profoundly altered during the development of the lesser sac and its associated viscera. However, despite the extensive areas of fusion, virtually the whole of the gastric greater curvature (other than a small suboesophageal area) and its topographical continuation (the inferior border of the first 2-3 cm of the duodenum), retain true mesenteric derivatives of the secondary dorsal mesogastrium and its continuation, the dorsal mesoduodenum. Although regional names are used to assist identification and description, it must be emphasized that they are all merely subregions of one continuous sheet. The upper (oesophagophrenic) part of the lesser omentum arches across the diaphragm. As this bilaminar mesentery approaches the oesophageal hiatus its laminae diverge, skirting the margins of the hiatus. They then descend for a limited distance and with variable inclination, to enclose reciprocally shaped areas on the dorsum of the gastric fundus and diaphragm. The area may be roughly triangular to quadrangular; it contains areolar tissue and constitutes the bare area of the stomach or, when large, the left extraperitoneal space. Its right lower angle is the base of the left gastropancreatic fold, and its left lower angle reconstitutes the bilaminar mesentery. The root of the latter arches downwards and to the left across the diaphragm and suprarenal gland and gives the gastrophrenic ligament to the gastric fundus. It continues to arch across the ventral surface of the upper part of the left kidney, and its layers part to receive the pancreatic tail: they initially extend to the hilum of the spleen as the splenorenal (lienorenal) ligament (Figs 90.6, 90.9). The left half of this bilaminar 'ligament' provides an almost complete peritoneal tunic for the spleen. It then reunites with its fellow at the opposite rim of the splenic hilum, and continues to the next part of the gastric greater curvature as the gastrosplenic ligament. The remaining part (perhaps twothirds) of the gastric greater curvature and its short duodenal extension provide attachment for the anterior, 'descending', bilaminar stratum of the greater omentum. Its returning, posterior, bilaminar stratum continues to its parietal root (which extends from the inferior limit assigned to the splenorenal ligament), and curves caudally and to the right along the anterior border of the body of the pancreas, immediately cranial to the line of attachment of the transverse mesocolon. Crossing the neck of the pancreas, the same curve is followed for a few centimetres on to its head; the omental root then sharply recurves cranially and to the left, to reach the inferior border of the duodenum. Thus it reaches that part of the lesser sac provided by cleavage of the dorsal mesoduodenum from the greater sac. It enters the epiploic foramen, traverses the epiploic canal between the caudate hepatic process and proximal duodenum, then crosses the right gastropancreatic fold, and descends behind the proximal duodenum to enter the
right marginal strip enclosed by the greater omentum. The definitive origins of the peritoneum from the posterior abdominal wall are shown in the adult in Figs 69.1, 69.4 and 69.5. Peritoneum associated with the mid- and hindgut
It is convenient to consider the mesenteries of the small and large intestine after rotation and the principal growth patterns have been achieved and the developing pancreas is becoming retroperitoneal. Small intestine
Most of the duodenal loop encircles the head of the pancreas and is retroperitoneal. The peritoneum principally covers its ventral and convex aspects. Areas not covered are a short initial segment of the superior (first) part, which is more completely peritonealized because it gives attachment to the right margins of the greater and lesser omenta; where the transverse colon is closely apposed to the descending (second) part, or where the latter is crossed by the root of the transverse mesocolon; where the mesentery crosses the transverse (third) part, and descends across the ascending (fourth) part from its upper extremity at the duodenojejunal flexure. These regions are illustrated in the adult in Fig. 69.1. In addition, one or more of up to six different duodenal recesses may develop. Their variations in shape and size, their intestinal, mesenteric and vascular relations, and, when adequately recorded, the frequencies and disposition of their orifices, are given on page 1137. From a mesenteric standpoint, the succeeding small intestine, from the duodenojejunal flexure to the ileocaecal junction, undergoes less modification of its embryonic form than other gut regions. Its early dorsal mesentery is a continuous, single (but structurally bilaminar) sheet, with a midline parietal attachment (line of reflexion, or 'root'). The attachment of the root becomes an oblique narrow band from the left aspect of the second lumbar vertebra to the cranial aspect of the right sacroiliac joint. Ascending colon
The caecum and vermiform appendix arise as a diverticulum from the antimesenteric border of the caudal limb of the midgut loop, consequently the caecum does not possess a primitive mesocaecum. These regions of the gut undergo long periods of growth, often asymmetrical, and their final positions, dimensions and general topography show considerable variation. The vermiform appendix is almost wholly clothed with visceral peritoneum, derived from the diverging layers of its rather diminutive mesoappendix. The latter should perhaps be regarded as a direct derivative of the primitive dorsal mesentery, and perhaps a similar status for the vascular fold of the caecum should be considered. The colonic gut retains its primitive dorsal mesentery, the mesocolon, until the differential growth, rotation and circumabdominal displacement of this part of the gut tube nears completion. Its original root is still vertical in the dorsal midline, although the mesocolon diverges from it widely as an incomplete, flattened pyramid, to reach its colonic border at the future taenia mesocolica. During the fourth and fifth months substantial areas of the primitive mesocolon adhere to, then fuse with, the parietal peritoneum. In this way, some colonic segments become sessile while others have a shorter mesocolon with an often profoundly altered parietal line of attachment. The mesocolon of the transverse and sigmoid segments normally persists, while the ascending colon, right (hepatic) flexure and descending colon become sessile: the ascending or descending, or both, colonic segments may also retain a mesocolon which varies from a localized 'fold' to a complete mesocolon. When sessile, the ventral, medial and lateral aspects of the ascending or descending colon are clothed with peritoneum, and the protrusion of the viscus produces medial and lateral peritoneal paracolic gutters on each side. This form of apposition to underlying structures proceeds from the ascending colon to include the right colic (hepatic) flexure, and thence continues anteroinferiorly to the left, so involving the right-sided initial segment of the
transverse colon. Transverse colon
The right extremity of the transverse colon is sessile, and is separated by fibroareolar tissue from the anterior aspect of the descending (second) part of the duodenum and the corresponding aspect of most of the head of the pancreas. The remainder of the transverse colon, up to and including the left (splenic) colic flexure, is almost completely peritonealized by the diverging layers of the transverse mesocolon. The root of the latter reaches the neck and whole extent of the anterior border of the body of the pancreas. The long axis of the definitive pancreas lies obliquely. The splenic colonic flexure is considerably more rostral than the hepatic flexure and consequently the root of the mesocolon curves obliquely upwards as it crosses the upper abdomen from right to left. As it expands, the posteroinferior wall of the greater omental part of the lesser sac gradually covers, and becomes closely applied to, the transverse mesocolon and its contained colon, finally projecting beyond the latter. Craniocaudal adherence now occurs between the omental wall and the pericolonic and mesocolonic layers. Descending colon
The left colic flexure receives much of its peritoneal covering from the left extremity of the transverse mesocolon. It is also often connected to the parietal peritoneum of the diaphragm over the tenth and eleventh ribs by a phrenicocolic ligament. The latter sometimes blends with a presplenic fold that radiates from the gastrosplenic ligament. The descending colon becomes sessile. The process of fusion and obliteration of both ascending and descending mesocolons starts laterally and progresses medially. Sigmoid colon
The sigmoid colon is most variable in its length and disposition. It retains its dorsal mesocolon, but the initial midline dorsal attachment of its root is considerably modified in its definitive state. Rectum page 1266 page 1267
The rectum continues from the ventral aspect of the third sacral vertebra to its anorectal (perineal) flexure anteroinferior to the tip of the coccyx: the distance changes with age. All aspects of the rectum are encased by mesenchyme, and the early dorsally placed mass is named, by some authorities, the dorsal mesorectum. However, the latter does not form a true mesentery: with progressive skeletal development it is reduced to a woven fibroareolar sheet which displays patterned variations in thickness and fibre orientation. The sheet is closely applied to the ventral concavity of the sacrum and coccyx, and encloses numerous fibromuscular and neurovascular elements. The rectum therefore becomes sessile, and visceral peritoneum is restricted to its lateral and ventral surfaces (Fig. 109.4). With the disappearance of the postanal gut by the end of the fifth week, the ventrolateral peritoneum reaches the superior surface of the pelvic floor musculature: this condition persists until late in the fourth month. In the male the ventral rectal peritoneum is reflected over the posterior surface of the prostate, bladder trigone and associated structures. In the female the ventral peritoneum initially receives a reflection which covers almost the whole posterior aspect of the vagina, and is continued over the uterus. Subsequently, the closely apposed walls of these deep peritoneal pouches fuse over much of their caudal extent, their mesothelia are lost, and the viscera are separated by an intervening, bilaminar (surgically separable), fibrous stratum. In the male this becomes the rectovesical fascia and posterior wall of the prostatic sheath. In the female it becomes the rectovaginal septum between the lower part of the vagina and the rectum (Fig. 11.5). The proximal third of the rectum is covered by peritoneum ventrolaterally: the lateral extensions of this tunic are triangular and deep proximally, but taper to
an acute angle by the middle third of the rectum. The middle third of the rectum is covered by peritoneum only on its ventral surface, where it forms the posterior wall of the shallower rectovesical or rectovagino-uterine pouch. The remaining rectum and anal canal are extraperitoneal.
NEONATAL PERITONEAL CAVITY The fully formed peritoneal cavity, although complex topographically, remains a single cavity with numerous intercommunicating regions, pouches and recesses (Fig. 11.5). The only small peritoneal sacs to separate completely from the main cavity are the infracardiac bursa (Fig. 90.7D) and the tunica vaginalis testis (Fig. 109.17). In fetal life the greater omental cavity extends to the internal aspect of the lateral and caudal edges of the omentum. Postnatally a slow but progressive fusion of the internal surfaces occurs with obliteration of the most dependent part of the cavity: this proceeds rostrally and, when mature, the cavity does not usually extend appreciably beyond the transverse colon. Transverse mesocolon-greater omentum fusion begins early while the umbilical hernia of the midgut has not returned. It starts between the right margin of the early greater omentum and near the root of the presumptive mesocolon, and later spreads to the left. In the neonate the peritoneal cavity is ovoid (Figs 11.4, 65.5). It is fairly shallow from anterior to posterior because the bilateral posterior extensions on each side of the vertebral column, which are prominent in the adult, are not present (Fig. 11.5). Two factors lead to the protuberance of the anterior abdominal wall in the neonate and infant. The diaphragm is flatter in the newborn, which produces a caudal displacement of the viscera. The pelvic cavity is very small in the neonate, which means that organs which are normally pelvic in the adult, i.e. urinary bladder, ovaries and uterus, all extend superiorly into the abdomen (Fig. 11.5). The pelvic cavity is joined to the abdominal cavity at less of an acute angle in the neonate because there is no lumbar vertebral curve and only a slight sacral curve. The peritoneal attachments are similar to the adult. However, the greater omentum is relatively small: its constituent layers of peritoneum may not be completely fused, and it does not extend much below the level of the umbilicus (Fig. 11.5). Generally the length of the mesentery of the small intestine and of the transverse and sigmoid mesocolons are longer than in the adult, whereas the area of attachment of the ascending and descending colons is relatively smaller. The peritoneal mesenteries and omenta contain little fat. UPDATE Date Added: 28 June 2005 Abstract: Congenital segmental dilatation of the colon with anorectal malformation. Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=15300559&query_hl=11 Congenital segmental dilatation of the colon with anorectal malformation.
© 2008 Elsevier
SPLEEN The spleen appears about the sixth week as a localized thickening of the coelomic epithelium of the dorsal mesogastrium near its cranial end (Figs 90.6, 90.7, 90.9). The proliferating cells invade the underlying angiogenetic mesenchyme, which becomes condensed and vascularized. The process occurs simultaneously in several adjoining areas which soon fuse to form a lobulated spleen of dual origin (from coelomic epithelium and from mesenchyme of the dorsal mesogastrium). The enlarging spleen projects to the left, so that its surfaces are covered by the peritoneum of the mesogastrium on its left aspect, which forms a boundary of the general extrabursal (greater) sac. When fusion occurs between the dorsal wall of the lesser sac and the dorsal parietal peritoneum, it does not extend to the left as far as the spleen, which remains connected to the dorsal abdominal wall by a short splenorenal ligament. Its original connection with the stomach persists as the gastrosplenic ligament. The earlier lobulated character of the spleen disappears, but is indicated by the presence of notches on its upper border in the adult. The histogenesis of the spleen has attracted relatively little attention. The vascular reticulum is well developed at 8-9 weeks, and contains immature reticulocytes and numerous closely spaced thin-walled vascular loops. Differentiation of blood cells, macrophages, and of arteries, veins, capillaries and sinusoids has occurred by the eleventh to twelfth week. Initially the capsule consists of cuboidal cells bearing cilia and microvilla. The spleen displays various developmental anomalies, including complete agenesis, multiple spleens or polysplenia, isolated small additional spleniculi and persistent lobulation. Asplenia and polysplenia are associated with other anomalies especially those involving the cardiac and pulmonary systems. Accessory spleens are very common in neonates, located in the greater omentum. At birth the spleen weighs, on average, 13 g (Fig. 11.4). It doubles its weight in the first postnatal year and triples it by the end of the third year.
© 2008 Elsevier
SUPRARENAL GLANDS The suprarenal (adrenal) cortex is formed during the second month by a proliferation of the coelomic epithelium. Cells pass into the underlying mesenchyme between the root of the dorsal mesogastrium and the mesonephros (Fig. 14.11). The proliferating tissue, which extends from the level of the sixth to the twelfth thoracic segments, is soon disorganized dorsomedially by invasion of neural crest cells from somite levels 18-24, which form the medulla, and also by the development of venous sinusoids. The latter are joined by capillaries, which arise from adjacent mesonephric arteries and penetrate the cortex in a radial manner. When proliferation of the coelomic epithelium stops the cortex is enveloped ventrally, and later dorsally, by a mesenchymal capsule which is derived from the mesonephros. The subcapsular nests of cortical cells are the rudiment of the zona glomerulosa: they proliferate cords of cells which pass deeply between the capillaries and sinusoids. The cells in these cords degenerate in an erratic fashion as they pass towards the medulla, becoming granular, eosinophilic and ultimately autolysed. These cords of cells constitute the fetal cortex, which undergoes rapid involution during the first two years after birth. The fascicular and reticular zones of the adult cortex are proliferated from the glomerular zone after birth. The most common abnormality of suprarenal gland development is congential adrenal hyperplasia, which occurs in 1:5000-1:15000 births. This condition is caused by a group of autosomal recessive disorders in which there are deficiencies in enzymes required for the synthesis of cortisol. In 90% of cases the cause is deficiency of the enzyme 21-hydroxylase, producing an accumulation of 17-hydroxyprogesterone, which is converted to androgens. The levels of androgens increase by several hundred times, causing female embryos and fetuses to undergo external genital masculinization ranging from clitoral hypertrophy to formation of a phallus and scrotum: masculinization of the brain has also been suggested. In male embryos the levels do not cause any changes in external genitalia. Signs of androgen excess may appear in childhood with precocious masculinization and accelerated growth (Lewis, Yaron & Evans 1999).
SUPRARENAL GLANDS IN THE NEONATE The suprarenal glands are relatively very large at birth (Figs 11.4, 109.8) and constitute 0.2% of the entire body weight, compared with 0.01% in the adult. The left gland is heavier and larger than the right, as it is in the adult. At term each gland weighs c.4 g; the average weight of the two glands is 9 g (average in the adult is 7-12 g). The glands involute rapidly in the neonatal period when each gland loses 25% of its mass; the average weight of both glands is 5 g by the end of the second week, and 4 g by 3 months. Birth weight is not regained until puberty. The cortex of the suprarenal gland is thicker than in the adult and the medulla of the gland is small. Early studies on fetal suprarenal glands described extensive degeneration and necrosis of fetal zone cells; however, it is believed that these studies showed disease processes rather than the normal involution of the gland. With normal involution the fetal zone cells of the postnatal gland
become smaller and they assume the appearance and organization typical of zona fasciculata. page 1267 page 1268
REFERENCES Butzner JD, Befus AD 1989 Interactions among intraepithelial leucocytes and other epithelial cells in intestinal development and function. In: Lebenthal E (ed) Human Gastrointestinal Development, Chapter 37. New York: Raven Press. Collins P 2002 Embryology of the pancreas. In: Howard ER, Stringer MD, Colombani PM (eds) Surgery of the Liver, Bile Ducts and Pancreas in Children, Part 8. London: Arnold: 479-92. Covers pancreatic morphogenesis, the timescale of development, the origin of pancreatic cell lines and factors that regulate pancreatic development. Collins P 2002 Embryology of the liver and bile ducts. In: Howard ER, Stringer MD, Colombani PM (eds) Surgery of the Liver, Bile Ducts and Pancreas in Children, Part 3. London: Arnold: 91-102. Covers morphogenesis of the liver and early hepatic circulation, the origin of hepatic cell lines and the development of the extra- and intrahepatic biliary systems. Gershon MD 1987 Phenotypic expression by neural crest-derived precursors of enteric neurons and glia. In: Madreson PFA (ed) Developmental and Evolutionary Aspects of the Neural Crest. New York: John Wiley. Describes a mouse model of Hirschsprung's disease. Howard ER 2002 Biliary atresia: etiology, management and complications. In: Howard ER, Stringer MD, Colombani PM (eds) Surgery of the Liver, Bile Ducts and Pancreas in Children, Part 3. London: Arnold: 103-32. Reviews the aetiology and clinical presentation of biliary atresia, including the congenital, infective and anatomical factors that are related to the condition. Lebenthal E 1989 Concepts in gastrointestinal development. In: Lebenthal E (ed) Human Gastrointestinal Development, Chapter 1. New York: Raven Press. This chapter is the first in a volume dedicated to the development of structure and function of the gut, liver and pancreas. Includes the development of the immunological surveillance mechanisms and gastrointestinal flora. Lewis P, Yaron Y, Evans MI 1999 Fetal endocrine disorders. In: Rodeck CH, Whittle MJ (eds) Fetal Medicine: Basic Sciences and Clinical Practice, Chapter 62. Edinburgh: Churchill Livingstone: 829-34. Reviews the diagnosis and treatment of common fetal endocrine disorders, particularly of the suprarenal and thyroid glands. Streeter GL 1942 Developmental horizons in human embryos. Descriptions of age group XI, 13 to 20 somites, and age group XII, 21 to 29 somites. Contrib Embryol Carnegie Inst Washington 30: 211-45. Whittle MJ 1999 Gastrointestinal abnormalities. In: Rodeck CH, Whittle MJ (eds) Fetal Medicine: Basic Sciences and Clinical Practice, Chapter 54. Edinburgh: Churchill Livingstone: 703-714. Reviews the diagnosis and treatment of abnormalities of the gastrointestinal tract.
© 2008 Elsevier
91 KIDNEY AND URETER Kidney The kidneys excrete the end products of metabolism and excess water. Both of these actions are essential to the control of concentrations of various substances in the body fluids, e.g. maintaining electrolyte and water balance approximately constant in the tissue fluids. The kidneys also have endocrine functions producing and releasing erythropoietin which affects red blood cell formation, renin which influences blood pressure, 1,25-hydroxycholecalciferol, which is involved in the control of calcium metabolism and is a derivative of vitamin D, and perhaps modifies the action of the parathyroid hormone, and various other soluble factors with metabolic actions.
Figure 91.1 Dissection to show the relations of structures on the posterior abdominal wall (female subject).
The kidneys in the fresh state are reddish-brown. They are situated posteriorly behind the peritoneum on each side of the vertebral column and are surrounded by adipose tissue. Superiorly they are level with the upper border of the twelfth thoracic vertebra, inferiorly with the third lumbar vertebra. The right is usually slightly inferior to the left, probably reflecting its relationship to the liver. The left is a little longer and narrower than the right and lies nearer the median plane (Fig. 91.1). The long axis of each kidney is directed inferolaterally and the transverse axis posteromedially. Hence the anterior and posterior aspects usually described are in fact anterolateral and posteromedial. The transpyloric plane passes through the superior part of the right renal hilum and the inferior part of the left (p. 1099). UPDATE Date Added: 19 April 2005 Abstract: Ureteroscopy: anatomic and physiologic considerations. Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=15040397 Ureteroscopy: anatomic and physiologic considerations.
Each kidney is c.11 cm in length; 6 cm in breadth and 3 cm in anteroposterior dimension. The left kidney may be 1.5 cm longer than the right; it is rare for the right kidney to be more than 1 cm longer than the left. The average weight is c.150 g in men and 135 g in women. In thin individuals with a lax abdominal wall the lower pole may just be felt in full inspiration by bimanual lumbar examination, but this is unusual. In the fetus and newborn, the kidney has c.12 lobules (Fig. 91.2). These are fused in adults to present a smooth surface although traces of lobulation may remain. UPDATE Date Added: 12 April 2005 Abstract: Human renal dysplasia: understanding evolving concepts. Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=15034102 Human renal dysplasia: understanding evolving concepts. Absent and ectopic kidneys
page 1269 page 1270
Figure 91.2 The kidneys and suprarenal glands of a newborn infant: anterior aspect. Note the lobulation of the renal surface and relative size of the organs.
A single absent kidney is seen in c.1 in 1200 individuals and results from failure of metanephric blastema to join with a ureteric bud on the affected side. It has no clinical sequelae but may frequently be associated with absence of the ipsilateral vas deferens and/or epididymis and may be associated with other congenital anomalies including imperforate anus, cardiac valvular anomalies and oesophageal atresia. A single kidney often shows compensatory hypertrophy. The life expectancy of individuals with a single kidney is the same as those with two kidneys. Failure of the kidney to ascend in utero to the correct position in the renal fossa results in renal ectopia. Most commonly the kidney is found in the pelvis: this occurs in c.1 in 2500 live births. Kidneys so placed often have associated malrotation anomalies, and may have marked fetal lobulation. Pelvic kidneys frequently become hydronephrotic as a result of an anterior placed ureter and an anomalous arterial supply. An associated pelviureteric junction obstruction is often present. Very rarely and despite the normal location of the ureteric orifices within the bladder, the two renal masses may be on the same side. This is termed crossed renal ectopia and usually the two renal masses are fused in such circumstances. A solitary crossed renal ectopia may be associated with skeletal and other genitourinary anomalies. Horseshoe kidney Horseshoe kidneys are found in 1 in 400 individuals. A transverse bridge of renal tissue, the isthmus, which usually but not invariably contains functioning renal substance, connects the two renal masses. The isthmus lies between the inferior poles, most commonly anterior to the great vessels. The ureters curve anterior to
the connection and may have a high insertion into the renal pelvis. The blood supply to horseshoe kidneys is variable. One vessel to each moiety is seen in 30% of horseshoe kidneys. Multiple anomalous vessels are common and the isthmus may be supplied by a vessel directly from the aorta or from branches of the inferior mesenteric, common iliac or external iliac arteries. In view of this variable arterial anatomy, angiography is very helpful when planning renal surgery on horseshoe kidneys. Horseshoe kidneys can have an associated congenital pelviureteric junction obstruction in up to 30%. Anomalous vessels crossing the ureter and the abnormal course of the ureter as it passes over renal substance may also cause obstruction. Horseshoe kidneys have an increased incidence of stone disease, probably as a consequence of areas of inefficient drainage.
© 2008 Elsevier
PERIRENAL FASCIA (Figs 91.3, (91.4) The perirenal fascia is a dense, elastic connective tissue sheath which envelops each kidney and suprarenal gland together with a layer of surrounding perirenal fat. The kidney and its vessels are embedded in perirenal fat, which is thickest at the renal borders and prolonged at the hilum into the renal sinus.
Figure 91.3 Sagittal section through the posterior abdominal wall showing the relations of the renal fascia of the right kidney.
Figure 91.4 Transverse section, showing the relations of the renal fascia.
The perirenal fascia was originally described as being made up of two separate entities, the posterior fascia of Zuckerkandl and the anterior fascia of Gerota which fused laterally to form the lateral conal fascia. According to this view, the lateral conal fascia continued anterolaterally behind the colon to blend with the parietal peritoneum. page 1270 page 1271
However, work by Mitchell (1950) showed that the perirenal fascia is not made up of distinct fused fasciae, but is in fact a single multilaminated structure which is fused posteriomedially with the muscular fasciae of psoas major and quadratus lumborum. It then extends anterolaterally behind the kidney as a bilaminated sheet, which at a variable point divides into a thin lamina which passes around the front of the kidney as the anterior perirenal fascia, and a thicker posterior lamina which continues anterolaterally as the lateral conal fascia (and fuses with the parietal peritoneum. Classically, the anterior perirenal fascia was thought to blend into the dense mass of connective tissue surrounding the great vessels in the root of the mesentery behind the duodenum and pancreas, thereby preventing communication between perirenal spaces across the midline. However inspection of CT and anatomical sections of cadavers following the injection of small volumes of contrast and coloured latex respectively into the perirenal space revealed that fluid could
extend across the midline at the third to fifth lumbar levels through a narrow channel measuring 2-10 mm in AP dimension. In the midline the anterior and posterior renal fasciae fuse superiorly and are attached to the crus of their respective hemidiaphragms. Inferiorly the fasciae separate for a variable craniocaudal distance along most of the length of each kidney. The posterior perirenal fascia fuses with the muscular fascia of psoas major whilst the anterior perirenal fascia extends across the midline in front of the great vessels and so communication between the two sides is permitted. Below this level the two fasciae once again merge and are attached to the great vessels or iliac vessels. The containment of fluid to one side of the perirenal space that is observed in over two thirds of clinical cases is attributed to the presence of fibrous septae. Above the suprarenal glands the anterior and posterior perirenal fasciae were previously said to fuse with each other and to the diaphragmatic fascia. This description of a closed superior cone is not universally accepted. Cadaveric experiments have shown the superior aspect of the perirenal space to be open and in continuity with the bare area of the liver on the right and the subphrenic extraperitoneal space on the left. The posterior fascial layer blends bilaterally with the fascia of psoas major and quadratus lumborum as well as the inferior phrenic fascia. The anterior fascial layer on the right blends with the right inferior coronary ligament at the level of the upper pole of the kidney and bare area of the liver. On the left the anterior layer fuses with the gastrosplenic ligament at the level of the suprarenal gland. There is some debate concerning the inferior fusion of the perirenal fascia. Many investigators believe that inferiorly the anterior and posterior leaves of the perirenal fascial fuse to produce an inverted cone which is open to the pelvis at its apex. Laterally the anterior and posterior leaves fuse with the iliac fascia, and medially they fuse with the periureteric connective tissue. The inferior apex of the cone is open anatomically towards the iliac fossa but rapidly becomes sealed in inflammatory disease. An alternative view is based on the dissection of recently deceased cadavers after injections of coloured latex into the perirenal space: these have shown that the anterior and posterior perirenal fasciae merge to form a single multilaminar fascia which contains the ureter in the iliac fossa. Anteriorly this common fascia is loosely connected to the parietal peritoneum, and so denies free communication between the perirenal space and the pelvis, and also denies communication between the perirenal and pararenal spaces. A simple nephrectomy for benign disease removes the kidney from within perirenal fascia; a radical nephrectomy for cancer removes the entire contents of the perirenal space including the perirenal fascia, in order to give adequate clearance around the tumour.
© 2008 Elsevier
RENAL RELATIONS (Fig. 91.5)
Figure 91.5 Multislice CT scan of the kidneys. A, Coronal oblique reformat showing both kidneys and the suprarenal glands. B, Sagittal oblique of the right kidney lying posterior to the right lobe of the liver, duodenum and right colic flexure. C, Sagittal oblique of the left kidney lying posterior to the stomach, pancreas and the splenic vessels.
The superior poles of both kidneys are thick and round and each is related to its suprarenal gland. The inferior poles are thinner and extend to within 2.5 cm of the iliac crests. The lateral borders are convex. The left kidney is covered superiorly by peritoneum, which separates it from the spleen, and below this is in contact with the descending colon (Fig. 91.6A). The peritoneum of the greater sac separates the lateral border of the right kidney from the right lobe of the liver. The medial borders are convex adjacent to the poles, concave between them and slope inferolaterally. In each a deep vertical fissure opens anteromedially as the hilum, which is bounded by anterior and posterior lips and contains the renal vessels and nerves and the renal pelvis. The relative positions of the main hilar structures are the renal vein anterior, the renal artery intermediate and the pelvis of the kidney posterior. Usually an arterial branch from the main renal artery runs over the superior margin of the renal pelvis to enter the hilum on the posterior aspect of the pelvis, and a renal venous tributary often leaves the hilum in the same plane. Above the hilum the medial border is related to the suprarenal gland and below to the origin of the ureter. The convex anterior surface of the kidney actually faces anterolaterally and its relations differ on the right and left. Likewise the posterior surface of the kidneys in reality faces posteromedially. Its relations are similar on both sides of the body.
ANTEROLATERAL SURFACE OF RIGHT KIDNEY (Fig. 91.6A) page 1271 page 1272
Figure 91.6 A, The anterior surfaces of the kidneys, showing the areas related to neighbouring viscera. Areas coloured pale blue are separated from adjacent viscera by the peritoneum. B, The posterior surfaces of the kidneys, showing the areas of relation to the posterior abdominal wall.
A small area of the superior pole is in contact with the right suprarenal gland, indeed the suprarenal gland may overlap the upper part of the medial border of the superior pole. A large area below this (about three-quarters of the anterior surface) is immediately related to the renal impression on the right lobe of the liver. A narrow medial area is related to the descending part of the duodenum. Inferiorly the anterior surface is in contact laterally with the right colic flexure and medially with part of the small intestine. The areas related to the small intestine and in contact with the liver are covered by peritoneum which overlies the renal fascia, whereas the suprarenal, duodenal and colic areas are devoid of peritoneum.
ANTEROLATERAL SURFACE OF LEFT KIDNEY (Fig. 91.6A) A small medial area of the superior pole is related to the left suprarenal gland. Approximately the upper two-thirds of the lateral half of the anterior surface is related to the spleen. A central quadrilateral area lies in contact with the pancreas and the splenic vessels. Above this a small variable triangular region, between the suprarenal and splenic areas, is in contact with the stomach. Below the pancreatic and splenic areas, a narrow lateral strip which extends to the lateral border of the kidney is related to the left colic flexure and the beginning of the descending colon. An extensive medial area is related to loops of jejunum. The gastric area is covered with the peritoneum of the lesser sac (omental bursa) and the splenic and jejunal areas are covered by the peritoneum of the greater sac. Behind the peritoneum covering the jejunal area, branches of the left colic vessels are related to the kidney. The suprarenal, pancreatic and colic areas are devoid of peritoneum.
POSTEROMEDIAL SURFACE OF BOTH KIDNEYS (Figs 91.6B, 91.7)
The posteromedial surface of the kidneys is embedded in fat and devoid of peritoneum (Figs 91.3, 91.4). It is anterior to the diaphragm, the medial and lateral arcuate ligaments, psoas major, quadratus lumborum and the aponeurotic tendon of transversus abdominis, the subcostal vessels and subcostal, iliohypogastric, and ilioinguinal nerves. The upper pole of the right kidney is level with the twelfth rib, and that of the left with the eleventh and twelfth ribs. The diaphragm separates the kidney from the pleura, which descends to form the costodiaphragmatic recess. Sometimes its muscle is defective or absent in a triangle immediately above the lateral arcuate ligament, and this allows perirenal adipose tissue to contact the diaphragmatic pleura.
page 1272 page 1273
Figure 91.7 The right kidney (posterior exposure). The blue area represents the pleura, the broken red line the upper part of the kidney. The subcostal nerve has been displaced downwards. Parts of the diaphragm and quadratus lumborum have been resected.
Figure 91.8 Longitudinal section through a kidney to show the normal macroscopic appearance: note the pelvis of the ureter and its division into calyces. The pelvis and major calyces have not been opened.
© 2008 Elsevier
GENERAL RENAL STRUCTURE (Fig. 91.8) The postnatal kidney has a thin capsule, easily removed, composed of collagenrich tissue with some elastic and smooth muscle fibres. In renal disease it may become adherent. The kidney itself can be divided into an internal medulla and external cortex. The renal medulla consists of pale, striated, conical renal pyramids, their bases peripheral, their apices converging to the renal sinus. At the renal sinus they project into calyces as papillae. The renal cortex is subcapsular, arching over the bases of the pyramids and extending between them towards the renal sinus as renal columns. The peripheral regions are cortical arches and are traversed by radial, lighter-coloured medullary rays, separated by darker tissue, the convoluted part. The rays taper towards the renal capsule and are peripheral prolongations from the bases of renal pyramids. The cortex is histologically divisible into outer and inner zones; the inner is demarcated from the medulla by tangential blood vessels (arcuate arteries and veins), which lie at the junction of the two, but a thin layer of cortical tissue (subcortex) appears on the medullary side of this zone. The cortex close to the medulla is sometimes termed juxtamedullary.
© 2008 Elsevier
RENAL CALYCES AND PELVIS (Fig. 91.9)
page 1273 page 1274
Figure 91.9 A left retrograde pyelogram. Contrast medium which has been introduced into the calyces, pelvis and upper ureter via a ureteric catheter can be clearly identified. This technique affords considerably better visualization of the calyces than can be achieved by the intravenous method. Note the relation of the ureter to the tips of the lumbar transverse processes and the characteristic 'cupping' or 'champagne glass' profiles of the tips of the lesser calyces where they surround the renal pyramids. 'Calyx' means a cup and such 'cupping' of the minor calyces is the normal appearance. (The major calyces are not cups and are hence inappropriately named.)
Figure 91.10 Multislice CT renal angiogram. A, Coronal reformat. B, Axial reformat.
The hilum of the kidney leads into a central renal sinus, lined by the renal capsule and almost filled by the renal pelvis and vessels, the remaining space being filled by fat. Within the renal sinus, the collecting tubules of the nephrons of the kidney open onto the summits of the renal papillae to drain into minor calyces, funnelshaped expansions of the upper urinary tract (Figs 91.9, 91.15). The renal capsule covers the external surface of the kidney and continues through the hilum to line the sinus and fuse with the adventitial coverings of the minor calyces. Each minor calyx surrounds either a single papilla or, more rarely, groups of two or three papillae. The minor calyces unite with their neighbours to form two or possibly three larger chambers, the major calyces. The calyces of each kidney are usually arranged in seven pairs (seven ventral and seven dorsal) although there is wide variation. The calyces drain into the infundibula. The renal pelvis is normally formed from the junction of two infundibula, one from the upper and one from the lower pole calyces, but there may be a third, draining the calyces in the mid-portion of the kidney. The calyces are usually grouped so that three pairs drain into the upper pole infundibulum and four pairs into the lower pole infundibulum. If there is a middle infundibulum, the distribution is normally three pairs at the upper pole, two in the middle, and two at the lower pole. There is considerable variation in the arrangement of the infundibula and in the extent to which the pelvis is intrarenal or extrarenal. The funnel-shaped renal pelvis tapers
as it passes inferomedially, traversing the renal hilum to become continuous with the ureter (Figs 92.1, 91.8, 91.9, 91.15). It is rarely possible to determine precisely where the renal pelvis ceases and the ureter begins: the region is usually extrahilar and normally lies adjacent to the lower part of the medial border of the kidney. Rarely, the entire renal pelvis has been found to lie inside the sinus of the kidney so that the pelviureteric region occurrs either in the vicinity of the renal hilum or completely within the renal sinus. The calyces, renal pelvis and ureter are well-demonstrated radiologically following an intravenous injection of radio-opaque contrast which is excreted in the urine (intravenous urography - IVU) (Fig. 92.1); or after the introduction of radioopaque contrast into the ureter by catheterization through a cystoscope (ascending or retrograde pyelography (Fig. 91.9). Normal cupping of the minor calyces by projecting renal papillae may be obliterated by conditions that cause hydronephrosis, chronic distension of the ureter and renal pelvis due to upper or lower urinary tract obstruction resulting in elevated intrapelvic pressure.
RENAL CALCULI An understanding of intrarenal and ureteric anatomy is essential when managing patients with calculi, particularly now that minimal invasive techniques are available to treat this common pathology. Smaller renal calculi are treated with extracorporeal shock wave lithotripsy. Stones in the lower pole of the kidney clear less well if the angle between the infundibulum of the calyx containing the stone and the ureter is acute, or if there is a particularly long and narrow infundibulum. Percutaneous stone extraction is most frequently achieved by puncturing a posterior calyx with a needle. Posterior calyces are seen to lie more medially when looking at an intravenous urogram because of the normal rotation of the kidney. Ureteric calculi tend to be arrested in their descent in either the pelviureteric region, the point where the ureter passes over the pelvic brim, or the vesicoureteric junction, because these are the three areas where the ureter is narrowest.
© 2008 Elsevier
VASCULAR SUPPLY AND LYMPHATIC DRAINAGE ARTERIES (Figs 91.1, 91.10) The paired renal arteries take c.20% of cardiac output to supply organs that represent less than one-hundredth of total body weight. They supply the kidneys through a number of subdivisions described sequentially as segmental, lobar, interlobar, and arcuate arteries. These are end arteries with no anastomoses. The arcuate arteries further divide into interlobular arteries which give rise to the afferent arteries to the glomeruli. The renal arteries branch laterally from the aorta just below the origin of the superior mesenteric artery. Both cross the corresponding crus of the diaphragm at right angles to the aorta (Figs 91.1, 91.11). The right renal artery is longer and often higher, passing posterior to the inferior vena cava, right renal vein, head of the pancreas and descending part of the duodenum. The left renal artery is a little lower and passes behind the left renal vein, the body of the pancreas and splenic vein. It may be crossed anteriorly by the inferior mesenteric vein. A single renal artery to each kidney is present in c.70% of individuals. The arteries vary in their level of origin and in their calibre, obliquity and precise relations. In its extrarenal course each renal artery gives off one or more inferior suprarenal arteries, a branch to the ureter and branches which supply perinephric tissue, the renal capsule and the pelvis. Near the renal hilum, each artery divides into an anterior and a posterior division, and these divide into segmental arteries supplying the renal vascular segments. Accessory renal arteries are common (30% of individuals), and usually arise from the aorta above or below the main renal artery and follow it to the renal hilum. They are regarded as persistent embryonic lateral splanchnic arteries. Accessory vessels to the inferior pole cross anterior to the ureter and may, by obstructing the ureter, cause hydronephrosis. Rarely, accessory renal arteries arise from the coeliac or superior mesenteric arteries near the aortic bifurcation or from the common iliac arteries. Segmental arteries (Fig. 91.12) page 1274 page 1275
Figure 91.11 CT renal venogram. Acquired from a multislice CT examination and reconstructed as a 3D surface shaded reformat.
The segmental arteries branch successively into lobar, interlobar, arcuate and interlobular arteries, afferent and efferent glomerular arterioles and cortical intertubular capillary plexuses. The cortical venous radicles drain them and also the vasa recta and associated capillary plexuses of the medulla into the renal vein (Figs 91.12, 91.15). Renal vascular segmentation was originally recognized by John Hunter in 1794, but the first detailed account of the primary pattern was produced in the 1950s from casts and radiographs of injected kidneys. Five arterial segments have been identified. The apical segment occupies the anteromedial region of the superior pole. The superior (anterior) segment includes the rest of the superior pole and the central anterosuperior region. The inferior segment encompasses the whole lower pole. The middle (anterior) segment lies between the anterior and inferior segments. The posterior segment includes the whole posterior region between the apical and inferior segments. This is the pattern most commonly seen and although there can be considerable variation it is the pattern that clinicians most frequently encounter when performing partial nephrectomy. Whatever pattern is present, it must be emphasized that vascular segments are supplied by virtual end arteries. In contrast, larger intrarenal veins have no segmental organization and anastomose freely. Brödel (1911) described a relatively avascular longitudinal zone (the 'bloodless' line of Brödel) along the convex renal border, which was proposed as the most suitable site for surgical incision. However, many vessels cross this zone, and it is far from 'bloodless': planned radial or intersegmental incisions are preferable. Knowledge of the vascular anatomy of the kidney is important when undertaking partial nephrectomy for renal cell cancers. In this surgery the branches of the
renal artery are defined so that the surgeon may safely excise the renal substance containing the tumour whilst not compromising the vascular supply to the remaining renal tissue. Lobar, interlobar, arcuate and interlobular arteries
Initial branches of segmental arteries are lobar, usually one to each renal pyramid. Before reaching the pyramid they subdivide into two or three interlobar arteries, extending towards the cortex around each pyramid. At the junction of the cortex and medulla, interlobar arteries dichotomize into arcuate arteries which diverge at right angles. As they arch between cortex and medulla, each divides further, ultimately supplying interlobular arteries which diverge radially into the cortex. The terminations of adjacent arcuate arteries do not anastomose but end in the cortex as additional interlobular arteries. Though most interlobular arteries come from arcuate branches, some arise directly from arcuate or even terminal interlobar arteries (Fig. 91.15). Interlobular arteries ascend towards the superficial cortex or may branch occasionally en route (Fig. 91.15). Some are more tortuous and recurve towards the medulla at least once before proceeding towards the renal surface. Others traverse the surface as perforating arteries to anastomose with the capsular plexus (which is also supplied from the inferior suprarenal, renal and gonadal arteries). Afferent and efferent arterioles
Afferent glomerular arterioles are mainly the lateral rami of interlobular arteries. A few arise from arcuate and interlobar arteries when they vary their direction and angle of origin: deeper ones incline obliquely back towards the medulla, the intermediate pass horizontally, and the more superficial approach the renal surface obliquely before ending in a glomerulus (Figs 91.13, 91.15). From most glomeruli (except those at juxtamedullary and some at intermediate cortical levels) efferent glomerular arterioles soon divide to form a dense peritubular capillary plexus around the proximal and distal convoluted tubules. In the main renal cortical circulation there are thus two sets of capillaries in series, glomerular and peritubular, linked by efferent glomerular arterioles. From the venous ends of the peritubular plexuses fine radicles converge to join interlobular veins, one with each interlobular artery. Many interlobular veins begin beneath the fibrous renal capsule by the convergence of several stellate veins, which drain the most superficial zone of the renal cortex and so are named from their surface appearance. Interlobular veins pass to the corticomedullary junction. They also receive some ascending vasa recta and end in arcuate veins, which accompany arcuate arteries, and anastomose with neighbouring veins. Arcuate veins drain into interlobar veins, which anastomose and form the renal vein. page 1275 page 1276
Figure 91.12 Segmental arterial anatomy of the right kidney. (By permission from Walsh PC, Retik AB, Vaughan ED et al (eds) 2002 Campbell's Urology, 8th edn. Philadelphia: Saunders.)
The vascular supply of the renal medulla is largely from efferent arterioles of juxtamedullary glomeruli, supplemented by some from more superficial glomeruli, and 'aglomerular' arterioles (probably from degenerated glomeruli). Efferent glomerular arterioles passing into the medulla are relatively long, wide vessels, and contribute side branches to neighbouring capillary plexuses before entering the medulla, where each divides into 12-25 descending vasa recta. As their name suggests, these run straight to varying depths in the renal medulla, contributing side branches to a radially elongated capillary plexus (Fig. 91.15) applied to the descending and ascending limbs of renal loops and to collecting ducts. The venous ends of capillaries converge to the ascending vasa recta, which drain into arcuate or interlobular veins. An essential feature of the vasa recta (particularly in the outer medulla) is that both ascending and descending vessels are grouped into vascular bundles, within which the external aspects of both types are closely apposed, bringing them close to the limbs of renal loops and collecting ducts. As these bundles converge centrally into the renal medulla they contain fewer vessels: some terminate at successive levels in neighbouring capillary plexuses. This proximity of descending and ascending vessels with each other and adjacent ducts provides the structural basis for the countercurrent exchange and multiplier
phenomena (Figs 91.13, 91.15). These complex renal vascular patterns show regional specializations which are closely adapted to the spatial organization and functions of renal corpuscles, tubules and ducts (Figs 91.13, 91.14, 91.15).
Figure 91.13 'Microfil' injection of the arterial tree of human kidney, high power micrograph. The juxtamedullary efferent arterioles leave the glomeruli to form medullary vascular bundles (descending vasa recta). (Preparation provided by DB Moffat, Department of Anatomy, University College of Wales, Cardiff.)
Renal, interlobar and arcuate arteries are typical large muscular arteries and the interlobular vessels resemble small muscular arteries. Afferent glomerular vessels have a typical arteriolar structure with a muscular coat two to three cells thick; this coat and the connective tissue components of the wall diminish near a glomerulus until a point 30-50 µm proximal to it where arteriolar cells begin to show modifications typical of the juxtaglomerular apparatus (p. 1283). The efferent arterioles from most cortical glomeruli have thicker walls and a narrower calibre than corresponding afferents. Although the afferent arteriole is generally considered to be solely responsible for tubuloglomerular feedback, the role of the efferent arteriole in this process has been reviewed by Davis (1991). The peritubular and medullary capillaries possess a well-defined basal lamina and their endothelial cells have typically fenestrated cytoplasm, as in ascending vasa recta, whereas the descending vasa recta have thicker, continuous endothelium.
VEINS (Fig. 91.11) The large renal veins lie anterior to the renal arteries and open into the inferior vena cava almost at right angles. The left is three times longer than the right (7.5 cm and 2.5 cm) and for this reason the left kidney is the preferred side for live donor nephrectomy. It runs from its origin in the renal hilum, posterior to the splenic vein and the body of pancreas, and then across the anterior aspect of the aorta, just below the origin of the superior mesenteric artery. The left gonadal vein enters it from below and the left suprarenal vein, usually receiving one of the left inferior phrenic veins, enters it above but nearer the midline. The left renal vein enters the inferior vena cava a little superior and to the right. The right renal vein is behind the descending duodenum and sometimes the lateral part of the head of the pancreas. page 1276 page 1277
Figure 91.14 Resin corrosion cast of human kidneys. Ureter, pelvis and calyces are yellow; aorta, renal arteries and their branches are red. Compare with Figs 91.9, 91.10A. (Prepared by the late DH Tompsett of the Royal College of Surgeons of England. By permission of the Museums of The Royal College of Surgeons.)
The left renal vein may be double, one vein passing posterior, the other anterior, to the aorta before joining the inferior vena cava. This is sometimes referred to as persistence of the 'renal collar'. The anterior vein may be absent so that there is a single retroaortic left renal vein. Because of its close relationship with the aorta, the left renal vein may be ligated during surgery for aortic aneurysm. This seldom results in any harm to the kidney, provided that the ligature is placed to the right of the draining gonadal and suprarenal veins, because these usually provide adequate collateral venous drainage. The right renal vein has no significant collateral drainage and cannot be ligated with impunity.
LYMPHATIC DRAINAGE Renal lymphatic vessels begin in three plexuses, around the renal tubules, under the renal capsule, and in the perirenal fat (the latter two connect freely). Collecting
vessels from the intrarenal plexus form four or five trunks which follow the renal vein to end in the lateral aortic nodes; as they leave the hilum the subcapsular collecting vessels join them. The perirenal plexus drains directly into the same nodes.
© 2008 Elsevier
INNERVATION A dense plexus of autonomic nerves around the renal artery is formed by rami from the coeliac ganglion and plexus, aorticorenal ganglion, lowest thoracic splanchnic nerve, first lumbar splanchnic nerve and aortic plexus. Small ganglia occur in the renal plexus, the largest usually behind the origin of the renal artery. The plexus continues into the kidney around the arterial branches to supply the vessels, renal glomeruli, and tubules, especially the cortical tubules. Axons from plexuses around the arcuate arteries innervate juxtamedullary efferent arterioles and vasa recta, which control the blood flow between the cortex and medulla without affecting the glomerular circulation. Axons from the renal plexus contribute to ureteric and gonadal plexuses. The ureteric plexus receives, in its upper part, branches from the renal and aortic plexuses, in its intermediate part, branches from the superior hypogastric plexus and hypogastric nerve, and in its lower part, branches from the hypogastric nerve and inferior hypogastric plexus. This supply influences the inherent motility of the ureter.
© 2008 Elsevier
MICROSTRUCTURE (Fig. 91.15) The kidney is composed of many tortuous, closely packed uriniferous tubules, bounded by a delicate connective tissue in which run blood vessels, lymphatics and nerves. Each tubule consists of two embryologically distinct parts (p. 1373), the nephron, which produces urine, and the collecting duct, which completes the concentration of urine and through which urine passes out of the kidney to the ureter and urinary bladder. The nephron consists of a renal corpuscle, concerned with filtration from the plasma, and a renal tubule, concerned with selective resorption from the filtrate to form the urine. Collecting ducts carry fluid from several renal tubules to a terminal papillary duct, opening into a minor calyx at the apex of a renal papilla (Fig. 91.15A). Papillary surfaces show numerous minute orifices of these ducts and pressure on a fresh kidney expresses urine from them.
RENAL CORPUSCLE (Figs 91.16, 91.17) Renal corpuscles are small rounded structures averaging c.0.2 mm in diameter, visible in the renal cortex deep to a narrow peripheral cortical zone (Figs 91.15, 91.16). There are one to two million renal corpuscles in each kidney, their number decreasing with age. Each has a central glomerulus of vessels and a glomerular (Bowman's) capsule, from which the renal tubule originates. Glomerulus
A glomerulus is a collection of convoluted capillary blood vessels, united by a delicate mesangial matrix and supplied by an afferent arteriole which enters the capsule opposite the urinary pole, where the filtrate enters the tubule. An efferent arteriole emerges from the same point, the vascular pole of the corpuscle. Glomeruli are simple in form until late prenatal life; some remain so for about 6 months after birth, the majority maturing by 6 years and all by 12 years (p. 1373). Bowman's capsule (Figs 91.16, 91.18, 91.19, 91.20)
Bowman's capsule is the blind expanded end of a renal tubule, and is deeply invaginated by the glomerulus. It is lined by a simple squamous epithelium on its outer (parietal) wall; its glomerular, juxtacapillary (visceral) wall is composed of specialized epithelial podocytes. Between the two walls of the capsule is a flattened urinary space, continuous with the proximal convoluted tubule (Figs 91.15B, 91.16). The basal lamina of the visceral capsular podocytes is shared with that of the glomerular endothelium. page 1277 page 1278
Figure 91.15 The structural and functional organization of the kidney. The major structures in the kidney cortex and medulla (left), the position of cortical and juxtamedullary nephrons (middle) and the major blood vessels (right). B, The regional microstructure and principal activities of a kidney nephron and collecting duct. For clarity, a nephron of the long loop (juxtamedullary) type is shown.
Podocytes surrounding the capillary loops are stellate cells, whose major (primary) foot processes curve around capillaries. These branch to form
secondary processes which are applied closely to the basal lamina and they, or tertiary processes, give rise to the terminal pedicels (Figs 91.18, 91.19, 91.20). Pedicels of one cell alternate with those of an adjacent cell and interdigitate tightly with each other. Pedicels are separated by narrow (25 nm) gaps, the filtration slits (Figs 91.18, 91.20). The latter are covered by a dense, membranous slit diaphragm, through which filtrate must pass to enter the urinary space. The glomerular endothelium is finely fenestrated. The principal barrier to the passage of fluid from capillary lumen to urinary space is the shared endothelial and podocyte basal lamina (Fig. 91.20). This is c.0.33 µm thick in man, and is produced by the fusion of endothelial and podocyte laminae; it is finely fibrillar and shows three layers. The first layer, towards the endothelial surface, and the third layer are pale-staining (lamina rara interna and lamina rara externa, respectively); the middle layer is dense and fibrous (lamina densa). This arrangement differs from basal laminae elsewhere, and reflects its dual origin. The glomerular basal lamina acts as a selective filter, allowing the passage from blood, under pressure, of water and various small molecules and ions in the circulation. Haemoglobin may cross the filter, but larger molecules and those of similar size with a negative charge, are largely retained. Most protein that does enter the filtrate is selectively resorbed and degraded by cells of the proximal convoluted tubule. Irregular mesangial cells, with phagocytic and contractile properties, lie within the delicate supportive mesangial matrix (mesangium) of the glomerulus, which they secrete. The mesangium is a specialized connective tissue which binds the loop of glomerular capillaries and fills the spaces between endothelial surfaces that are not invested by podocytes (Fig. 91.15B). Mesangial cells are related to vascular pericytes (p. 146) and are concerned with the turnover of glomerular basal lamina. They clear the glomerular filter of, e.g. immune complexes and cellular debris, and their contractile properties help to regulate blood flow. Similar cells, the extraglomerular mesangial (lacis) cells, lie outside the glomerulus at the vascular pole and form part of the juxtaglomerular apparatus.
RENAL TUBULE A renal or uriniferous tubule consists of a glomerular capsule leading into a proximal convoluted tubule, connected to the capsule by a short neck which continues into a sinuous or coiled convoluted part (Fig. 91.15B). This straightens as it approaches the medulla and becomes the descending thick limb of the loop of Henle which is connected to the ascending limb by an abrupt U-turn. The limbs of the loop of Henle are narrower and thin-walled as they traverse the deeper medullary tissue, forming the descending and ascending thin segments. The ascending thick limb continues into the distal tubule. The tubule wall shows a focal thickening, the macula densa, where it comes close to the vascular pole of its parent glomerulus at the start of the convoluted part of the distal tubule. The nephron finally straightens once more as the connecting tubule, which ends by joining a collecting duct. Collecting ducts originate in the cortical medullary rays and join others at intervals. They finally open into wider papillary ducts which open on to a papilla; their numerous orifices form a perforated area cribrosa on the surface at its tip (Fig. 91.15B). page page page page
1278 1279 1279 1280
Figure 91.16 A renal corpuscle in the kidney cortex (trichrome-stained), showing a glomerulus (centre) within its capsule, enclosing the urinary space (left). An arteriole containing erythrocytes is seen at the vascular pole of the capsule (arrow), where it is associated with the macula densa of a distal convoluted tubule (DCT, top). The macula densa comprises a group of more closely packed, taller cells in part of the DCT wall adjacent to the glomerulus. Profiles of proximal convoluted tubules (with brush borders obscuring their lumen) are also visible (e.g. top right). (Photograph by Kevin Fitzpatrick on behalf of GKT School of Medicine, London.)
Renal tubules are lined throughout by a single-layered epithelium (Figs 91.15B, 91.16, 91.21). The type of epithelial cell varies according to the functional roles of the different regions, e.g. active transport and passive diffusion of various ions and water into and out of the tubules; reabsorption of organic components such as glucose and amino acids; uptake of any proteins which leak through the glomerular filter.
Figure 91.17 Renal cortex and medulla. The cortex contains renal corpuscles (RC), renal tubules (T), blood vessels (V), and medullary rays (MR). The medulla contains many tubules of the loops of Henle and collecting ducts (CD), sectioned in transverse, oblique and longitudinal planes, and islands of capillaries, the vasa recta (VR). (By permission from Dr JB Kerr, Monash University, from Kerr JB 1999 Atlas of Functional Histology. London: Mosby.)
The proximal convoluted tubule is lined by cuboidal or low columnar epithelium and has a brush border of tall microvilli on its luminal surface. The shape of the cells depends on tubular fluid pressure, which in life distends the lumen and flattens the cells; their shape becomes taller when glomerular blood pressure falls postmortem or at biopsy. The cytoplasm of proximal tubular cells is strongly eosinophilic and their nuclei are euchromatic and central. By light microscopy their bases show faint striations, which ultrastructurally are seen to be due to a complex series of infoldings of the basal plasma membrane, between which numerous mitochondria are orientated perpendicularly. The lateral surfaces of adjacent epithelial cells interdigitate to increase the complexity of the basolateral plasma membrane. Taking into account the microvilli on their luminal surfaces, these cells possess large areas of plasma membrane in contact with tubular fluid and the extratubular space: this arrangement facilitates the transport of ions and small molecules against steep concentration gradients. The abundant mitochondria supply the energy, as ATP, needed for this process. Sodium/potassium adenosine triphosphatase (Na/K ATPase) is located in apical and basal membranes, and the cytoplasm contains numerous other enzymes concerned with ion transport. Water and other solutes pass between cells (paracellular transport) passively along osmotic and electrochemical gradients, probably through leaky apical tight junctions. Pinocytotic vesicles are found near the apical surface, and represent the means by which small proteins and peptides from the filtrate are internalized and degraded by associated lysosomes. Peroxisomes and lipid droplets abound in the cytoplasm. page 1280 page 1281
Figure 91.18 Scanning electron micrograph showing podocytes forming the visceral layer of Bowman's capsule in the renal corpuscle. Podocyte cell bodies (P) send out primary processes which branch several times to end in fine pedicels which wrap tightly around the glomerular capillaries (C), interdigitating with similar pedicels from a neighbouring podocyte. The pedicels and their underlying basal lamina form an important part of the glomerular filtration apparatus. (By permission of Igaku-Shoin, from Fujita T, Tanaka K, Tokunaga J 1981 SEM Atlas of Cells and Tissues, Vol 3.)
The loop of Henle consists of a thin segment (c.30 µm in diameter), lined by low cuboidal to squamous cells, and a thick segment (c.60 µm in diameter) composed of cuboidal cells like those in the distal convoluted tubule. The thin segment forms most of the loop in juxtamedullary nephrons which reach deep into the medulla. Few organelles appear in cells lining the thin segment, indicating that these cells play a passive, rather than an active, role in ion transport. The thick segment is composed of cuboidal epithelium with many mitochondria, deep basolateral folds and short apical microvilli, indicating a more active metabolic role. The thick limb of the loop of Henle is the source of Tamm-Horsfall protein in normal urine. Cells of the distal tubule are cuboidal and resemble those in the proximal tubule. They have few microvilli, and so the tubular lumen has a more distinct outline. The basolateral folds containing mitochondria are deep, almost reaching the luminal aspect (Fig. 91.15B). Enzymes concerned in active transport of sodium, potassium and other ions are abundant. At the junction of the straight and convoluted regions the distal tubule comes close to the vascular pole of its parent renal corpuscle. Here, tubular cells form a sensory structure, the macula densa, which is concerned with the regulation of blood flow and thus filtration rate. Cells in the terminal part of the distal tubule have fewer basal folds and mitochondria and constitute a connecting duct formed from metanephric mesenchyme during embryogenesis. Collecting ducts are lined by simple cuboidal or columnar epithelium, which increases in height from the cortex, where they receive the contents of distal tubules, to the wide papillary ducts which discharge at the area cribrosa. The pale-staining principal cells have relatively few organelles or lateral interdigitations and only occasional microvilli. A second cell type, intercalated or dark cells (also present in smaller numbers in the distal convoluted tubule), have longer microvilli and more mitochondria. These secrete H+ into the filtrate and function in the maintenance of acid-base homeostasis.
PRODUCTION OF URINE (Fig. 91.22) Glomerular filtration
Glomerular filtration is the passage of water containing dissolved small molecules from the blood plasma to the urinary space in the glomerular capsule. Larger molecules, e.g. plasma proteins above c.70 kilodaltons and those with a net negative charge, polysaccharides and lipids, are largely retained in blood by the selective permeability of the glomerular basal lamina.
page 1281 page 1282
Figure 91.19 Capillary loops (C) of the renal glomerulus; one profile contains an electron-dense erythrocyte. Capillaries are lined by a fenestrated endothelium (F) and endothelial cell nuclei (E) are seen bulging into the capillary lumina. The nuclei of several epithelial podocytes (P) of the visceral layer of Bowman's capsule can be seen. Their primary processes (P1 ) give rise to numerous secondary foot processes (P2 ) and these, or tertiary processes, rest on the glomerular basal lamina (BL). A mesangial stalk is shown at the top right, comprising mesangial cells (M) and a dense mesangial matrix (MM), which supports the capillary loops. The mesangium is separated from the capillary lumen only by the endothelial cell cytoplasm, whereas the podocytes and their basal laminae continue around the mesangial stalk and separate it from the urinary (Bowman's) space (BS), which ramifies throughout the glomerulus. Part of the outer parietal layer of Bowman's capsule (BC) is seen to the left. (By permission from Young B, Heath JW 2000 Wheater's Functional Histology. Edinburgh: Churchill Livingstone.)
Figure 91.20 Filtration apparatus of the renal corpuscle, formed by the fenestrated capillary endothelial cells, the filtration slits between podocyte pedicels and their thick, shared basal lamina. BL, basal lamina; C, capillary loops; E, endothelial cell cytoplasm; F, fenestrations; P, podocyte, P 1 , primary processes and P 2 , secondary foot processes which rest on the glomerular basal lamella. (By permission from Young B, Heath JW 2000 Wheater's Functional Histology. Edinburgh: Churchill Livingstone.)
Figure 91.21 Part of the renal medulla (trichrome stained) in cross-section. Note large collecting ducts and small thin segments of the loop of Henle, interspersed with vasa recta (V).
Filtration occurs along a steep pressure gradient between the large glomerular capillaries and the urinary space, the principal structure separating the two being the glomerular basal lamina. This gradient far exceeds the colloid osmotic pressure of blood which opposes the outward flow of filtrate. In the peripheral renal cortex the arteriolar pressure gradient is enhanced by the higher calibre of afferent, compared with efferent, glomerular arterioles. In all glomeruli the rate of filtration can be altered by changes in the tone of the glomerular arterioles. When first formed, the glomerular filtrate is isotonic with glomerular blood and has an identical concentration of ions and small molecules. Selective resorption
Selective resorption from the filtrate is an active process and occurs mainly in the proximal convoluted tubules, which resorb glucose , amino acids , phosphate, chloride, sodium, calcium and bicarbonate; they also take up small proteins by endocytosis. Cells of the proximal tubules are permeable to water, which passes out of the tubules passively, so that the filtrate remains locally isotonic with blood. The rest of the tubule reabsorbs most of the water (variable, but up to 95%), such that when it reaches the calyces, urine is generally much reduced in volume and hypertonic to blood. This depends on the establishment of high osmolality in the medullary interstitium, which exerts considerable osmotic pressure on waterpermeable regions of the tubule (Fig. 91.22). Countercurrent multiplier mechanism
The countercurrent multiplier mechanism is responsible for producing a high osmolality in the extratubular interstitial tissue of the renal medulla. Water passes freely from the tubular lumen into the adjacent medullary interstitium along the descending limb of the loop of Henle. This part of the tubule is less permeable to solutes. In the thick segment of the ascending limb, sodium and chloride ions are actively transported from the tubule lumen to interstitial spaces, whilst the tubular epithelium remains impermeable to water. The increased interstitial osmolality causes water to be withdrawn from the descending part of the loop, thus concentrating the filtrate. Tubular fluid flows in a countercurrent on its descent into and ascent out of the medulla: it is augmented by new isotonic fluid entering the loop and depleted by hypotonic fluid leaving the loop, as solutes are actively resorbed. In this way the osmotic gradient within the interstitium is multiplied from the corticomedullary boundary to the medullary pyramids, where it reaches an equilibrium of four to five times the osmolality of plasma. Urea contributes c.50% of the medullary osmotic strength, mainly contributed passively by the medullary part of the collecting ducts. These are generally highly permeable to urea , and permeability is enhanced by antidiuretic hormone (ADH, vasopressin ). Although the tonicity of the tubular fluid changes during its passage through the steep osmotic gradient within the medulla, the osmotic gradient between ascending and descending limbs at each level never exceeds 200 mOsm/kg, a force which can be sustained by the cells of the tubular wall. Countercurrent exchange mechanism (Fig. 91.13B)
Rapid removal of ions from the renal medulla by the circulation of blood is minimized by another looped countercurrent system. This is the countercurrent
exchange mechanism, in which arterioles entering the medulla pass for long distances parallel to the venules leaving it, before ending in capillary beds around tubules. This close apposition of oppositely flowing blood allows the direct diffusion of ions from outflowing to inflowing blood, so that the vasa recta (Figs 91.13, 91.15A) conserve the high osmotic pressure in the medulla. Concentration of urine
Because sodium and chloride ions are selectively resorbed by the cells of the ascending limbs and distal tubules under aldosterone control (p. 1247), the filtrate at the distal end of the convoluted tubules is hypotonic. As it reaches the collecting ducts, fluid descends again through the medulla and thus re-enters a region of high osmotic pressure. The cells lining the collecting ducts are variably permeable to water, under the influence of neurohypophyseal ADH. Water follows an osmotic gradient into the adjacent extratubular spaces, so that the tonicity of the filtrate gradually rises along collecting ducts, until at the tip of the renal pyramids it is above that of blood. As much as 95% of water in the original glomerular filtrate is thus resorbed into blood. This complex system is highly flexible and the balance between the rate of filtration and absorption can be varied to meet current physiological demands. Control of hydrogen and ammonium ion concentrations is essential to the regulation of acids and bases in the blood; secretion of various ions occurs at several sites. Over 91% of ingested potassium is excreted in urine, largely through secretion by cells of the distal tubule and collecting duct. For further details of renal physiology see Davies et al 2001. page 1282 page 1283
Figure 91.22 The inter-relationships between the countercurrent multiplier and exchange mechanisms which operate in the renal medulla. The movements of ions and water and the action of antidiuretic hormone (ADH) are indicated. For further details see text. (By permission from Stevens A, Lowe JS 1996 Human Histology, 2nd edn. London: Mosby.)
JUXTAGLOMERULAR APPARATUS The juxtaglomerular apparatus provides a tubuloglomerular feedback system which maintains systemic arterial blood pressure during a reduction in vascular volume and decrease in filtration rate. The afferent and efferent arterioles at the vascular pole of a glomerulus and the macula densa of the distal tubule of the same nephron lie in close proximity, enclosing a small cone of tissue populated by extraglomerular mesangial (lacis) cells (Fig. 91.15B). The cells of the tunica media of the afferent and, to a lesser extent, efferent, arterioles differ from typical smooth muscle cells. They are large, rounded myoepithelioid cells and their cytoplasm contains many mitochondria and dense, renin-containing vesicles, 1040 nm in diameter. These juxtaglomerular cells form one element of the juxtaglomerular apparatus. The second element of the juxtaglomerular apparatus is the sensory component, the macula densa of the distal tubule. Up to 40 cells in the tubule wall form a cluster of taller, more tightly packed cells with large, oval nuclei (Fig. 91.16). Their mitochondria are concentrated apically. Macula densa cells are osmoreceptors, sensing the NaCl content of the filtrate after its passage through the loop of Henle. When NaCl concentrations in the filtrate change, tubuloglomerular feedback mechanisms operate to maintain the inverse relationship between salt concentration and glomerular filtration rate. Juxtaglomerular cells release renin, an enzyme which acts on circulating angiotensinogen (a liver protein) to activate the cascade whereby angiotensin II increases blood pressure (and therefore filtration rate), stimulates aldosterone and ADH release and increases sodium ion and water resorption, primarily from the distal tubule, to increase plasma volume. Macula densa cells are thought to respond to high salt concentration in the distal tubule by releasing nitric oxide , which inhibits the tubuloglomerular feedback response and reduces filtration rate. The role of macula densa cells in the stimulation of renin release to increase filtration rate is less well understood. The third element of the juxtaglomerular apparatus is a population of extraglomerular mesangial cells which form a network (or lace, hence their alternative name of lacis cells) of stellate cells connecting the macula densa sensory cells with the juxtaglomerular effector cells. It is likely that extraglomerular mesangial cells transmit the sensory signal, possibly through gap junctions. They may also signal to contractile glomerular mesangial cells and effect vasoconstriction directly within the glomerulus. Adrenergic nerve fibres occur in small numbers among these cells.
RENAL CALYCES AND PELVIS page 1283 page 1284
The wall of the proximal part of the urinary tract is composed of three layers, an outer connective tissue adventitia, an intermediate layer of smooth muscle and an inner mucosa. The mucosal lining of the renal calyces and pelvis is identical in structure to that of the ureter (p. 1288) and will not be considered further here. The adventitia consists of loose fibroelastic connective tissue which merges with retroperitoneal areolar tissue. Proximally the coat fuses with the fibrous capsule of the kidney lining the renal sinus. The smooth muscle of the renal calyces and pelvis is composed of two distinct types of smooth muscle cell. One type of muscle cell is identical to that described for the ureter and can be traced proximally through the pelviureteric region and renal pelvis as far as the minor calyces. The other type of cell forms the muscle coat of each minor calyx and continues into the major calyces and pelvis where it forms a distinct inner layer. The cells also form a thin sheet of muscle which covers each minor calyx and extends across the renal parenchyma between the attachments of neighbouring minor calyces, thereby linking each minor calyx to its neighbours. This discrete inner layer of atypical smooth muscle ceases in the pelviureteric region so that the proximal ureter lacks such an inner layer. Pacemaker cells that initiate renal pelvic and ureteric peristalsis are sited within the calyces. These allow coordinated peristalsis of the ureter c.6 times a minute.
OTHER RENAL CELLS Other cells essential to renal structure and function lie between the renal tubules and blood vessels. Connective tissue is inconspicuous in the cortex but prominent in the medulla, particularly in the papillae. Medullary interstitial cells, which may be modified fibroblasts, form vertical stacks of tangentially orientated cells
between the more distal collecting ducts, like the rungs of a ladder. These cells secrete prostaglandins and may contribute, with cortical tubular cells, to the renal source of erythropoietin. REFERENCES Brodel M 1911 The intrinsic blood-vessels of the kidney and their significance in nephrotomy. John Hopkins Hosp Bull 12: 10-13. Original description of a relatively avascular longitudinal zone within the kidney, proposed as a site for surgical incision. Burkhill GJC, Healy JC 2000 Anatomy of the peritoneum. Imaging 12: 10-20. Review of the imaging literature describing the contentious anatomy of the perirenal fascia. Davies A, Blakeley AGH, Kidd C 2001 The renal system. In: Human Physiology, Chapter 8. Edinburgh: Churchill Livingstone: 713-97. Davies JM 1991The role of the efferent arteriole in tubuloglomerular feedback. Kidney Int (Suppl) 32: S713. Gosling JA, Dixon JS 1974 Species variation in the location of upper urinary tract pacemaker cells. Invest Urol 11: 418. Early paper describing the identification of pacemaker cells in various species. Medline Similar articles Merklin RJ, Michels NA 1958 The variant renal and suprarenal blood supply with data on the inferior phrenic, ureteral, and gonadal arteries: a statistical analysis based on 185 dissections and a review of the literature. J Int Coll Surg; 29: 41-76. A review of renal vascular anatomy in almost 11,000 kidneys. Medline Similar articles Mitchell GAG 1950 The renal fascia. Br J Surg 37: 257-66. Demonstrating that the perirenal fascia is a multilaminate structure rather than a single fused fascia. Medline Similar articles Full article Novick AC 1998 Anatomic approaches in nephron-sparing surgery for renal cell carcinoma. Atlas Urol Clin North Am 6: 39.
© 2008 Elsevier
92 Ureter The ureters are muscular tubes whose peristaltic contractions convey urine from the kidneys to the urinary bladder (Fig. 92.1). Each measures 25-30 cm in length and is thick-walled, narrow, and continuous superiorly with the funnel-shaped renal pelvis. Each descends slightly medially anterior to psoas major, and enters the pelvic cavity where it curves laterally, then medially, as it runs down to open into the base of the urinary bladder. Its diameter is c.3 mm but is slightly less at its junction with the renal pelvis, at the brim of the lesser pelvis near the medial border of psoas major, and where it runs within the wall of the urinary bladder, which is its narrowest part. These are the commonest sites for renal stone impaction. The renal pelvis has already been described (p. 1274).
© 2008 Elsevier
RELATIONS (Fig. 92.2) In the abdomen the ureter descends posterior to the peritoneum on the medial part of psoas major, which separates it from the tips of he lumbar transverse processes. During surgery on intraperitoneal structures, the ureter can be tented up as the peritoneum is drawn anteriorly, resulting in inadvertent ureteric injury. Anterior to psoas major it crosses in front of the genitofemoral nerve and is obliquely crossed by the gonadal vessels. It enters the lesser pelvis anterior to either the end of the common iliac or the start of the external iliac vessels. The inferior vena cava is medial to the right ureter while the left ureter is lateral to the aorta. The inferior mesenteric vein has a long retroperitoneal course lying close to the medial aspect of the left ureter. At its origin the right ureter is usually overlapped by the descending part of the duodenum. It descends lateral to the inferior vena cava, and is crossed anteriorly by the right colic and ileocolic vessels. Near the superior aperture of the lesser pelvis it passes behind the lower part of the mesentery and terminal ileum. The left ureter is crossed by the gonadal and left colic vessels (Fig. 92.3). It passes posterior to loops of jejunum and sigmoid colon and its mesentery in the posterior wall of the intersigmoid recess. In the pelvis the ureter lies in extraperitoneal areolar tissue. At first it descends posterolaterally on the lateral wall of the lesser pelvis along the anterior border of the greater sciatic notch. Opposite the ischial spine it turns anteromedially into fibrous adipose tissue above levator ani to reach the base of the bladder. On the pelvic wall it is anterior to the internal iliac artery and the beginning of its anterior trunk, posterior to which are the internal iliac vein, lumbosacral nerve and sacroiliac joint. Laterally it lies on the fascia of obturator internus. It progressively crosses to become medial to the umbilical, inferior vesical, and middle rectal arteries. In males (Fig. 92.2), the pelvic ureter hooks under the vas deferens (Fig. 98.1), then passes in front of and slightly above the upper pole of the seminal vesicle to traverse the bladder wall obliquely before opening at the ipsilateral trigonal angle (Fig. 98.2). Its terminal part is surrounded by tributaries of the vesical veins. In females, the pelvic part at first has the same relations as in males, but anterior to the internal iliac artery it is immediately behind the ovary, forming the posterior boundary of the ovarian fossa (p. 1321). In the anteromedial part of its course to the bladder it is related to the uterine artery, uterine cervix and vaginal fornices. It is in extraperitoneal connective tissue in the inferomedial part of the broad ligament of the uterus where it may be damaged during hysterectomy. In the broad ligament, the uterine artery is anterosuperior to the ureter for 2.5 cm and then crosses to its medial side to ascend alongside the uterus. The ureter turns forwards slightly above the lateral vaginal fornix and is generally 2 cm lateral to the supravaginal part of the uterine cervix in this location. It then inclines medially to reach the bladder, with a variable relation to the front of the vagina. As the uterus is commonly deviated to one side, one ureter, usually the left, may be more extensively apposed to the vagina, and may cross the midline. The distal 1-2 cm of each ureter is surrounded by an incomplete collar of nonstriated muscle, which forms a sheath (of Waldeyer). The ureters pierce the posterior aspect of the bladder and run obliquely through its wall for a distance of 1.5-2.0 cm before terminating at the ureteric orifices. This arrangement is believed to assist in the prevention of reflux of urine into the ureter, since the intramural
ureters are thought to be occluded during increases in bladder pressure. There is no evidence of a classic ureteral sphincter mechanism in man. The longitudinally oriented muscle bundles of the terminal ureter continue into the bladder wall and at the ureteric orifices become continuous with the superficial trigonal muscle. In the distended bladder, in both sexes, the ureteric openings are c.5 cm apart, and c.2.5 cm apart when the bladder is empty. Duplex ureters In 1 in 125 individuals, two ureters drain the renal pelvis on one side; this is termed a duplex system. Bilateral duplex ureters occur in c.1 in 800 cases. The duplex ureters derive from two ureteric buds arising from the mesonephric duct. They are contained in a single fascial sheath and may fuse at any point along their course or may be separate until they insert through separate ureteric orifices into the bladder. Care must be taken not to compromise the blood supply of the second ureter when excising or reimplanting a single ureter of a duplex. The ureter from the upper pole of the kidney (the longer ureter) inserts more medially and caudally in the bladder than the ureter from the lower pole (the shorter ureter). This reflects their embryological development: the ureteric bud which is initially more proximal on the mesonephric duct has a shorter time to be pulled cranially in the bladder and so it inserts more distally in the mature bladder. The ureter from the lower pole has a shorter intramural course than the longer ureter and is prone to reflux. Ectopic ureters Single ureters and more commonly the longer ureter of a duplex system can insert more caudally and medially than normal in some individuals. In the male the ureter can insert at the bladder neck or posterior urethra, or rarely into the seminal vesicle, but it always inserts cranial to the external urethral sphincter. In the female, ectopic insertion can be distal to the external urethral sphincter in the urethra, or into the vagina, resulting in persistent childhood incontinence. Uretoroceles A ureterocele is a cystic dilatation of the lower end of the ureter: the ureteric orifice is covered by a membrane which expands as it is filled with urine and then deflates as it empties. Uretoroceles can vary in size with resultant obstructive 'back pressure' changes seen in the ureter and pelvicalyceal system proximally. They usually do not cause bladder outflow obstruction except for the rare prolapsing uretorocele. Prolapsing uretoroceles, though small, prolapse from their position around the uretero-vesical junction region in to the urethra, causing intermittent bladder outflow obstruction. They are identified antenatally with ultrasound and can result in obstruction to the ureter. In adults uretoroceles tend to be bilateral and small. They can produce obstruction to the ureter but commonly produce no clinical manifestations. Retrocaval ureter
page 1285 page 1286
page 1286 page 1287
Figure 92.1 A, Conventional intravenous urogram showing the renal calyces, pelvis and both ureters. B and C, Multislice CT urogram. B, Coronal reformat showing the enhancing renal parenchyma and both ureters along their entire length. C, 3D-surface shaded reformat showing the kidneys, ureters, bladder and surrounding bony anatomy.
Figure 92.2 Relations of lower right ureter in male. (By permission from Walsh PC, Retik AB, Vaughan ED et al (eds) 2002 Campbell's Urology, 8th edn. Philadelphia: Saunders.)
A persistence of the posterior cardinal vein, associated with high confluence of the right and left common iliac veins or a double inferior vena cava, may result a retrocaval ureter which passes behind the inferior vena cava before it emerges in front of it to pass from medial to lateral. Retrocaval ureter occurs in c.1 in 1500 individuals. Most commonly it has no clinical sequelae although it can result in upper ureteric obstruction.
© 2008 Elsevier
VASCULAR SUPPLY AND LYMPHATIC DRAINAGE ARTERIES AND VEINS (Fig. 92.3) The ureter is supplied by branches from the renal, gonadal, common iliac, internal iliac, vesical and uterine arteries and the abdominal aorta. The pattern of distribution is subject to much variation. There is a good longitudinal anastomosis between these branches on the wall of the ureter, which means that the ureter can be safely transected at any level intraoperatively, and a uretero-ureterostomy performed, without compromising its viability. The branches from the inferior vesical artery are constant in their occurrence and supply the lower part of the ureter as well as a large part of the trigone of the bladder. The branch from the renal artery is also constant and is preserved whenever possible in renal transplantation to ensure good vascularity of the ureter. The venous drainage generally follows the arterial supply.
LYMPHATIC DRAINAGE Lymph vessels begin in submucosal, intramuscular and adventitial plexuses, which all communicate. Collecting vessels from the upper ureter may join the renal collecting vessels or pass directly to the lateral aortic nodes near the origin of the gonadal artery; those from its lower abdominal part go to the common iliac nodes; and those from its pelvic part end in the common, external or internal iliac nodes.
© 2008 Elsevier
INNERVATION The ureter is supplied from the lower three thoracic, first lumbar, and the second to fourth sacral segments of the spinal cord by branches from the renal and aortic plexuses, and the superior and inferior hypogastric plexuses. The ureteric nerves consist of relatively large bundles of axons which form an irregular plexus in the adventitia of the ureter. Numerous smaller branches penetrate the ureteric muscle coat. Some of the adventitial nerves accompany the blood vessels and branch with them as they extend into the muscle layer. Others are unrelated to the vascular supply and lie free in the adventitial connective tissue around the circumference of the ureter. There is a gradual increase in innervation from the renal pelvis and upper ureter (which has a sparse distribution of autonomic nerves) to a maximum density in the juxtavesical segment. There are at least three different phenotypes: cholinergic, noradrenergic and peptidergic (substance P). Other neurotransmitters also exist, so the control mechanisms are likely to be complex. The functional significance of these different types of autonomic nerve fibres in relation to ureteric smooth muscle activity is not fully understood, however, it is recognized that nerves are not essential for the initiation and propagation of ureteric contraction waves. A branching plexus of fine axons occurs within the lamina propria and extends from the inner aspect of the muscle coat towards the base of the urothelium. These axons are cholinergic and, while some form perivascular plexuses, others lie in isolation from the vascular supply. A similar distribution of noradrenergic and peptidergic nerves has been observed throughout the lamina propria. The functional significance of these nerves, which are not related to blood vessels, remains unclear, but it seems probable that at least some of them are sensory in function. page 1287 page 1288
Figure 92.3 Course of the left ureter, showing how the proximal part takes its blood supply medially, and the distal part is supplied laterally. (By permission from Walsh PC, Retik AB, Vaughan ED et al (eds) 2002 Campbell's Urology, 8th edn. Philadelphia: Saunders.)
REFERRED PAIN Excessive distension of the ureter or spasm of its muscle may be caused by a stone (calculus) and provokes severe pain (ureteric colic, which is commonly, but mistakenly, called renal colic). The pain, spasmodic and agonizing, particularly if the obstruction is gradually forced down the ureter by the muscle spasm, is referred to cutaneous areas innervated from spinal segments which supply the ureter, mainly T11-L2. It shoots down and forwards from the loin to the groin and scrotum or labium majus and may extend into the proximal anterior aspect of the thigh by projection to the genitofemoral nerve (L1, 2). The cremaster, which has the same innervation, may reflexly retract the testis.
© 2008 Elsevier
MICROSTRUCTURE The wall of the ureter is composed of an external adventitia, a smooth muscle layer and an inner mucosal layer (Fig. 92.4). The last consists of the urothelium and an underlying connective tissue lamina propria.
Figure 92.4 Transverse section of ureter. The walls are muscular and are lined by a specialized urothelium. (By permission from Stevens A, Lowe JS 1996 Human Histology, 2nd edn. London: Mosby.)
The ureteric adventitial blood vessels and connective tissue fibres are orientated parallel to the long axis of the ureter. Throughout its length, the muscle coat of the ureter is fairly uniform in thickness and in cross-section measures c.750-800 µm in width. The muscle bundles which constitute this coat are frequently separated from one another by relatively large amounts of connective tissue. However, branches which interconnect muscle bundles are common and there is frequent interchange of muscle fibres between adjacent bundles. As a consequence of this extensive branching, individual muscle bundles do not spiral around the ureter, but form a complex meshwork of interweaving bundles. In addition, unlike the gut (Chapter 72), the muscle bundles are so arranged that morphologically distinct longitudinal and circular layers cannot be clearly distinguished. In the upper part of the ureter, the inner muscle bundles tend to lie longitudinally while those on the outer aspect have a circular or oblique orientation. In its middle and lower parts, there are additional outer longitudinally orientated fibres. As the ureterovesical junction is approached, the muscle coat consists predominantly of longitudinally orientated muscle bundles.
The mucosa of the ureter consists of an epithelium, the urothelium, on the deep aspect of which is a layer of subepithelial fibroelastic connective tissue lamina propria. The latter varies in thickness from 350-700 µm and is a conduit for small blood vessels and bundles of non-myelinated nerve fibres. Occasional lymphocytes may be present in the lamina propria but their aggregation into definitive lymph nodules is rare. The urothelium is usually extensively folded, giving the ureteric lumen a stellate outline.
© 2008 Elsevier
URETERIC PERISTALSIS Under normal conditions contraction waves originate in the proximal part of the upper urinary tract and are propagated in an anterograde direction towards the bladder. Atypical smooth muscle cells in the wall of the minor calyces act individually or collectively as pacemaker sites. A peristaltic wave begins at one (or possibly more) of these sites. Once initiated, the contraction is propagated through the wall of the adjacent major calyx and activates the smooth muscle of the renal pelvis. Contraction waves are propagated away from the kidney, and so undesirable pressure rises are not directed against the renal parenchyma. Since several potential pacemaker sites exist, the initiation of contraction waves is unimpaired by partial nephrectomy: the minor calyces spared by the resection remain in situ to continue their pacemaking function. Experimental evidence indicates that autonomic nerves do not play a major part in the propagation of peristalsis. It seems more likely that they play a modulatory role on the contractile events occurring in the musculature of the upper urinary tract. The most likely mechanism to account for impulse propagation is myogenic conduction as a result of the electrotonic coupling of one muscle cell to its immediate neighbours by means of intercellular 'gap' junctions. There are numerous regions of close approach between ureteric smooth muscle cells and also between both types of muscle cell in the renal pelvis and calyces. It is therefore reasonable to assume that in the upper urinary tract this type of intercellular junction may be responsible for the conduction of excitation from one myocyte to the next. UPDATE Date Added: 08 March 2005 Helen Elizabeth Wiggett, PhD (Dianthus Medical Limited) Update: Cajal-like cells in the human upper urinary tract Interstitial cells of Cajal (ICCs) play an important role in the control of gut motility. ICCs have also been identified in the upper urinary tract of animals where they may both trigger motility and be responsible for pacemaker activity in this tissue. A recent study has investigated whether ICCs exist in the human upper urinary tract by analyzing the localization and distribution of the c-kit CD117 receptor, which is found in the ICC cell membrane. Human ureter segments were obtained from 7 cadavers (mean age 54 years) and 49 patients who underwent renal tumor removal (mean age 49 years). Tissue sections were stained using the alkaline phosphatase-anti alkaline phosphatase (APAAP) technique with five monoclonal antibodies and two polyclonal antibodies to c-kit and with antibodies to common cell markers for comparison. The tissue was separated into four segments: the pyelon, the proximal ureter 5 cm distal from the pyelon, the intermediate ureter, and the distal ureter 5 cm proximal from the bladder. Expression of the c-kit receptor was detected on three cell types, one of which was subsequently identified as mast cells expressing both c-kit and CD34. Spindle cells expressing c-kit were identified throughout the ureteral wall among the septa of the inner and outer smooth muscle layers and in the lamina propria. The highest concentration of c-kit expressing cells was in the pyelon, and there was a slightly decreased concentration in the distal ureteral segments. Vertical cells expressing c-kit were
identified in the urothelium: the highest concentration was found in the intermediate section of the ureter and the lowest concentration in the pyelon. There were cytologic differences between these ICC-like cells and ICC cells found in the human duodenum. In particular, the ureteral ICC-like cells did not show a typical staghorn appearance. This study has demonstrated the presence of ICC-like cells in the human ureter. Further functional studies are necessary to determine the physiologic importance and pathologic significance of these findings. Metzger R, Schuster T, Till H, Stehr M, Franke FE, Dietz HG. Cajal-like cells in the human upper urinary tract. J Urol. 2004 Aug;172(2):769-72.
© 2008 Elsevier
93 BLADDER, PROSTATE AND URETHRA Bladder The urinary bladder is a reservoir (Figs 93.1, 93.2, 98.1). It varies in size, shape, position and relations, according to its content and the state of neighbouring viscera. When empty, it lies entirely in the lesser pelvis but as it distends it expands anterosuperiorly into the abdominal cavity. When empty, it is somewhat tetrahedral and has a base (fundus), neck, apex, a superior and two inferolateral surfaces. The base (fundus) of the bladder is triangular and posteroinferior. In females it is closely related to the anterior vaginal wall (Fig. 102.1); in males it is related to the rectum although it is separated from it above by the rectovesical pouch and below by the seminal vesicle and vas deferens on each side (Figs 93.2, 98.1). In a triangular area between the vasa deferentia, the bladder and rectum are separated only by rectovesical fascia, commonly known as Denonvillier's fascia. The inferior part of this area may be obliterated by approximation of the ampullae of the vas deferens above the prostate. The neck is the lowest region and is also the most fixed. It is 3-4 cm behind the lower part of the symphysis pubis, which is a little above the plane of the inferior aperture of the lesser pelvis. The bladder neck is the internal urethral orifice and alters little in position with varying conditions of the bladder and rectum. In males the neck rests on, and is in direct continuity with, the base of the prostate; in females it is related to the pelvic fascia, which surrounds the upper urethra.
Figure 93.1 Median sagittal section to show male internal and external genitalia, bladder. A number of structures (e.g. obturator vessels, ureter) are only faintly visible through the overlying peritoneum.
The vesical apex in both sexes faces towards the upper part of the symphysis pubis. The median umbilical ligament (urachus, Fig. 109.1) ascends behind the anterior abdominal wall from the apex to the umbilicus, covered by peritoneum to form the median umbilical fold. The triangular superior surface is bounded by lateral borders from the apex to the ureteric entrances and by a posterior border, which joins them. In males the superior surface is completely covered by peritoneum, which extends slightly onto the base and continues posteriorly into the rectovesical pouch and anteriorly into the median umbilical fold (p. 1133). It is in contact with the sigmoid colon and the terminal coils of the ileum. In females the superior surface is largely covered by peritoneum, which is reflected posteriorly onto the uterus at the level of the internal os (i.e. the junction of the uterine body and cervix), to form the vesicouterine pouch. The posterior part of the superior surface, devoid of peritoneum, is separated from the supravaginal cervix by fibroareolar tissue (p. 1133). In males, each inferolateral surface is separated anteriorly from the pubis and puboprostatic ligaments by the (potential) retropubic space. In females the relations are similar, except that the pubovesical ligaments replace the
puboprostatic ligaments. The inferolateral surfaces are not covered by peritoneum.
page 1289 page 1290
Figure 93.2 Anterior aspect of the interior of the urinary bladder.
As the bladder fills it becomes ovoid. In front it displaces the parietal peritoneum from the suprapubic region of the abdominal wall. Its inferolateral surfaces become anterior and rest against the abdominal wall without intervening peritoneum for a distance above the symphysis pubis which varies with the degree of distension, but is commonly 5-7 cm. The distended bladder may be punctured just above the symphysis pubis without traversing the peritoneum (suprapubic cystostomy): surgical access to the bladder through the anterior abdominal wall is usually by this route. The summit of the full bladder points up and forwards above the attachment of the median umbilical ligament, so that the peritoneum forms a supravesical recess of varying depth between the summit and the anterior abdominal wall: this recess often contains coils of small intestine. A distended bladder may be ruptured in lower abdominal or pelvic injuries, either extraperitoneally or, if the superior surface is involved, with tearing of the peritoneum and escape of urine into the peritoneal cavity. At birth, the bladder is relatively higher than in the adult, and the internal urethral
orifice is level with the upper symphyseal border. The bladder is then abdominal rather than pelvic, and extends about two-thirds of the distance towards the umbilicus. Urine samples may therefore be obtained in children by performing suprapubic needle puncture. The bladder progressively descends, and reaches the adult position shortly after puberty. Congenital abnormalities of the bladder are described on page 1379.
© 2008 Elsevier
LIGAMENTS OF BLADDER (Fig. 69.9) The bladder is anchored inferiorly by condensations of pelvic fascia which attach it to the pubis, lateral pelvic side-walls, and rectum. In both sexes stout bands of fibromuscular tissue extend from the bladder neck to the inferior aspect of the pubic bones. These structures are the pubovesical ligaments. They constitute the superior extensions of the pubourethral ligaments in the female or the puboprostatic ligaments in the male. The pubovesical ligaments lie on each side of the median plane, leaving a midline hiatus through which numerous small veins pass. A number of other so-called ligaments have been described in relation to the base of the urinary bladder. The reflections of the peritoneum from the bladder to the side-walls of the pelvis form the lateral ligaments, and the sacrogenital folds constitute the posterior ligaments: they are not true ligaments, but condensations of connective tissue around the major neurovascular structures. They are described as ligaments in routine clinical use. The apex of the bladder is connected to the umbilicus by the remains of the urachus, which forms the median umbilical ligament. The lumen of the lower part of the urachus may persist throughout life and communicate with the cavity of the bladder. From the superior surface of the bladder the peritoneum is carried off in a series of folds, the 'false' ligaments of the bladder. Anteriorly there are three folds, the median umbilical fold over the median umbilical ligament and two medial umbilical folds over the obliterated umbilical arteries.
© 2008 Elsevier
BLADDER INTERIOR VESICAL MUCOSA (Fig. 93.2) The vesical mucosa is attached only loosely to subjacent muscle for the most part: it folds when the bladder empties, and the folds are effaced as it fills. Over the trigone, immediately above and behind the internal urethral orifice, it is adherent to the subjacent muscle layer and always smooth (Fig. 93.2). The anteroinferior angle of the trigone is formed by the internal urethral orifice, its posterolateral angles by the ureteric orifices. The superior trigonal boundary is a slightly curved interureteric crest, which connects the two ureteric orifices and is produced by the continuation into the vesical wall of the ureteric internal longitudinal muscle. Laterally this ridge extends beyond the ureteric openings as ureteric folds, produced by the terminal parts of the ureters which run obliquely through the bladder wall. At cystoscopy the interureteric crest appears as a pale band and is a guide to the ureteric orifices in catheterization.
TRIGONE The smooth muscle of the trigone consists of two distinct layers, sometimes termed the superficial and deep trigonal muscles. The latter is composed of muscle cells, indistinguishable from those of the detrusor, and is simply the posteroinferior portion of the detrusor muscle proper. Confusion might be avoided if the term deep trigonal muscle was abandoned in favour of the more accurate term trigonal detrusor muscle. The superficial trigonal muscle represents a morphologically distinct component of the trigone, which, unlike the detrusor, is composed of relatively small diameter muscle bundles continuous proximally with those of the intramural ureters. The superficial trigonal muscle is relatively thin but is generally described as becoming thickened along its superior border to form the interureteric crest. Similar thickenings occur along the lateral edges of the superficial trigone. In both sexes the superficial trigone muscle becomes continuous with the smooth muscle of the proximal urethra, and extends in the male along the urethral crest as far as the openings of the ejaculatory ducts.
URETERIC ORIFICES The slit-like ureteric orifices are placed at the posterolateral trigonal angles (Fig. 93.2). In empty bladders they are c.2.5 cm apart, and c.2.5 cm from the internal urethral orifice; in distension these measurements may be doubled.
INTERNAL URETHRAL ORIFICE The internal urethral orifice is sited at the trigonal apex, the lowest part of the bladder, and is usually somewhat crescentic in section. There is often an elevation immediately behind it in adult males (particularly past middle age) which is caused by the median prostatic lobe, sometimes known as the uvula of the bladder.
© 2008 Elsevier
BLADDER NECK The smooth muscle of the bladder neck is histologically, histochemically and pharmacologically distinct from the detrusor muscle proper and so the bladder neck should be considered as a separate functional unit. The arrangement of smooth muscle in this region is quite different in males and females, and therefore each sex will be described separately.
FEMALE The female bladder neck consists of morphologically distinct smooth muscle. The large diameter fasciculi characteristic of the detrusor are replaced in the region of the bladder neck by small diameter fasciculi which extend obliquely or longitudinally into the urethral wall. In the normal female the bladder neck sits above the pelvic floor supported predominantly by the pubovesical ligaments, the endopelvic fascia of the pelvic floor and levator ani. These support the urethra at rest; with elevated intraabdominal pressure the levators contract, increasing urethral closure pressure to maintain continence. This anatomical arrangement commonly alters after parturition and with increasing age, such that the bladder neck lies beneath the pelvic floor, particularly when the intra-abdominal pressure rises: the mechanism described above then fails to maintain continence (stress incontinence as a result of urethral hypermobility).
MALE In the male, the bladder neck is completely surrounded by a circular collar of smooth muscle which extends distally to surround the preprostatic portion of the urethra. Because of the location and orientation of its constituent fibres, the term preprostatic sphincter is suitable for this particular component of urinary tract smooth muscle. This is a genital sphincter mechanism with a well-defined adrenergic innervation, which ensures anterograde ejaculation by closing the bladder neck during seminal emission. Distally, this muscle merges with and becomes indistinguishable from the musculature in the stroma and capsule of the prostate gland. Whether this preprostatic sphincter replaces, or is additional to, the bladder neck muscle pattern seen in the female is unclear, but it is probably additional.
BLADDER OUTFLOW OBSTRUCTION In progressive chronic obstruction to micturition, e.g. by prostatic enlargement or urethral stricture, bladder muscle hypertrophies. The muscle fasciculi increase in size and, because they interlace in all directions, a thick-walled 'trabeculated bladder' is produced. Mucosa between the fascicles forms 'diverticula'. When outflow is thus obstructed, emptying is not complete: some urine remains and may become infected, and infection may ascend to the kidneys. Back pressure from a chronically distended bladder may gradually dilate the ureters and renal pelves (so-called 'hydronephrosis') and even the renal collecting tubules, which can result in progressive renal impairment.
© 2008 Elsevier
VASCULAR SUPPLY AND LYMPHATIC DRAINAGE ARTERIES (Fig. 108.4) The bladder is supplied principally by the superior and inferior vesical arteries, derived from the anterior trunk of the internal iliac artery, supplemented by the obturator and inferior gluteal arteries. In the female additional branches are derived from the uterine and vaginal arteries. Superior vesical artery
The superior vesical artery supplies many branches to the fundus of the bladder. The artery to the vas deferens often originates from one of these and accompanies the vas deferens to the testis, where it anastomoses with the testicular artery. Other branches supply the ureter. The beginning of the superior vesical artery is the proximal, patent section of the fetal umbilical artery. Inferior vesical artery
The inferior vesical artery often arises with the middle rectal artery from the internal iliac artery. It supplies the base of the bladder, prostate, seminal vesicles and lower ureter. Prostatic branches communicate across the midline. The inferior vesical artery may sometimes provide the artery to the vas deferens.
VEINS (Fig. 93.3) The veins which drain the bladder form a complicated plexus on its inferolateral surfaces and pass backwards in the lateral ligaments of the bladder to end in the internal iliac veins.
LYMPHATIC DRAINAGE (Fig. 108.6)
Figure 93.3 The veins of the right half of the male pelvis. (After Spalteholtz W 1924 Die Arterien der Herzwand. Anatomische Untersuchungen an Menschen and Tieren. Leipzig: Hirzel.)
Lymphatics which drain the bladder begin in mucosal, intermuscular and serosal plexuses. There are three sets of collecting vessels, most of which end in the external iliac nodes. Vessels from the trigone emerge on the exterior of the bladder to run superolaterally. Vessels from the superior surface of the bladder converge to the posterolateral angle and pass superolaterally to the external iliac nodes (one may go to the internal or common iliac group). Vessels from the inferolateral surface of the bladder ascend to join those from the superior surface or run to the lymph nodes in the obturator fossa. Minute nodules of lymphoid tissue may occur along the vesical lymph vessels.
Innervation The nerves supplying the bladder arise from the pelvic plexuses, which are a mesh of autonomic nerves and ganglia on the lateral aspects of the rectum,
internal genitalia and bladder base. They consist of both sympathetic and parasympathetic components, each of which contains both efferent and afferent fibres. The innervation of the bladder has been reviewed in some detail by Mundy (1999).
EFFERENT FIBRES Parasympathetic fibres arise from the second to the fourth sacral segments of the spinal cord and enter the pelvic plexuses on the posterolateral aspects of the rectum as the pelvic splanchnic nerves or nervi erigentes. The sympathetic fibres are derived from the lower three thoracic and upper two lumbar segments of the spinal cord. These form the coeliac and mesenteric plexuses around the great vessels in the abdomen from which the hypogastric plexuses descend into the pelvis as fairly discrete nerve bundles within the extraperitoneal connective tissue posterior to the ureter on each side. The anterior part of the pelvic plexus is known as the vesical plexus. Small groups of autonomic neurones occur within the plexus and throughout all regions of the bladder wall. These multipolar intramural neurones are rich in acetylcholinesterase (AChE) and occur in ganglia consisting of up to 20 nerve cell bodies. Numerous preganglionic autonomic fibres form both axosomatic and axodendritic synapses with the ganglionic neurones. The majority of the preganglionic nerve terminals correspond morphologically to presumptive cholinergic fibres. Noradrenergic terminals also relay on cell bodies in the pelvic plexus: it is not known whether similar nerves synapse on intramural bladder ganglia. The urinary bladder (including the trigonal detrusor muscle) is profusely supplied with nerves which form a dense plexus among the detrusor muscle cells. The majority of these nerves contain AChE and occur in abundance throughout the muscle coat of the bladder. Axonal varicosities adjacent to detrusor muscle cells possess features which are considered to typify cholinergic nerve terminals and contain clusters of small (50 nm diameter) agranular vesicles together with occasional large (80-160 nm diameter) granulated vesicles and small mitochondria. Terminal regions approach to within 20 nm of the surface of the muscle cells and may be partially surrounded by Schwann cell cytoplasm, or more often are naked nerve endings. The human detrusor muscle possesses a sparse supply of sympathetic noradrenergic nerves which generally accompany the vascular supply and only rarely extend among the myocytes. Nonadrenergic, noncholinergic nerves have been identified, and a number of other neurotransmitters or neuromodulators have been detected in intramural ganglia, including the peptide somatostatin. The superficial trigonal muscle is associated with more noradrenergic (sympathetic) fibres than cholinergic (parasympathetic) nerves. This difference supports the view that the superficial trigonal muscle should be regarded as 'ureteric' rather than 'vesical' in origin. However it must be emphasized that the superficial trigonal muscle forms a very minor part of the total muscle mass of the bladder neck and proximal urethra in either sex and is probably of little significance in the physiological mechanisms which control these regions. page 1291 page 1292
The smooth muscle of the bladder neck in males is predominantly orientated obliquely or circularly and is sparsely supplied with cholinergic (parasympathetic) nerves but possesses a rich noradrenergic (sympathetic) innervation. A similar distribution of autonomic nerves also occurs in the smooth muscle of the prostate gland, seminal vesicles and vasa deferentia. Stimulation of sympathetic nerves causes contraction of smooth muscle in the wall of the genital tract resulting in seminal emission. Concomitant sympathetic stimulation of the proximal urethral smooth muscle causes sphincteric closure of the preprostatic sphincter, thereby preventing reflux of ejaculate into the bladder. Although this genital function of the bladder neck of the male is well established, it is not known whether the smooth muscle of this region plays an active role in maintaining urinary continence. In contrast, the smooth muscle of the bladder neck of the female receives relatively few noradrenergic nerves but is richly supplied with presumptive cholinergic fibres. The sparse supply of sympathetic nerves presumably relates to the absence of a functioning 'genital' portion of the wall of the female urethra. The lamina propria of the fundus and inferolateral walls of the bladder is virtually devoid of autonomic nerve fibres, apart from some noradrenergic and occasional presumptive cholinergic perivascular nerves. However, as the urethral orifice is approached, the density of nerves unrelated to blood vessels increases. At the bladder neck and trigone a nerve plexus extends throughout the lamina propria. The constituent nerves are cholinesterase positive and run through the connective tissue independent of blood vessels. Some of the larger diameter axons are myelinated and others lie adjacent to the basal urothelial cells. As in the ureter, the subepithelial nerve plexus of the bladder is assumed to subserve a sensory function in the absence of any obvious effector target sites.
AFFERENT FIBRES Vesical nerves are also concerned with pain and awareness of distension and are stimulated by distension or spasm due to a stone, inflammation or malignant disease; they travel in sympathetic and parasympathetic nerves, predominantly the latter. Division of the sympathetic paths (e.g. 'presacral neurectomy'), or of the superior hypogastric plexus, therefore does not materially relieve vesical pain, whereas considerable relief follows bilateral anterolateral cordotomy. Since nerve fibres mediating awareness of distension travel in the posterior columns (fasciculus gracilis), the patient still retains awareness of the need to micturate after anterolateral cordotomy. The nerve endings detecting noxious stimuli are probably of more than one type: a subepithelial plexus of fibres containing dense vesicles, probably afferent endings, has been described.
MECHANISM AND CONTROL OF MICTURITION (Fig. 93.4) Micturition consists of a storage phase and a voiding phase. During the storage phase the bladder accommodates an increasing volume of urine without any change in intravesical pressure. This is partly because of its viscoelastic properties, and partly because a gateing mechanism in the spinal cord reflexly inhibits preganglionic parasympathetic (efferent) activity, and a similar mechanism in the pelvic ganglia prevents the transmission of preganglionic activity to postganglionic parasympathetic neurones until preganglionic activity reaches a
threshold level (giving an all-or-none effect). These properties are augmented by sufficient activity of the distal urethral sphincter to maintain urethral closure. This activity is controlled centrally by a storage centre within the rostral pons (called the L-centre because of its lateral location, p. 349). Mean bladder capacity in male adults varies around 400 ml. Micturition commonly occurs at smaller volumes. Filling to c.500 ml may be tolerated, but beyond this pain is caused by tension in the wall, leading to the urgent desire to micturate. Pain is referred to the cutaneous areas supplied by spinal segments supplying the bladder (T10-L2, S24), including the lower anterior abdominal wall, perineum and penis. With threshold afferent stimulation, efferent impulses from the micturition centre in the rostral pons (also called the M-centre because of its relative medial location) (p. 349). activate descending spinal pathways to the intermediolateral grey column of the second, third and fourth sacral spinal segments where the cell bodies of the preganglionic parasympathetic nerves are located. Their axons run to the pelvic (inferior hypogastric) plexuses as the pelvic splanchnic nerves, and synapse on postganglionic parasympathetic neurones in ganglia within the plexuses and within the wall of the bladder. Postganglionic fibres ramify throughout the thickness of the detrusor smooth muscle coat. The profuse distribution of these motor nerves emphasizes the importance of the parasympathetic nervous system in initiating and sustaining bladder contraction during micturition. Just before the onset of voiding, the distal urethral sphincter is relaxed by central inhibition of its motor neurones (which are also located in the second, third and fourth sacral spinal segments), by the same nerve pathway which activates the preganglionic parasympathetic nerves. Activation of the parasympathetic innervation of the bladder causes the release of acetylcholine. This activates muscarinic receptors in the detrusor layer of the bladder wall and this causes bladder contraction. Relaxation of the urethra is currently believed to be due to the release of nitric oxide from these same parasympathetic nerves in the bladder neck and urethra. The central integration of the nervous control of the bladder and urethra is essential for normal micturition. UPDATE Date Added: 20 December 2005 Abstract: Cortical representation of the urge to void: a functional magnetic resonance imaging study. Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve=pubmed=Abstract_uids=16145475_hl=3 Cortical representation of the urge to void: a functional magnetic resonance imaging study. KuhtzBuschbeck JP, van der Horst C, Pott C et al: J Urol 174(4 Pt. 1):1477-1481, 2005. UPDATE Date Added: 20 December 2005 Abstract: An fMRI study of the role of suprapontine brain structures in the voluntary voiding control induced by pelvic floor contraction. Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve=pubmed=Abstract_uids=15588608_hl=5 An fMRI study of the role of suprapontine brain structures in the voluntary voiding control induced by pelvic floor contraction. Zhang H, Reitz A, Kollias S et al: Neuroimage 24(1):174-180,
2005. UPDATE Date Added: 13 December 2005 Abstract: Brain control of normal and overactive bladder. Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=16217325&query_hl=6 Brain control of normal and overactive bladder. Griffiths D, Derbyshire S, Stenger A et al. J Urol. 2005 Nov;174(5):1862-7, 2005. UPDATE Date Added: 07 September 2005 Abstract Abstract: Pathophysiology of adult urinary incontinence. Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=14978635&query_hl=11 Pathophysiology of adult urinary incontinence. Delancey JO, Ashton-Miller JA: Gastroenterology. Jan;126 (1 Suppl 1):S23-32, 2004.
© 2008 Elsevier
MICROSTRUCTURE (Fig. 93.4) The wall of the urinary bladder consists of three layers, an outer adventitial layer of soft connective tissue (which in some regions possesses a serosal covering of peritoneum), a smooth muscle coat (the detrusor muscle), and an inner mucosal layer which lines the interior of the bladder. The serous layer is restricted to the superior and, in males, part of the posterior, surfaces of the bladder. It consists of mesothelium and underlying connective tissue as elsewhere in the peritoneum. The detrusor muscular layer is composed of relatively large diameter interlacing bundles of smooth muscle cells arranged as a complex meshwork. Three ill-defined layers are present and arranged in such a way that longitudinally orientated muscle bundles predominate on the inner and outer aspects of a substantial middle circular layer. Posteriorly, some of the outer longitudinal bundles pass over the bladder base and fuse with the capsule of the prostate or with the anterior vaginal wall. Other bundles extend to the anterior aspect of the rectum as the rectovesical muscle. Anteriorly some of the outer longitudinal bundles continue into the pubovesical ligaments and contribute to the muscular component of these structures. As in the muscle coat of the ureter, exchange of fibres between adjacent muscle bundles frequently occurs within the bladder wall. Functionally, therefore, the detrusor acts as a single unit of interlacing smooth muscle. The mucosa has a structure similar to that of the ureters. The mucosal lamina propria forms a relatively thick layer, varying in depth from 500 µm in the fundus and inferolateral walls to c.100 µm in the trigone. Small bundles of smooth muscle cells form an incomplete and rudimentary muscularis mucosae. An extensive network of blood vessels is present throughout the lamina propria and supplies a plexus of thin-walled fenestrated capillaries which lie in grooves at the base of the urothelium. In addition to the urothelium, the epithelium of the bladder neck and trigone also contains a cell which is characterized by the presence of numerous large membrane-bound vesicles each containing a central dense granule: these cells belong to the dispersed neuroendocrine system (p. 180). Several morphological variations have been described in the mucosa of the bladder. Since they occur in otherwise normal healthy adults, they are not considered to represent pathological conditions. One of the commonest epithelial variants found in bladder biopsy samples or at postmortem are so-called Von Brunn's nests, proliferations of morphologically normal basal urothelial cells which project into the underlying connective tissue of the lamina propria and are particularly frequent in the trigone. They may undergo central involution which results in areas of oedematous-looking mucosa termed cystitis cystica. Mucussecreting glands with single or branched ducts are frequently observed in the bladder mucosa, and are particularly numerous near the ureteric and internal urethral orifices. Non-keratinizing squamous metaplasia of the vaginal type frequently occurs in the urinary bladder mucosa, especially over the trigone, commonly in adult females, and rarely in males. It is of no pathological significance when confined to the trigone. UPDATE Date Added: 10 January 2006 Publication Services, Inc. Update: Specialized interstitial cells of the urinary tract: an assessment of current knowledge. Interstitial cells of the urinary tract are similar to the interstitial cells of Cajal (ICC) that occur in the digestive tract. Urinary interstitial cells stain immunopositive for Kit, a tyrosine kinase receptor in ICC, and for vimentin, an intermediate filament expressed by ICC but not by smooth muscle cells. ICC-like cells in the urethra possess slow wave activity mediated via ryanodinesensitive calcium-activated chloride conductance channels. Although a direct coupling of these cells to smooth muscle activity has not been observed, periodic depolarization of smooth muscle cells may help maintain urethral tone and urinary continence.
In the bladder, densely innervated specialized cells form structural networks in the detrusor layers. Spontaneous, but poorly synchronized, parasympathetic activation of smooth muscle bundles maintains bladder tone and shape, and coordinated contraction facilitates voiding. Novel cell types, probably having both afferent and efferent innervation, occur beneath the urothelium. They stain positive for both Kit and vimentin and possess many of the ultrastructural properties that characterize ICCs, including frequent connexin-43-positive gap junctions. Atypical smooth muscle cells of the upper urinary tract contain contractile filaments and are the pacemaker cells of the proximal renal pelvis. ICC-like cells just distal to the upper urinary tract conduct impulses from the pacemaker cells to smooth muscle cells. Kit-positive cells are found in the inner and outer smooth muscle layers and in the lamina propria; their numbers decrease from the proximal to the distal ureteral segments. Similar cells may also occur in the uretero-pelvic junction. Kit-positive, connexin-43-positive cells are also present in the prostate gland where they may play a role in generating the contractions for flow of prostatic fluid from the acini to the ducts. Similar cells have been found in the corpora cavernosa and vas deferens and may function in semen transport. Urinary tract ICCs, phenotypically different from smooth muscle cells, may be promising targets for therapeutic manipulation in the treatment of urinary tract and sexual dysfunction. Brading AF, McCloskey KD: Mechanisms of disease: specialized interstitial cells of the urinary tract-an assessment of current knowledge. Nature Clinical Practice Urology 2:546-554, 2005. page 1292 page 1293
page 1293 page 1294
Figure 93.4 Control of micturition. The micturition control centre is in the paramedian pontine reticular formation on each side. It comprises a medially placed micturition centre, 'M' and laterally placed storage centre, 'L'. Neurones project from the 'M' centre and to the storage centre 'L' to parasympathetic neurones in segments 2-4 of the sacral spinal cord, and to Onuf's nucleus, which is in the same segments, and which innervates the external urethral sphincter. At higher levels, neurones in the right prefrontal and anterior cingulate cortex, right preoptic nucleus and periaqueductal grey matter, are involved in the control of micturition. Vesical afferents from stretch
receptors in the detrusor and trigonal mucosa relay the extent of bladder filling to the brain stem and thalamus via spinoreticulothalamic fibres (1). Activity in the sympathetic system that maintains bladder compliance increases (via ! 2 receptors detrusor fibres) and parasympathetic activity is inhibited (2). Spinoreticular fibres synapsing in the 'L' nucleus in the pons activate Onuf's nucleus to increase the tone of the external sphincter (3). If micturition is deferred, fibres projecting from the inferior frontal gyrus inhibit the right anterior cingulate gyrus, preoptic area and periaqueductal grey matter) (4). Voluntary contraction of the pelvic floor musculature, controlled by the prefrontal cortex driving the perineal 'area' of the motor cortex (5) cannot be long sustained once filling is complete. (By permission from FitzGerald MJT, Folan-Curren J 2001 Clinical Neuroanatomy, 4th edn. London: Saunders.)
REFERENCES Klutke CG, Siegel CL 1995 Functional female pelvic anatomy. Urol Clin North Am 22(3): 487-98. Medline Similar articles Mundy AR, Fitzpatrick J, Neal D, George N 1999 Structure and function of the lower urinary tract. In: The Scientific Basis of Urology, Chapter 11. Oxford: Isis Medical Media: 217-42. Explains the neurological components of bladder function from the higher cortical centres to molecular events within the cells of the detrusor muscle.
© 2008 Elsevier
94 BLADDER, PROSTATE AND URETHRA Male urethra The male urethra is18-20 cm long and extends from the internal orifice in the urinary bladder to the external opening, or meatus, at the end of the penis. It may be considered in two parts (Figs 93.1, 94.1). The relatively long anterior urethra (c.16 cm long) lies within the perineum (proximally) and the penis (distally) surrounded by the corpus spongiosum and is functionally a conduit. The relatively short posterior urethra (4 cm) lies in the pelvis proximal to the corpus spongiosum and is acted upon by the urogenital sphincter mechanisms and also acts as a conduit. The anterior urethra is subdivided into a proximal component, the bulbar urethra, which is surrounded by the bulbospongiosus and is entirely within the perineum, and a pendulous or penile component, which continues on to the tip of the penis. The posterior urethra is divided into preprostatic, prostatic, and membranous segments.
Figure 94.1 The whole length of the lumen of the male urethra exposed by an incision extending into it from its dorsal aspect. Note openings of prostatic utricle and ejaculatory ducts on the colliculus seminalis (verumontanum).
In the flaccid penis, the urethra as a whole presents a double curve (Fig. 93.1). Except during the passage of fluid along it, the urethral canal is a mere slit: in transverse section, the slit is transversely arched in the prostatic part, in the preprostatic and membranous portions it is stellate, in the bulbar and penile portions it is transverse, while at the external orifice it is sagittal.
© 2008 Elsevier
PREPROSTATIC PART The preprostatic urethra is c.1-1.5 cm in length, extending almost vertically from the bladder neck to the superior aspect of the verumontanum. In addition to the smooth muscle bundles which run in continuity from the bladder neck down to the prostatic urethra, and distinct from the smooth muscle within the prostate, smooth muscle bundles surround the bladder neck and preprostatic urethra: they are arranged as a distinct circular collar which has its own distinct adrenergic innervation. The bundles which form this 'preprostatic sphincter' are small in size compared with the muscle bundles of the detrusor and are separated by a relatively larger connective tissue component rich in elastic fibres. Unlike the detrusor and the rest of the urethral smooth muscle (common to both sexes), the preprostatic sphincter is almost totally devoid of parasympathetic cholinergic nerves but is richly supplied with sympathetic noradrenergic nerves. Contraction of the preprostatic sphincter serves to prevent the retrograde flow of ejaculate through the proximal urethra into the bladder, and can maintain continence when the external sphincter has been damaged. It is extensively disrupted in the vast majority of men undergoing bladder neck surgery, e.g. transurethral resection of the prostate, which results in retrograde ejaculation.
© 2008 Elsevier
PROSTATIC PART The prostatic urethra is c.3-4 cm in length and tunnels through the substance of the prostate closer to the anterior than the posterior surface of the gland. It is continuous above with the preprostatic part and emerges from the prostate slightly anterior to its apex (the most inferior point of the prostate). The urethra turns anteriorly as it passes through the prostate making an angle of c.35°. Throughout most of its length the posterior wall possesses a midline ridge, the urethral crest, which projects into the lumen causing it to appear crescentic in transverse section. On each side of the crest there is a shallow depression, termed the prostatic sinus, the floor of which is perforated by the orifices of c.15-20 prostatic ducts. An elevation, the verumontanum (colliculus seminalis), at about the middle of the length of the urethral crest, contains the slit-like orifice of the prostatic utricle. On both sides of, or just within, this orifice are the two small openings of the ejaculatory ducts. The prostatic utricle is a cul-de-sac c.6 mm long, which runs upwards and backwards in the substance of the prostate behind its median lobe. Its walls are composed of fibrous tissue, muscular fibres and mucous membrane; the latter is pitted by the openings of numerous small glands. The prostatic utricle develops from the paramesonephric ducts or urogenital sinus, and is thought to be homologous with the vagina of the female (p. 1385). It is sometimes called the 'vagina masculina', but the more usual view is that it is a uterine homologue and hence the term 'utricle'. The lowermost part of the prostatic urethra is fixed by the puboprostatic ligaments and is therefore immobile.
© 2008 Elsevier
MEMBRANOUS PART page 1295 page 1296
The membranous part of the urethra is the shortest (c.1.5 cm), least dilatable and, with the exception of the external orifice, the narrowest, section of the urethra. It descends with a slight ventral concavity from the prostate to the bulb of the penis (Fig. 93.1), passing through the perineal membrane, c.2.5 cm posteroinferior to the pubic symphysis. The wall of the membranous urethra consists of a muscle coat, separated from the epithelial lining by a narrow layer of fibroelastic connective tissue. The muscle coat consists of a relatively thin layer of bundles of smooth muscle, which are continuous proximally with those of the prostatic urethra, and a prominent outer layer of circularly orientated striated muscle fibres, which form the external urethral sphincter, as it is commonly known. The bladder neck is sometimes called the proximal sphincter mechanism, to distinguish it from the distal sphincter mechanism. This latter term has considerable value because it recognizes that the sphincter-active membranous urethra consists of several components, namely, urethral smooth muscle; urethral striated muscle (rhabdosphincter), which is the most important component; and the periurethral part of levator ani, which is important to resist surges of intra-abdominal pressure (e.g. on coughing or exercise). The external sphincter represents the point of highest intraurethral pressure in the normal, contracted state. The intrinsic striated muscle component is devoid of muscle spindles. The striated muscle fibres themselves are unusually small in cross-section (15-20 µm diameter), and are physiologically of the slow twitch type, unlike the pelvic floor musculature, which is a heterogeneous mixture of slow and fast twitch fibres of larger diameter. The slow twitch fibres of the external sphincter are capable of sustained contraction over relatively long periods of time and actively contribute to the tone, which closes the urethra and maintains urinary continence. The urethral sphincter mechanism is described on page 1367.
© 2008 Elsevier
ANTERIOR PART The anterior or spongiose part of the urethra lies within the corpus spongiosum penis (p. 1315). It is c.15 cm long when the penis is flaccid and extends from the end of the membranous urethra to the external urethral orifice on the glans penis. It starts below the perineal membrane at a point anterior to the lowest level of the symphysis. This part of the anterior urethra is surrounded by bulbospongiosus and is called the bulbar urethra. It is the widest part of the urethra. The bulbourethral glands open into this section of the urethra c.2.5 cm below the perineal membrane. From here, when the penis is flaccid, the urethra curves downwards as the penile urethra. It is a narrow, transverse slit when empty, and has a diameter of c.6 mm when passing urine. It is dilated at its termination within the glans penis where it is known as the navicular fossa. The external urethral orifice is the narrowest part of the urethra, and is a sagittal slit, c.6 mm long, bounded on each side by a small labium. The epithelium of the urethra, particularly in the bulbar and distal penile segments, presents the orifices of numerous small mucous glands and follicles situated in the submucous tissue and named the urethral glands. It also contains a number of small pit-like recesses, or lacunae, of varying sizes whose orifices are directed forwards. One, larger than the rest, the lacuna magna, is situated on the roof of the navicular fossa.
© 2008 Elsevier
TRAUMATIC INJURY TO THE URETHRA The urethra may be ruptured by a fall-astride (or straddle) injury to the bulbar urethra in the perineum, or by an injury related to a pelvic fracture. These injuries usually affect the junction of the membranous with the bulbar segments across the perineal membrane. One complication of such injuries is extravasation of urine. After an injury to the bulbar urethra, urine usually extravasates between the perineal membrane and the membranous layer of the superficial fascia (clinically known as Colles' fascia). As both of these are attached firmly to the ischiopubic rami, extravasated fluid cannot pass posteriorly because the two layers are continuous around the superficial transverse perineal muscles. Laterally, the spread of urine is blocked by the pubic and ischial rami. Urine cannot enter the lesser pelvis through the perineal membrane if this remains intact, so it makes its way anteriorly into the loose connective tissue of the scrotum and penis and thence to the anterior abdominal wall. If the posterior urethra is injured, urine is extravasated into the pelvic extraperitoneal tissue: if the perineal membrane is torn as well extravasation may also occur into the perineum.
© 2008 Elsevier
CONGENITAL ANOMALIES OF THE URETHRA Hypospadias, found in c.1 in 300 boys, most often results in the urethra opening in the distal penis; sometimes it opens on the ventral aspect of the penis or more proximally up to the perineum. There is also an associated abnormality of the prepuce, which is longer dorsally and lacking ventrally, and often an associated chordee which causes a ventral curvature of the penis. It is important that this anomaly be identified prior to circumcision, because the abnormal foreskin is sometimes used for surgical correction of the deformity. Posterior urethral valves occur in 1 in 5000 to 8000 males and the most common cause of urinary outflow obstruction in male infants. The commest type (Type I) are believed to occur if the Wolffian ducts open too anteriorly onto the primitive prostatic urethra. This abnormal migration of the ducts leaves behind thick vestigial tissue that forms rigid valve cusps extending caudally from the verumontanum. Very rarely urethral duplication is seen. When present the two urethrae almost invariably lie on top of each other rather than side by side. It is possible that one of the urethrae, most likely the more dorsal, may be blind ending.
© 2008 Elsevier
VASCULAR SUPPLY AND LYMPHATIC DRAINAGE URETHRAL ARTERY The urethral artery arises from the internal pudendal artery or common penile artery just below the perineal membrane and travels through the corpus spongiosum, to reach the glans penis. It supplies the urethra and erectile tissue around it. In addition, the urethra is supplied by the dorsal penile artery, via its circumflex branches on each side and, retrograde from the glans, by its terminal branches. The blood supply through the corpus spongiosum is so plentiful that the urethra can be divided without compromising its vascular supply.
VEINS The venous drainage of the anterior urethra is to the dorsal veins of the penis and internal pudendal veins, which drain to the prostatic plexus. The posterior urethra drains into the prostatic and vesical venous plexuses, which drain into the internal iliac veins.
LYMPHATIC DRAINAGE Vessels from the posterior urethra pass mainly to the internal iliac nodes; a few may end in the external iliac nodes. Vessels from the membranous urethra accompany the internal pudendal artery. Vessels from the anterior urethra accompany those of the glans penis, ending in the deep inguinal nodes. Some may end in superficial nodes; others may traverse the inguinal canal to end in the external iliac nodes.
© 2008 Elsevier
INNERVATION The prostatic plexus supplies the smooth muscle of the prostate and prostatic urethra. It is derived from the pelvic plexus on each side and lies on the posterolateral aspect of the seminal vesicle and prostate on each side. Lesser cavernous nerves pierce the bulb of the corpus spongiosum proximally to supply the penile urethra. The greater cavernous nerves carry the sympathetic supply, which causes contraction of the preprostatic sphincter during ejaculation and prevents reflux of ejaculate into the bladder. The parasympathetic preganglionic fibres come from the second to fourth sacral segments. The nerve supply of the intrinsic muscle striated component (or rhabdosphincter) of the distal sphincter mechanism (or external sphincter) is controversial. It is generally believed to be supplied by neurones in Onuf's nucleus and by perineal branches of the pudendal nerve lying on the perineal aspect of the pelvic floor. In both instances the axons arise from neurones in S2, 3 and 4. Fibres from Onuf's nucleus (which is somatic) travel with the pelvic plexus on each side until they branch off and run on the pelvic aspect of the pelvic floor to enter the membranous urethra.
© 2008 Elsevier
MICROSTRUCTURE The epithelium lining the preprostatic urethra and the proximal part of the prostatic urethra is a typical urothelium. It is continuous with that lining the bladder, and with the epithelium lining the ducts of the prostate and bulbourethral glands, the seminal vesicles, and the vasa deferentia and ejaculatory ducts. These relationships are important in the spread of urinary tract infections. Below the openings of the ejaculatory ducts the epithelium changes to a pseudostratified or stratified columnar type, which lines the membranous urethra and the major part of the penile urethra. Mucus-secreting cells are common throughout this epithelium and frequently occur in small clusters in the penile urethra. Branching tubular paraurethral glands secrete protective mucus onto the urethral epithelial lining and are especially numerous on its dorsal aspect. In older men many of the deep recesses of the urethral mucosa contain concretions similar to those found within prostatic glands (p. 1302). Towards the distal end of the penile urethra the epithelium changes once again, becoming stratified squamous in type with well-defined connective tissue papillae. This epithelium also lines the navicular fossa and becomes keratinized at the external meatus. The epithelial cells lining the navicular fossa are glycogen-rich. This may provide a substrate for commensal lactobacilli which, as in the female vagina (p. 1353), provide a defence against pathogenic organisms. REFERENCE Chancellor MB, Yoshimura N 2002 Physiology and pharmacology of the bladder and urethra. In: Walsh PC et al (eds) Campbell's Urology Study Guide, 2nd edn. Philadelphia: Saunders: Chapter 23.
UPDATE Date Added: 30 August 2005 Abstract Abstract: The structure and innervation of the male urethra. Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=15817107&query_hl=11 The structure and innervation of the male urethra: histological and immunohistochemical studies with three-dimensional reconstruction. Karam I, Moudouni S, Droupy S et al: J Anat. 206(4):395-403, 2005. page 1297 page 1298
© 2008 Elsevier
95 BLADDER, PROSTATE AND URETHRA Female urethra The female urethra is c.4 cm long and 6 mm in diameter. It begins at the internal urethral orifice of the bladder, approximately opposite the middle of the symphysis pubis, and runs anteroinferiorly behind the symphysis pubis, embedded in the anterior wall of the vagina. It crosses the perineal membrane and ends at the external urethral orifice as an anteroposterior slit with rather prominent margins situated directly anterior to the opening of the vagina and c.2.5 cm behind the glans clitoridis (Fig. 108.10). It sometimes opens into the anterior vaginal wall. Except during the passage of urine, the anterior and posterior walls of the urethra are in apposition and the epithelium is thrown into longitudinal folds, one of which, on the posterior wall of the canal, is termed the urethral crest. Many small mucous urethral glands and minute pit-like recesses or lacunae open into the urethra. On each side, near the lower end of the urethra, a number of these glands are grouped together and open into a duct, named the para-urethral duct: each duct runs down in the submucous tissue and ends in a small aperture on the lateral margin of the external urethral orifice.
© 2008 Elsevier
VASCULAR SUPPLY, LYMPHATIC DRAINAGE AND INNERVATION Arteries The blood supply to the female urethra is derived from the vesical and vaginal arteries, principally the latter. Veins The venous plexus around the urethra drains into the vesical venous plexus around the bladder neck, and into the internal pudendal veins. An erectile plexus of veins along the length of the urethra is continuous with the erectile tissue of the vestibular bulb. Lymphatic drainage The urethral lymphatics drain into the internal and external iliac nodes. Innervation Parasympathetic preganglionic fibres from the second to fourth segments of the sacral spinal cord run in the pelvic splanchnic nerves and synapse in the vesical plexus in or near the bladder wall (p. 1291): postganglionic fibres are distributed to the smooth muscle of the urethral wall. Somatic fibres to the striated muscle are derived from the same sacral segments, and run in the pelvic splanchnic nerves but do not synapse in the vesical plexus. Sensory fibres run in the pelvic splanchnic nerves to the second to fourth segments of the sacral spinal cord. Postganglionic sympathetic fibres arise from the plexus around the vaginal arteries.
© 2008 Elsevier
MICROSTRUCTURE (Fig. 95.1) The wall of the female urethra consists of an outer muscle coat and an inner mucosa, which lines the lumen and is continuous with that of the bladder. The muscle coat consists of an outer sheath of striated muscle (external urethral sphincter or distal sphincter mechanism together with an inner coat of smooth muscle fibres. The female external urethral sphincter is anatomically separate from the adjacent periurethral striated muscle of the anterior pelvic floor. The fibres of the sphincter are arranged like a signet ring. They form a sleeve which is thickest anteriorly in the middle one-third of the urethra, and is relatively deficient posteriorly. The striated muscle extends into the anterior wall of both the proximal and distal thirds of the urethra but is deficient posteriorly in these regions. The muscle cells forming the external urethral sphincter are all small diameter, slow twitch, fibres. The smooth muscle coat extends throughout the length of the urethra and consists of slender muscle bundles, the majority of which are orientated obliquely or longitudinally. A few circularly arranged muscle fibres occur in the outer aspect of the non-striated muscle layer and intermingle with the skeletal muscle fibres forming the inner part of the external urethral sphincter. Proximally the urethral smooth muscle extends as far as the bladder neck where it is replaced by fascicles of detrusor smooth muscle. This region in the female lacks a welldefined circular smooth muscle component comparable with the preprostatic sphincter of the male. Distally, urethral smooth muscle bundles terminate in the subcutaneous adipose tissue surrounding the external urethral meatus.
page 1299 page 1300
Figure 95.1 Transverse section through the female urethra. The lumen (centre field) is lined by a stratified squamous non-keratinizing epithelium, continuous with the keratinized epithelium of the genital vestibule (below) into which it opens. The urethral mucosa is surrounded by a circular smooth muscular layer. (By kind permission of Anthony R Mundy.)
The smooth muscle of the female urethra receives an extensive presumptive cholinergic parasympathetic nerve supply, but contains relatively few noradrenergic nerves. In the absence of an anatomical sphincter, competence of the female bladder neck and proximal urethra is unlikely to be totally dependent on smooth muscle activity, and is more probably related to the support provided by the ligamentous structures which surround them. The innervation and longitudinal orientation of most of the muscle fibres suggest that urethral smooth muscle in the female is active during micturition, serving to shorten and widen the urethral lumen. The mucosa lining the female urethra consists of a stratified epithelium and a supporting lamina propria of loose fibroelastic connective tissue. The latter is bulky and well-vascularized, and contains numerous thin-walled veins. Its abundant elastic fibres are orientated both longitudinally and circularly around the urethra. The lamina propria contains a fine nerve plexus, believed to be derived from sensory branches of the pudendal nerves. The proximal part of the urethra is lined by urothelium, identical in appearance to that of the bladder neck. Distally the epithelium changes into a non-keratinizing stratified squamous type which lines the major portion of the female urethra. This epithelium is keratinized at the external urethral meatus where it becomes continuous with the skin of the vestibule. REFERENCE Chancellor MB, Yoshimura N 2002 Physiology and pharmacology of the bladder and urethra. In: Walsh PC et al (eds) Campbell's Urology Study Guide, 2nd edn. Philadelphia; Saunders: Chapter 23.
UPDATE Date Added: 07 September 2005 Abstract Abstract: Neuroanatomy of the human female lower urogenital tract. Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=15201770&query_hl=9 Neuroanatomy of the human female lower urogenital tract. Yucel S, De Souza A Jr, Baskin LS: J Urol. Jul;172(1):191-5, 2004. UPDATE Date Added: 30 August 2005 Abstract Abstract: An anatomical description of the male and female urethral sphincter complex. Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=15076301&query_hl=8 An anatomical description of the male and female urethral sphincter complex. Yucel S, Baskin LS: J Urol 171(5):1890-7, 2004.
© 2008 Elsevier
96 BLADDER, PROSTATE AND URETHRA Prostate The prostate is a pyramidal fibromuscular gland which surrounds the prostatic urethra from the bladder base to the membranous urethra and is itself surrounded by a thin but tough connective tissue capsule (Figs 94.1, 98.1, 101.1, 96.1, 96.2). It lies at a low level in the lesser pelvis, behind the inferior border of the symphysis pubis and pubic arch and anterior to the rectal ampulla, through which it may be palpated. Being somewhat pyramidal, it presents a base or vesical aspect superiorly, an apex inferiorly and posterior, anterior and two inferolateral surfaces. The prostatic base measures about 4 cm transversely. The gland is c.2 cm in anteroposterior and 3 cm in its vertical diameters, and weighs c.8 g in youth, but almost invariably enlarges with the development of benign prostatic hyperplasia (BPH), weighing usually c.40 g, but sometimes as much as 150 g or even more after the first five decades of life (p. 1301). Superiorly the base is largely contiguous with the neck of the bladder. The urethra enters the prostate near its anterior border. The apex is inferior, surrounding the junction of the prostatic and membranous parts of the posterior urethra. The anterior surface lies in the arch of the pubis, separated from it by a venous plexus (Santorini's plexus) and loose adipose tissue. It is transversely narrow and convex, extending from the apex to the base. Near its superior limit it is connected to the pubic bones by the puboprostatic ligaments. The urethra emerges from this surface anterosuperior to the apex of the gland. The anterior part of the prostate is relatively deficient in glandular tissue and is largely composed of fibromuscular tissue. The inferolateral surfaces are related to the muscles of the pelvic sidewall: the anterior fibres of levator ani embrace the prostate in the pubourethral sling or pubourethralis. These muscles are separated from the prostate by a thin layer of connective tissue. The posterior surface is separated from the rectum by the prostatic capsule and by Denonvillier's fascia, a dense condensation of pelvic fascia which develops by obliteration of the rectovesical peritoneal pouch. It is obliterated from below upwards as fetal life progresses so that at birth this fascia separates the prostate, the seminal vesicles and the ampullae of the vasa deferentia from the rectum. The posterior surface is transversely flat and vertically convex. Near its superior (juxtavesical) border is a depression where it is penetrated by the two ejaculatory ducts. Below this is a shallow, median sulcus, usually considered to mark a partial separation into right and left lateral lobes. The anterior and lateral aspects of the prostate are covered by a layer of fascia derived from the endopelvic fascia on each side. The prostatic venous plexus (Fig. 93.3) lies between this extension of the endopelvic fascia and the capsule of the prostate. Anteroinferiorly the fascia and the capsule of the prostate merge and blend with the puboprostatic ligaments.
The prostate is traversed by the urethra and ejaculatory ducts, and contains the prostatic utricle. The urethra usually passes between its anterior and middle thirds. The ejaculatory ducts pass anteroinferiorly through its posterior region to open into the prostatic urethra (p. 1295).
© 2008 Elsevier
ZONAL ANATOMY OF THE PROSTATE (Figs 96.1, 96.2)
page 1301 page 1302
Figure 96.1 Zonal anatomy of the prostate. (By permission from Walsh PC, Retik AB, Vaughan ED et al (eds) 2002 Campbell's Urology, 8th edn. Philadelphia: Saunders.)
Figure 96.2 T2-weighted axial MRI scan of the prostate in a young man showing the normal high signal of the peripheral zone, intermediate signal central and transitional zones, and verumontanum in the central gland.
The prostate gland was initially thought to be divided into five anatomical lobes, but it is now recognized that five lobes can only be distinguished in the fetal gland prior to 20 weeks' gestation. Between then and the onset of BPH, three lobes are recognizable, two lateral and a median lobe. This simplified view of prostatic lobation is retained because clinicians refer to left and right 'lobes' when describing rectally palpable and endoscopically visible abnormalities in the diseased state when prostatic anatomy is distorted by BPH. From an anatomical and particularly from a morbid anatomical perspective, the glandular tissue may be subdivided into three distinct zones, peripheral (70% by volume), central (25% by volume), and transition (5% by volume) (Fig. 96.2). Non-glandular tissue (fibromuscular stoma) fills up the space between the
peripheral zones anterior to the preprostatic urethra. The central zone surrounds the ejaculatory ducts posterior to the preprostatic urethra and is more or less conical in shape with its apex at the verumontanum. The transition zone lies around the distal part of the preprostatic urethra just proximal to the apex of the central zone and the ejaculatory ducts. Its ducts enter the prostatic urethra just below the preprostatic sphincter and just above the ducts of the peripheral zone. The peripheral zone is cup-shaped and encloses the central transition zone and the preprostatic urethra except anteriorly, where the space is filled by the anterior fibromuscular stoma. Simple mucus-secreting glands lie in the tissue around the preprostatic urethra, above the transition zone and surrounded by the preprostatic sphincter. These simple glands are similar to those in the female urethra and unlike the glands of the prostate. The zonal anatomy of the prostate is clinically important because most carcinomas arise in the peripheral zone, whereas BPH affects the transition zone, which may grow to form the bulk of the prostate. BPH begins as micronodules in the transition zone; these grow and coalesce to form macronodules around the inferior margin of the preprostatic urethra, just above the verumontanum. Macronodules in turn compress the surrounding normal tissue of the peripheral zone posteroinferiorly thereby creating a 'false capsule' around the hyperplastic tissue, which coincidentally provides a plane of cleavage for surgical enucleation of the hyperplastic tissue. As the transition zone grows, it produces the appearance of 'lobes' on either side of the urethra above. These lobes may, in due course, compress or distort the preprostatic and prostatic parts of the urethra to produce symptoms. The central zone surrounding the ejaculatory ducts is rarely involved in any disease. It shows certain histochemical characteristics which are different from the rest of the prostate and is thought to be derived from the Wolffian duct system (much like the epididymi, vasa deferentia and seminal vesicles), whereas the rest of the prostate is derived from the urogenital sinus (p. 1393). The zonal anatomy may be distinguished to some extent on radiological imaging. On transrectal ultrasonography (TRUS) the central and peripheral zones are generally of uniform low-level echogenicity, although slight differences may be appreciated. The preprostatic urethra is surrounded by a less echogenic area which corresponds to the preprostatic sphincter, periurethral glandular tissue and transition zone. It is often possible to see the ejaculatory ducts coursing to the prostatic urethra on sagittal scans of the gland. The seminal vesicles are hypoechoic/anechoic sacculated structures which lie superoposterior to the gland. On magnetic resonance (MR) imaging the prostate gland has a zonal anatomy on T2-weighted images (Fig. 96.2). The normal peripheral zone has high signal intensity, as does fluid within the seminal vesicles. The central and transition zones have relatively low signal and are often referred to as the 'central gland'. The verumontanum may be seen as high signal within the central gland. The relationship of the zones of the gland normally changes with age. The central zone atrophies, and the transition zone enlarges secondary to BPH. This often produces a low signal band at the margin of the hypertrophied transition and
compressed peripheral zones, the surgical pseudocapsule, which is well seen on T2-weighted MR images. The fluid-containing seminal vesicles appear very high signal on T2-weighted images
© 2008 Elsevier
VASCULAR SUPPLY AND LYMPHATIC DRAINAGE ARTERIES (Fig. 108.4) The prostate is supplied by branches from the inferior vesical, internal pudendal and middle rectal arteries. They perforate the gland along a posterolateral line from the junction of the prostate with the bladder down to the apex of the gland.
VEINS The veins run into a plexus around the anterolateral aspects of the prostate, posterior to the arcuate pubic ligament and the lower part of symphysis pubis, anterior to the bladder and prostate (Fig. 93.3). The chief tributary is the deep dorsal vein of the penis. The plexus also receives anterior vesical and prostatic rami (which connect with the vesical plexus and internal pudendal vein), and drains into vesical and internal iliac veins.
LYMPHATIC DRAINAGE Collecting vessels from the vas deferens end in the external iliac nodes, while those from the seminal vesicle drain to the internal and external iliac nodes. Prostatic vessels end mainly in internal iliac, sacral and obturator nodes. A vessel from the posterior surface accompanies the vesical vessels to the external iliac nodes and one from the anterior surface reaches the internal iliac group by joining vessels which drain the membranous urethra.
© 2008 Elsevier
ARTERIES (Fig. 108.4) The prostate is supplied by branches from the inferior vesical, internal pudendal and middle rectal arteries. They perforate the gland along a posterolateral line from the junction of the prostate with the bladder down to the apex of the gland.
VEINS The veins run into a plexus around the anterolateral aspects of the prostate, posterior to the arcuate pubic ligament and the lower part of symphysis pubis, anterior to the bladder and prostate (Fig. 93.3). The chief tributary is the deep dorsal vein of the penis. The plexus also receives anterior vesical and prostatic rami (which connect with the vesical plexus and internal pudendal vein), and drains into vesical and internal iliac veins.
LYMPHATIC DRAINAGE Collecting vessels from the vas deferens end in the external iliac nodes, while those from the seminal vesicle drain to the internal and external iliac nodes. Prostatic vessels end mainly in internal iliac, sacral and obturator nodes. A vessel from the posterior surface accompanies the vesical vessels to the external iliac nodes and one from the anterior surface reaches the internal iliac group by joining vessels which drain the membranous urethra.
© 2008 Elsevier
INNERVATION The prostate has an abundant nerve supply from the inferior hypogastric (pelvic) plexus. The prostatic capsule is covered by numerous nerve fibres and ganglia, which form a periprostatic nerve plexus. The greatest density of nerves is found in the preprostatic sphincter, followed by the anterior fibromuscular stroma, and the peripheral zone is the least densely innervated. Nerves containing neuropeptide Y and vasointestinal polypeptide (VIP) are localized in the subepithelial connective tissue, in the smooth muscle layers of the gland, and in the walls of its blood vessels. Neurovascular bundles containing nerves which supply the prostate, seminal vesicles, prostatic urethra, ejaculatory ducts, corpora cavernosa, corpus spongiosum, membranous and penile urethra and bulbourethral glands are closely applied to, but separable from, the posterolateral margins of the prostate. These nerves are frequently damaged during radical prostate surgery for organconfined prostate cancer, producing impotence (p. 1317).
© 2008 Elsevier
MICROSTRUCTURE (Figs 96.3, 96.4) The prostate is grey to reddish in colour according to its activity, and is very dense. It is enclosed by a thin, but strong fibrous capsule within a sheath derived from pelvic fascia and containing a venous plexus. The capsule is firmly adherent to the gland and is continuous with a median septum and with numerous fibromuscular septa which divide the glandular tissue into indistinct lobules. page 1302 page 1303
Figure 96.3 Main prostatic glands with several prostatic concretions (corpora amylacea) as seen centre right. The lining epithelium is typically variable, from cuboidal to pseudostratified columnar, with complex infolded regions. Several ducts are seen, top left. (Photograph by Sarah-Jane Smith.)
The muscular tissue is mainly smooth. Anterior to the urethra a layer of smooth muscle merges with the main mass of muscle in the fibromuscular septa. Superiorly it blends with vesical smooth muscle. Anterior to this smooth muscle a transversely crescent-shaped mass of skeletal muscle is continuous inferiorly with the sphincter urethrae in the deep perineal pouch. Its fibres pass transversely internal to the capsule, and are attached to it laterally by diffuse collagen bundles; other collagen bundles pass posteromedially, merging with the prostatic fibromuscular septa and the septum of the urethral crest. This muscle, supplied by the pudendal nerve, probably compresses the urethra but it may pull the urethral crest back and the prostatic sinuses forwards, dilating the urethra. Glandular contents may be expelled simultaneously into the urethra when it has expanded in this way, so that it contains 3-5 ml seminal fluid prior to ejaculation. The glandular tissue consists of numerous follicles with frequent internal papillae. Follicles open into elongated canals which join to form 12-20 main ducts. The follicles are separated by loose connective tissue, supported by extensions of the fibrous capsule and muscular stroma and enclosed in a delicate capillary plexus. Follicular epithelium is variable but predominantly columnar and either singlelayered or pseudostratified. Prostatic ducts open mainly into the prostatic sinuses in the floor of the prostatic urethra. They have a bilayered epithelium, the luminal layer is columnar and the basal layer is populated by small cuboidal cells. Small colloid amyloid bodies are frequent in the follicles (Fig. 96.3). Prostatic and seminal vesicular secretions form the bulk of seminal fluid. Prostatic secretions are slightly acid, and contain acid phosphatase, amylase, prostate specific antigen and fibrinolysin as well as zinc. Numerous neuroendocrine cells, containing neurone-specific enolase, chromogranin and serotonin, are present in the glandular epithelium. Their numbers decline after middle age and their function is unknown. Histological sections just above the level of the verumontanum show two
concentric, partially circumurethral, zones of glandular tissue (Fig. 96.4). The larger outer zone is the peripheral zone which has long, branched glands, whose ducts open mainly into the prostatic sinuses. The inner zone is the transition zone and consists of glands whose ducts open on the floor of the prostatic sinuses and colliculus seminalis and a group of simple mucosal glands, surrounding the preprostatic urethra. Anteriorly, in the prostatic isthmus, the peripheral zone and submucosal glands are absent. Carcinomas arise almost exclusively in the peripheral zone, whereas the transition zone is prone to BPH.
Figure 96.4 A, Microstructure of the prostate. B, The prostate gland is shown in transverse hemisection at the level of the urethral crest. Periurethral submucosal glands are commonly involved in benign prostatic hypertrophy, whereas the peripheral glands are the usual site of origin of carcinoma. (By permission from Kierszenbaum AL 2002 Histology and Cell Biology. St Louis: Mosby.)
© 2008 Elsevier
AGE CHANGES IN THE PROSTATE At birth, the prostate has a system of ducts embedded in a stroma which forms a large part of the gland. Follicles are represented by small end-buds on the ducts. Before birth there is hyperplasia and squamous metaplasia of the epithelium of the ducts, colliculus seminalis and prostatic utricle, possibly due to maternal oestrogens in the fetal blood. This subsides after birth and is followed by a period of quiescence lasting for 12-14 years. At puberty, between the ages of approximately 14 and 18 years, the prostate gland enters a maturation phase: it more than doubles in size during this time. Growth is almost entirely due to follicular development, partly from end-buds on ducts, and partly from modification of the ductal branches. Morphogenesis and differentiation of the epithelial cords starts in an intermediate part of the epithelial anlage and proceeds to the urethral and subcapsular parts of the gland; the latter is reached by the age of 17-18 years. The glandular epithelium is initially multilayered squamous or cuboidal, and is transformed into a pseudostratified epithelium consisting of basal, exocrine secretory (including mucous) and neuroendocrine cells. The mucous cells are temporary, and are lost as the gland matures. The remaining exocrine secretory cells produce a number of products including acid phosphatase, prostate-specific antigen and !-microseminoprotein. This growth of the secretory component is associated with a condensation of the stroma, which diminishes relative to the glandular tissue. These changes are probably a response to the secretion of testosterone by the testis. During the third decade the glandular epithelium grows by irregular multiplication of the epithelial infoldings into the lumen of the follicles. After the third decade the size remains virtually unaltered until 45-50 years, when the epithelial foldings tend to disappear, follicular outlines become more regular, and amyloid bodies increase in number. All these changes are signs of prostatic involution. After 45-50 years the prostate tends to develop BPH. The nature of BPH has been outlined earlier in this chapter. It is an age-related condition: if a man lives long enough then it is inevitable, although it is not always symptomatic. REFERENCE Mundy AR, Fitzpatrick J, Neal D, George N (eds) 1999 The prostate and benign prostatic hyperplasia. In: The Scientific Basis of Urology, Chapter 13. Oxford: Isis Medical Media: 257-76. Includes a review of prostatic zonal anatomy.
© 2008 Elsevier
97 MALE REPRODUCTIVE SYSTEM Testes and epididymes The testes are the primary reproductive organs or gonads in the male. They are ovoid reproductive and endocrine organs responsible for sperm production and are suspended in the scrotum by scrotal tissues including the non-striated dartos muscle and the spermatic cords. Average testicular dimensions are 4-5 cm in length, 2.5 cm in breadth and 3 cm in anteroposterior diameter; their weight varies from 10.5-14 g. The left testis usually lies lower than the right testis. Each testis lies obliquely within the scrotum, its upper pole tilted anterolaterally and the lower posteromedially (Fig. 97.1). The anterior aspect is convex, the posterior nearly straight, with the spermatic cord attached to it. Anterior, medial and lateral surfaces and both poles are convex, smooth and covered by the visceral layer of the serosal tunica vaginalis, which separates them from the parietal layer and the scrotal tissues external to this. Between these two layers there is always a very fine film of fluid. This fluid layer can increase on occasions, creating a hydrocele. The posterior aspect is only partly covered by tunica serosa; the epididymis adjoins its lateral part.
© 2008 Elsevier
TESTIS The testis is invested by three coats, which are, from outside inwards, the tunica vaginalis, tunica albuginea and tunica vasculosa. Each testis is separated from its fellow by a fibrous median raphe, which is deficient superiorly.
TUNICA VAGINALIS
Figure 97.1 The left testis, exposed by incising and laying open the cremasteric fascia and parietal layer of the tunica vaginalis on the lateral aspect of the testis.
The tunica vaginalis (Figs 97.1, 99.1.) is the lower end of the peritoneal processus vaginalis, whose formation precedes the descent of the fetal testis from the abdomen to the scrotum (p. 1388). After this migration, the proximal part of the tunica, from the internal inguinal ring almost to the testis, contracts and is obliterated, leaving a closed distal sac into which the testis is invaginated. The tunica is reflected from the testis onto the internal surface of the scrotum, so forming the visceral and parietal layers of the tunica. The visceral layer covers all aspects of the testis except most of the posterior aspect. Posteromedially it is reflected forwards to the parietal layer. Posterolaterally it passes to the medial aspect of the epididymis and lines the epididymal sinus, and then passes laterally to its posterior border where it is reflected forwards to become continuous with the parietal layer. The visceral and parietal layers are continuous at both poles but at the upper pole the visceral layer surmounts the head of the epididymis before reflexion. The more extensive parietal layer reaches below the testis and ascends in front of and medial to the spermatic cord. The inner surface of the tunica vaginalis has a smooth, moist mesothelium: the potential space between its visceral and parietal layers is termed the cavity of the tunica vaginalis.
TUNICA ALBUGINEA The tunica albuginea is a dense, bluish-white covering for the testis. It is composed mainly of interlacing bundles of collagen fibres, and is covered
externally by the visceral layer of the tunica vaginalis, except at the epididymal head and tail and the posterior aspect of the testis, where vessels and nerves enter. It covers the tunica vasculosa and, at the posterior border of the testis, projects into the testicular interior as a thick, incomplete, fibrous septum, the mediastinum testis, which extends from the upper to the lower end of the testis (Fig. 97.2). Testicular vessels run within the mediastinum testis.
page 1305 page 1306
Figure 97.2 Vertical section through the testis and epididymis, showing the arrangement of the ducts of the testis and the mode of formation of the vas deferens.
TUNICA VASCULOSA The tunica vasculosa contains a plexus of blood vessels and delicate loose connective tissue, and extends over the internal aspect of the tunica albuginea, covering the septa and therefore all the testicular lobules.
UNDESCENDED TESTIS (See also p. 1306.) In the early fetal period the testes are located posteriorly in the abdominal cavity. Their descent to the scrotum appears to be under hormonal control (gonadotropins and androgens) (p. 1388). Testicular descent may be arrested at any point along its route into the scrotum and a clinically undescended testis may be in the abdomen, at the deep inguinal ring, in the inguinal canal, or between the superficial inguinal ring and the scrotum. A unilateral undescended testis is present in 3% of boys at birth and 1% of boys by 3 months of age. Bilateral maldescent is seen in just over 1% of male births. Undescended testes are associated with infertility. There is evidence that surgical correction of an undescended testis at any age may not improve its
spermatogenesis, but significant impairment of fertility is probably only seen in men with bilateral undescended testis. Leydig cell function is not usually affected by maldescent, so androgen production usually remains within the normal range and erectile potency function is unaffected. Patients with an undescended testis are at increased risk of testicular tumour, particularly seminoma, and require surgical intervention as early as possible to ensure its correct location. The risk is highest in abdominal testes. Surgery may not reduce the risk of tumour development, but maximizes the chance of early detection of any tumour. An undescended testis can usually be found by ultrasonography if it is in or close to the inguinal canal. Laparoscopy is more reliable in the pelvis. Retention in the inguinal canal is often complicated by congenital hernia, because the processus vaginalis remains patent. The testis may traverse the canal but reach an abnormal site.
OBLITERATED PART OF THE PROCESSUS VAGINALIS The obliterated part of the processus vaginalis is often seen as a fibrous thread in the anterior part of the spermatic cord, extending from the internal end of the inguinal canal - where it is connected to the peritoneum - as far as the tunica vaginalis. Sometimes it disappears within the cord. However, its proximal part may remain patent, so that the peritoneal cavity communicates with the tunica vaginalis, or the proximal processus may persist, although it may be shut off distally from the tunica. Occasionally its cavity may persist at an intermediate level as a cyst. When patent, its cavity may admit a loop of intestine, to form an indirect inguinal hernia (p. 1111). The processus is usually obliterated by 18 months of age.
HYDROCELE, SPERMATOCELE, EPIDIDYMAL CYST In congenital hydrocele the fluid is in the tunical sac, which communicates with the peritoneal cavity through a non-obliterated processus vaginalis. Infantile hydrocele occurs when the processus is obliterated only at or near the deep inguinal ring. It resembles vaginal hydrocele, but fluid extends up the cord into the inguinal canal. If the processus is obliterated at both the deep inguinal ring and above the epididymis, leaving a central open part, this may distend as an encysted hydrocele of the cord. A spermatocele is a cyst related to the caput epididymis: it may contain spermatozoa and it is probably a retention cyst of one of the seminiferous tubules. Removal is usually unnecessary and may result in epididymal obstruction. The same applies to a simple epididymal cyst, which may have a similar aetiology to a spermatocele but remains free of sperm.
© 2008 Elsevier
EPIDIDYMIS (Fig. 97.2) The epididymis lies posteriorly and slightly lateral to the testis, with the vas deferens along its medial side (Fig. 97.2), (Ch. 98). It has an expanded head or globus major superiorly, a body (corpus) and a tail (cauda or globus minor). Its overall length is 6-7 cm and it consists of the single convoluted ductus epididymis formed by the union of the efferent ducts of the testis, which attach to the rete testis. From the tail, the vas deferens ascends medially to the deep inguinal ring, within the spermatic cord. The epididymis is invested by tunica vaginalis, some what less closely applied than it is to the testis, except at its posterior margin. Laterally there is a deep groove, the sinus epididymis, between the epididymis and the testis.
© 2008 Elsevier
TESTICULAR AND EPIDIDYMAL APPENDICES At the upper extremities of the testis and epididymis are two small, stalked bodies, the appendix testis and appendix epididymis. They are developmental remnants of the paramesonephric ducts (Müllerian) duct and the mesonephros, respectively (Figs 97.1, 99.1), (p. 1385). They are liable to undergo torsion.
© 2008 Elsevier
TESTICULAR TORSION The testis and epididymis are usually fixed to their surrounding tissues. In some patients this fixation may be insufficient, a condition which allows the structures to twist within the tunica vaginalis. This is termed testicular torsion and normally results in severe scrotal pain. This is a surgical emergency. Histopathological changes leading to gangrene occur in the testis if the twist is not reversed within 4-6 hours. The injury results from venous and then arterial occlusion. Fertility may be affected by an episode of torsion. Other structures may also twist within the scrotum, e.g. the testicular appendix (otherwise termed the hydatid of Morgagni) and the appendix epididymis. Torsion of these structures may result in scrotal pain, which is usually far more localized than the pain of testicular torsion.
© 2008 Elsevier
VASCULAR SUPPLY AND LYMPHATIC DRAINAGE TESTICULAR ARTERIES The testicular arteries are two long, slender vessels, which arise anteriorly from the aorta a little inferior to the renal arteries. Each passes inferolaterally under the parietal peritoneum on psoas major. The right testicular artery lies anterior to the inferior vena cava and posterior to the horizontal part of the duodenum, right colic and ileocolic arteries, root of the mesentery and terminal ileum. The left testicular artery lies posterior to the inferior mesenteric vein, left colic artery and lower part of the descending colon. Each artery crosses anterior to the genitofemoral nerve, ureter and the lower part of the external iliac artery and passes to the deep inguinal ring to enter the spermatic cord and travel via the inguinal canal to enter the scrotum. At the posterosuperior aspect of the testis the testicular artery divides into two branches on its medial and lateral surfaces: these pass through the tunica albuginea and ramify in the tunica vasculosa. Terminal branches enter the testis over its surface. Some pass into the mediastinum testis and loop back before reaching their distribution. Capillaries lying next to seminiferous tubules penetrate the layers of interstitial tissue and are of interest as part of the 'bloodtestis' barrier. They run either parallel to the tubules or across them but do not enter their walls. They are separated from germinal and supporting cells by a basement membrane and variable amounts of fibrous tissue containing interstitial cells: selective exchange phenomena involving androgens and immune substances occur here. In the abdomen the testicular artery supplies perirenal fat, the ureter and iliac lymph nodes, and in the inguinal canal it supplies cremaster. Sometimes the right testicular artery passes posterior to the inferior vena cava. The testicular arteries represent persistent lateral splanchnic aortic branches which enter the mesonephros and cross ventral to the supracardinal vein, but dorsal to the subcardinal vein. Normally the lateral splanchnic artery - which persists as the right testicular artery - passes caudal to the suprasubcardinal anastomosis, which forms part of the inferior vena cava. When it passes cranial to the anastomosis, the right testicular artery is behind the inferior vena cava. page 1306 page 1307
The testis also receives blood from the cremasteric branch of the inferior epigastric artery, and from the artery to the vas deferens. Interference with the testicular artery high in the abdomen therefore usually leaves the testis unharmed, whereas interruption in the region of the spermatic cord may interfere with all of these vessels and lead to infarction. Indeed it has been proposed that the correct way of treating varicoceles is to ligate both the testicular artery and vein high up, which also has the advantage of ligating the venae comitantes of the artery. These small veins anastomose with the internal spermatic veins and can be responsible for recurrence of the varicocele.
TESTICULAR VEINS (Fig. 97.3) The testicular veins emerge posteriorly from the testis, drain the epididymis and
unite to form the pampiniform plexus, which is a major component of the spermatic cord, and ascends anterior to the vas deferens. In the inguinal canal the plexus is drained by three or four veins which run into the abdomen through the deep inguinal ring. Within the abdomen these veins coalesce into two veins, which ascend on each side of the testicular artery, anterior to psoas major and the ureter, and behind the peritoneum. The left veins pass behind the lower descending colon and inferior margin of the pancreas and are crossed by the left colic vessels, and the right veins pass behind the terminal ileum and horizontal part of the duodenum and are crossed by the root of the mesentery, ileocolic and right colic vessels. The veins join to form single right or left testicular veins: the right testicular vein opens into the inferior vena cava at an acute angle just inferior to the level of the renal veins, and the left testicular vein opens into the left renal vein at a right angle. The testicular veins contain valves. 1. Right testicular vein. 2. Inferior vena cava. 3. Left renal vein. 4. Left testicular vein.
Figure 97.3 Multislice CT of the inferior vena cava showing the left testicular vein draining to the left renal vein and the right testicular vein draining directly to the inferior vena cava.
The testicular veins in the scrotum and inguinal canal are frequently varicose. Varicocele formation, which is almost always on the left, is perhaps due to the orthogonal junction of the left testicular and renal veins. There is evidence that
the presence of a varicocele raises testicular temperature and impairs fertility. In fact most varicoceles are treated (surgically) for pain rather than infertility. Varicoceles can also be treated by radiological embolization of the left testicular vein via a right femoral vein approach. After ligation of a varicocele, venous return is by the small veins of the vas deferens, cremaster and scrotal tissues.
LYMPHATIC DRAINAGE OF THE TESTIS Testicular vessels start in a superficial plexus under the tunica vaginalis, and a deep plexus in the substance of the testis and epididymis. Four to eight collecting trunks ascend in the spermatic cord and accompany the testicular vessels on psoas major, ending in the lateral aortic and pre-aortic nodes.
© 2008 Elsevier
INNERVATION Testicular nerves accompany the testicular vessels and are derived from the tenth and eleventh thoracic spinal segments via the renal and aortic autonomic plexuses. Catecholaminergic nerve fibres form plexuses around smaller blood vessels and among the interstitial cells in the testis and epididymis.
© 2008 Elsevier
MICROSTRUCTURE TESTIS (Figs 97.4, 97.5) The surface of the testis is covered closely by the visceral tunica vaginalis, a layer of flat mesothelial cells similar to and continuous with the peritoneal lining. It is separated from the parietal tunica vaginalis, the outer layer of the double fold of peritoneum, which accompanies the descending testis (p. 1388), by a potential space containing serous fluid, which acts as a lubricant and allows movement of the testis within the scrotum. The testicular capsule proper, the tunica albuginea, is tough and collagenous and thickened posteriorly as the mediastinum testis. Beneath the tunica albuginea is a thin layer of connective tissue containing the superficial blood vessels. Blood vessels, lymphatics and the genital ducts enter or leave the body of the testis at the mediastinum.
page 1307 page 1308
Figure 97.4 Seminiferous tubules (cut in various planes of section), and the interstitial tissue of the testis. The seminiferous tubules are highly convoluted and lined by a stratified epithelium which consists of cells in various stages of spermatogenesis and spermiogenesis (collectively referred to as the spermatogenic series). Nonspermatogenic cells are the Sertoli cells. (By permission from Young B, Heath JW 2000 Wheater's Functional Histology. Edinburgh: Churchill Livingstone.)
Figure 97.5 Human seminiferous tubule showing the differentiation sequence of spermatozoa from basally situated spermatogonia (SG). Large primary spermatocytes (SC) have characteristic threadlike chromatin in various stages of prophase of the first meiotic division. Smaller haploid spermatids (ST) have round nuclei initially, but mature to possess the dense, elongated nuclei and flagellae of spermatozoa (SZ). Sertoli cells (S) are identified from their oval or pear-shaped nuclei orientated perpendicular to the basal lamina, and prominent nucleoli. The tubule is surrounded by peritubular myoid cells (M) and clusters of large endocrine Leydig cells (L) are seen in the interstitial connective tissue. (By permission from Dr JB Kerr, Monash University, from Kerr JB 1999 Atlas of Functional Histology. London: Mosby.)
Septa from the mediastinum extend internally to partition the testis into c.250 lobules (Fig. 97.2). These differ in size, and are largest and longest in the centre. Each lobule contains one to four convoluted seminiferous tubules, which are much-coiled loops whose free ends both open into channels (the rete testis) within the mediastinum. The loose connective tissue between seminiferous tubules contains several layers of contractile peritubular myoid cells, and clusters of steroid-producing interstitial (Leydig) cells (Fig. 97.5). There are 400-600 seminiferous tubules in each testis and the length of each is 70-80 cm. Their diameter varies from 0.12-0.3 mm. They are pale in early life, but in old age they contain much fat and are deep yellow. Each tubule is surrounded by a basal lamina, on which rests a complex, stratified seminiferous epithelium containing spermatogenic cells and supportive Sertoli cells. When active, the spermatogenic cells include basally situated spermatogonia and their progeny in the adluminal compartment, spermatocytes, spermatids and mature spermatozoa (Fig. 97.5). Among the spermatids may be residual bodies, which are spherical structures derived from surplus spermatid cytoplasm shed during maturation and phagocytosed by Sertoli cells. For a review of the ultrastructural features of the human testis, with emphasis on spermatogenesis and the cytology of the Leydig cells, see Kerr (1991). Spermatogonia
Spermatogonia, the stem cells for all spermatozoa, are descended from primordial germ cells which migrate into the genital cords of the developing testis (p. 1386). In the fully differentiated testis they are located along the basal laminae of the seminiferous tubules. Several types of spermatogonia are recognized on the
basis of cell and nuclear dimensions, distribution of nuclear chromatin (dark, condensed or pale, euchromatic) and histochemical and ultrastructural data. The three basic groups of spermatogonia are dark type A (Ad), pale type A (Ap), and type B. Ad cells divide mitotically to maintain the population of spermatogonia which, before puberty, is small but increases under androgenic stimulation. Some divisions give rise to Ap cells which also divide mitotically but remain linked in clusters by fine cytoplasmic bridges. These are the precursors of type B cells, which are committed to the spermatogenic sequence. At about the time type B cells enter a final round of DNA synthesis, without undergoing cytokinesis, they leave the basal compartment and cross the blood-testis barrier to enter meiotic prophase as primary spermatocytes. These coordinated processes are under the control of Sertoli cells. Primary and secondary spermatocytes
Primary spermatocytes have a diploid chromosome number but duplicated sister chromatids (DNA content is thus 4N, where N is the DNA content of haploid spermatozoa). They are all at some stage of a long meiotic prophase (p. 24) of c.3 weeks. Primary spermatocytes are characteristically large cells (Fig. 97.5) with large round nuclei in which the nuclear chromatin is condensed into dark, threadlike, coiled chromatids at different stages in the process of crossing over and genetic exchange between chromatids of maternal and paternal homologues. These cells give rise to secondary spermatocytes with a haploid chromosome complement (but 2N DNA content), the reduction division is designated as meiosis I. Few secondary spermatocytes are seen in tissue sections because they rapidly undergo the second meiotic (equatorial) division, where sister chromatids separate (DNA content is now N), to form haploid spermatids. Theoretically each primary spermatocyte produces four spermatids, but some degenerate during maturation so that the yield is lower. Spermatids
Spermatids do not divide again but gradually mature into spermatozoa by a series of nuclear and cytoplasmic changes known as spermiogenesis. All of these maturational changes take place while the spermatids remain closely associated with Sertoli cells and linked by cytoplasmic bridges with each other. The first phase of spermiogenesis is the Golgi phase, during which hydrolytic enzymes accumulate in Golgi vesicles, which coalesce into a single large acrosomal vesicle, close to the nucleus. The pair of centrioles migrates to the opposite posterior pole. The distal centriole begins to generate the axoneme, a circular arrangement of nine microtubule doublets surrounding a central pair. In the cap phase, which follows, the acrosomal vesicle flattens and envelops the anterior half of the nucleus to form an acrosomal cap. This comes to occupy the presumptive anterior pole of the spermatozoon, furthest from the tubule lumen. During the acrosome phase, nuclear chromatin condenses and the nucleus elongates to a spearhead shape. The anterior cytoplasmic volume reduces considerably, bringing the wall of the acrosomal vesicle into contact with the plasma membrane. A perinuclear sheath of microtubules develops from the posterior edge of the acrosome to form the manchette, extending towards the posterior pole. The axonemal complex continues to extend into the developing tail
region, which protrudes into the tubule lumen. Prominent mitochondria migrate through a neck region, which forms at the posterior pole of the nucleus and contains the centrioles, and along the axoneme into the developing middle piece. Here they assemble into a helical sheath of mitochondria which surrounds a ring of nine coarse fibres forming around the axonemal complex along its length in the developing tail. In the final phase of maturation, excess cytoplasm is detached as a residual body which is phagocytosed and degraded by Sertoli cells. During the formation of residual bodies, spermatids lose their cytoplasmic bridges and separate from each other before being released into their tubule. Spermatozoa (Fig. 97.6)
As it is released from the wall of the seminiferous tubule into the lumen, the spermatozoon is non-motile but structurally mature (Fig. 97.6). Its expanded head contains little cytoplasm and is connected by a short constricted neck to the tail. The tail is a complex flagellum, which greatly exceeds the head region in volume, and is divided into middle, principal and end pieces. The head has a maximum length of c.4 µm and a maximum diameter of 3 µm, and contains the elongated, flattened nucleus with condensed, deeply staining chromatin and the acrosomal cap anteriorly. The latter contains acid phosphatase, hyaluronidase , neuraminidase and proteases necessary for fertilization (p. 185). The neck is c.0.3 µm long. In its centre is a well-formed centriole, corresponding to the proximal centriole of the spermatid from which it differentiated. The axonemal complex is derived from the distal centriole. A small amount of cytoplasm exists in the neck, covered by a plasma membrane continuous with that of the head and tail. page 1308 page 1309
Figure 97.6 The main ultrastructural features of a mature spermatozoon.
The middle piece of the tail is a long cylinder, c.1 µm in diameter and 7 µm long. It consists of an axial bundle of microtubules, the axoneme, outside which is a cylinder of nine dense outer fibres, surrounded by a helical mitochondrial sheath. At the caudal end of the middle piece is an electron-dense body, the anulus. The principal piece of the tail becomes the motile part of the cell. It is c.40 µm long and 0.5 µm in diameter and forms the majority of the spermatozoon. The axoneme and the surrounding dense fibres are continuous from the neck region through the whole length of the tail except for its terminal 5-7 µm, in which the axoneme alone persists. In this terminal end piece the tail thus has the typical structure of a flagellum, with a simple nine plus two arrangement of microtubules. Sertoli cells
Sertoli cells are the supporting, non-spermatogenic cells of the seminiferous tubules and form a major cellular component of the tubule before puberty, and in the elderly. They are variable in overall cell shape, but they all contact the basal lamina and their cytoplasm extends to the tubule lumen. Here, their apical plasma membranes form complex recesses which envelop spermatids and spermatozoa until the latter are mature enough for release. Long cytoplasmic processes also extend between the spermatogonia in the basal compartment and spermatocytes in the adluminal compartment of the tubule. Adjacent Sertoli cell processes are joined at this level by tight junctions, which create a diffusion barrier between the
extratubular and intratubular compartments. This is the blood-testis barrier which, if breached by traumatic or inflammatory events, can allow immune responses to develop against sperm antigens, resulting in subfertility. For a general review of the blood-testis barrier see Johnson and Gomes (1977). The Sertoli cell nucleus is euchromatic and irregular or pear-shaped, and contains one or two prominent nucleoli. It is usually aligned perpendicular to the basal lamina. The cytoplasm contains many lysosomes, consistent with its phagocytic phenotype. Sertoli cells provide trophic support for the surrounding germ cells, secrete androgen-binding protein and play an important role in controlling spermatocyte and spermatid differentiation and maturation. The proteinaceous fluid they secrete into the tubule lumen provides nutrients and facilitates the transport of spermatozoa into the excurrent duct system. Sertoli cells change considerably during the spermatogenic cycle and respond to the hypophyseal hormones, luteinizing hormone (LH) and follicle-stimulating hormone (FSH). Spermatogenic cycle
At any locus in a seminiferous tubule, generation of germ cells occurs in a cycle, with a periodicity of c.16 days. Stages in the cycle are characterized by the presence of different combinations of cells within the spermatogenic sequence. The generation of a mature spermatozoon from a spermatogonium takes four such cycles, or c.64 days. In cross-section, the seminiferous tubule shows more than one phase of the cycle around its circumference, as waves of progression through a spermatogenic cycle occur in spirals down the length of the tubule. Testicular interstitial tissue
The tissues between the seminiferous tubules include various connective tissue components, peritubular myoid cells, vessels and nerves. The myoid cells are contractile and their rhythmic activity propels non-motile spermatozoa through the tubule towards the rete testis and excurrent ductal system. Clusters of steroidsecreting interstitial Leydig cells lie between the tubules. Leydig cells are large polyhedral cells with an eccentric nucleus containing one to three nucleoli. Their pale staining cytoplasm contains a considerable amount of smooth endoplasmic reticulum, lipid droplets and unique needle-shaped crystalloid inclusions up to 20 µm long (crystals of Reinke), of unknown function. Leydig cells synthesize and secrete androgens, stimulated by LH and prolactin, which induces expression of the LH receptor. The activity of Leydig cells varies with age: they are active in fetal life in the development of the genital tract but decline in function postpartum until the onset of puberty.
EFFERENT DUCTULES AND EPIDIDYMIS (Fig. 97.7) The process of spermatogenesis described above occurs in the highly coiled parts of the seminiferous tubules. As the latter reach the apical part of the lobule towards the mediastinum they become much less convoluted. They form the short tubuli recti, lined by cuboidal epithelium lacking spermatogenic cells. Tubuli recti enter the fibrous tissue of the mediastinum testis as a close network of anastomosing tubes, the rete testis, lined by a flat epithelium. At the upper pole of
the mediastinum, 12-20 efferent ductules (ductuli efferentes) perforate the tunica albuginea and leave the testis for the epididymis. The efferent ductules are lined by a ciliated columnar epithelium which also contains shorter, actively endocytic, non-ciliated cells. External to the epithelium, the ductules are surrounded by a thin circular coat of smooth muscle. Initially the ductules are straight. They become enlarged and very convoluted and form the conical lobules of the epididymis, which make up its head (caput). Each lobular duct, 15-20 cm in length, opens into the single duct of the epididymis, whose coils form the epididymal body (corpus) and tail (cauda). When the coils are unravelled, the tube measures more than 6 metres, increasing in thickness as it approaches the epididymal tail, where it becomes the vas deferens. The coils are held together by bands of fibrous connective tissue. In the epididymal duct the muscle is thicker and the epithelium is composed of columnar pseudostratified cells (Fig. 97.7). The muscle undergoes peristaltic contractions to propel spermatozoa towards the tail region, where they are stored.
page 1309 page 1310
Figure 97.7 The lining of the epididymis. The pseudostratified epithelium consists of two major cell types, principal cells with long apical stereocilia, and basal cells. (By permission from Kierszenbaum AL 2002 Histology and Cell Biology. St Louis: Mosby.)
The epithelium contains two main cell types, principal and basal cells, and the less common apical cells and clear cells. Principal cells are tall columnar cells with basally located, elongated, oval nuclei. They bear long (15 µm) regular apical microvilli termed stereocilia, because they were once thought to be immotile cilia. They function to resorb fluid from the testicular secretions and c.90% of the total is absorbed in the epididymis. These cells also secrete glycoproteins essential for the maturation of spermatozoa and endocytose various other components of the seminal fluid. Basal cells lie between the bases of the principal cells and are
thought to be the precursors of principal cells. Apical cells have numerous mitochondria and are most abundant in the head of the epididymis. Clear cells are columnar cells, most numerous in the tail region, with few microvilli but numerous endocytic vesicles and lipid droplets. Their functions are unknown.
MATURATION OF SPERMATOZOA Functional maturation is a complex process. Spermatozoa show little independent motility while still in the male genital tract, though when removed from the epididymis they may display circular or even forward directional movements if taken from the cauda epididymis near the beginning of the vas deferens. Apart from this immature motility, spermatozoa are largely transported through the genital tract first by ciliary action and then by peristaltic contractions of the duct walls. Human spermatozoa do not undergo any demonstrable structural changes during their passage through the epididymis, but there is evidence of biochemical and functional modifications. Evidence from restorative surgery after vasectomy indicates that at least part of the human epididymis is essential for acquisition of mature motile activity. Motility of spermatozoa
It is now generally accepted that the tail executes undulatory movements in one plane. It has also been suggested that a helical component is superimposed upon these movements, and that there are perhaps two separable mechanisms to account for sperm motility, one involving flat waves travelling along the tail, the other associated with torsional activity. The latter type of movement has been linked with the unequal size and distribution of the dense fibres, and the asymmetry of the spermatozoan head. It has also been suggested that the central pair of fine fibrils act as axial stiffeners. As soon as they are ejaculated the spermatozoa display their full pattern of motility. The factors which trigger these movements are not yet clear: the other constituents of semen, which are derived from the epididymis, testis, seminal vesicle and prostate, are generally considered to exert an activating influence. Spermatozoa have been recovered in a motile state in human cervical mucus several days after insemination and will survive in this condition for as long as 7 days when implanted into such secretions in vitro. In view of the speed with which spermatozoa reach the infundibulum of the uterine tube, and the brevity of their fertilizing power, these survival periods may be of little significance. Spermatozoa have been shown to reach their tubal destination in a manner of minutes after ejaculation in some mammals; experiments on recently excised human uteri and tubes indicate a time of c.70 minutes. The conclusion must be that factors other than their own motility (1.5-3.0 mm/min) are responsible for the transport of spermatozoa from the site of deposition in the vaginal fornix to the ovarian end of the uterine tube: there is considerable evidence that contraction of the uterine and tubal musculature is responsible. Capacitation
After ejaculation into the female genital tract, spermatozoa undergo the final step in their maturation, a process known as capacitation. This normally requires
exposure to the secretions of the uterine tube: a spermatozoon is unable to penetrate the corona radiata to fertilize an ovum until it has been within the female genital tract for a period of time, usually a few hours.
AGE CHANGES IN THE TESTIS Functionally, the fetal testis is predominantly an endocrine gland which produces testosterone and a specifically fetal gonadal hormone, the anti-Müllerian hormone. These two hormones play crucial roles in the induction and regulation of male sexual differentiation. The seminiferous tubules do not become canalized until approximately the seventh month of gestation, although this may occur later. Postnatally the testis gradually changes its role, but retains the ability to manufacture testosterone and other regulatory materials, e.g. the peptide hormone oxytocin , which act in either an endocrine or a paracrine fashion. At puberty, the testis becomes primarily a source of spermatozoa. The fetal Leydig cells, which are responsible for the androgen-induced differentiation of the male genitalia, degenerate after birth, and are replaced, during puberty, by an adult population of androgen-producing cells which persist throughout adult life. The testes grow slowly until the age of c.10 or 11 years, at which time there is a marked acceleration of growth rate, and spermatogenesis begins. There is no definite age for the onset of the progressive testicular involution associated with advancing age. Testicular size, sperm quality and quantity, and the numbers of Sertoli cells and Leydig cells, have all been reported to decrease in the elderly. Leydig cell activity is driven by LH: the decrease in Leydig cell function in the elderly, as part of what has been described as the normal ageing process, may be affected by changes in the secretion of LH, which is controlled by the hypothalamus. The volume occupied by the seminiferous tubules decreases, whereas that occupied by interstitial tissue remains approximately constant. The most frequently observed histological change in the ageing testis is variation in spermatogenesis in different seminiferous tubules, so that it is complete, though reduced, in some, but absent in others, when sclerosis may occur. In tubules where spermatogenesis is complete, morphological abnormalities may be observed in the germ cells, including multinucleation. Germ cell loss generally begins with the spermatids, but progressively affects the earlier germ cell types, i.e. the spermatocytes and spermatogonia. Sertoli cells are also affected by ageing, and show a range of morphological changes including dedifferentiation, mitochondrial metaplasia and multinucleation. In the Leydig cells there is a decrease in the quantity of smooth endoplasmic reticulum and mitochondria, while lipid droplets, crystalline inclusions and residual bodies increase and some cells become multinucleate. Tubules in which the entire epithelium has been lost have been observed in testes where other tubules appeared normal. The development of tubular involution with advancing age is similar to that observed after experimental ischaemia, suggesting that vascular lesions may be involved in age-related testicular atrophy. However, there is no abrupt change in testicular function equivalent to the female climacteric. REFERENCES
Kerr JB 1991 Ultrastructure of the seminiferous epithelium and intertubular tissue of the human testis. J Electron Microsc Tech 19: 215-40. Reviews the ultrastructural features of the human testis, with emphasis on spermatogenesis and cytology. Medline Similar articles Full article Johnson AD, Gomes WR (eds) 1977 The Testis, Vol 4. Advances in Physiology, Biochemistry and Function. New York: Academic Press. A general review of the blood-testis barrier.
© 2008 Elsevier
98 MALE REPRODUCTIVE SYSTEM Vas deferens and ejaculatory ducts VAS DEFERENS (Fig. 98.1) The vas deferens is a muscular tube, 45 cm long, which conveys sperm to the ejaculatory ducts, and is the distal continuation of the epididymis, starting at the epididymal tail (Fig. 97.2). At first it is very tortuous, but it becomes straighter, and ascends along the posterior aspect of the testis, medial to the epididymis. From the superior pole of the testis it ascends in the posterior part of the spermatic cord, and traverses the inguinal canal. At the internal (deep) inguinal ring the vas deferens leaves the cord, curves round the lateral side of the inferior epigastric artery and ascends for c.2.5 cm anterior to the external iliac artery. It then turns back and inclines slightly down and obliquely across the external iliac vessels to enter the lesser pelvis, where, situated retroperitoneally, it continues posteriorly, medial to the obliterated umbilical artery, the obturator nerve and vessels, and the vesical vessels (Fig. 93.1). It crosses the ureter (Fig. 98.1) and bends acutely to pass anteromedially between the posterior surface of the bladder and the upper pole of the seminal vesicle. It then descends in contact with the seminal vesicle, gradually approaching the opposite duct. Here it lies between the base of the bladder and the rectum, from which it is separated by Denonvillier's fascia. It finally descends to the base of the prostate, where it joins the duct of the seminal vesicle at an acute angle to form the ejaculatory duct (Fig. 101.1). It feels cord-like when grasped because of its thick wall and small lumen. Posterior to the bladder the lumen becomes dilated and tortuous and is termed the ampulla; beyond this, where it joins the duct of the seminal vesicle, it is again greatly diminished in calibre (Fig. 98.1).
Figure 98.1 Posterosuperior aspect of the male internal urogenital organs.
The vasa deferentia may be congenitally absent, most commonly in association with the presence of one or two abnormal gene loci at the cystic fibrosis site. This condition results in azoospermia.
ABERRANT DUCTULES A narrow, blind, caudal aberrant ductule often occurs, usually connected with the caudal part of the epididymal duct or with the start of the vas deferens. Uncoiled, it varies in length from 5 to 35 cm; it may be dilated near its end, but is otherwise uniform in calibre. In structure it is similar to the vas deferens. Occasionally it is not connected with the epididymis. A rostral aberrant ductule may occur in the epididymal head, connected with the rete testis. Aberrant ductules are derived from mesonephric tubules (p. 1375).
PARADIDYMIS The paradidymis is a small collection of convoluted tubules, found anteriorly in the spermatic cord above the epididymal head. The tubules are lined by ciliated columnar epithelium and probably represent the remains of the mesonephros (p. 1375).
EJACULATORY DUCTS
The ejaculatory ducts are formed on each side by the union of the duct of the seminal vesicle with the ampulla of the vas (Figs 93.1, 101.1). Each is almost 2 cm in length, starts from the base of the prostate, runs anteroinferiorly between its median and right or left lobes, and skirts the prostatic utricle to end on the verumontanum at two slit-like orifices on, or just within, the utricular opening. The ducts diminish and converge towards their ends.
© 2008 Elsevier
VAS DEFERENS (Fig. 98.1) The vas deferens is a muscular tube, 45 cm long, which conveys sperm to the ejaculatory ducts, and is the distal continuation of the epididymis, starting at the epididymal tail (Fig. 97.2). At first it is very tortuous, but it becomes straighter, and ascends along the posterior aspect of the testis, medial to the epididymis. From the superior pole of the testis it ascends in the posterior part of the spermatic cord, and traverses the inguinal canal. At the internal (deep) inguinal ring the vas deferens leaves the cord, curves round the lateral side of the inferior epigastric artery and ascends for c.2.5 cm anterior to the external iliac artery. It then turns back and inclines slightly down and obliquely across the external iliac vessels to enter the lesser pelvis, where, situated retroperitoneally, it continues posteriorly, medial to the obliterated umbilical artery, the obturator nerve and vessels, and the vesical vessels (Fig. 93.1). It crosses the ureter (Fig. 98.1) and bends acutely to pass anteromedially between the posterior surface of the bladder and the upper pole of the seminal vesicle. It then descends in contact with the seminal vesicle, gradually approaching the opposite duct. Here it lies between the base of the bladder and the rectum, from which it is separated by Denonvillier's fascia. It finally descends to the base of the prostate, where it joins the duct of the seminal vesicle at an acute angle to form the ejaculatory duct (Fig. 101.1). It feels cord-like when grasped because of its thick wall and small lumen. Posterior to the bladder the lumen becomes dilated and tortuous and is termed the ampulla; beyond this, where it joins the duct of the seminal vesicle, it is again greatly diminished in calibre (Fig. 98.1).
Figure 98.1 Posterosuperior aspect of the male internal urogenital organs.
The vasa deferentia may be congenitally absent, most commonly in association with the presence of one or two abnormal gene loci at the cystic fibrosis site. This condition results in azoospermia.
ABERRANT DUCTULES A narrow, blind, caudal aberrant ductule often occurs, usually connected with the caudal part of the epididymal duct or with the start of the vas deferens. Uncoiled, it varies in length from 5 to 35 cm; it may be dilated near its end, but is otherwise uniform in calibre. In structure it is similar to the vas deferens. Occasionally it is not connected with the epididymis. A rostral aberrant ductule may occur in the epididymal head, connected with the rete testis. Aberrant ductules are derived from mesonephric tubules (p. 1375).
PARADIDYMIS The paradidymis is a small collection of convoluted tubules, found anteriorly in the spermatic cord above the epididymal head. The tubules are lined by ciliated columnar epithelium and probably represent the remains of the mesonephros (p. 1375).
EJACULATORY DUCTS
The ejaculatory ducts are formed on each side by the union of the duct of the seminal vesicle with the ampulla of the vas (Figs 93.1, 101.1). Each is almost 2 cm in length, starts from the base of the prostate, runs anteroinferiorly between its median and right or left lobes, and skirts the prostatic utricle to end on the verumontanum at two slit-like orifices on, or just within, the utricular opening. The ducts diminish and converge towards their ends.
© 2008 Elsevier
VASCULAR SUPPLY, LYMPHATIC DRAINAGE AND INNERVATION Each vas deferens has its own artery, usually derived from the superior vesical artery, which anastomoses with the testicular artery to supply the epididymis and testis. Veins drain from the vas deferens and seminal vesicles to the pelvic venous plexus. Lymphatic vessels drain to the external and internal iliac nodes. The vasa deferentia are innervated by a rich autonomic plexus composed mainly of sympathetic nerve fibres derived from the pelvic plexus.
© 2008 Elsevier
MICROSTRUCTURE The wall of the vas deferens has loose connective tissue externally, a thick middle muscular layer and an internal mucosal layer. The muscular layer is composed of smooth muscle fibres arranged in external longitudinal and internal circular sheets. An additional internal longitudinal layer is present at the origin of the duct where it leaves the tail of the epididymis, however all muscle layers intermingle. The mucosa is folded longitudinally and its epithelium is columnar and nonciliated through most of the duct. Towards its distal end a pseudostratified columnar epithelium appears, the tallest cells of which bear non-motile stereocilia (elongated microvilli), similar to those of the epididymis. The connective tissue of the lamina propria contains elastic fibres. The walls of the ejaculatory ducts are thin. They contain an outer fibrous layer, which is much reduced beyond their entrance into the prostate, a thin layer of smooth muscle fibres with an outer circular and inner longitudinal orientation, and a mucosa lined by columnar epithelium. The ducts dilate during ejaculation. page 1311 page 1312
© 2008 Elsevier
99 MALE REPRODUCTIVE SYSTEM Spermatic cords and scrotum SPERMATIC CORD As the testis traverses the abdominal wall into the scrotum during early life, it carries its vessels, nerves and vas deferens with it. These meet at the deep inguinal ring to form the spermatic cord, which suspends the testis in the scrotum and extends from the deep inguinal ring to the posterior aspect of the testis. The left cord is a little longer than the right. Between the superficial ring and testis the cord is anterior to the rounded tendon of adductor longus. It is crossed anteriorly by the superficial and posteriorly by the deep external pudendal arteries respectively. The cord traverses the inguinal canal with its walls as relations: the ilioinguinal nerve is inferior. In the canal the cord acquires coverings from the layers of the abdominal wall, which extend into the scrotal wall as the internal spermatic, cremasteric and external spermatic fasciae. The internal spermatic fascia is a thin, loose layer around the spermatic cord and it is derived from the transversalis fascia. The cremasteric fascia contains fasciculi of skeletal muscle united by loose connective tissue to form the cremaster, which is continuous with internal oblique. The external spermatic fascia, a thin fibrous stratum continuous above with the aponeurosis of external oblique, descends from the crura of the superficial ring. The spermatic cord contains the vas deferens; testicular artery and veins, cremasteric artery (a branch of the inferior epigastric artery) and artery to the vas deferens (from the superior vesical artery); genital branch of the genitofemoral nerve, cremasteric nerve and sympathetic components of the testicular plexus, which are joined by filaments from the pelvic plexus accompanying the artery to the vas deferens; 4-8 lymph vessels draining the testis. All of these structures are conjoined by loose connective tissue.
© 2008 Elsevier
SPERMATIC CORD As the testis traverses the abdominal wall into the scrotum during early life, it carries its vessels, nerves and vas deferens with it. These meet at the deep inguinal ring to form the spermatic cord, which suspends the testis in the scrotum and extends from the deep inguinal ring to the posterior aspect of the testis. The left cord is a little longer than the right. Between the superficial ring and testis the cord is anterior to the rounded tendon of adductor longus. It is crossed anteriorly by the superficial and posteriorly by the deep external pudendal arteries respectively. The cord traverses the inguinal canal with its walls as relations: the ilioinguinal nerve is inferior. In the canal the cord acquires coverings from the layers of the abdominal wall, which extend into the scrotal wall as the internal spermatic, cremasteric and external spermatic fasciae. The internal spermatic fascia is a thin, loose layer around the spermatic cord and it is derived from the transversalis fascia. The cremasteric fascia contains fasciculi of skeletal muscle united by loose connective tissue to form the cremaster, which is continuous with internal oblique. The external spermatic fascia, a thin fibrous stratum continuous above with the aponeurosis of external oblique, descends from the crura of the superficial ring. The spermatic cord contains the vas deferens; testicular artery and veins, cremasteric artery (a branch of the inferior epigastric artery) and artery to the vas deferens (from the superior vesical artery); genital branch of the genitofemoral nerve, cremasteric nerve and sympathetic components of the testicular plexus, which are joined by filaments from the pelvic plexus accompanying the artery to the vas deferens; 4-8 lymph vessels draining the testis. All of these structures are conjoined by loose connective tissue.
© 2008 Elsevier
SCROTUM (Fig. 99.1) The scrotum is a cutaneous fibromuscular sac containing the testes and lower parts of the spermatic cords. It hangs below the pubic symphysis between the anteromedial aspects of the thighs. It is divided into right and left halves by a cutaneous raphe, which continues ventrally to the inferior penile surface and dorsally along the midline of the perineum to the anus. The raphe indicates the bilateral origin of the scrotum from the genital swellings. The left side of the scrotum is usually lower because the left spermatic cord is longer. The external appearance varies. When warm, and in the elderly and debilitated, the scrotum is smooth, elongated and flaccid. When cold, and in the young and robust, it is short, corrugated and closely applied to the testes because of the contraction of the dartos muscle. It consists of skin, dartos muscle and external spermatic, cremasteric and internal spermatic fasciae. The internal spermatic fascia is loosely attached to the parietal layer of the tunica vaginalis (Fig. 99.1). The scrotal skin is thin, pigmented and often rugose. It bears thinly scattered, crisp hairs, whose roots are visible through the skin. It has sebaceous glands, whose secretion has a characteristic odour, and also numerous sweat glands, pigment cells and nerve endings. These nerve endings respond to mechanical stimulation of the hairs and skin and to variations in temperature. There is no subcutaneous adipose tissue.
Figure 99.1 Transverse section through the left half of the scrotum and the left testis. The tunica vaginalis is represented as artificially distended (as would occur in a hydrocoele) to show its visceral and parietal layers.
The dartos muscle is a thin layer of smooth muscle, which is continuous beyond the scrotum with the superficial inguinal and perineal fasciae. It extends into the scrotal septum, which connects the raphe to the inferior surface of the penile radix and divides the scrotum into two cavities. The septum contains all the layers of scrotal wall except skin. The dartos muscle is closely united to the skin, but is connected to subjacent parts by delicate loose connective tissue, giving it marked independence. A fibromuscular 'scrotal ligament' extends from the dartos sheet to the inferior testicular pole, and may play a role in thermoregulation of the testis.
© 2008 Elsevier
VASCULAR SUPPLY AND LYMPHATIC DRAINAGE OF THE SCROTUM Arteries supplying the scrotum (Figs 108.4, 100.3) The arteries supplying the scrotum include the external pudendal branches of the femoral artery, the scrotal branches of the internal pudendal artery, and a cremasteric branch from the inferior epigastric artery. Dense subcutaneous plexuses of scrotal vessels carry a substantial blood flow, which facilitates heat loss. Arteriovenous anastomoses of a simple but large-calibre type are also prominent. Veins (Fig. 93.3) The veins follow the corresponding arteries. Lymphatic drainage The skin of the scrotum is drained by vessels, which accompany the external pudendal blood vessels to the superficial inguinal nodes.
© 2008 Elsevier
INNERVATION page 1313 page 1314
The scrotum is innervated by the ilioinguinal nerve, the genital branch of the genitofemoral nerve, the two posterior scrotal branches of the perineal nerve, and the perineal branch of the posterior femoral cutaneous nerve. The anterior third of the scrotum is supplied mainly from the first lumbar spinal segment (by way of the ilioinguinal and genitofemoral nerves), while the posterior two-thirds are innervated principally from the third sacral spinal cord segment (via the perineal and posterior femoral cutaneous nerves). The ventral axial line of the lower limb passes between these areas. A spinal anaesthetic, therefore, must be injected much higher to anaesthetize the anterior region.
© 2008 Elsevier
100 MALE REPRODUCTIVE SYSTEM Penis The penis, the male copulatory organ, consists of an attached root (radix) in the perineum and a free, normally pendulous, body (corpus), which is completely enveloped in skin.
© 2008 Elsevier
SKIN The penile skin is remarkably thin, dark and loosely connected to the tunica albuginea. At the corona of the penis it is folded to form the prepuce or foreskin, which variably overlaps the glans. The internal preputial layer is confluent at the neck with the thin skin covering and adhering firmly to the glans, and by this with the urethral mucosa at the external urethral orifice. On the urethral aspect of the glans a median fold, the frenulum, passes from the deep surface of the prepuce to the glans immediately proximal to the orifice. Cutaneous sensitivity, which is high over the surface of the glans, is accentuated near the frenulum. The prepuce and glans penis enclose a potential cleft, the preputial sac, and two shallow fossae flank the frenulum.
© 2008 Elsevier
ROOT (Fig. 100.1) The root of the penis consists of three masses of erectile tissue in the urogenital triangle, namely the two crura and the bulb, firmly attached to the pubic arch and perineal membrane respectively. The crura are the posterior regions of the corpora, and the bulb is the posterior end of the corpus spongiosum. Each penile crus (Fig. 100.1) starts behind as a blunt, elongate but rounded process, attached firmly to the everted edge of the ischiopubic ramus and covered by ischiocavernosus. Anteriorly it converges towards its fellow and is slightly enlarged posterior to this. Near the inferior symphyseal border the two crura come together and continue as the corpora cavernosa of the body of the penis. The bulb of the penis (Fig. 100.1) lies between the crura and is firmly connected to the inferior aspect of the perineal membrane, from which it receives a fibrous covering. Oval in section, the bulb narrows anteriorly into the corpus spongiosum, down and forwards at this point. Its convex superficial surface is covered by bulbospongiosus. Its flattened deep surface is pierced above its centre by the urethra, which traverses it to reach the corpus spongiosum.
© 2008 Elsevier
BODY (Fig. 100.2) The body of the penis contains three elongated erectile masses, capable of considerable enlargement when engorged with blood during erection. When flaccid the penis is cylindrical, but when erect it is triangular with rounded angles. The surface which is posterosuperior during erection, is termed the dorsum of the penis and the opposite aspect is the ventral surface. The erectile masses are the right and left corpora cavernosa, and the median corpus spongiosum, which are continuations of the crura and bulb of the penis respectively.
CORPORA CAVERNOSA
Figure 100.1 Ventral aspect of the constituent erectile masses of the penis in erect position. The glans penis and the distal part of the corpus spongiosum are shown detached from the corpora cavernosa penis and turned to the left.
The corpora cavernosa of the penis form most of the body. In close apposition throughout, they share a common fibrous envelope and are separated only by a median fibrous septum. On the urethral surface their combined mass has a wide median groove, adjoining the corpus spongiosum (Fig. 100.2); dorsally a similar but narrower groove contains the deep dorsal vein. The corpora end distally in the hollow, proximal aspect of the glans penis in a rounded cone, on which each has a small terminal projection (Fig. 100.1). They are enclosed in a strong fibrous tunica albuginea, consisting of superficial and deep strata. The superficial fibres are longitudinal, and form a single tube round both corpora. The deep fibres are circularly orientated and surround each corpus separately, joining together as a median septum of the penis. The median septum is thick and complete proximally so that the corporal bodies can be separated proximally for 5-7 cm. Distally it consists of a pectiniform (comb-like) series of bands and is called the pectiniform septum, which is incomplete and allows cross-circulation of blood between the two corpora.
CORPUS SPONGIOSUM The corpus spongiosum of the penis is traversed by the urethra. It adjoins the median groove on the urethral surface of the conjoined corpora cavernosa. It is cylindrical, tapering slightly distally, and surrounded by a tunica albuginea. Near the end of the penis it expands into a somewhat conical enlargement, the glans penis (Figs 100.1, 100.2). page 1315 page 1316
Figure 100.2 Transverse section of penis. (By permission from Eardley I, Sethia K 1998 Erectile Dysfunction. London: Mosby.)
The glans penis projects dorsally over the end of the corpora cavernosa, and has a shallow concave surface to which they are attached. The corona glandis projects from its base, overhanging an obliquely grooved neck of the penis. Numerous small preputial glands on the corona glandis and penile neck secrete
sebaceous smegma. The navicular fossa of the urethra is in the glans and opens by a sagittal slit on or near its apex.
© 2008 Elsevier
SUPERFICIAL PENILE FASCIA The superficial penile fascia is devoid of fat, and consists of loose connective tissue, invaded by a few fibres of dartos muscle from the scrotum (p. 1313). Indeed, clinically, it is commonly call the dartos layer. As in the suprapubic abdominal wall, the deepest layer is condensed to form a distinct tough fascial sheath known as Buck's fascia. It surrounds both corpora cavernosa and splits to enclose the corpus spongiosum, separating the superficial and deep dorsal veins. At the penile neck it blends with the fibrous covering of all three corpora. Proximally, it is continuous with the dartos muscle and with the fascia covering the urogenital region of the perineum (p. 1367).
© 2008 Elsevier
SUSPENSORY LIGAMENTS OF PENIS The body of the penis is supported by two ligaments, the fundiform and triangular ligaments, which are continuous with its fascia and consist largely of elastin fibres. The fundiform ligament, stems from the lowest part of the linea alba, and splits into two lamellae which skirt the penis and unite below with the scrotal septum. The triangular suspensory ligament, deep to the fundiform ligament, is attached above to the front of the pubic symphysis, and blends below, on each side, with the fascia penis.
© 2008 Elsevier
VASCULAR SUPPLY AND LYMPHATIC DRAINAGE PERINEAL ARTERY (Fig. 100.3) The perineal artery leaves the internal pudendal artery (Fig. 108.4) near the anterior end of its canal and approaches the scrotum in the superficial perineal region, between bulbospongiosus and ischiocavernosus. Beyond the perineal membrane, and near its base, a small transverse branch passes medially, inferior to the superficial transverse perineal muscle, to anastomose with its contralateral fellow, and with the posterior scrotal and inferior rectal arteries; collectively these vessels supply tissues between the anus and the penile bulb. The posterior scrotal arteries, distributed to the scrotal skin, dartos and perineal muscles, are usually terminal branches of the perineal artery but may also arise from its transverse branch.
ARTERY OF THE BULB OF THE PENIS The artery of the bulb of the penis is short but wide. It runs medially through the deep transverse perineal muscle to the penile bulb, which it penetrates. It supplies the corpus spongiosum and the bulbourethral gland.
Figure 100.3 The superficial branches of the internal pudendal artery in the male.
CAVERNOSAL ARTERY (DEEP ARTERY OF THE PENIS) The cavernosal artery is a terminal branch of the internal pudendal artery. It passes through the perineal membrane to enter the crus penis. It runs the length of the corpus cavernosum and supplies its erectile tissue. Within the corpus the cavernosal arteries divide into branches running in the trabeculae. Some end directly in capillary networks which open into the cavernous spaces, and others
become convoluted and somewhat dilated helicine arteries, which then open into the cavernous spaces. Helicine arteries are most abundant in the posterior regions of the corpora cavernosa.
DORSAL ARTERY OF THE PENIS The dorsal artery of the penis is the other terminal branch of the internal pudendal artery. It runs between the crus penis and pubic symphysis, and then pierces the suspensory ligament of the penis to run along its dorsum to the glans, where it forks into branches to the glans and prepuce. In the penis it lies deep to Buck's fascia between the dorsal nerve and deep dorsal vein, the latter being most medial. It supplies penile skin by branches which run through the dartos layer. It gives off circumflex branches which run around the shaft of the penis deep to and then within Buck's fascia to supply the tunica albuginea of the corpus cavernosum, anastomosing through the tunica with the cavernosal system. These vessels also supply the corpus spongiosum.
DORSAL VEINS OF THE PENIS (Fig. 93.3) The veins which drain the corpora cavernosa leave the corpora by passing obliquely through the tunica albuginea via a series of small vessels. These small veins run into the circumflex veins which run circumferentially around the shaft of the penis from its ventral aspect, where they receive tributaries from the corpus spongiosum, to its dorsal aspect, where they drain into the deep dorsal vein. The dorsal veins, superficial and deep, are unpaired. The superficial dorsal vein drains the prepuce and penile skin. It runs back in subcutaneous tissue and inclines right or left, before it opens into one of the external pudendal veins. The deep dorsal vein lies deep to Buck's fascia. It receives blood from the glans penis and corpora cavernosa penis, and courses back in the midline between the paired dorsal arteries. Near the root of the penis it passes deep to the suspensory ligament and through a gap between the arcuate pubic ligament and anterior margin of the perineal membrane. It divides into right and left branches which connect below the symphysis pubis with the internal pudendal veins and ultimately enter the prostatic plexus. page 1316 page 1317
LYMPHATIC DRAINAGE OF THE PENIS The penile skin is drained by vessels, which, with those of the perineal skin, accompany the external pudendal blood vessels to the superficial inguinal nodes. Lymph vessels from the glans pass to the deep inguinal and external iliac nodes. Lymph vessels from the erectile tissue and penile urethra pass to the internal iliac lymph nodes.
© 2008 Elsevier
INNERVATION The nerves to the corpora cavernosa form two groups, the lesser and greater cavernous nerves, which arise from the front of the pelvic (inferior hypogastric) plexus and join branches from the pudendal nerve before passing below the pubic arch. Lesser cavernous nerves pierce the fibrous penile sheath proximally to supply the erectile tissue of the corpus spongiosum and penile urethra. Greater cavernous nerves proceed on the dorsum of the penis, where they connect with the dorsal nerve, and supply the erectile tissue: some filaments reach the erectile tissue of the corpus spongiosum. Stimulation of the sympathetic supply to the male genital organs produces vasoconstriction (the parasympathetic is vasodilator), contraction of the seminal vesicles and prostate, and seminal emission. Parasympathetic fibres are vasodilator and come from the second, third and fourth sacral spinal segments via the pudendal nerve and pelvic plexuses. On the glans and bulb of the penis some cutaneous filaments innervate lamellated corpuscles and many terminate in characteristic end bulbs.
© 2008 Elsevier
MICROSTRUCTURE The internal surfaces of the fibrous sheaths of the corpora cavernosa and their dividing septum give rise to numerous trabeculae. These cross the corpora cavernosa in all directions and divide them into a series of cavernous spaces, which gives them a spongy appearance (Fig. 100.2). The trabeculae are composed of collagen and elastin fibres, and smooth muscle cells. They contain numerous vessels and nerves. The cavernous spaces are filled with blood during erection, but many are empty in the flaccid penis. They are lined by flat nonfenestrated endothelial cells. The fibrous tunica albuginea of the corpus spongiosum is thinner, whiter and more elastic than that of the corpora cavernosa. It is formed partly of smooth muscle cells: a layer of the same tissue surrounds the urethral epithelium and the paraurethral glands.
© 2008 Elsevier
ERECTION AND EJACULATION Erection is purely vascular. It occurs in response to parasympathetic stimulation and is independent of compression by the ischiocavernosi and bulbospongiosus, although these may contribute to maximum rigidity. Sexual arousal leads to rapid inflow from the helicine arteries following relaxation of the smooth muscle of the corpora cavernosa, an event which is dependent on the production of nitrous oxide and cyclic GMP. This inflow of blood fills the cavernous spaces leading to tumescence. The resulting distension converts tumescence to erection by pressure on the subtunical veins, which drain the erectile tissue, thereby obstructing them. The pressure within the corpora cavernosa is maintained at 100 mmHg to maintain penile erection. Continuing cutaneous stimulation of the glans and frenulum contributes significantly to maintaining erection and initiating orgasm and ejaculation. Erection is thus dependent on a normal psychogenic response to stimulation, intact parasympathetic nerves, corporal smooth muscle capable of relaxation, patent arteries capable of delivering blood at the required rate, and a normal venous system. Ejaculation consists of two processes: emission and ejaculation. Emission is the transmission of seminal fluid from the vasa, prostate and seminal vesicles into the prostatic urethra under sympathetic control. Ejaculation is the onward transmission of seminal fluid from the prostatic urethra to the exterior. This has autonomic and somatic components. The first discernible part of the process is contraction of bulbospongiosus, which contracts about six times under somatic control. The way in which seminal fluid crosses the external urethral sphincter into the bulbar urethra is not clear: it is known to be under autonomic control and is timed such that from the second to the final contraction of bulbospongiosus the ejaculate appears from the external meatus, in some younger men in a pulsatile fashion. Failure to achieve tumescence with adequate stimulation is termed impotence or more recently and politically correctly, erectile dysfunction. The mechanism of erection is complex: failure in any of the previously mentioned components can result in impotence. The commonest causes include psychogenic disturbance with failure to relax cavernous smooth muscle; arterial insufficiency, as a result of atheromatous disease; and damage to the parasympathetic nervous system secondary to diabetes or following pelvic surgery such as radical prostatectomy, radical cystectomy or bowel resection. Pharmacotherapy is predominantly directed at achieving cavernosal smooth muscle relaxation. Detumescence is effected through the sympathetic pathway. Failure of an erection to detumesce is termed priapism. This can occur spontaneously but is most commonly seen with conditions that impair blood flow by increasing its viscosity such as sickle cell anaemia or leukaemia, or as a consequence of drug treatment when given by injection. These conditions result in ischaemia of the corporal smooth muscle, which causes pain within the penis. Peyronie's disease produces a bend in the erect penis. This is most commonly a
dorsal curvature and results from a localized thickening or plaque of the corpora cavernosa which prevents expansion of a segment during erection. REFERENCES Lepor H, Gregerman M, Crosby R, Mostofi FK, Walsh PC 1985 Precise localization of the autonomic nerves of the pelvic plexus to the corpora cavernosa: a detailed anatomical study of the adult male pelvis. J Urol 133: 207-212. Medline Similar articles Mundy AR, Fitzpatrick J, Neal D, George N (eds) 1999 Male sexual function. In: The Scientific Basis of Urology, Chapter 12. Isis Medical Media: 243-53. page 1317 page 1318
© 2008 Elsevier
101 MALE REPRODUCTIVE SYSTEM Accessory glandular structures SEMINAL VESICLES (Figs 101.1, 101.2) The two seminal vesicles are sacculated, contorted tubes located between the bladder and rectum (Figs 98.1, 101.1). Each vesicle is c.5 cm long, somewhat pyramidal, the base being directed up and posterolaterally. Essentially, the seminal vesicle is a single coiled tube with irregular diverticula (Fig. 101.1); the coils and diverticula are connected by fibrous tissue. The diameter of the tube is 3-4 mm and its uncoiled length is 10-15 cm. The upper pole is a cul-de-sac, the lower narrows to a straight duct, which joins the vas deferens to form the ejaculatory duct. The anterior surface contacts the posterior aspect of the bladder, and extends from near the entry of the ureter to the prostatic base. The posterior surface is related to the rectum, from which it is separated by Denonvillier's fascia. The seminal vesicles diverge superiorly. They are related to the vas deferens and the terminations of the ureters, and are partly covered by peritoneum. Each has a dense, fibromuscular sheath. Along the medial margin of each vesicle is the ampulla of the vas deferens. The veins of the prostatic venous plexus, which drain posteriorly to the internal iliac veins, lie laterally.
© 2008 Elsevier
SEMINAL VESICLES (Figs 101.1, 101.2) The two seminal vesicles are sacculated, contorted tubes located between the bladder and rectum (Figs 98.1, 101.1). Each vesicle is c.5 cm long, somewhat pyramidal, the base being directed up and posterolaterally. Essentially, the seminal vesicle is a single coiled tube with irregular diverticula (Fig. 101.1); the coils and diverticula are connected by fibrous tissue. The diameter of the tube is 3-4 mm and its uncoiled length is 10-15 cm. The upper pole is a cul-de-sac, the lower narrows to a straight duct, which joins the vas deferens to form the ejaculatory duct. The anterior surface contacts the posterior aspect of the bladder, and extends from near the entry of the ureter to the prostatic base. The posterior surface is related to the rectum, from which it is separated by Denonvillier's fascia. The seminal vesicles diverge superiorly. They are related to the vas deferens and the terminations of the ureters, and are partly covered by peritoneum. Each has a dense, fibromuscular sheath. Along the medial margin of each vesicle is the ampulla of the vas deferens. The veins of the prostatic venous plexus, which drain posteriorly to the internal iliac veins, lie laterally.
© 2008 Elsevier
BULBOURETHRAL GLANDS The two bulbourethral glands are small round yellow somewhat lobulated masses c.1 cm in diameter (Fig. 109.19). They lie lateral to the membranous urethra above the perineal membrane and penile bulb and are enclosed by fibres of the urethral sphincter. The excretory duct of each, almost 3 cm long, passes obliquely forwards external to the mucosa of the membranous urethra and penetrates the perineal membrane. It opens by a minute orifice on the floor of the bulbar urethra c.2.5 cm below the perineal membrane. In later decades the glands generally diminish in size.
Figure 101.1 Anterior aspect of the seminal vesicles, terminal parts of the vasa deferentia and the prostate. The lamina of the right seminal vesicle, the ampulla of the right vas deferens and of the prostatic part of the urethra have been exposed by appropriate removal of tissues.
© 2008 Elsevier
VASCULAR SUPPLY AND LYMPHATIC DRAINAGE The arteries to the seminal vesicles are derived from the inferior vesical and middle rectal arteries. The veins and lymphatics accompany these arteries.
© 2008 Elsevier
INNERVATION The innervation of the seminal vesicles and bulbourethral glands is derived from the pelvic plexuses.
© 2008 Elsevier
MICROSTRUCTURE SEMINAL VESICLE The paired seminal vesicles, together with the ampulla of the vas deferens and the ejaculatory ducts, form a functional unit which develops slowly until the onset of puberty. After puberty the vesicles form sac-like structures which contribute up to 85% of the seminal fluid. They are mainly concerned with secretion of seminal coagulating proteins, fructose, prostaglandins and other specific proteins in an alkaline, viscous yellowish fluid. The wall of the seminal vesicle is composed of an external connective tissue layer, a middle smooth muscle layer - which is thinner than in the vas deferens and arranged in external longitudinal and internal circular layers - and an internal mucosal layer with a highly folded, labyrinthine structure. The cuboidal to pseudostratified columnar epithelium of the mucosa shows features typical of protein-secreting cells. The vesicles are not reservoirs for spermatozoa as their name suggests, because spermatozoa are stored mainly in the epididymis. They contract during ejaculation, and their secretion forms most of the ejaculate.
page 1319 page 1320
Figure 101.2 Axial oblique MRI (STIR sequence) demonstrates normal high signal within prominent seminal vesicles lying above the prostate in a young man.
BULBOURETHRAL GLANDS Each bulbourethral gland consists of several lobules held together by a fibrous capsule. The secretory units are mainly tubulo-alveolar in form. The glandular epithelium, which is columnar, secretes acid and neutral mucins into the penile (spongiose) urethra prior to ejaculation, and these have a lubricating function. The main secretory duct is lined by a stratified columnar epithelium. Diffuse lymphoid tissue (MALT; p. 77) is associated with the glands.
© 2008 Elsevier
102 FEMALE REPRODUCTIVE SYSTEM Ovaries The ovaries are paired ovoid structures, homologous with the testes, but smaller (Figs 102.1, 102.2, 102.3, 102.4, 103.2). They have an average volume of 11 cm3 in reproductively mature women. The ovaries are dull white in colour and consist of dense fibrous tissue in which ova are embedded. Before regular ovulation begins they have a smooth surface, but thereafter their surfaces are distorted by scarring that follows the degeneration of successive corpus lutea. In embryonic and early fetal life the ovaries are situated in the lumbar region near the kidneys, and they gradually descend into the lesser pelvis. In the neonate, they are c.1.3 cm long, 0.6 cm wide, and 0.4 cm thick. Prior to the first menstrual period (menarche) the ovaries are about a third of the normal reproductive adult size. They gradually increase in size with body growth. They are mobile structures and may change their position to some extent according to the state of the surrounding organs such as the intestines.
Figure 102.1 A, Median sagittal section through a human female pelvis. Peritoneum: blue. B, Sagittal section showing the peritoneal attachments of the ovary.
During pregnancy, the ovaries are lifted high in the pelvis and, by 14 weeks of gestation, become partly abdominal structures. By the third trimester they are totally abdominal structures and lie vertically behind and lateral to the parous uterus. The ovaries more than double their size by term. Following childbirth, the ovarian position varies. They are displaced in the first pregnancy and usually never return to their original location. During the early menopause, the average size of the ovary is 2.0 ! 1.5 ! 0.5 cm and this reduces to 1.5 ! 7.5 ! 0.5 cm in late menopause. Accessory ovaries may occur in the mesovarium or in the adjacent part of the broad ligament. The description that follows refers to the ovarian condition in nulliparous women, except where otherwise stated.
© 2008 Elsevier
RELATIONS (Fig. 102.4) page 1321 page 1322
Figure 102.2 Posterosuperior aspect of the uterus and the left broad ligament. The ligament has been spread out and the ovary is displaced downwards.
Figure 102.3 Posterior aspect of a cleared injected specimen to show the distribution of the left uterine and ovarian arteries of a female aged 17 years. (Prepared by Hamilton Drummond.)
The ovaries lie on each side of the uterus close to the lateral pelvic wall (Fig. 102.4). They are suspended in the pelvic cavity by a double fold of peritoneum, the mesovarium, which attaches to the upper limit of the posterior aspect of the broad uterine ligament (Fig. 102.2). The long axis of each ovary is vertical in the erect position. During pregnancy they may be horizontal or oblique. Compare Figs 102.1 and 102.2.
Figure 102.4 Axial T2-weighted images through the female pelvis showing the anteverted uterus with ovaries on either side close to the pelvic wall. Both ovaries contain multiple small high-signal follicles.
The ovary has lateral and medial surfaces, superior and inferior extremities, or poles, and anterior and posterior borders. The lateral surface of the ovary contacts parietal peritoneum in the ovarian fossa. Behind the ovarian fossa are extraperitoneal structures, including the ureter, internal iliac vessels, obturator vessels and nerve, and the origin of the uterine artery. The medial surface faces the uterus and uterine vessels in the broad ligament, and the peritoneal recess here is termed the ovarian bursa. Above the superior extremity are the fimbria and distal section of the uterine tube. The inferior extremity points downwards towards the pelvic floor. The anterior border faces the posterior leaf of the broad ligament and contains the mesovarium. The posterior border is free and faces the peritoneum, which overlies the upper part of the internal iliac artery and vein, and the ureter. On the right side, superior and lateral to the ovary, are the ileocaecal junction, caecum and appendix. On the left side, the sigmoid colon passes over the superior pole of the ovary and joins the rectum, which lies between the medial surfaces of both ovaries.
© 2008 Elsevier
PERITONEAL ATTACHMENTS (Fig. 102.2) SUSPENSORY (INFUNDIBULOPELVIC) LIGAMENT page 1322 page 1323
The suspensory (infundibulopelvic) ligament of the ovary is a peritoneal fold, which is attached to the upper part of the lateral surface of the ovary. It contains the ovarian vessels and nerves and passes superiorly over the external iliac vessels, genitofemoral nerve and ureter (Fig. 102.1) to join the peritoneum, which covers psoas major. On the right side the suspensory ligament is attached to a fold of peritoneum that is posterior and inferior to the caecum and appendix. On the left side the peritoneal attachment is higher than on the right, and is lateral to the junction of the descending and sigmoid colons.
OVARIAN LIGAMENTS The ovarian ligament attaches the uterine (inferiomedial) extremity of the ovary to the lateral angle of the uterus, posteroinferior to the uterine tube. It lies in the posterior leaf of the broad ligament and contains some smooth muscle cells. The ovarian ligament is continuous with the medial border of the round ligament (p. 1389), both of which are remnants of the gubernaculum (Ch. 109).
MESOVARIUM The mesovarium is a short peritoneal fold, which attaches the ovary to the back of the broad ligament. It carries blood vessels and nerves to the ovarian hilum. The uterine tube arches over the ovary, and ascends in relation to its mesovarian border, then curves over its tubal end and passes down on its posterior, free border and medial surface (Fig. 102.1).
© 2008 Elsevier
VASCULAR SUPPLY AND LYMPHATIC DRAINAGE (Fig. 102.3) ARTERIES The ovarian arteries are branches of the abdominal aorta and originate below the renal arteries. They correspond to the testicular arteries (Fig. 91.1). Each descends behind the peritoneum in the paracolic gutter, and at the brim of the pelvis crosses the external iliac artery and vein to enter the true pelvic cavity. Here the artery turns medially in the ovarian suspensory ligament and continues into the uterine broad ligament, below the uterine tube. At the ovarian level it passes back in the mesovarium and divides into branches that supply the ovary. Branches also supply the uterine tube, and, on each side, a branch passes lateral to the uterus to unite with the uterine artery. Other branches accompany the round ligaments through the inguinal canal to the skin of the labium majus and the inguinal region. Early in intrauterine life the ovaries flank the vertebral column inferior to the kidneys, and so the ovarian arteries are relatively short - the arteries gradually lengthen as the ovaries descend into the pelvis. The vessels supplying the ovary dilate during pregnancy as their anastamoses form part of the uterine and tubal circulation (Fig. 102.3), and this results in a swollen mesovarium and ovary.
VEINS The ovarian veins emerge from the ovary as a plexus (pampiniform plexus) in the mesovarium and suspensory ligament. Two veins issue from the plexus and ascend with the ovarian artery. Their further course is similar to that of the testicular veins. They usually merge into a single vessel before entering either the inferior vena cava on the right side, or the renal vein on the left side. They may contain valves. The ovarian veins are much enlarged in pregnancy.
LYMPHATIC DRAINAGE Lymph drainage is via three main routes. Lymph vessels ascend along the ovarian artery to preaortic and lateral aortic nodes situated near the origin of the renal arteries. Lymph may drain directly to lumboaortic, iliac and obturator nodes, which then drain into para-aortic nodes, or it may drain along the round liagament to the inguinal nodes and across the uterus to the contralateral pelvic nodes.
© 2008 Elsevier
INNERVATION The ovarian plexuses consist of postganglionic sympathetic, parasympathetic and visceral afferent fibres. The efferent sympathetic fibres are derived from the tenth and eleventh thoracic spinal segments and are probably vasoconstrictor, whereas the parasympathetic fibres, from the inferior hypogastric plexuses, are probably vasodilator. Little is known of their actual distribution or function. Histochemical studies have demonstrated the presence of cholinergic and adrenergic nerve fibres in both the ovaries and ovarian ligaments. The nerves accompany the ovarian artery to the ovary and uterine tube. The upper part of the ovarian plexus is formed from branches of the renal and aortic plexuses, and the lower part is reinforced from the superior and inferior hypogastric plexuses. Autonomic fibres do not reach the ovarian follicles and are not required for ovulation.
REFERRED PAIN Sensory fibres accompany the sympathetic nerves, and so ovarian pain can be periumbilical. It is often perceived in the right or left iliac fossa due to local inflammation. Ovarian pain can also be perceived on the medial side of the thigh in the cutaneous distribution of the obturator nerve, presumably because the ovary lies close to the obturator nerve in the ovarian fossa, and so any inflammation of the ovary or peritoneum in the ovarian fossa may affect the obturator nerve.
© 2008 Elsevier
MICROSTRUCTURE In young females the surface of the ovary is covered by a single layer of cuboidal epithelium, which contains some flatter cells. It appears dull white, in contrast to the shiny smooth peritoneal mesothelial covering of the mesovarium, with which it is continuous. A white line around the anterior mesovarian border usually marks the transition between peritoneum and ovarian epithelium. The surface epithelium is also termed the germinal epithelium, but this is a misnomer, because it is not the source of germ cells (p. 1386). In adults, c.85% of ovarian cancers arise from neoplastic changes in the surface epithelium. Immediately beneath the epithelium there is a tough collagenous coat, the tunica albuginea. The ovarian tissue it surrounds is divisible into a cortex, which contains the ovarian follicles, and a medulla, which receives the ovarian vessels and nerves at the hilum (Fig. 102.5).
OVARIAN CORTEX Before puberty, the cortex forms c.35%, the medulla c.20%, and interstitial cells up to 45%, of the volume of the ovary. After puberty the cortex forms the major part of the ovary, and encloses the medulla except at the hilum. It contains the ovarian follicles at various stages of development, and corpora lutea and their degenerative remnants, depending on age or stage of the menstrual cycle. The follicles and structures derived from them are embedded in a dense stroma composed of a meshwork of thin collagen fibres and fusiform fibroblast-like cells arranged in characteristic whorls. Stromal cells differ from fibroblasts in general connective tissue in that they contain lipid droplets, which accumulate in pregnancy. Stromal cells give rise to the thecal layers of maturing ovarian follicles. The theca interna becomes steroid-secreting in the corpus luteum.
MEDULLA The medulla is highly vascular, much more so than the cortex. It contains numerous veins and spiral arteries, which enter the hilum from the mesovarium and lie within a loose connective tissue stroma. Small numbers of cells (hilus cells) with characteristics similar to interstitial (Leydig) cells in the testis are found in the medulla at the hilum, and they may be a source of androgens.
OVARIAN FOLLICLES Primordial follicle
The formation of the female gamete is a complex process. At birth, the ovarian cortex contains a superficial zone of primordial follicles. These consist of primary oocytes c.25 µm in diameter, each surrounded by a single layer of flat follicular cells (Fig. 102.6). The oocyte nuclei are slightly eccentric and have a characteristically prominent nucleolus. They contain the diploid number of chromosomes (duplicated as sister chromatids), arrested at the diplotene stage of meiotic prophase since before birth. The prenatal development of primordial follicles is described on page 1386. Many follicles degenerate either during prepubertal (including prenatal) life, or through atresia at some stage after beginning the process of cyclical maturation during the child-bearing period. Their remnants are visible as atretic follicles, the remains of which accumulate throughout the period of reproductive life. After puberty, a cohort of up to 20 primordial follicles, (fewer with advancing age) become activated in each menstrual cycle. Their development takes a number of cycles. Of the follicles activated in each cohort, usually only one follicle from one or other ovary becomes dominant, reaches maturity and releases its oocyte at ovulation.
page 1323 page 1324
Figure 102.5 The microstructure of the ovary and follicles at various stages in their cyclical development and the formation of corpora lutea and albicans. Note that in the human ovary, developing follicles are rarely seen.
Figure 102.6 High power micrograph of primordial follicles in the ovarian cortex of a 32-week fetus. A single layer of flattened folliclular cells surrounds each large primary oocyte. (By permission from Stevens A, Lowe JS 1996 Human Histology, 2nd edn. London: Mosby.)
Primary follicle
Figure 102.7 Primary follicle (rat) showing the zona pellucida between the oocyte and a single layer of cuboidal follicular (granulosa) cells.
The first sign of activation is a change in the follicle cells from flattened to cuboidal. This is followed by their proliferation to give rise to a multilayered follicle consisting of granulosa cells surrounded by a thick basal lamina. Stromal cells immediately surrounding the follicle begin to differentiate into spindle-shaped cells, which constitute the theca folliculi (the presumptive theca interna). They are later accompanied by a more fibrous theca externa. At the same time the oocyte increases in size and secretes a thick layer of extracellular proteoglycan-rich material, the zona pellucida, between its plasma membrane and the surrounding granulosa cells of the early follicle (Fig. 102.7). This is important for the process of fertilization (p. 185). The granulosa cells in contact with the zona pellucida send cytoplasmic processes radially inwards and these contact oocyte microvilli at gap junctions, indicating communication between them (for a review see Buccione et al 1990). The follicular cells, in particular the granulosa cells, which are in functional contact with each other through gap junctions, continue to proliferate and so the thickness of the late primary follicle wall increases. Secondary (antral) follicle page 1324 page 1325
Figure 102.8 Early antral (secondary) follicle showing the development of a fluid-filled
antrum (follicular fluid, FF) within the granulosa cell layer. The oocyte (OC) is separated from the granulosa cells (G) by a pale-staining zona pellucida (ZP). Note the relative size of the oocyte and its nucleus and prominent nucleolus, compared with follicular cells. The follicle is surrounded by a basal lamina (BL) and a thecal cell layer (T). (By permission from Dr JB Kerr, Monash University, from Kerr JB 1999 Atlas of Functional Histology. London: Mosby.)
As the number of granulosa cells continues to increase, cavities begin to form between them (Fig. 102.8) filled with a clear fluid (liquor folliculi) containing hyaluronate, growth factors, and steroid hormones secreted by the granulosa cells. The follicle is now about 200 µm in diameter and usually lies deep in the cortex. These cavities coalesce to form one large fluid-filled space, the antrum. This is surrounded by a thin, uniform layer of granulosa cells, except at one pole of the follicle where a thickened granulosa layer envelopes the eccentrically placed oocyte, to form the cumulus oophorus (Fig. 102.9). The oocyte has now reached its maximum size of about 80 µm and the inner and outer thecae are clearly differentiated. As follicles mature, the theca interna becomes more prominent and its cells more rounded and typical of steroid-secreting endocrine cells. These cells produce androstenedione from which the granulosa cells synthesize oestrogens (primarily oestradiol). Follicular development is stimulated by follicle-stimulating hormone (FSH). Tertiary follicle
Although a number of follicles may progress to the secondary stage by about the first week of a menstrual cycle, usually only one follicle, from either one of the two ovaries, proceeds to the tertiary stage, and the remainder become atretic. The surviving follicle increases in size considerably as the antrum takes up fluid from the surrounding tissues and expands to a diameter of c.2 cm. The cumulus oophorus surrounding the oocyte thins. The term Graafian follicle is often used to describe this mature follicular stage. The oocyte and a surrounding ring of tightly adherent cells, the corona radiata, breaks away from the follicle wall and floats freely in the follicular fluid. The primary oocyte, which has remained in the first meiotic prophase since fetal life, completes its first meiotic division to produce the almost equally large secondary oocyte and a minute first polar body with very little cytoplasm. The secondary haploid oocyte immediately begins its second meiotic division, but when it reaches metaphase, the process is arrested until fertilization has occurred. The follicle moves to the superficial cortex, causing the surface of the ovary to bulge. The tissues at the point of contact (the stigma) with the tough tunica albuginea and ovarian surface epithelium are eroded until the follicle ruptures and its contents are released into the peritoneal cavity for capture by the fimbria of the uterine (Fallopian) tube (p. 1329). The oocyte at ovulation is still surrounded by its zona pellucida and corona radiata of granulosa cells. If fertilization does not occur, it begins to degenerate after 24 to 48 hours.
Figure 102.9 A mature tertiary (Graafian) follicle, with an eccentrically located oocyte surrounded by a cumulus oophorus (of granulosa cells) within the fluid-filled antrum. (By permission from Stevens A, Lowe JS 1996 Human Histology, 2nd edn. London: Mosby.)
Corpus luteum
After ovulation, the walls of the empty follicle collapse and fold. The granulosa cells increase in size and synthesize a cytoplasmic carotenoid pigment (lutein) giving them a yellowish colour (hence corpus luteum). These large (30-50 µm) granulosa lutein cells form most of the corpus luteum (Fig. 102.10). The basal lamina surrounding the follicle breaks down, and numerous smaller theca lutein cells infiltrate the folds of the cellular mass, accompanied by capillaries and connective tissue. Extravasated blood from thecal capillaries accumulates in the centre as a small clot, but this rapidly resolves and is replaced by connective tissue. Ultrastructurally, all lutein cells have a cytoplasm filled with abundant smooth endoplasmic reticulum, characteristic of steroid synthesizing endocrine cells. Granulosa lutein cells secrete progesterone and oestradiol (from aromatization of androstenedione, which is synthesized by theca lutein cells). The two cell types also respond differently to circulating gonadotropins. Theca lutein cells express receptors for, and respond to, human chorionic gonadotrophin (hCG). If the oocyte is not fertilized, the corpus luteum (of menstruation) functions for c.12 to 14 days after ovulation, then atrophies. The lutein cells undergo fatty degeneration, autolysis, removal by macrophages and gradual replacement with fibrous tissue. Eventually, after c.2 months, a small, whitish scar-like corpus albicans is all that remains. If fertilization does occur, implantation of the blastocyst into the uterine endometrium usually begins seven days later and the embryonic trophoblast then starts to produce hCG. The chorionic gonadotrophin stimulates the corpus luteum of menstruation to grow, and it becomes a corpus luteum of pregnancy. It normally increases in size from c.10 mm in diameter to 25 mm around 8 weeks of gestation (p. 1323) and can be seen clearly on the ovary on ultrasound. It secretes progesterone , oestrogen and relaxin, and functions throughout pregnancy, although it gradually regresses as its endocrine functions are largely
replaced by the placenta after c.8 weeks' gestation. Its diameter is reduced by the end of pregnancy to c.1 cm. In the next few months it degenerates, like the corpus luteum of menstruation, to form a corpus albicans. page 1325 page 1326
Figure 102.10 Low-power micrograph showing the human ovary in reproductive maturity. The large structure on the right is an active corpus luteum and several other corpora lutea are at various stages of degeneration and the formation of corpora albicans. The uterine tube is seen in transverse section to the right of the ovary. (By permission from Young B, Heath JW 2000 Wheater's Functional Histology. Edinburgh: Churchill Livingstone.)
© 2008 Elsevier
OVARY IN THE MENOPAUSE At the menopause (usually in the 45-55-year-age range), ovulation ceases and various microscopic changes ensue within the ovarian tissues. The stroma becomes denser, the tunica albuginea thickens and the ovarian surface epithelium thins. However, many follicles persist within the cortex. Some lack oocytes, and others contain oocytes, providing the possibility of ovulation if the hormonal changes were to be reversed. Some abnormal follicles may become cystic as age progresses, and this is quite a common feature in later years.
© 2008 Elsevier
OVARIAN CYSTS An ovary may contain a functional or pathological cyst. Prior to ovulation a Graafian follicle can be seen on the surface of the ovary and can be as large as 1 cm. After ovulation, the corpus luteum develops, and this can also be visualized on the surface of the ovary. These normal functional cysts can sometimes be confused with ovarian cystic pathology on ultrasound. A Graafian follicle can fail to rupture and continue to enlarge, reaching sizes of 5 or 6 cm. When rupture eventually occurs the process can cause severe pain. Corpus luteal cysts can also develop into larger than average structures. This is especially so in the first trimester of pregnancy when large cysts can be identified on ultrasound scan and cause concern. They usually regress as pregnancy advances and rarely cause symptoms. Benign and malignant neoplasms of the ovary can produce fluid (usually a serous, mucinous, or sebaceous fluid) and develop into cysts. Sometimes these cysts become very large and can reach sizes of 30 or 40 cm. When a pathological cyst becomes very large, movement within the pelvis and abdomen can result in torsion around the suspensory ligament. This causes ischaemic pain, and the ovary becomes gangrenous and requires surgical excision. Polcystic ovaries occur in one in five women and are part of a genetically inherited condition associated with subfertility, irregular periods, androgenic signs, early baldness in the woman's father, insulin resistance, and late-onset diabetes. Structurally, polycystic ovaries are about one and a half times the size of a normal ovary and have a glistening white surface. They consist of 20 to 40 small (2-5 mm diameter) cysts on the surface of the ovary, which encircle a denser than normal stroma. The classic description of this appearance on ultrasound is that of a string of pearls. For an extensive review on this and other aspects of polycystic ovaries see Dunaif and Thomas (2002). REFERENCES Buccione R, Schroeder AC, Eppig JJ 1990 Interactions between somatic cells and germ cells throughout mammalian oogenesis. Biol Reprod 43: 543-7. Medline Similar articles Full article Dunaif A, Thomas A 2001 Current concepts in the polycystic ovary syndrome. Annu Rev Med 52: 401-19. Medline Similar articles Full article
© 2008 Elsevier
103 FEMALE REPRODUCTIVE SYSTEM Uterine tubes The two uterine (Fallopian) tubes lie on each side of the uterus in the upper margin of the broad ligament (mesosalpinx) (Figs 103.1, 103.2, 103.3). They are c.10cm long and are pinkish red in colour.
Figure 103.1 Digitally subtracted hysterosalpingogram. Radiopaque contrast is introduced via a catheter inserted through the cervical os. The catheter is introduced using a vaginal speculum. The contrast fills the triangular uterine cavity. The lumina of the narrow intramural and isthmic parts of the uterine tubes may be traced inferolaterally from the superior angles. They expand into the wider ampullae. Some contrast media has escaped into the pelvic cavity from the abdominal ostia. (By kind permission from Dr Julia Hillier, Chelsea and Westminster Hospital, London.)
The medial opening of the tube (the uterine os) is located at the superior angle of the uterus. The tube passes laterally and superiorly and consists of four main parts: intramural; isthmus; ampulla; and fimbria. The intramural part is c.0.7mm wide, 1cm long, and lies within the myometrium. It is continuous laterally with the isthmus, which is 1 to 5mm wide and 3cm long, and is rounded, muscular and firm. Its lumen is narrow and exhibits three to five longitudinal major folds with a
variable degree of relatively simple secondary folding. The isthmus is continuous laterally with the ampulla, the widest portion of the tube with a maximum luminal diameter of c.1cm. The ampulla is c.5cm long and has a thin wall and a tortuously folded luminal surface marked by 4 to 5 major longitudinal ridges on which lie secondary folds, creating an extensive, labyrinthine surface area (see Fig. 103.5). Typically, fertilization takes place in its lumen. The ampulla opens into the funnel-shaped infundibulum at the abdominal os. Numerous mucosal fingerlike folds, c.1mm wide, the fimbriae, are attached to the ends of the infundibulum (Fig. 102.2). They extend from its inner circumference beyond the muscular wall of the tube. One of these, the ovarian fimbria, is longer and more deeply grooved than the others, and is typically applied to the tubal pole of the ovary. At the time of ovulation the fimbriae swell and extend as a result of engorgement of the vessels in the lamina propria, which aids capture of the released oocyte. All fimbriae are covered, like the mucosal lining throughout the tube, by a ciliated epithelium whose cilia beat towards the ampulla.
TUBAL PREGNANCY AND BLOCKAGE After fertilization the segmenting zygote normally enters the uterus. Occasionally it may adhere to and develop in the tube. A tubal pregnancy is the commonest variety of ectopic gestation. Ectopic gestations usually end with extrusion through the abdominal ostium or natural death and resorption. Occasionally they can continue to expand and rupture through the uterine tube causing severe haemorrhage. Uterine tubes may be blocked intrinsically or extrinsically by scar tissue. The most common causes of tubal blockage include infection, endometriosis, and adhesions from previous surgery. The two most common methods for assessing tubal patency are laparoscopic dye insufflation (LDI) (Fig. 103.4) and hysterosalpingography (HSG) (Fig. 103.1). LDI involves a general anaesthetic and the insertion of an endoscope (laparoscope) into the abdomen. Blue dye is inserted through the cervical os and is seen to spill from the fimbriated end in a patent uterine tube (Fig. 103.4). This technique also allows inspection of any abnormal features that might cause extrinsic blockage of the tube, e.g. endometriosis or adhesions. HSG involves the injection of a radiopaque dye through the cervical os (Fig. 103.1) and assessing tubal patency radiologically. It has the advantage of allowing assessment of the cavity of the uterus and uterine tubes.
page 1327 page 1328
Figure 103.2 Coronal T2-weighted MRI in a young female showing the anteverted uterus lying above the bladder. The right ovary is clearly seen at the lateral margin of the right uterine tube. Both ovaries contain high signal follicles.
Figure 103.3 The female pelvis and its contents.
Tubal inflammation sometimes results in the fimbriated end becoming blocked by adhesions. Pus may collect in the tube causing a pyosalpinx, or fluid may accumulate as a result of mucosal inflammation and this causes a hydrosalpinx. Pelvic peritonitis is said to occur more frequently in females because infection of the vagina, uterus or of the uterine tube may spread directly to the peritoneum via the abdominal ostium.
© 2008 Elsevier
RELATIONS The mesosalpinx is inferior, and the ovarian ligament is inferior and medial, to the tube (Fig. 102.2). The mesovarium and ovary lie inferiorly at its fimbrial end. The round ligament is anterior to the tube. The superior and posterior surfaces of the tube lie free in the peritoneal cavity (Figs 102.1, 102.2).
© 2008 Elsevier
PERITONEAL ATTACHMENTS The uterine tube is attached on its inferior surface to a double fold of peritoneum in the broad ligament called the mesosalpinx (p. 1332). The vessels, nerves and lymphatics that supply the uterine tube lie within the mesosalpinx (p. 1328).
© 2008 Elsevier
VESTIGIAL STRUCTURES
Figure 103.4 Photograph taken at laparoscopy demonstrating the infundibulum of the uterine tube with its fimbriae. Blue dye has been injected through the cervix and is seen spilling from the infundibulum. This is a test of tubal patency in a woman with infertility.
The epoophoron (Fig. 102.2) lies in the lateral part of the mesosalpinx between the ovary and uterine tube. It consists of 10 to 15 short, blind-ending transverse ductules, which converge towards the ovary. Their other ends open into a rudimentary longitudinal duct, the duct of the epoophoron (duct of Gartner). This runs medially in the broad ligament, parallel with the lateral part of the uterine tube (p. 1332). There are often one or more small cystic vesicular appendices between the epoophoron and the fimbriated end of the tube. The duct of the epoophoron can occasionally be followed along the uterus nearly to the internal os where it penetrates the muscular wall of the uterus and descends in the wall of the cervix, gradually approaching the mucosa without actually reaching it. It then descends in the lateral wall of the vagina to end near or at the free margin of the hymen. The paroophoron consists of a few rudimentary tubules scattered in the broad ligament between the epoophoron and uterus, and is most easily seen in children. Both the epoophoron and paroophoron are remnants of mesonephric tubules. The duct of the epoöphoron is a persistent part of the mesonephric duct (p. 1375).
© 2008 Elsevier
VASCULAR SUPPLY AND LYMPHATIC DRAINAGE ARTERIES The blood supply to the uterine tubes is derived from ovarian and uterine stems. The lateral third of the tube is supplied by the ovarian artery (p. 1323), which continues in the mesosalpinx to anastomose with branches from the uterine artery (p. 1333). The branches from the uterine artery supply the medial two-thirds of the tube.
VEINS Venous drainage is similar to the arterial supply. The venous drainage for the lateral two-thirds of the uterine tube is via the pampiniform plexus to the ovarian veins. The latter open into the inferior vena cava on the right side and the renal vein on the left side. The medial two-thirds of the tube drain via the uterine plexus to the internal iliac vein.
LYMPHATIC DRAINAGE Lymph drainage is via ovarian vessels to the para-aortic nodes (p. 1323) and uterine vessels to the internal iliac chain (p. 1323). It is possible for lymph to reach the inguinal nodes via the round ligament.
© 2008 Elsevier
INNERVATION Nerve fibres are distributed largely with the ovarian and uterine arteries. Most of the tube has a dual sympathetic and parasympathetic supply. Preganglionic parasympathetic fibres are derived from the vagus for the lateral half of the tube, and pelvic splanchnic nerves for the medial half. Sympathetic supply is derived from the tenth thoracic to the second lumbar spinal segments. Visceral afferent fibres travel with the sympathetic nerves, and enter the cord through corresponding dorsal roots. They may also accompany parasympathetic fibres. The ampullary submucosa contains modified Pacinian corpuscles.
REFERRED PAIN page 1328 page 1329
Pain from tubal disease is classically described as occurring in the iliac fossa as a result of local peritoneal irritation. As with pain from the ovary (p. 1323), this can sometimes cause discomfort in the distribution of the obturator nerve on the medial aspect of the thigh.
© 2008 Elsevier
MICROSTRUCTURE The walls of the uterine tubes show typical visceral mucosal, muscular and serosal layers (pp. 41, 41). The mucosa is thrown into longitudinal folds (Fig. 103.5), which are most pronounced distally at the infundibulum and decrease to shallow bulges in the intrauterine (intramural) portion. The mucosa is lined by a single-layered, tall, columnar epithelium, which contains mainly ciliated cells and secretory (peg) cells (so-called because they project into the lumen further than their ciliated neighbours), and occasional intraepithelial lymphocytes. In the tube, ciliated cells predominate distally (Fig. 103.5) and secretory cells proximally. Their activities vary with the stage in the menstrual cycle and with age. Secretory cells are most active around the time of ovulation. Their secretions include nutrients for the gametes and aid capacitation of the spermatozoa. Ciliated cells increase in height and develop more cilia in the oestrogenic first half of the menstrual cycle. Their cilia waft the oocyte from the open-ended infundibulum towards the uterus in fluids secreted by the peg cells. The epithelium regresses in height towards the end of the cycle and postmenopausally, when ciliated cells are reduced in number. The lamina propria provides vascular connective tissue support and abundant lymphatic drainage vessels. The smooth muscle of the muscularis is arranged as an inner circular, or spiral, layer and an outer, thinner, longitudinal layer. Together, their contractile activity produces peristaltic movements of the tube, which assist propulsion of the gametes and the fertilized ovum. The uterine tubes are covered externally by a highly vascular serosa.
page 1329 page 1330
Figure 103.5 Ampullary region of uterine tube showing the two layers of smooth muscle (S) surrounding the mucosa (M), with its complex folds projecting into the lumen. (By permission from Dr JB Kerr, Monash University, from Kerr JB 1999 Atlas of Functional Histology. London: Mosby.)
© 2008 Elsevier
104 FEMALE REPRODUCTIVE SYSTEM Uterus The uterus is a hollow, thick-walled and muscular organ. It is normally situated in the lesser pelvis between the urinary bladder and the rectum (Fig. 103.3). The uterus is divided into two main regions. The body of the uterus (corpus uteri) forms the upper two-thirds, and the cervix (cervix uteri) forms the lower third (Figs 102.1, 104.1). The body of the uterus is pear shaped and contains a lumen that is flat anteroposteriorly. The cervix is narrower and is cylindrical in shape. The uterine tubes are attached to the upper part of the body of the uterus with their ostia opening into the lumen (Ch. 103). The lower portion of the cervix continues into the vagina. The adult non-pregnant uterus is c.7.5 cm long, 5.0 cm in breadth, 2.5 cm thick, and weighs between 30 and 40 gm.
© 2008 Elsevier
UTERINE BODY (CORPUS UTERI) (Figs 104.2, 104.3) The body of the uterus extends from the fundus at its uppermost part to the cervix inferiorly (Fig. 104.1). Near its upper end, the body receives uterine tubes on both sides (Figs 103.1, 103.2). The point of fusion between the uterine tube and body is called the uterine cornu. Inferoanterior to the cornu is the round ligament and inferoposterior is the ovarian ligament (Figs 103.3, 102.1, 102.2). The fundus is superior to the entry points of the uterine tubes, and the uterine body narrows as it extends towards the cervix. The dome-like fundus is covered by peritoneum, which is continuous with that of neighbouring surfaces. It is contacted by coils of small intestine and occasionally by distended sigmoid colon. The lateral margins of the body are convex, and on each side their peritoneum is reflected laterally to form the broad ligament, which extends as a flat sheet to the pelvic wall. The anterior surface of the uterine body is covered by peritoneum, which is reflected onto the bladder at the uterovesical fold. This normally occurs at the level of the internal os, which is the most inferior margin of the body of the uterus. Between the bladder and uterus there is the vesico-uterine pouch, which is obliterated when the bladder is distended. When the bladder is empty, the vesicouterine pouch is usually empty, but it may be occupied by part of the small intestine.
Figure 104.1 Sectional diagram showing the interior divisions of the uterus and its continuity with the vagina.
Figure 104.2 Sagittal MRI in a young female. On T2-weighted MRI the uterus displays a zonal anatomy, with three distinct zones: the endometrium; junctional zone; and myometrium. The endometrium and uterine cavity appear as a high-signal stripe. A band of low signal, the junctional zone, borders the endometrium. This represents the inner myometrium, which blends with the low-signal band of fibrous cervical stroma at the level of the internal os.
page 1331
page 1332
Figure 104.3 Transabdominal ultrasound of the female pelvis showing the uterus and both ovaries.
The posterior surface of the uterus is convex transversely. Its peritoneal covering continues down to the cervix and upper vagina and is then reflected back to the rectum (Fig. 102.1) along the surface of the recto-uterine pouch (of Douglas), which lies posterior to the uterus. The sigmoid colon lies posterior to the uterus, although the terminal ileal coil usually separates the two. The cavity of the uterine body measures c.6 cm from the external os of the cervix to the wall of the fundus and is flat in its anteroposterior plane (Figs 103.1, 104.1). In coronal section (Fig. 104.1) it is triangular, broad above where the two uterine tubes join the uterus, and narrow below at the internal os of the cervix.
© 2008 Elsevier
CERVIX (CERVIX UTERI) The adult, non-pregnant cervix is c.2.5 cm long. It is narrower and more cylindrical than the corpus, is widest at its midlevel, and round in section. The upper end communicates with the uterine body via the internal os and the lower end opens into the vagina at the external os. In nulliparous women, the external os is usually a circular aperture, whereas after childbirth it is a transverse slit. There are two longitudinal ridges, one each on its anterior and posterior walls, that give off small oblique palmate folds. These ascend laterally like the branches of a tree (arbor vitae uteri). The folds on opposing walls interdigitate to close the canal. The narrower isthmus of the cervix forms the upper third. Although unaffected in the first month of pregnancy, it is gradually taken up into the uterine body during the second month to form the 'lower uterine segment'. In nonpregnant women the isthmus undergoes menstrual changes, although these are less pronounced than those in the uterine body. The external end of the cervix bulges into the anterior wall of the vagina, which divides it into supravaginal and vaginal regions (Figs 102.1, 104.4). The supravaginal part of the cervix is separated in front from the bladder by cellular connective tissue, the parametrium, which also passes to the sides of the cervix and laterally between the two layers of the broad ligaments. The uterine arteries flank the cervix in this tissue and the ureters descend forwards in it c.2 cm from the cervix, curving under the arch formed by the uterine arteries. The relation of the arteries to the ureters is not always symmetrical. Posteriorly the supravaginal cervix is covered by peritoneum, which continues caudally on to the posterior vaginal wall and is then reflected onto the rectum via the recto-uterine recess (Fig. 102.1). Posteriorly, it is related to the rectum, from which it may be separated by a terminal ileal coil. The vaginal part of the cervix projects into the vaginal cavity forming grooves around its perimeter termed vaginal fornices.
© 2008 Elsevier
VARIATIONS Sometimes there is failure in fusion of the paramesonephric (Müllerian) ducts. This results in a uterus that is not pear shaped and has a varying degree of septation. The most extreme example is often associated with a septate vagina, two cervices, and two discrete uteri each with one uterine tube (Fig. 109.15). This is called a didelphine uterus. Sometimes there is just a septum (septate uterus) or partial clefting of the uterus (bicornuate uterus).
© 2008 Elsevier
RELATIONS AND POSITIONS (Fig. 104.4) The bladder and uterovesical space are anterior to the uterus. The rectum and the recto-uterine pouch are posterior. The broad ligaments are lateral to the uterus. The long axis usually lies along the axis of the pelvic inlet, but since the uterus is movable, its position varies with distension of the bladder and rectum. Except when displaced by a much distended bladder, the long axis of the uterus is nearly at right angles to that of the vagina; the axis of the vagina corresponds to the axis of the pelvic outlet. In the adult nulliparous state the uterus normally tilts forward along its long axis (Fig. 104.4), a state which is normally described as anteflexed. With the bladder empty the whole uterus leans forwards at an angle to the vagina, and in this position is described as anteverted. In 10 to 15% of women the whole uterus leans backwards at an angle to the vagina and is said to be retroverted. A uterus that angles backwards on the cervix is described as retroflexed.
Figure 104.4 Variations in uterine position.
© 2008 Elsevier
UTERINE LIGAMENTS The uterus is continuous with a number of 'ligaments'. Some of these are true ligaments in that they have a fibrous composition and supply support to the uterus. Some ligaments provide no support to the uterus, and others only consist of folds of peritoneum.
PERITONEAL FOLDS The peritoneal folds are the anterior and posterior ligaments of the uterus, and the broad ligament. Uterovesical fold
The uterovesical fold or anterior ligament consists of peritoneum reflected onto the bladder from the uterus at the junction of its cervix and body. Rectovaginal fold
The rectovaginal fold or posterior ligament consists of peritoneum reflected from the posterior vaginal fornix on to the front of the rectum, thereby forming the deep recto-uterine pouch. It is bounded anteriorly by the uterus, supravaginal cervix and posterior vaginal fornix. Posteriorly it is bounded by the rectum. Laterally it is bounded by the uterosacral ligaments (p. 1333). Broad ligament
The broad ligaments extend, one from each side of the uterus, to the lateral walls of the pelvis, where they become continuous with the peritoneum. The upper border is free. The lower border is continuous with the peritoneum over the bladder, rectum and pelvic sidewall. They are continuous with each other at the free edge via the uterine fundus and diverge below near the superior surfaces of levatores ani. In the free border on either side lies a uterine tube. The uterus and the broad ligament form a septum across the lesser pelvic cavity, dividing it into an anterior part containing the bladder and a posterior part containing the rectum, terminal ileum and part of the sigmoid colon. With the bladder empty, the uterus and broad ligament are inclined forwards so that the posterior surface faces up and back, and the anterior surface faces down and forwards. As the bladder fills, the plane of the ligament tilts backwards and the free borders become superior so that the layers face anteriorly and posteriorly. The broad ligament is divided into an upper mesosalpinx, a posterior mesovarium and an inferior mesometrium. Mesosalpinx
page 1332 page 1333
The mesosalpinx is attached above to the uterine tube and posteroinferiorly to the mesovarium. Superior and laterally it is attached to the suspensory ligament of the ovary and medially it is attached to the ovarian ligament. The fimbria of the tubal infundibulum projects from its free lateral end. Between the ovary and uterine tube the mesosalpinx contains vascular anastomoses between the uterine and ovarian vessels, the epoophoron, and the paroophoron (p. 1384).
Mesovarium The mesovarium projects from the posterior aspect of the broad ligament, of which it is the smaller part. It is attached to the hilum of the ovary and carries vessels and nerves to the ovary. Mesometrium The mesometrium is the largest part of the broad ligament, and extends from the pelvic floor to the ovarian ligament and uterine body. The uterine artery passes between its two peritoneal layers c.1.5 cm lateral to the cervix. The uterine artery crosses the ureter shortly after its origin from the internal iliac artery and gives off a branch that passes superiorly to the uterine tube, where it anastomoses with the ovarian artery. Between the pyramid formed by the infundibulum of the tube, the upper pole of the ovary, and the lateral pelvic wall, the mesometrium contains ovarian vessels and nerves within a fibrous suspensory ligament of the ovary (infundibulopelvic ligament). This ligament continues laterally over the external iliac vessels as a distinct fold. The mesometrium also encloses the proximal part of the round ligament of the uterus, as well as smooth muscle and loose connective tissue.
LIGAMENTS These consist of the round, uterosacral, transverse cervical and pubocervical ligaments. Round ligaments
The round ligaments are narrow somewhat flattened bands 10 to 12 cm long, which pass diagonally down and laterally within the mesometrium from the upper part of the uterus to the pelvic floor. They are attached superiorly to the uterine wall just below and anterior to the lateral cornua. Each ligament continues laterally downwards across the vesical, obturator and external iliac vessels, the obturator nerve, and the obliterated umbilical artery. At the start of the inferior epigastric artery, the round ligament enters the deep inguinal ring. It traverses the inguinal canal and finally splits into strands that merge with surrounding connective tissue terminating in the mons pubis above the labium majus. Near the uterus the round ligament contains much smooth muscle but this gradually diminishes until the terminal part is purely fibrous. It contains blood vessels, nerves and lymphatics. The latter drain the uterine region around the entry of the uterine tube to the superficial inguinal lymph nodes. Uterine neoplasms may spread by this route. In the fetus a projection of the peritoneum (processus vaginalis) is carried with the round ligament for a short distance into the inguinal canal. This is generally obliterated in adults, although it is sometimes patent even in old age. In the canal the ligament receives the same coverings as the spermatic cord, although they are thinner and blend with the ligament itself, which may not reach the mons pubis. The round and ovarian ligaments both develop from the gubernaculum (p. 1389) and are continuous. Uterosacral ligaments
The uterosacral ligaments are recto-uterine folds and contain fibrous tissue and smooth muscle. They pass back from the cervix and uterine body on both sides of
the rectum, and they are attached to the front of the sacrum. The ligaments can be palpated laterally on rectal examination. On vaginal examination they can be felt as thick bands of tissue passing downwards on both sides of the posterior fornix. Transverse cervical ligaments
The transverse cervical ligaments (cardinal ligaments, ligaments of Mackenrodt) extend from the side of the cervix and lateral fornix of the vagina to attach extensively on the pelvic wall. At the level of the cervix, some fibres interdigitate with fibres of the uterosacral ligaments. They are continuous with the fibrous tissue around the lower parts of the ureters and pelvic blood vessels. Pubocervical ligament
Fibres of the pubocervical ligament pass forward from the anterior aspect of the cervix and upper vagina to diverge around the urethra. These fibres attach to the posterior aspect of the pubic bones.
UTERINE SUPPORT While the uterosacral and transverse cervical ligaments may act in varying measure as mechanical supports of the uterus, levatores ani and coccygei, the urogenital diaphragm and the perineal body appear at least as important in this respect. There has been renewed interest in the supporting structures of the pelvis, and the subject has been reviewed in detail by DeLancey (2000). Occasionally, the muscular and ligamentous support of the uterus fails, which results in prolapse of the uterus through the vagina. In the most extreme example, procidentia, it prolapses all the way through the vagina. Uterine prolapse usually occurs in association with prolapse of other organs such as the vagina and rectum. It is more common in women who have gone through childbirth and is the result of successive weakening of the pelvic floor during parturition. There is also an inherited component to pelvic-floor prolapse.
© 2008 Elsevier
VASCULAR SUPPLY AND LYMPHATIC DRAINAGE ARTERIES The uterine artery arises as a branch of the anterior division of the internal iliac artery and crosses anterior to the ureter to give off two main branches. These pass superiorly and inferiorly along the lateral surface of the uterus. The uterine artery supplies the uterine body, the uterine cervix, the uterine tubes, and the upper part of the vagina. From its origin, the uterine artery crosses the ureter anteriorly in the broad ligament before branching at the level of the uterus. One major branch ascends the uterus tortuously within the broad ligament until it reaches the region of the ovarian hilum where it anastomoses with branches of the ovarian artery. Another branch descends to supply the cervix and anastomoses with branches of the vaginal artery to form two median longitudinal vessels, the azygos arteries of the vagina, which descend anterior and posterior to the vagina. Although there are anastomoses with the ovarian and vaginal arteries, the dominance of the uterine artery is indicated by its marked hypertrophy during pregnancy. The tortuosity of the vessels as they ascend in the broad ligaments is repeated in their branches within the uterine wall. Each uterine artery gives off numerous branches. These enter the uterine wall, divide and run circumferentially as groups of anterior and posterior arcuate arteries. They ramify and narrow as they approach the anterior and posterior midline so that no large vessels are present in these regions. However, the left and right arterial trees anastomose across the midline and unilateral ligation can be performed without serious effects. The arcuate arteries supply many tortuous radial branches, which pass centripetally through the deeper myometrial layers, supplying these en route, to reach the endometrium. Terminal branches in the uterine muscle are tortuous and are called helicine arterioles. They provide a series of dense capillary plexuses in the myometrium and endometrium. From the arcuate arteries many helical arteriolar rami pass into the endometrium. Their detailed appearance changes during the menstrual cycle. In the proliferative phase helical arterioles are less prominent, whereas they grow in length and calibre, becoming even more tortuous in the secretory phase.
VEINS The venous drainage of the uterus is via the uterine veins, which extend laterally in the broad ligaments and drain into the internal iliac veins. The uterine veins run a course adjacent to the arteries in the broad ligament and pass over the ureters. The uterine venous plexus anastomoses with the vaginal and ovarian venous plexuses.
LYMPHATIC DRAINAGE page 1333 page 1334
Uterine lymphatics exist in the superficial (subperitoneal) and deep parts of the
uterine wall. Collecting vessels from the cervix pass laterally in the parametrium to the external iliac nodes, posterolaterally to the internal iliac nodes, and posteriorly to the rectal and sacral nodes. Some cervical efferents may reach the obturator or gluteal nodes. Vessels from the lower part of the uterine body pass mostly to the external iliac nodes, with those from the cervix. From the upper part of the body, the fundus and the uterine tubes, vessels accompany those of the ovaries to the lateral aortic and pre-aortic nodes. A few pass to the external iliac nodes. The region surrounding the isthmic part of the uterine tube is drained along the round ligament to the superficial inguinal nodes. Uterine lymph vessels enlarge greatly during pregnancy.
© 2008 Elsevier
INNERVATION Uterine nerves arise predominantly from the inferior hypogastric plexus. Some branches descend with the vaginal arteries. Other branches pass directly to the cervix uteri. Some branches ascend with, or near, the uterine arteries in the broad ligament. Nerves to the cervix form a plexus that contains small paracervical ganglia. Sometimes one ganglion is larger and is termed the uterine cervical ganglion. Nerves accompanying the uterine arteries supply the uterine body and tube, and connect with tubal nerves from the inferior hypogastric plexus and with the ovarian plexus. The uterine nerves ramify in the myometrium and endometrium, and usually accompany the vessels. Efferent preganglionic sympathetic fibres are derived from the last thoracic and first lumbar spinal segments. The sites of their postganglionic neurones are unknown. Preganglionic parasympathetic fibres arise in the second to fourth sacral spinal segments and relay in the paracervical ganglia. Sympathetic activity may produce uterine contraction and vasoconstriction and parasympathetic activity may produce uterine inhibition and vasodilatation, but these activities are complicated by hormonal control of uterine functions.
© 2008 Elsevier
MICROSTRUCTURE The uterine wall is composed of three main layers. From its lumen outwards these are the endometrium (mucosa), myometrium (smooth muscle layer) and perimetrium (serosa) or adventitia, depending on region. The myometrium is by far the largest component.
ENDOMETRIUM The mucosal layer forming the endometrium is continuous below with the vaginal mucosa through the external os, and with the peritoneum through the abdominal os of the uterine tubes. It is formed by a layer of connective tissue, the endometrial stroma, which supports a single-layered columnar epithelium continuous with large numbers of tubular endometrial (uterine) glands running perpendicular to the luminal surface or slightly coiled (Fig. 104.5). These glands penetrate as far as the boundary with the myometrium. In the uterine body the surface epithelium is ciliated and cuboidal before puberty but becomes columnar and is usually non-ciliated over large areas in the adult uterus. The endometrial glands are composed largely of columnar cells secreting glycoproteins and glycogen, variably with the stages of the menstrual cycle (p. 1335). The stroma consists of a highly cellular connective tissue between the endometrial glands, and contains blood and lymphatic vessels. The endometrium undergoes a number of changes during the menstrual cycle (p. 1335). For a review of endometrial structure see Spornitz (1992). Endometriosis
In some women, endometrial tissue can exist outside the uterus, a condition called endometriosis. This can occur anywhere, including the lung and bowel, but is most common on the pelvic peritoneum and ovary. Endometriosis may result in a number of symptoms which correlate with the cyclical hormonal influences that occur on endometrial cells. When endometriosis occurs in the pelvis, women develop painful periods and painful intercourse. When endometriosis occurs on the ovary, a cyst of endometrial tissue, an endometrioma, can occur. These are sometimes called chocolate cysts because they contain old blood from repeated monthly haemorrhage into the cyst. The cause of endometriosis is unknown. However, a number of theories exist, including retrograde menstruation and abnormal cell development.
MYOMETRIUM (Fig. 104.5) The myometrium is a fibromuscular layer that forms most of the uterine wall. In nulliparous women it is dense and c.1.3 cm thick at the uterine midlevel and fundus but thin at the tubal orifices. It is composed largely of smooth muscle fasciculi mingled with loose connective tissue, blood vessels, lymphatic vessels and nerves.
Figure 104.5 A, Low-power micrograph of the uterine wall, including endometrium, glands and myometrium. B, Higher-power micrograph of the endometrium, showing the coiled glands typical of the secretory second half of the menstrual cycle. (By permission from Kierszenbaum AL 2002 Histology and Cell Biology. St Louis: Mosby.)
The body of the uterus is often described as having four more or less distinct muscular layers. The innermost layer (submucosal layer) is composed mostly of longitudinal and some oblique smooth muscle. Where the lumen of the uterine tube passes through the uterine wall, this layer forms a circular muscle coat. External to the submucosal layer is the vascular layer, a zone rich in blood vessels as well as longitudinal muscle. Next is a layer of predominantly circular muscle, the supravascular layer. The outer, thin, longitudinal muscle layer, the subserosal layer, lies adjacent to the perimetrium. The fibromuscular fascicles of the outer two layers converge at the lateral angles of the uterus, and continue into the uterine tubes and the round and ovarian
of the uterus, and continue into the uterine tubes and the round and ovarian ligaments. Some fascicles enter the broad ligaments, others turn back into the uterosacral ligaments. At the junction between the body and the cervix, the smooth muscle merges with dense irregular connective tissue containing both collagen and elastin, and forms the majority of the cervical wall. Bilateral longitudinal fascicles extend in the lateral submucosal layer from the fundal angle to the cervix. Their muscle fibres differ structurally from those of typical myometrium, and they may provide fast conducting pathways which coordinate the contractile activities of the uterine wall. During pregnancy the muscle hypertrophies, and the individual fibres are greatly enlarged. Their numbers also increase by proliferative hyperplasia. The numbers of gap junctions coupling adjacent fibres also increase, indicating greater coordination of their contractility. During pregnancy, contractility is inhibited by relaxin, which is secreted by the ovarian corpus luteum and placenta (p. 1325). Leiomyoma
Benign tumours of myometrial cells are are called leiomyomata (fibroids). The aetiology of these benign tumours is unknown. However, they are more common in women of Afro-Caribbean origin than in Caucasian women. Over a third of women of reproductive age have fibroids. If they project predominantly into the cavity of the uterus they are termed submucous fibroids. Submucous fibroids increase the surface area of the endometrium and can result in heavy menstrual bleeding. Fibroids that are predominantly on the surface of the uterus are called subserous and those that exist predominantly within the myometrium are called intramural. Most women have no symptoms from their fibroids but occasionally they can get pain from a degenerating fibroid as well as excessive menstrual blood loss. Large fibroids can cause particular problems in pregnancy where they are more likely to degenerate and cause pain and can also obstruct labour. Occasionally a fibroid may develop a blood supply from another organ such as the omentum. It then loses its supply from the uterus and becomes a parasytic fibroid. Adenomyosis page 1334 page 1335
Sometimes areas of endometrial tissue can exist deep within the myometrium. This tissue responds to cyclical changes in oestrogen and progesterone. However, as the endometrial lining cannot be expelled through the cervix, and is in effect trapped, it causes painful periods. This condition is known as adenomyosis and is characterized by ill-defined thickening of the low signal junctional zone and sometimes areas of high signal on MRI scans.
PERIMETRIUM The perimetrium (serosa) is composed of peritoneum (mesothelium overlying a thin connective tissue lamina propria), which covers the uterine body and supravaginal cervix posteriorly, but anteriorly covers only the body. Over the most inferior quarter of the uterine length the peritoneum is separated posteriorly from the underlying uterus by loose cellular connective tissue and large veins. Beneath the peritoneum there is a subserous layer of loose, more fibrous tissue.
UTERINE CERVIX (Fig. 104.6) Except for the lower part, the uterine cervix is lined by a single-layered, columnar epithelium with tubular glands, which overlies a fibroelastic connective tissue
stroma containing relatively little smooth muscle. The elastic component of the cervical stroma is essential to the stretching capacity of the cervix during childbirth. Its lower region (approximately one-third, but variable), including its vaginal surface, is covered by non-keratinizing stratified squamous epithelium, which contains glycogen. None of the mucosa is shed during menstruation and so, unlike the body of the uterus, it is not divided into functional and basal layers and lacks spiral arteries. For most of the upper part of the endocervical canal the mucosa is c.3 mm thick. It is lined by a deeply folded surface epithelium of columnar mucous cells continuous with branched tubular cervical glands, which are lined by a similar secretory epithelium (Fig. 104.6). Ciliated cells are also present in patches at the surface. The cervical glands extend obliquely upwards and outwards from the canal. They secrete clear, alkaline mucus, which is relatively viscous except at the midpoint of the menstrual cycle, when their more copious, less viscous, secretions favour sperm motility. Not uncommonly, the aperture of a gland becomes blocked and it then fills with mucus to form a Nabothian cyst (or follicle), up to 5 mm or more in diameter.
Figure 104.6 The transformation zone of the uterine cervix. The single-layered columnar epithelium of the endocervical canal (above left) changes abruptly to the stratified squamous non-keratinizing epithelium (below) of the external os and ectocervix. (Photograph by Sarah-Jane Smith.)
The squamocolumnar junction between the columnar secretory epithelium of the endocervical canal and the stratified squamous covering of the ectocervix is abrupt (Fig. 104.6). Its precise location varies with age and, to a lesser extent, with the stage in the menstrual cycle. It approximates to the position of the external os, which is its usual prepubertal location. At puberty, the endocervical
columnar epithelium responds to oestrogenic stimulation and extends distally on the ectocervix. This area of columnar cells on the ectocervix forms an area that is red and raw in appearance called an ectropion (cervical erosion). It is then exposed to the acidic environment of the vagina (p. 1353) and through a process of squamous metaplasia, a stratified squamous ectocervical epithelium effectively regrows over the exposed area, resulting in a transformation zone. Other hyperoestrogenic states such as pregnancy and the use of oral contraceptive pills can also result in an ectropion. This is important clinically as this area is the most usual site of epithelial abnormalities that may progress to malignancy. In addition, the new stratified epithelium frequently occludes the ducts of endocervical glands in the region and Nabothian cysts develop. In postmenopausal women, the squamocolumnar junction recedes into the endocervical canal.
© 2008 Elsevier
CYCLICAL CHANGES IN THE UTERUS Throughout the period of reproductive life (except during pregnancy and lactation), a series of closely interrelated cyclical changes occur in the ovary, uterus and vagina. Each cycle extends over a period of c.28 days. In the ovarian cycle one follicle usually reaches full maturity, ruptures and releases its secondary oocyte. The wall of the follicle is then transformed into an important endocrine gland, the corpus luteum. About 10 days after ovulation the corpus luteum begins to regress, then ceases to function and is replaced by fibrous tissue. The changes of the uterine cycle (menstrual cycle) chiefly involve the lining endometrium of the body and fundus of the uterus and may be divided into three phases: menstrual; proliferative; and secretory.
MENSTRUAL PHASE (Fig. 104.7) Prior to menstruation three strata can be recognized in the endometrium (Fig. 104.7). These are the stratum compactum, spongiosum and basale. In the stratum compactum, next to the free surface, the necks of the glands are only slightly expanded and the stromal cells show a distinct decidual reaction. In the stratum spongiosum, the uterine glands are tortuous, dilated and ultimately only separated from one another by a small amount of interglandular tissue. The stratum basale, next to the uterine muscle, is thin and contains the tips of the uterine glands embedded in an unaltered stroma. The upper two strata are often grouped together as the functional layer, stratum functionalis, of the endometrium, and the lower (basal) layer is the stratum basalis. As regression of the corpus luteum occurs, those parts of the stroma showing a decidual reaction together with the glandular epithelium undergo degenerative changes and the endometrium often diminishes in thickness. Blood escapes from the superficial vessels of the endometrium forming small haematomata beneath the surface epithelium. The superficial part of the endometrium, next to the free surface, is shed piecemeal, leaving mainly the basal zone, adjacent to the uterine muscle (Fig. 104.7). Approximately two-thirds to three-quarters of the thickness of the endometrium may be shed. Blood and necrotic endometrium then begin to appear in the uterine lumen, and are discharged from the uterus through the vagina. This menstrual flow lasts from 3 to 6 days. The amount of tissue lost is variable, but usually the stratum compactum and most of the spongiosum are desquamated.
PROLIFERATIVE PHASE (Fig. 104.8) page 1335 page 1336
Figure 104.7 Stages in the menstrual cycle. The top panel shows the variation in thickness of endometrium during an idealized 28-day cycle in which ovulation occurs at day 0. Measurements were made by transvaginal ultrasound. The five lower panels are histological sections of endometrium at the cycle times indicated. (By permission from Buckley CH, Fox H 1989 Biopsy Pathology of the Endometrium. London: Chapman and Hall Medical.)
In the early proliferative phase, and even before the menstrual flow ceases, the epithelium from the persisting basal parts of the uterine glands grows luminally over the denuded surface of the endometrium. Re-epithelialization is complete by 5 to 6 days after the start of menstruation. Initially the tissue is only 1-2 mm thick and lined by low cuboidal epithelium. The glands are straight and narrow with short columnar cells. The apical cell surface contains microvilli, and some ciliated cells are present. The stroma is dense and contains small numbers of lymphocytes among the larger population of mesenchymally derived cells. During days 10 to12 of the proliferative phase the endometrium grows in response to the presence in the bloodstream of oestrogen produced by the ovary (Fig. 104.8), which acts through receptors present on both the stromal and epithelial cells of the endometrium. Mitoses are seen and the glands become distinctly tortuous. Their lining epithelial cells become tall columnar (Fig. 104.7).
SECRETORY PHASE (Fig. 104.9) The secretory phase coincides with the luteal phase of the ovarian cycle. Ovulation occurs c.14 days before the onset of the next menstrual flow. The changes occurring in the secretory phase depend upon the presence in the bloodstream of progesterone and oestrogens, secreted by the corpus luteum (Fig. 104.8). Steroid receptors in the endometrium respond by activating a programme of new gene expression to produce, in the following 7 days, a highly regulated sequence of differentiative events, which are presumably required to prepare the tissue for blastocyst implantation. Part of the response is direct, but there is evidence that some of the effects may be mediated through growth factors. The first morphological effects of progesterone are evident 24 to 36 hours after ovulation. In the early secretory phase, glycogen masses (known incorrectly as 'subnuclear vacuoles') appear in the basal cytoplasm of the epithelial cells lining the glands, where they are often associated with lipid. Nuclei are thus displaced towards the centre of the cells (Fig. 104.9). Giant mitochondria appear and are associated with semi-rough endoplasmic reticulum. A prominent nuclear channel system is present. A notable increase in polarization of the gland cells occurs and Golgi apparatus and secretory vesicles accumulate in the supranuclear cytoplasm. Nascent secretory products may be detected immunohistochemically within the gland cells. Progestational effects on the stroma are also evident in the early secretory phase. Nuclear enlargement occurs
and the packing density of the resident mesenchymal cells increases, due in part to the increase in volume of gland lumens and onset of secretory activity in the epithelial compartment. By the mid-secretory phase the endometrium may be up to 6 mm deep. The basal epithelial glycogen mass is progressively transferred to the apical cytoplasm, allowing the return of nuclei to the cell base. The Golgi apparatus becomes dilated and products including glycogen, mucin and other glycoproteins are released from the glandular epithelium into the lumen by a combination of apocrine and exocrine mechanisms: this activity reaches a maximum c.6 days after ovulation. These secretory changes are considerably less pronounced in the basal gland cells and the luminal epithelium than in the glandular cell population of the stratum functionalis. There is a notable stromal oedema and a corresponding decrease in the density of collagen fibrils. At the same time the endoplasmic reticulum and Golgi apparatus become more prominent, and there is evidence for the synthesis of collagen as well as its endocytosis and degradation, presumably reflecting matrix remodelling. page 1336 page 1337
Figure 104.8 Some salient features of the female reproductive cycle. Note the periodic changes that occur during the non-pregnant state of the ovarian cycle and the concomitant endometrial changes of the menstrual (uterine) cycle, and variations in circulating plasma hormone levels, and the consequences of pregnancy. FSH, follicle stimulating hormone; LH, luteinizing hormone. (Modified from Wendell-Smith CP, Williams PL, Treadgold S 1984 Basic Human Embryology, 3rd edn. Baltimore: Urban and Schwarzenberg.)
In the late secretory phase glandular secretory activity declines. Decidual differentiation occurs in the superficial stromal cells surrounding blood vessels. This transformation includes rounding of the nucleus and an increase in the cytoplasmic volume with a concurrent increase and dilatation of the rough endoplasmic reticulum and Golgi systems and cytoplasmic accumulation of lipid droplets and glycogen. The cells begin to produce basal lamina components including laminin and type IV collagen.
VASCULAR CHANGES DURING THE MENSTRUAL CYCLE
Figure 104.9 Section of human endometrium at about day 17 of the menstrual cycle (early secretory phase). The accumulation of secretory material in the basal parts of the epithelial cells lining the glands has displaced the nuclei towards the lumen of the gland. (Lent by Gordon Museum, London.)
The vascular bed of the endometrium undergoes significant changes during the menstrual cycle. The arteries to the endometrium arise from a myometrial plexus and consist of short, straight vessels to the basal portion of the endometrium and more muscular spiral arteries to its superficial two-thirds. The capillary bed consists of an endothelium with a basal lamina that is discontinuous in the proliferative phase, becoming more distinct by the mid-secretory phase. Pericytes are present, some of which resemble smooth muscle cells, and these are sometimes enclosed within the basal lamina. The pericytes make contact with the endothelial cells by means of cytoplasmic extensions that project through the basal lamina. Enlargement of the pericytes starts in the early secretory phase and leads to a conspicuous cuff of cells in the mid- and late-secretory phases. The venous drainage, consisting of narrow perpendicular vessels that anastomose by cross branches, is common to both the superficial and basal layers of the endometrium. The arterial supply to the basal part of the endometrium remains unchanged during the menstrual cycle. However, the spiral arteries to the superficial strata lengthen disproportionately. They become increasingly coiled and their tips approach more closely the uterine epithelium during the secretory phase of the menstrual cycle. This leads to a slowing of the circulation in the superficial strata with some vasodilation. Immediately before the menstrual flow these vessels begin to constrict intermittently, causing stasis of the blood and anaemia of the superficial strata. During the periods of relaxation of the vessels, blood escapes from the devitalized capillaries and veins, thus causing the menstrual blood loss. During the proliferative, early, and mid-secretory phases of the cycle, the bonemarrow derived cells present in the endometrium are mainly macrophages and classic T cells, and there are very few B cells. In the late secretory phase, an unusual, large, granular lymphocyte population is recruited to the tissue and is found mainly in the stromal compartment. If fertilization of the ovum does not occur, the corpus luteum undergoes degeneration. The breakdown of the endometrium that follows this cessation of function is due to the reducing levels of progesterone and oestrogen.
© 2008 Elsevier
MAGNETIC RESONANCE IMAGING OF THE UTERUS (Fig. 104.2) On T2-weighted magnetic resonance imaging (MRI) (Fig. 104.2), the uterus displays a zonal anatomy, with three distinct zones: the endometrium; junctional zone; and myometrium. The endometrium and uterine cavity appear as a highsignal stripe the thickness of which varies with the menstrual cycle. In the early proliferative phase it measures up to 5 mm, widening to up to 1 cm in the midsecretory phase. A band of low signal, the junctional zone, borders the endometrium. This represents the inner myometrium, which blends with the lowsignal band of fibrous cervical stroma at the level of the internal os. The junctional zone appears low signal because the cells of the myometrium have a low water content when compared to those of the outer myometrium. The junctional zone is of constant thickness and signal throughout the menstrual cycle, measuring c.5 mm. The outer myometrium is of medium signal intensity in the proliferative phase, becoming of high signal intensity in the mid-secretory phase because of the increased vascularity and prominence of the arcuate vessels. In prepubertal females, the uterus is smaller (only 4 cm in length) and on T2weighted images the endometrium is minimal or absent, with an indistinct junctional zone. In postmenopausal women, the corpus decreases in size and the zonal anatomy is indistinct. page 1337 page 1338
On T2-weighted MRI the cervix has an inner low-signal stroma continuous with the junctional zone of the uterus. Often this is surrounded by an outer zone of intermediate signal intensity, which is continuous with the outer myometrium. The appearances do not change with the menstrual cycle or oral contraceptive pill use. The central stripe is very high signal and is a consequence of the secretions produced by the endocervical glands. REFERENCES DeLancey JO 2000 Anatomy. In: Stanton SL, Monga A (eds) Clinical Urogynaecology, 2nd edition. London: Churchill Livingstone. An overview of the support of the pelvic floor in relation to female prolapse and incontinence. Spornitz UM 1992 The functional morphology of the human endometrium and decidua. Adv Anat Embryol Cell Biol 124: 1-99. Medline Similar articles
© 2008 Elsevier
105 FEMALE REPRODUCTIVE SYSTEM Implantation, placentation, pregnancy and parturition FERTILIZATION AND IMPLANTATION Fertilization usually occurs in the lateral or ampullary part of the uterine tube, and is followed c.26-40 hours later by the first cleavage. The dividing preimplantation embryo is conveyed along the tube to the uterine cavity by ciliary action and is aided by muscular tubal contractions. The journey takes c.3 days. The blastocyst adheres to the endometrium after it hatches from the zona pellucida. The outer cells of the blastocyst, the trophoblast or trophectoderm, are flattened polyhedral cells, which possess ultrastructural features typical of a transporting epithelium. The trophoblast covering the inner cell mass is the polar trophoblast and that surrounding the blastocyst cavity is the mural trophoblast. Implantation involves the initial attachment of polar trophectoderm to endometrial luminal epithelium. The trophectoderm then penetrates the epithelium and underlying basal lamina and implants into the stroma using a combination of motile and locally degradative activities. There are two distinct cell arrangements in the trophoblast: an inner cytotrophoblast of cuboidal cells; and an outer multinucleated mass of cytoplasm, the syncytiotrophoblast. The latter penetrates the uterine luminal epithelium and sends finger-like projections between adjacent epithelial cells towards the underlying basal lamina. The two layers become interlocked by numerous tight junctions. Preimplantation embryos produce proteases (responsible for degrading basal lamina extracellular matrix molecules), which probably mediate penetration of the subepithelial basal lamina by the syncytiotrophoblast. Implantation continues with erosion of maternal vascular endothelium and glandular epithelium, and phagocytosis of secretory products, until the blastocyst occupies an uneven implantation cavity in the stroma (interstitial implantation). In the early postimplantation phase, the maternal surface is resealed by re-epithelialization and the formation of a plug, which may contain fibrin. The syncytiotrophoblast secretes a hormone, human chorionic gonadotrophin (hCG), which may be detected in the urine from as early as 10 days after fertilization, and forms the basis for tests for early pregnancy. HCG prolongs the life of the corpus luteum, which continues to secrete progesterone and oestrogens during approximately the first two months of pregnancy. Thereafter, these and other hormones are produced by the placenta. Menstruation ceases on successful implantation. The endometrium, known as the decidua of pregnancy, thickens to form a suitable nidus for the conceptus. Decidualization of the endometrial stroma may occur without an intrauterine pregnancy, e.g. in the presence of an ectopic pregnancy, after prolonged treatment with progesterone , and in the late secretory phase of a nonconception cycle. Decidual differentiation is not evident in the stroma at the earliest stages of
implantation, and it may not be until a week later that fully differentiated cells are present. During decidualization the interglandular tissue increases in quantity. It contains a substantial population of leukocytes (large granular lymphocytes, macrophages and T cells) distributed amongst large decidual cells. Decidual cells are mesenchymally derived stromal cells which contain varying amounts of glycogen, lipid, and vimentin-type intermediate filaments in their cytoplasm. They are generally rounded, but their shape may vary depending on the local packing density. They may contain one, two or sometimes three nuclei and frequently display rows of club-like cytoplasmic protrusions enclosing granules. The cells are associated with a characteristic capsular basal lamina. Decidual cells produce a range of secretory products, including insulin-like growth factor, binding protein 1 (IGF-BP1) and prolactin, which may be taken up by the trophoblast. These secretions probably play a role in the maintenance and growth of the conceptus in the early part of postimplantational development, and can be detected in amniotic fluid in the first trimester of pregnancy. Extracellular matrix, growth factors and protease inhibitors produced by the decidua all probably modulate the degradative activity of the trophoblast and support placental morphogenesis and placental accession to the maternal blood supply. Formation of the haemochorial placenta requires a developmental progression (where the develop-ment proceeds in a specific order over time), which is specified in the trophoblast but is dependent on the maternal environment for its correct expression. Immunological rejection of the semiallogenec conceptus does not occur because the trophoblast expresses human leukocyte antigen-G (HLA-G) which downregulates the maternal immune response. Once implantation is complete, distinctive names are applied to different regions of the decidua. The part covering the conceptus is the decidua capsularis; that between the conceptus and the uterine muscular wall is the decidua basalis (where the placenta subsequently develops); that which lines the remainder of the body of the uterus is the decidua parietalis (Fig. 105.1). There is no evidence that their respective resident maternal cell populations exhibit site-specific properties.
© 2008 Elsevier
FERTILIZATION AND IMPLANTATION Fertilization usually occurs in the lateral or ampullary part of the uterine tube, and is followed c.26-40 hours later by the first cleavage. The dividing preimplantation embryo is conveyed along the tube to the uterine cavity by ciliary action and is aided by muscular tubal contractions. The journey takes c.3 days. The blastocyst adheres to the endometrium after it hatches from the zona pellucida. The outer cells of the blastocyst, the trophoblast or trophectoderm, are flattened polyhedral cells, which possess ultrastructural features typical of a transporting epithelium. The trophoblast covering the inner cell mass is the polar trophoblast and that surrounding the blastocyst cavity is the mural trophoblast. Implantation involves the initial attachment of polar trophectoderm to endometrial luminal epithelium. The trophectoderm then penetrates the epithelium and underlying basal lamina and implants into the stroma using a combination of motile and locally degradative activities. There are two distinct cell arrangements in the trophoblast: an inner cytotrophoblast of cuboidal cells; and an outer multinucleated mass of cytoplasm, the syncytiotrophoblast. The latter penetrates the uterine luminal epithelium and sends finger-like projections between adjacent epithelial cells towards the underlying basal lamina. The two layers become interlocked by numerous tight junctions. Preimplantation embryos produce proteases (responsible for degrading basal lamina extracellular matrix molecules), which probably mediate penetration of the subepithelial basal lamina by the syncytiotrophoblast. Implantation continues with erosion of maternal vascular endothelium and glandular epithelium, and phagocytosis of secretory products, until the blastocyst occupies an uneven implantation cavity in the stroma (interstitial implantation). In the early postimplantation phase, the maternal surface is resealed by re-epithelialization and the formation of a plug, which may contain fibrin. The syncytiotrophoblast secretes a hormone, human chorionic gonadotrophin (hCG), which may be detected in the urine from as early as 10 days after fertilization, and forms the basis for tests for early pregnancy. HCG prolongs the life of the corpus luteum, which continues to secrete progesterone and oestrogens during approximately the first two months of pregnancy. Thereafter, these and other hormones are produced by the placenta. Menstruation ceases on successful implantation. The endometrium, known as the decidua of pregnancy, thickens to form a suitable nidus for the conceptus. Decidualization of the endometrial stroma may occur without an intrauterine pregnancy, e.g. in the presence of an ectopic pregnancy, after prolonged treatment with progesterone , and in the late secretory phase of a nonconception cycle. Decidual differentiation is not evident in the stroma at the earliest stages of implantation, and it may not be until a week later that fully differentiated cells are present. During decidualization the interglandular tissue increases in quantity. It contains a substantial population of leukocytes (large granular lymphocytes,
macrophages and T cells) distributed amongst large decidual cells. Decidual cells are mesenchymally derived stromal cells which contain varying amounts of glycogen, lipid, and vimentin-type intermediate filaments in their cytoplasm. They are generally rounded, but their shape may vary depending on the local packing density. They may contain one, two or sometimes three nuclei and frequently display rows of club-like cytoplasmic protrusions enclosing granules. The cells are associated with a characteristic capsular basal lamina. Decidual cells produce a range of secretory products, including insulin-like growth factor, binding protein 1 (IGF-BP1) and prolactin, which may be taken up by the trophoblast. These secretions probably play a role in the maintenance and growth of the conceptus in the early part of postimplantational development, and can be detected in amniotic fluid in the first trimester of pregnancy. Extracellular matrix, growth factors and protease inhibitors produced by the decidua all probably modulate the degradative activity of the trophoblast and support placental morphogenesis and placental accession to the maternal blood supply. Formation of the haemochorial placenta requires a developmental progression (where the develop-ment proceeds in a specific order over time), which is specified in the trophoblast but is dependent on the maternal environment for its correct expression. Immunological rejection of the semiallogenec conceptus does not occur because the trophoblast expresses human leukocyte antigen-G (HLA-G) which downregulates the maternal immune response. Once implantation is complete, distinctive names are applied to different regions of the decidua. The part covering the conceptus is the decidua capsularis; that between the conceptus and the uterine muscular wall is the decidua basalis (where the placenta subsequently develops); that which lines the remainder of the body of the uterus is the decidua parietalis (Fig. 105.1). There is no evidence that their respective resident maternal cell populations exhibit site-specific properties.
© 2008 Elsevier
FETAL MEMBRANES
page 1339 page 1340
Figure 105.1 The gravid uterus in the second month. A placental site precisely in the uterine fundus as indicated in the plan is, however, rather unusual. The dorsal, ventral or lateral wall of the corpus uteri is more usual.
The implanting conceptus consists of three cavities and surrounding epithelia. The original blastocyst cavity, now termed the extraembryonic coelom or chorionic cavity, is large compared to the embryo within (Figs 10.9, 10.10). The trophoblast becomes lined by a mesothelium derived from cells proliferating close to the inner cell mass. These two layers collectively form the chorion. The part of the chorion overlying the implantation site that will give rise to the placenta is termed the chorion frondosum, reflecting the abundance of proliferating villi. The early villi degenerate as the placenta enlarges, initially at the abembryonic pole, and then later over most of the enlarging chorion, leaving the smooth chorion laeve. The embryo is formed from upper epiblast and lower hypoblast layers (Ch. 10). A small amniotic cavity develops above the epiblast and a larger yolk sac develops beneath the hypoblast. Both cavities are covered externally by a mesothelium, which is continuous with the inner lining of the trophoblast. The amnion is thus composed of epiblast-derived cells and mesothelium. The yolk sac wall is composed of hypoblast derived cells, extraembryonic mesenchyme and mesothelium. A fourth cavity, the allantois, develops as a caudal hypoblastic diverticulum that
becomes embedded within the extraembryonic mesenchyme, which forms the connecting stalk of the embryo. It does not have a direct mesothelial covering. In the early stages the amnion and yolk sac are suspended within the chorionic cavity (see Fig. 105.3). With embryonic and fetal growth the amniotic cavity expands until it fills the chorionic cavity and abuts against the chorion.
AMNION (CHORIO-AMNION) Between the tenth and twelfth weeks of pregnancy the amniotic cavity expands and the chorion frondosum regresses to form the chorion laeve, which is in turn apposed to the decidua capsularis. During the same period the amnion and chorion fuse to form the chorio-amnion, and this avascular membrane persists to term. The original amniotic cells develop from the edges of the epiblast of the embryonic disc and are ultimately connected directly to the skin at the umbilical region. As the fetus grows and the amniotic cavity expands, the amniotic membrane extends along the connecting stalk and forms the outer covering of the umbilical cord. For details of the formation of the umbilical cord see page 1259. After birth, the site of this embryonic/extraembryonic junction is important, because the extraembryonic cell lines will die, causing the umbilical cord to degenerate and detach from the body wall. In cases of anomalous development of the ventral body wall, e.g. gastroschisis and exomphalos, there may be insufficient skin in this region and a large ventral wall defect may result. The inner surface of the amnion consists of a simple cuboidal epithelium with a microvillous apical surface beneath which is a cortical web of intermediate filaments and microfilaments. There are no tight junctional complexes between adjacent cells and cationic dyes penetrate between the cells as far as the basal lamina. The intercellular clefts present scattered desmosomes, but elsewhere the clefts widen and contain interlacing microvilli. These features are consistent with selective permeability properties. The epithelium synthesizes and deposits extracellular matrix into the compact layer of acellular stroma located beneath the basal lamina, as well as the basal lamina itself. Human amniotic epithelial cells are thought to be pluripotential because they arise so early from the conceptus. They can be distinguished from the epiblast cells from day 8. Recent studies have shown that amniotic cells lack the major histocompatibility complex antigen. The amnion can therefore be exposed to the maternal immune system without eliciting a maternal immune response. Cultured human amniotic epithelial cells express a range of neural and glial markers, including glial fibrillary acidic protein (GFAP), myelin basic protein, vimentin and neurofilament proteins, suggesting that these cells may supply neurotrophic factors to the amniotic fluid. Amnion is used in the repair of corneas after trauma and as a graft material for reconstructing vaginas in women with cloacal abnormalities. Towards the end of gestation increasing numbers of amniotic cells undergo apoptosis. Apoptotic cells become detached from the amnion and are found in the amniotic cavity at term. The highest incidence is in weeks 40-41, independent of
the onset of labour. Apoptosis may play a role in the fragility and rupture of the fetal membranes at term. The chorion at term consists of an inner cellular layer containing fibroblasts and a reticular layer of fibroblasts and Hofbauer cells, which resembles the mesenchyme of an intermediate villus. The outer layer consists of cytotrophoblast 3 to 10 cells deep, resting on a pseudobasal lamina, which extends beneath and between the cells. Occasional obliterated villi within the trophoblast layer are the remnants of villi present in the chorion frondosum of the first trimester. Although the interface between the trophoblast and decidua parietalis is uneven, no trophoblast infiltration of the parietalis occurs.
AMNIOTIC FLUID The amniotic fluid, or liquor amnii, is derived from multiple sources throughout gestation. These include secretions from amnion epithelium, filtration of fluid from maternal vessels via the parietal decidua and amniochorion, filtration from the fetal vessels via the chorionic plate or the umbilical cord, and fetal urine. In early pregnancy, diffusion from intracorporeal vessels via fetal skin provides another source. Once the gut is formed, fetal swallowing of amniotic fluid is a normal occurrence. The fluid is absorbed into the fetal circulation and passes via the placental barrier into the maternal circulation. There is rapid exchange between the amniotic fluid and maternal and fetal circulations via the placenta and fetal kidneys. In the early stages amniotic fluid resembles blood plasma in composition and is probably formed largely by transport across the amniotic membrane. As pregnancy advances, it becomes progressively more dilute, partly by the addition of fetal urine. It contains less than 2% of solids, including urea , inorganic salts, a small amount of protein and frequently a trace of sugar. Glycoprotein secretions from amniotic epithelium include fibronectin. There is experimental evidence of a considerable and rapid flux of water across the amniotic membrane. By the end of the third month the expanding amnion has extensive contact with the chorion laeve, and only these thin membranes separate the amniotic fluid from the decidua parietalis, the tissues and vessels of which provide another route for the exchange of water and dissolved substances. Secretory products of maternal decidua, including prolactin and insulin-like growth factor binding protein 1 (IGF-BP1), are present in the liquor. The amount of amniotic fluid increases in quantity up to the sixth or seventh month and then diminishes slightly. At the end of pregnancy it is usually somewhat less than a litre. It provides a buoyant medium, which supports the delicate tissues of the young embryo and allows free movement of the fetus during the later stages of pregnancy. It also diminishes the risk to the fetus of injury from without. A volume of amniotic fluid in excess of 2 litres is generally considered to be abnormal and constitutes polyhydramnios. A deficiency is termed oligohydramnios. Both conditions may be associated with fetal abnormalities, e.g. fetuses with agenesis of the kidneys or atresia of the lower urinary tract are often associated with oligohydramnios. Similarly, pulmonary hypoplasia at birth may be
caused by severe congenital urinary obstruction because a reduction in the amniotic fluid pressure permits fluid produced in the fetal lungs to escape and this compromises lung expansion. Cases of oesophageal atresia or anencephaly, in which swallowing is impossible or impaired, and open spina bifida, are often associated with polyhydramnios. Impaired swallowing combined with these neural defects is accompanied by direct discharge of cerebrospinal fluid (CSF) into the amniotic liquor. In fetuses with spina bifida and some other neural tube defects the concentration of !-fetoprotein in the amniotic fluid is exceptionally high and is used to diagnose these abnormalities.
YOLK SAC As the secondary yolk sac forms it delineates a cavity lined with parietal, and perhaps visceral, hypoblast, which is continuous with the developing endoderm from the primitive streak (Ch. 10). The yolk sac becomes coated with extraembryonic mesenchyme, which forms mesenchymal and mesothelial layers. The inner cells of the yolk sac (denoted endoderm in many studies, although this layer is restricted to the embryo itself) are columnar with numerous microvilli and pinocytotic vesicles. They contain abundant rough endoplasmic reticulum, welldeveloped Golgi apparatus, and numerous mitochondria. The outer mesothelium also exhibits microvilli, which are covered by a mucus layer. The cells appear less active than the inner yolk sac cells. The mesothelium may provide a protective coat to prevent damage caused by compression or friction of the yolk sac wall against the chorionic cavity lining. The intervening mesenchymal layer is the main site for blood vessel and cell formation in the early embryo. page 1340 page 1341
Figure 105.2 A fetus of c.8 weeks, enclosed in the amnion, magnified c.2.5 diameters. A part of the chorion frondosum with its branching villous stems is shown in the lower part of the figure. The villous stems have been detached from the basal plate, which is not shown here.
The yolk sac remains within the extraembryonic coelom (chorionic cavity) throughout gestation. It becomes located between the amnion and chorion as they fuse near the placental attachment of the umbilical cord. It continues to grow slowly and is sometimes found at term in this site as a small vesicle, usually less than 5 mm in diameter. The yolk stalk and contained endodermal duct gradually elongate with the growth in length of the umbilical cord (Fig. 105.2).
ALLANTOIS The allantoenteric diverticulum (Fig. 105.3) arises early in the third week as a solid, endodermal outgrowth from the dorsocaudal part of the yolk sac into the
mesenchyme of the connecting stalk. It soon becomes canalized. When the hindgut is developed, the proximal (enteric) part of the diverticulum is incorporated in its ventral wall. The distal (allantoic) part remains as the allantoic duct and is carried ventrally to open into the ventral aspect of the cloaca or terminal part of the hindgut (Fig. 105.3). The allantois is a site of angiogenesis, which gives rise to the umbilical vessels and placental circulation. The extraembryonic mesenchyme around the allantois forms the connecting stalk, which is later incorporated into the umbilical cord. In the fetus, the allantoic duct, which is confined to the proximal end of the umbilical cord, elongates and thins. However, it may persist as an interrupted series of epithelial strands at term, in which case the proximal strand is often continuous at the umbilicus with the median intra-abdominal urachus, and this in turn continues into the apex of the bladder.
UMBILICAL CORD The formation of the connecting stalk is described in Chapter 10, and the early formation of the umbilical cord is described on page 1259 (Fig. 105.3). The umbilical cord ultimately consists of an outer covering of flattened amniotic epithelial cells and an interior mass of mesenchyme of diverse origins (Fig. 10.22), which contains two endodermal tubes, the yolk and allantoic ducts, and their associated vitelline and allantoic (umbilical) blood vessels. The mesenchymal core is derived from the somatopleuric extraembryonic mesenchyme covering the amniotic folds, splanchnopleuric extraembryonic mesenchyme of the yolk stalk (which carries the vitelline vessels and clothes the endodermal yolk duct), and similar allantoic mesenchyme of the connecting stalk (which clothes the allantoic duct and initially carries two umbilical arteries and two umbilical veins). These various mesenchymal compartments fuse and are gradually transformed into the loose connective tissue (Wharton's jelly), which characterizes the more mature cord. The tissue consists of widely spaced elongated fibroblasts separated by a delicate three-dimensional meshwork of fine collagen fibres, which contains a variety of hydrated glycosaminoglycans, and is particularly rich in hyaluronic acid. The yolk stalk and endodermal duct remain within the cord. The allantois extends only into the proximal part of the umbilical cord. It may remain as a series of epithelial strands. The vitelline and allantoic (umbilical) vessels, which are initially symmetrical, become modified as a result of changes in the circulation. The vitelline vessels involute, whereas most of the allantoic (umbilical) vessels persist. The right umbilical vein disappears but the two umbilical arteries normally remain; occasionally one umbilical artery may disappear. The vessels of the umbilical cord are rarely straight, and are usually twisted into either a right- or left-handed cylindrical helix. The number of turns involved ranges from a few to over 300. This conformation may be produced by unequal growth of the vessels, or by torsional forces imposed by fetal movements. Its functional significance is obscure. Perhaps the pulsations and contractions of the helical vessels assist the
venous return to the fetus in the umbilical vein. Mature umbilical vessels, particularly the arteries, have a strong muscular coat, which contracts readily in response to mechanical stimuli. The outermost bundles pursue an interlacing spiral course, and when they contract they produce shortening of the vessel and thickening of the media, with folding of the interna and considerable narrowing of the lumen. This action may account for the periodic sharp constrictions of contour, the so-called valves of Hoboken, which often characterize these vessels. The fully developed umbilical cord is on average some 50 cm long and 1-2 cm in diameter. Its length varies from 20-120 cm. Exceptionally short or long cords are associated with fetal problems and complications during labour. The cord usually attaches to the placenta, but in a minority of cases velamentous insertion is observed (i.e. into the membranes) and this may be associated with vulnerability to injury and fetal haemorrhage.
© 2008 Elsevier
DEVELOPMENT OF THE PLACENTA The human placenta is initially labyrinthine as the early villous stems are formed, and becomes villous as generations of terminal villi develop. Maternal blood bathes the surfaces of the chorion that bound the intervillous space and the placenta is therefore defined as haemochorial. Different grades of fusion exist between the maternal and fetal tissues in many other mammals (e.g. epitheliochorial, syndesmochorial, endotheliochorial). The chorion is vascularized by the allantoic blood vessels of the body stalk and so the human placenta is also termed chorio-allantoic (whereas in some mammals a choriovitelline placenta either exists alone or supplements the chorio-allantoic variety). In addition, the human placenta is said to be dedicuate because maternal tissue is shed with the placenta and membranes at term as part of the afterbirth. As the blastocyst implants, the syncytiotrophoblast invades and digests the uterine tissues, including the glands and walls of maternal blood vessels (Figs 10.8, 10.9). The syncytiotrophoblast increases rapidly in thickness over the embryonic pole. A progressively thinner layer covers the rest of the wall towards the abembryonic pole. Microvillus-lined clefts and lacunar spaces develop in the syncytiotrophoblastic envelope (days 9-11 of pregnancy) and establish communications with one another. Initially, many of these spaces contain maternal blood derived from dilated uterine capillaries and veins, as the walls of the vessels are partially destroyed. As the conceptus grows, the lacunar spaces enlarge, and become confluent to form an intervillous space. page 1341 page 1342
Figure 105.3 Diagrams showing: A, an early stage in development of the human blastocyst; B, blastocyst sectioned so the embryo is longitudinally sectioned showing the early formation of the allantois and the connecting stalk; C, longitudinal section of embryo at a later stage of development; the pericardial cavity can be see at the most rostral part of the embryonic area; D, longitudinal section of embryo at a later stage showing formation of the head and tail folds, the expansion of the amnion and the delimitation of the umbilicus; E, a transverse section along the line a-b in D. Observe that the intraembryonic coelom communicates freely with the extraembryonic coelom; F, longitudinal section of embryo at a later stage showing full expansion of the amniotic cavity and the umbilical cord.
The projections of syncytiotrophoblast into the maternal decidua are called primary villi. They are invaded first with cytotrophoblast and then with mesenchyme (days 13-15) to form secondary placental villi. Fetal capillaries develop in the mesenchymal core of the villi. The cytotrophoblast within the villi continues to grow through the invading syncytiotrophoblast and makes direct contact with the decidua basalis, forming anchoring villi. Further cytotrophoblast proliferation occurs laterally so that neighbouring outgrowths meet to form a spherical cytotrophoblastic shell around the conceptus (Figs 10.9, 105.4). Lateral projections from the main stem villus form true and terminal villi (Fig. 105.4). As secondary villi form, single mononuclear cells become detached from the distal cytotrophoblast and infiltrate the maternal decidua. This process occurs in two phases: an initial infiltration of the decidua basilis, when interstitial extravillous trophoblast tends to accumulate in the vicinity of maternal spiral arteries; and a second wave of migration so that extravillous trophoblast reaches the inner onethird of the myometrium. At the same time, cytotrophoblast from the shell penetrates into and migrates along the inner walls of maternal spiral arteries (endovascular extravillous trophoblast) so that by the 18th week it has reached the inner myometrial segments. The interstitially migrating cells appear to have the capacity to invade arteries from their adventitia, and are presumably involved in the remodelling of the maternal arteries. The latter lose smooth muscle and associated elastic and collagenous matrix, which is replaced with non-resistive fibrinoid, an arrangement that permits an expansion of the vessels and as much as a 20-fold increase in the flow of blood into the intervillous space. Common pregnancy pathologies, including intrauterine growth retardation, pre-eclampsia and spontaneous abortion, are all associated with incomplete vascular remodelling, which probably reflects a failure of penetration by extravillous trophoblast.
With the onset of the embryonic heartbeat, a primitive circulation exists between the yolk sac, the embryo and the chorio-allantoic placenta. The embryonic side of the placenta is termed the chorionic plate and the maternal side the basal plate. Growing free villi permeate the intervillous space and are spanned by the early villous stems and their branches. page 1342 page 1343
page 1343 page 1344
Figure 105.4 Nutrition of oocyte, zygote, morula, free and implanted blastocyst, embryo and fetus throughout gestation. Embryonic and placental development proceed from left to right. Aspects of mature placental structure and circulation are shown below.
Figure 105.5 The general structure of the implanting blastocyst and its relationship to the tissues of the endometrium on the fifteenth day after fertilization. Note the arrangement and gradation in thickness of the syncytial trophoblast, which has eroded the maternal tissues. Some of the deeper trophoblastic lacunae already contain maternal blood.
Expansion of the entire conceptus is accompanied by radial growth of the villi and, simultaneously, an integrated tangential growth and expansion of the trophoblastic shell. Eventually each villous stem forms a complex consisting of a single trunk attached by its base to the chorion, from which second and third order branches (intermediate and terminal villi) arise distally. Terminal villi are specialized for exchange between fetal and maternal circulations. Each one starts as a syncytial outgrowth and is invaded by cytotrophoblastic cells, which then develop a core of fetal mesenchyme as the villus continues to grow. The core is vascularized by fetal capillaries (i.e. each villus passes through primary, secondary and tertiary grades of histological differentiation). The germinal cytotrophoblast continues to add cells that fuse with the overlying syncytium and so contribute to the expansion of the haemochorial interface. Terminal villi continue to form and branch within the confines of the definitive placenta throughout gestation, projecting in all directions into the intervillous space (Fig. 105.4). From the third week until about the second month of pregnancy, the entire chorion is covered with villous stems. They are thus continuous peripherally with the trophoblastic shell, which is in close apposition with both the decidua capsularis and the decidua basalis. The villi adjacent to the decidua basilis are stouter, longer and show a greater profusion of terminal villi. As the conceptus continues to expand, the decidua capsularis is progressively compressed and
thinned, the circulation through it is gradually reduced and adjacent villi slowly atrophy and disappear. This process starts at the abembryonic pole, and by the end of the third month, the abembryonic hemisphere of the conceptus is largely denuded. Eventually the whole chorion apposed to the decidua capsularis is smooth (the chorion laeve). In contrast, the villous stems of the disc-shaped region of chorion apposed to the decidua basalis increase greatly in size and complexity (the chorion frondosum), and together with the decidua basalis constitutes the definitive placental site. Coincidentally with the growth of the embryo and the expansion of the amnion, the decidua capsularis is thinned and distended (Figs 105.5, 105.1) and the space between it and the decidua parietalis gradually obliterated. By the second month of pregnancy the three endometrial strata recognizable in the premenstrual phase, compactum, spongiosum and basale, are better differentiated and easily distinguished. In the spongiosum the glands are compressed and appear as oblique slit-like fissures lined by low cuboidal cells. By the beginning of the third month of pregnancy the decidua capsularis and decidua parietalis are in contact, while by the fifth month the decidua capsularis is greatly thinned, and during the succeeding months (Fig. 105.6) it virtually disappears.
INTERVILLOUS SPACE page 1344 page 1345
Figure 105.6 A full-term human fetus in utero, including a sectional view of the placenta, the amnion (mauve), chorion (green), uterine wall and cervix (yellow), the cervix with a plug of mucus in the cervical canal, the umbilical cord and its contained vessels, and the rugose vaginal wall. Note the characteristic flexed posture of the fetus and its limbs, and the overall position within the uterus that the fetus commonly occupies. The single umbilical vein carries oxygenated blood, the two umbilical arteries carry deoxygenated blood. These vessels arborize in the chorionic plate (seen through the overlying amnion) and their branches pass into the villous stems. The latter span the intervillous space where they branch into intermediate and terminal villi. Incomplete placental septa project from the basal plate towards the chorionic plate.
The intervillous space, at first spanned by the early villous stems and their branches, is increasingly permeated by growing free villi (Figs 105.4, 105.7, 105.8). It contains the circulating maternal blood. On its fetal aspect, it is bounded by a chorionic plate, which consists of syncytial, cytotrophoblastic and mesenchymal layers of the chorion. The latter carry radicles of the umbilical vessels and fuse laterally with the mesenchyme of the expanding amnion. On its maternal aspect it is bounded by a basal plate, which consists of an incomplete peripheral syncytium with an outer cytotrophoblastic shell and columns of
cytotrophoblast, which extend deeper into the maternal decidual stroma. The trophoblast and adjacent decidua are enmeshed in layers of fibrinoid and basement membrane-like extracellular matrix to form a complex junctional zone. Where a discrete layer of fibrinoid is present between the trophoblastic shell and decidual stroma it is known as Nitabuch's layer. The intervillous space from chorionic to basal plates contains the main trunks of the villous stems dividing into their intermediate and terminal villi. The trunk and its branches may be regarded as the essential structural, functional and growth unit of the developing placenta. The maternal blood vessels approach and reach the intervillous space through the various layers of the basal plate. The spiral arteries of the endometrium open through gaps in the cytotrophoblastic shell and peripheral syncytium. They probably do not open directly into the intervillous space until as late as the tenth week. At term, from the inner myometrium to the intervillous space, the walls of most spiral arteries consist of fibrinoid matrix within which cytotrophoblast is embedded. This arrangement allows expansion of the arterial diameter (and so an increased blood flow) independent of the local action of vasocontrictive agents. Endothelial cells, where present, are often hypertrophic. The veins that drain the blood away from the intervillous space pierce the basal plate and join tributaries of the uterine veins. The presence of a marginal venous sinus, which hitherto has been described as a constant feature occupying the peripheral margin of the placenta and communicating freely with the intervillous space, has not been confirmed. Experimental studies have shown that radio-opaque material injected into the aorta passes in spurts or jets to the intervillous space and at sufficient pressure to drive it towards the chorion, thus preventing a short circuit of arterial blood into the venous openings. The openings of the coiled arteries show intermittent activity. Myometrial contractions alter the pressure in the intervillous space and promote placental venous drainage.
CHORIONIC PLATE The chorionic plate is covered on its fetal aspect by the amniotic epithelium, on the stromal side of which is a connective tissue layer carrying the main branches of the umbilical vessels (Figs 105.4, 105.7). Subjacent to this is a diminishing layer of cytotrophoblasts and then the inner syncytial wall of the intervillous space. The connective tissue layer is formed by fusion between the mesenchymecovered surfaces of amnion and chorion. It is more fibrous and less cellular than Wharton's jelly (of the umbilical cord), except near the larger vessels. The vessels radiate and branch from the cord attachment, with variations in the branching pattern, until they reach the bases of the trunks of the villous stems and then arborize within the intermediate and terminal villi. There are no anastomoses between vascular trees of adjacent stems. The two umbilical arteries are normally joined at, or just before they enter, the chorionic plate, by some form of substantial transverse anastomosis, (Hyrtl's anastomosis).
BASAL PLATE The basal plate, from fetal to maternal aspect, consists of the outer wall of the intervillous space. In different places, this may contain syncytium, cytotrophoblast or fibrinoid matrix, Rohr's stria of fibrinoid, remnants of the cytotrophoblastic shell, Nitabuch's stria of fibrinoid, and maternal decidua (Figs 105.4, 105.7). Nitabuch's stria and the decidua basilis contain cytotrophoblast and multinucleate trophoblast giant cells derived from the mononuclear cytotrophoblast population, which infiltrates the decidua basilis during the first 18 weeks of pregnancy. These cells penetrate as far as the inner one-third of the myometrium, but can often be observed at or near the decidual-myometrial junction. They are not found in the decidua parietalis or the adjacent myometrium, from which it may be inferred that the placental bed giant cell represents a differentiative end stage in the extravillous trophoblast lineage. The striae of fibrinoid are irregularly interconnected and variable in prominence. Strands pass from Nitabuch's stria into the adjacent decidua, which contains basal remnants of the endometrial glands and large and small decidual cells scattered in a connective tissue framework. The latter supports an extensive venous plexus. Throughout the second half of pregnancy the basal plate becomes thinned and progressively modified. There is a relative diminution of the decidual elements, increasing deposition of fibrinoid, and admixture of fetal and maternal derivatives.
STRUCTURE OF A VILLUS Chorionic villi are the essential structures involved in exchanges between mother and fetus. The villous tissues separating fetal and maternal blood are therefore of crucial functional importance. From the chorionic plate, progressive branching occurs into the villous tree, as stem villi give way to intermediate and terminal villi. Each villus has a core of connective tissue containing collagen types I, III, V and VI, as well as fibronectin. Cross-banded fibres (30-35 nm) of type I collagen often occur in bundles. Type III collagen is present as thinner (10-15 nm) beaded fibres, which form a meshwork that often encases the larger fibres. Collagens V and VI are present as 6-10 nm fibres closely associated with collagens I and III. Laminin and collagen type IV are present in the stroma associated with basal laminae surrounding fetal vessels and in the trophoblast basal lamina. Overlying this matrix are ensheathing cyto- and syncytial trophoblast cells bathed by the maternal blood in the intervillous space (Figs 105.4, 105.7, 105.9). Cohesion between the cells of the cytotrophoblast and also between the cytotrophoblast and the syncytium is provided by numerous desmosomes between the apposed plasma membranes. page 1345 page 1346
Figure 105.7 The arrangement of the placental tissues. The intervillous space is spanned by a villous stem and its divisions. The sectioned surfaces show the disposition of the fetal and maternal blood vessels, the amniotic epithelium, the cellular and syncytial layers of trophoblast and the complex junctional zone between the fetal and maternal tissues in the basal plate, which contains deposits of fibrinoid material and isolated masses of peripheral syncytium. Note surface syncytial sprouts and Hofbauer cells (large phagocytic cells) associated with a terminal villus. Syncytial fusion is occurring between the tips of two terminal villi.
Figure 105.8 A syncytial sprout.
In earlier stages, the cytotrophoblast forms an almost continuous layer on the basal lamina. After the fourth month it gradually expends itself producing syncytium, which comes to lie on the basal lamina over an increasingly large area and becomes progressively thinner. A few cytotrophoblastic cells, usually disposed singly, persist until term. In the first and second trimester cytotrophoblastic sprouts, covered in syncytium, are present; they represent a stage in the development of new villi (Fig. 105.8). Cytotrophoblast columns at the tips of anchoring villi extend from the villous basal lamina to the maternal decidual stroma. The cells of the villous cytotrophoblast (Langhans cells) are pale-staining with a slight basophilia. Ultrastructurally, they have a rather electron-translucent cytoplasm, and relatively few organelles. These cells contain intermediate filaments particularly in association with desmosomes. Between the desmosomes, the membranes of adjacent cells are separated by c.20 nm. Sometimes the intercellular gap widens to accommodate microvillous projections from the cell surfaces, and it occasionally contains patches of fibrinoid. A smaller population of intermediate cytotrophoblast may also be found in the chorionic villi. This postmitotic population represents a state of partial differentiation between the cytotrophoblast stem cell and the overlying syncytium. The syncytium is an intensely active tissue layer across which most transplacental transport must occur. It secretes a range of placental hormones into the maternal circulation. Syncytial cytoplasm is more strongly basophilic than that of the Langhans cells and is packed with organelles consistent with its secretory phenotype. Where the plasma membrane adjoins basal lamina it is often infolded into the cytoplasm, whereas the surface bordering the intervillous space is set with numerous long microvilli, which constitute the brush border seen by light microscopy. Glycogen is thought to be present in both layers of the trophoblast at all stages, although it is not always possible to demonstrate its presence histochemically. Lipid droplets occur in both layers and are free in the core of the villus. In the trophoblast they are found principally within the cytoplasm, but they also occur in the extracellular space between cytotrophoblast and syncytium, and between the individual cells of the cytotrophoblast. The droplets diminish in number with advancing age and may represent fat in transit from mother to fetus, and/or a pool of precursors for steroid synthesis. Membrane-bound granular bodies of moderate electron density occur in the cytoplasm, particularly in the syncytium, some of which are probably secretion granules. Other membrane-bound bodies, lysosomes and phagosomes, are involved in the degradation of materials engulfed from the intervillous space. In the immature placenta, syncytial sprouts (Fig. 105.8) represent the first stages in the development of new terminal villi, which later become invaded by cytotrophoblast and villous mesenchyme. Occasionally, adjacent syncytial sprouts make contact and fuse to form slender syncytial bridges. The sprouts may become detached, forming maternal syncytial emboli, which pass to the lungs. It has been computed that some 100,000 sprouts pass daily into the maternal circulation. In the lungs they provoke little local reaction and apparently disappear by lysis. However, they may occasionally form foci for neoplastic growth. Syncytial sprouts are present in the term placenta, but are usually degenerating. Syncytial knots are aggregates of degenerating nuclei, which may represent a sequestration phenomenon by which senescent nuclear material is removed from adjacent metabolically active areas of syncytium. Fibrinoid deposits are frequently found on the villous surface in areas lacking syncytiotrophoblast. They may constitute a repair mechanism in which the fibrinoid forms a wound surface that is subsequently re-epithelialized by trophoblast. The extracellular matrix glycoprotein tenascin has been localized in the stroma adjacent to these sites. The core of a villus contains small and large reticulum cells, fibroblasts, and large phagocytic Hofbauer cells, which are more numerous in early pregnancy. Early mesenchymal cells probably differentiate into small reticulum cells, which in turn produce fibroblasts or large reticulum cells. The small reticulum cells appear to delimit a collagen-free stromal channel system through which Hofbauer cells migrate. Mesenchymal collagen increases from a network of fine fibres in early mesenchymal villi to a densely fibrous stroma within stem villi in the second and third trimester. After approximately 14 weeks, the stromal channels found in
immature intermediate villi are infilled by collagen to give the fibrous stroma characteristic of the stem villus. page 1346 page 1347
Figure 105.9 A, Chorionic villus and its arterio-capillary-venous system carrying fetal blood. The artery carries deoxygenated blood and waste products from the fetus, whereas the vein carries oxygenated blood and nutrients to the fetus. Sections through a chorionic villus at 10 weeks, B, and at full term, C. The villi are bathed externally in maternal blood. The placental membrane, composed of fetal tissues, separates the maternal blood from the fetal blood.
Fetal vessels include arterioles and capillaries. Pericytes may be found in close association with the capillary endothelium. From late first trimester the vessels are surrounded externally by a basal lamina. From the second trimester (and a little later in terminal villi), dilated thin-walled capillaries are found immediately adjacent to the villous trophoblast; their respective basal laminae apparently fuse to produce a vasculo-syncytial interface.
© 2008 Elsevier
MATURATION AND FUNCTIONS OF THE PLACENTA In the early stages of placental development the blood in the fetal vessels is separated from the maternal blood in the intervillous space by the fetal vascular endothelial cells, the connective tissue of the villus, the subepithelial basal lamina and its covering of cyto- and syncytial trophoblast. These constitute a placental barrier interposed between the bloodstreams. It is a selectively permeable barrier and allows water, oxygen and other nutritive substances and hormones to pass from mother to fetus, and some of the products of excretion to pass from fetus to mother. Throughout pregnancy, the placenta increases its surface area and thickness, and there are concomitant increases in the size, length and complexity of branching of the villous stems. At term, the placental diameter varies from 200 to 220 mm, the mean placental weight is 470 g, its mean thickness is 25 mm and the total villous surface area exceeds 10 m 2. The placental barrier becomes reduced in thickness during gestation. After the fourth month the villous syncytium comes into direct apposition with the subepithelial basal lamina over an increasing area (80% at term) and it becomes thinner. The fetal capillaries approach the surface of the terminal villi and become dilated. The mechanism of transfer of substances across the placental barrier is complex. The volume of maternal blood circulating through the intervillous space has been assessed at 500 ml per minute. Simple diffusion suffices to explain gaseous exchange. Transfer of ions and other water-soluble solutes is by paracellular and transcellular diffusion and transport, although the relative importance of each of these for most individual solutes is unknown, and the paracellular pathway is morphologically undefined. Glucose transfer involves facilitated diffusion, while active transport mechanisms carry calcium and at least some amino acids . The fat-soluble and water-soluble vitamins are likely to pass the placental barrier with different degrees of facility. The water-soluble vitamins B and C pass readily. Water is interchanged between fetus and mother (in both directions) at c.3.5 litres per hour. The transfer of substances of high molecular weight, such as complex sugars, some lipids and hormonal and non-hormonal proteins, varies greatly in rate and degree, and is not so well understood. Energy-dependent selective transport mechanisms including receptor-mediated transcytosis are likely to be involved. Lipids may be transported unchanged through and between the cells of the trophoblast to the core of the villus. The passage of maternal antibodies (immunoglobulins) across the placental barrier confers some degree of passive immunity on the fetus. In this instance it is widely accepted that transfer is by micropinocytosis. Investigation of transplacental mechanisms is complicated by the fact that the trophoblast itself is the site of synthesis and storage of certain substances, e.g. glycogen. The placenta is an important endocrine organ. Some steroid hormones, various oestrogens, !-endorphins, progesterone , hCG and human chorionic
somatomammotropin (hCS), which is also known as placental lactogen (hPL), are synthesized and secreted by the syncytium. The trophoblast also contains enzyme systems that are associated with the synthesis of steroid hormones. It has been suggested that leukocytes may migrate from the maternal blood through the placental barrier into the fetal capillaries. It has also been shown that some fetal and maternal red blood cells may cross the barrier. The former may have important consequences, e.g. in Rhesus incompatibility. page 1347 page 1348
The majority of drugs are small molecules, which are sufficiently lipophilic to pass the barrier. Many are tolerated by the fetus, but some may exert grave teratogenic effects on the developing embryo (e.g. thalidomide ). A well-documented association exists between maternal alcohol ingestion and fetal abnormalities. Addiction of the fetus can occur to substances of maternal abuse such as cocaine and heroin. A wide variety of bacteria, spirochaetes, protozoa and viruses, including human immunodeficiency virus (HIV), are known to pass the placental barrier from mother to fetus, although the mechanism of transfer is uncertain. The presence of maternal rubella in the early months of pregnancy is of especial importance in relation to the production of congenital anomalies.
© 2008 Elsevier
THE PLACENTA AT TERM After delivery of the fetus the placenta becomes separated from the uterine wall and, together with the so-called 'membranes', is expelled as the afterbirth. Separation takes place along the plane of the stratum spongiosum and extends beyond the placental area, detaching the villous placenta, with associated fibrinoid matrix and small amounts of decidua basale, and the chorio-amnion, together with a superficial layer of the fused decidua capsularis and decidua parietalis. The process of separation ruptures many uterine vessels. However, under normal circumstances postpartum haemorrhage is limited after delivery of the placenta and membranes because the firm contraction of the muscular wall of the uterus closes the torn ends of the vessels. When the placenta and membranes have been expelled, a thin layer of stratum spongiosum is left as a lining for the uterus.It soon degenerates and is cast off in the early part of the puerperium. A new epithelial lining for the uterus is regenerated from the remaining stratum basale. The chorio-amnion is continuous with the placenta at its margin and constitutes the 'membranes' familiar in obstetrics. When ligature of the umbilical cord is delayed, the blood volume of the child is, on average, appreciably greater than it is when the ligature is applied at the earliest possible moment. It appears that in the former case much of the blood in the fetal placental vessels is transferred from the placenta to the fetus.
Figure 105.10 The fetal surface of a recently delivered placenta. The spiral umbilical vessels in the umbilical cord, and their radiating branches shine through the transparent amnion. The maternal surface is exposed in the lower and right corner of the figure. Note the fringes of amnion and chorion, the majority of which have been
cut away near the placental margin. (Drawn from a coloured photograph provided by EF Gibberd.)
The expelled placenta is a flattened discoidal mass with an approximately circular or oval outline (Fig. 105.10). It has an average volume of 500 ml (range 200-950 ml), an average weight of 470 g (range 200-800 g), an average diameter of 185 mm (range 150-200 mm), an average thickness of 23 mm (range 10-40 mm), and an average surface area of c.30000 mm2. Thickest at its centre (the original embryonic pole), it rapidly thins towards its periphery where it continues as the chorion laeve. Macroscopically, the fetal or inner surface, covered by amnion, is smooth, shiny and transparent, so that the mottled appearance of the subjacent chorion, to which it is closely applied, can be seen. The umbilical cord is usually attached near the centre of the fetal surface, and branches of the umbilical vessels radiate out under the amnion from this point; the veins being deeper and larger than the arteries. The remains of the yolk sac can sometimes be identified beneath the amnion and close to the attachment of the cord, as a minute vesicle, up to 5 mm in diameter. A fine thread, which is a vestige of the yolk stalk, is attached to it. The maternal surface of the placenta is finely granular and mapped into some 1530 lobes by a series of fissures or grooves. The placental lobes correspond in large measure to the major branches of distribution of the umbilical vessels, and this is particularly well seen in specimens which have been X-rayed after intravascular injection of radio-opaque media. The lobes are often somewhat loosely termed cotyledons. The grooves correspond to the bases of incomplete placental septa, which become increasingly prominent from the third month onwards. They extend from the maternal aspect of the intervillous space (the basal plate) towards, but do not quite reach, the chorionic plate. The septa are complex structures composed of components of the cytotrophoblastic shell and residual syncytium, together with maternally derived material including decidual cells, occasional blood vessels and gland remnants, collagenous and fibrinoid extracellular matrix and, in the later months of pregnancy, foci of degeneration. The nature of the maternal surface of the expelled placenta is determined by the tissue plane of separation of the placenta at parturition.
PLACENTAL VARIATIONS The placenta is usually attached to the posterior wall of the uterus near the fundus, with its centre in or near the median plane. The site of attachment is determined by the point where the blastocyst becomes embedded, but the factors on which this depends are not understood. The placenta may be attached at any point on the uterine wall, offering no complications to a normal labour unless it is so low down that it overlies the internal os, in which case serious antepartum haemorrhage may occur, especially if it is nearly central in position. This occurs in c.1 in 400 pregnancies and is known as placenta praevia. (Extrauterine sites of implantation are described in Ch. 10.) The umbilical cord, although usually attached near the centre of the placenta, may
reach it at any point between its centre and margin; the latter condition is known as a battledore placenta. Occasionally the cord fails to reach the placenta itself and ends in the membranes in its vicinity. When insertion of the cord is so velamentous, the larger branches of the umbilical vessels traverse the membranes before they reach and ramify on the placenta. A small accessory (succenturiate) placental lobe is occasionally present, connected to the main organ by membranes and blood vessels. It may be retained in utero after delivery of the main placental mass and prolong postpartum haemorrhage. Occasionally other degrees of division occur (bipartite or tripartite placentae). Other variations include placenta membranacea, in which villous stems and their branches persist over the whole chorion, and placenta circumvallata, where the placental margin is undercut by a deep groove. Pathological forms of adherence or penetration include: placenta accreta, which displays exceptional adherence to the decidua basalis; placenta increta, in which the myometrium is invaded; and placenta percreta, when the invasion by placental tissue passes completely through the uterine wall.
© 2008 Elsevier
NUTRITION OF THE EMBRYO page 1348 page 1349
In early development the blastomeres derive their nourishment in part from stores laid down in the cytoplasm of the primary oocyte. These stores are not as extensive as those found in the yolk of most non-mammalian species, and it is assumed that the human embryo also derives nutrition from tubal and uterine secretions. The cleaving embryo uses pyruvate rather than glucose as an energy substrate, but switches to utilizing glucose at the blastocyst stage. New protein production occurs during the preimplantation phase, but ongoing protein breakdown is responsible for a slight net decrease in protein content. During the process of implantation, breakdown products derived from lysed uterine tissues may provide a source of nutrition. There follows a period of about two weeks during which the embryonic disc is dependent on nutrients obtained from the fluid-filled cavities of the amnion, the coelom and the yolk sac. These fluids contain products arising from extravasated maternal blood absorbed by trophoblasts. However, early in development these sources of supply diminish. After gastrulation the lumen of the neural tube is isolated by closure of the neuropores, the extraembryonic coelom becomes greatly reduced (Fig. 105.3) and is later shut off from the intraembryonic coelom, and the yolk sac is separated from the gut by the narrowing of the yolk duct. Absorption of nutrients over the surface of the embryo becomes inadequate as the surface-to-volume ratio decreases. It therefore becomes imperative that some other source of nutrients should be available at an early stage. This involves the maternal circulation coming into close, although indirect, apposition with the developing embryonic circulation. The differentiating angioblastic mesenchyme in which the embryonic vessels and erythrocytes develop is first formed early in the third week from the extraembryonic mesenchyme beneath the mesothelium that clothes the yolk sac. Slightly later, angioblastic mesenchyme appears around the allantois in the connecting stalk, within the mesenchyme of the chorion, and, later still, within the embryonic area. Spaces form within the angioblastic mesenchyme, and the cells lining these spaces differentiate into typical flattened endothelial cells. Neighbouring spaces join to form capillary plexuses. Meanwhile, small localized groups of mesenchymal cells project into the spaces and become cut off to form blood islands, their cells differentiating into embryonic erythrocytes. The development of extraembryonic vessels around the chorion (Fig. 105.4) and within the early placenta, which form an intimate relationship with the maternal circulation, occurs ahead of embryonic development. Thus chorionic villi and the intervillous space are enlarging when the primitive streak forms. By the time the body plan stage is attained and the heart beats, an early circulation is present.
© 2008 Elsevier
ANATOMY OF PREGNANCY AND PARTURITION During pregnancy many morphological changes occur in the female reproductive system and associated abdominal structures. The uterus enlarges to accommodate the developing fetus and placenta, and there are various alterations in the pelvic walls, floor and contents which allow for this expansion, and also anticipate parturition. At the end of this period dramatic changes take place, which facilitate the passage of the baby through the birth canal. These alterations are often neglected by anatomists, but they represent an important aspect of normal reproductive morphology, and have considerable clinical significance.
UTERINE SIZE IN PREGNANCY The uterus grows dramatically during pregnancy and increases in weight from c.50 g at the beginning of pregnancy to up to 1 kg at term. Most of this gain in weight results from increases in the vascularity and tissue fluid of the uterine wall, together with myometrial growth (Ch. 104). The increased growth of the uterine wall is driven by a combination of mechanical stretching as the conceptus grows, and the stimulus of oestrogen and progesterone . The smooth muscle mass of the myometrium is thought to increase mainly by hypertrophy, although some hyperplasia occurs early on in pregnancy. As pregnancy proceeds, the uterus expands out of the pelvic basin and is usually palpable just above the pubic symphysis by the twelfth week, unless the uterus is retroverted (Ch. 104), in which case it may not be palpable in the abdomen until a little later. In past obstetric practice, anatomical surface landmarks such as the pubic symphysis, umbilicus and xiphisternum were used to estimate uterine size and therefore gestational age. So, for example, by the twentieth week of pregnancy the uterine fundus has usually risen to the level of the umbilicus, and by 36 weeks the uterine fundus has reached the xiphisternum. However, with the advent of diagnostic ultrasound (which allows both accurate dating of pregnancy and detection of many fetal anomalies), it has become clear that there is great variation in uterine size for a given gestation, and that clinical estimates based on anatomical landmarks are of limited value. In late pregnancy, fetal size and growth can be assessed by serial measurement of the distance between the pubic symphysis and the uterine fundus. The symphysis-fundus measurements act primarily as a screening method, and more accurate ultrasound assessment of fetal biometry is usually held in reserve (Fig. 105.11).
RELATIONS OF THE UTERUS IN PREGNANCY With uterine expansion, the ovaries and uterine tubes are displaced upwards and laterally. The round ligaments become hypertrophied and their course from the cornual regions of the uterus down to the internal inguinal ring becomes more vertical. The broad ligament tends to open out to accommodate the massive increase in the sizes of the uterine and ovarian vessels. The uterine veins in particular can reach c.1 cm in diameter and, for this reason, they appear to act as a significant reservoir for blood after uterine contraction. Lymphatics and nerves similarly proliferate, although the significance of the increased innervation is not clear, because paraplegic women are able to labour normally, albeit painlessly. The uterine fundus comes into contact with the anterior abdominal wall at c.16-20 weeks' gestation. Later in pregnancy the increase in intra-abdominal pressure
produced by the gravid uterus may produce eversion of the umbilicus. On the skin over the abdomen, a combination of stretching and hormonal changes may produce stretch marks (striae gravidarum). In multiparous patients, separation of right and left rectus abdominis may allow the uterine fundus to fall forwards to some extent. In the supine position, the pregnant woman in late pregnancy is vulnerable to aortocaval compression, as the enlarged uterus presses on, and reduces blood flow in, the great vessels. Symptoms of nausea and faintness may be obvious, and uteroplacental blood flow may be impaired in some cases. The jejunum, ileum and transverse colon tend to be displaced upwards by the enlarging uterus, whereas the caecum and appendix are displaced to the right, and the sigmoid colon posteriorly and to the left. Upward and lateral displacement of the appendix in later pregnancy can cause difficulties in the diagnosis of appendicitis. The ureters are pushed laterally by the enlarging uterus and in late pregnancy can be compressed at the level of the pelvic brim, resulting in hydronephrosis and loin pain. However, mild ureteric dilatation is normal in pregnancy, caused by progesterone-induced relaxation of smooth muscle in the ureteric walls. The axis of the uterus is shifted or dextrorotated by the presence of the sigmoid colon and this may lead to inadvertent incision into large uterine vessels at the time of lower segment caesarean section unless the operator is aware of any such rotation.
MATERNAL RESPIRATION, MICTURITION AND COLORECTAL CONTROL As the uterus grows there is some outward displacement of the chest, with flaring of the ribs. Although vital capacity is unchanged, tidal air is said to increase by 200 ml and residual volume to fall by the same amount. The respiratory rate may increase somewhat even in early pregnancy, possibly due to the effect of progesterone , while in late pregnancy, uterine expansion may limit diaphragmatic excursion. The bladder becomes hyperaemic in early pregnancy, and there is also an increase in the frequency of micturition because of the raised glomerular filtration rate at that time, and later due to pressure from the presenting part on the bladder. Such pressure may also provoke urinary stress incontinence in the third trimester. Changes in colorectal control are not very significant during pregnancy itself, although some women are troubled by constipation. Rectal sphincter damage can occur during childbirth and may lead to significant faecal incontinence if not recognized and adequately repaired at the time.
PELVIC CHANGES IN PREGNANCY page 1349 page 1350
Figure 105.11 Transabdominal ultrasound on a 16-week fetus. A, Biparietal diameter (an accurate method of dating the pregnancy). B, Longitudinal view of the thoracic spine. C, Four-chamber view of the heart (used to detect foetal anomalies). (By kind permission from Carien Laubscher, Superintendent in Ultrasound, Chelsea and Westminster Hospital, London.)
Figure 105.12 Sagittal T2-weighted scan, showing a normal cephalic presentation and the positions of the pelvic inlet and outlet at 30 weeks' pregnancy. (By kind permission from Kelly Wimpey, Superintendent in MRI, Chelsea and Westminster Hospital, London.)
The presence of a pregnant uterus results in a change in the centre of gravity of the body, especially in late pregnancy (Fig. 105.12). In order to compensate for this, the mother tends to straighten her cervical and thoracic spine, and throw her shoulders back, resulting in a compensatory lumbar lordosis. There is also a softening of the pubic symphysis and sacroiliac joints, caused by production of relaxin and other pregnancy hormones. This increased mobility produces a form of pelvic instability so that the pregnant woman tends to walk with a waddling gait. The result of this softening is an increase in pelvic diameter, which is of benefit during the time of labour. Significant joint relaxation can be associated with pain, sometimes called pelvic arthropathy, and in severe cases radiographs show that when a woman stands on one leg the two halves of the symphysis are almost at different levels. Rotation of the sacrum at the sacroiliac joint may increase the diameter of the pelvic outlet.
BIRTH CANAL AND PERINEUM DURING PARTURITION
The uterine cervix is required to serve two functions in relation to pregnancy and parturition. For 9 months the cervix is a relatively rigid fibromuscular structure, which retains the products of conception within the uterus, and yet, within a few hours during active labour, it has to dilate rapidly to allow the fetus to descend through the birth canal. In fact this transition is not as abrupt as might first appear, and there is considerable softening and shortening of the cervix in the weeks before the onset of labour. A corresponding increase in uterine activity is usually apparent during this prelabour period. The rigidity of the cervix appears to be related to the orientation of its collagen fibres within a regular connective tissue matrix. Softening of the cervix prior to and during labour is associated with a loss of this pattern of fibre distribution and a large increase in tissue water.
LABOUR page 1350 page 1351
The onset of labour is defined as the combination of regular and usually painful uterine contractions of sufficient intensity to produce progressive effacement and dilatation of the cervix. It is often difficult to define the exact time of the onset of labour, except retrospectively. The pain of labour contractions is thought to be caused by myometrial ischaemia produced by a reduction in uterine blood flow during the peak of a contraction. Uterine contractions direct the fetus against the cervix and at the same time result in retraction of the upper uterine segment, drawing the fibromuscular cervix upwards past the presenting part. The process of labour is described as having three main stages, as follows. First stage The first stage is defined as the period during which the cervix dilates as it is drawn up into the lower portion of the uterus until there is no longer any cervix palpable on vaginal examination, and thus no further impediment to the descent of the fetus through the birth canal. Second stage The second stage begins once the cervix is fully dilated, and ends with the delivery of the baby. Uterine contractions produce the descent of the fetal presenting part. Pressure at this stage on the pelvic diaphragm and rectum produces in the mother a reflex desire to 'bear down'. Thus involuntary maternal effort using the diaphragm and abdominal musculature augments uterine activity to help deliver the child. The head of the baby usually enters the pelvis with the occiput facing laterally. Further descent of the head results in the occiput contacting the gutter-shaped pelvic floor formed by levator ani. This promotes flexion and rotation of the occiput to the anterior position. With further descent the occiput escapes under the symphysis pubis and the head is born by extension. At this point, the baby's head regains its normal relationship with its shoulders, and slight rotation (or restitution) of the head is seen. Further external rotation occurs as the leading shoulder is directed medially by the maternal pelvic floor. The body of the baby is now born by lateral flexion as one shoulder slips underneath the symphysis and the posterior shoulder is drawn over the frenulum. Third stage The third stage is defined as the time from delivery of the fetus until delivery of the placenta. This process is usually expedited by the administration of oxytocic drugs in an attempt to limit maternal blood loss.
FACTORS AFFECTING THE PROGRESS OF LABOUR Many factors affect the progress of labour including the quality of uterine contractions, the size of the maternal bony pelvis, the size and position of the baby's head and the extent to which the skull will mould to the shape of the pelvis (Fig. 105.12). The complex nature of these interactions means that it is generally not possible to predict the outcome of labour with any degree of accuracy, and most pregnant women will be offered a trial of labour if there is any doubt, providing the baby is in cephalic presentation. Poor progress during labour is predominantly a problem in the first labour because of the combination of incoordinate uterine action and increased soft tissue resistance at the level of the cervix, pelvic diaphragm and perineum. Treatment of such primigravidae is aimed at improving uterine activity through the combination of artificial rupture of the fetal membranes and administration of an intravenous infusion of oxytocin . Over 80% of cases will respond to such a regime, the remainder usually requiring caesarean delivery. In multiparous women, slow progress in labour may be due to inadequate uterine contractions but is more commonly due to mechanical problems such as a large baby. Use of oxytocin in the context of such genuine mechanical difficulties can lead to uterine rupture.
OBSTETRIC EMERGENCIES Fetal distress The unborn baby derives its oxygen from the mother via the placenta and umbilical cord. Fetal hypoxia may occur if the uteroplacental circulation is inadequate, if the placenta separates from the uterine wall or if the cord is compressed. During labour, uterine contractions tend to reduce placental perfusion. In addition, cord compression may occur during contractions, particularly if the amniotic fluid volume is reduced. Indirect evidence of fetal hypoxia can be inferred from certain changes in the fetal heart rate such as reduced variation and decelerations occurring in the rate after uterine contractions. Confirmation of hypoxia can be achieved by obtaining a blood sample from the fetal scalp and measuring the acid-base balance of the specimen. Confirmation of significant hypoxia acidosis during labour (pH < 7.20) is an indication for immediate delivery. Prolapsed cord The umbilical cord may prolapse through the cervix into the vagina once the fetal membranes rupture. Conditions that prevent the fetal head from fully occupying the maternal pelvis will predispose to this problem, i.e. pelvic tumours (fibroids), ovarian cysts, placenta praevia and prematurity. Compression of the cord by the presenting part of the fetus, or an umbilical artery spasm will lead to fetal hypoxia and death if untreated. The treatment is either funic replacement (pushing the cord back above the fetal head) or immediate caesarean section, and the risk of perinatal death rises as the interval from diagnosis to delivery increases. Antepartum haemorrhage The two most serious causes of antepartum haemorrhage are placenta praevia and placental abruption. Placenta praevia In early pregnancy the placental disc occupies a large proportion of the uterine cavity and will often appear to be situated near the internal os, on ultrasonographic or MRI examination (Fig. 105.13). In the majority of cases where the placenta appears low in early pregnancy, growth and stretching of the uterus
will usually draw the placenta upwards away from the cervix by the end of pregnancy. In c.1% of pregnancies the position of the placenta will remain over, or in close proximity to, the internal cervical os at the end of pregnancy. This condition is called placenta praevia and is associated with vaginal bleeding during pregnancy and labour. The blood loss can be life-threatening for the mother. The diagnosis is confirmed by ultrasound examination and the usual therapeutic goal is to prolong pregnancy with hospitalization and if necessary provide blood transfusion until the fetus is of sufficient maturity to be delivered. Caesarean section is required and the procedure may be very haemorrhagic because of the increased vascularity of the lower uterine segment. Placental abruption Placental abruption, i.e. premature separation of the placenta from the uterine wall, is an emergency that may occur in either pregnancy or labour, and remains a significant cause of intrauterine death. The diagnosis is suggested by the onset of constant and severe abdominal pain, with the uterus appearing rigid and tender on abdominal palpation. Placental separation is usually accompanied by bleeding but the blood may initially be contained within the uterus and not be obvious on external examination. Release of thromboplastin from the damaged placenta into the maternal circulation causes disseminated intravascular coagulation and consumption of clotting factors, which may predispose to further maternal haemorrhage at the time of delivery. Transfusion of blood and clotting factors may be required to resuscitate the mother. If the fetus is showing signs of distress on presentation, caesarean section is usually undertaken. Postpartum haemorrhage Prior to separation of the placenta, a large proportion of the mother's cardiac output passes through the uterine circulation. After separation in the third stage of labour, exsanguination is only prevented by marked uterine contraction, with crisscrossing myometrial fibres acting as a tourniquet, restricting blood flow to the area that was the placental site. Therefore, any condition that predisposes to poor uterine contraction, such as retained placental tissue or blood clot within the uterus, will increase the likelihood of haemorrhage immediately after delivery. page 1351 page 1352
Figure 105.13 Sagittal T2-weighted scan showing a placenta praevia overlying the internal cervical os at 36 weeks' pregnancy. (By kind permission from Kelly Wimpey, Superintendent in MRI, Chelsea and Westminster Hospital, London.)
The other major cause of postpartum bleeding is that of trauma to the genital tract. Tearing will be found most frequently in the perineum and vagina but on occasion cervical laceration or even uterine rupture may be responsible for bleeding. Primary management of postpartum haemorrhaging is aimed at administering oxytocic drugs and resuscitating the mother with intravenous fluid. If this fails to stem the bleeding, then exploration of the genital tract under anaesthesia is undertaken to exclude retained placental tissue or genital tract trauma.
© 2008 Elsevier
106 FEMALE REPRODUCTIVE SYSTEM Vagina The vagina is a fibromuscular tube lined by non-keratinized stratified epithelium (Figs 102.1A,B, 102.2, 104.1, 104.4, 106.3). It extends from the vestibule (the cleft between the labia minora) to the uterus. The vagina ascends posteriorly and superiorly at an angle of over 90° to the uterine axis. This angle varies with the contents of the bladder and rectum, and the width increases as it ascends. Above the level of the hymen, the inner surfaces of the anterior and posterior vaginal walls are ordinarily in contact with each other forming a transverse slit. Its anterior wall is 7.5 cm in length and the posterior wall is 9 cm long on average. The upper end of the vagina surrounds the vaginal projection of the uterine cervix. The vaginal mucosa is attached to the uterine cervix higher on the posterior cervical wall than on the anterior. The annular recess between the cervix and vagina forms four fornices. These are called the anterior, posterior and, two lateral fornices. Although the different parts of this recess are given separate names, the recess is essentially continuous. Occasionally remnants of the duct of Gartner (p. 1384), Gartner's cysts, can be seen protruding through the lateral fornices or lateral parts of the vagina. These normal embryological remnants can sometimes be confused with cancerous lesions and cause clinical concern during a routine cervical screening examination. They are normally asymptomatic.
© 2008 Elsevier
RELATIONS The bladder and urethra are anterior to the vagina. The rectum and anal canal are posterior and separated from the upper part by the recto-uterine pouch (p. 1332). The anterior wall of the vagina is related to the urethra (which is embedded in it) inferiorly, and to the base of the bladder in its middle and upper portions. The posterior wall is covered by peritoneum in its upper quarter. It is separated from the rectum by the recto-uterine pouch superiorly, and by moderately loose connective tissue in its middle half (Denonvillier's fascia). In its lower quarter it is separated from the anal canal by the musculofibrous perineal body. Laterally are levator ani muscles (p. 1358) and pelvic fascia. As the ureters pass anteromedially to reach the fundus of the bladder, they pass close to the lateral fornices. As they enter the bladder the ureters are usually anterior to the vagina (p. 1375). At this point, each ureter is crossed transversely by a uterine artery (p. 1333).
VAGINAL FISTULA Childbirth can be complicated by cephalopelvic disproportion where the pelvic cavity is too small for the size of the fetal head. This results in obstructed labour, which is treated by Caesarean section. In the developing world, where early recourse to Caesarean section is not available, this can cause a prolonged labour and necrosis of the vagina anteriorly due to pressure of the fetal head. This necrosis can cause a connection between the bladder and vagina (vesicovaginal fistula), which results in urinary incontinence. Other causes of a vesicovaginal fistula include cancer, post-radiation therapy, and trauma. A connection can also occur between the vagina and rectum, resulting in a rectovaginal fistula.
© 2008 Elsevier
VASCULAR SUPPLY AND LYMPHATIC DRAINAGE ARTERIES Arterial supply is derived from the vaginal, uterine, internal pudendal and middle rectal branches of the internal iliac arteries. The vaginal artery often gives off two or three branches that correspond to the inferior vesical artery in males. They descend on the vagina and supply the mucous membrane. Branches are also sent to the vestibular bulb, vesical fundus, and adjacent part of the rectum.
VEINS The vaginal veins, one on each side, form from lateral plexuses that connect with uterine, vesical and rectal plexuses and drain to the internal iliac veins. The uterine and vaginal plexuses may provide collateral venous drainage to the lower limb.
LYMPHATIC DRAINAGE Vaginal lymphatic vessels link with those of the cervix uteri, rectum and vulva. They form three groups but the regions drained are not sharply demarcated. Upper vessels accompany the uterine artery to the internal and external iliac nodes. Intermediate vessels accompany the vaginal artery to the internal iliac nodes. Vaginal vessels below the hymen, and from the vulva and perineal skin, pass to the superficial inguinal nodes.
© 2008 Elsevier
INNERVATION Innervation is derived from the vaginal plexuses and pelvic splanchnic nerves (p. 1126). The lower vagina is supplied by the pudendal nerve (p. 1457). Many nerve fibres in the lamina propria and muscle are probably cholinergic. Vaginal nerves from the lower parts of the inferior hypogastric and uterovaginal plexuses follow the vaginal arteries to supply the vaginal walls, the erectile tissue of the vestibular bulbs and clitoris (cavernous nerves of the clitoris), the urethra and the greater vestibular glands. The nerves contain many parasympathetic fibres, which are vasodilatory to the erectile tissue.
© 2008 Elsevier
SPECULUM EXAMINATION (Figs 106.1, 106.2) The vagina is examined using either a bivalve or angled (Sim's) speculum. A bivalve speculum (Fig. 106.1A) exposes the cervix, vaginal fornices, and lateral walls of the vagina. The angled (Sim's) speculum (Fig. 106.1B) is used to assess for uterine and vaginal prolapse.
© 2008 Elsevier
MICROSTRUCTURE The vagina has an inner mucosal and an external muscular layer. The lamina propria of the mucosa contains many thin-walled veins.
MUCOSA (Figs 106.2) page 1353 page 1354
Figure 106.1 A, bivalve speculum (Cuscos') used for exposing the cervix and lateral vaginal walls during clinical examination. B, An angled speculum (Sim's) used for exposing the vaginal walls during clinical examination.
The mucosa adheres firmly to the muscular layer. There are two median longitudinal ridges on its epithelial surface, one anterior and the other posterior. Numerous transverse bilateral rugae extend from these vaginal columns. They are divided by sulci of variable depth, giving an appearance of conical papillae, which are most numerous on the posterior wall and near the orifice, and which are especially well developed before parturition. The epithelium is non-keratinized, stratified, squamous similar to, and continuous with, that of the ectocervix (p. 1334). After puberty it thickens and its superficial cells accumulate glycogen, which gives them a clear appearance in histological preparations.
Figure 106.2 Section through the mucosa of the vagina, showing the non-keratinized stratified squamous epithelium and the lamina propria beneath. Note the papillated interface between the epithelium and underlying connective tissue. Haematoxylin and eosin.
The vaginal epithelium does not change markedly during the menstrual cycle, but its glycogen content increases after ovulation and then diminishes towards the end of the cycle. Natural vaginal bacteria, particularly Lactobacillus acidophilus, break down glycogen in the desquamated cellular debris to lactic acid . This produces a highly acidic (pH 3) environment, which inhibits the growth of most other microorganisms. The amount of glycogen is less before puberty and after the menopause, when vaginal infections are more common. There are no mucous glands, but a fluid transudate from the lamina propria and mucus from the cervical glands lubricate the vagina.
MUSCULAR LAYERS The muscular layers are composed of smooth muscle and consist of a thick outer longitudinal and an inner circular layer. Longitudinal fibres are continuous with the superficial muscle fibres of the uterus, and the strongest fasciculi are those attached to the rectovesical fascia on each side. The two layers are not distinct but connected by oblique decussating fasciculi. The lower vagina is also surrounded by the skeletal muscle fibres of bulbospongiosus. A layer of loose connective tissue, containing extensive vascular plexuses, surrounds the muscle layers.
© 2008 Elsevier
107 FEMALE REPRODUCTIVE SYSTEM Female external genital organs The female external genitalia (Figs 107.1, 107.2) include the mons pubis, labia majora, labia minora, clitoris, vestibule, vestibular bulb and the greater vestibular glands. The term pudendum, or vulva, includes all these parts.
© 2008 Elsevier
SKIN MONS PUBIS The mons pubis is the rounded eminence that is anterior to the pubic symphysis. It is formed by a mass of subcutaneous adipose connective tissue. In adults, the mons is covered by coarse hair. This hair is usually limited above by a horizontal boundary. (In males there is a continuation of pubic hair to the umbilicus.)
LABIA MAJORA (Fig. 107.1)
Figure 107.1 The female external genitalia.
The labia majora are two prominent, longitudinal, cutaneous folds extending back from the mons pubis to the perineum. They form the lateral boundaries of the pudendal cleft, into which the vagina and urethra open. Each labium has an external, pigmented surface, covered with crisp hairs and a pink, smooth, internal surface with large sebaceous follicles. Between these surfaces there is much loose connective and adipose tissue, intermixed with smooth muscle resembling the scrotal dartos muscle, together with vessels, nerves and glands. The uterine
round ligament may end in the adipose tissue and skin in the front part of the labium. A persistent processus vaginalis and congenital inguinal hernia may also reach a labium. The labia are thicker in front, where they join to form the anterior commissure. Posteriorly they do not join but merge into neighbouring skin, ending near and almost parallel to each other. The connecting skin between them posteriorly forms a ridge called the posterior commissure. This overlies the perineal body and is the posterior limit of the vulva. The interval between this and the anus is c.2.5 to 3cm thick and is termed the 'gynaecological' perineum.
LABIA MINORA (Figs 107.1, 107.2) The labia minora are two small cutaneous folds, devoid of fat, that lie between the labia majora. They extend from the clitoris obliquely down, laterally and back for c.4cm, flanking the vaginal orifice. In virgins their posterior ends may be joined by the cutaneous frenulum of the labia minora. Anteriorly, each labium minus bifurcates. The upper layer passes above the clitoris to form with its fellow a fold, the prepuce, which overhangs the glans of the clitoris. The lower layer passes below the clitoris to form with its fellow the frenulum of the clitoris. Sebaceous follicles are numerous on the apposed labial surfaces. Sometimes an extra labial fold (labium tertium) is found on one or both sides between the labia minora and majora.
page 1355 page 1356
Figure 107.2 The female perineum. On the right side the bulb of the vestibule and greater vestibular gland are shown. On the left side the muscles superficial to these
structures are shown.
VESTIBULE (Fig. 107.2) The vestibule is the cavity that lies between the labia minora. It contains the vaginal and external urethral orifices and the openings of the two greater vestibular glands and those of numerous, mucous, lesser vestibular glands. There is a shallow vestibular fossa between the vaginal orifice and the frenulum of the labia minora.
CLITORIS (Fig. 107.2) The clitoris is an erectile structure, homologous with the penis, which lies posteroinferior to the anterior commissure. It is partially enclosed by the anterior bifurcated ends of the labia minora. The corpus clitoridis has two corpora cavernosa, which are composed of erectile tissue and enclosed in dense fibrous tissue separated medially by an incomplete fibrous pectiniform septum. Each corpus cavernosum is connected to its ischiopubic ramus by a crus. The glans of the clitoris is a small round tubercle of spongy erectile tissue. Its epithelium has high cutaneous sensitivity, which is important in sexual responses. The clitoris, like the penis, has a 'suspensory' ligament and two small muscles, ischiocavernosi, attached to its crura (p. 1316). In many anatomical details it is a small version of the penis, but differs from it in being separate from the urethra.
VAGINAL ORIFICE (INTROITUS) (Fig. 107.2) The introitus is usually a sagittal slit positioned posteroinferior to the urethral meatus. Its size varies. It is capable of great distension during parturition and to a lesser degree during coitus.
HYMEN VAGINAE The hymen is a thin fold of mucous membrane situated just within the vaginal orifice. The internal surfaces of the folds are normally in contact each other and the vaginal orifice appears as a cleft between them. The hymen varies greatly in shape and area. When stretched, it is annular and widest posteriorly. Sometimes it is semilunar, concave towards the mons pubis. Occasionally it is cribriform or fringed. It may be absent or form a complete, imperforate hymen. When it is ruptured, small round carunculae hymenales (also known as carunculae myrtiformis) are its remnants. It has no established function.
EXTERNAL URETHRAL ORIFICE (URINARY MEATUS) (Fig. 107.2) The urethra opens into the vestibule c.2.5cm inferiorly to the clitoris and anterior to the vaginal orifice. The meatus is usually a short, sagittal cleft with slightly raised margins and is very distensible. It varies in shape, and the aperture may exhibit either rounded, slit-like, crescentic or stellate forms.
BULBS OF THE VESTIBULE (Fig. 107.2) The bulbs of the vestibule are homologues of the single penile bulb and corpus
spongiosum. They are two elongate erectile masses, flanking the vaginal orifice and united in front of it by a narrow commissura bulborum (pars intermedia). Each lateral mass is c.3cm in length. Their posterior ends are expanded and are in contact with the greater vestibular glands. Their anterior ends are tapered and joined to one another by a commissure, and to the clitoris by two slender bands of erectile tissue. Their deep surfaces contact the inferior aspect of the urogenital diaphragm. Superficially each is covered by the bulbospongiosus posteriorly. Thus the female corpus spongiosum is split into bilateral masses, except in its most anterior region, by the vestibule and the vaginal and urethral orifices.
GREATER VESTIBULAR GLANDS (GLANDS OF BARTHOLIN) (Fig. 107.2) The greater vestibular glands (glands of Bartholin)are homologues of the male bulbourethral glands. They consist of two small, round or oval reddish-yellow bodies, flanking the vaginal orifice, in contact with, and often overlapped by, the posterior end of the vestibular bulb. Each opens into the vestibule, by a duct of c.2cm, in the groove between the hymen and a labium minus. The glands are composed of tubulo-acinar tissue. The secretory cells are columnar and secrete a clear or whitish mucus with lubricant properties. They are stimulated by sexual arousal. Ducts connecting the greater vestibular glands with the vagina may become blocked with proteinaceous material. This can cause the secretion from the gland to accumulate within it forming a cyst (Bartholin's cyst), which can appear as a pea- to grape-sized swelling bulging unilaterally from the lower part of the vulvovaginal margin. Occasionally a cyst becomes infected, forming an abscess. Treatment is surgical, by excision or marsupialization.
© 2008 Elsevier
VASCULAR SUPPLY AND LYMPHATIC DRAINAGE ARTERIES The arterial blood supply of the female external genitalia resembles that of homologous structures in males. Thus it is derived from the superficial and deep external pudendal branches of the femoral artery and the internal pudendal artery on each side. The blood supply is substantial and consequently haemorrhage from vulval injuries may be severe.
VEINS Venous drainage of the vulval skin is via external pudendal veins to the long saphenous vein. Venous drainage of the clitoris mirrors that of the penis and is via deep dorsal veins to the internal pudendal vein and superficial dorsal veins to the external pudendal and long saphenous veins (p. 1203).
LYMPHATIC DRAINAGE Lymphatic drainage of the vulva is via a meshwork of connecting vessels that emerge into three or four collecting trunks around the mons pubis. They drain to superficial inguinal nodes, then deep femoral nodes and eventually to pelvic nodes. The last of the deep femoral nodes lies under the inguinal ligament and is often called Cloquet's node. Lymph vessels in the perineum and lower part of the labia majora drain to the rectal lymphatic plexus (p. 1203). Lymph vessels from the clitoris and labia minora drain to deep inguinal nodes and direct clitoral efferents may pass to the internal iliac nodes.
© 2008 Elsevier
INNERVATION The sensory innervation of the anterior and posterior parts of the labium majus differ, as they do in the scrotum. The anterior third of the labium majus is supplied by the ilioinguinal nerve (L1). The posterior two-thirds are supplied by the labial branches of the perineal nerve (S3). The lateral aspect of the labium majus also receives nervous innervation from the perineal branch of the posterior cutaneous nerve of the thigh (S2).
© 2008 Elsevier
AGE-RELATED CHANGES Before adolescence, the mons pubis is relatively flat and the labia minora are poorly formed. The hymenal ring is normally narrow and well formed. During adolescence, coarse hair forms over the mons and labia majora. The labia minora also become more prominent and flap like. The hymenal ring normally ruptures after first sexual intercourse but can rupture earlier due to other non-sexual physical activity. After the menopause, pubic hair thins and labial tissue atrophies slightly.
© 2008 Elsevier
108 TRUE PELVIS, PELVIC FLOOR AND PERINEUM True pelvis, pelvic floor and perineum TRUE PELVIS AND PELVIC FLOOR The true pelvis is a bowl-shaped structure formed from the sacrum, pubis, ilium, ischium, the ligaments which interconnect these bones and the muscles which line their inner surfaces. The true pelvis is considered to start at the level of the plane passing through the promontory of the sacrum, the arcuate line on the ilium, the iliopectineal line and the posterior surface of the pubic crest. This plane, or 'inlet' lies at an angle of between 35 and 50° up from the horizontal and above this the bony structures are sometimes referred to as the false pelvis. They form part of the walls of the lower abdomen. The floor or 'outlet' of the true pelvis is formed by the muscles of levator ani. Although the floor is gutter shaped, it generally lies in a plane between 5 and 15° up from the horizontal. This difference between the planes of the inlet and outlet is the reason why the true pelvis is said to have an axis (lying perpendicular to the plane of both inlet and outlet) which progressively changes through the pelvis from above downwards. The details of the topography of the bony and ligamentous pelvis is considered fully on page 1428.
Muscles and fasciae of the pelvis PELVIC MUSCLES (Figs 108.1, 108.2)
Figure 108.1 Muscles of the male pelvis - lateral view. The superior gluteal and obturator vessels and nerves have been divided close to their exit from the pelvis. The rectum, bladder and upper prostate have been omitted for clarity.
The muscles arising within the pelvis form two groups. Piriformis and obturator internus, although forming part of the walls of the pelvis, are considered as primarily muscles of the lower limb. Levator ani and coccygeus form the pelvic diaphragm and delineate the lower limit of the true pelvis. The fasciae investing the muscles are continuous with visceral pelvic fascia above, perineal fascia below and obturator fascia laterally.
PIRIFORMIS (See also p. 1357.)
Piriformis forms part of the posterolateral wall of the true pelvis. It is attached to the anterior surface of the sacrum, the gluteal surface of the ilium near the posterior inferior iliac spine, the capsule of the adjacent sacroiliac joint and sometimes to the upper part of the pelvic surface of the sacrotuberous ligament. It passes out of the pelvis through the greater sciatic foramen. Within the pelvis, the anterior surface of piriformis is related to the rectum (especially on the left), the sacral plexus of nerves and branches of the internal iliac vessels. The posterior surface lies against the sacrum.
OBTURATOR INTERNUS (See also p. 1357.) page 1357 page 1358
Figure 108.2 Muscles of the female pelvis viewed from above. The sacral nerve roots have been divided close to the sacral foramina. The anorectal junction, vagina and urethra have been divided at the level of the pelvic floor.
Obturator internus and the fascia over its upper inner (pelvic) surface form part of the anterolateral wall of the true pelvis. It is attached to the structures surrounding the obturator foramen; the inferior ramus of the pubis, the ischial ramus, the pelvic surface of the hip bone below and behind the pelvic brim, and the upper part of the greater sciatic foramen. It also attaches to the medial part of the pelvic surface of the obturator membrane. The muscle is covered by a thick fascial layer and the fibres themselves cannot be seen directly from within the pelvis. This fascia gives attachment to some of the fibres of levator ani and thus only the upper portion of the muscle lies lateral to the contents of the true pelvis, whilst the lower portion forms part of the boundaries of the ischioanal fossa. In the male, the upper portion lies lateral to the bladder, the obturator and vesical vessels, and the obturator nerve. In the female, the attachments of the broad ligament of the uterus, the fallopian end of the uterine tubes, and the uterine vessels, also lie medial to obturator internus and its fascia.
LEVATOR ANI (ISCHIOCOCCYGEUS, ILIOCOCCYGEUS, PUBOCOCCYGEUS) Levator ani is a broad muscular sheet of variable thickness attached to the internal surface of the true pelvis and forms a large portion of the pelvic floor. The muscle is subdivided into named portions according to their attachments and the pelvic viscera to which they are related. These parts are often referred to as separate muscles, but the boundaries between each part cannot be easily distinguished and they perform many similar physiological functions. The separate parts are referred to as ischiococcygeus, iliococcygeus and pubococcygeus. Pubococcygeus is often subdivided into separate parts according to the pelvic
viscera to which they relate, i.e. pubourethralis and puborectalis in the male, pubovaginalis and puborectalis in the female. Levator ani arises from each side of the walls of the pelvis. Fibres from ischiococcygeus attach to the sacrum and coccyx but the remaining parts of the muscle converge in the midline. The fibres of iliococcygeus join by a partly fibrous intersection and form a raphe posterior to the anorectal junction. Closer to the anorectal junction and elsewhere in the pelvic floor, the fibres are more nearly continuous with those of the opposite side and the muscle forms a sling (puborectalis and pubovaginalis or pubourethralis). Attachments
Ischiococcygeus The ischiococcygeal part may be referred to as a separate muscle, sometimes named coccygeus. It lies as the most posterosuperior portion of levator ani and arises as a triangular musculotendinous sheet with its apex attached to the pelvic surface and tip of the ischial spine. The base of the muscle is attached to the lateral margins of the coccyx and the fifth sacral segment. Ischiococcygeus is rarely absent, but may be nearly completely tendinous rather than muscular. It lies on the pelvic aspect of the sacrospinous ligament and may be fused with it, particularly if mostly tendinous. The sacrospinous ligament may represent a degenerate part or an aponeurosis of the muscle since the muscle and ligament are coextensive. Iliococcygeus The iliococcygeal part is attached to the inner surface of the ischial spine below and anterior to the attachment of ischiococcygeus and to the obturator fascia as far forward as the obturator canal (Fig. 108.1). The most posterior fibres are attached to the tip of the sacrum and coccyx but most join with fibres from the opposite side to form a raphe. This raphe is effectively continuous with the fibroelastic anococcygeal ligament, which is closely applied to its inferior surface and some muscle fibres may attach into the ligament. The raphe provides a strong attachment for the pelvic floor posteriorly and must be divided to allow wide excisions of the anorectal canal during abdominoperineal excisions for malignancy. An accessory slip may arise from the most posterior part and is sometimes referred to as iliosacralis. Pubococcygeus
Figure 108.3 Fasciae of the pelvis and perineum. Median sagittal section in the male. The deep fascia of the abdominal wall, the layers of the urogenital fascia and the mesorectal fascia are in green, the peritoneum in blue, the superficial fascia of the abdominal wall and perineum in red. Muscles are shown in brown.
page 1358 page 1359
The pubococcygeal part is attached to the back of the body of the pubis and passes back almost horizontally. The most medial fibres run directly lateral to the urethra and its sphincter as it passes through the pelvic floor. In males these fibres therefore lie lateral and inferior to the prostate and are referred to as pubourethralis. They form part of the urethral sphincter complex together with the intrinsic striated and smooth musculature of the urethra and fibres decussate across the midline directly behind the urethra. In females the fibres of this part of the muscle run further back to from a sling around the posterior wall of the vagina and are referred to as pubovaginalis. In both sexes fibres from this part of pubococcygeus attach to the perineal body and a few elements also attach to the anorectal junction. Some of these fibres, sometimes called puboanalis, decussate and blend with the longitudinal rectal muscle and fascial elements to contribute to the conjoint longitudinal coat of the anal canal. Behind the rectum some fibres of pubococcygeus form a tendinous intersection as part of the levator raphe but a thick muscular sling, puborectalis, wraps around the anorectal junction. Some fibres blend with those of the external anal sphincter. Relations
The superior, pelvic surface of levator ani is separated only by fascia (superior pelvic diaphragmatic, visceral and extraperitoneal (p. 1359) from the urinary bladder, prostate or uterus and vagina, rectum and peritoneum. Its inferior, perineal, surface forms the medial wall of the ischioanal fossa and the superior wall of the anterior recess of the fossa, both being covered by inferior pelvic diaphragmatic fascia. The posterior border is separated from the coccyx by areolar tissue. The medial borders of the two levator muscles are separated by the visceral outlet, through which pass the urethra, vagina, and anorectum. Vascular supply
Levator ani is supplied by branches of the inferior gluteal artery, the inferior vesical artery and the pudendal artery. Innervation
Fibres originating mainly in the second, third and fourth sacral spinal segments reach levator ani from below and above by a variety of routes (Wendell-Smith & Wilson 1991). Most commonly pubococcygeus (puborectalis and pubovaginalis or pubourethralis) is supplied by second and third sacral spinal segments via the pudendal nerve, and ischiococcygeus and iliococcygeus by direct branches of the sacral plexus from third and fourth sacral spinal segments. Actions
Pubococcygeus is a lateral compressor of the various visceral canals which cross the pelvic floor. The puborectalis part also reinforces the external anal sphincter and helps to create the anorectal angle. It also reduces the anteroposterior dimension of the ano-urogenital hiatus. Iliococcygeus and, to a lesser extent, the less muscular ischiococcygeus, assist puborectalis in contributing to anorectal and urinary continence. It is well recognized that levator ani must relax appropriately to permit expulsion of urine and particularly faeces. Levator ani also forms much of the basin-shaped muscular pelvic diaphragm, which supports the pelvic viscera and it contracts with abdominal muscles and the abdominothoracic diaphragm to raise intra-abdominal pressure. Like the abdominothoracic diaphragm, but unlike abdominal muscles, levator ani is also active in the inspiratory phase of quiet respiration. In the pregnant female, the shape of the pelvic floor may help to direct the fetal head into the anteroposterior diameter of the pelvic outlet.
PELVIC FASCIAE (Fig. 108.3) The pelvic fasciae may be conveniently divided into the parietal pelvic fascia, which mainly forms the coverings of the pelvic muscles, and the visceral pelvic fascia, which forms the coverings of the pelvic viscera, their supplying vessels and nerves. The visceral pelvic fascia is described with the pelvic viscera.
PARIETAL PELVIC FASCIA The parietal pelvic fascia on the pelvic surface of obturator internus is well differentiated as the obturator fascia. Above, it is connected to the posterior part of the arcuate line of the ilium, and is continuous with iliac fascia. Anterior to this, as it follows the line of origin of obturator internus, it is gradually separated from the attachment of the iliac fascia and a portion of the periosteum of the ilium and pubis spans between them. It arches below the obturator vessels and nerve, investing the obturator canal, and is attached anteriorly to the back of the pubis.
Behind the obturator canal the fascia is markedly aponeurotic and gives a firm attachment to levator ani. Below the attachment of levator ani it is thin and forms part of the lateral wall of the ischioanal fossa in the perineum. It is continuous with the pelvic periosteum and thus the fascia over piriformis. page 1359 page 1360
FASCIA OVER PIRIFORMIS The fascia of piriformis is very thin, and fuses with the periosteum on the front of the sacrum at the margins of the anterior sacral foramina. It ensheathes the sacral anterior primary rami which emerges from these foramina, and the nerves are often described as lying behind the fascia. The internal iliac vessels lie in front of the fascia over piriformis and their branches draw out sheaths of the fascia and extraperitoneal tissue into the gluteal region, above and below piriformis.
FASCIA OF THE PELVIC DIAPHRAGM The fascia of the pelvic diaphragm covers both of the surfaces of the pelvic diaphragm. On the lower surface is the thin inferior fascia of the pelvic diaphragm, which is continuous with the obturator fascia laterally. It covers the medial wall of the ischioanal fossa and blends below with fasciae on the urethral sphincter and the external anal sphincter. On the upper surface is the superior fascia of the pelvic diaphragm which is generally known clinically as the endopelvic fascia. It is attached anteriorly to the back of the body of the pubis, c.2 cm above its lower border, and extends laterally across the superior ramus of the pubis, blending with the obturator fascia and continuing along an irregular line to the spine of the ischium. It is continuous posteriorly with the fascia over piriformis and the anterior sacrococcygeal ligament. Medially, the superior fascia of the pelvic diaphragm blends with the visceral pelvic fascia. The fascia over obturator internus above the attachment of levator ani is therefore composed of the obturator fascia itself, the superior and inferior pelvic diaphragmatic fasciae and fibres from levator ani. The thickening where these structures fuse is the tendinous arch of levator ani. Below it, within the superior fascia, is the tendinous arch of the pelvic fascia, a thick white band extending from the lower part of the symphysis pubis to the inferior margin of the spine of the ischium (arcus tendineous fasciae pelvis). This is the attachment of the lateral, 'true' ligament of the urinary bladder. Anteriorly the same fascia forms two thick bands, the paired puboprostatic ligaments in the male, or the pubourethral ligaments in the female.
PRESACRAL FASCIA The presacral fascia lies between the posterior aspect of the mesorectal fascia and the superior pelvic diaphragmatic fascia. It is a hammock-like sheet extending between the tendinous arches of the pelvic fascia on either side. Below, it extends to the anorectal junction, where it fuses with the posterior aspect of the mesorectal fascia at the level of the anorectal junction. Above, it can be traced to the origin of the superior hypogastric plexus where it becomes progressively thinner over the promontory of the sacrum and becomes continuous with the retroperitoneal tissues. The right and left hypogastric nerves and inferior hypogastric plexuses lie on its surface and the presacral veins lie immediately posterior to it. It forms a distinct layer which can be seen both on magnetic resonance images of the pelvis and during surgery. The fascia provides an important landmark because extension of rectal tumours through it signifiantly reduces the chance of curative resectional surgery being possible. Dissection in the plane posterior to it may result in bleeding from the presacral veins and, since the adventitia of the veins is partly attached to the posterior surface of the fascia, the haemorrhage may be severe because the veins are unable to contract down properly. UPDATE Abstract Abstract: Clinical anatomy of the pelvic floor.
Date Added: 30 August 2005
Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=15152384&query_hl=10 Clinical anatomy of the pelvic floor. Fritsch H, Lienemann A, Brenner E et al: Adv Anat Embryol Cell Biol. 175:III-IX, 1-64, 2004.
Vascular supply and lymphatic drainage of the pelvis The true pelvis contains the internal iliac arteries and veins as well as the lymphatics draining the majority of the pelvic viscera. The common and external
iliac vessels as well as the lymphatics draining the lower limb lie along the pelvic brim and in the lower retroperitoneum, but are conveniently discussed together with the vessels of the true pelvis.
ARTERIES OF THE PELVIS COMMON ILIAC ARTERIES The abdominal aorta bifurcates into the right and left common iliac arteries anterolateral to the left side of the fourth lumbar vertebral body. These arteries diverge as they descend to divide at the level of the sacroiliac joint into external and internal iliac arteries. The external iliac artery is the principal artery of the lower limb and the internal iliac artery provides the principal supply to the pelvic viscera and walls, the perineum and the gluteal region. Right common iliac artery The right common iliac artery is approximately 5 cm long and passes obliquely across part of the fourth and the fifth lumbar vertebral bodies. The sympathetic rami to the pelvic plexus and, at its division, the ureter, cross anterior to it. It is covered by the parietal peritoneum, which separates it from the coils of the small intestine. Posteriorly, it is separated from the fourth and fifth lumbar vertebral bodies and their intervening disc by the right sympathetic trunk, the terminal parts of the common iliac veins and the start of the inferior vena cava, the obturator nerve, lumbosacral trunk and iliolumbar artery. Lateral to its upper part are the inferior vena cava and the right common iliac vein and lower down is the right psoas major. The left common iliac vein is medial to the upper part. Left common iliac artery The left common iliac artery is shorter than the right and is approximately 4 cm long. Lying anterior to it are the sympathetic rami to the pelvic plexus, the superior rectal artery and, at its terminal bifurcation, the ureter. The sympathetic trunk, the fourth and fifth lumbar vertebral bodies and intervening disc, the obturator nerve, lumbosacral trunk and iliolumbar artery are all posterior to it. The left common iliac vein is posteromedial to the artery while the left psoas major lies lateral to it. Branches In addition to the external iliac and internal iliac terminal branches, each common iliac artery gives small branches to the peritoneum, psoas major, ureter, adjacent nerves and surrounding areolar tissue. Occasionally the common iliac artery gives rise to the iliolumbar artery and accessory renal arteries if the kidney is low lying.
INTERNAL ILIAC ARTERIES (Fig. 108.4)
page 1360 page 1361
Figure 108.4 Arteries of the male pelvis. The internal iliac vein and its tributaries and the rectum have been omitted for clarity.
Each internal iliac artery, c.4 cm long, begins at the common iliac bifurcation, level
with the lumbosacral intervertebral disc and anterior to the sacroiliac joint. It descends posteriorly to the superior margin of the greater sciatic foramen where it divides into an anterior trunk, which continues in the same line towards the ischial spine, and a posterior trunk, which passes back to the greater sciatic foramen. Anterior to the artery are the ureter and, in females, the ovary and fimbriated end of the uterine tube. The internal iliac vein, lumbosacral trunk and sacroiliac joint are posterior. Lateral is the external iliac vein, between the artery and psoas major and inferior to this is the obturator nerve. The parietal peritoneum is medial, separating it from the terminal ileum on the right and the sigmoid colon on the left. Tributaries of the internal iliac vein are also medial. In the fetus the internal iliac artery is twice the size of the external and is the direct continuation of the common iliac artery. The main trunk ascends on the anterior abdominal wall to the umbilicus, converging on the contralateral artery. The two arteries run through the umbilicus to enter the umbilical cord as the umbilical arteries. At birth, when placental circulation ceases, only the pelvic segment remains patent as the internal iliac artery and part of the superior vesical artery; the remainder becomes a fibrous medial umbilical ligament. In males, the patent part usually gives off an artery to the vas deferens. Posterior trunk branches
Iliolumbar artery The iliolumbar artery is the first branch of the posterior trunk and ascends laterally anterior to the sacroiliac joint and lumbosacral nerve trunk. It lies posterior to the obturator nerve and external iliac vessels and reaches the medial border of psoas major, dividing behind it into the lumbar and iliac branches. The lumbar branch supplies psoas major and quadratus lumborum and anastomoses with the fourth lumbar artery. It sends a small spinal branch through the intervertebral foramen between the fifth lumbar and first sacral vertebrae, to supply the cauda equina. The iliac branch supplies iliacus; between the muscle and bone it anastomoses with the iliac branches of the obturator artery. A large nutrient branch enters an oblique canal in the ilium. Other branches runs around the iliac crest, contribute to the supply of the gluteal and abdominal muscles, and anastomose with the superior gluteal, circumflex iliac and lateral circumflex femoral arteries. Lateral sacral arteries The lateral sacral arteries are usually double or if single divide rapidly into superior and inferior branches. The superior and larger artery passes medially into the first or second anterior sacral foramen, supplies the sacral vertebrae and contents of the sacral canal and leaves the sacrum via the corresponding dorsal foramen to supply the skin and muscles dorsal to the sacrum. The inferior or lateral sacral artery crosses obliquely anterior to piriformis and the sacral anterior spinal rami, then descends lateral to the sympathetic trunk to anastomose with its fellow and the median sacral artery anterior to the coccyx. Its branches enter the anterior sacral foramina and are distributed like those of the superior artery. Superior gluteal artery The superior gluteal artery is the largest branch of the internal iliac and effectively forms the main continuation of its posterior trunk. It runs posteriorly between the lumbosacral trunk and the first sacral ramus or between the first and second rami, then turns slightly inferiorly leaving the pelvis by the greater sciatic foramen above piriformis and dividing into superficial and deep branches. In the pelvis it supplies piriformis, obturator internus and a nutrient artery to the ilium. The superficial branch enters the deep surface of gluteus maximus. Its numerous branches supply the muscle and anastomose with the inferior gluteal branches while others perforate the tendinous medial attachment of the muscle to supply the skin over the sacrum where they anastomose with the posterior branches of the lateral sacral arteries. The deep branch of the superior gluteal artery passes between gluteus medius and the bone, soon dividing into superior and inferior branches. The superior branch skirts the superior border of gluteus minimus to the anterior superior iliac spine and anastomoses with the deep circumflex iliac artery and the ascending branch of the lateral circumflex femoral artery. The inferior branch runs through gluteus minimus obliquely, supplies it and gluteus medius and anastomoses with the lateral circumflex femoral artery. A branch enters the trochanteric fossa to join the inferior gluteal artery and ascending branch of the medial circumflex femoral artery while other branches run through gluteus minimus to supply the hip joint. The superior gluteal artery occasionally arises directly from the internal iliac artery with the inferior gluteal artery and sometimes from the internal pudendal artery. Anterior trunk branches
Superior vesical artery (See also p. 1361.) The superior vesical artery is the first large branch of the anterior trunk. It lies on the lateral wall of the pelvis just below the brim and runs anteroinferiorly medial to the periosteum of the posterior surface of the pubis. It supplies the distal end of the ureter, the bladder, the proximal end of the vas deferens and the seminal vesicles. It also gives origin to the umbilical artery in the foetus, which remains as a fibrous cord, the medial umbilical ligament, in the adult. This vessel occasionally remains patent as a small artery supplying the umbilicus. Inferior vesical artery (See also p. 1361.) The inferior vesical artery may arise as a common branch with the middle rectal artery. In the female it is often replaced by the vaginal artery. It supplies the bladder, the prostate, the seminal vesicles and the vas deferens. Middle rectal artery (See also p. 1361.) The middle rectal artery is often multiple and may be small. It runs into the lateral fascial coverings of the mesorectum. It occasionally arises close to or in common with the origin of the inferior vesical artery in males. Vaginal artery (See also p. 1361.) In females the vaginal artery may replace the inferior vesical artery. It may arise from the uterine artery close to its origin. Obturator artery The obturator artery runs anteroinferiorly from the anterior trunk on the lateral pelvic wall to the upper part of the obturator foramen. It leaves the pelvis via the obturator canal and divides into anterior and posterior branches. In the pelvis it is related laterally to the fascia over obturator internus and is crossed on its medial aspect by the ureter and, in the male, by the vas deferens. In the nulliparous female the ovary lies medial to it. The obturator nerve is above the artery, the obturator vein below it. In the pelvis the obturator artery provides iliac branches to the iliac fossa. These supply the bone and iliacus and anastomose with the iliolumbar artery. A vesical branch runs medially to the bladder and sometimes replaces the inferior vesical branch of the internal iliac artery. A pubic branch usually arises just before the obturator artery leaves the pelvis, and ascends over the pubis to anastomose with the contralateral artery and the pubic branch of the inferior epigastric artery. Outside the pelvis the anterior and posterior terminal branches encircle the foramen between obturator externus and the obturator membrane. The anterior branch curves anteriorly on the membrane and then inferiorly along its anterior margin to supply branches to obturator externus, pectineus, the femoral adductors and gracilis. It anastomoses with the posterior branch and the medial circumflex femoral artery. The posterior branch follows the posterior margin of the foramen and turns anteriorly on the ischial part to anastomose with the anterior branch. It supplies the muscles attached to the ischial tuberosity and anastomoses with the inferior gluteal artery. An acetabular branch enters the hip joint at the acetabular notch, ramifies in the fat of the acetabular fossa and sends a branch along the ligament of the femoral head. Occasionally the obturator artery is replaced by an enlarged pubic branch of the inferior epigastric artery (p. 1101) which descends almost vertically to the obturator foramen. It usually lies near the external iliac vein, lateral to the femoral ring, and is rarely injured during inguinal or femoral hernia surgery. Sometimes it curves along the edge of the lacunar part of the inguinal ligament, partly encircling the neck of a hernial sac, and may be inadvertently cut during enlargement of the femoral ring in reducing a femoral hernia. Uterine artery (See also p. 1361.) The uterine artery is an additional branch in females. It is a large branch which arises below the obturator artery on the lateral wall of the pelvis and runs inferomedially into the broad ligament of the uterus. Internal pudendal artery (in the pelvis) (See also p. 1361.)
page 1361 page 1362
The internal pudendal artery arises just below the origin of the obturator artery. It descends laterally to the inferior rim of the greater sciatic foramen, where it leaves the pelvis between piriformis and ischiococcygeus to enter the gluteal region. It then curves around the dorsum of the ischial spine to enter the perineum by the lesser sciatic foramen. This course effectively allows the nerve to wrap around the posterior limit of levator ani at its attachment to the ischial spine and so gain access to the perineum. In the pelvis the internal pudendal artery crosses anterior
to piriformis, the sacral plexus and the inferior gluteal artery. Behind the ischial spine it is covered by gluteus maximus, with the pudendal nerve medial and the nerve to obturator internus lateral to it. Several muscular branches leave the pudendal artery in the pelvis and gluteal region to supply the adjacent muscles and nerves. Inferior gluteal artery The inferior gluteal artery is the larger terminal branch of the anterior internal iliac trunk and principally supplies the buttock and thigh. It descends posteriorly, anterior to the sacral plexus and piriformis but posterior to the internal pudendal artery. It passes between the first and second or second and third sacral anterior spinal nerve rami, then between piriformis and ischiococcygeus. It runs through the lower part of the greater sciatic foramen to reach the gluteal region. The artery runs inferiorly between the greater trochanter and ischial tuberosity with the sciatic and posterior femoral cutaneous nerves deep to gluteus maximus. It continues down the thigh, supplying the skin and anastomosing with branches of the perforating arteries. The inferior gluteal and internal pudendal arteries often arise as a common stem from the internal iliac, sometimes with the superior gluteal artery. Inside the pelvis the inferior gluteal artery gives branches to piriformis, ischiococcygeus and iliococcygeus. Occasionally it contributes to the middle rectal arterial supply and, in the male, supplies vessels to the seminal vesicles and prostate.
EXTERNAL ILIAC ARTERIES The external iliac arteries are of larger calibre than the internal iliac artery. Each artery descends laterally along the medial border of psoas major from the common iliac bifurcation to a point midway between the anterior superior iliac spine and the symphysis pubis. It enters the thigh posterior to the inguinal ligament to become the femoral artery. On the right the artery is separated from the terminal ileum and, usually, the appendix by the parietal peritoneum and extraperitoneal tissue. On the left the artery is separated from the sigmoid colon and coils of the small intestine lie anteromedially. At its origin the artery may be crossed by the ureter. It is also crossed by the gonadal vessels, the genital branch of the genitofemoral nerve, the deep circumflex iliac vein and the vas deferens (male) or round ligament (female). Posterior to the artery the iliac fascia separates it from the medial border of psoas major. The external iliac vein lies partly posterior to its upper part but is more medial to it below. Laterally, it is related to psoas major which is covered by the iliac and psoas fascia. Numerous lymph vessels and nodes lie on its front and sides. The external iliac artery is principally the artery of the lower limb and as such has few branches in the pelvis. Apart from very small vessels to psoas major and neighbouring lymph nodes, the artery has no branches until it gives off the inferior epigastric and deep circumflex iliac arteries which arise near to its passage under the inguinal ligament. Deep circumflex iliac artery The deep circumflex iliac artery branches laterally from the external iliac artery almost opposite the origin of the inferior epigastric artery. It ascends and runs laterally to the anterior superior iliac spine behind the inguinal ligament in a sheath formed by the union of the transversalis and iliac fasciae. There it anastomoses with the ascending branch of the lateral circumflex femoral artery, pierces the transversalis fascia and skirts the internal lip of the iliac crest. About halfway along the iliac crest it runs through transversus abdominis and then between transversus and internal oblique to anastomose with the iliolumbar and superior gluteal arteries. At the anterior superior iliac spine it gives off a large ascending branch, which runs between internal oblique and transversus abdominis. It supplies both muscles and anastomoses with the lumbar and inferior epigastric arteries. Inferior epigastric artery (See also p. 1362.) The inferior epigastric artery originates from the external iliac artery posterior to the inguinal ligament. It curves forwards in the anterior extraperitoneal tissue and ascends obliquely along the medial margin of the deep inguinal ring where it continues as an artery of the anterior abdominal wall.
VEINS OF THE PELVIS The true pelvis contains a large number of veins which drain the wall and most of the viscera contained within the pelvis and carry venous blood from the gluteal
region, thigh the hip. The external iliac veins, lying close to the brim of the pelvis, carry the venous drainage of most of the lower limb. There is considerable variation in the venous drainage of the pelvis and although the major veins frequently follow their named arterial counterparts, the small tributaries exhibit a great deal of variation between individuals.
COMMON ILIAC VEINS (Fig. 108.5) The common iliac vein is formed by the union of external and internal iliac veins, anterior to the sacroiliac joints. It ascends obliquely to end at the right side of the fifth lumbar vertebra, uniting at an acute angle with the contralateral vessel to form the inferior vena cava. The right common iliac vein is shorter and more nearly vertical, lying posterior then lateral to its artery. The right obturator nerve passes posterior. The left common iliac vein is longer and more oblique and lies first medial, then posterior to its artery. It is crossed anteriorly by the attachment of the sigmoid mesocolon and superior rectal vessels. Each vein receives iliolumbar and sometimes lateral sacral veins. The left common iliac usually drains the median sacral vein. There are no valves in these veins. The left common iliac vein occasionally ascends to the left of the aorta to the level of the kidney where it receives the left renal vein and crosses anterior to the aorta to join the inferior vena cava. This vessel represents the persistent caudal half of the left postcardinal or supracardinal vein. Median sacral veins The medial sacral veins accompany the corresponding artery anterior to the sacrum, and unite to form a single vein which usually ends in the left common iliac vein. Sometimes it ends at the common iliac junction. Internal pudendal veins The internal pudendal veins are venae comitantes of the internal pudendal artery. They unite as a single vessel ending in the internal iliac vein. They receive veins from the penile bulb and the scrotum (males) or clitoris and labia (females) and the inferior rectal veins.
page 1362 page 1363
Figure 108.5 Venogram showing the veins of the pelvis and groin. Contrast medium has been injected into the bodies of the pubic bones. (Provided by M Lea Thomas.)
INTERNAL ILIAC VEIN
The internal iliac vein is formed by the convergence of several veins above the greater sciatic foramen. It does not have the predictable trunks and branches of the internal iliac artery but its branches drain the same territories. It ascends posteromedial to the internal iliac artery to join the external iliac vein, forming the common iliac vein at the pelvic brim, anterior to the lower part of the sacroiliac joint. It is covered anteromedially by parietal peritoneum. Its tributaries are the gluteal, internal pudendal and obturator veins, which originate outside the pelvis; the lateral sacral veins which run from the anterior surface of the sacrum; and the middle rectal, vesical, uterine and vaginal veins which originate in the venous plexuses of the pelvic viscera. The venous drainage of the leg may be blocked by thrombosis involving the external iliac systems and the inferior vena cava. Under these circumstances, the pelvic veins, particularly the internal iliac tributaries, enlarge and provide a major avenue of venous return from the femoral system. Surgical interference with these veins may seriously compromise venous drainage and precipitate oedema of one or both legs. Superior gluteal veins The superior gluteal veins are the venae comitantes of the superior gluteal artery. They receive branches corresponding to branches of the artery and enter the pelvis via the greater sciatic foramen, above piriormis. They join the internal iliac vein, frequently as a single trunk. Inferior gluteal veins The inferior gluteal veins are venae comitantes of the inferior gluteal artery. They begin proximally and posterior in the thigh, where they anastomose with the medial circumflex femoral and first perforating veins. They enter the pelvis low in the greater sciatic foramen, joining to form a vessel opening into the distal (lower) part of the internal iliac vein. They connect with the superficial gluteal veins by perforating veins (Doyle 1970) analogous to the sural perforating veins. They probably have a venous 'pumping' role, and provide collaterals between the femoral and internal iliac veins. Obturator vein The obturator vein begins in the proximal adductor region and enters the pelvis via the obturator foramen. It runs posteriorly and superiorly on the lateral pelvic wall below the obturator artery and between the ureter and internal iliac artery to end in the internal iliac vein. It is sometimes replaced by an enlarged pubic vein, which joins the external iliac vein. Lateral sacral veins The lateral sacral veins accompany the lateral sacral arteries, and are interconnected by a sacral venous plexus. Middle rectal vein The middle rectal vein begins in the rectal venous plexus and drains the rectum and mesorectum. It often receives tributaries from the bladder and the prostate and seminal vesicle (males) and the posterior aspect of the vagina (females). It is variable in size and runs laterally on the pelvic surface of levator ani to end in the internal iliac vein.
EXTERNAL ILIAC VEIN The external iliac vein is the proximal continuation of the femoral vein. It begins posterior to the inguinal ligament, ascends along the pelvic brim and ends anterior to the sacroiliac joint by joining the internal iliac vein to form the common iliac vein. On the right it lies medial to the external iliac artery, gradually inclining behind it as it ascends. On the left it is wholly medial. Disease of the external iliac artery may cause it to adhere closely to the vein at the point where it is in contact, and, particularly on the right side, the walls of the vessels may become fused, making dissection hazardous. Medially the external iliac vein is crossed by the ureter and internal iliac artery. In males it is crossed by the vas deferens, in females by the round ligament and ovarian vessels. Lateral to it lies psoas major, except where the artery intervenes. The vein is usually valveless, but may contain a single valve. It tributaries are the inferior epigastric, deep circumflex iliac and pubic veins. Inferior epigastric vein One or two inferior epigastric veins accompany the artery and drain into the external iliac vein a little above the inguinal ligament. Deep circumflex iliac vein
The deep circumflex vein is formed from venae comitantes of the corresponding artery. It joins the external iliac vein a little above the inferior epigastric veins after crossing anterior to the external iliac artery. Pubic vein The pubic vein connects the external iliac and the obturator vein. It ascends on the pelvic surface of the pubis with the pubic branch of the inferior epigastric artery. It sometimes replaces the normal obturator vein.
LYMPHATIC DRAINAGE OF THE PELVIS COMMON ILIAC NODES (Fig. 108.6) The common iliac nodes are grouped around the artery, and one or two lie inferior to the aortic bifurcation and anterior to the fifth lumbar vertebra or sacral promontory. They drain the external and internal iliac nodes and connect to the lateral aortic nodes. They usually lie in medial, lateral and anterior chains around the artery, the lateral being the main route. Since they receive drainage from both internal and external iliac nodes, the common iliac nodes receive the entire lymphatic drainage of the lower limb.
EXTERNAL ILIAC NODES The external iliac nodes usually form three subgroups, lateral, medial and anterior to the external iliac vessels. The medial nodes are considered the main channel of drainage, collecting lymph from the lower limb via the inguinal nodes, the deeper layers of the infra-umbilical abdominal wall, the adductor region of the thigh, the glans penis or clitoris, the membranous urethra, prostate, fundus of the bladder, uterine cervix and upper vagina. Their efferents pass to the common iliac nodes. Inferior epigastric and circumflex iliac nodes The inferior epigastric and circumflex iliac nodes are associated with their vessels and drain the corresponding areas to the external iliac nodes.
INTERNAL ILIAC NODES (Fig. 108.7)
page 1363 page 1364
Figure 108.6 Lymphatic drainage of the male pelvis and urinary bladder.
Figure 108.7 Lymphatic drainage of the female pelvis. (After Cuneo and Marcille.)
The internal iliac nodes surround the branches of the internal iliac vessels and receive afferents from most of the pelvic viscera (with the exception of the gonads and the rectum), deeper parts of the perineum and the gluteal and posterior femoral muscles. They drain to the common iliac nodes. The individual groups are considered in the description of the viscera. There are frequent connections between the right and left groups particularly when they lie close to the anterior and posterior midlines.
Innervation of the pelvis The pelvis contains the lumbosacral nerve trunk, the sacral plexus, the coccygeal plexus and the pelvic parts of the sympathetic and parasympathetic systems. These supply the somatic and autonomic innervation to the majority of the pelvic visceral organs, the pelvic floor and perineum, the gluteal region and the lower limb. The ventral rami of the sacral and coccygeal spinal nerves form the sacral and coccygeal plexuses. The upper four sacral ventral rami enter the pelvis by the anterior sacral foramina, the fifth between the sacrum and coccyx, while that of the coccygeal nerve curves forwards below the rudimentary transverse process of the first coccygeal segment. The first and second sacral ventral rami are large, the third to fifth diminish progressively and the coccygeal is the smallest. Each receives a grey ramus communicans from a corresponding sympathetic ganglion. Visceral efferent rami leave the second to fourth sacral rami as pelvic splanchnic nerves, containing parasympathetic fibres which reach minute ganglia in the walls of the pelvic viscera.
LUMBOSACRAL TRUNK AND SACRAL PLEXUS (Fig. 111.41) The sacral plexus is formed by the lumbosacral trunk, the first to third sacral ventral rami and part of the fourth, the remainder of the last joining the coccygeal plexus. The lumbar part of the lumbosacral trunk contains part of the fourth and all the fifth lumbar ventral rami; it appears at the medial margin of psoas major, and descends over the pelvic brim anterior to the sacroiliac joint to join the first sacral ramus. The greater part of the second and third sacral rami converge on the inferomedial aspect of the lumbosacral trunk in the greater sciatic foramen to form the sciatic nerve. The ventral and dorsal divisions of the nerves do not separate physically from each other but the fibres remain separate within the rami, and ventral and dorsal divisions of each contributing root join within the sciatic nerve.
The fibres of the dorsal divisions will go on to form the common peroneal nerve and the ventral division fibres form the tibial nerve. The sciatic nerve occasionally divides into common peroneal and tibial nerves inside the pelvis. In these cases the common peroneal nerve usually runs through piriformis. The sacral plexus lies against the posterior pelvic wall anterior to piriformis, posterior to the internal iliac vessels and ureter, and behind the sigmoid colon on the left. The superior gluteal vessels run between the lumbosacral trunk and first sacral ventral ramus or between the first and second sacral rami, while the inferior gluteal vessels lie between the first and second or second and third sacral rami (Fig. 108.8). The sacral plexus is not commonly involved in malignant tumours of the pelvis because in lies behind the relatively dense presacral fascia which resists all but locally very advanced malignant infiltration. When it occurs, there is intractable pain in the distribution of the branches of the plexus which may be very difficult to treat. The plexus may also be involved in the reticuloses or be affected by plexiform neuromas. UPDATE Date Added: 07 March 2006 Publication Services, Inc. Abstract: Urinary and sexual function after total mesorectal excision. Recent results Click on the following line to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=15865034&query_hl=6&itool=pubmed_docsum Urinary and sexual function after total mesorectal excision. Recent results. Maurer CA: Cancer Res 165:196-204, 2005.
BRANCHES OF THE SACRAL PLEXUS Ventral divisions Nerve to quadratus femoris and gemellus L4,5, S1 inferior Nerve to obturator internus and gemellus L5, S1,2 superior Nerve to piriformis Superior gluteal nerve Inferior gluteal nerve Posterior femoral cutaneous nerve S2,3 Tibial (sciatic) nerve L4,5, S1,2,3 Common peroneal (sciatic) nerve Perforating cutaneous nerve Pudendal nerve S2,3,4 Nerves to levator ani and external anal S4 sphincter Pelvic splanchnic nerves
Dorsal divisions
S2 (S1) L4,5, S1 L5, S1,2 S1,2 L4,5, S1,2 S2,3
S2,3 (S4)
The branches of the sacral plexus are: The course and distribution of most of the branches of the sacral plexus are covered fully on page 1456.
PUDENDAL NERVE (IN THE PELVIS) The pudendal nerve arises from the ventral divisions of the second, third and fourth sacral ventral rami and is formed just above the superior border of the sacrotuberous ligament and the upper fibres of ischiococcygeus. It leaves the pelvis via the greater sciatic foramen between piriformis and ischiococcygeus, enters the gluteal region and crosses the sacrospinous ligament close to its attachment to the ischial spine. The nerve lies medial to the internal pudendal vessels on the spine. It accompanies the internal pudendal artery through the lesser sciatic foramen into the pudendal (Alcock's) canal on the lateral wall of the ischioanal fossa. In the posterior part of the canal it gives rise to the inferior rectal nerve, the perineal nerve and the dorsal nerve of the penis or clitoris.
SACRAL VISCERAL BRANCHES These arise from the second to fourth sacral ventral rami to innervate the pelvic viscera; they are termed pelvic splanchnic nerves.
SACRAL MUSCULAR BRANCHES page 1364 page 1365
Figure 108.8 The lumbosacral plexus in the pelvis. The pelvic viscera have been omitted for clarity.
Several muscular branches arise from the fourth sacral ventral ramus to supply the superior surface of levator ani and the upper part of the external anal sphincter. The branches to levator ani enter the superior (pelvic) surface of the muscle whilst the branch to the external anal sphincter (also referred to as the perineal branch of the fourth sacral nerve) reaches the ischioanal fossa by running through ischiococcygeus or between ischiococcygeus and iliococcygeus. It supplies the skin between the anus and coccyx via its cutaneous branches.
COCCYGEAL PLEXUS The coccygeal plexus is formed by a small descending branch from the fourth sacral ramus and by the fifth sacral and coccygeal ventral rami. The fifth sacral ventral ramus emerges from the sacral hiatus, curves round the lateral margin of the sacrum below its cornu and pierces ischiococcygeus from below to reach its upper, pelvic surface. Here it is joined by a descending branch of the fourth sacral ventral ramus, and the small trunk so formed descends on the pelvic surface of ischiococcygeus. They join the minute coccygeal ventral ramus which emerges from the sacral hiatus and curves round the lateral coccygeal margin to pierce coccygeus to reach the pelvis. This small trunk is the coccygeal plexus. Anococcygeal nerves arise from it and form a few fine filaments which pierce the sacrotuberous ligament to supply the adjacent skin.
PELVIC PART OF THE SYMPATHETIC SYSTEM The pelvic sympathetic trunk lies in the extraperitoneal tissue anterior to the sacrum beneath the presacral fascia. It lies medial or anterior to the anterior sacral foramina and has four or five interconnected ganglia. Above, it is continuous with the lumbar sympathetic trunk. Below the lowest ganglia the two trunks converge to unite in the small ganglion impar anterior to the coccyx. Grey rami communicantes pass from the ganglia to sacral and coccygeal spinal nerves but there are no white rami communicantes. Medial branches connect across the midline and twigs from the first two ganglia join the inferior hypogastric plexus or the hypogastric 'nerve'. Other branches form a plexus on the median sacral artery.
VASCULAR BRANCHES Postganglionic fibres pass through the grey rami communicantes to the roots of the sacral plexus. Those forming the tibial nerve are conveyed to the popliteal artery and its branches in the leg and foot whilst those in the pudendal and
superior and inferior gluteal nerves accompany the same named arteries to the gluteal and perineal tissues. Branches may also supply the pelvic lymph nodes. Preganglionic fibres for the rest of the lower limb are derived from the lower three thoracic and upper two or three lumbar spinal segments. They reach the lower thoracic and upper lumbar ganglia through white rami communicantes and descend through the sympathetic trunk to synapse in the lumbar ganglia. Postganglionic fibres pass from these ganglia via grey rami communicantes to the femoral nerve which carries them to the distribution of the femoral artery and its branches. Some fibres descend through the lumbar ganglia to synapse in the upper two or three sacral ganglia, from which postganglionic axons join the tibial nerve to supply the popliteal artery and its branches in the leg and foot. Sympathetic denervation of vessels in the lower limb can be effected by removing or ablating the upper three lumbar ganglia and the intervening parts of the sympathetic trunk, which is rarely useful in treating vascular insufficiency of the lower limb.
© 2008 Elsevier
TRUE PELVIS AND PELVIC FLOOR The true pelvis is a bowl-shaped structure formed from the sacrum, pubis, ilium, ischium, the ligaments which interconnect these bones and the muscles which line their inner surfaces. The true pelvis is considered to start at the level of the plane passing through the promontory of the sacrum, the arcuate line on the ilium, the iliopectineal line and the posterior surface of the pubic crest. This plane, or 'inlet' lies at an angle of between 35 and 50° up from the horizontal and above this the bony structures are sometimes referred to as the false pelvis. They form part of the walls of the lower abdomen. The floor or 'outlet' of the true pelvis is formed by the muscles of levator ani. Although the floor is gutter shaped, it generally lies in a plane between 5 and 15° up from the horizontal. This difference between the planes of the inlet and outlet is the reason why the true pelvis is said to have an axis (lying perpendicular to the plane of both inlet and outlet) which progressively changes through the pelvis from above downwards. The details of the topography of the bony and ligamentous pelvis is considered fully on page 1428.
Muscles and fasciae of the pelvis PELVIC MUSCLES (Figs 108.1, 108.2)
Figure 108.1 Muscles of the male pelvis - lateral view. The superior gluteal and obturator vessels and nerves have been divided close to their exit from the pelvis. The rectum, bladder and upper prostate have been omitted for clarity.
The muscles arising within the pelvis form two groups. Piriformis and obturator internus, although forming part of the walls of the pelvis, are considered as primarily muscles of the lower limb. Levator ani and coccygeus form the pelvic diaphragm and delineate the lower limit of the true pelvis. The fasciae investing the muscles are continuous with visceral pelvic fascia above, perineal fascia below and obturator fascia laterally.
PIRIFORMIS (See also p. 1357.) Piriformis forms part of the posterolateral wall of the true pelvis. It is attached to the anterior surface of the sacrum, the gluteal surface of the ilium near the posterior inferior iliac spine, the capsule of the adjacent sacroiliac joint and sometimes to the upper part of the pelvic surface of the sacrotuberous ligament. It
passes out of the pelvis through the greater sciatic foramen. Within the pelvis, the anterior surface of piriformis is related to the rectum (especially on the left), the sacral plexus of nerves and branches of the internal iliac vessels. The posterior surface lies against the sacrum.
OBTURATOR INTERNUS (See also p. 1357.) page 1357 page 1358
Figure 108.2 Muscles of the female pelvis viewed from above. The sacral nerve roots have been divided close to the sacral foramina. The anorectal junction, vagina and urethra have been divided at the level of the pelvic floor.
Obturator internus and the fascia over its upper inner (pelvic) surface form part of the anterolateral wall of the true pelvis. It is attached to the structures surrounding the obturator foramen; the inferior ramus of the pubis, the ischial ramus, the pelvic surface of the hip bone below and behind the pelvic brim, and the upper part of the greater sciatic foramen. It also attaches to the medial part of the pelvic surface of the obturator membrane. The muscle is covered by a thick fascial layer and the fibres themselves cannot be seen directly from within the pelvis. This fascia gives attachment to some of the fibres of levator ani and thus only the upper portion of the muscle lies lateral to the contents of the true pelvis, whilst the lower portion forms part of the boundaries of the ischioanal fossa. In the male, the upper portion lies lateral to the bladder, the obturator and vesical vessels, and the obturator nerve. In the female, the attachments of the broad ligament of the uterus, the fallopian end of the uterine tubes, and the uterine vessels, also lie medial to obturator internus and its fascia.
LEVATOR ANI (ISCHIOCOCCYGEUS, ILIOCOCCYGEUS, PUBOCOCCYGEUS) Levator ani is a broad muscular sheet of variable thickness attached to the internal surface of the true pelvis and forms a large portion of the pelvic floor. The muscle is subdivided into named portions according to their attachments and the pelvic viscera to which they are related. These parts are often referred to as separate muscles, but the boundaries between each part cannot be easily distinguished and they perform many similar physiological functions. The separate parts are referred to as ischiococcygeus, iliococcygeus and pubococcygeus. Pubococcygeus is often subdivided into separate parts according to the pelvic viscera to which they relate, i.e. pubourethralis and puborectalis in the male, pubovaginalis and puborectalis in the female. Levator ani arises from each side of the walls of the pelvis. Fibres from ischiococcygeus attach to the sacrum and coccyx but the remaining parts of the muscle converge in the midline. The fibres
of iliococcygeus join by a partly fibrous intersection and form a raphe posterior to the anorectal junction. Closer to the anorectal junction and elsewhere in the pelvic floor, the fibres are more nearly continuous with those of the opposite side and the muscle forms a sling (puborectalis and pubovaginalis or pubourethralis). Attachments
Ischiococcygeus The ischiococcygeal part may be referred to as a separate muscle, sometimes named coccygeus. It lies as the most posterosuperior portion of levator ani and arises as a triangular musculotendinous sheet with its apex attached to the pelvic surface and tip of the ischial spine. The base of the muscle is attached to the lateral margins of the coccyx and the fifth sacral segment. Ischiococcygeus is rarely absent, but may be nearly completely tendinous rather than muscular. It lies on the pelvic aspect of the sacrospinous ligament and may be fused with it, particularly if mostly tendinous. The sacrospinous ligament may represent a degenerate part or an aponeurosis of the muscle since the muscle and ligament are coextensive. Iliococcygeus The iliococcygeal part is attached to the inner surface of the ischial spine below and anterior to the attachment of ischiococcygeus and to the obturator fascia as far forward as the obturator canal (Fig. 108.1). The most posterior fibres are attached to the tip of the sacrum and coccyx but most join with fibres from the opposite side to form a raphe. This raphe is effectively continuous with the fibroelastic anococcygeal ligament, which is closely applied to its inferior surface and some muscle fibres may attach into the ligament. The raphe provides a strong attachment for the pelvic floor posteriorly and must be divided to allow wide excisions of the anorectal canal during abdominoperineal excisions for malignancy. An accessory slip may arise from the most posterior part and is sometimes referred to as iliosacralis. Pubococcygeus
page 1358 page 1359
Figure 108.3 Fasciae of the pelvis and perineum. Median sagittal section in the male. The deep fascia of the abdominal wall, the layers of the urogenital fascia and the mesorectal fascia are in green, the peritoneum in blue, the superficial fascia of the abdominal wall and perineum in red. Muscles are shown in brown.
The pubococcygeal part is attached to the back of the body of the pubis and passes back almost horizontally. The most medial fibres run directly lateral to the urethra and its sphincter as it passes through the pelvic floor. In males these
fibres therefore lie lateral and inferior to the prostate and are referred to as pubourethralis. They form part of the urethral sphincter complex together with the intrinsic striated and smooth musculature of the urethra and fibres decussate across the midline directly behind the urethra. In females the fibres of this part of the muscle run further back to from a sling around the posterior wall of the vagina and are referred to as pubovaginalis. In both sexes fibres from this part of pubococcygeus attach to the perineal body and a few elements also attach to the anorectal junction. Some of these fibres, sometimes called puboanalis, decussate and blend with the longitudinal rectal muscle and fascial elements to contribute to the conjoint longitudinal coat of the anal canal. Behind the rectum some fibres of pubococcygeus form a tendinous intersection as part of the levator raphe but a thick muscular sling, puborectalis, wraps around the anorectal junction. Some fibres blend with those of the external anal sphincter. Relations
The superior, pelvic surface of levator ani is separated only by fascia (superior pelvic diaphragmatic, visceral and extraperitoneal (p. 1359) from the urinary bladder, prostate or uterus and vagina, rectum and peritoneum. Its inferior, perineal, surface forms the medial wall of the ischioanal fossa and the superior wall of the anterior recess of the fossa, both being covered by inferior pelvic diaphragmatic fascia. The posterior border is separated from the coccyx by areolar tissue. The medial borders of the two levator muscles are separated by the visceral outlet, through which pass the urethra, vagina, and anorectum. Vascular supply
Levator ani is supplied by branches of the inferior gluteal artery, the inferior vesical artery and the pudendal artery. Innervation
Fibres originating mainly in the second, third and fourth sacral spinal segments reach levator ani from below and above by a variety of routes (Wendell-Smith & Wilson 1991). Most commonly pubococcygeus (puborectalis and pubovaginalis or pubourethralis) is supplied by second and third sacral spinal segments via the pudendal nerve, and ischiococcygeus and iliococcygeus by direct branches of the sacral plexus from third and fourth sacral spinal segments. Actions
Pubococcygeus is a lateral compressor of the various visceral canals which cross the pelvic floor. The puborectalis part also reinforces the external anal sphincter and helps to create the anorectal angle. It also reduces the anteroposterior dimension of the ano-urogenital hiatus. Iliococcygeus and, to a lesser extent, the less muscular ischiococcygeus, assist puborectalis in contributing to anorectal and urinary continence. It is well recognized that levator ani must relax appropriately to permit expulsion of urine and particularly faeces. Levator ani also forms much of the basin-shaped muscular pelvic diaphragm, which supports the pelvic viscera and it contracts with abdominal muscles and the abdominothoracic diaphragm to raise intra-abdominal pressure. Like the abdominothoracic diaphragm, but unlike abdominal muscles, levator ani is also active in the inspiratory phase of quiet respiration. In the pregnant female, the shape of the pelvic floor may help to direct the fetal head into the anteroposterior diameter of the pelvic outlet.
PELVIC FASCIAE (Fig. 108.3) The pelvic fasciae may be conveniently divided into the parietal pelvic fascia, which mainly forms the coverings of the pelvic muscles, and the visceral pelvic fascia, which forms the coverings of the pelvic viscera, their supplying vessels and nerves. The visceral pelvic fascia is described with the pelvic viscera.
PARIETAL PELVIC FASCIA The parietal pelvic fascia on the pelvic surface of obturator internus is well differentiated as the obturator fascia. Above, it is connected to the posterior part of the arcuate line of the ilium, and is continuous with iliac fascia. Anterior to this, as it follows the line of origin of obturator internus, it is gradually separated from the attachment of the iliac fascia and a portion of the periosteum of the ilium and pubis spans between them. It arches below the obturator vessels and nerve, investing the obturator canal, and is attached anteriorly to the back of the pubis. Behind the obturator canal the fascia is markedly aponeurotic and gives a firm attachment to levator ani. Below the attachment of levator ani it is thin and forms part of the lateral wall of the ischioanal fossa in the perineum. It is continuous with the pelvic periosteum and thus the fascia over piriformis.
page 1359 page 1360
FASCIA OVER PIRIFORMIS The fascia of piriformis is very thin, and fuses with the periosteum on the front of the sacrum at the margins of the anterior sacral foramina. It ensheathes the sacral anterior primary rami which emerges from these foramina, and the nerves are often described as lying behind the fascia. The internal iliac vessels lie in front of the fascia over piriformis and their branches draw out sheaths of the fascia and extraperitoneal tissue into the gluteal region, above and below piriformis.
FASCIA OF THE PELVIC DIAPHRAGM The fascia of the pelvic diaphragm covers both of the surfaces of the pelvic diaphragm. On the lower surface is the thin inferior fascia of the pelvic diaphragm, which is continuous with the obturator fascia laterally. It covers the medial wall of the ischioanal fossa and blends below with fasciae on the urethral sphincter and the external anal sphincter. On the upper surface is the superior fascia of the pelvic diaphragm which is generally known clinically as the endopelvic fascia. It is attached anteriorly to the back of the body of the pubis, c.2 cm above its lower border, and extends laterally across the superior ramus of the pubis, blending with the obturator fascia and continuing along an irregular line to the spine of the ischium. It is continuous posteriorly with the fascia over piriformis and the anterior sacrococcygeal ligament. Medially, the superior fascia of the pelvic diaphragm blends with the visceral pelvic fascia. The fascia over obturator internus above the attachment of levator ani is therefore composed of the obturator fascia itself, the superior and inferior pelvic diaphragmatic fasciae and fibres from levator ani. The thickening where these structures fuse is the tendinous arch of levator ani. Below it, within the superior fascia, is the tendinous arch of the pelvic fascia, a thick white band extending from the lower part of the symphysis pubis to the inferior margin of the spine of the ischium (arcus tendineous fasciae pelvis). This is the attachment of the lateral, 'true' ligament of the urinary bladder. Anteriorly the same fascia forms two thick bands, the paired puboprostatic ligaments in the male, or the pubourethral ligaments in the female.
PRESACRAL FASCIA The presacral fascia lies between the posterior aspect of the mesorectal fascia and the superior pelvic diaphragmatic fascia. It is a hammock-like sheet extending between the tendinous arches of the pelvic fascia on either side. Below, it extends to the anorectal junction, where it fuses with the posterior aspect of the mesorectal fascia at the level of the anorectal junction. Above, it can be traced to the origin of the superior hypogastric plexus where it becomes progressively thinner over the promontory of the sacrum and becomes continuous with the retroperitoneal tissues. The right and left hypogastric nerves and inferior hypogastric plexuses lie on its surface and the presacral veins lie immediately posterior to it. It forms a distinct layer which can be seen both on magnetic resonance images of the pelvis and during surgery. The fascia provides an important landmark because extension of rectal tumours through it signifiantly reduces the chance of curative resectional surgery being possible. Dissection in the plane posterior to it may result in bleeding from the presacral veins and, since the adventitia of the veins is partly attached to the posterior surface of the fascia, the haemorrhage may be severe because the veins are unable to contract down properly. UPDATE Abstract Abstract: Clinical anatomy of the pelvic floor.
Date Added: 30 August 2005
Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=15152384&query_hl=10 Clinical anatomy of the pelvic floor. Fritsch H, Lienemann A, Brenner E et al: Adv Anat Embryol Cell Biol. 175:III-IX, 1-64, 2004.
Vascular supply and lymphatic drainage of the pelvis The true pelvis contains the internal iliac arteries and veins as well as the lymphatics draining the majority of the pelvic viscera. The common and external iliac vessels as well as the lymphatics draining the lower limb lie along the pelvic brim and in the lower retroperitoneum, but are conveniently discussed together with the vessels of the true pelvis.
ARTERIES OF THE PELVIS
COMMON ILIAC ARTERIES The abdominal aorta bifurcates into the right and left common iliac arteries anterolateral to the left side of the fourth lumbar vertebral body. These arteries diverge as they descend to divide at the level of the sacroiliac joint into external and internal iliac arteries. The external iliac artery is the principal artery of the lower limb and the internal iliac artery provides the principal supply to the pelvic viscera and walls, the perineum and the gluteal region. Right common iliac artery The right common iliac artery is approximately 5 cm long and passes obliquely across part of the fourth and the fifth lumbar vertebral bodies. The sympathetic rami to the pelvic plexus and, at its division, the ureter, cross anterior to it. It is covered by the parietal peritoneum, which separates it from the coils of the small intestine. Posteriorly, it is separated from the fourth and fifth lumbar vertebral bodies and their intervening disc by the right sympathetic trunk, the terminal parts of the common iliac veins and the start of the inferior vena cava, the obturator nerve, lumbosacral trunk and iliolumbar artery. Lateral to its upper part are the inferior vena cava and the right common iliac vein and lower down is the right psoas major. The left common iliac vein is medial to the upper part. Left common iliac artery The left common iliac artery is shorter than the right and is approximately 4 cm long. Lying anterior to it are the sympathetic rami to the pelvic plexus, the superior rectal artery and, at its terminal bifurcation, the ureter. The sympathetic trunk, the fourth and fifth lumbar vertebral bodies and intervening disc, the obturator nerve, lumbosacral trunk and iliolumbar artery are all posterior to it. The left common iliac vein is posteromedial to the artery while the left psoas major lies lateral to it. Branches In addition to the external iliac and internal iliac terminal branches, each common iliac artery gives small branches to the peritoneum, psoas major, ureter, adjacent nerves and surrounding areolar tissue. Occasionally the common iliac artery gives rise to the iliolumbar artery and accessory renal arteries if the kidney is low lying.
INTERNAL ILIAC ARTERIES (Fig. 108.4)
page 1360 page 1361
Figure 108.4 Arteries of the male pelvis. The internal iliac vein and its tributaries and the rectum have been omitted for clarity.
Each internal iliac artery, c.4 cm long, begins at the common iliac bifurcation, level with the lumbosacral intervertebral disc and anterior to the sacroiliac joint. It descends posteriorly to the superior margin of the greater sciatic foramen where it divides into an anterior trunk, which continues in the same line towards the ischial spine, and a posterior trunk, which passes back to the greater sciatic foramen. Anterior to the artery are the ureter and, in females, the ovary and fimbriated end
of the uterine tube. The internal iliac vein, lumbosacral trunk and sacroiliac joint are posterior. Lateral is the external iliac vein, between the artery and psoas major and inferior to this is the obturator nerve. The parietal peritoneum is medial, separating it from the terminal ileum on the right and the sigmoid colon on the left. Tributaries of the internal iliac vein are also medial. In the fetus the internal iliac artery is twice the size of the external and is the direct continuation of the common iliac artery. The main trunk ascends on the anterior abdominal wall to the umbilicus, converging on the contralateral artery. The two arteries run through the umbilicus to enter the umbilical cord as the umbilical arteries. At birth, when placental circulation ceases, only the pelvic segment remains patent as the internal iliac artery and part of the superior vesical artery; the remainder becomes a fibrous medial umbilical ligament. In males, the patent part usually gives off an artery to the vas deferens. Posterior trunk branches
Iliolumbar artery The iliolumbar artery is the first branch of the posterior trunk and ascends laterally anterior to the sacroiliac joint and lumbosacral nerve trunk. It lies posterior to the obturator nerve and external iliac vessels and reaches the medial border of psoas major, dividing behind it into the lumbar and iliac branches. The lumbar branch supplies psoas major and quadratus lumborum and anastomoses with the fourth lumbar artery. It sends a small spinal branch through the intervertebral foramen between the fifth lumbar and first sacral vertebrae, to supply the cauda equina. The iliac branch supplies iliacus; between the muscle and bone it anastomoses with the iliac branches of the obturator artery. A large nutrient branch enters an oblique canal in the ilium. Other branches runs around the iliac crest, contribute to the supply of the gluteal and abdominal muscles, and anastomose with the superior gluteal, circumflex iliac and lateral circumflex femoral arteries. Lateral sacral arteries The lateral sacral arteries are usually double or if single divide rapidly into superior and inferior branches. The superior and larger artery passes medially into the first or second anterior sacral foramen, supplies the sacral vertebrae and contents of the sacral canal and leaves the sacrum via the corresponding dorsal foramen to supply the skin and muscles dorsal to the sacrum. The inferior or lateral sacral artery crosses obliquely anterior to piriformis and the sacral anterior spinal rami, then descends lateral to the sympathetic trunk to anastomose with its fellow and the median sacral artery anterior to the coccyx. Its branches enter the anterior sacral foramina and are distributed like those of the superior artery. Superior gluteal artery The superior gluteal artery is the largest branch of the internal iliac and effectively forms the main continuation of its posterior trunk. It runs posteriorly between the lumbosacral trunk and the first sacral ramus or between the first and second rami, then turns slightly inferiorly leaving the pelvis by the greater sciatic foramen above piriformis and dividing into superficial and deep branches. In the pelvis it supplies piriformis, obturator internus and a nutrient artery to the ilium. The superficial branch enters the deep surface of gluteus maximus. Its numerous branches supply the muscle and anastomose with the inferior gluteal branches while others perforate the tendinous medial attachment of the muscle to supply the skin over the sacrum where they anastomose with the posterior branches of the lateral sacral arteries. The deep branch of the superior gluteal artery passes between gluteus medius and the bone, soon dividing into superior and inferior branches. The superior branch skirts the superior border of gluteus minimus to the anterior superior iliac spine and anastomoses with the deep circumflex iliac artery and the ascending branch of the lateral circumflex femoral artery. The inferior branch runs through gluteus minimus obliquely, supplies it and gluteus medius and anastomoses with the lateral circumflex femoral artery. A branch enters the trochanteric fossa to join the inferior gluteal artery and ascending branch of the medial circumflex femoral artery while other branches run through gluteus minimus to supply the hip joint. The superior gluteal artery occasionally arises directly from the internal iliac artery with the inferior gluteal artery and sometimes from the internal pudendal artery. Anterior trunk branches
Superior vesical artery (See also p. 1361.) The superior vesical artery is the first large branch of the anterior trunk. It lies on the lateral wall of the pelvis just below the brim and runs anteroinferiorly medial to the periosteum of the posterior surface of the pubis. It supplies the distal end of
the ureter, the bladder, the proximal end of the vas deferens and the seminal vesicles. It also gives origin to the umbilical artery in the foetus, which remains as a fibrous cord, the medial umbilical ligament, in the adult. This vessel occasionally remains patent as a small artery supplying the umbilicus. Inferior vesical artery (See also p. 1361.) The inferior vesical artery may arise as a common branch with the middle rectal artery. In the female it is often replaced by the vaginal artery. It supplies the bladder, the prostate, the seminal vesicles and the vas deferens. Middle rectal artery (See also p. 1361.) The middle rectal artery is often multiple and may be small. It runs into the lateral fascial coverings of the mesorectum. It occasionally arises close to or in common with the origin of the inferior vesical artery in males. Vaginal artery (See also p. 1361.) In females the vaginal artery may replace the inferior vesical artery. It may arise from the uterine artery close to its origin. Obturator artery The obturator artery runs anteroinferiorly from the anterior trunk on the lateral pelvic wall to the upper part of the obturator foramen. It leaves the pelvis via the obturator canal and divides into anterior and posterior branches. In the pelvis it is related laterally to the fascia over obturator internus and is crossed on its medial aspect by the ureter and, in the male, by the vas deferens. In the nulliparous female the ovary lies medial to it. The obturator nerve is above the artery, the obturator vein below it. In the pelvis the obturator artery provides iliac branches to the iliac fossa. These supply the bone and iliacus and anastomose with the iliolumbar artery. A vesical branch runs medially to the bladder and sometimes replaces the inferior vesical branch of the internal iliac artery. A pubic branch usually arises just before the obturator artery leaves the pelvis, and ascends over the pubis to anastomose with the contralateral artery and the pubic branch of the inferior epigastric artery. Outside the pelvis the anterior and posterior terminal branches encircle the foramen between obturator externus and the obturator membrane. The anterior branch curves anteriorly on the membrane and then inferiorly along its anterior margin to supply branches to obturator externus, pectineus, the femoral adductors and gracilis. It anastomoses with the posterior branch and the medial circumflex femoral artery. The posterior branch follows the posterior margin of the foramen and turns anteriorly on the ischial part to anastomose with the anterior branch. It supplies the muscles attached to the ischial tuberosity and anastomoses with the inferior gluteal artery. An acetabular branch enters the hip joint at the acetabular notch, ramifies in the fat of the acetabular fossa and sends a branch along the ligament of the femoral head. Occasionally the obturator artery is replaced by an enlarged pubic branch of the inferior epigastric artery (p. 1101) which descends almost vertically to the obturator foramen. It usually lies near the external iliac vein, lateral to the femoral ring, and is rarely injured during inguinal or femoral hernia surgery. Sometimes it curves along the edge of the lacunar part of the inguinal ligament, partly encircling the neck of a hernial sac, and may be inadvertently cut during enlargement of the femoral ring in reducing a femoral hernia. Uterine artery (See also p. 1361.) The uterine artery is an additional branch in females. It is a large branch which arises below the obturator artery on the lateral wall of the pelvis and runs inferomedially into the broad ligament of the uterus. Internal pudendal artery (in the pelvis) (See also p. 1361.)
page 1361 page 1362
The internal pudendal artery arises just below the origin of the obturator artery. It descends laterally to the inferior rim of the greater sciatic foramen, where it leaves the pelvis between piriformis and ischiococcygeus to enter the gluteal region. It then curves around the dorsum of the ischial spine to enter the perineum by the lesser sciatic foramen. This course effectively allows the nerve to wrap around the posterior limit of levator ani at its attachment to the ischial spine and so gain access to the perineum. In the pelvis the internal pudendal artery crosses anterior to piriformis, the sacral plexus and the inferior gluteal artery. Behind the ischial spine it is covered by gluteus maximus, with the pudendal nerve medial and the nerve to obturator internus lateral to it. Several muscular branches leave the pudendal artery in the pelvis and gluteal region to supply the adjacent muscles
and nerves. Inferior gluteal artery The inferior gluteal artery is the larger terminal branch of the anterior internal iliac trunk and principally supplies the buttock and thigh. It descends posteriorly, anterior to the sacral plexus and piriformis but posterior to the internal pudendal artery. It passes between the first and second or second and third sacral anterior spinal nerve rami, then between piriformis and ischiococcygeus. It runs through the lower part of the greater sciatic foramen to reach the gluteal region. The artery runs inferiorly between the greater trochanter and ischial tuberosity with the sciatic and posterior femoral cutaneous nerves deep to gluteus maximus. It continues down the thigh, supplying the skin and anastomosing with branches of the perforating arteries. The inferior gluteal and internal pudendal arteries often arise as a common stem from the internal iliac, sometimes with the superior gluteal artery. Inside the pelvis the inferior gluteal artery gives branches to piriformis, ischiococcygeus and iliococcygeus. Occasionally it contributes to the middle rectal arterial supply and, in the male, supplies vessels to the seminal vesicles and prostate.
EXTERNAL ILIAC ARTERIES The external iliac arteries are of larger calibre than the internal iliac artery. Each artery descends laterally along the medial border of psoas major from the common iliac bifurcation to a point midway between the anterior superior iliac spine and the symphysis pubis. It enters the thigh posterior to the inguinal ligament to become the femoral artery. On the right the artery is separated from the terminal ileum and, usually, the appendix by the parietal peritoneum and extraperitoneal tissue. On the left the artery is separated from the sigmoid colon and coils of the small intestine lie anteromedially. At its origin the artery may be crossed by the ureter. It is also crossed by the gonadal vessels, the genital branch of the genitofemoral nerve, the deep circumflex iliac vein and the vas deferens (male) or round ligament (female). Posterior to the artery the iliac fascia separates it from the medial border of psoas major. The external iliac vein lies partly posterior to its upper part but is more medial to it below. Laterally, it is related to psoas major which is covered by the iliac and psoas fascia. Numerous lymph vessels and nodes lie on its front and sides. The external iliac artery is principally the artery of the lower limb and as such has few branches in the pelvis. Apart from very small vessels to psoas major and neighbouring lymph nodes, the artery has no branches until it gives off the inferior epigastric and deep circumflex iliac arteries which arise near to its passage under the inguinal ligament. Deep circumflex iliac artery The deep circumflex iliac artery branches laterally from the external iliac artery almost opposite the origin of the inferior epigastric artery. It ascends and runs laterally to the anterior superior iliac spine behind the inguinal ligament in a sheath formed by the union of the transversalis and iliac fasciae. There it anastomoses with the ascending branch of the lateral circumflex femoral artery, pierces the transversalis fascia and skirts the internal lip of the iliac crest. About halfway along the iliac crest it runs through transversus abdominis and then between transversus and internal oblique to anastomose with the iliolumbar and superior gluteal arteries. At the anterior superior iliac spine it gives off a large ascending branch, which runs between internal oblique and transversus abdominis. It supplies both muscles and anastomoses with the lumbar and inferior epigastric arteries. Inferior epigastric artery (See also p. 1362.) The inferior epigastric artery originates from the external iliac artery posterior to the inguinal ligament. It curves forwards in the anterior extraperitoneal tissue and ascends obliquely along the medial margin of the deep inguinal ring where it continues as an artery of the anterior abdominal wall.
VEINS OF THE PELVIS The true pelvis contains a large number of veins which drain the wall and most of the viscera contained within the pelvis and carry venous blood from the gluteal region, thigh the hip. The external iliac veins, lying close to the brim of the pelvis, carry the venous drainage of most of the lower limb. There is considerable variation in the venous drainage of the pelvis and although the major veins frequently follow their named arterial counterparts, the small tributaries exhibit a
great deal of variation between individuals.
COMMON ILIAC VEINS (Fig. 108.5) The common iliac vein is formed by the union of external and internal iliac veins, anterior to the sacroiliac joints. It ascends obliquely to end at the right side of the fifth lumbar vertebra, uniting at an acute angle with the contralateral vessel to form the inferior vena cava. The right common iliac vein is shorter and more nearly vertical, lying posterior then lateral to its artery. The right obturator nerve passes posterior. The left common iliac vein is longer and more oblique and lies first medial, then posterior to its artery. It is crossed anteriorly by the attachment of the sigmoid mesocolon and superior rectal vessels. Each vein receives iliolumbar and sometimes lateral sacral veins. The left common iliac usually drains the median sacral vein. There are no valves in these veins. The left common iliac vein occasionally ascends to the left of the aorta to the level of the kidney where it receives the left renal vein and crosses anterior to the aorta to join the inferior vena cava. This vessel represents the persistent caudal half of the left postcardinal or supracardinal vein. Median sacral veins The medial sacral veins accompany the corresponding artery anterior to the sacrum, and unite to form a single vein which usually ends in the left common iliac vein. Sometimes it ends at the common iliac junction. Internal pudendal veins The internal pudendal veins are venae comitantes of the internal pudendal artery. They unite as a single vessel ending in the internal iliac vein. They receive veins from the penile bulb and the scrotum (males) or clitoris and labia (females) and the inferior rectal veins.
page 1362 page 1363
Figure 108.5 Venogram showing the veins of the pelvis and groin. Contrast medium has been injected into the bodies of the pubic bones. (Provided by M Lea Thomas.)
INTERNAL ILIAC VEIN The internal iliac vein is formed by the convergence of several veins above the greater sciatic foramen. It does not have the predictable trunks and branches of the internal iliac artery but its branches drain the same territories. It ascends
posteromedial to the internal iliac artery to join the external iliac vein, forming the common iliac vein at the pelvic brim, anterior to the lower part of the sacroiliac joint. It is covered anteromedially by parietal peritoneum. Its tributaries are the gluteal, internal pudendal and obturator veins, which originate outside the pelvis; the lateral sacral veins which run from the anterior surface of the sacrum; and the middle rectal, vesical, uterine and vaginal veins which originate in the venous plexuses of the pelvic viscera. The venous drainage of the leg may be blocked by thrombosis involving the external iliac systems and the inferior vena cava. Under these circumstances, the pelvic veins, particularly the internal iliac tributaries, enlarge and provide a major avenue of venous return from the femoral system. Surgical interference with these veins may seriously compromise venous drainage and precipitate oedema of one or both legs. Superior gluteal veins The superior gluteal veins are the venae comitantes of the superior gluteal artery. They receive branches corresponding to branches of the artery and enter the pelvis via the greater sciatic foramen, above piriormis. They join the internal iliac vein, frequently as a single trunk. Inferior gluteal veins The inferior gluteal veins are venae comitantes of the inferior gluteal artery. They begin proximally and posterior in the thigh, where they anastomose with the medial circumflex femoral and first perforating veins. They enter the pelvis low in the greater sciatic foramen, joining to form a vessel opening into the distal (lower) part of the internal iliac vein. They connect with the superficial gluteal veins by perforating veins (Doyle 1970) analogous to the sural perforating veins. They probably have a venous 'pumping' role, and provide collaterals between the femoral and internal iliac veins. Obturator vein The obturator vein begins in the proximal adductor region and enters the pelvis via the obturator foramen. It runs posteriorly and superiorly on the lateral pelvic wall below the obturator artery and between the ureter and internal iliac artery to end in the internal iliac vein. It is sometimes replaced by an enlarged pubic vein, which joins the external iliac vein. Lateral sacral veins The lateral sacral veins accompany the lateral sacral arteries, and are interconnected by a sacral venous plexus. Middle rectal vein The middle rectal vein begins in the rectal venous plexus and drains the rectum and mesorectum. It often receives tributaries from the bladder and the prostate and seminal vesicle (males) and the posterior aspect of the vagina (females). It is variable in size and runs laterally on the pelvic surface of levator ani to end in the internal iliac vein.
EXTERNAL ILIAC VEIN The external iliac vein is the proximal continuation of the femoral vein. It begins posterior to the inguinal ligament, ascends along the pelvic brim and ends anterior to the sacroiliac joint by joining the internal iliac vein to form the common iliac vein. On the right it lies medial to the external iliac artery, gradually inclining behind it as it ascends. On the left it is wholly medial. Disease of the external iliac artery may cause it to adhere closely to the vein at the point where it is in contact, and, particularly on the right side, the walls of the vessels may become fused, making dissection hazardous. Medially the external iliac vein is crossed by the ureter and internal iliac artery. In males it is crossed by the vas deferens, in females by the round ligament and ovarian vessels. Lateral to it lies psoas major, except where the artery intervenes. The vein is usually valveless, but may contain a single valve. It tributaries are the inferior epigastric, deep circumflex iliac and pubic veins. Inferior epigastric vein One or two inferior epigastric veins accompany the artery and drain into the external iliac vein a little above the inguinal ligament. Deep circumflex iliac vein The deep circumflex vein is formed from venae comitantes of the corresponding artery. It joins the external iliac vein a little above the inferior epigastric veins after crossing anterior to the external iliac artery.
Pubic vein The pubic vein connects the external iliac and the obturator vein. It ascends on the pelvic surface of the pubis with the pubic branch of the inferior epigastric artery. It sometimes replaces the normal obturator vein.
LYMPHATIC DRAINAGE OF THE PELVIS COMMON ILIAC NODES (Fig. 108.6) The common iliac nodes are grouped around the artery, and one or two lie inferior to the aortic bifurcation and anterior to the fifth lumbar vertebra or sacral promontory. They drain the external and internal iliac nodes and connect to the lateral aortic nodes. They usually lie in medial, lateral and anterior chains around the artery, the lateral being the main route. Since they receive drainage from both internal and external iliac nodes, the common iliac nodes receive the entire lymphatic drainage of the lower limb.
EXTERNAL ILIAC NODES The external iliac nodes usually form three subgroups, lateral, medial and anterior to the external iliac vessels. The medial nodes are considered the main channel of drainage, collecting lymph from the lower limb via the inguinal nodes, the deeper layers of the infra-umbilical abdominal wall, the adductor region of the thigh, the glans penis or clitoris, the membranous urethra, prostate, fundus of the bladder, uterine cervix and upper vagina. Their efferents pass to the common iliac nodes. Inferior epigastric and circumflex iliac nodes The inferior epigastric and circumflex iliac nodes are associated with their vessels and drain the corresponding areas to the external iliac nodes.
INTERNAL ILIAC NODES (Fig. 108.7)
page 1363 page 1364
Figure 108.6 Lymphatic drainage of the male pelvis and urinary bladder.
Figure 108.7 Lymphatic drainage of the female pelvis. (After Cuneo and Marcille.)
The internal iliac nodes surround the branches of the internal iliac vessels and receive afferents from most of the pelvic viscera (with the exception of the gonads and the rectum), deeper parts of the perineum and the gluteal and posterior femoral muscles. They drain to the common iliac nodes. The individual groups are considered in the description of the viscera. There are frequent connections between the right and left groups particularly when they lie close to the anterior and posterior midlines.
Innervation of the pelvis The pelvis contains the lumbosacral nerve trunk, the sacral plexus, the coccygeal plexus and the pelvic parts of the sympathetic and parasympathetic systems. These supply the somatic and autonomic innervation to the majority of the pelvic visceral organs, the pelvic floor and perineum, the gluteal region and the lower limb. The ventral rami of the sacral and coccygeal spinal nerves form the sacral and coccygeal plexuses. The upper four sacral ventral rami enter the pelvis by the anterior sacral foramina, the fifth between the sacrum and coccyx, while that of the coccygeal nerve curves forwards below the rudimentary transverse process of the first coccygeal segment. The first and second sacral ventral rami are large, the third to fifth diminish progressively and the coccygeal is the smallest. Each receives a grey ramus communicans from a corresponding sympathetic ganglion. Visceral efferent rami leave the second to fourth sacral rami as pelvic splanchnic nerves, containing parasympathetic fibres which reach minute ganglia in the walls of the pelvic viscera.
LUMBOSACRAL TRUNK AND SACRAL PLEXUS (Fig. 111.41) The sacral plexus is formed by the lumbosacral trunk, the first to third sacral ventral rami and part of the fourth, the remainder of the last joining the coccygeal plexus. The lumbar part of the lumbosacral trunk contains part of the fourth and all the fifth lumbar ventral rami; it appears at the medial margin of psoas major, and descends over the pelvic brim anterior to the sacroiliac joint to join the first sacral ramus. The greater part of the second and third sacral rami converge on the inferomedial aspect of the lumbosacral trunk in the greater sciatic foramen to form the sciatic nerve. The ventral and dorsal divisions of the nerves do not separate physically from each other but the fibres remain separate within the rami, and ventral and dorsal divisions of each contributing root join within the sciatic nerve.
The fibres of the dorsal divisions will go on to form the common peroneal nerve and the ventral division fibres form the tibial nerve. The sciatic nerve occasionally divides into common peroneal and tibial nerves inside the pelvis. In these cases the common peroneal nerve usually runs through piriformis. The sacral plexus lies against the posterior pelvic wall anterior to piriformis, posterior to the internal iliac vessels and ureter, and behind the sigmoid colon on the left. The superior gluteal vessels run between the lumbosacral trunk and first sacral ventral ramus or between the first and second sacral rami, while the inferior gluteal vessels lie between the first and second or second and third sacral rami (Fig. 108.8). The sacral plexus is not commonly involved in malignant tumours of the pelvis because in lies behind the relatively dense presacral fascia which resists all but locally very advanced malignant infiltration. When it occurs, there is intractable pain in the distribution of the branches of the plexus which may be very difficult to treat. The plexus may also be involved in the reticuloses or be affected by plexiform neuromas. UPDATE Date Added: 07 March 2006 Publication Services, Inc. Abstract: Urinary and sexual function after total mesorectal excision. Recent results Click on the following line to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=15865034&query_hl=6&itool=pubmed_docsum Urinary and sexual function after total mesorectal excision. Recent results. Maurer CA: Cancer Res 165:196-204, 2005.
BRANCHES OF THE SACRAL PLEXUS Ventral divisions Nerve to quadratus femoris and gemellus L4,5, S1 inferior Nerve to obturator internus and gemellus L5, S1,2 superior Nerve to piriformis Superior gluteal nerve Inferior gluteal nerve Posterior femoral cutaneous nerve S2,3 Tibial (sciatic) nerve L4,5, S1,2,3 Common peroneal (sciatic) nerve Perforating cutaneous nerve Pudendal nerve S2,3,4 Nerves to levator ani and external anal S4 sphincter Pelvic splanchnic nerves
Dorsal divisions
S2 (S1) L4,5, S1 L5, S1,2 S1,2 L4,5, S1,2 S2,3
S2,3 (S4)
The branches of the sacral plexus are: The course and distribution of most of the branches of the sacral plexus are covered fully on page 1456.
PUDENDAL NERVE (IN THE PELVIS) The pudendal nerve arises from the ventral divisions of the second, third and fourth sacral ventral rami and is formed just above the superior border of the sacrotuberous ligament and the upper fibres of ischiococcygeus. It leaves the pelvis via the greater sciatic foramen between piriformis and ischiococcygeus, enters the gluteal region and crosses the sacrospinous ligament close to its attachment to the ischial spine. The nerve lies medial to the internal pudendal vessels on the spine. It accompanies the internal pudendal artery through the lesser sciatic foramen into the pudendal (Alcock's) canal on the lateral wall of the ischioanal fossa. In the posterior part of the canal it gives rise to the inferior rectal nerve, the perineal nerve and the dorsal nerve of the penis or clitoris.
SACRAL VISCERAL BRANCHES These arise from the second to fourth sacral ventral rami to innervate the pelvic viscera; they are termed pelvic splanchnic nerves.
SACRAL MUSCULAR BRANCHES page 1364 page 1365
Figure 108.8 The lumbosacral plexus in the pelvis. The pelvic viscera have been omitted for clarity.
Several muscular branches arise from the fourth sacral ventral ramus to supply the superior surface of levator ani and the upper part of the external anal sphincter. The branches to levator ani enter the superior (pelvic) surface of the muscle whilst the branch to the external anal sphincter (also referred to as the perineal branch of the fourth sacral nerve) reaches the ischioanal fossa by running through ischiococcygeus or between ischiococcygeus and iliococcygeus. It supplies the skin between the anus and coccyx via its cutaneous branches.
COCCYGEAL PLEXUS The coccygeal plexus is formed by a small descending branch from the fourth sacral ramus and by the fifth sacral and coccygeal ventral rami. The fifth sacral ventral ramus emerges from the sacral hiatus, curves round the lateral margin of the sacrum below its cornu and pierces ischiococcygeus from below to reach its upper, pelvic surface. Here it is joined by a descending branch of the fourth sacral ventral ramus, and the small trunk so formed descends on the pelvic surface of ischiococcygeus. They join the minute coccygeal ventral ramus which emerges from the sacral hiatus and curves round the lateral coccygeal margin to pierce coccygeus to reach the pelvis. This small trunk is the coccygeal plexus. Anococcygeal nerves arise from it and form a few fine filaments which pierce the sacrotuberous ligament to supply the adjacent skin.
PELVIC PART OF THE SYMPATHETIC SYSTEM The pelvic sympathetic trunk lies in the extraperitoneal tissue anterior to the sacrum beneath the presacral fascia. It lies medial or anterior to the anterior sacral foramina and has four or five interconnected ganglia. Above, it is continuous with the lumbar sympathetic trunk. Below the lowest ganglia the two trunks converge to unite in the small ganglion impar anterior to the coccyx. Grey rami communicantes pass from the ganglia to sacral and coccygeal spinal nerves but there are no white rami communicantes. Medial branches connect across the midline and twigs from the first two ganglia join the inferior hypogastric plexus or the hypogastric 'nerve'. Other branches form a plexus on the median sacral artery.
VASCULAR BRANCHES Postganglionic fibres pass through the grey rami communicantes to the roots of the sacral plexus. Those forming the tibial nerve are conveyed to the popliteal artery and its branches in the leg and foot whilst those in the pudendal and
superior and inferior gluteal nerves accompany the same named arteries to the gluteal and perineal tissues. Branches may also supply the pelvic lymph nodes. Preganglionic fibres for the rest of the lower limb are derived from the lower three thoracic and upper two or three lumbar spinal segments. They reach the lower thoracic and upper lumbar ganglia through white rami communicantes and descend through the sympathetic trunk to synapse in the lumbar ganglia. Postganglionic fibres pass from these ganglia via grey rami communicantes to the femoral nerve which carries them to the distribution of the femoral artery and its branches. Some fibres descend through the lumbar ganglia to synapse in the upper two or three sacral ganglia, from which postganglionic axons join the tibial nerve to supply the popliteal artery and its branches in the leg and foot. Sympathetic denervation of vessels in the lower limb can be effected by removing or ablating the upper three lumbar ganglia and the intervening parts of the sympathetic trunk, which is rarely useful in treating vascular insufficiency of the lower limb.
© 2008 Elsevier
PERINEUM Muscles and fasciae of the perineum The perineum is an approximately diamond-shaped region which lies below the pelvic floor, between the inner aspects of the thighs and anterior to the sacrum and coccyx. It is usually described as if from the position of an individual lying supine with the hip joints in abduction and partial flexion. The surface projection of the perineum and the form of the skin covering it varies considerably depending on the position of the thighs but the deep tissues themselves occupy relatively fixed positions. The perineum is bounded anteriorly by the pubic symphysis and its arcuate ligament, posteriorly by the coccyx, anterolaterally by the ischiopubic rami and the ischial tuberosities and posterolaterally by the sacrotuberous ligaments. The deep limit of the perineum is the inferior surface of the pelvic diaphragm and its superficial limit is the skin which is continuous with that over the medial aspect of the thighs and the lower abdominal wall. An arbitrary line joining the ischial tuberosities (the interischial line) divides the perineum into an anterior urogenital triangle and a posterior anal triangle. The urogenital triangle faces downwards and forwards, whereas the anal triangle faces downwards and backwards. The male urogenital triangle contains the bulb and attachments of the penis (Fig. 108.9, Chapter 100) and the female urogenital triangle contains the mons pubis, the labia majora, the labia minora, the clitoris and the vaginal and urethral orifices (Chapter 107). page 1365 page 1366
Figure 108.9 Muscles and fasciae of the male perineum. On the left side the skin and superficial fascia of the perineum only have been removed. The perineal (scrotal) artery has been shown as it runs forward into the scrotal tissues. On the right side, the corpora cavernosa and corpus spongiosum and their associated muscles, the superficial perineal muscles and inferior layer of the urogenital fascia have been removed to reveal the underlying deep muscles and arteries of the perineum. All veins and nerves have been omitted for clarity.
ANAL TRIANGLE The structure of the anal triangle is similar in males and females. The main difference reflects the wider transverse dimension of the triangle in females as a result of the larger size of the pelvic outlet (p. 1430). The anal triangle contains the anal canal and its sphincters, and the ischioanal fossa and its contained
nerves and vessels. It is lined by superficial and deep fascia. Superficial fascia
The superficial fascia of the region is thin and is continuous with the superficial fascia of the skin of the perineum, thighs and buttocks. Deep fascia
The deep fascia lines the inferior surface of levator ani and is continuous at its lateral origin with the fascia over obturator internus below the attachment of levator. It lines the deep portion of the ischioanal fossa and its lateral walls. Ischioanal fossa
The ischioanal fossa is an approximately horse-shoe shaped region filling the majority of the anal triangle. The 'arms' of the horseshoe are triangular in cross section because levator ani slopes downwards towards the anorectal junction (Fig. 84.1). Although it is often referred to as a space, it is filled with loose adipose tissue and occasional blood vessels. The anal canal and its sphincters lie in the centre of the horse-shoe. Above them the deep medial limit of the fossa is formed by the deep fascia over levator ani. The outer boundary of the fossa is formed anterolaterally by the deep fascia over obturator internus deeply and the periosteum of the ischial tuberosities more superficially. Posterolaterally the outer boundary is formed by the lower border of gluteus maximus and the sacrotuberous ligament. Anteriorly, the superficial boundary of the fossa is formed by the posterior aspect of the muscles of the urogenital triangle. Deep to this there is no fascial boundary between the fossa and the tissues deep to the perineal membrane (p. 1367) as far anteriorly as the posterior surface of the pubis below the attachment of levator ani. Posteriorly, the fossa contains the attachment of the external anal sphincter to the tip of the coccyx: above and below this the adipose tissue of the fossa is uninterrupted across the midline. These continuations of the ischioanal fossa mean that infections, tumours and fluid collections within may not only enlarge relatively freely to the side of the anal canal but may also spread with little resistance to the opposite side and deep to the perineal membrane. The internal pudendal vessels and accompanying nerves lie in the lateral wall of the ischioanal fossa, enclosed in fascia to form the pudendal canal. The inferior rectal vessels and nerves cross the fossa from the pudendal canal and often branch within it. The fossa is an important surgical plane during resections of the anal canal and anorectal junction for malignancy. It provides an easy plane of dissection with relatively few vessels encountered, which encompasses all of the muscular structures of the anal canal. It leads to the inferior surface of levator ani through which the dissection is carried. External anal sphincter (See also p. 1366.) page 1366 page 1367
The external anal sphincter is an oval tube of striated muscle which surrounds the lowest part of the anal canal. The upper most (deepest) fibres blend with the lowest fibres of puborectalis and the two are seen to be continuous on endoanal ultrasound and magnetic resonance imaging. Anteriorly some of these upper fibres decussate into the superficial transverse perineal muscles and posteriorly, some fibres are attached to the anococcygeal raphe. The majority of the middle fibres of the external anal sphincter surround the lower part of the internal sphincter. This portion is attached anteriorly to the perineal body and posteriorly to the coccyx via the anococcygeal ligament. Some fibres from each side of the sphincter decussate in these areas to form a sort of commissure in the anterior and posterior midline. The anterior and posterior attachments of the external anal sphincter give the muscular tube an oval profile lying anteroposteriorly. The lower fibres lie below the level of the internal anal sphincter and are separated from the lowest anal epithelium only by submucosa. Anococcygeal ligament
The anococcygeal ligament is a layered musculotendinous structure running between the middle portion of the external anal sphincter and the coccyx. The lowest portion of the presacral fascia lies above the deep part of the ligament and between the two lie the most posterior fibres of the raphe of iliococcygeus. These three structures together are sometimes referred to as the postanal plate. Division of the anococcygeal raphe may cause descent of the anal canal and a lowering of the posterior part of the anal triangle but does not demonstrably interfere with the process of defaecation.
UROGENITAL TRIANGLE The urogenital triangle is bounded posteriorly by the interischial line which usually
overlies the posterior border of the transverse perineal muscles. Anteriorly and laterally it is bounded deeply by the symphysis pubis and ischiopubic rami. In males, the urogenital triangle extends superficially to encompass the scrotum and the root of the penis. In females, it extends to the lower limit of the labia and mons pubis. The urogenital triangle is divided into two parts by a strong perineal membrane. The deep perineal space lies above the membrane and below it is the superficial perineal space. The female urogenital triangle includes muscles, fasciae and spaces similar to those in the male. There are some differences in size and disposition caused by the presence of the vagina and female external genitalia.
DEEP PERINEAL SPACE The deep perineal space was long regarded as an anatomical region between the urogenital diaphragm and the perineal membrane which contained the urethra and urethral sphincter. However, the urogenital diaphragm does not exist and the urethral sphincter, previously thought to be the principle content of the deep perineal space where it was described as surrounding the urethra, is now recognized to be contained within the urethra. The deep perineal space is bounded deeply by the endopelvic fascia of the pelvic floor and superficially by the perineal membrane. Between these two fascial layers lie the deep transverse perinei, superficial to the urethral sphincter mechanism and pubourethralis, and in females superficial to the compressor urethrae and sphincter urethrovaginalis. These muscles do not form a true diaphragmatic sheet as such because fibres from several parts extend through the visceral outlet in the pelvic floor into the lower reaches of the pelvic cavity. Perineal membrane (inferior fascia of the urogenital diaphragm)
The perineal membrane is a triangular membrane which stretches almost horizontally across the urogenital triangle. It is attached laterally to the periosteum of the ischiopubic rami and its apex is attached to the arcuate ligament of the pubis. It is particularly thick in this area and is referred to as the transverse perineal ligament. The posterior border is fused with the deep part of the perineal body and is continuous with the fascia over the deep transverse perinei. In the male, the perineal membrane is crossed by the urethra, 2-3 cm behind the inferior border of the symphysis pubis; the vessels and nerves to the bulb of the penis; the ducts of the bulbourethral glands, posterolateral to the urethral orifice; the deep dorsal vessels and dorsal nerves of the penis, behind the pubic arch in the midline; and the posterior scrotal vessels and nerves, anterior to the transverse perinei. In the female the perineal membrane is less well defined than in the male. It is divided almost into two halves by the vagina and urethra such that it forms a triangle on each side of these structures and the pubourethral ligament (the female equivalent of the transverse perineal ligament in males) links the two sides anteriorly behind the pubic arch. It is crossed by the urethra, 2-3 cm behind the inferior border of the symphysis pubis; the vagina, centrally; the ducts of Bartholin's glands, posterolateral to the urethral orifice; the deep dorsal vessels and dorsal nerves of the clitoris, behind the pubic arch in the midline; the posterior labial vessels and nerves, anterior to the transverse perinei. Deep transverse perinei
The deep transverse perinei form an incomplete sheet of muscle extending across the urogenital triangle from the medial aspects of the ischiopubic rami. Posteriorly the sheet is attached to the perineal body where its fibres decussate with those of the opposite side. Anteriorly, the muscles are deficient and the visceral structures pass across the endopelvic fascia and the perineal membrane. Some fibres pass to the deep part of external anal sphincter posteriorly and sphincter urethrae anteriorly. Together with the superficial transverse perinei the muscles act to tether the perineal body in the median plane and may help support the visceral canals which pass through them. They are supplied by the perineal branches of the pudendal vessels and nerves. The urethral sphincter mechanism
The urethral sphincter mechanism consists of the intrinsic striated and smooth muscle of the urethra and the pubourethralis component of levator ani which surrounds the urethra at the point of maximum concentration of those muscles. It surrounds the membranous urethra in the male and the middle and lower thirds of the urethra in the female. In the male, fibres also reach up to the lowest part of the neck of the bladder and, between the two, fibres lie on the surface of the prostate.
The bulk of the fibres surround the membranous urethra and some fibres are attached to the inner surface of the ischiopubic ramus. In the female, the sphincter mechanism surrounds more than the middle third of the urethra. It blends above with the smooth muscle of the bladder neck and below with the smooth muscle of the lower urethra and vagina. Actions The urethral sphincter mechanism compresses the urethra, particularly when the bladder contains fluid. Its location around the region of highest urethral closing pressure suggests that it plays an important role in the continence of urine. Like bulbospongiosus, it is relaxed during micturition, but it contracts to expel final drops of urine, or of semen in the male, from the bulbar urethra. It may be stimulated via a vaginal tampon electrode as has been used in the treatment of stress incontinence of urine in females. Innervation The urethral sphincter mechanism extends from the perineum through the urogenital hiatus into the pelvic cavity. It probably receives innervation via the perineal branch of the pudendal nerve from below and direct branches from the sacral plexus and the pelvic splanchnic nerves from above (Wendell-Smith & Wilson 1991). All these nerves originate in the second, third and fourth sacral spinal segments. Compressor urethrae
Compressor urethrae exists in females and arises from the ischiopubic rami of each side by a small tendon. Fibres pass anteriorly to meet their contralateral counterparts in a flat band which lies anterior to the urethra, below sphincter urethrae (Fig. 108.10). A variable number of fibres from the same origin fan medially to reach the lower walls of the vagina. These fibres may rarely reach as far posteriorly as the perineal body. Sphincter urethrovaginalis
Sphincter urethrovaginalis exists in females and arises from the perineal body. Its fibres pass forwards on either side of the vagina and urethra to meet their contralateral counterparts in a flat band, anterior to the urethra, below compressor urethrae (Fig. 108.10). Actions The direction of the fibres of compressor urethrae and sphincter urethrovaginalis suggests that they produce elongation as well as compression of the membranous urethra and thus aid continence in females.
SUPERFICIAL PERINEAL SPACE (Fig. 108.11) The superficial perineal space lies below the perineal membrane and is limited superficially by the superficial perineal fascia. It contains the corpora cavernosa and corpus spongiosum, ischiocavernosus, bulbospongiosus and the superficial transverse perinei and branches of the pudendal vessels and nerves. In the female it is crossed by the urethra and vagina and contains the clitoris. In the male it contains the urethra which runs in the root of the penis. page 1367 page 1368
Figure 108.10 Muscles of the female perineum. On the right side, the membranous layer of superficial fascia has been removed (note the cut edge). On the left side, superficial perineal muscles and overlying fascia have been removed to show the deep perineal muscles. The smaller figure illustrates the continuity of the deep perineal muscles with sphincter urethrae.
page 1368 page 1369
Figure 108.11 Muscles and fasciae of the male perineum - coronal view. The section passes through the bulb of the penis at the level of the urethra. The deep perineal space is continuous with the ischioanal fossa posteriorly. The layers of the urogenital fascia and the superficial fascia of the perineum are in green and muscles are shown in brown.
Superficial perineal fascia
The superficial fascia of the urogenital triangle (Colles' fascia) forms a clear, surgically recognizable plane beneath the skin of the anterior perineum. It is firmly attached posteriorly to the fascia over the superficial transverse perinei and the posterior limit of the perineal membrane. Laterally, it is attached to the margins of the ischiopubic rami as far back as the ischial tuberosities. From here it runs more superficially to the skin of the urogenital triangle, lining the external genitalia before running anteriorly into the skin of the lower abdominal wall where it is continuous with the membranous fascia (Scarpa's fascia). In the male, the superficial perineal fascia covers the corpora cavernosa from their attachment to the ischiopubic rami and is continuous with the fascia of the penis. It is also continuous with the fascial layer in the skin of the scrotum containing the dartos muscle. In females the fascia follows the same limits but is much less extensive in the labia majora. Since the superficial perineal fascia is in continuity with the fascia of the anterior abdominal wall but tethered firmly by its posterolateral attachments, fluid, blood or pus may track freely from the lower anterior abdominal wall into the superficial perineal space. Similarly blood, urine or fluid accumulating in the superficial space due to trauma or surgery on the urogenital triangle will spread throughout the tissues of the triangle including the scrotum or labia majora but cannot pass posteriorly into the anal triangle or laterally into the medial thigh. Deep perineal fascia
The deep perineal fascia is attached to the ischiopubic rami and to the posterior margin of the perineal membrane and perineal body over the membranous layer. In front it fuses with the suspensory ligament of the penis or clitoris and the fasciae of external oblique and the rectus sheath. Perineal body
The perineal body is a poorly defined aggregation of fibromuscular tissue located in the midline at the junction between the anal and urogenital triangles. It is attached to many structures in both the deep and superficial urogenital spaces. Posteriorly fibres from the middle part of the external anal sphincter and the conjoint longitudinal coat merge with the body. Superiorly it is continuous with the rectoprostatic or rectovaginal septum including fibres from levator ani
(puborectalis or pubovaginalis). Anteriorly, the deep transverse perinei, the superficial transverse perinei and bulbospongiosus also attach and contribute fibres to it (Fig. 108.3). The perineal body is continuous with the perineal membrane and the superficial perineal fascia. Since the superficial perineal fascia runs forward into the skin of the perineum, the perineal body is tethered to the central perineal skin, which is often puckered over it. In males this is continuous with the perineal raphe in the skin of the scrotum. In females, the perineal body lies directly posterior, and is attached, to the posterior commissure of the labia majora and the introitus of the vagina. The anus can be surgically detached from the perineal body without any clinical consequences. However, spontaneous lacerations of the body sustained during childbirth are often associated with damage to the anterior fibres of the external anal sphincter. The deliberate division of the perineal body to facilitate delivery (episiotomy) is angled laterally to avoid such injuries. The perineal body is often used for the positioning of radiological markers used to determine the amount of descent the perineum undergoes during straining in order to assess pelvic floor dysfunction. Superficial transverse perinei
The superficial transverse perinei are narrow strips of muscle which run more or less transversely across the superficial perineal space anterior to the anus. The muscles are occasionally small and rarely absent. On each side the muscle is attached to the medial and anterior aspects of the ischial tuberosity. Medially the fibres mostly run into the perineal body although some may pass into bulbospongiosus or external anal sphincter on the same side. Bulbospongiosus
Bulbospongiosus differs between the sexes. In the male it lies in the midline, anterior to the perineal body and consists of two symmetrical parts united by a median fibrous raphe. The fibres attach to the perineal body, in which they decussate, and are attached to the transverse superficial perinei and the external anal sphincter. The fibres diverge like the halves of a feather from the median raphe. A thin layer of posterior fibres unites with the posterior portion of the perineal membrane. The majority of the middle fibres encircle the bulb of the penis and adjacent corpus spongiosum and attach to an aponeurosis on the dorsal surfaces. The anterior fibres spread out over the sides of the corpora cavernosa, ending partly in them, anterior to ischiocavernosus, and partly in a tendinous expansion which covers the dorsal vessels of the penis. Actions Bulbospongiosus helps to empty the urethra of urine after the bladder has emptied. It may assist in the final stage of erection as the middle fibres compress the erectile tissue of the bulb and the anterior fibres contribute by compressing the deep dorsal vein of the penis. It contracts six or seven times during ejaculation, assisting in the expulsion of semen. In the female bulbospongiosus also attaches to the perineal body, but the muscle on each side is separate and covers the superficial parts of the vestibular bulbs and greater vestibular glands. They run anteriorly on either side of the vagina to attach to the corpora cavernosa clitoridis. A few fibres cross over the dorsum of the body of the clitoris. The muscle acts to constrict the vaginal orifice and express the secretions of the greater vestibular glands. Anterior fibres contribute to erection of the clitoris by compressing its deep dorsal vein. Ischiocavernosus (Figs 108.12, 108.13)
In the male ischiocavernosus covers the crus penis and is attached by tendinous and muscular fibres to the medial aspect of the ischial tuberosity behind, and to the ischial ramus on both sides of the crus. The fibres end in an aponeurosis attached to the sides and under surface of the crus penis. In the female, ischiocavernosus is related to a smaller crus clitoris and has a much smaller attachment to the ischiopubic ramus, but is otherwise similar to the corresponding muscle in the male. Action Ischiocavernosus compresses the crus penis in males and may help to maintain penile erection. The muscles form a triangle on each side of the midlines with bulbospongiosus medially and the superficial transverse perinei posteriorly attached to the perineal membrane. When contracted, ischiocavernosi act together to stabilize the erect penis. In the female ischiocavernosus may help to promote clitoral erection.
Vascular supply and lymphatic drainage of the perineum (Figs 108.4, 108.5) ARTERIES OF THE PERINEUM Internal pudendal artery (in the perineum)
The internal pudendal artery enters the perineum around the posterior aspect of the ischial spine. It runs on the lateral wall of the ischioanal fossa in the pudendal (Alcock's) canal with the pudendal veins and the pudendal nerve. The canal is formed by the connective tissue binding the vessels and nerve to the perineal surface of the obturator internus fascia. The canal lies c.4 cm above the lower limit of the ischial tuberosity. Approaching the margin of the ischial branch, it proceeds above or below the inferior fascia of the urogenital diaphragm along the medial margin of the inferior pubic ramus and ends behind the inferior pubic ligament. In the male the internal pudendal artery gives a branch to the bulb of the penis before it divides into the cavernosal and dorsal arteries of the penis. The internal pudendal artery distal to its perineal branch has been named the artery of the penis in view of its distribution. The artery to the bulb supplies the corpus spongiosum, the cavernoid artery to the penis supplies the corpus cavernosa on each side and the dorsal artery runs on the dorsal aspect of the penis, supplying circumflex branches to the corpora cavernosa and corpus spongiosum before they end by anastomosing in the coronal sulcus to supply the glans penis and its overlying skin. In the female the artery to the bulb is distributed to the erectile tissue of the vestibular bulb and vagina. The cavernosal artery is much smaller and supplies the corpora cavernosa of the clitoris and the dorsal artery supplies the glans and prepuce of the clitoris. Branches of the internal pudendal artery are sometimes derived from an accessory pudendal artery, which is usually a branch of the pudendal before its exit from the pelvis. page 1369 page 1370
Figure 108.12 Muscles and fasciae of the female perineum - coronal view. The section passes through the bulb of the clitoris at the level of the urethra. The deep perineal space is continuous with the ischioanal fossa posteriorly. The layers of the perineal fascia and the superficial fascia of the perineum are in green and muscles are shown in brown.
Figure 108.13 Muscles and fasciae of the female perineum - coronal T2-weighted MRI. (Provided by Dr J Lee and Ms K Wimpey, Chelsea and Westminster Hospital, London.)
Inferior rectal artery (See also p. 1370.)
The inferior rectal artery arises just after the pudendal artery enters the canal on the lateral wall of the ischioanal fossa. It runs anteromedially through the adipose tissue of the ischioanal fossa to reach the deep portion of the external anal sphincter. It often branches before reaching the sphincter. During dissections of the anal canal, particularly during perineal excisions of the anorectum, the inferior rectal vessels are encountered in the ischioanal fossa and must be secured before division or they tend to retract laterally to the canal, where they can cause troublesome bleeding. Perineal artery page 1370 page 1371
The perineal artery is a branch of the internal pudendal artery near the anterior end of the pudendal canal, and runs through the inferior fascia of the urogenital diaphragm. In the male it approaches the scrotum in the superficial perineal space, between bulbospongiosus and ischiocavernosus. A small transverse branch passes medially, inferior to the superficial transverse perineal muscle, to anastomose with the contralateral artery and with the posterior scrotal and inferior rectal arteries. It supplies the transverse perinei, the perineal body and the posterior attachment of the bulb of the penis. The posterior scrotal arteries are usually terminal branches of the perineal artery but may also arise from its transverse branch. They are distributed to the scrotal skin and dartos muscle in the male and supply the perineal muscles. In the female the perineal artery runs an almost identical course and gives rise to similar branches. The posterior scrotal arteries are replaced by posterior labial arteries.
VEINS OF THE PERINEUM: INTERNAL PUDENDAL VEINS The internal pudendal veins are venae comitantes of the internal pudendal artery and unite as a single vessel ending in the internal iliac vein. The perineal tributaries receive veins from the penile bulb and the scrotum (males) or clitoris and labia (females) and the inferior rectal veins join towards the posterior end of the pudendal canal.
LYMPHATIC DRAINAGE OF THE PERINEUM The lymphatics from the skin of the penis and scrotum (male) or skin of the clitoris and labia (female) drain together with lymphatics from the perineal skin to the superficial inguinal nodes and thence to the deep inguinal nodes. The glans, corpora cavernosa and corpus spongiosum of the penis or clitoris drain directly to the deep inguinal nodes.
Innervation of the perineum: pudendal nerve (in the perineum) The pudendal nerve gives rise to the inferior rectal, perineal and dorsal nerves of
the penis or clitoris. The pudendal nerve is readily found in its very constant position over the ischial spine. It may be 'blocked' by infiltration with a local anaesthetic applied via a needle passed through the lateral wall of the vagina to cause anaesthesia of the perineal and anal skin. It may also be palpated here through the lateral wall of the rectum and motor terminal latencies measured.
INFERIOR RECTAL NERVE The inferior rectal nerve runs through the medial wall of the pudendal canal with the inferior rectal vessels. It crosses the ischioanal fossa to supply the external anal sphincter, the lining of the lower part of the anal canal, and the circumanal skin. It frequently breaks into terminal branches just before reaching the lateral border of the sphincter. Its cutaneous branches distributed around the anus overlap the perineal branch of the posterior femoral cutaneous nerve and the scrotal or labial nerves. The inferior rectal nerve occasionally arises directly from the sacral plexus and crosses the sacrospinous ligament or reconnects with the pudendal nerve. In females the inferior rectal nerve may supply sensory branches to the lower part of the vagina.
PERINEAL NERVE The perineal nerve is the inferior and larger terminal branch of the pudendal nerve in the pudendal canal. It runs forwards below the internal pudendal artery and accompanies the perineal artery, dividing into posterior scrotal or labial and muscular branches. The posterior scrotal or labial nerves are usually double and have medial and lateral branches which run over the perineal membrane and pass forwards in the lateral part of the urogenital triangle with the scrotal (or labial) branches of the perineal artery. They supply the skin of the scrotum or labia majora, overlapping the distribution of the perineal branch of the posterior femoral cutaneous and inferior rectal nerve. In females the posterior labial branches also supply sensory fibres to the skin of the lower vagina. Muscular branches arise directly from the pudendal nerve to supply the superficial transverse perinei, bulbospongiosus, ischiocavernosus, deep transverse perinei, sphincter urethrae and the anterior parts of the external anal sphincter and levator ani. In males a nerve to the bulb of the urethra leaves the nerve to the bulbospongiosus, piercing it to supply the corpus spongiosum penis and ends in the urethral mucosa.
DORSAL NERVE OF THE PENIS OR CLITORIS The dorsal nerve of the penis or clitoris runs anteriorly above the internal pudendal artery along the ischiopubic ramus deep to the inferior fascia of the urogenital diaphragm. It supplies the corpus cavernosum and accompanies the dorsal artery of the penis or clitoris between the layers of the suspensory ligament. In males it runs on the dorsum of the penis to end in the glans. In females the dorsal nerve of the clitoris is very small. REFERENCES Doyle JF 1970 The perforating veins of the gluteus maximus. Ir J Med Sci 3: 285-8. Medline Similar articles Wendell-Smith C P, Wilson PM 1991 The vulva, vagina and urethra and the musculature of the pelvic floor. In: Philipp E, Setchell M, Ginsburg J (eds) Scientific Foundations of Obstetrics and Gynaecology. Oxford: Butterworth-Heinemann: 84-100. page 1371 page 1372
© 2008 Elsevier
109 DEVELOPMENT OF THE UROGENITAL SYSTEM Development of the urogenital system URINARY SYSTEM The urinary and reproductive systems develop from intermediate mesenchyme and are intimately associated with one another especially in the earlier stages of their development. The urinary system develops ahead of the reproductive or genital systems. Intermediate mesenchyme is disposed longitudinally in the trunk, subjacent to the somites (in the folded embryo), at the junction between the splanchnopleuric mesenchyme (adjacent to the gut medially) and the somatopleuric mesenchyme (subjacent to the ectoderm laterally) (Fig. 109.1). In lower vertebrates, intermediate mesenchyme typically develops serial, segmental epithelial diverticuli termed nephrotomes. Each nephrotome encloses a cavity, the nephrocoele, which communicates with the coelom through a peritoneal funnel, the nephrostome (Fig. 109.2). The dorsal wall of a nephrotome evaginates as a nephric tubule. The dorsal tips of the cranial nephric tubules bend caudally and fuse to form a longitudinal primary excretory duct, which grows caudally and curves ventrally to open into the cloaca. The more caudally placed, and therefore chronologically later, tubules open secondarily into this duct or into tubular outgrowths from it. Glomeruli, specific arrangements of capillaries and overlying coelomic epithelium, arise from the ventral wall of the nephrocoele (internal glomeruli) or the roof of the coelom adjacent to the peritoneal funnels (coelomic or external glomeruli), or in both situations (Fig. 109.2).
Figure 109.1 A, Major epithelial populations within a stage 10 embryo, viewed from a ventrolateral position. B, Position of pronephros and mesonephros on the posterior thoracic and abdominal wall. C, Position of mesonephros and metanephros.
It has been customary to regard the renal excretory system as three organs, the pronephros, mesonephros and metanephros, succeeding each other in time and space, such that the last to develop is retained as the permanent kidney (Fig. 109.1, 109.2). However, it is difficult to provide reliable criteria by which to distinguish these stages or to define their precise limits in embryos.
PRONEPHROS The intermediate mesenchyme becomes visible in stage 10 embryos and can be distinguished as a nephrogenic cord when 10 somites are present. A pronephros is present in human embryos only as clusters of cells in the most cranial portions of the nephrogenic cord (Figs 109.1, 109.2). More caudally, similar groups of cells appear and become vesicular. The dorsal ends of the most caudal of the
vesicles join the primary excretory duct. Their central ends are connected with the coelomic epithelium by cellular strands, which probably represent rudimentary peritoneal funnels. Glomeruli do not develop in association with these cranially situated nephric tubules, which ultimately disappear. It is doubtful whether external glomeruli develop in human embryos. Primary excretory duct page 1373 page 1374
page 1374 page 1375
Figure 109.2 Principal features of the primitive vertebrate nephric system for comparison with the development of the human nephric system. A considerable period of embryonic and fetal life has necessarily been compressed into a single diagram. (Modified from Williams PL, Wendell-Smith CP, Treadgold S 1969 Basic Human Embryology, 2nd edn. Philadelphia: Lippincott.)
In stage 11 embryos of c.14 somites, the primary excretory duct can be seen as a solid rod of cells in the dorsal part of the nephrogenic cord. Its cranial end is about the level of the ninth somite and its caudal tip merges with the undifferentiated mesenchyme of the cord. It differentiates before any nephric tubules, and when the latter appear it is at first unconnected with them. In older embryos the duct has lengthened and its caudal end becomes detached from the nephrogenic cord to lie immediately beneath the ectoderm. From this level it grows caudally, independent of the nephrogenic mesenchyme, and then curves ventrally to reach the wall of the cloaca. It becomes canalized progressively from its caudal end to form a true duct, which opens into the cloaca in embryos at stage 12 (Fig. 109.1C). Clearly, up to this stage the name 'duct' is scarcely appropriate.
MESONEPHROS From stage 12 mesonephric tubules, which develop from the intermediate mesenchyme between somite levels 8-20, begin to connect to the primary excretory duct, which is now renamed the mesonephric duct. More caudally, a continuous ridge of nephrogenic mesenchyme extends to the level of somite 24. The mesonephric tubules (nephrons) are not metameric - there may be two or more mesonephric tubules opposite each somite. Within the mesonephros, each tubule first appears as a condensation of mesenchyme cells, which epithelialize and form a vesicle. One end of the vesicle grows towards and opens into the mesonephric duct, while the other dilates and invaginates. The outer stratum forms the glomerular capsule, while the inner cells differentiate into mesonephric podocytes, which clothe the invaginating capillaries to form a glomerulus. The capillaries are supplied with blood through lateral branches of the aorta. It has been estimated that 70-80 mesonephric tubules and a corresponding number of glomeruli develop. However, these tubules are not all present at the same time, it is rare to find more than 30-40 in an individual embryo, because the cranial tubules and glomeruli develop and atrophy before the development of those situated more caudally. By the end of the sixth week each mesonephros is an elongated, spindle-shaped organ that projects into the coelomic cavity, one on each side of the dorsal mesentery, from the level of the septum transversum to the third lumbar segment. This whole projection is called the mesonephric ridge, mesonephros, or Wolffian body (Fig. 109.1B, C). It develops subregions, and a gonad develops on its medial surface (p. 1381). There are striking similarities in structure between the mesonephros and the permanent kidney or metanephros, but the mesonephric nephrons lack a segment that corresponds to the descending limb of the loop of Henle. The mesonephros is believed to produce urine by stage 17. A detailed comparison of the development and function of the mesonephros and metanephros in staged human embryos is not available. In stage 18 embryos (13-17 mm) the mesonephric ridge extends cranially to about the level of rib 9. In both sexes the cranial end of the mesonephros atrophies, and in embryos 20 mm in length (stage 19) a mesonephros is found only in the first three lumbar segments, although it may still possess as many as 26 tubules. The most cranial one or two tubules persist as rostral aberrant ductules (Fig. 109.13); the succeeding five or six tubules develop into either the efferent ductules of the testis and lobules of the head of the epididymis (male), or the tubules of the epoöphoron (female); the caudal tubules form the caudal aberrant ductules and the paradidymis (male), or the paroöphoron (female) (p. 1384). Mesonephric duct
Once mesonephric nephrons connect to the primary excretory duct it is renamed the mesonephric duct. This runs caudally in the lateral part of the nephric ridge, and at the caudal end of the ridge it projects into the cavity of the coelom in the substance of a mesonephric fold (Fig. 109.3). As the mesonephric ducts from each side approach the urogenital sinus the two mesonephric folds fuse, between the bladder ventrally and the rectum dorsally, forming a transverse partition across the cavity of the pelvis, which is somewhat inappropriately called the genital cord (Fig. 109.3). In the male the peritoneal fossa between the bladder and the genital cord becomes obliterated, but it persists in the female as the
uterovesical pouch. The mesonephric duct itself becomes the canal of the epididymis, vas deferens and ejaculatory duct (p. 1382). Urogenital sinus
Figure 109.3 Arrangement of mesonephric duct from mesonephros to urogenital sinus. The duct runs within the tubal fold with the paramesonephric duct. For later development, see Fig. 109.14. (Redrawn from Tuchmann-Duplessis H, Haegel P 1972 Illustrated Human Embryology, Vol 2 Organogenesis. London: Chapman and Hall.)
The primitive hindgut ends in a cloacal region. This is connected ventrally with a blind-ending diverticulum, the allantois, which is intimately related to the development of the caudal portion of the urinary system. The enteric and allantoic portions of the hindgut are separated by the proliferation of the urorectal septum, a partition of mesenchyme and endoderm in the angle of the junction of hindgut and allantois (Fig. 109.4). The endodermal epithelium beneath the mesenchyme of the urorectal septum approaches and fuses with the cloacal membrane, thereby dividing the membrane into anal (dorsal) and urogenital (ventral) membranes, and the cloacal region into dorsal and ventral portions. The dorsal portion of the cloacal region is the putative rectum. The ventral portion can be further divided into: a cranial vesicourethral canal, continuous above with the allantoic duct; a middle, narrow channel, the pelvic portion; and a caudal, deep, phallic section, which is closed externally by the urogenital membrane. The second and third parts together constitute the urogenital sinus.
METANEPHROS The pronephros and mesonephros are linear structures. They both contain stacks of tubules distributed along the craniocaudal axis of the embryo, an arrangement that results in the production of hypotonic urine. In marked contrast, the tubules in the metanephric kidney are arranged concentrically, and the loops of Henle are directed towards the renal pelvis. This arrangement allows different concentration gradients to develop within the kidney and results in the production of hypertonic urine. Metanephric nephrons do not join with the existing mesonephric duct but with an evagination of that duct, which branches dichotomously to produce a characteristic pattern of collecting ducts. page 1375 page 1376
Figure 109.4 The division of the hindgut into urinary and enteric parts. Left ventrolateral view of the intraembryonic coelom and corresponding midsagittal sections. A, Early cloaca. B, Proliferation of urorectal septum. C, Complete separation of urethra and anal canal, and position of perineal body (also includes a sagittal section which permits a view into the bladder and rectum).
The metanephric kidney develops from three sources. An evagination of the mesonephric duct, the ureteric bud, and a local condensation of mesenchyme, the metanephric blastema, form the nephric structure (Fig. 109.5). Angiogenic mesenchyme migrates into the metanephric blastema slightly later to produce the glomeruli and vasa recta. It is possible that an intact nerve supply is also required for metanephric kidney induction. An epithelial/mesenchymal interaction between the duct system and the surrounding mesenchyme occurs in both mesonephric and metanephric systems. In the mesonephric kidney, development proceeds in a craniocaudal progression, and cranial nephrons degenerate before caudal ones are produced. In the metanephric kidney a proportion of the mesenchyme remains as stem cells that continue to divide and which enter the nephrogenic pathway later when the individual collecting ducts lengthen. The temporal development of the metanephric kidney is patterned radially, such that the outer cortex is the last part to be formed. The following interactions occur in the development of the metanephric
kidney (Fig. 109.5). The ureteric bud undergoes a series of bifurcations within the surrounding metanephric mesenchyme, and forms smaller ureteric ducts. At the same time the metanephric mesenchyme condenses around the dividing ducts to form S-shaped clusters, which transform into epithelia and fuse with the ureteric ducts at their distal ends. Blood vessels invade the proximal ends of the Sshaped clusters to form vascularized glomeruli. page 1376 page 1377
Figure 109.5 Overview of metanephric kidney development. A, The ureteric bud arises from the mesonephric duct. The metanephric mesenchyme proliferates and separates with each subdivision of the ureteric bud. B, The metanephric mesenchyme converts to epithelia, forms comma and S-shaped vesicles, which become metanephric nephrons. C, All stages of metanephric development are present concurrently. The most recently formed are on the outer aspect of the kidney.
The ureteric bud bifurcates when it comes into contact with the metanephric blastema in response to extracellular matrix molecules synthesized by the mesenchyme. Both chondroitin sulphate proteoglycan synthesis and chondroitin sulphate glycosaminoglycan processing are necessary for the dichotomous branching of the ureteric bud. In metanephric culture, incubation of fetal kidneys in !-D-xyloside, an inhibitor of chrondroitin sulphate synthesis, dramatically inhibits ureteric bud branching. Subsequent divisions of the ureteric bud and associated mesenchyme define the gross structure of the kidney and the major and minor calyces, the distal branches of the ureteric ducts that will form the collecting ducts of the kidney. As the collecting ducts elongate the metanephric mesenchyme condenses around them. An adhesion molecule, syndecan, can be detected between the mesenchymal cells in the condensate. The cells switch off expression of N-CAM (cell adhesion molecules), fibronectin and collagen I, and start to synthesize LCAM (also called E cadherin) and the basal lamina constituents laminin and collagen IV. The mesenchymal clusters are thus converted to small groups of epithelial cells, which undergo complex morphogenetic changes. Each epithelial group elongates, and forms first a comma-shaped, then an S-shaped, body, which continues to elongate and subsequently fuses with a branch of the ureteric duct at its distal end, while expanding as a dilated sac at its proximal end. The latter involutes, and cells differentiate locally such that the outer cells become the parietal glomerular cells, while the inner ones become visceral epithelial podocytes. The podocytes develop in close proximity to invading capillaries derived from angiogenic mesenchyme outside the nephrogenic mesenchyme. This third source of mesenchyme produces the endothelial and mesangial cells within the glomerulus. The (metanephric-derived) podocytes and the angiogenic mesenchyme produce fibronectin and other components of the glomerular basal lamina. The isoforms of type-IV collagen within this layer follow a specific programme of maturation as the filtration of macromolecules from the plasma becomes restricted. page 1377
page 1378
Platelet derived growth factor (PDGF) !-chain and the PDGF receptor !-subunit (PDGFR !) have been detected in developing human glomeruli between 54 and 109 days' gestation. PDGF !-chain is localized in the differentiating epithelium of the glomerular vesicle during its comma and S-shaped stages, while PDGFR ! is expressed in the undifferentiated metanephric blastema, vascular structures and interstitial cells. Both PDGF !-chain and PDGFR ! are expressed by mesangial cells, which may promote further mesangial cell proliferation. Metanephric mesenchyme will develop successfully in vitro, which makes experimental perturbation of kidney development comparatively easy to evaluate. Early experimental studies demonstrated that other mesenchymal populations, and spinal cord, were able to induce ureteric bud division and metanephric development. Nerves enter the developing kidney very early, travelling along the ureter. If developing kidney rudiments are incubated with antisense oligonucleotides, which neutralize nerve growth factor receptor (NGF-R) mRNA, nephrogenesis is completely blocked, suggesting that metanephric mesenchyme induction is a response to innervation. The powerful inductive effect of the spinal cord on metanephric mesenchyme may be a further expression of this phenomenon. All stages of nephron differentiation are present concurrently in the developing metanephric kidney (Fig. 109.5). Antigens for the brush border of the renal tubule appear when the S-shaped body has formed. They appear first in the inner cortical area. The metanephric kidney is lobulated throughout fetal life, but this condition usually disappears during the first year after birth (Fig. 109.8, p. 1378). Varying degrees of lobulation occasionally persist through-out life. The growth of left and right kidneys is well matched during development. Fetal kidney volume increases most during the second trimester in both sexes. For reasons that are not understood, male fetuses show greater values for renal volume than female fetuses from the third trimester onwards. Endocrine development of the kidney
The kidney functions not only as an excretory organ, but also as an endocrine organ, secreting hormones that are concerned with renal haemodynamics. Before birth homeostasis is controlled by the placenta. The fetal kidney produces amniotic fluid. The kidneys of premature babies of less than 36 weeks are immature. They contain incompletely differentiated cortical nephrons, which compromise their ability to maintain homeostasis. Problems of immaturity are further compounded by the effects of hypoxia and asphyxia, which modify renal hormones. Renal hormones include the renin-angiotensin system, renal prostaglandins, the kallikrein-kinin system, and renal dopamine. Renin is found in the smooth muscle cells of arterioles, interlobular arteries and branches of the renal artery, and has also been described in the distal convoluted tubule cells. Kallikrein has been demonstrated in rat fetal kidney, and prostaglandins have been demonstrated in the renal medulla and renal tubule. Renal dopamine is produced (mainly) by the enzymatic conversion of L-dopa to dopamine in the early segments of the proximal convoluted tubule, and is also sourced locally from dopaminergic nerves. Other renal hormones include an antihypertensive lipid, which is produced in the interstitial cells of the renal medulla, and, possibly, histamine and serotonin. Growth factors produced by human embryonic kidney cells include erythropoietin and interleukin ! (which stimulate megakaryocyte maturation) and transforming growth factor-!. Ascent of the kidney
The metanephric kidney is initially sacral. As the ureteric outgrowth lengthens, it becomes positioned more and more cranially. The metanephric pelvis lies on a level with the second lumbar vertebra when the embryo reaches a length of c.13 mm. During this period the ascending kidney receives its blood supply sequentially from arteries in its immediate neighbourhood, i.e. the middle sacral and common iliac arteries. The definitive renal artery is not recognizable until the beginning of the third month. It arises from the most caudal of the three suprarenal arteries, all of which represent persistent mesonephric or lateral splanchnic arteries. Additional renal arteries are relatively common, and may enter at the hilum or at the upper or lower pole of the gland - they also represent persistent mesonephric arteries.
Ureter
The wall of the ureter is initially highly permeable. Its lumen later becomes obliterated and is subsequently recanalized. Both of these processes begin in intermediate portions of the ureter and proceed cranially and caudally. Recanalization is not associated with metanephric function, but perhaps reflects the rapid elongation of the ureter as the embryo grows. Two fusiform enlargements appear at the lumbar and pelvic levels of the ureter at 5 and 9 months, respectively (the pelvic is inconstant). As a result the ureter shows a constriction at its proximal end (pelviureteric region) and another as it crosses the pelvic brim. A third narrowing is always present at its distal end and is related to the growth of the bladder wall. At first the distal end of the ureter is connected to the dorsomedial aspect of the mesonephric duct, but, as a result of differential growth, this connection comes to lie lateral to the duct. Urinary bladder
The urinary bladder develops from the cranial vesicourethral canal, which is continuous above with the allantoic duct (Figs 109.4, 109.6, 109.7). The mesonephric ducts open into the urogenital sinus early in development. The ureters develop as branches of the mesonephric ducts, which attain their own access to the developing bladder, and their orifices open separately into the bladder on the lateral side of the opening of the mesonephric ducts. Later the two orifices become separated still further and, although the ureter retains its point of entry into the bladder, the mesonephric duct opens into that part of the urogenital sinus that subsequently becomes the prostatic urethra (Fig. 109.6). The triangular region of absorption of the mesonephric ducts contributes to the trigone of the bladder and dorsal wall of the proximal half of the prostatic urethra, i.e. as far as the opening of the prostatic utricle and ejaculatory ducts, or its female homologue, the whole female urethral dorsal wall. The remainder of the vesicourethral canal forms the body of the bladder and urethra, and its apex is prolonged to the umbilicus as a narrow canal, the urachus. The fetal bladder can be identified by ultrasound examination at 9-11 weeks' gestation and the absence of a bladder image is considered abnormal at 13 weeks or later. UPDATE Date Added: 07 March 2006 Publication Services, Inc. Abstract: Fetal development of the female external urinary sphincter complex: An anatomical and histological study Click on the following line to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=15821572&query_hl=11&itool=pubmed_docsum Fetal development of the female external urinary sphincter complex: an anatomical and histological study. Sebe P, Fritsch H, Oswald J et al: Urol 173:1738-1742, 2005. UPDATE Date Added: 07 March 2006 Publication Services, Inc. Abstract: Fetal development of striated and smooth muscle sphincters of the male urethra from a common primordium and modifications due to the development of the prostate: An anatomic and histologic study Click on the following line to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=15821572&query_hl=10&itool=pubmed_docsum Fetal development of striated and smooth muscle sphincters of the male urethra from a common primordium and modifications due to the development of the prostate: an anatomic and histologic study. Sebe P, Schwentner C, Oswald J et al: Prostate 62:388-393, 2005. UPDATE Date Added: 21 February 2006 Publication Services, Inc. Abstract: The development of the external urethral sphincter in humans Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=11298059&query_hl=9&itool=pubmed_docsum The development of the external urethral sphincter in humans. Ludwikowski B, Oesch Hayward I, Brenner E, Fritsch H: BJU Int 87(6):565-568, 2001. UPDATE Date Added: 30 August 2005 Abstract Abstract: Innervation of the female human urethral sphincter: 3D reconstruction of immunohistochemical studies in the fetus. Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?
cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=15826754&query_hl=9 Innervation of the female human urethral sphincter: 3D reconstruction of immunohistochemical studies in the fetus. Karam I, Droupy S, Abd-Alsamad I: Urol. 47(5):627-33, 2005.
NEONATAL URINARY SYSTEM At birth the two kidneys weigh c.23 g. They function early in development and produce the amniotic fluid that surrounds the fetus. The lobulated appearance of fetal kidneys is still present at birth (Figs 11.4, 109.8). Addition of new cortical nephrons continues in the first few months of postnatal life after which general growth of the glomeruli and tubules results in the disappearance of lobulation. The renal blood flow is lower in the neonate - adult values are attained by the end of the first year. The glomerular filtration rate at birth is c.30% of the adult value, which is attained by 3-5 months of age. The neonatal urinary bladder is egg-shaped and the larger end is directed downwards and backwards (Figs 11.4, 11.5, 109.9, 109.10). Although described as an abdominal organ, nearly one half of the neonatal bladder lies below a line drawn from the promontory of the sacrum to the upper edge of the pubic symphysis, i.e. within the cavity of the true pelvis. From the bladder neck, the bladder extends anteriorly and slightly upwards in close contact with the pubis until it reaches the anterior abdominal wall. The apex of the contracted bladder lies at a point midway between the pubis and the umbilicus. When the bladder is filled with urine the apex may extend up to the level of the umbilicus. It is therefore possible to obtain urine by inserting a needle, connected to a syringe, into the bladder through the abdominal wall c.2 cm above the symphysis pubis and aspirating the contents into the sterile syringe. The success rate of the procedure is variable and depends upon the bladder being full. Recently, a much higher success rate has been reported by using an ultrasound scanner to locate the bladder and confirm that it contains urine prior to the insertion of the needle. There is no true fundus in the fetal bladder as there is in the adult. Although the anterior surface is not covered with peritoneum, peritoneum extends posteriorly as low as the level of the urethral orifice. Because the apex of the bladder is relatively high, pressure on the lower abdominal wall will express urine from an infant bladder. Moreover, because the bladder remains connected to the umbilicus by the obliterated remains of the urachus (Fig. 65.5), stimulation of the umbilicus can initiate micturition in babies. The elongated shape of the bladder in neonates means that the ureters are correspondingly reduced in length and they lack a pelvic portion. The bladder does not gain its adult, pelvic, position until about the sixth year. A distinct interureteric fold is present in the contracted neonatal bladder. page 1378 page 1379
Figure 109.6 Development of the urinary part of the urogenital sinus and formation of the trigone of the bladder. A-C and E, Posterior views. D, Male and female, midsagittal sections. (Redrawn from Tuchmann-Duplessis H, Haegel P 1972 Illustrated Human Embryology, Vol 2 Organogenesis. London: Chapman and Hall.)
ANOMALIES OF THE URINARY SYSTEM Anomalies of the urinary system are relatively common (3% of live births). Renal agenesis is the absence of one or both kidneys. In unilateral renal agenesis, the remaining kidney exhibits compensatory hypertrophy and produces a nearly normal functional mass of renal tissue. Problems with kidney ascent can result in a pelvic kidney. Alternatively, the kidneys may fuse together at their caudal poles producing a horseshoe kidney, which cannot ascend out of the pelvic cavity because the inferior mesenteric artery prevents further migration. It was thought that renal cysts arose from clumps of vesicular cells, which persisted when the tips of branches from the ureteric diverticulum failed to fuse with metanephrogenic cap tissue. It is now believed that they are wide dilatations of a part of otherwise continuous nephrons. In most cases, autosomal dominant polycystic kidney disease results from mutations of PKD1 or PKD2 genes which are expressed in human embryos from 5-6 weeks of development within the mesonephros and later the metanephros (Chauvet et al 2002). In this condition the cystic dilatation may affect any part of the nephron, from Bowman's capsule to collecting tubules. Less common is infantile cystic renal disease, inherited as a recessive trait, where the proximal and distal tubules are dilated to some degree but the collecting ducts are grossly affected. Abnormalities of the ventral body wall caudal to the umbilicus, especially with inappropriate siting of the genital tubercle (p. 1393) can result in exstrophy of the bladder (Fig. 109.11). In this condition the urorectal septum (internal) is
associated with the genital tubercle (external), which means that the urogenital and anal membranes are widely separated. When the urogenital membrane involutes, the posterior surface of the bladder is exposed to the anterior abdominal wall. The lower part of the abdominal wall is therefore occupied by an irregularly oval area, covered with mucous membrane, on which the two ureters open. The periphery of this extroverted area, which is covered by urothelium, becomes continuous with the skin (Fig. 109.19). The routine use of ultrasound as an aid to in-utero diagnosis of abnormalities has revealed a prevalence of 1-2 abnormal fetuses per 1000 ultrasound procedures. Of these, 20-30% are anomalies of the genitourinary tract, and can be detected as early as 12-15 weeks' gestation. However, the decision to be made after such a diagnosis is by no means clear. Urinary obstruction is considered an abnormality, yet transient modest obstruction is considered normal during canalization of the urinary tract, and has been reported in 10-20% of fetuses in the third trimester. A delay in canalization, or in the rupture of the cloacal membrane, can produce a dilatation. Similarly, the closure of the urachus at 32 weeks may be associated with high-resistance outflow for the system, which again produces transient obstruction. The degree to which obstruction may cause renal parenchymal damage cannot be assessed in a developing kidney, which may have primary nephrogenic dysgenesis. page 1379 page 1380
page 1380 page 1381
Figure 109.7 A, The caudal end of a human embryo, c.4 weeks, showing the left lateral aspects of the spinal cord, notochord and endodermal cloaca. B, The endodermal cloaca of a human embryo, near the end of the fifth week. Part of the left wall of the cloaca, including the left mesonephric duct, has been removed, together with the adjoining portions of the walls of the developing bladder and rectum. A piece of the ectoderm around the cloacal membrane has been left in situ. A wire is shown passing along the right mesonephric duct into the cloaca. C, The caudal end of a human embryo, c.5 weeks, showing the endodermal cloaca. D, The caudal end of a human embryo, c.6 weeks. The cloaca is becoming divided by the urorectal septum. E, The caudal end of a female human fetus, 8"-9 weeks, from the left-hand side showing structures in and near the median plane. The cloaca is now completely divided into urogenital and intestinal segments. F, Part of the vesicourethral portion of the endodermal cloaca of a female human fetus, 8"-9 weeks. The sinus tubercle is the elevation on the posterior wall of the urogenital sinus, caused by the fusion with
the paramesonephric ducts.
Figure 109.8 Posterior abdominal wall of a full-term neonate. Note the lobulated kidneys and relatively wide calibre of the ureters. (After Crelin ES 1969 Anatomy of the Newborn. Philadelphia: Lea and Febiger.)
The volume of amniotic fluid is used as an indicator of renal function, but, because other sources produce amniotic fluid in early gestation, amniotic volume does not reflect fetal urinary output until the second trimester. Too little amniotic fluid is termed oligohydramnios, too much, hydramnios. Although variation in the amount of amniotic fluid may suggest abnormalities of either the gut or kidneys, it is not always possible to correlate even severe oligohydramnios with renal dysfunction. There is an important relationship between the volume of amniotic fluid, lung development and maturity, and oligohydramnios has been shown to be associated with pulmonary hypoplasia.
© 2008 Elsevier
URINARY SYSTEM The urinary and reproductive systems develop from intermediate mesenchyme and are intimately associated with one another especially in the earlier stages of their development. The urinary system develops ahead of the reproductive or genital systems. Intermediate mesenchyme is disposed longitudinally in the trunk, subjacent to the somites (in the folded embryo), at the junction between the splanchnopleuric mesenchyme (adjacent to the gut medially) and the somatopleuric mesenchyme (subjacent to the ectoderm laterally) (Fig. 109.1). In lower vertebrates, intermediate mesenchyme typically develops serial, segmental epithelial diverticuli termed nephrotomes. Each nephrotome encloses a cavity, the nephrocoele, which communicates with the coelom through a peritoneal funnel, the nephrostome (Fig. 109.2). The dorsal wall of a nephrotome evaginates as a nephric tubule. The dorsal tips of the cranial nephric tubules bend caudally and fuse to form a longitudinal primary excretory duct, which grows caudally and curves ventrally to open into the cloaca. The more caudally placed, and therefore chronologically later, tubules open secondarily into this duct or into tubular outgrowths from it. Glomeruli, specific arrangements of capillaries and overlying coelomic epithelium, arise from the ventral wall of the nephrocoele (internal glomeruli) or the roof of the coelom adjacent to the peritoneal funnels (coelomic or external glomeruli), or in both situations (Fig. 109.2).
Figure 109.1 A, Major epithelial populations within a stage 10 embryo, viewed from a ventrolateral position. B, Position of pronephros and mesonephros on the posterior thoracic and abdominal wall. C, Position of mesonephros and metanephros.
It has been customary to regard the renal excretory system as three organs, the pronephros, mesonephros and metanephros, succeeding each other in time and space, such that the last to develop is retained as the permanent kidney (Fig. 109.1, 109.2). However, it is difficult to provide reliable criteria by which to distinguish these stages or to define their precise limits in embryos.
PRONEPHROS The intermediate mesenchyme becomes visible in stage 10 embryos and can be distinguished as a nephrogenic cord when 10 somites are present. A pronephros is present in human embryos only as clusters of cells in the most cranial portions of the nephrogenic cord (Figs 109.1, 109.2). More caudally, similar groups of cells appear and become vesicular. The dorsal ends of the most caudal of the vesicles join the primary excretory duct. Their central ends are connected with the coelomic epithelium by cellular strands, which probably represent rudimentary peritoneal funnels. Glomeruli do not develop in association with these cranially
situated nephric tubules, which ultimately disappear. It is doubtful whether external glomeruli develop in human embryos. Primary excretory duct page 1373 page 1374
page 1374 page 1375
Figure 109.2 Principal features of the primitive vertebrate nephric system for comparison with the development of the human nephric system. A considerable period of embryonic and fetal life has necessarily been compressed into a single diagram.
(Modified from Williams PL, Wendell-Smith CP, Treadgold S 1969 Basic Human Embryology, 2nd edn. Philadelphia: Lippincott.)
In stage 11 embryos of c.14 somites, the primary excretory duct can be seen as a solid rod of cells in the dorsal part of the nephrogenic cord. Its cranial end is about the level of the ninth somite and its caudal tip merges with the undifferentiated mesenchyme of the cord. It differentiates before any nephric tubules, and when the latter appear it is at first unconnected with them. In older embryos the duct has lengthened and its caudal end becomes detached from the nephrogenic cord to lie immediately beneath the ectoderm. From this level it grows caudally, independent of the nephrogenic mesenchyme, and then curves ventrally to reach the wall of the cloaca. It becomes canalized progressively from its caudal end to form a true duct, which opens into the cloaca in embryos at stage 12 (Fig. 109.1C). Clearly, up to this stage the name 'duct' is scarcely appropriate.
MESONEPHROS From stage 12 mesonephric tubules, which develop from the intermediate mesenchyme between somite levels 8-20, begin to connect to the primary excretory duct, which is now renamed the mesonephric duct. More caudally, a continuous ridge of nephrogenic mesenchyme extends to the level of somite 24. The mesonephric tubules (nephrons) are not metameric - there may be two or more mesonephric tubules opposite each somite. Within the mesonephros, each tubule first appears as a condensation of mesenchyme cells, which epithelialize and form a vesicle. One end of the vesicle grows towards and opens into the mesonephric duct, while the other dilates and invaginates. The outer stratum forms the glomerular capsule, while the inner cells differentiate into mesonephric podocytes, which clothe the invaginating capillaries to form a glomerulus. The capillaries are supplied with blood through lateral branches of the aorta. It has been estimated that 70-80 mesonephric tubules and a corresponding number of glomeruli develop. However, these tubules are not all present at the same time, it is rare to find more than 30-40 in an individual embryo, because the cranial tubules and glomeruli develop and atrophy before the development of those situated more caudally. By the end of the sixth week each mesonephros is an elongated, spindle-shaped organ that projects into the coelomic cavity, one on each side of the dorsal mesentery, from the level of the septum transversum to the third lumbar segment. This whole projection is called the mesonephric ridge, mesonephros, or Wolffian body (Fig. 109.1B, C). It develops subregions, and a gonad develops on its medial surface (p. 1381). There are striking similarities in structure between the mesonephros and the permanent kidney or metanephros, but the mesonephric nephrons lack a segment that corresponds to the descending limb of the loop of Henle. The mesonephros is believed to produce urine by stage 17. A detailed comparison of the development and function of the mesonephros and metanephros in staged human embryos is not available. In stage 18 embryos (13-17 mm) the mesonephric ridge extends cranially to about the level of rib 9. In both sexes the cranial end of the mesonephros atrophies, and in embryos 20 mm in length (stage 19) a mesonephros is found only in the first three lumbar segments, although it may still possess as many as 26 tubules. The most cranial one or two tubules persist as rostral aberrant ductules (Fig. 109.13); the succeeding five or six tubules develop into either the efferent ductules of the testis and lobules of the head of the epididymis (male), or the tubules of the epoöphoron (female); the caudal tubules form the caudal aberrant ductules and the paradidymis (male), or the paroöphoron (female) (p. 1384). Mesonephric duct
Once mesonephric nephrons connect to the primary excretory duct it is renamed the mesonephric duct. This runs caudally in the lateral part of the nephric ridge, and at the caudal end of the ridge it projects into the cavity of the coelom in the substance of a mesonephric fold (Fig. 109.3). As the mesonephric ducts from each side approach the urogenital sinus the two mesonephric folds fuse, between the bladder ventrally and the rectum dorsally, forming a transverse partition across the cavity of the pelvis, which is somewhat inappropriately called the genital cord (Fig. 109.3). In the male the peritoneal fossa between the bladder and the genital cord becomes obliterated, but it persists in the female as the uterovesical pouch. The mesonephric duct itself becomes the canal of the epididymis, vas deferens and ejaculatory duct (p. 1382). Urogenital sinus
Figure 109.3 Arrangement of mesonephric duct from mesonephros to urogenital sinus. The duct runs within the tubal fold with the paramesonephric duct. For later development, see Fig. 109.14. (Redrawn from Tuchmann-Duplessis H, Haegel P 1972 Illustrated Human Embryology, Vol 2 Organogenesis. London: Chapman and Hall.)
The primitive hindgut ends in a cloacal region. This is connected ventrally with a blind-ending diverticulum, the allantois, which is intimately related to the development of the caudal portion of the urinary system. The enteric and allantoic portions of the hindgut are separated by the proliferation of the urorectal septum, a partition of mesenchyme and endoderm in the angle of the junction of hindgut and allantois (Fig. 109.4). The endodermal epithelium beneath the mesenchyme of the urorectal septum approaches and fuses with the cloacal membrane, thereby dividing the membrane into anal (dorsal) and urogenital (ventral) membranes, and the cloacal region into dorsal and ventral portions. The dorsal portion of the cloacal region is the putative rectum. The ventral portion can be further divided into: a cranial vesicourethral canal, continuous above with the allantoic duct; a middle, narrow channel, the pelvic portion; and a caudal, deep, phallic section, which is closed externally by the urogenital membrane. The second and third parts together constitute the urogenital sinus.
METANEPHROS The pronephros and mesonephros are linear structures. They both contain stacks of tubules distributed along the craniocaudal axis of the embryo, an arrangement that results in the production of hypotonic urine. In marked contrast, the tubules in the metanephric kidney are arranged concentrically, and the loops of Henle are directed towards the renal pelvis. This arrangement allows different concentration gradients to develop within the kidney and results in the production of hypertonic urine. Metanephric nephrons do not join with the existing mesonephric duct but with an evagination of that duct, which branches dichotomously to produce a characteristic pattern of collecting ducts. page 1375 page 1376
Figure 109.4 The division of the hindgut into urinary and enteric parts. Left ventrolateral view of the intraembryonic coelom and corresponding midsagittal sections. A, Early cloaca. B, Proliferation of urorectal septum. C, Complete separation of urethra and anal canal, and position of perineal body (also includes a sagittal section which permits a view into the bladder and rectum).
The metanephric kidney develops from three sources. An evagination of the mesonephric duct, the ureteric bud, and a local condensation of mesenchyme, the metanephric blastema, form the nephric structure (Fig. 109.5). Angiogenic mesenchyme migrates into the metanephric blastema slightly later to produce the glomeruli and vasa recta. It is possible that an intact nerve supply is also required for metanephric kidney induction. An epithelial/mesenchymal interaction between the duct system and the surrounding mesenchyme occurs in both mesonephric and metanephric systems. In the mesonephric kidney, development proceeds in a craniocaudal progression, and cranial nephrons degenerate before caudal ones are produced. In the metanephric kidney a proportion of the mesenchyme remains as stem cells that continue to divide and which enter the nephrogenic pathway later when the individual collecting ducts lengthen. The temporal development of the metanephric kidney is patterned radially, such that the outer cortex is the last part to be formed. The following interactions occur in the development of the metanephric
kidney (Fig. 109.5). The ureteric bud undergoes a series of bifurcations within the surrounding metanephric mesenchyme, and forms smaller ureteric ducts. At the same time the metanephric mesenchyme condenses around the dividing ducts to form S-shaped clusters, which transform into epithelia and fuse with the ureteric ducts at their distal ends. Blood vessels invade the proximal ends of the Sshaped clusters to form vascularized glomeruli. page 1376 page 1377
Figure 109.5 Overview of metanephric kidney development. A, The ureteric bud arises from the mesonephric duct. The metanephric mesenchyme proliferates and separates with each subdivision of the ureteric bud. B, The metanephric mesenchyme converts to epithelia, forms comma and S-shaped vesicles, which become metanephric nephrons. C, All stages of metanephric development are present concurrently. The most recently formed are on the outer aspect of the kidney.
The ureteric bud bifurcates when it comes into contact with the metanephric blastema in response to extracellular matrix molecules synthesized by the mesenchyme. Both chondroitin sulphate proteoglycan synthesis and chondroitin sulphate glycosaminoglycan processing are necessary for the dichotomous branching of the ureteric bud. In metanephric culture, incubation of fetal kidneys in !-D-xyloside, an inhibitor of chrondroitin sulphate synthesis, dramatically inhibits ureteric bud branching. Subsequent divisions of the ureteric bud and associated mesenchyme define the gross structure of the kidney and the major and minor calyces, the distal branches of the ureteric ducts that will form the collecting ducts of the kidney. As the collecting ducts elongate the metanephric mesenchyme condenses around them. An adhesion molecule, syndecan, can be detected between the mesenchymal cells in the condensate. The cells switch off expression of N-CAM (cell adhesion molecules), fibronectin and collagen I, and start to synthesize LCAM (also called E cadherin) and the basal lamina constituents laminin and collagen IV. The mesenchymal clusters are thus converted to small groups of epithelial cells, which undergo complex morphogenetic changes. Each epithelial group elongates, and forms first a comma-shaped, then an S-shaped, body, which continues to elongate and subsequently fuses with a branch of the ureteric duct at its distal end, while expanding as a dilated sac at its proximal end. The latter involutes, and cells differentiate locally such that the outer cells become the parietal glomerular cells, while the inner ones become visceral epithelial podocytes. The podocytes develop in close proximity to invading capillaries derived from angiogenic mesenchyme outside the nephrogenic mesenchyme. This third source of mesenchyme produces the endothelial and mesangial cells within the glomerulus. The (metanephric-derived) podocytes and the angiogenic mesenchyme produce fibronectin and other components of the glomerular basal lamina. The isoforms of type-IV collagen within this layer follow a specific programme of maturation as the filtration of macromolecules from the plasma becomes restricted. page 1377
page 1378
Platelet derived growth factor (PDGF) !-chain and the PDGF receptor !-subunit (PDGFR !) have been detected in developing human glomeruli between 54 and 109 days' gestation. PDGF !-chain is localized in the differentiating epithelium of the glomerular vesicle during its comma and S-shaped stages, while PDGFR ! is expressed in the undifferentiated metanephric blastema, vascular structures and interstitial cells. Both PDGF !-chain and PDGFR ! are expressed by mesangial cells, which may promote further mesangial cell proliferation. Metanephric mesenchyme will develop successfully in vitro, which makes experimental perturbation of kidney development comparatively easy to evaluate. Early experimental studies demonstrated that other mesenchymal populations, and spinal cord, were able to induce ureteric bud division and metanephric development. Nerves enter the developing kidney very early, travelling along the ureter. If developing kidney rudiments are incubated with antisense oligonucleotides, which neutralize nerve growth factor receptor (NGF-R) mRNA, nephrogenesis is completely blocked, suggesting that metanephric mesenchyme induction is a response to innervation. The powerful inductive effect of the spinal cord on metanephric mesenchyme may be a further expression of this phenomenon. All stages of nephron differentiation are present concurrently in the developing metanephric kidney (Fig. 109.5). Antigens for the brush border of the renal tubule appear when the S-shaped body has formed. They appear first in the inner cortical area. The metanephric kidney is lobulated throughout fetal life, but this condition usually disappears during the first year after birth (Fig. 109.8, p. 1378). Varying degrees of lobulation occasionally persist through-out life. The growth of left and right kidneys is well matched during development. Fetal kidney volume increases most during the second trimester in both sexes. For reasons that are not understood, male fetuses show greater values for renal volume than female fetuses from the third trimester onwards. Endocrine development of the kidney
The kidney functions not only as an excretory organ, but also as an endocrine organ, secreting hormones that are concerned with renal haemodynamics. Before birth homeostasis is controlled by the placenta. The fetal kidney produces amniotic fluid. The kidneys of premature babies of less than 36 weeks are immature. They contain incompletely differentiated cortical nephrons, which compromise their ability to maintain homeostasis. Problems of immaturity are further compounded by the effects of hypoxia and asphyxia, which modify renal hormones. Renal hormones include the renin-angiotensin system, renal prostaglandins, the kallikrein-kinin system, and renal dopamine. Renin is found in the smooth muscle cells of arterioles, interlobular arteries and branches of the renal artery, and has also been described in the distal convoluted tubule cells. Kallikrein has been demonstrated in rat fetal kidney, and prostaglandins have been demonstrated in the renal medulla and renal tubule. Renal dopamine is produced (mainly) by the enzymatic conversion of L-dopa to dopamine in the early segments of the proximal convoluted tubule, and is also sourced locally from dopaminergic nerves. Other renal hormones include an antihypertensive lipid, which is produced in the interstitial cells of the renal medulla, and, possibly, histamine and serotonin. Growth factors produced by human embryonic kidney cells include erythropoietin and interleukin ! (which stimulate megakaryocyte maturation) and transforming growth factor-!. Ascent of the kidney
The metanephric kidney is initially sacral. As the ureteric outgrowth lengthens, it becomes positioned more and more cranially. The metanephric pelvis lies on a level with the second lumbar vertebra when the embryo reaches a length of c.13 mm. During this period the ascending kidney receives its blood supply sequentially from arteries in its immediate neighbourhood, i.e. the middle sacral and common iliac arteries. The definitive renal artery is not recognizable until the beginning of the third month. It arises from the most caudal of the three suprarenal arteries, all of which represent persistent mesonephric or lateral splanchnic arteries. Additional renal arteries are relatively common, and may enter at the hilum or at the upper or lower pole of the gland - they also represent persistent mesonephric arteries.
Ureter
The wall of the ureter is initially highly permeable. Its lumen later becomes obliterated and is subsequently recanalized. Both of these processes begin in intermediate portions of the ureter and proceed cranially and caudally. Recanalization is not associated with metanephric function, but perhaps reflects the rapid elongation of the ureter as the embryo grows. Two fusiform enlargements appear at the lumbar and pelvic levels of the ureter at 5 and 9 months, respectively (the pelvic is inconstant). As a result the ureter shows a constriction at its proximal end (pelviureteric region) and another as it crosses the pelvic brim. A third narrowing is always present at its distal end and is related to the growth of the bladder wall. At first the distal end of the ureter is connected to the dorsomedial aspect of the mesonephric duct, but, as a result of differential growth, this connection comes to lie lateral to the duct. Urinary bladder
The urinary bladder develops from the cranial vesicourethral canal, which is continuous above with the allantoic duct (Figs 109.4, 109.6, 109.7). The mesonephric ducts open into the urogenital sinus early in development. The ureters develop as branches of the mesonephric ducts, which attain their own access to the developing bladder, and their orifices open separately into the bladder on the lateral side of the opening of the mesonephric ducts. Later the two orifices become separated still further and, although the ureter retains its point of entry into the bladder, the mesonephric duct opens into that part of the urogenital sinus that subsequently becomes the prostatic urethra (Fig. 109.6). The triangular region of absorption of the mesonephric ducts contributes to the trigone of the bladder and dorsal wall of the proximal half of the prostatic urethra, i.e. as far as the opening of the prostatic utricle and ejaculatory ducts, or its female homologue, the whole female urethral dorsal wall. The remainder of the vesicourethral canal forms the body of the bladder and urethra, and its apex is prolonged to the umbilicus as a narrow canal, the urachus. The fetal bladder can be identified by ultrasound examination at 9-11 weeks' gestation and the absence of a bladder image is considered abnormal at 13 weeks or later. UPDATE Date Added: 07 March 2006 Publication Services, Inc. Abstract: Fetal development of the female external urinary sphincter complex: An anatomical and histological study Click on the following line to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=15821572&query_hl=11&itool=pubmed_docsum Fetal development of the female external urinary sphincter complex: an anatomical and histological study. Sebe P, Fritsch H, Oswald J et al: Urol 173:1738-1742, 2005. UPDATE Date Added: 07 March 2006 Publication Services, Inc. Abstract: Fetal development of striated and smooth muscle sphincters of the male urethra from a common primordium and modifications due to the development of the prostate: An anatomic and histologic study Click on the following line to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=15821572&query_hl=10&itool=pubmed_docsum Fetal development of striated and smooth muscle sphincters of the male urethra from a common primordium and modifications due to the development of the prostate: an anatomic and histologic study. Sebe P, Schwentner C, Oswald J et al: Prostate 62:388-393, 2005. UPDATE Date Added: 21 February 2006 Publication Services, Inc. Abstract: The development of the external urethral sphincter in humans Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=11298059&query_hl=9&itool=pubmed_docsum The development of the external urethral sphincter in humans. Ludwikowski B, Oesch Hayward I, Brenner E, Fritsch H: BJU Int 87(6):565-568, 2001. UPDATE Date Added: 30 August 2005 Abstract Abstract: Innervation of the female human urethral sphincter: 3D reconstruction of immunohistochemical studies in the fetus. Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?
cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=15826754&query_hl=9 Innervation of the female human urethral sphincter: 3D reconstruction of immunohistochemical studies in the fetus. Karam I, Droupy S, Abd-Alsamad I: Urol. 47(5):627-33, 2005.
NEONATAL URINARY SYSTEM At birth the two kidneys weigh c.23 g. They function early in development and produce the amniotic fluid that surrounds the fetus. The lobulated appearance of fetal kidneys is still present at birth (Figs 11.4, 109.8). Addition of new cortical nephrons continues in the first few months of postnatal life after which general growth of the glomeruli and tubules results in the disappearance of lobulation. The renal blood flow is lower in the neonate - adult values are attained by the end of the first year. The glomerular filtration rate at birth is c.30% of the adult value, which is attained by 3-5 months of age. The neonatal urinary bladder is egg-shaped and the larger end is directed downwards and backwards (Figs 11.4, 11.5, 109.9, 109.10). Although described as an abdominal organ, nearly one half of the neonatal bladder lies below a line drawn from the promontory of the sacrum to the upper edge of the pubic symphysis, i.e. within the cavity of the true pelvis. From the bladder neck, the bladder extends anteriorly and slightly upwards in close contact with the pubis until it reaches the anterior abdominal wall. The apex of the contracted bladder lies at a point midway between the pubis and the umbilicus. When the bladder is filled with urine the apex may extend up to the level of the umbilicus. It is therefore possible to obtain urine by inserting a needle, connected to a syringe, into the bladder through the abdominal wall c.2 cm above the symphysis pubis and aspirating the contents into the sterile syringe. The success rate of the procedure is variable and depends upon the bladder being full. Recently, a much higher success rate has been reported by using an ultrasound scanner to locate the bladder and confirm that it contains urine prior to the insertion of the needle. There is no true fundus in the fetal bladder as there is in the adult. Although the anterior surface is not covered with peritoneum, peritoneum extends posteriorly as low as the level of the urethral orifice. Because the apex of the bladder is relatively high, pressure on the lower abdominal wall will express urine from an infant bladder. Moreover, because the bladder remains connected to the umbilicus by the obliterated remains of the urachus (Fig. 65.5), stimulation of the umbilicus can initiate micturition in babies. The elongated shape of the bladder in neonates means that the ureters are correspondingly reduced in length and they lack a pelvic portion. The bladder does not gain its adult, pelvic, position until about the sixth year. A distinct interureteric fold is present in the contracted neonatal bladder. page 1378 page 1379
Figure 109.6 Development of the urinary part of the urogenital sinus and formation of the trigone of the bladder. A-C and E, Posterior views. D, Male and female, midsagittal sections. (Redrawn from Tuchmann-Duplessis H, Haegel P 1972 Illustrated Human Embryology, Vol 2 Organogenesis. London: Chapman and Hall.)
ANOMALIES OF THE URINARY SYSTEM Anomalies of the urinary system are relatively common (3% of live births). Renal agenesis is the absence of one or both kidneys. In unilateral renal agenesis, the remaining kidney exhibits compensatory hypertrophy and produces a nearly normal functional mass of renal tissue. Problems with kidney ascent can result in a pelvic kidney. Alternatively, the kidneys may fuse together at their caudal poles producing a horseshoe kidney, which cannot ascend out of the pelvic cavity because the inferior mesenteric artery prevents further migration. It was thought that renal cysts arose from clumps of vesicular cells, which persisted when the tips of branches from the ureteric diverticulum failed to fuse with metanephrogenic cap tissue. It is now believed that they are wide dilatations of a part of otherwise continuous nephrons. In most cases, autosomal dominant polycystic kidney disease results from mutations of PKD1 or PKD2 genes which are expressed in human embryos from 5-6 weeks of development within the mesonephros and later the metanephros (Chauvet et al 2002). In this condition the cystic dilatation may affect any part of the nephron, from Bowman's capsule to collecting tubules. Less common is infantile cystic renal disease, inherited as a recessive trait, where the proximal and distal tubules are dilated to some degree but the collecting ducts are grossly affected. Abnormalities of the ventral body wall caudal to the umbilicus, especially with inappropriate siting of the genital tubercle (p. 1393) can result in exstrophy of the bladder (Fig. 109.11). In this condition the urorectal septum (internal) is
associated with the genital tubercle (external), which means that the urogenital and anal membranes are widely separated. When the urogenital membrane involutes, the posterior surface of the bladder is exposed to the anterior abdominal wall. The lower part of the abdominal wall is therefore occupied by an irregularly oval area, covered with mucous membrane, on which the two ureters open. The periphery of this extroverted area, which is covered by urothelium, becomes continuous with the skin (Fig. 109.19). The routine use of ultrasound as an aid to in-utero diagnosis of abnormalities has revealed a prevalence of 1-2 abnormal fetuses per 1000 ultrasound procedures. Of these, 20-30% are anomalies of the genitourinary tract, and can be detected as early as 12-15 weeks' gestation. However, the decision to be made after such a diagnosis is by no means clear. Urinary obstruction is considered an abnormality, yet transient modest obstruction is considered normal during canalization of the urinary tract, and has been reported in 10-20% of fetuses in the third trimester. A delay in canalization, or in the rupture of the cloacal membrane, can produce a dilatation. Similarly, the closure of the urachus at 32 weeks may be associated with high-resistance outflow for the system, which again produces transient obstruction. The degree to which obstruction may cause renal parenchymal damage cannot be assessed in a developing kidney, which may have primary nephrogenic dysgenesis. page 1379 page 1380
page 1380 page 1381
Figure 109.7 A, The caudal end of a human embryo, c.4 weeks, showing the left lateral aspects of the spinal cord, notochord and endodermal cloaca. B, The endodermal cloaca of a human embryo, near the end of the fifth week. Part of the left wall of the cloaca, including the left mesonephric duct, has been removed, together with the adjoining portions of the walls of the developing bladder and rectum. A piece of the ectoderm around the cloacal membrane has been left in situ. A wire is shown passing along the right mesonephric duct into the cloaca. C, The caudal end of a human embryo, c.5 weeks, showing the endodermal cloaca. D, The caudal end of a human embryo, c.6 weeks. The cloaca is becoming divided by the urorectal septum. E, The caudal end of a female human fetus, 8"-9 weeks, from the left-hand side showing structures in and near the median plane. The cloaca is now completely divided into urogenital and intestinal segments. F, Part of the vesicourethral portion of the endodermal cloaca of a female human fetus, 8"-9 weeks. The sinus tubercle is the elevation on the posterior wall of the urogenital sinus, caused by the fusion with
the paramesonephric ducts.
Figure 109.8 Posterior abdominal wall of a full-term neonate. Note the lobulated kidneys and relatively wide calibre of the ureters. (After Crelin ES 1969 Anatomy of the Newborn. Philadelphia: Lea and Febiger.)
The volume of amniotic fluid is used as an indicator of renal function, but, because other sources produce amniotic fluid in early gestation, amniotic volume does not reflect fetal urinary output until the second trimester. Too little amniotic fluid is termed oligohydramnios, too much, hydramnios. Although variation in the amount of amniotic fluid may suggest abnormalities of either the gut or kidneys, it is not always possible to correlate even severe oligohydramnios with renal dysfunction. There is an important relationship between the volume of amniotic fluid, lung development and maturity, and oligohydramnios has been shown to be associated with pulmonary hypoplasia.
© 2008 Elsevier
REPRODUCTIVE SYSTEM Development of the reproductive organs from the intermediate mesenchyme starts from stage 14, c.10 days later than the urinary system. Bilateral paramesonephric (Müllerian) ducts develop alongside the mesonephric ducts, and the midportion of each mesonephros undergoes thickening to form the gonadal ridge. Although the primordial germ cells are delineated very early in development, they are sequestered in the extraembryonic tissues until the gonadal ridge is ready to receive them. It was thought that development to one or other sexual phenotype occurred after migration of the primordial germ cells to the indifferent gonads. However, it is now recognized that the development of male or female gonads, genital ducts and external genitalia is far more complicated, and is the result of a complex interplay between genetic expression, timing of development and the influence of sex hormones. As development proceeds, a significant proportion of early embryonic urinary tissue is incorporated into the reproductive tracts, especially in the male. The earliest stage of reproductive development, prior to the arrival of the primordial germ cells into the gonad, is termed the indifferent or ambisexual stage.
EARLY GONADAL DEVELOPMENT (AMBISEXUAL OR INDIFFERENT STAGE) Essentially four different cell lineages contribute to the gonads. Cells are derived from: proliferating coelomic epithelium on the medial side of the mesonephros; underlying mesonephric mesenchyme; invading angiogenic mesenchyme already present in the mesonephros; and primordial germ cells that arise from the epiblast very early in development and later migrate from the allantoic wall.
page 1381 page 1382
Figure 109.9 Midsagittal section through the pelvis of a full-term female neonate. Note the abdominal position of the urinary bladder and uterus. (After Crelin ES 1969 Anatomy of the Newborn. Philadelphia: Lea and Febiger.)
Figure 109.10 Midsagittal section through the pelvis of a full-term male neonate. Note the abdominal position of the urinary bladder. (After Crelin ES 1969 Anatomy of the Newborn. Philadelphia: Lea and Febiger.)
The formation of the gonads is first indicated by the appearance of an area of thickened coelomic epithelium on the medial side of the mesonephric ridge in the fifth week, stage 16 (Figs 109.12, 109.13, 109.16). Elsewhere on the surface of the ridge the coelomic epithelium is one or two cells thick, but over this gonadal area it becomes multilayered. Thickening rapidly extends in a longitudinal direction until it covers nearly the whole of the medial surface of the ridge. The thickened epithelium continues to proliferate, displacing the renal corpuscles of the mesonephros in a dorsolateral direction and forms a projection into the coelomic cavity, the gonadal ridge. Surface depressions form along the limits of the ridge, which is thus connected to the mesonephros by a broad mesentery, the mesogenitale. In this way the mesonephric ridge becomes subdivided into a lateral part, the tubal fold, containing the mesonephric and paramesonephric ducts, and a medial part, the gonadal fold. The tubal fold also contains the nephric tubules and glomeruli at its base (Fig. 109.3). Up to the seventh week the ambisexual gonad possesses no sexually differentiating feature. From stage 16 the proliferating coelomic epithelium forms a number of cellular epithelial cords (sometimes called primary sex cords), separated by mesenchyme. The cords remain at the periphery of the primordium and form a cortex. Proliferation and labyrinthine cellular condensation of the
mesonephric mesenchyme, including angiogenic mesenchyme, produces a central medulla (Fig. 109.16).
REPRODUCTIVE DUCTS page 1382 page 1383
Figure 109.11 Bladder exstrophy. Misalignment of the genital tubercle and urogenital swellings with the urogenital membrane during early development results in subsequent malposition of the bladder, urethra and associated sphincters. The disappearance of the urogenital membrane exposes the posterior wall of the bladder, and the urethral opening is on the superior side of the penis or clitoris. (Redrawn from Tuchmann-Duplessis H, Haegel P 1972 Illustrated Human Embryology, Vol 2 Organogenesis. London: Chapman and Hall.)
page 1383 page 1384
Figure 109.12 A, Position of the gonads on the posterior abdominal wall, anteromedial to the mesonephros. B, Transverse section of figure A, through the line X-Y.
Figure 109.13 A, Indifferent or ambisexual stage of development. B, Male. The mesonephric ducts are retained (left) and the paramesonephric ducts involute (right). C, Female. The paramesonephric (Müllerian) ducts are retained (right) and the mesonephric ducts involute (left).
The paramesonephric Müllerian ducts develop in embryos of both sexes, but
become dominant in the development of the female reproductive system. They are not detectable until the embryo reaches a length of 10-12 mm (early sixth week). Each begins as a linear invagination of the coelomic epithelium, the paramesonephric groove, on the lateral aspect of the mesonephric ridge near its cranial end. The blind caudal end continues to grow caudally into the substance of the ridge as a solid rod of cells, which becomes canalized as it lengthens. Throughout the extent of the mesonephros it is lateral to the mesonephric duct, which acts as a guide for it. The paramesonephric duct reaches the caudal end of the mesonephros in the eighth week. It turns medially and crosses ventral to the mesonephric duct to enter the genital cord, where it bends caudally in close apposition with its fellow from the opposite side (Fig. 109.13). The two ducts reach the dorsal wall of the urogenital sinus during the third month, and their blind ends produce an elevation called the Müllerian sinus tubercle (Fig. 109.19). At the end of the indifferent stage each paramesonephric duct consists of vertical cranial and caudal parts and an intermediate horizontal region. The mesonephric ducts course caudally, medial to the paramesonephric ducts, and both duct systems open into the urogenital sinus. The genital ducts possess an external serosa on some surfaces derived from coelomic epithelium, a smooth muscle muscularis derived from underlying mesenchyme, and an internal mucosa derived from either the mesonephric duct or from the invaginated tube of coelomic epithelium that forms the paramesonephic duct. The layers are invaded by angiogenic mesenchyme and by nerves. Uterus and uterine tubes
In the female the mesonephric duct is vestigial. Cranially it becomes the longitudinal duct of the epoöphoron, while caudally it is referred to as Gartner's duct (Table 109.1). The cranial part of the paramesonephric ducts forms the uterine tubes, and the original coelomic invagination remains as the pelvic opening of the tube. The fimbriae become defined as the cranial end of the mesonephros degenerates. The caudal vertical parts of the two ducts fuse with each other to form the uterovaginal primordium (Figs 109.13, 109.14, 109.19). This gives rise to the lower part of the uterus and, as it enlarges, it takes in the horizontal parts to form the fundus and most of the body of the adult uterus. A constriction between the body of the uterus and the cervix can be found at 9 weeks. The stroma of the endometrium and the uterine musculature develop from the surrounding mesenchyme of the genital cord. Failure of fusion of the two paramesonephric ducts can lead to a range of anomalies summarized in Fig. 109.15. These fusions can also contribute to anomalies of vaginal development. At birth the uterus is 2.5-5 cm long (average 3.5 cm), 2 cm wide between the uterine tubes, and c.1.3 cm thick (Figs 11.4, 11.5, 109.9). The body of the uterus is smaller than the uterine cervix, which forms two-thirds or more of the length. The isthmus between the body and the cervix is absent. The fetal female reproductive tract is affected by maternal hormones and undergoes some enlargement in the fetus. The endocervical glands are active before birth and the cervical canal is usually filled with mucus. The uterus is relatively large at birth, and subsequently involutes to about one-third of its length and more than half of its weight. The neonatal size and weight of the uterus are not regained until puberty. The uterine tubes are relatively short and wide. The position of the uterus in the pelvic cavity depends to a great extent on the state of the bladder anteriorly and the rectum posteriorly. If the bladder contains only a small amount of urine the uterus may be anteverted but it is often in a direct line with the vagina (Figs 109.9, 109.19). Vagina
At c.60 mm crown rump (CR) length an epithelial proliferation, the sinuvaginal bulb, arises from the dorsal wall of the urogenital sinus in the region of the sinus tubercle. Its origin marks the site of the future hymen. The proliferation gradually extends cranially as a solid anteroposteriorly flattened plate inside the tubular condensation of the uterovaginal primordium, which will eventually become the fibromuscular vaginal wall. The caudal tip of the paramesonephric duct epithelium recedes until, at about the 140 mm stage, its junction with the epithelial proliferation lies in the cervical canal. page 1384 page 1385
Figure 109.14 Relative movements of the gonads and associated tubes. A, Gonads and mesonephros move caudally, the metanephros ascends. B, Posterior view of the mesonephric ducts (ureters) and the fused paramesonephric ducts (uterovaginal canal) in the female. For earlier development, see Fig. 109.3. (Redrawn from Tuchmann-Duplessis H, Haegel P 1972 Illustrated Human Embryology, Vol 2 Organogenesis. London: Chapman and Hall.)
Starting from its caudal end, and gradually extending cranially through its whole extent, the solid plate formed by the sinus proliferation enlarges into a cylindrical structure. Thereafter the central cells desquamate to establish the vaginal lumen. The extent to which mesonephric and paramesonephric ducts contribute directly to the formation of the vagina is controversial. As the upper end of the vaginal plate enlarges it grows up to embrace the cervix, and then is excavated to produce the vaginal fornices. Anomalies of paramesonepric duct fusion can produce related vaginal anomalies (Fig. 109.15). The urogenital sinus undergoes relative shortening craniocaudally to form the vestibule, which opens on the surface through the cleft between the genital folds. The lower end of the vaginal plate grows caudally so that in 109 mm embryos the vaginal rudiment approaches the vestibule. In fetuses of 162 mm the vaginal lumen is complete except at the cephalic end where the fornices are still solid; they are hollow by 170 mm. At approximately half way through gestation (180 mm) the genital canal is continuous with the exterior. During the later months of fetal life the vaginal epithelium is greatly hypertrophied, apparently under the influence of maternal hormones, but after birth it assumes the inactive form of childhood. In the neonate, the vagina is c.2.5-3.5 cm long and 1.5 cm wide at the fornices. The uterine cervix extends into the vagina for c.1 cm. The posterior vaginal wall is longer than the anterior wall, which gives the vagina a distinct curve (Figs 11.5, 109.9, 109.19). The cavity is filled with longitudinal columns covered with a thick layer of cornified, stratified squamous epithelium. These cells slough off after birth when the effect exerted by maternal hormones is removed. The orifice of the vagina is surrounded by a thick elliptical ring of connective tissue, the hymen (Fig. 109.9). During childhood the hymen becomes a membranous fold along the posterior margin of the vaginal lumen. Should the fold form a complete diaphragm across the vaginal lumen it is termed an imperforate hymenal membrane. Reproductive ducts in the male
In the male, the most paramesonephric ducts atrophy (Fig. 109.13) under the influence of anti-Müllerian hormone (AMH), which is released locally by the Sertoli cells of the testis. Vestigial structures are therefore most likely to persists cranially and/or caudally, at the limits of the local effects of AMH. A vestige of the cranial end of the duct persists as the appendix testis (Figs 109.13, 109.19, Table 109.1). The fused caudal ends of the two ducts are connected to the wall of the urogenital sinus by a solid utricular cord of cells, which soon merges with a proliferation of sinus epithelium, the sinu-utricular cord. The latter is similar to, but less extensive than, the sinus proliferation in the female. The proliferating epithelium is claimed to be an intermingling of the endoderm of the urogenital sinus with the lining epithelia of the mesonephric and paramesonephric ducts, which have extended on to the surface of the sinus tubercle. As the sinu-utricular
cord grows, so the utricular cord recedes from the tubercle. In the second half of fetal life the composite cord acquires a lumen and becomes dilated to form the prostatic utricle, the lining of which consists of hyperplastic stratified squamous epithelium (Figs 109.13, 109.19). The sinus tubercle becomes the colliculus seminalis. The main reproductive ducts in the male are derived from the mesonephric ducts, which are subsumed into the male reproductive system as the metanephric kidney develops (p. 1373). The mesonephric duct gives rise to the canal of the epididymis, vas deferens and ejaculatory duct. The seminal vesicle and the ampulla of the vas deferens appear as a common swelling at the end of the mesonephric duct during the end of the third and into the fourth month. Their appearance coincides with degeneration of the paramesonephric ducts, although no causal relation between the two events has been established. Separation into two rudiments occurs at c.125 mm crown-heel length. The seminal vesicle elongates, its duct is delineated and hollow diverticula bud from its wall. About the sixth month (300 mm crown-heel length) the growth rate of both vesicle and ampulla is greatly increased. Figure 109.10 shows the position of the ampulla of the vas deferens in the neonate, possibly in response to increased secretion of prolactin by the fetal or maternal hypophysis, or to the effects of placental hormones. The tubules of the prostate show a similar increase of growth rate at this time.
PRIMORDIAL GERM CELLS The primordial germ cells are formed very early from the epiblast. They are large cells, 12 to 20 µm in diameter, in comparison with most somatic cells. They are characterized by vesicular nuclei with well-defined nuclear membranes and by a tendency to retain yolk inclusions long after these have disappeared from somatic cells. It is not yet clear whether the primordial germ cells are derived from particular blastomeres during cleavage, if they constitute a clonal line from a single blastomere, or if they are the product of a progressive concentration of the nucleus of the fertilized ovum by unequal partition at successive mitoses. Primordial germ cells spend the early stages of development within the extraembryonic tissues near the end of the primitive streak and in the connecting stalk. In this situation they are away from the inductive influences to which the majority of the somatic cells are subjected during early development. page 1385 page 1386
Figure 109.15 Uterovaginal malformations. (Redrawn from Tuchmann-Duplessis H, Haegel P 1972 Illustrated Human Embryology, Vol 2 Organogenesis. London: Chapman and Hall.)
Primordial germ cells can be identified in human embryos in stage 11 when the number of cells is probably not more than 20-30. When the tail fold has formed they appear within the endoderm and the splanchnopleuric mesenchyme and epithelium of the hindgut as well as in the adjoining region of the wall of the yolk sac. They migrate dorsocranially in the mesentery, by amoeboid movements and by growth displacement, and pass around the dorsal angles of the coelom (medial coelomic bays) to reach the genital ridges from stage 15 (Fig. 109.16). It is believed that the genital ridges exert long-range effects on the migrating primordial germ cells, in terms of controlling their direction of migration and supporting the primordial germ-cell population. Primordial germ cells proliferate both during and after migration to the mesonephric ridges. Cells which do not complete this migration degenerate. After segregation, when they are often termed primary gonocytes, they divide to form secondary gonocytes.
DEVELOPMENT OF THE GONADS The factors that lead to formation of either testis or ovary are described below and in Fig. 109.16. The morphological events which occur in each type of gonadal development are presented first. UPDATE Date Added: 13 December 2005 Publication Services, Inc. Abstract: The retroperitoneal anastomoses of the gonadal veins in human foetuses. Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=16121322&query_hl=4 The retroperitoneal anastomoses of the gonadal veins in human foetuses. Szpinda M, Frackiewicz P, Flisinski P et al. Folia Morphol (Warsz). 2005 May;64(2):72-7.
Testis
Most studies support the hypothesis that the seminiferous tubules are formed from cords of epithelial cells derived from the proliferating coelomic epithelium (Figs 109.13, 109.16). The cords lengthen, partly by addition from the coelomic epithelium, and encroach on the medulla, where they unite with a network of cells derived from the mesonephric mesenchyme destined to become the rete testis. Primordial germ cells are incorporated into the cords, which later become enlarged and canalized to form the seminiferous tubules. The cells derived from the surface of the early gonad form the supporting Sertoli cells. Sertoli cells proliferate throughout development and perhaps into childhood. When they stop dividing they mature and cannot be reactivated. Each Sertoli cell can only support a fixed number of germ cells during their development into spermatozoa, i.e. the number of Sertoli cells produced at this time determines the maximal limit of sperm output. Because the germ cells make up the bulk of the adult testis, the number of Sertoli cells is a major determinant of the size to which the testes will grow (factors which impair the process of spermatogenesis, resulting in the loss of germ cells, will also affect testicular size). Variation in Sertoli cell number is probably the most important factor in accounting for the enormous variation in sperm counts between individual men, whether fertile or infertile. Indeed, the available data for adult men indicate that Sertoli cell numbers vary across a fiftyfold range. Although some of this variation may result from attrition of Sertoli cell numbers because of ageing, the major differences in Sertoli cell numbers will have been determined by events in fetal and/or childhood life. The interstitial cells of the testis are derived from mesenchyme and possibly also from coelomic epithelial cells that do not become incorporated into the tubules. Among other cell lines they form the embryonic and fetal cells of Leydig, which secrete testosterone . A later migration of mesenchyme beneath the coelomic epithelium forms the tunica albuginea of the testis. The cords of the rete testis become connected to the glomerular capsules in the persisting part of the mesonephros. Ultimately they become connected to the mesonephric duct by the five to twelve most cranial persisting mesonephric tubules. These become exceedingly convoluted and form the lobules of the head of the epididymis. The mesonephric duct, which was the primitive 'ureter' of the mesonephros, becomes the canal of the epididymis and the vas deferens of the testis. The seminiferous tubules do not acquire lumina until the seventh month; the tubules of the testicular rete become canalized somewhat earlier. Disorders of development of the testis and reproductive tract in the male fetus seem to be increasing in incidence. Testicular maldescent (cryptorchidism) and hypospadias appear to have doubled or trebled in incidence in the last 30-50 years, while testicular cancer has increased by an even greater margin and is now the commonest cancer of young men. Although testicular cancer is primarily a disease of young men (95% of cases affect 15- to 45-year-old males) it is now established that this age-incidence reflects activation of premalignant carcinomain-situ (CIS) cells, which are present at birth and which almost certainly arise during fetal life. It has been suggested that CIS cells are primordial germ cells that have failed to develop normally. Abnormalities of development of the testis and reproductive tract (e.g. gonadal dysgenesis, cryptorchidism, small testes) are important risk factors for the development of testicular cancer. However, the most dramatic change that appears to have occurred in the relatively recent past is a fall in sperm counts of around 40-50% (1% per year over the last 50 years). Although this dramatic decrease is obviously manifest only in adulthood, as with testicular cancer, it is thought that an explanation for this is impaired testicular development during fetal or childhood life. Ovary page 1386 page 1387
page 1387 page 1388
Figure 109.16 Development of the gonads and associated ducts as seen in transverse section to show the fate of the primordial sex cells, mesonephric duct and tubules and paramesonephric duct in the two sexes. (Modified from Williams PL, Wendell-Smith CP, Treadgold S 1969 Basic Human Embryology, 2nd edn, Philadelphia: Lippincott.)
In its earliest stages, the ovary closely resembles the testis, although it is slower to differentiate its characteristically female features (Figs 109.13, 109.16). Few, if any, of the epithelial cords invade the medulla. The majority remain in the cortex, where they may be joined by a second proliferation from the coelomic epithelium
overlying the gonad. In histological sections of ovaries from the third and subsequent months, the epithelial cords appear as clusters of cells, which may contain primitive germ cells, separated by fine septa of undifferentiated mesenchyme. An ovarian rete condenses in the medullary mesenchyme and some of its cords may join mesonephric glomeruli. The medulla subsequently regresses, and connective tissue and blood vessels from this region invade the cortex to form the ovarian stroma. During this invasion the clusters of epithelial cortical cells break into individual groups of supporting cells (now identified as granulosa cells), which surround the primordial germ cells (now identified as primary oocytes) that have entered the prophase of the first meiotic division. Primary oocytes are derived from a mitotic division of the primordial germ cells (naked oogonia). Their epithelial capsules consist of flattened pregranulosa cells derived from proliferations of coelomic epithelium. The ovary now has its full complement of primary oocytes. The majority undergo atresia at various stages during their development, but the remainder resume development after puberty, when they complete the first meiotic division shortly before ovulation (p. 1323). The granulosa cells at this time enlarge and multiply to form the stratum granulosum and, as they do so, they become surrounded by thecal cells, which differentiate from the stroma. Only the middle part of the gonadal ridge produces the ovary. Its cranial part is sterile and becomes the suspensory ligament of the ovary (infundibulopelvic fold of peritoneum). Its caudal region, also sterile, is incorporated into the ovarian ligament. Sex determination in the embryo
It was believed that the gonads were indifferent or ambisexual until the arrival of the primordial germ cells in the gonadal ridge, at which point the sex of the embryo was 'turned on' by the presence of the male or female germ cells. It now seems that the germ cells may be essentially irrelevant to testis determination; embryos in which the genital ridges are devoid of germ cells may still have morphologically normal testis development. It is not clear if the germ cells are necessary for ovarian determination. They are required for the proper organization and differentiation of the ovary, and their absence results in the development of 'streak gonads', where only lines of follicular cells can be seen, as in Turner's syndrome. The processes of sex determination and differentiation involve interacting pathways of gene activity, which lead to the total patterning of the embryo as either male or female. In one model of sex determination in humans, the female pathway is considered to be the default pathway. According to this model, the Y chromosome of a male embryo diverts development into the testicular pathway, and the resulting changes that convert an indifferent gonad to a testis produce a range of local and widely acting hormones, which collectively generate all the secondary sexual characteristics. The possession of a Y chromosome is usually associated with a male developmental pathway. The male-determining region of the Y chromosome, which is located near its tip, is termed the testis-determining factor (TDF), and is regarded by some as the 'master switch' that programmes the direction of sexual development. It has been suggested that the TDF acts initially within the epithelial cords of cells derived from the coloemic epithelium of the ambisexual gonad. These cells can potentially differentiate into either Sertoli or granulosa cells (the supporting cells for the germ cells in the testis and ovary respectively). TDF directs their development into Sertoli cells, which then influence the differentiation of the other cell types in the testicular pathway, so that Leydig cells appear later, and the connective tissue becomes organized into a male pattern. The germ cells are also affected by this environment. When they arrive they become enclosed within the Sertoli cells and enter mitotic arrest (which is characteristic of spermatogenesis), instead of entering meiosis and meiotic arrest (which characterizes oogenesis). Thus the development of male characteristics follows the expression of TDF, and female characteristics develop in its absence. Subsequent development of the male phenotype requires fetal secretion of both testosterone and anti-Müllerian hormone (AMH), (also called Müllerian inhibiting substance or MIS), and the development of the appropriate cytoplasmic testosterone-binding protein. Sertoli cells synthesize AMH, which causes the regression of the Müllerian ducts, and Leydig cells produce testosterone , which promotes the development of the mesonephric ducts, sets into process the development of male external genitalia, and sensitizes other tissues to
testosterone (p. 1388). Absence of the testosterone-binding protein results in XY individuals with testes and degenerated Müllerian ducts, but because they cannot respond to the circulating testosterone produced by their testes they develop female secondary sexual characteristics. Studies on the exact position of the TDF have been based on deletion mapping the Y chromosome in a class of XX males arising from abnormal X:Y interchange at meiosis. In all mammals tested a conserved sequence that mapped to the Y chromosome was found. The sequence formed part of a gene in the sexdetermining region of the Y chromosome, and was therefore termed SRY. It is believed to be genetically and functionally equivalent to TDF. This gene is first expressed in cells located centrally in the developing gonad and then later in the cranial and caudal poles in supporting cells, called pre-Sertoli cells, which are derived from the coelomic epithelium. Studies indicate that SRY initiates testis formation from the indifferent gonad by directing the development of supportingcell precursors as Sertoli rather than granulosa cells (Albrecht & Eicher 2001). The possession of a Y chromosome expressing SRY and TDF may, therefore, underlie the switch to development of the male phenotype, by initiating Sertoli cell differentiation. An alternative view is that the possession of TDF accelerates the development of the gonads in XY embryos generally, so that testes are larger and more advanced than ovaries of the same age. Male human fetuses are generally bigger than females from 12 weeks' gestation, indeed males are already slightly ahead of females at six weeks' gestation, just prior to testicular differentiation. It has been suggested that this difference in the growth rate is encoded in the sex chromosomes. Once gonadal development has started, the difference in size between testes and ovaries becomes proportionately much greater than the overall size difference between XY and XX fetuses. (Interestingly, the right gonad develops slightly ahead of the left, an observation which may be correlated with the finding that testes are more often on the right side, and ovaries on the left, in hermaphrodites.) Although a fetus is exposed to maternal hormones, the accelerated development of the testis at early embryological stages ensures arrest of meiosis of the germ cells, and the production of local hormones that masculinize the male embryo before the development of the reproductive tract and ovaries of the female. The range of intersex conditions, of phenotypic sex that is not correlated to genotype, and the effect of multiple X chromosomes in males, suggests that the male developmental pathway involves many testis determining genes, whereas only a single X chromosome determines the female default pathway. Once testicular differentiation and male hormone secretion have begun, other Y-chromosomal genes are required to maintain spermatogenesis and complete spermiogenesis. The impairment of oogenesis by other chromosomal abnormalities is much less severe than the impairment of spermatogenesis.
DESCENT OF THE GONADS The gonads develop on the posterior abdominal wall bilaterally along the central portion of the mesonephros. This region receives a rich blood supply, which is directed to the gonads as the mesonephros involutes. Both gonads descend, the testis to lie outside the abdominal cavity, and the ovary to the pelvis, however, they retain their early blood supply from the dorsal aorta. Descent of the testis
Each testis initially lies on the dorsal abdominal wall. As it enlarges, its cranial end degenerates and the remaining organ therefore occupies a more caudal position. It is attached to the mesonephric fold by the mesorchium (the mesogenitale of the undifferentiated gonad), a peritoneal fold that contains the testicular vessels and nerves and a quantity of undifferentiated mesenchyme. It also acquires a secondary attachment to the ventral abdominal wall, which has a considerable influence on its subsequent movements. At the point where the mesonephric fold bends medially to form the genital cord (Fig. 109.3), it becomes connected to the lower part of the ventral abdominal wall by an inguinal fold of peritoneum. The mesenchymal cells occupying the core of the inguinal fold condense as another cord, the gubernaculum (Figs 109.3, 109.13, 109.17). This extends from the epidermal ectoderm, which will later form the scrotum, through the inguinal fold and the mesorchium to the caudal pole of the testis. It travels through the site of the future inguinal canal, which is formed around it by the differentiating muscles of the abdominal wall. At the end of the second month the caudal part of the ventral abdominal wall is horizontal but, after the return of the intestine to the peritoneal cavity, it grows in length and becomes progressively more vertical. As the umbilical artery runs ventrally from the dorsal to the ventral
wall, it pulls up a falciform peritoneal fold, which forms the medial boundary of a peritoneal fossa, the saccus vaginalis or lateral inguinal fossa, into which the testis projects. This lower end of the fossa protrudes down the inguinal canal, along the ventrosuperior aspect of the gubernaculum, as the processus vaginalis (Figs 109.17, 109.19). page 1388 page 1389
Figure 109.17 Descent of the testis. The testis is always retroperitoneal. (Redrawn from Tuchmann-Duplessis H, Haegel P 1972 Illustrated Human Embryology, Vol 2 Organogenesis. London: Chapman and Hall.)
The mechanism of testicular descent has been variously ascribed to shortening and active contraction of the gubernaculum, increased intra-abdominal pressure, a simple growth process, and the effect on the convex surface of the gland of the active contraction of the lower fibres of internal oblique, which squeeze it through the canal. (For a review of testicular descent in the human see Barteczko & Jacob 2000.) The gubernaculum precedes the testis both spatially and in rate of growth, and forms a tapering column of soft tissue with the diminutive testis at its cranial pole. It continues to grow until the seventh month, by which time its caudal part has filled the future inguinal canal and has begun to expand the developing scrotum. In this it also precedes the processus vaginalis. It does not develop attachments to skin, nor is there any evidence that it produces the radiating extensions into suprapubic, perineal or femoral sites that are often cited to explain the various forms of ectopic testis. By virtue of its soft consistency, gubernacular tissue (which in the early stage is formed mainly of hyaluronic acid) may offer a route of low resistance to the descending testis. It stops growing in the last two months of gestation, and this, coupled with an accelerating rate of growth in the testis and epididymis, may be a factor in testicular descent as far as the inguinal canal. The mechanism of the final rapid descent of the testis into the scrotum is not yet clear, although endocrine effects are certainly important. Experimentally, division of the genitofemoral nerve prevents both inguinoscrotal testicular descent and differentiation and migration of the gubernaculum. It has been suggested that androgens act on neurones, the axons of which run in the genitofemoral nerve, stimulating release of neurotransmitters, which might act as second messengers for androgens, from the nerve endings. A peptide neurotransmitter, calcitonin gene-related peptide (CGRP), is present in the genitofemoral nerve and its cell bodies in the spinal cord. CGRP causes the gubernacula from newborn male mice to contract rhythmically, whereas CGRP antagonists inhibit this contraction. The caudal pole of the testis is retained in apposition with the deep inguinal ring by the gubernaculum during the sixth and seventh months. The testes finally descend into the scrotum before birth (Fig. 11.4), the left testis usually migrating ahead of the right. In full-term male neonates 90% have descended testes. In premature babies descent may not be complete. As the testis descends it is preceded by the processus vaginalis. The distal end of the processus vaginalis, into which the testis projects, forms the tunica vaginalis testis. The portion associated with the spermatic cord in the scrotum and inguinal canal normally becomes obliterated, and usually leaves a fibrous remnant. At birth the processus vaginalis is collapsed but not necessarily obliterated. It remains patent for up to 14 days in nearly 70% of male infants, but by 20 days after birth it is partially (or completely) obliterated in 80% of male infants, the left side before the right. UPDATE Date Added: 27 July 2005 Abstract: Autonomic nervous system plays role in obliteration of processus vaginalis. Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=14730435&query_hl=8 Autonomic nervous system plays role in obliteration of processus vaginalis. In a few cases the original cavity within the processus may persist, in whole or in part, in any location. These variations may form the walls of hernial sacs or encysted fluid sites. When a patent processus retains a connection with the general peritoneal cavity it provides a preformed sac for a potential oblique inguinal hernia. It may be occluded at its upper end and be shut off from the tunica vaginalis and yet remain patent in the intervening section. The patent portion may become distended with fluid, and present as an encysted hydrocoele of the spermatic cord. The spermatic cord is relatively large in the neonate, as are the seminal vesicles and adjacent ampullae of the vas deferens. In aberrant testicular descent the testis may remain in the abdomen, or it may fail to reach the scrotum and may then lie in the perineum, at the root of the penis, at the superficial inguinal ring, or in the upper part of the thigh. The testis must follow the processus vaginalis. Should the latter, for any reason, follow a structure other than the scrotal extension of the gubernaculum, malposition of the testis will result. Traditionally, these malpositions have been associated with certain additional extensions of gubernacular tissue. The largest extension normally passes to the scrotum. Lesser extensions have been
described as gaining attachment to the perineum, the root of the penis, the pubis, the inguinal ligament, and the neighbourhood of the saphenous opening. However, there is considerable doubt about these lesser expansions (the socalled 'tails of Lockwood'). They may reflect premature and abnormal fibrous partitioning of the gubernacular mesenchyme. Descent of the ovary page 1389 page 1390
Figure 109.18 Descent of the ovary. A, Early developing left ovary and uterine tube. B, Start of posterior movement of ovary. C, Left ovary in definitive posterior position. D, Parasagittal section of the ligaments associated with the ovary viewed from a left lateral position. (Redrawn from Tuchmann-Duplessis H, Haegel P 1972 Illustrated Human Embryology, Vol 2 Organogenesis. London: Chapman and Hall.)
The relative movements of the ovary are less extensive than those of the testis. Like the testis, the ovary ultimately reaches a lower level than it occupies in the early months of fetal life, but it does not leave the pelvis to enter the inguinal canal, except in certain anomalies. The ovary is connected to the medial aspect of the mesonephric fold by the mesovarium (homologous with the mesorchium), and to the ventral abdominal wall by the inguinal fold (Figs 109.13, 109.18). A mesenchymatous gubernaculum develops in this fold but, as it traverses the mesonephric fold, it acquires an additional attachment to the lateral margin of the uterus near the entrance of the uterine tube. Its lower part, caudal to this uterine attachment, becomes the round ligament of the uterus and the part cranial to this becomes the ovarian ligament. Collectively these structures are homologous with the gubernaculum testis in the male (Figs 109.18, 109.19). This new uterine attachment may be correlated with the restricted ovarian descent. At first the ovary is attached to the medial side of the mesonephric fold but, in accordance with the manner in which the two mesonephric folds form the genital cord (Figs 109.3, 109.13), it is finally connected to the posterior layer of the broad ligament of the uterus. The gubernacula thus persist in the female, unlike the male, as bilateral fibrous bands or ligaments. Experimentally, they do not contract in response to application of CGRP (see above). They do not extend into the labia majora, as frequently described, but to the connective tissue just external to the external ring of the inguinal canal. The saccus vaginalis is present in the female.
Its prolongation into the inguinal canal (sometimes termed the canal of Nuck) is normally completely obliterated, but may remain patent and form the sac of a potential indirect inguinal hernia. page 1390 page 1391
Figure 109.19 The development of the urogenital system from the indifferent stage to the definitive male and female conditions. (Modified from Williams PL, Wendell-Smith CP, Treadgold S 1969 Basic Human Embryology, 2nd edn. Philadelphia: Lippincott.)
In the neonate the ovaries lie in the lower part of the iliac fossae. The long axis of the ovary is almost vertical. It becomes temporarily horizontal during descent, but regains the vertical when it reaches the ovarian fossa. The ovaries complete their descent into the ovarian fossae in early childhood. Thus, at birth the ovary and the lateral end of the corresponding uterine tube lie above the pelvic brim. They do not sink into the lesser pelvis until the latter enlarges sufficiently to contain both of them and the other pelvic viscera, including the bladder. The combined weight of the ovaries at birth is c.0.3 g, which is relatively large, and much larger than the combined weight of the testes (Figs 11.4, 109.9). The ovaries double in weight during the first 6 postnatal weeks. They bear surface furrows, which disappear during the second and third postnatal months. All of the primary oocytes for the reproductive life of a female are present in her ovaries by the end of the first trimester of pregnancy. Of the 7,000,000 primary oocytes estimated to
be present at the fifth month of gestation, 1,000,000 remain at birth, 40,000 by puberty, and only 400 are ovulated during reproductive life.
CLOACA AND EXTERNAL GENITALIA The cloaca is that region at the end of the primitive hindgut, which is continuous with the allantois, a ventral diverticulum (Fig. 109.7). The allantois passes into the connecting stalk of the early embryo prior to tail folding and is then drawn into the body cavity after stage 10. It retains an extension into the connecting stalk, and later into the umbilicus, throughout embryonic life. The cloaca is a slightly dilated cavity lined with endoderm. It is initially connected cranially to the enteric hindgut, and ventrocaudally is in contact with overlying ectoderm at the cloacal membrane. Proliferation of mesenchyme at the angle of the junction of the hindgut and allantois produces a urorectal septum, which grows caudally and eventually produces fusion of the endodermal epithelium with the cloacal membrane (Figs 109.4, 109.7, 109.20). The cloaca is thus divided into a presumptive rectum and anal canal dorsally, and a presumptive urinary bladder and urogenital sinus ventrally, and the cloacal membrane is divided into anal and urogenital membranes respectively. The nodal centre of division is the site of the future perineal body. The urogenital sinus receives the mesonephric and paramesonephric ducts. page 1391 page 1392
Figure 109.20 The development of the external genitalia from the indifferent stage to the definitive male (left) and female (right) conditions.
Anomalies of cloacal subdivision may result in a range of defects. In extroversion of the cloaca (ectopia cloacae), the urorectal septum does not develop. The defect is complicated by a failure of mesenchymal migration around the ventral body wall to support the umbilical cord and this results in a large abdominal
defect with a central colonic portion and bilateral bladder components. With only partial development of the urorectal septum, the urogenital sinus may remain with a high confluence of bladder, vagina and rectum. The cloacal membrane may be abnormally elongated and prematurely ruptured throughout its whole extent, prior to the formation of the urorectal septum, or, in some cases, there may be only a small sinus opening externally at the skin. The anal musculature is often present but not associated with the anal canal. Pelvic floor
The pelvic floor consists of the ligamentous supports of the cervix, and the pelvic and urogenital diaphragms, and constitutes another partition that traverses the body cavity. Little is known about pelvic floor development in the human. The striated muscle is derived from the somatic epithelial plates in a similar manner to the muscles of the ventrolateral body wall. Puborectalis appears in 20-30 mm embryos, following opening of the anal membrane, and striated muscle fibres can be seen at 15 weeks. The smooth muscle of the urethral sphincter is also present at this time. Urethra
The urethra is derived from endoderm, as are the prostate gland and vagina (both outgrowths of the lower urogenital sinus or urethra), and the other small glandular structures that develop around the caudal body orifices. In the male, the prostatic urethra proximal to the orifice of the prostatic utricle is derived from the vesicourethral part of the cloaca and the incorporated caudal ends of the mesonephric ducts. The remainder of the prostatic part, the membranous part, and probably the part within the bulb, are all derived from the urogenital sinus. The succeeding section, as far as the glans, is formed by the fusion of the genital folds (Fig. 109.20) and so is contained in the shaft of the penis. The short section within the glans is formed from ectoderm, which invaginates into the glans. In the female, the urethra is derived entirely from the vesicourethral region of the cloaca, including the dorsal region derived from the mesonephric ducts. It is homologous with the part of the prostatic urethra proximal to the orifices of the prostatic utricle and the ejaculatory ducts. The region of the early urethra remains open to form the vestibule into which the definitive urethra and vagina open. It is believed that these regions are invaded by ectoderm because they are innervated by somatic nerves. Urethral defects caused by arrests of development are not uncommon in the male. In epispadias the urethra opens on the dorsal aspect of the penis at its junction with the anterior abdominal wall. This anomaly is considered to be a less severe form of exstrophy of the bladder. In the simplest form of hypospadias, the urethra may open on the ventral (perineal) aspect of the penis at the base of the glans, and the part of the urethra that is normally within the glans is absent. In more severe cases the genital folds fail to fuse, and the urethra opens on the ventral aspect of a malformed penis just in front of the scrotum. A still greater degree of this malformation is accompanied by failure of the genital swellings to unite with each other. In these cases the scrotum is divided and, since the testes are also frequently undescended, the resemblance to the labia majora is very striking. Male children suffering from this deformity are often mistaken for girls. In such cases it is important to determine at the earliest stage not only the chromosomal sex of the infant but also the internal anatomy and stage of development of the internal genital tract. Sex assignment and rearing will depend on these factors. The urethral sphincter first forms as a mesenchymal condensation around the urethra in 12-15 mm (stage 18) embryos, after division of the cloaca. The mesenchyme proliferates and becomes defined at the bladder neck in 31 mm embryos, and along the anterior part of the urethra by 69 mm. The muscle fibres differentiate after 15 weeks' gestation, at which time both smooth and striated fibres are present. In females there is continuity between the smooth muscle of the urethral wall and of the bladder. In males the muscle fibres are less abundant because of the local development of the prostate. Striated muscle fibres form around the smooth muscle initially in the anterior wall of the urethra and later encircle the smooth muscle layer. The origin of the striated muscle is not known it could be derived from the myogenic cells from which puborectalis develops. The smooth and striated components of the urethral sphincter are closely related, but there is no mixing of fibres as occurs in the anorectal sphincter. page 1392 page 1393
Prostate gland
The prostate gland arises during the third month from interactions between the urogenital sinus mesenchyme and the endoderm of the proximal part of the urethra. Early outgrowths, some 14 to 20 in number, arise from the endoderm around the whole circumference of the tube, but mainly on its lateral aspects and excluding the dorsal wall above the utricular plate. They give rise to the outer glandular zone of the prostate (p. 1301). Later outgrowths from the dorsal wall above the mesonephric ducts arise from the epithelium of mixed urogenital, mesonephric and possibly paramesonephric, origin covering the cranial end of the sinus tubercle. They produce the internal zone of glandular tissue. The outgrowths, which are at first solid, branch, become tubular and invade the surrounding mesenchyme. The latter is differentiating into smooth muscle, associated blood and lymphatic vessels and connective tissue and is invaded by autonomic nerves. Similar outgrowths occur in the female but remain rudimentary. The urethral glands correspond to the mucosal glands around the upper part of the prostatic urethra, and the para-urethral glands correspond to the true prostatic glands of the external zone. The bulbourethral glands in the male, and the greater vestibular glands in the female, arise as diverticula from the epithelial lining of the urogenital sinus. External genitalia
Patterning of the external genitalia may be achieved by mechanisms similar to those that pattern the face and limb. In the cranial region neural crest mesenchyme makes an important contribution to the organization of the pharyngeal arches and the regions around the upper sphincters. Neural crest also arises from the tail-bud region, specifically from a population of cells termed the caudoneural hinge, which share the same molecular markers as the primitive node (p. 241). The neural tube at this level is derived from a mesenchymal/epithelial transformation of caudoneural-hinge cells, which form a cylinder. Neural crest cells delaminate from the dorsal surface of the cylinder in a rostrocaudal direction. It is not known whether neuronal neural crest arising from secondary neurulation processes contributes to the caudal interface between endoderm and ectoderm. The external genitalia, like the gonads, pass through an indifferent state before distinguishing sexual characters appear (Fig. 109.20). From stage 13, primordia of the external genitalia, composed of underlying proliferating mesenchyme covered with ectoderm, arise around the cloacal membrane, between the primitive umbilical cord and the tail. During stage 15 the cloacal membrane is divided by the urorectal septum into a cranial urogenital membrane and a caudal anal membrane (Figs 109.4, 109.14). Local ectodermal/mesenchymal interactions give rise to the anal sphincter, which will develop without the presence of the urorectal septum or the anal canal. A surface elevation, the genital tubercle, appears at the cranial end of the urogenital membrane and two lateral ridges, the genital or urethral folds, form each side of the membrane (Fig. 109.20). The genital tubercle forms a distinct primordium, which will become the glans of either the penis or the clitoris. Elongation of the genital folds and urogenital membrane produces a primitive phallus. As this structure grows it is described as having a cranial surface analogous to the dorsum of the penis, and a caudal surface analogous to the perineal surface of both sexes. The urogenital sinus, contiguous with the internal aspect of the urogenital membrane, becomes attenuated within the elongating phallus forming the primitive urethra. The urogenital membrane breaks down at about stage 19 (20 mm, 6.5 weeks) allowing communication of ectoderm and endoderm at the edges of the disrupted membrane and continuity of the urogenital sinus with the amniotic cavity. Urine can escape from the urinary tract from this time. The endodermal layer of the attenuated distal portion of the urogenital sinus, which is now displayed on the caudal aspect of the phallus, is termed the urethral plate. As mesenchyme proliferates within the genital folds, the urethral plate sinks into the body of the phallus forming a primary urethral groove. The genital folds meet proximally in a transverse ridge immediately ventral to the anal membrane. While these changes are in progress two labioscrotal (genital) swellings appear, one on each side of the base of the phallus, and extend caudally, separated from the genital folds by distinct grooves (Figs 109.20, 109.21). As a general rule, epithelium, which can be touched easily and has a somatic innervation, is derived from ectoderm. In the buccal cavity and pharynx the ectoderm/endoderm zone is towards the posterior third of the tongue - touch here
usually elicits the gag reflex. In the anal canal the outer portion, distal to the anal valves, is derived from ectoderm and has a somatic innervation, whereas the epithelium proximal to the valves is derived from endoderm and has an autonomic innervation. Homologies of the parts of the urogenital system are shown in Table 109.1. Male genitalia
The growth of male external characteristics is stimulated by androgens regardless of the genetic sex. The male phallus enlarges to form the penis. The genital swellings meet each other ventral to the anus and unite to form the scrotum (Fig. 109.20). The genital folds fuse with each other from behind forwards, enclosing the phallic part of the urogenital sinus behind to form the bulb of the urethra and closing the definitive urethral groove in front to form the greater part of the spongiose urethra. Fusion of the folds results in the formation of a median raphe and occurs in such a way that the lining of the postglandular urethra is mainly, perhaps wholly, endodermal in origin. Thus, as the phallus lengthens, the urogenital orifice is carried onwards until it reaches the base of the glans at the apex of the penis. From the tip of the phallus an ingrowth of surface ectoderm occurs within the glans to meet and fuse with the penile urethra. Subsequent canalization of the ectoderm permits a continuation of the urethra within the glans.
page 1393 page 1394
Figure 109.21 Scanning electron micrographs of early human external genitalia. A, Indifferent stage in a human embryo estimated as 42 postovulatory days. B, A human female embryo at 12 weeks' development. The genital folds are not fused. C, A human male embryo at 12 weeks. Fusion of the genital folds has occurred. (Photographs by P Collins.)
Table 109-1. Homologies of the parts of the urogenital system in male and female
Gonad Gubernacular cord Mesonephros (Wolffian body)
Mesonephric duct (Wolffian duct)
Paramesonephric (or Müllerian) duct
Allantoic duct Cloaca: dorsal part ventral part
urogenital sinus
Genital folds Genital tubercle
Testis Gubernaculum testis
Ovary Ovarian and round ligaments Appendix of epididymis Appendices (?) vesiculosae (?) Efferent ductules Epoöphoron Lobules of epididymis Paradidymis Paroöphoron Aberrant ductules Duct of epididymis Duct of epoöphoron (Gartner's duct) Vas deferens Ejaculatory duct Part of bladder and Part of bladder and prostatic urethra urethra Appendix of testis Uterine tube
Prostatic utricle Urachus
Uterus Vagina (?) Urachus
Rectum and upper part Rectum and upper part of anal canal of anal canal Most of bladder Most of bladder and the Part of prostatic urethra urethra Prostatic urethra distal to utricle Bulbo-urethral glands Greater vestibular glands Rest of urethra to glans Vestibule Ventral penis Labia minora Glans penis Clitoris Urethra in glans
The glans and shaft of the penis are recognizable by the third month. The prepuce also begins to develop in the third month, when the primary external orifice of the urethra is still at the base of the glans. A ridge consisting of a mesenchymal core covered by epithelium appears proximal to the neck of the penis and extends forwards over the glans. A solid lamella of epithelium deep to this ridge extends backwards to the base of the glans. The ventral extremities of the ridge curve backwards to become continuous with the genital folds at the margins of the urethral orifice. As the urethral folds meet to form the terminal part of the urethra, the ventral horns of the ridge fuse to form the frenulum. The epithelial lamella breaks down over the dorsum and sides of the glans to form the preputial sac, and thus free the prepuce from the surface of the glans. Thereafter the prepuce grows as a free fold of skin, which covers the terminal part of the glans. Although the prepuce and glans begin to separate from the fifth month in utero, they may still be joined at birth. The preputial sac may not be complete until 6-12 months or more after birth and, even then, the presence of some connecting strands may still interfere with the retractability of the prepuce. The mesenchymal core of the phallus is comparatively undifferentiated in the first two months, but the blastemata of the corpora cavernosa become defined during the third month. Nerves are present in the differentiating mesenchyme from the seventh week. Despite containing less smooth muscle and elastic tissue than the adult, the neonatal penis is capable of erection. The scrotum is formed by proliferation of the genital swellings, which are the anchoring points for the gubernaculum testis. The genital swellings fuse across the midline covering the base of the penis. The testes descend into the scrotum prior to birth. In the neonate the penis and scrotum are relatively large. The
scrotum has a broad base which does not narrow until after the first year. Both the septum and the walls of the scrotum are relatively thicker than in adults. Female genitalia
The female phallus, which exceeds the male in length in the early stages, becomes the clitoris. The genital swellings remain separate as the labia majora and the genital folds also remain separate, forming the labia minora (Fig. 109.20). The perineal orifice of the urogenital sinus is retained as the cleft between the labia minora, above which the urethra and vagina open. The prepuce of the clitoris develops in the same way as its male homologue. By the fourth month the female external genitalia can no longer be masculinized by androgens. At birth neonatal females have relatively enlarged labia minora, clitoris and labia majora. The labia majora are united by a posterior labial commissure and each contains the distal end of the round ligament of the uterus, the gubernaculum ovarii (Fig. 109.18). There is evidence that in certain tissues, e.g. urogenital sinus and genital swellings, testosterone is converted into 5!-dihydrotestosterone. In XY individuals with a genetic deficiency of the enzyme responsible for this conversion, not only functioning testes but also female external genitalia with an enlarged clitoris and a small vaginal pouch, are present, suggesting that external genital development is under the control of 5!-dihydrotestosterone. Such individuals are usually raised as girls. However, at puberty the external genitalia become responsive to testosterone , which causes masculinization at this stage.
MATURATION OF THE REPRODUCTIVE ORGANS AT PUBERTY Until the adolescent growth spurt the reproductive organs grow very slowly. Generally the changes occur over a time period termed puberty. The sequence of these events is much less variable than the age at which they take place. The sequence of puberty in girls and boys is shown in Fig. 109.22. In girls the appearance of the breast bud is usually the first sign of puberty. The uterus and vagina develop simultaneously with the breast. Menarche occurs after the peak of the height spurt - onset is more closely related to radiological than to chronological age. It has been suggested that the menarche occurs as a critical weight of c.50 kg is attained, and certainly sports and excessive restriction of diet, which may reduce weight below this level, can cause amenorrhoea in women who were previously menstruating normally. Tall girls reach sexual maturity earlier than short ones, but girls with a late adolescent growth spurt and later puberty are ultimately taller on the average than those who pass through the menarche early, for they have longer to grow. A girl who has begun to menstruate can be predicted to grow a further 7.5 cm at most. Menarche marks a definitive stage of uterine development but does not mean attainment of full reproductive function. Many of the early menstrual cycles may not involve ovulation.
page 1394 page 1395
Figure 109.22 Events which occur at adolescence in average girls and boys. The figures beneath the bars indicate the range of ages within which each event may begin and end. Figures within the bars indicate the developmental stage. (Adapted with permission from Tanner JM 1962 Growth at Adolescence, 2nd edn. Oxford: Blackwell Publishing.)
The earliest sign of puberty in boys is the growth of the testes and scrotum. The volume of the testes may be estimated - the average adult volume is 20 ml - and a volume of 6 ml indicates that puberty has started. Later the penis, prostate and seminal vesicles begin to enlarge. Increased testosterone levels produced by the Leydig cells of the testes promote changes in the larynx, skin and distribution of bodily hair. REFERENCES Albrecht KH, Eicher EM 2001 Evidence that SRY is expressed in pre-Sertoli cells and Sertoli and granulosa cells have a common precursor. Dev Biol 240(1): 92-107. Medline Similar articles Full article Barteczko KJ, Jacob MI 2000 The testicular descent in human. Origin, development and fate of the gubernaculum Hunteri, processus vaginalis peritonei and gonadal ligaments. Adv Anat Embryol Cell Biol 156: III-X, 1-98. Springer. Includes scanning electron microscopy images and three-dimensional reconstructions of the human testis from stage 14. The caudal ligaments of the ovary and uterus are also considered. Medline Similar articles Chauvet V, Qian F, Boute N et al 2002 Expression of PKD1 and PKD2 transcripts and proteins in human embryo and during normal kidney development. Am J Pathol 160: 973-83. Medline Similar articles Sharpe RM, McKinnell C, Kivlin C, Fisher JS 2003 Proliferation and functional maturation of Sertoli cells, and the relevance to disorders of testis function in adulthood. Reproduction 125: 769-84. Medline Similar articles Full article page 1395 page 1396
© 2008 Elsevier
SECTION 8 PELVIC GIRDLE AND LOWER LIMB Andrew Williams (Lead Editor) Richard LM Newell (Editor) Mark S Davies (Editor, chapter 115) Patricia Collins (Embryology, Growth and Development) Critical reviewers: Paul Cartwright (chapter 110), Thomas Ind (111) page 1397 page 1398
© 2008 Elsevier
110 General organization and surface anatomy of the lower limb This chapter is divided into two sections. The first is an overview of the general organization of the lower limb, with particular emphasis on the fascial skeleton, distribution of the major blood vessels and lymphatic channels, and of the branches of the lumbar and sacral plexuses: it is intended to complement the detailed regional anatomy described in Chapters 111 to 115. The second section describes the surface anatomy of the lower limb. The structure of the lower limb is determined by its adaptations for weightbearing, locomotion and the maintenance of equilibrium (stability). Indeed, the adaptations for weightbearing and for stability, and the differing developmental histories of the limbs account for the major structural and functional differences between the lower limb and the upper limb. There are two important anatomical junctional or transitional zones between the trunk and the lower limb through which longitudinally running nerves and vessels pass in both directions. These zones are the inguinal (pelvicrural) and the gluteal (buttock) regions. The latter includes the junctional zones between the limb and the abdominopelvic cavity (via the greater sciatic foramen) and between the limb and the perineum (via the lesser sciatic foramen).
© 2008 Elsevier
SKIN, FASCIA AND SOFT TISSUES In the young adult the skin of the lower limb is generally stronger and thicker than that of the upper limb: weightbearing skin, e.g. of the sole of the foot, is particularly thickened. The skin of the buttocks and posterior thigh bears weight in the sitting position. The skin over the anterior aspect of the lower leg is particularly fragile and vulnerable in the elderly. Body hair is usually well developed in all areas except the sole and posterior ankle.
FASCIAL SKELETON The well-defined 'fascial skeleton' of the lower limb forms a tough circumferential 'stocking-like' structure that constrains the musculature. Septa pass from this outer fascial sheath to the bones within, confining the functional muscle groups within osteofascial compartments. The tough fascia gives additional areas of attachment to the muscles and ensures that they work to maximal effect. Thickenings in the ensheathing layer may act as additional tendons and form fibrous retinacula where tendons cross joints. Although these fascial layers and planes are particularly prominent in the embalmed cadaver, they are also very real and significant structures in the living. The pattern of soft-tissue organization directs the physiological action of the muscles and is crucial for efficient venous return from the limb. The fascial planes also control and direct the spread of pathological fluids (blood, pus) within the limb and play an important part in determining the degree and direction of displacement seen in long bone fractures. Fasciocutaneous system
The fascial septa dictate the pathways of cutaneous arteries, which subsequently perforate and ramify on the fascial 'stocking' before supplying the skin.
OSTEOFASCIAL COMPARTMENTS IN THE LOWER LIMB There are three functional compartments in the thigh: anterior (extensor), posterior (flexor) and medial (adductor) (p. 1461). Only the anterior and posterior compartments are separated by distinct fascial septa. A very definite fascial separation into anterior (extensor), posterior (flexor) and lateral (evertor) compartments exists in the leg: compartment syndrome is most common in this region (p. 1489). Osteofascial compartments in the foot are described on page 1509. All osteofascial compartments are traversed by vessels and nerves that also supply the muscles within the compartments (pp. 1461, 1489). Compartment syndrome The limiting walls of osteofascial compartments are largely inelastic: any condition that leads to an increase in the volume of the compartmental contents is therefore likely to cause an increase in intracompartmental pressure. Such conditions include muscle swelling caused by trauma or overuse, haemorrhage and local infection. If untreated, this increased pressure will lead to damage to the nerves
and vessels traversing the compartment - which will have severe consequences for those parts of the limb distal to the compartment - and produce necrosis of the intracompartmental muscles.
© 2008 Elsevier
BONES AND JOINTS (Figs 110.1, 110.2) The bones of the lower limb are: the three fused components of the pelvic girdle; the femur and its associated patella (thigh); the tibia and fibula (leg); the tarsus, metatarsus and phalanges (foot). The innominate bone (especially the ilium and ischium) and the femur, tibia and bones of the hindfoot are strong and their external (cortical) and internal (trabecular) structure are adapted for weightbearing. The pelvic girdle connects the lower limb to the axial skeleton via the sacroiliac joint, an originally synovial joint in which mobility has been sacrificed for stability and strength for effective weight transmission from the trunk to the lower limb. The anterior joint of the pelvic girdle is the pubic symphysis, a secondary cartilaginous joint that may move slightly during hip and sacroiliac movement and during childbirth. The hip joint, a synovial ball-and-socket joint, exhibits a very effective compromise between mobility and stability, allowing movement in all three orthogonal planes. The more distal joints have gained mobility at the expense of stability. The knee joint anatomically includes the patellofemoral joint, a synovial joint allowing the patella to move over the distal femur. The main component of the knee joint is a bicompartmental synovial joint between femur and tibia allowing flexion, extension and some medial and lateral rotation of the leg. It is not a true hinge joint as its axes of flexion and extension are variable and there is coupled rotation. The tibia and fibula articulate with each other at the superior and inferior tibiofibular joints. The superior joint, a plane synovial joint, allows slight gliding movement only. The inferior joint, a fibrous joint, lies just above the ankle and allows significant rotation of the fibula linked to ankle motion. The ankle (talocrural) joint is formed by the inferior tibia and fibula 'gripping' the talus. It allows dorsiflexion and plantarflexion. There are multiple joints in the foot: these can be simplified by considering the hindfoot, midfoot and forefoot. These joints allow the complex movements of which the foot is capable, making the foot well adapted to provide a platform for standing and for shock absorption and propulsion in gait. Both knee and ankle are commonly subject to closed injuries: the virtually subcutaneous position of the knee makes it liable to open injury and infection. Although the ankle is a frequently injured weightbearing joint, the prevalence of degenerative arthritis as a clinical presentation is surprisingly low when compared with that found in the hip and knee. page 1399 page 1400
Figure 110.1 Overview of bones of the lower limb: posterior aspect. (From Aids to the Examination of the Peripheral Nervous System. 2000. 4th edn. London: Saunders. With permission of Guarantors of Brain.)
Figure 110.2 Overview of bones of the lower limb: anterior aspect. (From Aids to the Examination of the Peripheral Nervous System. 2000. 4th edn. London: Saunders. With permission of Guarantors of Brain.)
© 2008 Elsevier
MUSCLES The effects of developmental extension and medial rotation of the limb are most evident in the relative positions of the muscle groups in the thigh and the leg, and in the adult pattern of segmental innervation (dermatomes) (page 175). In general terms, the anterior aspect of the adult limb is the extensor aspect, while the flexors lie posteriorly; the reverse is true at the hip. As a result of this rotation, the dermatomes tend to spiral around the limb rather than lie parallel to its long axis. The role of the muscles of the lower limb in the maintenance of equilibrium during locomotion and while standing has often been overlooked. Many of the muscles act frequently or predominantly from their distal attachments. During both stance and locomotion, it is often the distal attachment that is fixed and the proximal that is mobile, as in the predominant action of gluteus medius as a pelvic stabilizer rather than as a hip abductor. In contrast, in the upper limb for most of the time the proximal muscle attachments are fixed and the distal attachments are mobile as the hand moves in space. The lower limb contains many muscles that act upon more than one joint: indeed it is unusual for any joint of the lower limb to move in isolation. Muscles of the lower limb may be subdivided into muscles of the iliac region, gluteal region, thigh, leg and foot. (Note that, according to anatomical convention, 'leg' refers to the part of the lower limb between the knee and ankle.) In both thigh and leg, the functional groups of muscles are contained in osteofascial compartments that are particularly well defined and of major clinical significance in the leg. The muscles acting within these closed osteofascial compartments also assist and maintain drainage of the venous system against gravity. The main muscles of the iliac region are psoas major and iliacus, a functional unit comprising the major flexors of the hip and running from the lumbar spine and inner surface of the ilium respectively to the lesser trochanter of the femur. The much less important and inconstantly present psoas minor runs from the lumbar spine to the pubis. The muscles of the gluteal region include the named gluteal muscles and the deeper-lying short lateral rotators of the hip joint. Gluteus maximus lies most superficially, running from the posterior pelvis to the proximal femur and fascia lata. It is a powerful extensor of the hip joint, acting more often to extend the trunk on the femur than to extend the limb on the trunk. Gluteus medius and minimus, attaching proximally to the outer iliac surface and distally to the greater trochanter of the femur, are abductors of the hip whose most important action is to stabilize the pelvis on the femur during locomotion. They are helped in this function by tensor fasciae latae, a more anteriorly placed muscle arising from the anterolateral ilium and inserting into the fascia lata. Two of the short lateral rotators of the hip, piriformis and obturator internus, attach proximally within the pelvis, while the others - obturator externus, the gemelli and quadratus femoris attach externally. All attach distally to the proximal femur. The muscles of the thigh lie in three functional compartments: anterior (extensor),
medial (adductor) and posterior (flexor). The anterior or extensor compartment includes sartorius and the quadriceps group. Sartorius and rectus femoris attach proximally to the pelvis and can thus act on the hip joint as well as the knee. The remaining components of the quadriceps, the vastus muscles, attach proximally to the femur only, and, acting as a unit, are powerful knee extensors. The medial or adductor compartment contains the named adductor muscles and gracilis; pectineus may also be included. These muscles connect the anterior pelvis and the femur: adductor magnus also attaches proximally to the ischial tuberosity. The posterior ('hamstring') compartment includes semitendinosus, semimembranosus and biceps femoris. These muscles attach proximally to the pelvis (ischial tuberosity) and act both to extend the trunk on the femur and to flex and rotate the knee. Adductor magnus, as reflected by the extent of its proximal attachment and by its dual innervation, shares the first of these functions with the hamstrings. Biceps femoris is the only thigh muscle that attaches distally to the fibula: it has no tibial attachment. The muscles of the leg also comprise three functional compartments. The anterior or extensor compartment includes the extensors (dorsiflexors) of the ankle and the long extensors of the toes. Tibialis anterior, the main ankle dorsiflexor, also inverts the foot at the subtalar joint, while the smallest muscle of the compartment, peroneus (fibularis) tertius, is a dorsiflexor that everts the foot. The posterior or flexor (plantarflexor) compartment has superficial and deep components. The superficial component contains the gastrocnemius and soleus, powerful plantar flexors of the ankle, and the small and inconsistent plantaris. All attach distally via the calcaneal (Achilles) tendon. Popliteus, a rotator of the knee, lies most proximally in the deep leg. Gastrocnemius, plantaris and popliteus are the only leg muscles attached proximally to the femur and can thus act on the knee as well as the ankle. The remaining leg muscles attach proximally to the tibia, fibula or both, and to the interosseous membrane. The deep flexor compartment proper contains the long flexors of the toes together with tibialis posterior, the main invertor of the foot. The lateral compartment contains the main evertors of the foot, peroneus (fibularis) longus and brevis: both are also plantar flexors at the ankle. page 1400 page 1401
The muscles of the foot are arranged in layers. They facilitate action of the muscles originating from the leg acting via the long tendons passing through the foot and control foot posture during its various functions.
© 2008 Elsevier
VASCULAR SUPPLY AND LYMPHATIC DRAINAGE ARTERIAL SUPPLY (Figs 110.3, 110.4) The main artery of the lower limb distal to the inguinal ligament and the gluteal fold is the continuation of the external iliac artery, which starts as the femoral artery in the anterior compartment of the thigh. It becomes the popliteal artery in the posterior compartment of the thigh and then divides into its terminal branches in the posterior compartment of the leg. The obturator and inferior gluteal vessels also contribute to the supply of the proximal part of the limb. In the embryo the inferior gluteal artery supplied the main axial artery of the limb, which is represented in the adult by the arteria comitans nervi ischiadici (artery to the sciatic nerve, p. 1470). UPDATE Abstract: The persistence of the sciatic artery.
Date Added: 02 August 2005
Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=15712154&query_hl=3 The persistence of the sciatic artery. The bones of the lower limb receive their arterial supply from nutrient vessels, metaphyseal arterial branches of the peri-articular anastomoses, and the arteries supplying the muscles that attach to their periosteum. The pattern of arterial supply is particularly relevant to fracture healing, the spread of infection and malignancy, and the planning of reconstructive surgical procedures. For further details consult Cormack and Lamberty (1994), Taylor and Razaboni (1994) and Crock (1996).
VENOUS DRAINAGE (Figs 110.5, 110.6, 110.7, 110.8) The veins of the lower limb can be subdivided, like those of the upper limb, into superficial and deep groups. The superficial veins are subcutaneous and lie in the superficial fascia; the deep veins (beneath the deep fascia) accompany the major arteries. Both groups have valves, which are more numerous in the deep veins and also more numerous than in the veins of the upper limb. Venous plexuses occur within and between some of the lower limb muscles. The principal named superficial veins are the long and short saphenous veins. Their numerous tributaries are mainly unnamed. For details and variations consult Kosinski (1926).
Figure 110.3 The anatomical territories served by the cutaneous blood supply to the lower limb. (By permission from Cormack GC, Lamberty BGH 1994 The Arterial Anatomy of Skin Flaps, 2nd edn. Edinburgh: Churchill Livingstone.)
Deep veins of the lower limbs accompany the arteries and their branches. Plantar digital veins arise from plexuses in the plantar regions of the toes, connect with dorsal digital veins and unite into four plantar metatarsal veins. These run in the intermetatarsal spaces and connect by perforating veins with dorsal veins, then continue to form a deep plantar venous arch accompanying the plantar arterial arch. From this arch, medial and lateral plantar veins run near the corresponding arteries: they communicate with the long and short saphenous veins before forming the posterior tibial veins behind the medial malleolus. The posterior tibial veins accompany the posterior tibial artery. They receive veins from the calf muscles, especially the venous plexus in soleus, and connect with superficial veins and with the peroneal veins. The latter, running with their artery, receive branches from soleus and superficial veins. The anterior tibial veins are continuations of the venae comitantes of the dorsalis pedis artery. They leave the extensor region between the tibia and fibula, pass through the proximal end of the interosseous membrane, and unite with the posterior tibial veins to form the popliteal vein at the distal border of popliteus. Venous (muscle) pumps
In a standing position, venous return from the lower limb depends largely on muscular activity, especially contraction of the calf and foot muscles, known as
the 'muscle pump', whose efficiency is aided by the tight sleeve of deep fascia. 'Perforating' veins connect the long saphenous vein with the deep veins, particularly near the ankle, distal calf and knee. Their valves are arranged so as to prevent flow of blood from the deep to the superficial veins. At rest, pressure in a superficial vein is equal to the height of the column of blood extending from that vein to the heart. When calf muscles contract, blood is pumped proximally in the deep veins and is normally prevented from flowing into the superficial veins by the valves in the perforating veins. During relaxation, blood may be aspirated from the superficial into the deep veins. If the valves in the perforating veins become incompetent, these veins become 'high pressure leaks' during muscular contraction, and the superficial veins become dilated and varicose. Similar perforating connections occur in the anterolateral region, where varicosities may also occur. Veins connecting the long saphenous vein to the femoral vein in the adductor canal may become varicose (Dodd & Cockett 1976). Venous plexuses
Venous plexuses may be intramuscular (soleus) or intermuscular (in the foot and gluteal region). The plexuses communicate with the axially running deep veins and are components of the 'muscle pump' mechanism.
LYMPHATIC DRAINAGE (Fig. 110.9) Most lymph from the lower limb traverses a large intermediary inguinal group of nodes. Peripheral nodes are few and all are deeply sited. Except for an inconsistent node lying proximally on the interosseous membrane near the anterior tibial vessels, they occur only in the popliteal fossa. Enlarged popliteal nodes may be palpated along the line of the popliteal vessels while the passively supported knee is gradually moved from extension to semi-flexion. Inguinal nodes are found superficial and deep to the deep fascia. The deep nodes are few and lie alongside the medial aspect of the femoral vein. The superficial nodes may be divided into a lower vertical group that clothe the proximal part of the long saphenous vein, and an upper group that lie parallel to, but below, the inguinal ligament and which are related to the superficial circumflex iliac and superficial external pudendal vessels. Lymph from the lower limb passes from the inguinal nodes to the external and common iliac nodes, and ultimately drains to the lateral aortic group. Deep gluteal lymph reaches the same group through the internal and common iliac chains. Lymphatic drainage of superficial tissues page 1401 page 1402
Figure 110.4 Overview of arteries of the lower limb. A, Anterior. B, Posterior.
The superficial lymph vessels begin in subcutaneous plexuses. Collecting vessels leave the foot medially, along the long saphenous vein, or laterally with the short saphenous vein. Medial vessels are larger and more numerous. They start on the tibial side of the dorsum of the foot, and ascend anterior or posterior to the medial malleolus. Thereafter both sets converge on the long saphenous vein and accompany it to the distal superficial inguinal nodes. Lateral vessels begin on the fibular side of the dorsum of the foot, and some cross anteriorly in the leg to join the medial vessels and so pass to the distal superficial inguinal lymph nodes. Others accompany the short saphenous vein to the popliteal nodes. Superficial lymph vessels from the gluteal region run anteriorly to the proximal superficial inguinal nodes. Lymphatic drainage of deeper tissues
The deep vessels accompany the anterior and posterior tibial, peroneal, popliteal and femoral vessels. The deep vessels from the foot and leg are interrupted by popliteal nodes; those from the thigh pass to the deep inguinal nodes. The deep lymph vessels of the gluteal and ischial regions follow their corresponding blood vessels. Those accompanying the superior gluteal vessels end in a node near the intrapelvic part of the superior gluteal artery, near the superior border of the greater sciatic foramen, while those which follow the inferior gluteal vessels traverse one or two of the small nodes below piriformis and pass to the internal iliac nodes.
© 2008 Elsevier
INNERVATION Overview of the plexuses
The lumbar and sacral plexuses innervate the lower limb. The lumbar plexus lies deep within psoas major, anterior to the transverse processes of the first three lumbar vertebrae. The sacral plexus lies in the pelvis on the anterior surface of piriformis, deep to the pelvic fascia, which separates it from the inferior gluteal and pudendal vessels. The lumbosacral trunk (L4 and L5) emerges medial to psoas major and lies on the ala of the sacrum before crossing the pelvic brim to join the anterior primary ramus of S1.
LESIONS OF THE LUMBAR AND SACRAL PLEXUSES page 1402 page 1403
Figure 110.5 Overview of veins of the lower limb.
The deep and protected situation of the plexuses means that lesions are not common. The lumbar plexus may be involved in retroperitoneal pathology, and the sacral plexus may be invaded by spreading pelvic malignancy. Both may be involved in the reticuloses, affected by plexiform neuromas, or damaged in fractures of the lumbar spine and pelvis or in other conditions that cause severe retroperitoneal and pelvic haemorrhage. Temporary lesions may occur after pregnancy and childbirth, e.g. after difficult forceps delivery of a large baby. Pain, which may be diffuse, is the most common feature, and there is often clinical involvement of several roots.
OVERVIEW OF THE PRINCIPAL NERVES OF THE LOWER LIMB (Figs 110.10, 110.11) Femoral nerve (L2-4)
Figure 110.6 The long saphenous vein and its tributaries.
The femoral nerve is the nerve of the anterior compartment of the thigh. It arises from the posterior divisions of the second to fourth lumbar ventral rami, descends through psoas major and emerges on its lateral border to pass between psoas and iliacus and enter the thigh behind the inguinal ligament and lateral to the femoral sheath. Its terminal branches form in the femoral triangle c.2 cm distal to the inguinal ligament. In the abdomen the nerve supplies small branches to iliacus and a branch to the proximal part of the femoral artery. It subsequently supplies a
large cutaneous area on the anterior and medial thigh, medial leg and foot, and gives articular branches to the hip and knee. The femoral nerve is described in detail on page 1455. Obturator nerve (L2-4)
The obturator nerve is the nerve of the medial compartment of the thigh. It arises from the anterior divisions of the second to fourth lumbar ventral rami, descends through psoas major and emerges from its medial border at the pelvic brim. It crosses the sacroiliac joint behind the common iliac artery and lateral to the internal iliac vessels, runs along the lateral pelvic wall on obturator internus, and enters the thigh through the upper part of the obturator foramen. Near the foramen it divides into anterior and posterior branches, separated at first by part of obturator externus and more distally by adductor brevis. It gives articular branches to the hip and knee, and may supply skin on the medial thigh and leg. The obturator nerve is described in detail on page 1455. Sciatic nerve (L4, L5, S1-3) page 1403 page 1404
Figure 110.7 The short saphenous vein and its tributaries.
The sciatic nerve is the nerve of the posterior compartment of the thigh and, via its major branches, of all the compartments of the lower leg and foot. Formed in the pelvis from the ventral rami of the fourth lumbar to third sacral spinal nerves, it is 2 cm wide at its origin and is the thickest nerve in the body. It enters the lower limb via the greater sciatic foramen below piriformis and descends between the greater trochanter and ischial tuberosity. The nerve passes along the back of the
thigh, where it is crossed by the long head of biceps femoris, and divides into the tibial and common peroneal (fibular) nerves proximal to the knee. The actual level of division is very variable since the tibial and common peroneal nerves are structurally separate and only loosely connected throughout their proximal course. The sciatic gives off articular branches that supply the hip joint through its posterior capsule (these are sometimes derived directly from the sacral plexus) and the knee joint. All the hamstring muscles, including the ischial part of adductor magnus, but not the short head of biceps femoris, are supplied by the medial (tibial) component of the sciatic nerve. The short head of biceps is supplied by the lateral (common peroneal) component. The sciatic nerve is described in detail on page 1456. UPDATE Abstract: Sciatic nerve varices.
Date Added: 02 August 2005
Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=16023386&query_hl=6 Sciatic nerve varices. Tibial nerve (L4, L5, S1-3)
Figure 110.8 Venogram of the leg to show the deep veins; the valves are clearly demonstrated. (Provided by Sean Gallagher, GKT School of Medicine, London; photograph by Sarah-Jane Smith.)
The tibial nerve arises from the anterior division of the sacral plexus. It descends along the back of the thigh and popliteal fossa to the distal border of popliteus, then passes anterior to the arch of soleus with the popliteal artery and continues into the leg. In the popliteal fossa it lies lateral to the popliteal vessels, becomes superficial to them at the knee and crosses to the medial side of the artery. In the leg it is the nerve of the posterior compartment and descends with the posterior
tibial vessels to lie between the heel and the medial malleolus. It ends beneath the flexor retinaculum by dividing into the medial and lateral plantar nerves. The tibial nerve supplies articular branches to the knee and ankle. Its cutaneous area of supply, including its terminal branches, includes the back of the calf, the whole of the sole, the lateral border of the foot and the medial and lateral sides of the heel. The tibial nerve is described in detail on page 1504. Common peroneal nerve (L4, L5, S1, S2)
The common peroneal nerve (common fibular nerve) is derived from the posterior division of the sacral plexus. In the leg it is the nerve of the anterior and lateral compartments. It descends obliquely along the lateral side of the popliteal fossa to the fibular head, lying between the tendon of biceps femoris and the lateral head of gastrocnemius. It curves lateral to the neck of the fibula deep to peroneus longus and divides into superficial and deep peroneal (fibular) nerves: the common peroneal nerve is easily injured at the fibular neck. Before it divides, it gives off articular branches to the knee and the superior tibiofibular joints and cutaneous branches. Its cutaneous area of supply, including its terminal branches, includes the anterolateral and lateral surfaces of the leg and most of the dorsum of the foot. The common peroneal nerve is described in detail on page 1504. page 1404 page 1405
Figure 110.9 Lymphatics of the lower limb.
Figure 110.9 Lymphatics of the lower limb.
Gluteal nerves
The gluteal nerves arise from the posterior division of the sacral plexus. The superior gluteal nerve (L4, L5, S1) leaves the pelvis through the greater sciatic notch above piriformis and supplies gluteus medius, gluteus minimus, tensor fasciae latae and the hip joint. The inferior gluteal nerve (L5, S1, S2) passes through the greater sciatic notch below piriformis and supplies gluteus maximus. The gluteal nerves are described in detail on page 1456.
AUTONOMIC INNERVATION
Figure 110.10 Nerves on the anterior aspect of the lower limb, their cutaneous branches, and the muscles they supply. (From Aids to the Examination of the Peripheral Nervous System. 2000. 4th edition. London: Saunders. With permission of Guarantors of Brain.)
The autonomic supply to the limbs is exclusively sympathetic. The preganglionic sympathetic inflow to the lower limb is derived from neurones in the lateral horn of the lower thoracic (T10, T11) and upper lumbar (L1, L2) spinal cord segments. Fibres pass in white rami communicantes to the sympathetic chain and synapse in the lumbar and sacral ganglia. Postganglionic fibres pass in grey rami communicantes to enter the lumbar and sacral plexuses, and many are distributed via the cutaneous branches of the nerves derived from these plexuses. The blood vessels to the lower limb receive their sympathetic supply via adjacent peripheral nerves. Postganglionic fibres accompanying the iliac arteries are destined mainly for the pelvis but may supply vessels in the upper thigh. Surgical lumbar sympathectomy may be indicated in arterial disease. Surgical or chemical (phenol injection) sympathectomy may be used to treat rest pain or other troublesome sensory symptoms in arterial disease or in causalgia. The segment of the chain including the second and third lumbar ganglia is removed: preservation of the first lumbar ganglion is said to lessen the risk of ejaculatory problems. page 1405 page 1406
Figure 110.11 Nerves on the posterior aspect of the lower limb, their cutaneous branches, and the muscles they supply. (From Aids to the Examination of the Peripheral Nervous System. 2000. 4th edn. London: Saunders. With permission of Guarantors of Brain.)
© 2008 Elsevier
SURFACE ANATOMY SKELETAL LANDMARKS Pelvis (Fig. 110.12)
Figure 110.12 Inguinal region (bones and soft tissues) and femoral triangle (vessels and nerves). (Photograph by Sarah-Jane Smith. Artwork modified from Lumley JSP 2002 Surface Anatomy, 3rd edn. Edinburgh: Churchill Livingstone.)
An oblique skin crease, the fold of the groin, marks the junction of the front of the thigh with the anterior abdominal wall, and corresponds fairly accurately to the inguinal ligament. The anterior superior iliac spine lies at the lateral end of the fold and can always be palpated. At its medial end, the fold reaches the pubic tubercle. From the anterior superior iliac spine, the iliac crest is easily palpable along its entire length. It terminates posteriorly as the posterior superior iliac spine, which may be felt in the depression seen just above the buttock. This depression lies at the level of the second sacral segment, at the level of the middle of the sacroiliac joint and the termination of the spinal dural sac. The ischial tuberosity may be palpated in the lower part of the buttock. It is covered by gluteus maximus when the hip joint is extended, but can be identified without difficulty when the hip is flexed, e.g. in the sitting position, when the tuberosity emerges from under cover of the lower border of gluteus maximus and becomes subcutaneous, separated from the skin only by a pad of fat and the ischial bursa. In this position, the weight of the body is supported by the ischial tuberosities. Femur (Figs 110.13, 110.14, 110.15)
The greater trochanter of the femur, which is the only palpable part of the proximal portion of the femur, lies the breadth of the subject's hand below the midpoint of the iliac crest. It can be both seen and felt as a prominence in front of the hollow on the side of the gluteal region. The lower end of the femur is less deeply placed. When the knee is flexed passively, the medial surface of the medial condyle and the lateral surface of the lateral condyle of the femur may both be palpated, and portions of the femoral articular surface can be examined on each side of the lower part of the patella.
Patella
The patella can be readily identified. When the quadriceps is relaxed in the fully extended knee, the patella can be tilted and moved on the lower end of the femur. The lower limit of the patella lies more than 1-2 cm above the line of the knee joint. Knee (Figs 110.16, 110.17) page 1406 page 1407
Figure 110.13 Lateral aspect of the hip joint: bones. (Photograph by Sarah-Jane Smith. Artwork modified from Lumley JSP 2002 Surface Anatomy, 3rd edn. Edinburgh: Churchill Livingstone.)
The patella and the medial and lateral condyles of the femur have been described above. The tibial condyles form visible and palpable landmarks at the medial and lateral sides of the patellar tendon. The latter may be traced downwards from the apex of the patella to the tibial tubercle, which is easily seen as well as felt. When the knee is flexed, the anterior margins of the tibial condyles can be felt: each forms the lower boundary of a depression at the side of the patellar tendon. The lateral condyle is the more prominent of the two. The joint line of the tibiofemoral joint corresponds to the upper margins of the tibial condyles and can be represented by a line drawn round the limb at this level. The anterior horns of the menisci lie in the angles between this line and the edges of the patellar tendon. The iliotibial tract is attached to a prominence, Gerdy's tubercle, on the anterior aspect of the lateral condyle, 1 cm below the joint line and c.2 cm lateral to the tibial tubercle. The head of the fibula forms a slight surface elevation on the upper part of the posterolateral aspect of the leg. It lies vertically below the posterior part of the lateral condyle of the femur, not less than 1 cm below the level of the knee joint. Leg and ankle (Fig. 110.18)
The subcutaneous medial surface of the tibia corresponds to the flat anteromedial aspect of the leg. Above, this surface merges into the medial condyle of the tibia, and below it is continuous with the visible prominence of the medial malleolus of
the tibia. The sinuous anterior border of the tibia can be felt distinctly throughout most of its extent, but inferiorly it is somewhat masked by the tendon of tibialis anterior, which lies to its lateral side. The lateral malleolus of the fibula forms a conspicuous projection on the lateral side of the ankle: it descends to a more distal level than the medial malleolus and is placed on a more posterior plane. The lateral aspect of the lateral malleolus is continuous above with an elongated, subcutaneous, triangular area of the lower shaft of the fibula. The lateral part of the anterior margin of the lower end of the tibia can be detected immediately in front of the base of the lateral malleolus; the line of the ankle joint can be gauged from it. Foot (Figs 110.19, 110.20)
Figure 110.14 Gluteal region and posterior aspect of the thigh: bones. (Photograph by Sarah-Jane Smith. Artwork modified from Lumley JSP 2002 Surface Anatomy, 3rd edn. Edinburgh: Churchill Livingstone.)
On the dorsum of the foot, the anterior part of the upper surface of the calcaneus can be identified a little in front of the lateral malleolus. When the foot is passively inverted, the upper and lateral part of the head of the talus can be both seen and felt 3 cm anterior to the distal end of the tibia; it is obscured by the extensor tendons when the toes are dorsiflexed. The dorsal aspects of the bodies of the metatarsal bones can be felt more or less distinctly, although they tend to be obscured by the extensor tendons of the toes. The tuberosity on the base of the fifth metatarsal bone forms a distinct projection, which can be both seen and felt, half-way along the lateral border of the foot. The flat lateral surface of the calcaneus can be palpated on the lateral aspect of the heel and can be traced forwards below the lateral malleolus, where it is hidden by the tendons of peroneus longus and brevis. The peroneal tubercle, when sufficiently large, can be felt 2 cm below the tip of the lateral malleolus. A palpable depression just anterior to the lateral malleolus leads to the lateral end of
the sinus tarsi. On the medial side of the foot, the sustentaculum tali of the calcaneus can be felt 2 cm vertically below the medial malleolus. The medial aspect of the calcaneus can be felt (indistinctly) below and behind the sustentaculum tali. The most conspicuous bony landmark on the medial side of the foot is the tuberosity of the navicular bone, which is usually visible and can always be felt 2.5 cm anterior to the sustentaculum tali. Anterior to this, the medial cuneiform bone can be identified by tracing the tendon of tibialis anterior into it. The upper and medial parts of the joint between the medial cuneiform and the first metatarsal can be felt as a narrow groove. page 1407 page 1408
Figure 110.15 Anterior and medial aspect of the thigh: bones. (Photograph by SarahJane Smith. Artwork modified from Lumley JSP 2002 Surface Anatomy, 3rd edn. Edinburgh: Churchill Livingstone.)
When the foot is placed on the ground, it rests on the posterior part of the inferior surface of the calcaneus, the heads of the metatarsal bones and, to a lesser extent, on its lateral border. The instep, which corresponds to the medial longitudinal arch of the foot, is elevated from the ground. The medial and lateral tubercles of the calcaneus can be identified on the posterior part of the inferior surface of the calcaneus, but they are obscured by the tough fibro-fatty pad that covers them. The heads of the metatarsal bones are similarly covered by a thick pad, which forms the ball of the foot. The foot is at its widest at this level, reflecting the slight splay of the metatarsal bones as they pass anteriorly. The calcaneocuboid joint lies 2 cm behind the tubercle on the base of the fifth metatarsal bone and is practically in line with the talonavicular joint, whose position may be gauged from the position of the head of the talus. The tarsometatarsal joints lie on a line joining the tubercle of the fifth metatarsal bone to the tarsometatarsal joint of the great toe. When the latter joint cannot be felt on
the medial border of the foot, its position may be indicated 2.5 cm in front of the tuberosity of the navicular bone. The joint between the second metatarsal and the intermediate cuneiform lies some 2-3 mm behind the line of the other tarsometatarsal joints. The metatarsophalangeal joints lie 2.5 cm behind the webs of the toes.
MUSCULOTENDINOUS AND LIGAMENTOUS LANDMARKS Buttock and hip (Figs 110.13, 110.21)
The bulky prominence of the buttock is caused by the forward tilt of the pelvis (which throws the ischium backwards), the size of gluteus maximus, and the large amount of subcutaneous fat. The horizontal gluteal fold marks the upper limit of the posterior aspect of the thigh. It does not correspond to the lower border of gluteus maximus but is caused by fibrous connections between the skin and the deep fascia. The natal cleft, which separates the buttocks inferiorly, starts above at the third or fourth sacral spine.
Figure 110.16 Medial aspect of the flexed knee: bone and muscles. (Photograph by Sarah-Jane Smith. Artwork modified from Lumley JSP 2002 Surface Anatomy, rd edn. Edinburgh: Churchill Livingstone.)
page 1408 page 1409
Figure 110.17 Lateral aspect of the flexed knee: bone and soft tissues. (Photograph by Sarah-Jane Smith. Artwork modified from Lumley JSP 2002 Surface Anatomy, 3rd edn. Edinburgh: Churchill Livingstone.)
Figure 110.18 Anterior aspect of the lower leg: bones. (Photograph by Sarah-Jane Smith. Artwork modified from Lumley JSP 2002 Surface Anatomy, 3rd edn. Edinburgh: Churchill Livingstone.)
Figure 110.19 Left foot and ankle, lateral view. (Photograph by Sarah-Jane Smith.)
Figure 110.20 Sole of the foot: bones. (Photograph by Sarah-Jane Smith. Artwork modified from Lumley JSP 2002 Surface Anatomy, 3rd edn. Edinburgh: Churchill Livingstone.)
page 1409 page 1410
Figure 110.21 Gluteal region and posterior aspect of the thigh: superficial (left limb) and deep (right limb) muscles. (Photograph by Sarah-Jane Smith. Artwork modified
from Lumley JSP 2002 Surface Anatomy, 3rd edn. Edinburgh: Churchill Livingstone.)
The upper border of gluteus maximus begins on the iliac crest c.3 cm lateral to the posterior superior iliac spine and runs downwards and laterally to the apex of the greater trochanter. Its lower border corresponds to a line drawn from the ischial tuberosity, through the midpoint of the gluteal fold, to a point c.9 cm below the greater trochanter. Although gluteus maximus overlaps the ischial tuberosity in the standing position, on sitting it slides superiorly posterior to the tuberosity, leaving it free to bear weight. The muscle can be felt to contract when the hip is extended against resistance. Gluteus medius completely covers the underlying gluteus minimus. Both muscles lie in a slight depression superolateral to gluteus maximus and inferior to the anterior portion of the iliac crest. They constitute the major abductors of the hip and are demonstrated by asking the subject to stand on one limb. The ipsilateral muscles contract and tilt the pelvis in order to stabilize the centre of gravity, and the contralateral gluteal fold will rise. If the hip abductors are paralysed, e.g. in congenital dislocation of the hip or in a long-standing fracture of the neck of the femur, this mechanism is disturbed and the normal tilting of the pelvis does not occur. Indeed, when the patient stands on the affected hip, the pelvis tilts downwards on the contralateral side (Trendelenburg's sign). Thigh (Figs 110.12, 110.13, 110.16, 110.17, 110.22)
The inguinal ligament can be felt running between the anterior superior iliac spine and the pubic tubercle when the thigh is abducted and externally rotated. Just distal to the inguinal ligament, but running horizontally, is the hip flexure line (Holden's line), where the deep layer of superficial fascia of the abdominal wall meets the fascia lata of the thigh.
Figure 110.22 Anterior and medial aspect of the thigh: superficial (right limb) and deep (left limb) muscles. (Photograph by Sarah-Jane Smith. Artwork modified from Lumley JSP 2002 Surface Anatomy, 3rd edn. Edinburgh: Churchill Livingstone.)
The shallow depression lying immediately below the fold of the groin corresponds
to the femoral triangle (p. 1419). It is bounded on its lateral side by the strap-like sartorius. The latter can be both seen and felt in a reasonably thin and muscular subject when the hip is flexed in the sitting position, while keeping the knee extended, especially when the thigh is slightly abducted and rotated laterally. The muscle can be traced downwards and medially from the anterior superior iliac spine to approximately half-way down the medial side of the thigh. Distally, it may be identified as a soft longitudinal ridge passing towards the posterior part of the medial femoral condyle. The adductor group of muscles forms the bulky, fleshy mass at the upper part of the medial side of the thigh. The medial border of adductor longus forms the medial boundary of the femoral triangle and can be felt as a distinct ridge when the thigh is adducted against resistance. At its upper end, its tendon of origin immediately below the pubic tubercle can be identified and felt between the finger and thumb, which is a useful guide to this bony landmark. The forward convexity of the front of the thigh is caused by the curvature of the femur covered by the fleshy mass of quadriceps femoris. Rectus femoris appears as a ridge passing down the anterior aspect of the thigh when the sitting subject flexes the hip with the knee extended. Vastus medialis constitutes the bulge above and medial to the patella. Vastus lateralis forms the elevation above and lateral to the patella, more proximal and less pronounced than that of vastus medialis. Vastus intermedius is hidden by the other three muscles. The flattened appearance of the lateral aspect of the thigh is produced by the iliotibial tract, which is a thickened portion of the deep fascia of the thigh (fascia lata). It stands out as a strong, visible ridge on the lateral aspect of the knee when the knee is either extended against gravity or when the opposite limb is lifted from the floor while standing. Knee (Fig. 110.23)
page 1410 page 1411
Figure 110.23 Popliteal fossa: soft tissues. (By permission from Lumley JSP 2002 Surface Anatomy, 3rd edn. Edinburgh: Churchill Livingstone.)
The large depression that can be seen at the back of the knee when the joint is
actively flexed against resistance corresponds to the popliteal fossa. The transverse skin crease of the popliteal fossa is 2-3 cm above the tibiofemoral joint line. The fossa is bounded on the lateral side by the prominent tendon of biceps femoris, which can be felt between the finger and thumb and can be traced downwards to the head of the fibula. Three tendons can be felt on the medial side of the fossa. Semitendinosus is the most lateral and posterior, and gracilis is the most medial and anterior. These two tendons stand out sharply and can be seen when the knee is flexed against resistance and the limb actively adducted. The third tendon is that of semimembranosus: it is more deeply situated and can be felt in the interval between the tendons of semitendinosus and gracilis. It is much thicker than the other two tendons and broadens rapidly as it is traced upwards. Distally, in thin individuals, the upper border of the pes anserinus can be palpated. The upper borders of the two heads of gastrocnemius form the medial and lateral inferior boundaries of the popliteal fossa. The lateral collateral ligament may be felt passing from the tip of the fibular head to the lateral epicondyle of the femur when the knee is flexed and laterally directed pressure is applied to the medial side of the knee. The medial patellofemoral ligament may be felt overlying the medial femoral condyle in the flexed knee, running between the midpoint of the medial patella and medial femoral epicondyle. Leg (Figs 110.16, 110.17, 110.18, 110.24, 110.25, 110.26, 110.27)
Figure 110.24 Anterior aspect of the lower leg: muscles. (Photograph by Sarah-Jane Smith. Artwork modified from Lumley JSP 2002 Surface Anatomy, 3rd edn. Edinburgh: Churchill Livingstone.)
The muscles in the anterior osteofascial compartment of the leg form a gentle prominence over the upper two-thirds of its anterolateral aspect: this prominence is accentuated when the foot is actively dorsiflexed. In the lower third of the leg
these muscles are replaced by their tendons. The tendon of tibialis anterior can be seen just lateral to the anterior border of the tibia and traced downwards and medially across the front of the ankle. The other tendons cannot be examined satisfactorily above the ankle. Immediately above the medial malleolus and close to the medial border of the tibia, the tendons of tibialis posterior and flexor digitorum longus can be felt (rather indistinctly) when the foot is actively inverted and plantar flexed.
Figure 110.25 Left calf and ankle, posterior view with foot plantigrade. (Photograph by Sarah-Jane Smith.)
page 1411 page 1412
Figure 110.26 Left leg and foot, lateral view with ankle dorsiflexed. (Photograph by Sarah-Jane Smith.)
Figure 110.27 Left leg and foot, ankle plantar flexed, lateral view. (Photograph by Sarah-Jane Smith.)
On the lateral aspect of the leg, peroneus longus can be seen as a narrow ridge during active eversion and plantarflexion of the foot. It covers and hides peroneus brevis. Both muscles cover the lateral aspect of the fibula, which means that the shaft of the fibula can only be palpated indistinctly between its neck and its lower subcutaneous triangular area. The bulky prominence of the calf of the leg is formed by gastrocnemius and soleus, both of which can be identified either when the foot is plantar flexed against resistance, or when the heel is raised from the ground by standing on tiptoes. The two heads of gastrocnemius unite to form the inferior angle of the popliteal fossa. The medial head of gastrocnemius descends to a lower level than its lateral head. Soleus lies deep to gastrocnemius; when tensed, it bulges from under gastrocnemius, particularly on the lateral side, and its fleshy belly extends to a more distal level. Both muscles end below in the conspicuous calcaneal tendon, which can be gripped between the finger and thumb and followed downwards to its insertion into the posterior aspect of the calcaneus. Foot (Figs 110.19, 110.20, 110.26, 110.27)
When the toes are actively dorsiflexed, the belly of extensor digitorum brevis forms a small elevation on the dorsum of the foot, a little in front of the lateral malleolus. It is the only muscle that arises from the dorsum of the foot. The tendon of tibialis anterior stands out conspicuously on the medial side when active dorsiflexion of the toes is combined with inversion: it can be traced downwards and medially to the medial cuneiform bone. Also in dorsiflexion, the tendon of extensor hallucis longus can be identified lateral to tibialis anterior. Still more laterally, and immediately in front of the lateral part of the inferior end of the tibia, the tendons of extensor digitorum longus and peroneus tertius are crowded together as they pass through the fibrous loop of the inferior extensor retinaculum. More distally, these tendons diverge to their insertions. The tendon of tibialis posterior winds posterior to the medial malleolus and then curves forwards in the interval between this bony landmark and the sustentaculum tali to reach the tuberosity of the navicular bone. The tendon is thrown into relief when the foot is forcibly plantar flexed and inverted. The tendon of flexor digitorum longus lies midway between the medial margin of the calcaneal tendon and the margin of the medial malleolus. It curves forwards below tibialis posterior and lies on the medial aspect of the sustentaculum tali. From there, it passes forwards and laterally to the centre of the sole of the foot, where it breaks
up into tendons for the lateral four toes. The tendon of flexor hallucis longus lies inferior to, and grooves, the sustentaculum tali. As it passes forward to the great toe, it crosses the line of the flexor digitorum longus opposite the interval between the sustentaculum tali and the tuberosity of the navicular bone. Abductor hallucis may be seen in some subjects as a fleshy mass across the instep of the foot, passing from the medial calcaneal tubercle to the ball of the great toe. The plantar fascia is most easily palpated along its medial border, in the sole of the foot when the toes are maximally dorsiflexed. Its precise origin is difficult to palpate because of the thickness of the heel pad, but it is easily palpated just distal to the heel pad as far as the ball of the foot.
SURFACE MARKINGS OF VESSELS, PULSES AND NERVES Arteries
The femoral artery enters the thigh at the fold of the groin, at a point midway between the anterior superior iliac spine and the pubic symphysis, directly anterior to the hip joint. Its course can be represented by the upper two-thirds of a line joining that point to the adductor tubercle when the flexed thigh is abducted slightly and rotated laterally. The popliteal artery may be represented approximately by a line extending from the junction of the middle and lower thirds of the thigh, 2.5 cm medial to its posterior midline, to the midpoint between the femoral condyles, and continuing inferolaterally to the level of the tibial tuberosity, medial to the fibular neck. The artery bifurcates into the anterior and posterior tibial arteries at the lower border of popliteus. The anterior tibial artery may be represented by a line that begins 2.5 cm inferior to the medial side of the head of the fibula and runs downwards and slightly medially to the midpoint between the two malleoli. The posterior tibial artery corresponds to a line drawn from the level of the neck of the fibula to a point midway between the medial malleolus and calcaneal tendon, along the midline of the back of the calf. The same line represents the course of the tibial nerve. At first the nerve lies lateral to the popliteal artery, then gradually crosses the vessel to gain its medial side. The trunk of the medial plantar artery begins midway between the medial malleolus and heel (medial calcaneal tubercle) and extends towards the first interdigital cleft as far as the navicular bone. Pulses
Femoral artery pulse At its origin the pulsations of the femoral artery can be felt, and in this situation the vessel can be compressed against the superior ramus of the pubis or the head of the femur. Like the carotid pulse, the femoral pulse is of value in assessing whether there is any significant cardiac output in cases of circulatory collapse. It is a common site for radiological catheter insertion and for arterial puncture for blood gas analysis. Popliteal artery pulse The pulse of the popliteal artery is the most difficult of the peripheral pulses to feel because the artery lies deep in the popliteal fossa. It is best examined with the subject lying supine or prone, with the knee flexed, in order to relax the tense popliteal fascia that roofs the popliteal fossa. The popliteal pulse is then felt by deep pressure over the midline of the fossa against the popliteal surface of the femur. Posterior tibial artery pulse The pulse of the posterior tibial artery can be felt by gentle palpation behind the medial malleolus as the artery lies between the tendons of flexor hallucis longus and flexor digitorum longus. Dorsalis pedis arterial pulse The dorsalis pedis arterial pulse is found by palpation against the underlying tarsal bones immediately lateral to the tendon of extensor hallucis longus. Veins
The surface marking for the femoral vein is immediately medial to the femoral pulse. The long saphenous vein terminates c.2 cm inferior to the femoral pulse.
Its course may be marked out in the thigh by a line passing from this point downwards and backwards to a point the breadth of the subject's hand behind the medial border of the patella. page 1412 page 1413
The long saphenous vein can often be seen as it runs upwards and backwards across the medial surface of the tibia a little above and in front of the medial malleolus: this is a useful site for obtaining surgical venous access or for harvesting vein for cardiac bypass surgery. It is accompanied in this region by the saphenous nerve, which is usually anterolateral, but may be posterior to, the vein. The vein and nerve then run proximally and posteriorly along the medial border of the tibia to reach a point the breadth of the subject's hand posterior to the medial border of the patella. The short saphenous vein may be represented by a line which runs from the posterior surface of the lateral malleolus up the midline of the calf to the popliteal fossa: with the sural nerve it is the key anatomical guide to surgical dissection of the popliteal fossa. The dorsal venous arch forms a conspicuous feature on the dorsum of the foot and curves, convex forwards, across the metatarsus. The long saphenous vein arises from its medial end and runs upwards and backwards immediately in front of the medial malleolus. The short saphenous vein arises from the lateral end of the arch and passes backwards and inferior to, and then upwards and posterior to, the lateral malleolus. Femoral venous cannulation
While femoral venous puncture is relatively easy and supplies ready access to the right atrium, the use of this approach is relatively unpopular for long-term cannulation because of a higher incidence of thrombosis and sepsis. It is, however, a useful site for venous sampling in a patient with collapsed veins. For femoral venous cannulation the skin puncture site is approximately 1 cm medial to the femoral artery and just below the inguinal ligament. After skin puncture the needle is advanced with the syringe at an angle of 30° to the skin, aiming cephalad. Nerves (Fig. 110.28)
The surface marking for the femoral nerve in the inguinal region is immediately lateral to the femoral pulse. The surface marking for the course of the saphenous nerve is described with the course of the long saphenous vein (see above). The course of the sciatic nerve can be represented by a line that starts at a point midway between the posterior superior iliac spine and ischial tuberosity, curves outwards and downwards through a point midway between the greater trochanter and the ischial tuberosity, and then continues vertically downwards in the midline of the posterior aspect of the thigh to the upper angle of the popliteal fossa, where it divides into the tibial and common peroneal nerves (if it has not already done so at a higher level).
Figure 110.28 Surface markings of the sciatic nerve. The line of the nerve joins the midpoint between the ischial tuberosity and the posterior superior iliac spine with the midpoint between the ischial tuberosity and the greater trochanter and then continues vertically down the back of the thigh. (From Ellis H, Feldman S 1997 Anatomy for Anaesthetists, 7th edn. Oxford: Blackwell Science. By permission of Blackwell Publishing.)
The course of the common peroneal nerve can be indicated by a line that runs from the upper angle of the popliteal fossa, along the medial side of the tendon of biceps femoris and then curves downwards and forwards around the neck of the fibula. The nerve is palpable medial and, more distally to the tendon of biceps and over the neck of the fibula, although here it is less distinct as it passes deep to the origin of peroneus longus. At the neck of the fibula the nerve is at particular risk of damage either from a tightly applied plaster cast or from a fracture. The deep peroneal nerve starts on the lateral aspect of the neck of the fibula, passes downwards and medially, and rapidly becomes associated with the anterior tibial artery, which it accompanies to the ankle. The superficial peroneal nerve also begins on the lateral aspect of the neck of the fibula. It descends to a point on the anterior border of peroneus longus, at the junction of the middle and lower thirds of the leg, where it pierces the deep fascia and divides into medial and lateral branches. The latter gradually diverge as they descend to reach the dorsum of the foot and may be seen if the foot is pulled into maximal plantarflexion and adduction. With one exception, the cutaneous nerves of the foot are not normally visible. The superficial peroneal nerve is usually easily seen and palpated over the dorsolateral aspect of the ankle and midfoot when the ankle is plantar flexed and the fourth toe passively flexed. This is particularly true in individuals with little subcutaneous fat. Whilst the lateral branch is usually seen, the medial branch is rarely visible. These superficial nerves are at risk from surgery in the anterior ankle region (especially during arthroscopy): they can be identified reliably by passive plantarflexion and inversion of the foot whilst running the blunt end of a ballpoint pen across the anterior ankle. An easily palpable click will localize the nerves, even in patients who are not thin. Dermatomes (Figs 110.29, 110.30, 110.31)
Our knowledge of the extent of individual dermatomes, especially in the limbs, is largely based on clinical evidence. The dermatomes of the lower limb arise from spinal nerves T12 to S3.
page 1413 page 1414
Figure 110.29 Dermatomes of the lower limb. There is considerable variation and overlap between dermatomes, but the overlap across axial lines (heavy black) is minimal.
Figure 110.30 Dermatomes of the perineum.
The preaxial border starts near the midpoint of the thigh and descends to the knee. It then curves medially, descending to the medial malleolus and the medial side of the foot and hallux. The postaxial border starts in the gluteal region and descends to the centre of the popliteal fossa, then deviates laterally to the lateral malleolus and the lateral side of the foot. The ventral and dorsal axial lines exhibit corresponding obliquity. The ventral starts proximally at the medial end of the inguinal ligament and descends along the posteromedial aspect of the thigh and leg to end proximal to the heel. The dorsal axial line begins in the lateral gluteal region and descends posterolaterally in the thigh to the knee; it inclines medially and ends proximal to the ankle. Considerable overlap exists between adjacent
dermatomes innervated by nerves derived from consecutive spinal cord segments. Myotomes
Tables 110.1 to 110.4 summarize the predominant segmental origin of the nerve supply for each of the lower limb muscles and for movements that take place at the joints of the lower limb: damage to these segments or to their motor roots results in maximum paralysis. Data are based chiefly on clinical evidence (Sharrad 1995). Reflexes
Knee jerk (L2-4) With the patient supine and the knee supported and partially flexed, the patellar tendon is struck at its midpoint: this should elicit contraction of the quadriceps, which extends the knee. Ankle jerk (S1, 2) With the patient supine and the lower limb externally rotated and partially flexed at hip and knee, the foot is passively dorsiflexed to stretch the calcaneal tendon, which is then struck with a percussion hammer. Contraction of the calf muscles plantar flexes the ankle. The reflex can also be examined with the patient kneeling on a chair.
page 1414 page 1415
Figure 110.31 The cutaneous nerves of the right lower limb, their areas of distribution and segmental origins. A, Anterior aspect; B, sole of foot; C, posterior aspect. In C, the interrupted line represents the trunk of the posterior cutaneous nerve of the thigh, most of which lies deep to the fascia lata.
Table 110-1. Movements, muscles and segmental innervation in the lower limb Joint Movement Muscle Innervation L1 L2 L3 L4 L5 S1 S2 S3 HIP FLEXION Psoas major Spinal nn. L1-3 Iliacus Femoral n. Pectineus Femoral n. Rectus femoris Femoral n. Adductor longus Obturator n. Sartorius Femoral n. EXTENSION Gluteus maximus Inferior gluteal n. Adductor magnus Obturator & tibial nn. Hamstrings Mainly tibial nn. MEDIAL Iliacus Femoral n. ROTATION Gluteus medius & Superior minimus gluteal n. Tensor fasciae Superior latae gluteal n. LATERAL Superior & inferior Lumbosacral ROTATION gemelli plexus Quadratus femoris Lumbosacral plexus Piriformis Lumbosacral plexus Obturator internus Lumbosacral plexus Obturator externus Obturator n. Sartorius Femoral n. ADDUCTION Gracilis Obturator n. Adductor longus Obturator n. Adductor magnus Obturator & tibial nn. Adductor brevis Obturator n. Pectineus Femoral n. ABDUCTION Tensor fasciae Superior gluteal n. latae Gluteus medius & Superior gluteal n. minimus Piriformis Lumbosacral plexus KNEE FLEXION Hamstrings: Semimembranosus Tibial n. Semitendinosus Tibial n. Biceps femoris Tibial & common peroneal nn. Gastrocnemius Tibial n. EXTENSION Quadriceps femoris: Rectus femoris Femoral n. Vastus lateralis Femoral n. Vastus intermedius Femoral n. Vastus medialis Femoral n.
ANKLE
INVERSION EVERSION
TOES
EXTENSION
ABDUCTION
ADDUCTION
DORSIFLEXION
Tibialis anterior
Deep peroneal n. Extensor digitorum Deep longus peroneal n. Extensor hallucis Deep longus peroneal n. Peroneus tertius Deep peroneal n. PLANTARFLEXION Gastrocnemius Tibial n. Soleus Tibial n. Flexor digitorum Tibial n. longus Flexor hallucis Tibial n. longus Peroneus longus Superficial peroneal n. Tibialis posterior Tibial n. Tibialis anterior Deep peroneal n. Tibialis posterior Tibial n. Peroneus longus Superficial peroneal n. Peroneus tertius Deep peroneal n. Peroneus brevis Superficial peroneal n. FLEXION Flexor digitorum Tibial n. longus Flexor hallucis Tibial n. longus Flexor hallucis Medial brevis plantar n. Flexor digitorum Medial brevis plantar n. Flexor digitorum Lateral accessorius plantar n. Flexor digiti minimi Lateral brevis plantar n. Abductor hallucis Medial plantar n. Abductor digiti Lateral minimi plantar n. Lumbricals Medial & lateral plantar nn. Extensor digitorum Deep peroneal n. longus Extensor hallucis Deep peroneal n. longus Extensor digitorum Deep peroneal n. brevis Abductor hallucis Medial plantar n. Abductor digiti Lateral plantar n. minimi Dorsal interossei Lateral plantar n. Plantar interossei Lateral plantar n. Adductor hallucis Lateral plantar n. page 1415 page 1416
Table 110-2. Segmental innervation of the muscles of the lower limb L1 Psoas major, psoas minor L2 Psoas major, iliacus, sartorius, gracilis, pectineus, adductor longus, adductor brevis L3 Quadriceps, adductors (magnus, longus, brevis) L4 Quadriceps, tensor fasciae latae, adductor magnus, obturator externus, tibialis anterior, tibialis posterior L5 Gluteus medius, gluteus minimus, obturator internus, semimembranosus, semitendinosus, extensor hallucis longus, extensor digitorum longus, peroneus tertius, popliteus S1 Gluteus maximus, obturator internus, piriformis, biceps femoris, semitendinosus, popliteus, gastrocnemius, soleus, peronei (longus and brevis), extensor digitorum brevis S2 Piriformis, biceps femoris, gastrocnemius, soleus, flexor digitorum longus, flexor hallucis longus, some intrinsic foot muscles S3 Some intrinsic foot muscles (except abductor hallucis, flexor hallucis brevis, flexor digitorum brevis, extensor digitorum brevis)
Table 110.1 complements the mainly topographical description of muscles in Chapters 111-115 by bringing together information about the innervation and functions of the muscles of the lower limb. To achieve this, some simplification has been necessary. Movements. At the central nervous level of control, muscles are not recognized as individual actuators but as components of movement. Muscles may contribute to several types of movement, acting variously as prime movers, antagonists, fixators or synergists. A muscle that crosses two joints can produce more than one movement, and one or other of these functions may be emphasized when the proximal and distal attachments are fixed by the action of gravity or other muscles. Even a muscle that acts across one joint can produce a combination of movements, such as flexion with medial rotation, or extension with adduction, and some muscles have therefore been included in more than one place in the table. Nerve roots. The spinal roots listed as contributing to the innervation of muscles varies in different texts: this is a reflection of the often unreliable nature of the information available. The most positive identifications have been obtained by stimulating spinal roots electrically, and recording the evoked electromyographic activity in the muscles. However, this is a laborious process, and data of this quality are in limited supply. Much of the information in the table is based on neurological experience gained in examining the effects of lesions, and some of it is far from new. Major and minor contributions. Spinal roots have been given the same shading when they innervate a muscle to a similar extent or when differences in their contribution have not been described. Heavy shading has been used to indicate roots from which there is known to be a dominant contribution. From a clinical viewpoint, some of these roots may be regarded as innervating the muscle almost exclusively. Minor contributions have been retained in the table in order to increase its utility in other contexts, such as electromyography and comparative anatomy. Clinical testing. For diagnostic purposes, it is neither necessary nor possible to test every muscle, and the experienced neurologist can cover every clinical possibility with a much shorter list. Red has been used to highlight those muscles or movements that have diagnostic value. The emphasis in these tables is on the differentiation of lesions at different root levels. The preferred criteria for including a given muscle on this list are that it is visible and palpable; that its action is isolated or can be isolated by the examiner; that it is innervated by one peripheral nerve or (predominantly) one root; that is has a clinically elicitable reflex; and that it is useful in differentiating between different nerves, roots or levels of lesion. Table 110-3. Segmental innervation of joint movements of the lower limb Hip
Flexors, adductors, medial rotators Extensors, abductors, lateral rotators
L1-3 L5, S1
Knee Ankle Foot
Extensors Flexors Dorsiflexors Plantarflexors Invertors Evertors Intrinsic muscles
L3,4 L5, S1 L4, 5 S1, 2 L4, 5 L5, S1 S2, 3
Table 110-4. The movements and muscles tested to determine the location of a lesion in the lower limb Upper motor Movement Muscle neurone Root Reflex Nerve Hip flexion Iliopsoas ++ L1, Femoral 2 Hip Adductors L2, (+) Obturator adduction 3 Hip Gluteus maximus L5, Sciatic extension S1 Knee flexion Hamstrings + S1 Sciatic Knee Quadriceps L3, ++ Femoral extension 4 Ankle Tibialis anterior ++ L4 Deep dorsiflexion peroneal Ankle Peronei L5, Superficial eversion S1 peroneal Ankle Tibialis posterior L4, Tibial inversion 5 Ankle Gastrocnemius/soleus S1, ++ Tibial plantarflexion 2 Big toe Extensor hallucis L5 (Babinski Deep extension longus reflex) peroneal page 1416 page 1417
The muscles listed in the column Upper motor neurone are those which are preferentially affected in upper motor neurone lesions. The root level is the principal supply to a muscle.
Plantar reflex The plantar reflex is a superficial reflex whose elicitation forms an important part of the clinical examination of the central nervous system. With the foot relaxed and warm, the outer edge of the sole is stroked longitudinally with a hard object (traditionally the examiner's nail or a key). This should elicit flexion of the toes, although the normal adult response varies with the strength of the stimulus. In adults with upper motor neurone lesions, the response includes extension of the great toe (Babinski's sign).
© 2008 Elsevier
CLINICAL PROCEDURES NERVE BLOCKS Nerves can be effectively blocked with local anaesthetic injection for surgical or post-injury pain relief. The useful blocks in the lower limb are around the hip, for thigh and knee pain, and around the ankle, for foot surgery. For the deeper nerves around the hip, the use of a nerve stimulator needle is very helpful to localize the nerve precisely before infiltration of local anaesthetic. For example, to provide pain relief for total knee replacement, blockade of the sciatic nerve in the buttock, femoral nerve in the anterior groin, and obturator nerve in the medial groin, may be undertaken. To allow 'awake foot surgery' the posterior tibial nerve at the posteromedial ankle, the sural nerve at the posterolateral ankle and the superficial peroneal nerves at the anterior ankle, are blocked at the ankle.
DETERMINATION OF LOCATION OF A LESION The principles of testing muscle innervation in clinical neurological examination are given on page 814. Table 110.4 gives a list of movements and muscles of the lower limb chosen according to these principles. In practice, these tests would be combined with tests of sensory function.
INJECTION AND ASPIRATION Intramuscular injection
Intramuscular injections into the buttock should be avoided, to prevent iatrogenic damage to the sciatic nerve. If the buttock is to be used, the safe area is the true upper and outer quadrant, which is identified with the whole buttock exposed. The injection is then given mainly into gluteus medius rather than into gluteus maximus, provided a sufficiently long needle is used: most so-called 'intramuscular' injections given into the buttock are actually given into the fat. A safe alternative is to inject into the lateral aspect of the thigh (vastus lateralis). Joint injection and aspiration
Careful aseptic technique is essential for all joint aspirations and injections. Hip joint With the patient lying supine, and after the positions of the femoral artery and the anterior superior iliac spine have been marked out, the needle is introduced anteriorly c.5 cm distal to the anterior superior iliac spine and c.4 cm lateral to the femoral pulse, and passed posteriorly, a little proximally (cephalad), and medially. Knee joint The lateral retropatellar approach is used. With the patient supine and the knee extended, the needle is introduced at the level of the superior border of the patella and guided towards the suprapatellar pouch. Ankle joint The anterior approach entails introducing the needle between the tendons of tibialis anterior and extensor hallucis longus with the ankle partially plantar flexed,
keeping the needle tangential to the curve of the talus.
ARTHROSCOPY PORTALS The placement of portals for arthroscopy is critical, partly to maximize surgical access for visualization and for surgical instruments, but to also avoid damage to structures such as nerves and blood vessels. The knee is the joint most frequently examined in this way. Knee
For the knee, the standard portals are anterior. When the knee is flexed at a right angle, 'soft triangles' bordered by the patellar tendon, the femoral condyles and the tibia and anterior horns of the menisci are palpable on either side of the superior third of the patellar tendon. Small 'stab' incisions can be made in the apices of these triangular areas at about the level of the inferior pole of the patella for the anterolateral portal, and slightly lower for the anteromedial portal. These portals will allow passage of an arthroscope and instruments, with good access to most of the joint. The fat pad is close by: passage through it will hamper the view and subsequent surgery. The patellar tendon and the infrapatellar branch of the saphenous nerve are at risk from the incisions. The nerve is less vulnerable than the tendon: its position is variable but it usually lies below the appropriate portal sites. However, in the days of open meniscal surgery, the nerve was usually divided and painful neuroma formation was not uncommon. Posteromedial and posterolateral portals are useful when better access to the posterior knee is required. The saphenous and common peroneal nerves respectively are at risk with these approaches, and anatomical knowledge is vital to safe portal placement. Laterally the incision should be anterior to the tendon of biceps, which can easily be palpated. Medially the situation is more difficult because the saphenous nerve and vein run over the medial epicondylar region of the knee the breadth of the subject's hand posterior to the medial border of the patella. It is essential that sharp incision includes the skin only, and that dissection down to the joint capsule is undertaken bluntly. If the medial meniscus is sutured, the capsule must be exposed before tying off sutures to avoid ensnaring the nerve. In lateral meniscal repair a useful guide to avoid injury to the common peroneal nerve is never to allow needles or sutures to pass medial to the tendon of popliteus whilst viewing arthroscopically. Hip
Hip arthroscopy is not yet a common procedure but is increasingly being used. There are a number of portals. The anterolateral portal is sited on the skin c.4 cm lateral to the femoral pulse and 4 cm inferior to the inguinal ligament. A needle traversing the skin 30-45° cephalad is passed under X-ray control into the joint. Once the synovial cavity is entered the 'vacuum' is lost and the joint can be distracted. Other more lateral portals can be used: the lateral cutaneous nerve of the thigh and the sciatic nerve are potentially at risk. Ankle
The use of ankle arthroscopy is well established, especially in treating sportsrelated injuries. Anterior portals are standard. The anteromedial can be placed to
pass just medial to the tendon of tibialis anterior, in the palpable soft spot. However, this comes close to the long saphenous vein and nerve: limiting the sharp incision to the skin, followed by blunt deeper dissection reduces the risk. Alternatively the anteromedial portal can be chosen to pass between the lateral edge of tibialis anterior and extensor hallucis longus. The anterolateral portal is placed with the help of the arthroscope in the joint to check the position of a preliminary needle passed into the joint. The portal should pass lateral to peroneus tertius and extensor digitorum longus. The intermediate dorsal cutaneous branch of the superficial peroneal nerve is at risk as it crosses the ankle anterior to the lateral malleolus, and therefore it is wise to ink in the nerves and blood vessels before making any incision. The nerves can be felt at the front of the ankle even in obese patients. They are often seen if the ankle is maximally pulled into plantarflexion and are seen to flick as the barrel of a pen is pulled across the front of the ankle. Posterior ankle arthroscopy is controversial: there is a view that the proximity of the neurovascular bundle (medially) and the sural nerve (laterally) renders it too hazardous. Keeping lateral to the tendon of flexor hallucis will safeguard the medial neurovascular bundle.
SURGICAL INCISIONS Hip page 1417 page 1418
Hip joint surgery is usually undertaken for paediatric problems, trauma, or arthroplasty in arthritis. Anterior and anterolateral approaches, often used in children, put the lateral cutaneous nerve of the thigh at considerable risk. Even in more lateral approaches, the lateral cutaneous nerve of the thigh and the femoral and sciatic nerves are all at risk through traction. A popular anterolateral approach to the hip joint involves splitting and separating forwards the anterior part of gluteus medius and vastus lateralis as a single sheet of tissue for subsequent reattachment to the greater trochanter. This technique relies on the anatomical continuity of the tissue. If the splitting of gluteus medius is more than a few centimetres superior to the tip of the greater trochanter then the superior gluteal nerve and vessels are at risk, and weakness of hip abduction and a limp may result. Knee
Most open knee surgery can be undertaken through an anterior midline longitudinal incision, which gives good access and means that any future surgery can usually be undertaken via the same wound. New incisions run the risk of skin necrosis and poor wound healing as a consequence of interfering with the cutaneous blood supply. Inevitable interruption of the cutaneous nerves, including the infrapatellar branch of the saphenous nerve, means that there is always numbness lateral to a longitudinal incision. The extensile approach to the posterior knee is extensive. The key is to expose the sural nerve and short saphenous vein and to trace them proximally. This will lead the surgeon into the popliteal fossa and, after opening the deep fascia, safely to the neurovascular bundle. The wound crosses a flexure crease. A scar perpendicular to the crease might induce a fixed flexion contracture of the joint
and therefore, as at the anterior elbow, an S-shaped incision is employed where the transverse segment runs in the line of the flexure crease. Foot and ankle
Incisions around the foot and ankle frequently put cutaneous nerves at risk: such an injury can cause a distressing painful neuroma. The sural nerve at the ankle has a notorious tendency to form neuromas, often after repair of a ruptured calcaneal tendon. Bunion surgery is also problematic and can be undertaken through an internervous plane with a dorsomedial incision over the first metatarsophalangeal joint, rather than the traditional dorsal approach. REFERENCES Cormack GC, Lamberty BGH 1994 See Bibliography. Crock HV 1996 Atlas of Vascular Anatomy of the Skeleton and Spinal Cord. London: Martin Dunitz. Dodd H, Cockett FB 1976 The Pathology and Surgery of the Veins of the Lower Limb, 2nd edn. Edinburgh: Churchill Livingstone. Kosinski C 1926 Observations on the superficial venous system of the lower extremity. J Anat 60: 131-42. Medline Similar articles Sharrard WJW 1955 The distribution of the permanent paralysis in the lower limb in poliomyelitis; a clinical and pathological study. J Bone Joint Surg Br 37-B(4): 540-58. Taylor GI, Razaboni RM (eds) 1994 Michel Salmon: Anatomic Studies, Book 1. Arteries of the Muscles of the Extremities and Trunk. St. Louis: Quality Medical Publishing.
© 2008 Elsevier
111 Pelvic girdle, gluteal region and hip joint The term pelvic girdle is usually taken to be synonymous with a single 'innominate bone', though a girdle in other contexts is a complete ring. This implies that the human pelvis includes two pelvic girdles. Functionally it is more rational to consider a single pelvic 'girdle', which consists of the two innominate bones and the sacrum (strictly a part of the vertebral column). This strong pelvic girdle is virtually incapable of independent movement except during parturition in the female. It provides a weightbearing and protective structure, an attachment for trunk and limb muscles, and the skeletal framework of the birth canal. It is also best considered as a complete ring in the assessment of bony injuries. The gluteal region or buttock is an area encompassed by the gluteal fold inferiorly, a line joining the greater trochanter and the anterior superior iliac spine laterally, the iliac crest superiorly and the midline medially. It consists of a large bulk of skeletal muscle covering large and vulnerable neurovascular structures, and incorporates junctional (transitional) zones between the limb, pelvis and perineum at the sciatic foramina. Direct and indirect musculoskeletal injuries may entail damage to the sciatic nerve and gluteal vessels. The mobile yet very stable hip joint is deeply placed and well protected, but has important anatomical relations, especially anteriorly and posteriorly, which should be borne in mind during the management of trauma and degenerative arthritis, which commonly affect this joint. Its posterior relations lie in the gluteal region; anteriorly lies the junctional zone between pelvis and limb. This pelvicrural 'foramen', which lies between the inguinal ligament anteriorly and the pubis and ilium posteriorly, transmits numerous structures between the anterior thigh and pelvis.
© 2008 Elsevier
SKIN See also page 1399.
VASCULAR SUPPLY AND LYMPHATIC DRAINAGE Most of the skin of the buttock is supplied by musculocutaneous perforating vessels from the superior and inferior gluteal arteries. There are also small peripheral contributions from similar branches of the internal pudendal, iliolumbar, sacral, and first (profunda) perforating arteries. The arterial supply of the skin of the upper thigh is described on page 1461. For further detail consult Cormack and Lamberty (1994). Cutaneous veins are tributaries of vessels that correspond to the named arteries. Cutaneous lymphatic drainage is to the superficial inguinal nodes.
INNERVATION/DERMATOMES See page 1399.
© 2008 Elsevier
SOFT TISSUE FASCIA Superficial fascia
The superficial fascia of the buttock is continuous superiorly with that over the low back, and contains a considerable quantity of coarse fat. The superficial fascia of the thigh consists, as elsewhere in the limbs, of loose areolar tissue containing a variable amount of fat. In some regions, particularly near the inguinal ligament, it splits into recognizable layers, between which may be found the branches of superficial vessels and nerves. It is thick in the inguinal region, where its two layers enclose the superficial inguinal lymph nodes, long saphenous vein and other smaller vessels. Here the superficial layer is continuous with the abdominal superficial layer. The deep layer, a thin fibroelastic stratum, is most marked medial to the long saphenous vein and inferior to the inguinal ligament, and extends between the subcutaneous vessels and nerves and the deep fascia, with which it fuses a little below the ligament. (The line of fusion lies in the floor of the ventral flexure line of Holden or 'groove associated with the hip joint'.) This membranous fascia completes the saphenous opening, blending with its circumference and with the femoral sheath. Over the opening it is perforated by the long saphenous vein, by the superficial branches of the femoral artery except the superficial circumflex iliac (which perforates the fascia lata separately), and lymphatic vessels, hence the term cribriform fascia (Latin cribrum = a sieve). UPDATE Abstract: Structure of the lumbo-gluteal adipose body
Date Added: 30 May 2006
Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=14991238&query_hl=11&itool=pubmed_docsum The lumbo-gluteal adipose body. Kahn JL, Wolfram-Gabel R: Surg Radiol Anat 26:319-324, 2004. Deep fascia
The deep fascia covering the gluteal muscles varies in thickness. Over maximus it is thin, but over the anterior two-thirds of medius it forms the thick, strong gluteal aponeurosis. This is attached to the lateral border of the iliac crest superiorly, and splits anteriorly to enclose tensor fasciae latae and posteriorly to enclose gluteus maximus (p. 1444). Fascia lata
The fascia lata, the wide deep fascia of the thigh, is thicker in the proximal and lateral parts of the thigh where tensor fasciae latae and an expansion from gluteus maximus are attached to it. It is thin posteriorly and over the adductor muscles, but thicker around the knee, where it is strengthened by expansions from the tendon of biceps femoris laterally, sartorius medially, and quadriceps femoris anteriorly. The fascia lata is attached superiorly and posteriorly to the back of the sacrum and coccyx, laterally to the iliac crest, anteriorly to the inguinal ligament and superior ramus of the pubis, and medially to the inferior ramus of the pubis, the ramus and tuberosity of the ischium, and the lower border of the sacrotuberous ligament. From the iliac crest it descends as a dense layer over gluteus medius to the upper border of gluteus maximus, where it splits into two layers, one passing superficial and the other deep to the muscle, to reunite at its lower border. Iliotibial tract
Over the flattened lateral surface of the thigh, the fascia lata thickens to form a strong band, the iliotibial tract. The upper end of the tract splits into two layers, where it encloses and anchors tensor fasciae latae and receives, posteriorly, most of the tendon of gluteus maximus. The superficial layer ascends lateral to tensor fasciae latae to the iliac crest; the deeper layer passes up and medially, deep to the muscle, and blends with the lateral part of the capsule of the hip joint.
Distally, the iliotibial tract is attached to a smooth, triangular, anterolateral facet on the lateral condyle of the tibia (Gerdy's tubercle) where it is superficial to and blends with an aponeurotic expansion from vastus lateralis. When the leg is extended it stands out as a strong, visible ridge on the anterolateral aspect of the knee. Distally, the fascia lata is attached to all exposed bony points around the knee joint, such as the condyles of the femur and tibia, and the head of the fibula. On each side of the patella the deep fascia is reinforced by transverse fibres, which attach the vasti to it. The stronger lateral fibres are continuous with the iliotibial tract. Intermuscular septa page 1419 page 1420
The fascia lata is continuous with two intermuscular septa, which are attached to the whole of the linea aspera and its prolongations above and below. The lateral, stronger septum, which extends from the attachment of gluteus maximus to the lateral condyle, lies between vastus lateralis in front and the short head of biceps femoris behind and provides partial attachment for them. The medial, weaker septum lies between vastus medialis and the adductors and pectineus. Numerous smaller septa, such as that separating the thigh adductors and flexors, pass between the individual muscles, ensheathing them and sometimes providing partial attachment for their fibres.
SAPHENOUS OPENING (Fig. 111.1) The saphenous opening is an aperture in the deep fascia, lateral and a little distal to the medial part of the inguinal ligament, which allows passage to the long saphenous vein and other smaller vessels. The cribriform fascia, which is pierced by these structures, fills in the aperture and must be removed to reveal it. Adjacent subsidiary openings may exist to transmit venous tributaries, but these openings are more usually in the floor of the fossa. In the adult the approximate centre of the opening is c.3 cm inferior and 3 cm lateral to the pubic tubercle. It varies considerably in size, with a height of 1.5-9 cm and a width of 1-4 cm. The fascia lata in this part of the thigh displays superficial and deep strata (not to be confused with the superficial and deep layers of the superficial fascia described above). They lie respectively anterior and posterior to the femoral sheath; the somewhat spiral circumference of the saphenous opening is formed where the two are in continuity. The superficial stratum, lateral and superior to the saphenous opening, is attached to the crest and anterior superior spine of the ilium, to the whole length of the inguinal ligament, and to the pecten pubis together with the lacunar ligament. It is reflected inferolaterally from the pubic tubercle as the arched falciform margin, which forms the superior, lateral and inferior boundaries of the saphenous opening: this margin adheres to the anterior layer of the femoral sheath, and the cribriform fascia is attached to it. The falciform margin is considered to have superior and inferior cornua. The inferior cornu is well defined, and is continuous behind the long saphenous vein with the deep stratum of the fascia lata. The deep stratum is medial to the saphenous opening and is continuous with the superficial stratum at its lower margin. Traced upwards, it covers pectineus, adductor longus and gracilis, passes behind the femoral sheath, to which it is closely united, and continues to the pecten pubis.
FEMORAL SHEATH (Fig. 111.2)
Figure 111.1 The left saphenous opening, after removal of the cribriform fascia.
page 1420 page 1421
Figure 111.2 Structures passing beneath the left inguinal ligament.
The femoral sheath is formed by distal prolongations of the transversalis fascia anterior to the femoral vessels, and of the iliac fascia posteriorly, together forming a short funnel, wider proximally, its distal end fusing with the vascular fascia 3 or 4 cm distal to the ligament. At birth the sheath is shorter; it elongates when extension at the hips becomes habitual. The femoral branch of the genitofemoral
nerve perforates its vertical lateral wall. The medial wall slopes laterally and is pierced by the long (great) saphenous vein and lymphatic vessels. Like the carotid sheath, the femoral sheath encloses a mass of connective tissue in which the vessels are embedded. Three compartments are described: a lateral one containing the femoral artery, an intermediate one for the femoral vein, and most medial and smallest, the femoral canal, which contains the lymph vessels and a lymph node embedded in areolar tissue. The presence of this canal allows the femoral vein to distend. The canal is conical, c.1.25 cm in length. Its proximal end is the outer femoral ring, bounded in front by the inguinal ligament, behind by pectineus and its fascia, medially by the crescentic edge of the lacunar ligament and laterally by the femoral vein. The spermatic cord, or the round ligament, is just above its anterior margin, while the inferior epigastric vessels are near its anterolateral rim. It is larger in women than in men: this is due partly to the greater breadth of the pelvis and partly to the smaller size of the femoral vessels in women. The ring is filled by condensed extraperitoneal tissue, the femoral septum, covered by the parietal peritoneum. The femoral septum is traversed by numerous lymph vessels that connect the deep inguinal to the external iliac lymph nodes. Femoral hernia
Femoral hernia is described with other groin hernias on page 1111.
FEMORAL TRIANGLE The femoral triangle is a depressed area of the thigh lying distal to the inguinal fold. Its apex is distal, its limits are the medial margin of sartorius laterally, the medial margin of adductor longus medially and the inguinal ligament proximally (the base). Its floor is provided laterally by iliacus and psoas major, medially by pectineus and adductor longus. The femoral vessels, passing from midbase to apex, are in the deepest part of the triangle. Lateral to the artery the femoral nerve divides. The triangle also contains fat and lymph nodes.
BURSAE RELATED TO GLUTEUS MAXIMUS There are three bursae deep to gluteus maximus: the trochanteric, over the greater trochanter; the gluteofemoral, between the tendon of gluteus maximus and that of vastus lateralis; and the ischiofemoral, over the gluteal tuberosity, which is less commonly present.
OBTURATOR MEMBRANE (Fig. 111.3)
Figure 111.3 Oblique coronal section showing the left obturator membrane.
The obturator membrane is a thin aponeurosis that closes (obturates) most of the obturator foramen, leaving a superolateral aperture, the obturator canal, through which the obturator vessels and nerve leave the pelvis and enter the thigh. The membrane is attached to the sharp margin of the obturator foramen except at its inferolateral angle, where it is fixed to the pelvic surface of the ischial ramus, i.e. internal to the foramen. Its fibres are arranged mainly transversely in interlacing bundles; the uppermost bundle, which is attached to the obturator tubercles, completes the obturator canal. The two surfaces of the obturator membrane provide attachment for the two obturator muscles, internus and externus, and some fibres of the pubofemoral ligament of the hip joint are attached to the external surface.
© 2008 Elsevier
BONE INNOMINATE BONE (FigS 111.4, 111.5) Topography
The innominate or hip bone is large, irregular, constricted centrally and expanded above and below. Its lateral surface has a deep, cup-shaped acetabulum, articulating with the femoral head, anteroinferior to which is the large, oval or triangular obturator foramen. Above the acetabulum the bone widens into a plate with a sinuously curved iliac crest. The bone articulates in front with its fellow to form the pelvic girdle. Each has three parts, ilium, ischium and pubis, connected by cartilage in youth but united as one bone in adults. The principal union is in the acetabulum. The ilium includes the upper acetabulum and expanded area above it; the ischium includes the lower acetabulum and bone posteroinferior to it; the pubis forms the anterior acetabulum, separating the ilium from ischium, and the anterior median region where the pubes meet. Acetabulum (Fig. 111.6)
The acetabulum is an approximately hemispherical cavity central on the lateral aspect of the innominate bone, facing anteroinferiorly and surrounded by an irregular margin deficient inferiorly at the acetabular notch. The acetabular fossa forms the central floor and is rough and non-articular. The articular lunate surface is widest above (the 'dome'), where weight is transmitted to the femur. Fractures through this region therefore often lead to poor outcomes. All three innominate elements contribute to the acetabulum, but unequally. The pubis forms the anterosuperior fifth of the articular surface, the ischium forms the floor of the fossa and rather more than the posteroinferior two-fifths of the articular surface, and the ilium forms the remainder. A linear defect may cross the acetabular surface from the superior border to the acetabular fossa, but does not follow any junction between the main morphological parts of the innominate bone. UPDATE Date Added: 26 April 2006 Abstract: Anatomy and imaging of the acetabular labrum Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=16275573&query_hl=17&itool=pubmed_docsum Imaging of the acetabular labrum. Petersilge C: Magn Reson Imaging Clin N Am 13:641-652, 2005. UPDATE Date Added: 27 July 2005 Abstract: Modified ilioinguinal approach for acetabular fractures. Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=15213504&query_hl=6 Modified ilioinguinal approach for acetabular fractures. page 1421 page 1422
Figure 111.4 Left innominate bone, external aspect. B, The muscle attachments. (Photographs by Sarah-Jane Smith.)
Obturator foramen
The obturator foramen lies below and slightly anterior to the acetabulum, between the pubis and ischium. It is bordered above by the grooved obturator surface of the superior pubic ramus, medially by the pubic body and its inferior ramus, below by the ischial ramus, and laterally by the anterior border of the ischial body, including the margin of the acetabular notch. The foramen is almost closed by the obturator membrane (p. 1421), which is attached to its margins, except above, where a communication remains between the pelvis and thigh. This free edge is attached to an anterior obturator tubercle at the anterior end of the inferior border of the superior pubic ramus, and a posterior obturator tubercle on the anterior border of the acetabular notch; these tubercles are sometimes indistinct. Since the tubercles lie in different planes and the obturator groove crosses the upper border of the foramen, the acetabular rim is in fact a spiral. The foramen is large and oval in males, but smaller and nearly triangular in females. Structure
The thicker parts of the innominate bone are trabecular, encased by two layers of compact bone, while the thinner parts, as in the acetabulum and central iliac fossa, are often translucent and consist of one lamina of compact bone. In the upper acetabulum and along the arcuate line, i.e. the route of weight transmission from the sacrum to the femur, the amount of compact bone is increased and the subjacent trabecular bone displays two sets of pressure lamellae. These start together near the upper auricular surface and diverge to impinge on two strong buttresses of compact bone, from which two similar sets of lamellar arches start and converge on the acetabulum. The anterior iliac crest has been much studied as regards distribution of cortical and trabecular bone. Whitehouse (1977) has surveyed these studies: his own observations, by scanning electron micrography, indicate that the cortical bone is very porous, being only 75% bone, decreasing to 35% near the anterior superior iliac spine. Denser cortical bone starts at the margins of the crest and thickens rapidly below it on both aspects of the iliac blade.
Studies of the internal stresses within the innominate have revealed a systemic pattern of trabeculae that corresponds well with the expected patterns of theoretical stress trajectories (Holm 1980), though the patterns are much more complicated than in any other major bone. Stresses are higher in the acetabular than in the iliac region. In the ilium, stresses on the pelvic surface are much less than those on the gluteal surface. Muscle attachments
See individual bones. Vascular supply
In the infant, nutrient arteries are clearly demonstrable for each component of the innominate bone. Each nutrient artery branches in fan-like fashion within its bone of supply (Crock 1996). Later, a periosteal arterial network develops, with contributions from numerous local arteries (see under individual bones). Innervation
A periosteal network receives contributions from local nerves which supply muscles attaching to the periosteum and the joints in which the innominate bone is involved. Autonomic nerves accompany nutrient arteries and branch within the bone. page 1422 page 1423
Figure 111.5 Left innominate bone, internal (pelvic) aspect. B, The muscle attachments. (Photographs by Sarah-Jane Smith.)
Figure 111.6 Left acetabulum. (Photograph by Sarah-Jane Smith.)
page 1423 page 1424
Figure 111.7 The innominate bone at birth. Blue = unossified (cartilaginous) regions.
Figure 111.8 Anteroposterior radiograph of the pelvis of a boy aged 7.
Figure 111.9 The adolescent innominate bone. More heavily stippled areas indicate the secondary centres of ossification. Blue = unossified (cartilaginous) regions.
Ossification (Figs 111.7, 111.8, 111.9)
Ossification is by three primary centres, one each for the ilium, ischium and pubis. The iliac centre appears above the greater sciatic notch prenatally at about the ninth week, the ischial centre in its body in the fourth month, and the pubic centre in its superior ramus between the fourth and fifth months. At birth the whole iliac crest, the acetabular floor and inferior margin are still cartilaginous. The acetabulum is still a cartilaginous cup with a triradiate stem extending medially to the pelvic surface as a Y-shaped epiphyseal plate between the ilium, ischium and pubis, and including the anterior inferior iliac spine. Cartilage along the inferior margin also covers the ischial tuberosity, forms conjoined ischial and pubic rami, and continues to the pubic symphyseal surface and along the pubic crest to its tubercle. The ossifying ischium and pubis fuse to form a continuous body ramus at the seventh or eighth year. Secondary centres, other than for the acetabulum, appear
about puberty and fuse between the fifteenth and twenty-fifth years. There are usually two for the iliac crest (which rapidly fuse), and single centres for the ischial tuberosity (in cartilage close to the inferior acetabular margin and spreading forwards), anterior inferior iliac spine (although it may ossify from the triradiate cartilage), and symphyseal surface of the pubis (the pubic tubercle and crest may have separate centres). Progression of the ossification of the iliac crest in girls is a useful guide to skeletal maturity and helps the timing of surgery for spinal deformity. Between the ages of 8 and 9 years three major centres of ossification appear in the acetabular cartilage. The largest appears in the anterior wall of the acetabulum and fuses with the pubis, the second in the iliac acetabular cartilage superiorly, fusing with the ilium, and the third in the ischial acetabular cartilage posteriorly, fusing with the ischium. At puberty these epiphyses expand towards the periphery of the acetabulum and contribute to its depth (Ponseti 1978). Fusion between the three bones within the acetabulum occurs between the sixteenth and eighteenth years. Delaere et al (1992) have suggested that ossification of the ilium is similar to that of a long bone, possessing three cartilaginous epiphyses and one cartilaginous process, although it tends to undergo osteoclastic resorption comparable with that of cranial bones. During development the acetabulum increases in breadth at a faster rate than it does in depth. Avulsion fractures of pelvic apophyses may occur from excessive pull on tendons, usually in athletic adolescents. The most frequent injuries are to the ischial tuberosity (hamstrings) and anterior inferior iliac spine (rectus femoris).
PUBIS (Figs 111.4, 111.5) Topography page 1424 page 1425
The pubis is the ventral part of the innominate bone and forms a median cartilaginous pubic symphysis with its fellow. From its anteromedial body a superior ramus passes up and back to the acetabulum and an inferior ramus passes back, down and laterally to join the ischial ramus inferomedial to the obturator foramen. The body, anteroposteriorly compressed, has anterior, posterior and symphyseal (medial) surfaces and an upper border, the pubic crest. The anterior surface also faces inferolaterally; it is rough superomedially and smooth elsewhere, giving attachment to medial femoral muscles. The smooth posterior surface faces upwards and backwards as the oblique anterior wall of the lesser pelvis and is related to the urinary bladder. The symphyseal surface is elongated and oval, united by cartilage to its fellow at the pubic symphysis. Denuded of cartilage it has an irregular surface of small ridges and furrows or nodular elevations, varying considerably with age, features which are of forensic value. The pubic crest is the rounded upper border of the body which overhangs the anterior surface; its lateral end is the rounded pubic tubercle. Both crest and tubercle are palpable, the latter partly obscured in males by the spermatic cord that crosses above it from the scrotum to the abdomen. The pubic rami diverge posterolaterally from the lateral corners of the body. The anterior surface of the pubic body faces the femoral adductor region. The anterior pubic ligament attaches to its medial part and to a rough strip, which is wider in females. The posterior surface is separated from the urinary bladder by retropubic fat. The puboprostatic ligaments are attached medial to levator ani. Superior pubic ramus
The superior pubic ramus passes upwards, backwards and laterally from the body, superolateral to the obturator foramen, to reach the acetabulum. It is triangular in section and has three surfaces and borders. Its anterior, pectineal surface, tilted slightly up, is triangular in outline and extends from the pubic tubercle to the iliopubic eminence. It is bounded in front by the rounded obturator crest and behind by the sharp pecten pubis (pectineal line) which, with the crest, is the pubic part of the linea terminalis (i.e. anterior part of the pelvic brim). The posterosuperior, pelvic surface, medially inclined, is smooth and narrows into the posterior surface of the body, which is bounded above by the pecten pubis and below by a sharp inferior border. The obturator surface, directed down and back, is crossed by the obturator groove sloping down and forwards. Its anterior limit is the obturator crest and its posterior limit is the inferior border. Inferior pubic ramus
The inferior pubic ramus, an inferolateral process of the body, descends
inferolaterally to join the ischial ramus medial to, and below, the obturator foramen. The union may be locally thickened, but not obviously so in adults. The ramus has two surfaces and borders. The anteroexternal surface, continuous above with that of the pubic body, faces the thigh and is marked by muscles. It is limited laterally by the margin of the obturator foramen and, medially, by the rough anterior border. The posterointernal surface is continuous above with that of the body and is transversely convex: its medial part is often everted in males and connected to the crus of the penis. This surface faces the perineum medially, its smooth lateral part tilted up towards the pelvic cavity. The internal surface is indistinctly divided into medial, intermediate and lateral areas. The medial area faces inferomedially in direct contact with the crus of the penis or clitoris and is limited above and behind by an indistinct ridge for attachment of the fascia overlying the superficial perineal muscles. The medial margin of the ramus, strongly everted in males, provides attachment for the fascia lata and the membranous layer of the superficial perineal fascia. Pubic tubercle
The pubic tubercle provides a medial attachment for the inguinal ligament in the floor of the superficial inguinal ring and is crossed by the spermatic cord. Pecten pubis
The pecten pubis is the sharp, superior edge of the pectineal surface. The conjoint tendon and lacunar ligament are attached at its medial end. A strong fibrous pectineal ligament is attached along the rest of its surface. The smooth pelvic surface is separated from parietal peritoneum only by areolar tissue, in which the lateral umbilical ligament descends forwards across the ramus and, laterally, the vas deferens passes backwards. The obturator groove, converted to a canal by the upper borders of the obturator membrane and obturator muscles, transmits the obturator vessels and nerve from the pelvis to the thigh. Some fibres of the pubofemoral ligament are attached to the lateral end of the obturator crest. Muscle attachments
The tendon of adductor longus is attached on the anterior surface of the body, in the angle between its upper end and the pubic crest. Below adductor longus, gracilis attaches to a line near the medial border extending down to the inferior ramus. Lateral to gracilis, adductor brevis is attached to the body and inferior ramus. Obturator externus is attached laterally to the anterior surface, spreading onto both rami. Anterior fibres of levator ani are attached on the posterior surface of the body near its centre. More laterally, obturator internus is attached, extending onto both rami. Pectineus is attached to the pectineal surface of the superior ramus along its upper part. Gracilis, adductor brevis and obturator externus are attached in mediolateral order to the external surface of the inferior ramus. Adductor magnus usually extends from the ischial ramus on to the lower part of the inferior pubic ramus between adductor brevis and obturator externus. Some inner fibres of sphincter urethrae may be attached to the intermediate area of the internal surface, related to the dorsal nerve of the penis or clitoris, internal pudendal vessels and their fascial sheath. Fibres of obturator internus are attached to the lateral area. Ascending loops of cremaster are attached to the pubic tubercle. Lateral to the tubercle, on the pubic crest, the lateral part of rectus abdominis and, below it, pyramidalis are attached. Medially the crest is crossed by the medial part of rectus abdominis, ascending from ligamentous fibres interlacing in front of the pubic symphysis. Psoas minor, when present, is attached near the centre of the pecten pubis. Vascular supply
The pubis is supplied by a periosteal anastomosis of branches from the obturator, inferior epigastric and medial circumflex femoral arteries. The superficial and deep external pudendal arteries may also contribute. Multiple vascular foramina are present, mainly at the lateral (acetabular) end of the bone, but there is no consistently placed nutrient foramen. Innervation
The periosteum of the pubis is innervated by branches of nerves which supply
muscles attached to the bone, the hip joint and the symphysis pubis. Ossification
Ossification of the pubis is described on page 1421.
ILIUM (Figs 111.4, 111.5) Topography
The ilium has upper and lower parts and three surfaces. The smaller, lower part forms a little less than the upper two-fifths of the acetabulum. The upper part is much expanded, and has gluteal, sacropelvic and iliac (internal) surfaces. The posterolateral gluteal surface is an extensive rough area; the anteromedial iliac fossa is smooth and concave; the sacropelvic surface is medial and posteroinferior to the fossa, from which it is separated by the medial border. Iliac crest
The iliac crest is the superior border of the ilium, convex upwards but sinuously curved, internally concave in front, and convex behind. Its ends project as anterior and posterior superior iliac spines. The anterior superior iliac spine is palpable at the lateral end of the inguinal fold. The posterior superior iliac spine is not palpable but is often indicated by a dimple c.4 cm lateral to the second sacral spine above the medial gluteal region (buttock). The lateral end of the inguinal ligament is attached to the anterior superior iliac spine. page 1425 page 1426
The crest has ventral and dorsal segments: the ventral is slightly more than the anterior two-thirds of the crest and its prominence is associated with changes in iliac form as a result of the emergence of the upright posture; the dorsal segment, which occupies approximately the posterior third in man, exists in all land vertebrates. The ventral segment of the crest has internal and external lips; the rough intermediate zone is narrowest centrally. The tubercle of the crest projects onto the outer lip c.5 cm posterosuperior to the anterior superior spine. The dorsal segment has two sloping surfaces separated by a longitudinal ridge ending at the posterior superior spine. The summit of the crest, a little behind its midpoint, is level with the interval between the third and fourth lumbar spines. The interosseous and posterior sacroiliac ligaments arise from the medial margin of the dorsal segment. Anterior border
The anterior border descends to the acetabulum from the anterior superior spine. Superiorly it is concave forwards. Inferiorly is a rough anterior inferior iliac spine, immediately above the acetabulum, which is divided indistinctly into an upper area for the straight part of rectus femoris and a lower area extending laterally along the upper acetabular margin to form a triangular impression for the iliofemoral ligament. Posterior border
The posterior border is irregularly curved and descends from the posterior superior spine, at first forwards, with a posterior concavity forming a small notch. At the lower end of the notch is a wide, low projection, the posterior inferior iliac spine. Here the border turns almost horizontally forwards for c.3 cm then down and back to join the posterior ischial border. Together these borders form a deep greater sciatic notch, bounded above by the ilium and below by the ilium and ischium. The upper fibres of the sacrotuberous ligament are attached to the upper part of the posterior border. The superior rim of the notch is related to the superior gluteal vessels and nerve. The lower part of the border (i.e. the lower margin of the greater sciatic notch) is covered by piriformis and related to the sciatic nerve, which, however, largely adjoins the ischium. Medial border
The medial border separates the iliac fossa and the sacropelvic surface. It is indistinct near the crest, rough in its upper part, then sharp where it bounds an articular surface for the sacrum, and finally rounded. The latter part is the arcuate line, which inferiorly reaches the posterior part of the iliopubic (iliopectineal) eminence, marking the union of the ilium and pubis. Gluteal surface
The gluteal surface, facing inferiorly in its posterior part and laterally and slightly
downwards in front, is bounded above by the iliac crest, below by the upper acetabular border and by the anterior and posterior borders. It is rough and curved, convex in front, concave behind, and marked by three gluteal lines. The posterior gluteal line is shortest, descending from the external lip of the crest c.5 cm in front of its posterior limit and ending in front of the posterior inferior iliac spine. Above, it is usually distinct, but inferiorly it is ill-defined and frequently absent. The anterior gluteal line, the longest, begins near the midpoint of the superior margin of the greater sciatic notch and ascends forwards into the outer lip of the crest, a little anterior to its tubercle. The inferior gluteal line, rarely wellmarked, begins posterosuperior to the anterior inferior iliac spine, curving posteroinferiorly to end near the apex of the greater sciatic notch. Between the inferior gluteal line and the acetabular margin is a rough, shallow groove. Behind the acetabulum the lower gluteal surface is continuous with the posterior ischial surface, the unions marked by a low elevation. The articular capsule is attached to an area adjoining the acetabular rim, most of which is covered by gluteus minimus. Posteroinferiorly, near the union of the ilium and ischium, the bone is related to piriformis. Iliac fossa
The iliac fossa, the internal concavity of the ilium, faces anterosuperiorly. It is limited above by the iliac crest, in front by the anterior border and behind by the medial border, separating it from the sacropelvic surface. It forms the smooth and gently concave posterolateral wall of the greater pelvis. Below it is continuous with a wide shallow groove which is bounded laterally by the anterior inferior iliac spine and medially by the iliopubic eminence. The wide groove between the anterior inferior iliac spine and the iliopubic eminence is occupied by the converging fibres of iliacus laterally and the tendon of psoas major medially: the tendon is separated from bone by a bursa. The right iliac fossa contains the caecum, and often the vermiform appendix and terminal ileum. The left iliac fossa houses the end of the descending colon. Sacropelvic surface
The sacropelvic surface, the posteroinferior part of the medial iliac surface, is bounded posteroinferiorly by the posterior border, anterosuperiorly by the medial border, posterosuperiorly by the iliac crest and anteroinferiorly by the line of fusion of the ilium and ischium. It is divided into iliac tuberosity, auricular and pelvic surfaces. The iliac tuberosity, a large, rough area below the dorsal segment of the iliac crest, shows cranial and caudal areas separated by an oblique ridge and connected to the sacrum by the interosseous sacroiliac ligament. The sacropelvic surface gives attachment to the posterior sacroiliac ligaments and, behind the auricular surface, to the interosseous sacroiliac ligament. The iliolumbar ligament is attached to its anterior part. The auricular surface, immediately anteroinferior to the tuberosity, articulates with the lateral sacral mass. Shaped like an ear, its widest part is anterosuperior, its 'lobule' posteroinferior and on the medial aspect of the posterior inferior spine. Its edges are well defined, but the surface, though articular, is rough and irregular. It articulates with the sacrum and is reciprocally shaped. The anterior sacroiliac ligament is attached to its sharp anterior and inferior borders. The narrow part of the pelvic surface, between the auricular surface and the upper rim of the greater sciatic notch, often shows a rough preauricular sulcus for the lower fibres of the anterior sacroiliac ligament, more apparent in females. For the reliability of this feature as a sex discriminant see Finnegan (1978) and Brothwell and Pollard (2001). The pelvic surface is anteroinferior to the acutely recurved part of the auricular surface, contributing to the lateral wall of the lesser pelvis. Its upper part, facing down, is between the auricular surface and the upper limb of the greater sciatic notch. Its lower part faces medially and is separated from the iliac fossa by the arcuate line. Anteroinferiorly it extends to the line of union of the ilium and ischium. This is usually obliterated, but passes from the depth of the acetabulum to approximately the middle of the inferior limb of the greater sciatic notch. Muscle attachments
The attachment of sartorius extends down the anterior border below the anterior superior spine. The iliac crest gives attachment to lateral abdominal and dorsal muscles, and to fasciae and muscles of the lower limb. The fascia lata and iliotibial tract are attached to the outer lip and tubercle of its ventral segment. Tensor fasciae latae is attached anterior to the tubercle. The lower fibres of external oblique and, just behind the summit of the crest, the lowest fibres of latissimus dorsi are attached
to its anterior two-thirds. A variable interval exists between the most posterior attachment of external oblique and the most anterior attachment of latissimus dorsi, and here the crest is the base of the lumbar triangle. Internal oblique is attached to the intermediate area of the crest. Transversus abdominis is attached to the anterior two-thirds of the inner lip of the crest, and behind this to the thoracolumbar fascia and quadratus lumborum. The highest fibres of gluteus maximus are attached to the dorsal segment of the crest on its lateral slope. Erector spinae arises from the medial slope of the dorsal segment. The straight part of rectus femoris is attached to the upper area of the anterior inferior spine. Some fibres of piriformis are attached in front of the posterior inferior spine on the upper border of the greater sciatic notch. The gluteal surface is divided by three gluteal lines into four areas. Behind the posterior line, the upper rough part is for the attachment of the upper fibres of gluteus maximus and the lower, smooth region for part of the sacrotuberous ligament and iliac head of piriformis. Gluteus medius is attached between the posterior and anterior lines, below the iliac crest, and gluteus minimus is attached between the anterior and inferior lines. The fourth area, below the inferior line, contains vascular foramina. The reflected head of rectus femoris attaches to a curved groove above the acetabulum. Iliacus is attached to the upper two-thirds of the iliac fossa and is related to its lower third. page 1426 page 1427
The medial part of quadratus lumborum is attached to the anterior part of the sacropelvic surface, above the iliolumbar ligament. Piriformis is sometimes partly attached lateral to the preauricular sulcus, and part of obturator internus is attached to the more extensive remainder of the pelvic surface. Vascular supply
Branches of the iliolumbar artery run between iliacus and bone; one or more enter large nutrient foramina lying posteroinferiorly in the iliac fossa. The superior gluteal, obturator and superficial circumflex iliac arteries contribute to the periosteal supply. The obturator artery may supply a nutrient branch. Vascular foramina on the iliac gluteal aspect may lead into large vascular canals in the bone. Innervation
The periosteum is innervated by branches of nerves which supply muscles attached to the bone, the hip joint and the sacroiliac joint. Ossification
Ossification of the ilium is described on page 1421.
ISCHIUM (Figs 111.4, 111.5, 111.10) Topography
The ischium, the inferoposterior part of the innominate bone, has a body and ramus. The body has upper and lower ends and femoral, posterior and pelvic surfaces. Above, it forms the inferoposterior part of the acetabulum; below, its ramus ascends anteromedially at an acute angle to meet the descending pubic ramus and complete the obturator foramen. The ischiofemoral ligament is attached to the lateral border below the acetabulum.
Figure 111.10 The left ischial tuberosity: posterior aspect.
The femoral surface faces downwards, forwards and laterally towards the thigh. It is bounded in front by the margin of the obturator foramen. The lateral border, indistinct above but well defined below, forms the lateral limit of the ischial tuberosity. At a higher level the femoral surface is covered by piriformis, from which it is partially separated by the sciatic nerve and the nerve to quadratus femoris. The posterior surface, facing superolaterally, is continuous above with the iliac gluteal surface, and here a low convexity follows the acetabular curvature. Inferiorly, this surface forms the upper part of the ischial tuberosity, above which is a wide, shallow groove on its lateral and medial aspects. Above the ischial tuberosity the posterior surface is crossed by the tendon of obturator internus and the gemelli. The nerve to quadratus femoris lies between these structures and the bone. The ischial tuberosity is a large, rough area on the lower posterior surface and inferior extremity of the ischium. Though obscured by gluteus maximus in hip extension, it is palpable in flexion. It is 5 cm from the midline and about the same distance above the gluteal fold. It is elongated, widest above, and tapers inferiorly. The ischial posterior aspect lies between the lateral and posterior borders. The posterior border blends above with that of the ilium, helping to complete the inferior rim of the greater sciatic notch, the posterior end of which has a conspicuous ischial spine. Below this, the rounded border forms the floor of the lesser sciatic notch, between the ischial spine and tuberosity. The pelvic surface is smooth and faces the pelvic cavity; inferiorly it forms part of the lateral wall of the ischiorectal fossa. Ischial ramus
The ischial ramus has anteroinferior and posterior surfaces continuous with those of the inferior pubic ramus: the anteroinferior surface is roughened by the attachment of the medial femoral muscles. The smooth posterior surface is partly divided into perineal and pelvic areas, like the inferior pubic ramus. The upper border completes the obturator foramen; the rough lower border, together with the medial border of the inferior pubic ramus, bounds the subpubic angle and pubic arch. The fascia overlying the superficial perineal muscles is attached below the ridge between the perineal and pelvic areas of the posterior surface of the ischial ramus. Above the ridge are areas for the attachment of the crus of the penis or clitoris and sphincter urethrae. The lower border of the ramus is an attachment for the fascia lata and a membranous layer of the superficial perineal fascia. Ischial tuberosity
The ischial tuberosity is divided nearly transversely into upper and lower areas. The upper area is subdivided by an oblique line into a superolateral and an inferomedial part. The lower area, narrowing as it curves onto the inferior ischial aspect, is subdivided by an irregular vertical ridge into lateral and medial areas. The medial is covered by fibroadipose tissue, usually containing the ischial bursa
of gluteus maximus, which supports the body in sitting. Medially the tuberosity is limited by a curved ridge passing on to the ramus, to which the sacrotuberous ligament and its falciform process are attached. Ischial spine (Fig. 111.11)
The ischial spine projects downwards and a little medially. The sacrospinous ligament (p. 1439) is attached to its margins, separating the greater from the lesser sciatic foramen. The ligament is crossed posteriorly by the internal pudendal vessels and the nerve to obturator internus. Muscle attachments
Part of obturator externus is attached to the lower femoral surface of the ischial body. The anterior surface of the ischial ramus faces the adductor region. Obturator externus above, anterior fibres of adductor magnus and near the lower border, gracilis, are all attached here. Between adductor magnus and gracilis the attachment of adductor brevis may descend from the inferior pubic ramus. The posterior surface is divided into pelvic and perineal areas. The pelvic area, facing back, has part of obturator internus attached to it. The perineal area faces medially: its upper part is related to the crus of the penis or clitoris, and sphincter urethrae, ischiocavernosus and the transverse superficial perineal muscle are attached below this. The ischial tuberosity gives attachment to the posterior femoral muscles. Quadratus femoris is attached along the upper part of its lateral border. The upper area of the tuberosity is subdivided by an oblique line into a superolateral part for semimembranosus and an inferomedial part for the long head of biceps femoris and semitendinosus. The lower area is subdivided by an irregular vertical ridge into lateral and medial areas. The larger lateral area is for part of adductor magnus. Superomedial to the tuberosity the posterior surface has a wide, shallow groove, usually covered by hyaline cartilage, with a bursa between it and the tendon of obturator internus. Gemellus inferior is attached to the lower margin of the groove, near the tuberosity. Gemellus superior is attached to the upper margin, near the ischial spine. The pelvic surface of the ischial spine gives attachment to coccygeus (coextensive with the sacrospinous ligament) and to the most posterior fibres of levator ani. page 1427 page 1428
Figure 111.11 Joints and ligaments of the left half of the pelvis: anterior aspect.
Obturator internus is attached to the upper part of the smooth pelvic ischial surface, converging on the lesser sciatic notch (foramen) and covering the rest of this surface except the pelvic aspect of the ischial spine. The muscle and its fascia separate the bone from the ischiorectal fossa. Vascular supply
There are multiple vascular foramina at the acetabular margins, and a few are usually present on the pelvic surface. The bone is supplied by branches of the obturator, medial circumflex femoral and inferior gluteal arteries. Innervation
The periosteum is innervated by branches of nerves that supply the hip joint and muscles attached to the bone. Ossification
Ossification of the ischium is described on page 1421.
SACRUM See page 749.
COCCYX See page 754.
THE SKELETAL PELVIS AS A WHOLE (Fig. 111.12) The term pelvis ('basin') is applied vaguely to the skeletal ring formed by the innominate bones and the sacrum, the cavity therein, and even the entire region where the trunk and lower limbs meet. It is used here in the skeletal sense, to describe the irregular osseous girdle between the femoral heads and fifth lumbar vertebra. It is large because its primary function is to withstand the forces of body weight and musculature. In this section, its obstetric, forensic and anthropological significance will be considered. The pelvis can be regarded as having greater and lesser segments, the true and false pelves. The segments are arbitrarily divided by an oblique plane passing through the sacral promontory posteriorly and the lineae terminales elsewhere. Each linea terminalis includes the iliac arcuate line, iliopectineal line (pecten), and pubic crest. The segments are continuous, and the parts of the body cavity that
they enclose are also continuous through the pelvic inlet (superior pelvic aperture). The greater pelvis
The greater pelvis consists of iliac blades above the lineae terminales and the sacral base. This junctional zone is structurally massive and forms powerful arches from the acetabular fossae to the vertebral column around the visceral cavity, which is part of the abdomen. It has little anterior wall because of the pelvic inclination. The pelvic inlet (superior pelvic aperture)
The pelvic inlet may be round or oval, and is indented posteriorly by the sacral promontory. Its boundary, the pelvic brim, is obstetrically important and has also long been measured for anthropological reasons, as has the pelvic cavity. By convention, the pelvic inlet is described in three dimensions. The anteroposterior diameter (true conjugate) is measured between the midpoints of the sacral promontory and upper border of the symphysis pubis and is c.10 cm in the adult male and 11.2 cm in the adult female. The transverse diameter is the maximum distance between similar points (assessed by eye) on opposite sides of the pelvic brim and is c.12.5 cm in the male and 13.1 cm in the adult female. The oblique diameter is measured from the iliopubic eminence to the opposite sacroiliac joint and is c.12 cm in the adult male compared to 12.5 cm in the adult female. These measurements differ between racial groups. The articulated bony pelvis
The lesser pelvis encloses a true basin when soft tissues of the pelvic floor are in place. Skeletally it is a narrower continuation of the greater pelvis, with irregular but more complete walls around its cavity. Of obstetric importance, it has a curved median axis, and superior and inferior openings. The superior opening is occupied by viscera. The pelvic floor and its sphincters largely close the inferior opening. page 1428 page 1429
Figure 111.12 The diameters of the female lesser pelvis: A, pelvic inlet (superior
aperture); B, pelvic outlet (inferior aperture)-oblique diameter not shown.
Cavity of the lesser pelvis
The cavity of the lesser pelvis is short, curved, and markedly longer in its posterior wall. Anteroinferiorly it is bounded by pubic bones, their rami and symphysis. Posteriorly it is bounded by the concave anterior sacral surface and coccyx. Laterally on each side its margins are the smooth quadrangular pelvic aspect of the fused ilium and ischium. The region so enclosed is the pelvic cavity proper, through which pass the rectum, bladder and parts of the reproductive organs. The cavity in females must also permit passage of the fetal head. The pelvic cavity diameters are measured at approximately the mid level and also vary with different racial groups. The anteroposterior diameter is measured between the midpoints of the third sacral segment and posterior surface of the symphysis pubis and is c.10.5 cm in the male and 13 cm in the adult female. The transverse diameter is the widest transverse distance between the side walls of the cavity, and often the greatest transverse dimension in the whole cavity. It measures c.12 cm in the adult male and 12.5 cm in the adult female. The oblique diameter is the distance from the lowest point of one sacroiliac joint to the midpoint of the contralateral obturator membrane and measures c.11 cm in the male and 13.1 cm in the adult female. The pelvic outlet (inferior pelvic aperture)
Less regular in outline than the pelvic inlet, the pelvic outlet is indented behind by the coccyx and sacrum and bilaterally by the ischial tuberosities. Its perimeter thus consists of three wide arcs. Anteriorly is the pubic arch, between the converging ischiopubic rami. Posteriorly and laterally on both sides are the sciatic notches between the sacrum and ischial tuberosities. These are divided by the sacrotuberous and sacrospinous ligaments into greater and lesser sciatic foramina. With ligaments included, the pelvic outlet is rhomboidal. Its anterior limbs are the ischiopubic rami (joined by the inferior pubic ligament) and its posterior margins are the sacrotuberous ligaments, with the coccyx in the midline. The outlet is thus not rigid in its posterior half, being limited by ligaments and the coccyx, all slightly yielding. Even with the sacrum taken as the posterior midline limit (more reliable for measurement), there may be slight mobility at the sacroiliac joints. Note also that a plane of the pelvic outlet is merely conceptual. The anterior, ischiopubic part has a plane which is inclined down and back to a transverse line between the lower limits of the ischial tuberosities, and the posterior half has a plane approximating to the sacrotuberous ligaments, sloping down and forwards to the same line. Three measurements are made for the pelvic outlet. The anteroposterior diameter is usually measured from the coccygeal apex to the midpoint of the lower rim of the symphysis. The lowest sacral point may also be used (male 8 cm, female 12.5 cm). The transverse (bituberous) diameter is measured between the ischial tuberosities at the lower borders of their medial surfaces (male 8.5 cm, female 11.8 cm). The oblique diameter extends from the midpoint of the sacrotuberous ligament on one side to the contralateral ischiopubic junction (male 10 cm, female 11.8 cm). Other measurements
Apart from these main measurements, by consensus the basis of pelvic osteometry, other planes and measurements are used in obstetric practice. The plane of greatest pelvic dimensions is an obstetric concept. It represents the most capacious pelvic level, between the pelvic brim and midlevel plane, and corresponds with the latter anteriorly at the middle part of the symphysis pubis and posteriorly at the level of the second and third sacral segments. The plane of least dimensions is said to be at about midpelvic level. Its transverse diameter is between the apices of the ischial spines. This measurement is c.9.5 cm in an adult female and is just wide enough to allow passage of the biparietal diameter of a fetal head (c.9 cm). Not surprisingly, most difficulty in parturition occurs here. The above measurements are sometimes made in clinical practice using X-ray or MRI pelvimetry. Measurement is not possible without radiological techniques, and even these do not take into account the soft tissues. Anatomical measurements have been made in the past and were performed at physical and vaginal examination. However, these manual measurements have proven to be of little
clinical value and are now obsolete. Morphological classification of pelves
Interest in the dimensions described above is primarily obstetric and, less frequently, forensic. All pelvic measurements display individual variation and the values quoted are means from limited surveys. Sexual and racial differences also occur. These measurements have been analysed by many anatomists, anthropologists, obstetricians and radiologists in attempts to classify human pelves, especially female. The four most common terms used today are gynaecoid, anthropoid, platypelloid and android. The gynaecoid pelvis is the traditional Western female pelvis with a heart-shaped brim and the measurements described above. An anthropoid pelvis has a larger midcavity and a wide anteroposterior inlet which is oval in shape. An anthropoid pelvis is more common in women of African origin and may be associated with a 'high assimilation' pelvis where there is an additional lumbar vertebra. A platypelloid pelvis is flat and oval from side to side at the brim. It is a contracted pelvis that is rarely seen nowadays, having previously been associated with rickets. An android pelvis has a triangular brim and is the shape of a male pelvis. Pelvic axes and inclination (Fig. 111.13)
The axis of the superior pelvic aperture traverses its centre at right angles to its plane, directed down and backwards. When prolonged (projected) it passes through the umbilicus and midcoccyx. An axis is similarly established for the inferior aperture: projected upwards it impinges on the sacral promontory. Axes can likewise be constructed for any plane, and one for the whole cavity is a concatenation of an infinite series of such lines. It follows the curvature of the cavity, indicated by the profile of the sacrum and coccyx in lateral views. The form of this pelvic axis and the disparity in depth between the anterior and posterior contours of the cavity are prime factors in the mechanism of fetal transit in the pelvic canal. page 1429 page 1430
Figure 111.13 Median sagittal section through the female pelvis, showing the planes of the inlet and outlet and the axis of the pelvic cavity.
In the standing position the pelvic canal curves obliquely backwards relative to the trunk and abdominal cavity. The whole pelvis is tilted forwards, the plane of the pelvic brim making an angle of 50-60° with the horizontal. The plane of the pelvic outlet is tilted to c.15°. Therefore, the posterior parts of both planes are above the anterior. Strictly, the pelvic outlet has two planes, an anterior passing backwards from the pubic symphysis and a posterior passing forwards from the coccyx, both descending to meet at the intertuberous line. In standing, the pelvic aspect of the symphysis pubis faces as much upwards as backwards and the sacral concavity is directed anteroinferiorly. The front of the symphysis and anterior superior iliac spines are in the same vertical plane. In sitting, body weight is transmitted through inferomedial parts of the ischial tuberosities, with variable soft tissues intervening. The anterior superior iliac spines are in a vertical plane through the acetabular centres, and the whole pelvis is tilted back with the lumbosacral angle somewhat
diminished at the sacral promontory. Pelvic mechanism
The skeletal pelvis supports and protects the contained viscera, but is primarily part of the lower limbs, affording wide attachment for leg and trunk muscles. It constitutes the major mechanism for transmitting the weight of the head, trunk and upper limbs to the lower limbs. It may be considered as two arches divided by a vertical transacetabular plane. The posterior arch, chiefly concerned in transmitting weight, consists of the upper three sacral vertebrae and strong pillars of bone from the sacroiliac joints to the acetabular fossae. The anterior arch, formed by the pubic bones and their superior rami, connects these lateral pillars as a tie beam to prevent separation; it also acts as a compression strut against medial femoral thrust. The sacrum, as the summit of the posterior arch, is loaded at the lumbosacral joint. Theoretically this force has two components, one thrusting the sacrum downwards and backwards between the iliac bones, the other thrusting its upper end downwards and forwards. Sacral movements are regulated by osseous shape and massive ligaments. The first component therefore acts against the wedge, its tendency to separate iliac bones resisted by the sacroiliac and iliolumbar ligaments and symphysis pubis.
Figure 111.14 The anterior aspect of the pelvis: A, female; B, male.
Vertical coronal sections through the sacroiliac joints suggest division of the (synovial) articular region of the sacrum into three segments. In the anterosuperior segment, involving the first sacral vertebra, the articular surfaces are slightly sinuous and almost parallel. In the middle segment the posterior width between the articular markings is greater than the anterior, and centrally a sacral concavity fits a corresponding iliac convexity, an interlocking mechanism relieving the strain on the ligaments produced by body weight. In the posteroinferior segment the anterior sacral width is greater than the posterior and here its sacral surfaces are slightly concave. Anteroinferior sacral dislocation by the second component (of force) is prevented, therefore, mainly by the middle segment, owing to its cuneiform shape and interlocking mechanism. However, some rotation occurs, in which the anterosuperior segment tilts down and the
posteroinferior segment up. 'Superior' segmental movement is limited to a small degree by wedging but primarily by tension in the sacrotuberous and sacrospinous ligaments. In all movements the sacroiliac and iliolumbar ligaments and symphysis pubis resist separation of the iliac bones.
SEXUAL DIFFERENCES IN THE PELVIS (Fig. 111.14) The pelvis provides the most marked skeletal differences between male and female. Distinction can be made even during fetal life, particularly in the subpubic arch. In infancy, dimensions of the whole pelvis are greater in males than in females, but the size of the pelvic cavity is usually greater in females. This distinction prevails in childhood, but the difference is maximal at c.22 months. Sexual differences in adults are divisible into metrical and non-metrical features: the range of most features overlaps between the sexes. page 1430 page 1431
Differences are inevitably linked to function. While the primary pelvic function in both sexes is locomotor, the pelvis, particularly the lesser pelvis, is adapted to parturition in females, and these changes variably affect the proportions and dimensions of the greater pelvis. Since males are distinctly more muscular and therefore more heavily built, overall pelvic dimensions, such as the intercristal measurement (distance between the iliac crests), are greater, markings for muscles and ligaments more pronounced, and general architecture heavier. The male iliac crest is more rugged and more medially inclined at its anterior end; in females the crests are less curved in all parts. The iliac blades are more vertical in females, but do not ascend so far; the iliac fossae are therefore shallower and each iliopectineal line more vertical. These iliac peculiarities probably account for the greater prominence of female hips. The male is relatively and absolutely more heavily built above the pelvis, with consequent differences at the lumbosacral and hip joints. The sacral basal articular facet for the fifth lumbar vertebra and intervening disc is more than a third of the total sacral basal width in males but less than a third in females, in whom the sacrum is also relatively broader, accentuating this difference. The female has relatively broader sacral alae. The male acetabulum is absolutely larger, and its diameter is approximately equal to the distance between its anterior rim and symphysis pubis. In females, acetabular diameter is usually less than this distance, not only because it is absolutely smaller but also because the anterolateral wall of the cavity is comparatively and often absolutely wider. The height of the female symphysis and adjoining parts of the pubis and ischium, which form the anterior pelvic wall, are also absolutely less, producing a somewhat triangular obturator foramen, which is more ovoid in males. Differing pubic growth is also expressed in the subpubic arch below the symphysis and between the inferior pubic rami. It is more angular in males, being 50-60°; in females it is rounded, less easy to measure and usually 80-85°. A greater separation of the pubic tubercles in females contributes to the pubic width. The ischiopubic rami are also much more lightly built and narrowed near the symphysis; in males they bear a distinctly rough, everted area for attachment of the penile crura, the corresponding attachment for the clitoris being poorly developed. Ischial spines are closer in males and are more inturned. The greater sciatic notch is usually wider in females: mean values for males and females are 50.4° and 74.4°, respectively. The greater female values for angle and width are associated with increased backward sacral tilt and greater anteroposterior pelvic diameter, especially at lower levels. The sacrum also displays metrical sexual differences. Female sacra are less curved, the curvature being most marked between the first and second segments and the third and fifth, with an intervening flatter region. Male sacra are more evenly curved, relatively long and narrow and more often exceeding five segments (by addition of a lumbar or coccygeal vertebra). The sacral index compares sacral breadth (between the most anterior points on the auricular surfaces) with length (between midpoints on the anterior margins of the promontory and apex): average values for males and females are 105% and 115%. Auricular surfaces are relatively smaller and more oblique in females, but extend onto the upper three sacral vertebrae in both sexes. The dorsal auricular border is more concave in females. Many differences may be summarized in the generalization that the pelvic cavity is longer and more conical in males, shorter and more cylindrical in females; the axis is curved in both. Differences are greater at the inferior aperture than the brim, where in absolute measurements males are not as different from females as sometimes stated. The superior aperture is more likely to be anthropoid or android in males and gynaecoid or android in females, but there is overlap between the sexes.
In forensic practice, identification of human skeletal remains (which are sometimes fragmentary) usually involves diagnosis of sex, and this is most certainly established from the pelvis. Even parts of the pelvis may be useful. Several studies of metrical characteristics in various pelvic regions have been made, leading to the production of various indices. The ilium has received particular attention, e.g. one index compares the pelvic and sacroiliac parts of the bone. A line is extended back from the iliopectineal eminence to the nearest point on the anterior auricular margin and thence to the iliac crest. The auricular point divides this chilotic line into anterior (pelvic) and posterior (sacral) segments, each expressed as a percentage of the other. Chilotic indices display reciprocal values in the sexes: the pelvic part of the chilotic line is predominant in females, and the sacral part in males. Detailed metrical studies of the ilium have indicated its limited reliability in 'sexing' pelves. However, the higher incidence and definition of the female preauricular sulcus is recognized. The desirability of correlating all available metrical data is to be emphasized; when a range of pelvic data can be combined, especially if they are metrical, 95% accuracy should be achieved. Complete accuracy has been claimed when the rest of the skeleton is available. Assessment of sex from isolated and often incomplete human remains is less reliable. For details, consult Mays (1998) and Brothwell and Pollard (2001).
FEMUR Topography (Figs 111.15, 111.16)
The femur is the longest and strongest bone in the human body. Its length is associated with a striding gait, its strength with weight and muscular forces. Its shaft, almost cylindrical in most of its length and bowed forward, has a proximal round, articular head projecting mainly medially on its short neck, which is a medial curvature of the proximal shaft. The distal extremity is more massive and is a double 'knuckle' (condyle) that articulates with the tibia. In standing, the femoral shafts are oblique and their heads are separated by the pelvic width. The shafts converge downwards and medially to the knees and almost touch: they lie below the hip joints. Since the tibia and fibula descend vertically from the knees, the ankles are also in the line of body weight in standing or walking. Femoral obliquity varies but is greater in women, reflecting the relatively greater pelvic breadth and shorter femora. Proximally the femur consists of a head, neck, and greater and lesser trochanters. Femoral head (Fig. 111.17)
The femoral head faces anterosuperomedially to articulate with the acetabulum. The head, often described as rather more than half a 'sphere', is not part of a true sphere but is spheroidal and is part of the surface of an ovoid. Its smoothness is interrupted posteroinferior to its centre by a small, rough fovea. The head is intracapsular, encircled distal to its equator by the acetabular labrum. Its periphery is distinct, except anteriorly, where the articular surface extends to the neck. The ligamentum teres attaches to the fovea. The anterior surface of the head is separated inferomedially from the femoral artery by the tendon of psoas major, the psoas bursa, and the articular capsule. Femoral neck (Fig. 111.17)
The femoral neck is c.5 cm long, narrowest in its mid part and widest laterally, and connects the head to the shaft at an angle of c.125° (angle of inclination; neck-shaft angle): this facilitates movement at the hip joint, enabling the limb to swing clear of the pelvis. The neck also provides a lever for the action of the muscles acting about the hip joint, which are attached to the proximal femur. The neck-shaft angle is widest at birth and diminishes until adolescence; it is smaller in females. The neck is laterally rotated with respect to the shaft (angle of anteversion) some 10-15°, although values of this angle vary between individuals and between populations (Eckhoff et al 1994). The contours of the neck are rounded: the upper surface is almost horizontal and slightly concave, the lower is straighter but oblique, directed inferolaterally and backwards to the shaft near the lesser trochanter. On all aspects the neck expands as it approaches the articular surface of the head. The anterior surface of the neck is flat and marked at the junction with the shaft by a rough intertrochanteric line. The posterior surface, facing posteriorly and superiorly, is transversely convex, and concave in its long axis; its junction with the shaft is marked by a rounded intertrochanteric crest. There are numerous vascular foramina, especially anteriorly and posterosuperiorly. The anterior surface is intracapsular, the capsule attaching laterally to the
intertrochanteric line. Facets, often covered by extensions of articular cartilage, and various imprints frequently occur here. These facets may sometimes be associated with squatting. One such feature, the cervical fossa, may be a racial characteristic. On the posterior surface the capsule does not reach the intertrochanteric crest; little more than the medial half of the neck is intracapsular. The anterior surface adjoining the head and covered by cartilage is related to the iliofemoral ligament. A groove, produced by the tendon of obturator externus as it approaches the trochanteric fossa, spirals across the posterior surface of the neck in a proximolateral direction. Greater trochanter (Fig. 111.17) page 1431 page 1432
Figure 111.15 Left femur: anterior aspect. B, The muscle attachments. (Photographs by Sarah-Jane Smith.)
The greater trochanter is large and quadrangular, projecting up from the junction of the neck and shaft. Its posterosuperior region projects superomedially to overhang the adjacent posterior surface of the neck and here its medial surface
presents the rough trochanteric fossa. The proximal border of the trochanter lies the breadth of the subject's hand below the iliac tubercle, level with the centre of the femoral head. It has an anterior rough impression. Its lateral surface is divided by an oblique, flat strip, wider above, which crosses it down and forwards. This surface is palpable, especially when the muscles are relaxed. The trochanteric fossa occasionally presents a tubercle or exostosis. Lesser trochanter
The lesser trochanter is a conical posteromedial projection of the shaft at the posteroinferior aspect of its junction with the neck. Its summit and anterior surface are rough, but its posterior surface, at the distal end of the intertrochanteric crest, is smooth. It is not palpable. Intertrochanteric line
The intertrochanteric line, a prominent ridge at the junction of the anterior surfaces of the neck and shaft, descends medially from a superomedial tubercle on the anterior aspect of the greater trochanter to a point on the lower border of the neck, anterior to the lesser trochanter, where there may also be a tubercle. This line is the lateral limit of the hip joint capsule anteriorly. The upper and lower bands of the iliofemoral ligament are attached to its proximal and distal ends and tubercles. Distally it is continuous with the spiral line. Intertrochanteric crest
The intertrochanteric crest, a smooth ridge at the junction of the posterior surface of the neck with the shaft, descends from the posterosuperior angle of the greater trochanter medially down to the lesser trochanter. A little above its centre is a low, rounded quadrate tubercle. Above the tubercle it is covered by gluteus maximus, from which it is separated distally by quadratus femoris and the upper border of adductor magnus. page 1432 page 1433
Figure 111.16 Left femur: posterior aspect. B, The muscle attachments. (Photographs by Sarah-Jane Smith.)
Gluteal tuberosity
The gluteal tuberosity may be an elongated depression or a ridge. It may in part be prominent enough to be called a third trochanter. Shaft (Figs 111.15, 111.16)
The shaft is surrounded by muscles and is impalpable. The distal anterior surface, for 5-6 cm above the patellar articular surface, is covered by a suprapatellar bursa, between bone and muscle. The distal lateral surface is covered by vastus intermedius. The medial surface, devoid of attachments, is covered by vastus medialis. page 1433 page 1434
Figure 111.17 Proximal end of left femur: posterior aspect. (Photograph by SarahJane Smith.)
The shaft is narrowest centrally, expanding a little proximally, particularly towards its distal end. Its long axis makes an angle of c.10° with the vertical, and diverges c.5-7° from the long axis of the tibia. Its middle third has three surfaces and borders. The extensive anterior surface, smooth and gently convex, is between the lateral and medial borders, which are both round and indistinct. The posterolateral surface is bounded posteriorly by the broad, rough linea aspera, usually a crest with lateral and medial edges. Its subjacent compact bone is augmented to withstand compressive forces, which are concentrated here by the anterior curvature of the shaft. The linea aspera receives adductor longus, intermuscular septa and the short head of biceps femoris, inseparably blended at their attachment. Perforating arteries cross the linea laterally under tendinous arches in adductor magnus and biceps. Nutrient foramina, directed proximally, appear in the linea aspera, varying in number and site, one usually near its proximal end, a second usually near its distal end. The medial surface is posteromedial, smooth like the others, bounded in front by the indistinct medial border and behind by the linea aspera. In its proximal third the shaft has a fourth, posterior surface, bounded medially by a narrow, rough spiral line that is continuous proximally with the intertrochanteric line and distally with the medial edge of linea aspera. Laterally this surface is limited by the broad, rough, gluteal tuberosity, ascending a little laterally to the greater trochanter and descending to the lateral edge of the linea aspera. In its distal third the shaft also has a fourth, posterior surface, between the medial and lateral supracondylar lines, which is continuous above with the corresponding edges of the linea aspera. The lateral line is most distinct in its proximal two-thirds, where the short head of biceps femoris and lateral intermuscular septum are attached. Its distal third has a small rough area for the attachment of plantaris, often encroaching on the popliteal surface. The medial line is indistinct in its proximal two-thirds, where vastus medialis is attached. Proximally, the shaft is crossed by femoral vessels entering the popliteal fossa from the adductor canal. It is often sharp for 3 or 4 cm proximal to the adductor tubercle. The popliteal surface is also triangular. In its distal medial part it is rough and
slightly elevated. Forming the proximal floor of the popliteal fossa, it is covered by variable amounts of fat that separate the popliteal artery from bone. The superior medial genicular artery, a branch of the popliteal artery, arches medially above the medial condyle. It is separated from bone by the medial head of gastrocnemius. The latter is attached a little above the condyle; further distally there may be a smooth facet underlying a bursa for the medial head of gastrocnemius. More medially, there is often an imprint proximal to the articular surface: in flexion this is close to a rough tubercle on the medial tibial condyle for the attachment of semimembranosus. The superior lateral genicular artery arches up laterally proximal to the lateral condyle but is separated from bone by the attachment of plantaris to the distal part of the lateral supracondylar line.
Figure 111.18 Distal end of left femur: articular surface. (Photograph by Sarah-Jane Smith.)
Distal end (Fig. 111.18)
The distal end of the femur is widely expanded as a bearing surface for transmission of weight to the tibia. It has two massive condyles, which are partly articular. Anteriorly the condyles unite and continue into the shaft; posteriorly they are separated by a deep intercondylar fossa and project beyond the plane of the popliteal surface. The articular surface is a broad area, like an inverted U, for the patella and the tibia. The patellar surface extends anteriorly on both condyles, especially the lateral. It is transversely concave, vertically convex and grooved for the posterior patellar surface. The tibial surface is divided by the intercondylar fossa but is anteriorly continuous with the patellar surface. Its medial part is a broad strip on the convex inferoposterior surface of the medial condyle, and is gently curved with a medial convexity. Its lateral part covers similar aspects of the lateral condyle but is broader and passes straight back. The tibial surfaces are transversely convex in all directions. The anteroposterior curvature of both surfaces is not uniform. The exact pattern is controversial. One view is that in both tibial portions of the femoral condyles the sagittal radius of curvature is ever decreasing (a 'closing helix'). More recently it has been suggested that the medial articular surface describes arcs of two circles. The more posterior has a smaller radius. Laterally there may only be one arc of fixed curvature with a radius similar to that of the posterior arc of the medial femoral articular surface. These differences are important determinants of knee joint motion. Patellar surface (trochlear groove) The patellar surface extends more proximally on the lateral side. Its proximal border is therefore oblique and runs distally and medially, separated from the tibial surfaces by two faint grooves that cross the condyles obliquely. The lateral groove is the more distinct. It runs laterally and slightly forwards from the front of the intercondylar fossa and expands to form a faint triangular depression, resting on the anterior edge of the lateral meniscus with the knee fully extended. The medial groove is restricted to the medial part of the medial condyle and rests on the anterior edge of the medial meniscus in full extension. Where it ceases, the patellar surface continues back to the lateral part of the medial condyle as a semilunar area adjoining the anterior region of the intercondylar fossa. This area articulates with the medial vertical facet of the patella in full flexion; it is not distinct in outline in most femora. In habitual squatters articular cartilage may extend to the lateral aspect of the lateral condyle under vastus lateralis.
The trochlear groove helps to stabilize the patella. An abnormally shallow groove predisposes to instability. Intercondylar fossa
page 1434 page 1435
The intercondylar fossa separates the two condyles distally and behind. In front it is limited by the distal border of the patellar surface, and behind by an intercondylar line, separating it from the popliteal surface. It is intracapsular but largely extrasynovial. Its lateral wall, the medial surface of the lateral condyle, bears a flat posterosuperior impression which spreads to the floor of the fossa near the intercondylar line for the proximal attachment of the anterior cruciate ligament. The medial wall of the fossa, i.e. the lateral surface of the medial condyle, bears a similar larger area, but far more anteriorly, for the proximal attachment of the posterior cruciate ligament. Both impressions are smooth and largely devoid of vascular foramina, whereas the rest of the fossa is rough and pitted by vascular foramina. A bursal recess between the ligaments may ascend to the fossa. The capsular ligament and, laterally, the oblique popliteal ligament, are attached to the intercondylar line. The ligamentum mucosum (infrapatellar synovial fold or plica) is attached to the anterior border of the fossa. Lateral condyle (Figs 111.18, 111.19) The lateral condyle is larger anteroposteriorly than the medial. Its most prominent point is the lateral epicondyle to which the lateral collateral ligament is attached. A short groove, deeper in front, separates the lateral epicondyle inferiorly from the articular margin. This groove allows the tendon of popliteus to run deep to the lateral collateral ligament and insert inferior and anterior to the ligament insertion. Adjoining the joint margin is a strip of condyle, 1 cm broad. It is intracapsular and covered by synovial membrane except for the attachment of popliteus. The medial surface is the lateral wall of the intercondylar fossa. Its lateral surface projects beyond the shaft. Part of the lateral head of gastrocnemius is attached to an impression posterosuperior to the lateral epicondyle. Medial condyle The medial condyle has a bulging convex medial aspect, which is easily palpable. Proximally its adductor tubercle, which may only be a facet rather than a projection, receives the tendon of adductor magnus. The medial prominence of the condyle, the medial epicondyle, is anteroinferior to the tubercle. The lateral surface of the condyle is the medial wall of the intercondylar fossa. The condyle projects distally so that, despite the obliquity of the shaft, the profile of the distal end is almost horizontal. A curved strip, c.1 cm wide, adjoining the medial articular margin, is covered by synovial membrane and is inside the joint capsule. Proximal to this, the medial epicondyle receives the medial collateral ligament. UPDATE Abstract: Variations of femoral condyle shape
Date Added: 10 April 2006
Click on the following link to view the abstract: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd=Retrieve&db=pubmed&dopt=Abstract&list_uids=16417136&query_hl=3&itool=pubmed_docsum Variations of femoral condyle shape. Biscevic M, Hebibovic M, Smrke D: Coll Antropol 29:409-414, 2005. Structure (Fig. 111.20)
Figure 111.19 Distal end of left femur: lateral aspect. (Photograph by Sarah-Jane Smith.)
The femoral shaft is a cylinder of compact bone with a large medullary cavity. The wall is thick in its middle third, where the femur is narrowest and the medullary cavity most capacious. Proximally and distally the compact wall becomes progressively thinner, and the cavity gradually fills with trabecular bone. The extremities, especially where articular, consist of trabecular bone within a thin shell of compact bone, their trabeculae being disposed along lines of greatest stress. At the proximal end the main trabeculae form a series of plates orthogonal to the articular surface, converging to a central dense wedge, which is supported by strong trabeculae passing to the sides of the neck, especially along its upper and lower profiles. Force applied to the femoral head is therefore transmitted to the wedge and thence to the junction of the neck and shaft. This junction is strengthened by dense trabeculae extending laterally from the lesser trochanter to the end of the superior aspect of the neck, thus resisting tensile or shearing forces applied to the neck through the head. Tensile and compressive tests indicate that axial trabeculae of the femoral head withstand much greater stresses than peripheral trabeculae. A smaller bar across the junction of the greater trochanter with the neck and shaft resists shearing produced by muscles attached to it. These two bars are proximal layers of arches between the sides of the shaft and transmit to it forces applied to the proximal end. A thin vertical plate, the calcar femorale, ascends from the compact wall near the linea aspera into the trabeculae of the neck. Medially it joins the posterior wall of the neck; laterally it continues into the greater trochanter, where it disperses into general trabecular bone. It is thus in a plane anterior to the trochanteric crest and base of the lesser trochanter. Newell (1997), in a review of the calcar femorale, has described its three-dimensional anatomy in terms of the work of Dixon (1910), who was the first to recognize the deficiencies of the classic two-dimensional description of the architecture of the proximal femur. Dixon suggested that the trabecular framework of the proximal femur was spiral, and that the 'arches' were simplified sectional profiles of this spiral. At the distal end of the femur, trabeculae spring from the
entire internal surface of compact bone, descending perpendicular to the articular surface. Proximal to the condyles these are strongest and most accurately perpendicular. Horizontal planes of trabecular bone, arranged like crossed girders, form a series of cubical compartments. Muscle attachments
The greater trochanter provides attachment for the smaller gluteal muscles. Gluteus minimus is attached to its rough anterior impression and gluteus medius to its lateral oblique strip. The bone is separated from the tendon of gluteus medius by a bursa. The area behind is covered by deep fibres of gluteus maximus, with part of its trochanteric bursa interposed. The tendon of piriformis is attached to the upper border of the trochanter and the common tendon of obturator internus and the gemelli are attached to its medial surface. The trochanteric fossa receives the tendon of obturator externus. Psoas major is attached to the summit and anteromedial surface of the lesser trochanter. Iliacus is attached to the medial or anterior surface of its base, descending a little behind the spiral line as its tendon fuses with that of psoas. Adductor magnus (upper part) passes over its posterior surface, sometimes separated by an interposed bursa. The most proximal fibres of vastus lateralis are attached to the proximal end of the intertrochanteric line, and those of vastus medialis are attached distally. Quadratus femoris is attached to the quadrate tubercle and the immediately distal bone. Vastus intermedius is attached to the anterior and lateral surfaces of the shaft in their proximal three-quarters. Slips of articularis genu are attached distal to this. page 1435 page 1436
Figure 111.20 A, Coronal section through left hip joint: synovial membrane shown in blue. B, Oblique section through the proximal end of the left femur showing the trabecular architecture, calcar femorale, medullary cavity and variations in cortical thickness.
The gluteal tuberosity receives the deeper fibres of the distal half of gluteus maximus and, at its medial edge, pubic fibres of adductor magnus. Distal to this, adductor magnus is attached to the linea aspera and aponeurotically to the proximal part of the medial supracondylar ridge. Its remaining fibres form a large tendon attached to the adductor tubercle, with an aponeurotic expansion to the distal part of the medial supracondylar ridge. Pectineus and adductor brevis are attached to the posterior femoral surface between the gluteal tuberosity and spiral line. The pectineal attachment is a line, sometimes slightly rough, from the base of the lesser trochanter to the linea aspera. Adductor brevis is attached lateral to pectineus and beyond this to the proximal part of the linea aspera, medial to adductor magnus. Adductor longus, intermuscular septa and the short head of biceps femoris are attached to the linea aspera. Vastus lateralis has a linear attachment from the anterior surface of the base of the greater trochanter to the proximal end of the gluteal tuberosity, and along the lateral margin of the latter to the proximal half of the lateral edge of the linea aspera. Vastus medialis is attached from the distal end of the intertrochanteric line along the spiral line to the medial edge of the linea aspera and thence to the medial supracondylar line, which also receives many fibres from the aponeurotic attachments of adductor magnus. The medial head of gastrocnemius is attached to the posterior surface a little above the medial condyle. The short head of biceps femoris is attached to the proximal two-thirds of the lateral supracondylar line. Plantaris attaches to the line distally. There is an attachment of vastus medialis to the proximal two-thirds of the medial supracondylar line. Part of the lateral head of gastrocnemius is attached posterosuperiorly to the lateral epicondyle. Popliteus is attached anteriorly in the groove on the outer aspect of the lateral epicondyle. Its tendon passes deep to the lateral collateral ligament (Fig. 111.19). The tendon lies in the groove in full knee flexion; in
extension it crosses the articular margin and may form an impression on it. Vascular supply (Fig. 111.21)
For details consult Crock (1980, 1996). The blood supply of the femoral head is derived from an arterial ring around the neck, just outside the attachment of the fibrous capsule, constituted by the medial and lateral circumflex arteries with minor contributions from the superior and inferior gluteal vessels (see trochanteric anastomosis, p. 1453). From this ring, ascending cervical branches pierce the capsule (under its zona orbicularis, p. 1440) to ascend the neck beneath the reflected synovial membrane. These vessels become the retinacular arteries and form a subsynovial intra-articular ring. Here the vessels are at risk with a displaced fracture of the femoral neck. Interruption of blood supply in this way can lead to avascular necrosis of the femoral head. If the fracture is intra-articular then not only is the intraosseous blood supply damaged but the retinacular vessels can also be vulnerable. If the fracture is extracapsular, the retinacular vessels will remain intact and avascular necrosis of the femoral head does not occur. The ascending cervical vessels give off metaphyseal branches that enter the neck, while the intra-articular ring gives off lateral and inferior epiphyseal branches. A small medial epiphyseal supply, of importance in early childhood, reaches the head along the ligamentum teres (p. 1441) by the acetabular branches of the obturator and medial circumflex femoral arteries, which anastomose with the other epiphyseal vessels. During growth, the epiphyseal plate separates the territories of the metaphyseal and epiphyseal vessels; after osseous union of the head and neck, they anastomose freely. Observations on developmental patterns of this supply in late fetal and early postnatal periods have revealed that although medial and lateral circumflex arteries at first contribute equally, two major branches of the medial provide the final supply, both posterior to the neck. The supply from the lateral circumflex diminishes and the arterial ring is interrupted. As the femoral neck elongates, the extracapsular circle becomes more distant from the epiphyseal part of the head. The trochanteric regions and subtrochanteric shaft are supplied by the trochanteric and cruciate arterial anastomoses (p. 1453). More distally in the shaft, nutrient foramina, directed proximally, are found in the linea aspera, varying in number and site: one is usually near its proximal end and a second usually near its distal end. The main nutrient artery is usually derived from the second perforating artery (see profunda femoris, p. 1451). If two nutrient arteries occur, they may branch from the first and third perforators. Periosteal vessels arise from the perforators and from the profunda, and run circumferentially rather than longitudinally. The distal metaphysis has many vascular foramina. Arterial supply here is from the genicular anastomosis (p. 1486). page 1436 page 1437
Figure 111.21 Collateral circulation around the hip and upper thigh.
The pattern of venous drainage of the head and neck corresponds in general to that of the arteries, though there may be a single large cervical vein posteroinferiorly. Innervation
The periosteal innervation is derived proximally from nerves that supply the hip joint (p. 1440), distally from those supplying the knee (p. 1482), and in all areas from nerves that innervate muscles attached to the bone. Ossification (Figs 111.15, 111.16, 111.22)
The femur ossifies from five centres: in the shaft, head, greater and lesser trochanters and the distal end. Other than the clavicle, it is the first long bone to ossify. The process starts in the midshaft in the seventh prenatal week and extends to produce a miniature shaft that is largely ossified at birth. Secondary centres appear in the distal end (from which the condyles and epicondyles are formed) during the ninth month, in the head during the first six months after birth, in the greater trochanter during the fourth year and in the lesser between the twelfth and fourteenth year. The centre in the cartilaginous head is restricted to it until the tenth year, so that the epiphyseal line (Fig. 111.8) is horizontal and the inferomedial part of the articular surface is on the neck. The medial epiphyseal margin later grows over this part of the articular surface. Thus, the mature epiphysis is a hollow cup on the summit of the neck. The epiphyseal line follows the articular margin except where it is separated superiorly from the articular surface by a non-articular area where blood vessels enter the head (Trueta 1957). The epiphyses fuse independently: the lesser trochanter soon after puberty, then the greater, the head in the fourteenth year in females, seventeenth
in males, and the distal end in the sixteenth year in females, eighteenth in males. The distal epiphyseal plate traverses the adductor tubercle. Growth plate considerations
Trauma to any epiphyseal plate can lead to bony union between epiphysis and metaphysis, and so cause premature cessation of growth. Any surgery in the hip region in children can injure the growth plate, resulting in abnormal proximal femoral development. In the case of fractures involving the epiphysis, restoration of normal bony alignment as soon as possible is essential to minimize the risk of subsequent abnormal growth. The growth plate represents a line of weakness and predisposes to fracture from injury. Such injuries affecting the capital epiphysis are uncommon. page 1437 page 1438
Figure 111.22 Stages in ossification of the femur (not to scale). Blue = unossified (cartilaginous) regions.
As well as acute injury, a more chronic fracture through the capital epiphysis occurs in 'slipped upper femoral epiphysis'. The condition affects pubescent adolescents, especially males. Endocrinological abnormality may be related. The femoral head epiphysis displaces posteriorly off the femoral neck. If it heals in this position, lower limb deformity and restricted hip movement occur. A classic hallmark is obligatory external rotation of the femur as the hip is flexed. Treatment varies according to the time taken for the 'slip' to occur. Normal anatomical restoration is rarely possible in 'acute' cases. The position of the femoral head may be accepted as it is and fixed with screws in this position to stop further displacement. This treatment will deliberately cause premature growth plate fusion and so prevent future 'slippage'. Since the distal femoral growth plate produces the vast majority of femoral length, an acceptable limb length difference usually results. Infection of bone in neonates and young children tends to arise via bacteria in the blood stream which usually 'seed' in the metaphyseal region, probably as a consequence of the vascular 'arcade' arrangement of arteries in this part of the bone. The proximal femoral growth plate is intra-articular. As a result infection in the proximal femoral metaphysis can spread into the joint and result in a septic arthritis that can destroy the hip joint permanently. The distal end of the femur is the only epiphysis in which ossification consistently starts just before birth, a most reliable indicator that a dead newborn child was viable. Since the epiphyseal plate is level with the adductor tubercle, the epiphysis is partly extra-articular. Operations here may damage the distal epiphyseal cartilage in children and result in subsequent shortening of the leg.
© 2008 Elsevier
JOINTS PUBIC SYMPHYSIS The pubic bones meet in the midline at the pubic symphysis, a secondary cartilaginous joint (Fig. 111.3). Articulating surfaces
The articulating surfaces are the medial (symphyseal) surfaces of the pubic bones, each covered by a thin layer of tightly adherent hyaline cartilage (surface growth cartilage in the young). The junction is not flat but marked by reciprocal crests and papillae. Theoretically this would resist shearing. The surfaces of hyaline cartilage are connected by fibrocartilage, varying in thickness and constituting the interpubic disc. The symphysis often contains a cavity, probably due to absorption. It rarely appears before the tenth year and is non-synovial. The cavity, which is better developed in females, is usually posterosuperior but may reach the front or even occupy most of the cartilage. Ligaments
The interpubic disc is strengthened anteriorly by several interlacing collagenous fibrous layers, passing obliquely from bone to bone, decussating and interweaving with fibres of the external oblique aponeuroses and the medial tendons of the recti abdominis. These layers constitute the anterior pubic ligament. There are also less well-developed posterior fibres, sometimes named the posterior pubic ligament. The main ligaments of the joint are the superior and arcuate pubic ligaments. The superior pubic ligament connects the bones above, extending to the pubic tubercles. The arcuate pubic ligament, a thick arch of fibres, connects the lower borders of the symphyseal pubic surfaces bounding the pubic arch. Superiorly it blends with the interpubic disc and extends laterally attached to the inferior pubic rami. Its inferior edge is separated from the anterior border of the perineal membrane by an opening for the deep dorsal vein of the penis or clitoris. Vascular supply
The pubic symphysis is supplied by pubic branches of the obturator, superficial external pudendal and inferior epigastric arteries. Innervation
The pubic symphysis is innervated by branches from the iliohypogastric, ilioinguinal and pudendal nerves. Factors maintaining stability
The interpubic disc and the superior and arcuate ligaments are the main stabilizing factors of the pubic symphysis. Movements
Angulation, rotation and displacement are possible but slight and are likely during movement at the sacroiliac and hip joints. Excessive movement may occur as a sports injury. Some separation occurs late in gestation and during childbirth: on occasion this is considerable. Relations and 'at risk' structures
Anteriorly the pubic symphysis is related to superficial fascia and skin. Because of the obliquity of the joint, the proximal shafts of the penis and clitoris also lie anterior to its lower half. Inferiorly the urethra lies c.2.5 cm away in the male, and somewhat closer in the female, as it passes through the perineal membrane. Closer to the joint, the deep dorsal vein of the penis or clitoris passes between the arcuate ligament and the anterior border of the perineal membrane. Posteriorly the upper part of the joint is separated from the inferolateral surfaces of the bladder by the retropubic fat pad. Inferiorly in the male the prostatic venous plexus separates the prostate from the lower part of the joint. The region is
sometimes termed the retropubic space. Because of these relations traumatic disruption of the anterior bony pelvis may be associated with serious urogenital injury.
SACROILIAC JOINT (Fig. 111.23) The sacroiliac joint is a synovial articulation between the sacral and iliac auricular surfaces. Fibrous adhesions and gradual obliteration occur in both sexes, earlier in males, and after the menopause in females. Radiological evidence of obliteration in normal subjects is occasionally seen before 50 years, but is not uncommon thereafter. In old age the joint may be completely fibrosed and occasionally even ossified. Articulating surfaces
The surfaces are nearly flat only in infants; in adults they are irregular, often markedly so, and sometimes sinuous. The curvatures and irregularities, greater in males, are reciprocal: they restrict movements and contribute to the considerable strength of the joint in transmitting weight from the vertebral column to the lower limbs. The sacral surface is covered by hyaline cartilage, which is thicker anteriorly than posteriorly in adults. The thinner cartilage on the iliac surface, earlier thought to be fibrocartilage, is also hyaline as confirmed by the presence of type II collagen. page 1438 page 1439
Figure 111.23 Multislice CT of the sacroiliac joints in an adult female, reformatted in the coronal plane. (By kind permission from Dr Justin Lee, Chelsea and Westminster Hospital, London.)
Fibrous capsule
The capsule is attached close to both articular margins. Ligaments
Anterior sacroiliac ligament (Fig. 111.11) The anterior sacroiliac ligament, an anteroinferior capsular thickening, is particularly well-developed near the arcuate line and the posterior inferior iliac spine, where it connects the third sacral segment to the lateral side of the preauricular sulcus. It is thin elsewhere. Interosseous sacroiliac ligament The interosseous sacroiliac ligament is the major bond between the bones, filling the irregular space posterosuperior to the joint. It is covered superficially by the posterior sacroiliac ligament. Its deeper part has superior and inferior bands
passing from depressions posterior to the sacral auricular surface to those on the iliac tuberosity. These bands are covered by, and blend with, a more superficial fibrous sheet connecting the posterosuperior margin of a rough area posterior to the sacral auricular surface to the corresponding margins of the iliac tuberosity. This sheet is often partially divided into superior and inferior parts, the former uniting the superior articular process and lateral crest on the first two sacral segments to the neighbouring ilium as a short posterior iliac ligament (Fig. 45.49). The posterior sacroiliac ligament (Fig. 45.59) The posterior sacroiliac ligament lies over the interosseous ligament: the dorsal rami of the sacral spinal nerves and vessels intervene. It consists of several weak fasciculi connecting the intermediate and lateral sacral crests to the posterior superior iliac spine and posterior end of the internal lip of the iliac crest. Inferior fibres, from the third and fourth sacral segments, ascend to the posterior superior iliac spine and posterior end of the internal lip of the iliac crest: they may form a separate long posterior sacroiliac ligament. This ligament is continuous laterally with part of the sacrotuberous ligament and medially with the posterior lamina of the thoracolumbar fascia. Iliolumbar ligament See page 761. Sacrotuberous ligament (Figs 111.11, 45.59) The sacrotuberous ligament is broadly attached by its base to the posterior superior iliac spine. It is partly blended with the posterior sacroiliac ligaments to the lower transverse sacral tubercles and the lateral margins of the lower sacrum and upper coccyx. Its oblique fibres descend laterally, converging to form a thick, narrow band that widens again below and is attached to the medial margin of the ischial tuberosity. It then spreads along the ischial ramus as the falciform process, whose concave edge blends with the fascial sheath of the internal pudendal vessels and pudendal nerve. The lowest fibres of gluteus maximus are attached to the posterior surface of the ligament; superficial fibres of the lower part of the ligament continue into the tendon of biceps femoris. The ligament is pierced by the coccygeal branches of the inferior gluteal artery, the perforating cutaneous nerve and filaments of the coccygeal plexus. Sacrospinous ligament (Fig. 111.11) The thin, triangular sacrospinous ligament extends from the ischial spine to the lateral margins of the sacrum and coccyx anterior to the sacrotuberous ligament, with which it blends. Its anterior surface is coccygeus: muscle and ligament are coextensive and are the anterior and posterior aspects of the same structure. The sacrospinous ligament is often regarded as a degenerate part of coccygeus. Sciatic foramina (Fig. 111.11) The sacrotuberous and sacrospinous ligaments convert the sciatic notches into foramina. Greater sciatic foramen The greater sciatic foramen is bounded anterosuperiorly by the greater sciatic notch, posteriorly by the sacrotuberous ligament and inferiorly by the sacrospinous ligament and ischial spine. It is partly filled by the emerging piriformis, above which the superior gluteal vessels and nerve leave the pelvis. Below it, the inferior gluteal vessels and nerve, internal pudendal vessels and pudendal nerve, sciatic and posterior femoral cutaneous nerves and the nerves to obturator internus and quadratus femoris all leave the pelvis. Lesser sciatic foramen The lesser sciatic foramen is bounded anteriorly by the ischial body, superiorly by its spine and sacrospinous ligament, and posteriorly by the sacrotuberous ligament. It transmits the tendon of obturator internus, the nerve to obturator internus, and the internal pudendal vessels and pudendal nerve. Vascular supply
The arterial supply is derived from the iliolumbar, superior gluteal and superior lateral sacral arteries, with corresponding venous drainage. Lymphatic drainage follows the arteries, reaching the iliac and lumbar nodes. Innervation
The innervation of the sacroiliac joint is controversial and not generally agreed. Results of reported studies vary. It probably receives branches from the anterior and posterior primary rami of the first two sacral spinal nerves, and from the superior gluteal nerve. The obturator nerve and the lumbosacral trunk may also contribute. It is difficult to apply Hilton's law (p. 110) to this joint. Factors maintaining stability
The sacroiliac joint is one of the most stable joints in the body, and supports the weight of the trunk. The reciprocal irregularity of the joint surfaces allows very little movement. The tendency of the sacrum to be forced downwards by the trunk is resisted by the extremely strong posterior ligaments, while the iliolumbar ligaments help to resist displacement of the fifth lumbar vertebra over the sacrum. The sacrotuberous and sacrospinous ligaments oppose upward tilting of the lower part of the sacrum under downward thrust at its upper end. Movements
Primary movement of the sacroiliac joint is minimal: all muscles that cross the joint act on the lumbar spine or on the hip. Such movements as do occur are mainly imposed on the joint as the pelvis moves. Data from living subjects are technically difficult to obtain, and those based on plain radiographs are unreliable. Studies using implanted tantalum spheres and biplanar radiography have shown mean rotational ranges of less than 2°. Even when there is recordable movement, the direction of movement is irregular. Biplanar radiography has also shown that the axes of movement of the sacroiliac joint during hip movement are oblique, and that the axes differ in flexion and extension. During pregnancy the pelvic joints and ligaments loosen under the influence of the hormone relaxin. Movements in the joints increase. Relaxation renders the sacroiliac locking mechanism less effective, permitting greater rotation and perhaps allowing alterations in pelvic diameters at childbirth, although the effect is probably small. The impaired locking mechanism diverts the strain of weightbearing to the ligaments, with frequent sacroiliac strain after pregnancy. page 1439 page 1440
Relations and 'at risk' structures
The sacroiliac joints have many important anterior relations. The internal and external iliac veins join to form the common iliac veins immediately anteriorly, separating the joints from the bifurcations of the common iliac arteries and, more anteriorly, the ureters. The lumbosacral trunk and the obturator nerve cross the anterior aspect of the joint behind the vessels. Piriformis partly attaches to the anterior capsule, separating the joint from the upper part of the sacral plexus. Variants
Accessory sacroiliac articulations are not uncommon. They develop behind the articular surface between the lateral sacral crest and posterior superior iliac spine and iliac tuberosity, and are acquired fibrocartilaginous joints resulting from the stresses of weightbearing. They have a joint capsule, are saddle-shaped, and may be single, double, unilateral or bilateral (Weisl 1954).
HIP JOINT (Figs 111.24, 111.25, 111.26) The hip joint is a multiaxial synovial joint of ball-and-socket (spheroidal, cotyloid) type. Articular surfaces (Figs 111.3, 111.17, 111.25)
Figure 111.24 Anteroposterior radiograph of adult female pelvis.
Figure 111.25 Synovial cavity of left hip joint (distended): posterior aspect.
The femoral head articulates with the cup-shaped (cotyloid) acetabulum, its centre lying a little below the middle third of the inguinal ligament. (The profile of the anterior margin of the joint is parallel to the middle third of the inguinal ligament.) The articular surfaces are reciprocally curved but neither coextensive nor
completely congruent. The close-packed position is in full extension, with slight abduction and medial rotation. As in the shoulder joint, the surfaces are considered ovoid or spheroid rather than spherical, but this is controversial. Evidence favours spheroid and slightly ovoid surfaces, which become almost spherical with advancing age. The femoral head is covered by articular cartilage, except for a rough pit for the ligamentum teres. In front the cartilage extends laterally over a small area on the adjoining neck; it is thickest centrally. Cartilage thickness is maximal anterosuperiorly in the acetabulum and anterolaterally on the femoral head. The acetabular articular surface is an incomplete ring, the lunate surface, broadest above where the pressure of body weight falls in the erect posture, and narrowest in its pubic region. It is deficient inferiorly opposite the acetabular notch and covered by articular cartilage, which is thickest w