Comprehensive Management of Skull Base Tumors

about the book… The management of tumors in and adjacent to the skullbase is challenging given the complex and criticall...

Author: Ehab Y. Hanna | Franco DeMonte

64 downloads 1956 Views 67MB Size Report

This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form

DOWNLOAD PDF

about the book… The management of tumors in and adjacent to the skullbase is challenging given the complex and critically important anatomy of the region and the wide diversity of tumor pathologies that may be encountered. To help navigate the complexities of contemporary multidisciplinary management of these patients, Drs. Hanna and DeMonte bring you Comprehensive Management of Skull Base Tumors, a comprehensive guide filled with updated information from authorities around the world. Comprehensive Management of Skull Base Tumors is divided into three sections consisting of: • general principles • site specific surgery • tumor specific management Filled with scientific tables and lavishly illustrated, this text is written with an emphasis on surgery, radiation and chemotherapy, and will appeal to all neurosurgeons, otolaryngologists, plastic surgeons, maxillofacial surgeons, ophthalmologists, medical and radiation oncologists, and radiologists. about the editors... EHAB Y. HANNA is Professor of Head and Neck Surgery and Neurosurgery, The University of Texas M.D. Anderson Cancer Center, and Adjunct Professor of Otolaryngology, Baylor College of Medicine, Houston, Texas, USA. Dr. Hanna is also Vice Chair for Clinical Affairs and Medical Director of the Head and Neck Center at M.D. Anderson Cancer Center. He obtained his M.D. from Ain Shams University School of Medicine, Cairo, Egypt. Dr. Hanna currently serves on the editorial boards of several publications; he is Editor-in-Chief of Head and Neck and Editor of the Head and Neck Cancers Section of Current Oncology Reports. He is an involved member of the American Head and Neck Society, the North American Skull Base Society, the American Academy of Otolaryngology-Head and Neck Surgery, the American College of Surgeons, and the National Cancer Institute. Dr. Hanna has been nationally recognized as one of “America’s Top Doctors” and “America’s Top Doctors for Cancer”. He has been a frequent invited guest speaker at national and international conferences and lectures and has contributed to numerous peer-reviewed publications in the fields of head and neck cancer and skull base tumors. FRANCO DEMONTE is Professor of Neurosurgery and Head and Neck Surgery, Deputy Chairman of the Department of Neurosurgery, and Medical Director of the Brain and Spine Center, The University of Texas M.D. Anderson Cancer Center, Houston, Texas, USA. Dr. DeMonte is also an adjunct professor of neurosurgery at Baylor College of Medicine, Houston, Texas, USA. He received his M.D. from the University of Western Ontario, London, Canada. He is an active member of several organizations, including the Society of Neurological Surgeons, the American Association of Neurological Surgeons, and the Canadian Neurosurgical Society. He has served as president of the North American Skull Base Society and of the Houston Neurological Society. His clinical and educational activities have been recognized through his national and international presentations as well as by several teaching awards and inclusion in “Best Doctors in America” and “America’s Top Doctors”. Printed in the United States of America

Comprehensive Management of Skull Base Tumors

Otolaryngology/Neurosurgery

Comprehensive Management of Skull Base Tumors Edited by

Ehab Y. Hanna & Franco DeMonte

Hanna DK054X

•

DeMonte

Comprehensive Management of Skull Base Tumors

Comprehensive Management of Skull Base Tumors Edited by

Ehab Y. Hanna

The University of Texas M.D. Anderson Cancer Center Houston, Texas, USA

Franco DeMonte

The University of Texas M.D. Anderson Cancer Center Houston, Texas, USA

Informa Healthcare USA, Inc. 52 Vanderbilt Avenue New York, NY 10017 C

2009 by Informa Healthcare USA, Inc. Informa Healthcare is an Informa business No claim to original U.S. Government works Printed in the United States of America on acid-free paper 10 9 8 7 6 5 4 3 2 1 International Standard Book Number-10: 0-8493-4054-3 (Hardcover) International Standard Book Number-13: 978-0-8493-4054-3 (Hardcover) This book contains information obtained from authentic and highly regarded sources. Reprinted material is quoted with permission, and sources are indicated. A wide variety of references are listed. Reasonable efforts have been made to publish reliable data and information, but the author and the publisher cannot assume responsibility for the validity of all materials or for the consequence of their use. No part of this book may be reprinted, reproduced, transmitted, or utilized in any form by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying, microfilming, and recording, or in any information storage or retrieval system, without written permission from the publishers. For permission to photocopy or use material electronically from this work, please access www.copyright.com (http://www.copyright.com/) or contact the Copyright Clearance Center, Inc. (CCC) 222 Rosewood Drive, Danvers, MA 01923, 978-750-8400. CCC is a not-for-profit organization that provides licenses and registration for a variety of users. For organizations that have been granted a photocopy license by the CCC, a separate system of payment has been arranged. Trademark Notice: Product or corporate names may be trademarks or registered trademarks, and are used only for identification and explanation without intent to infringe.

Library of Congress Cataloging-in-Publication Data Comprehensive management of skull base tumors / edited by Ehab Y. Hanna, Franco DeMonte. p. ; cm. Includes bibliographical references and index. ISBN-13: 978-0-8493-4054-3 (hardcover : alk. paper) ISBN-10: 0-8493-4054-3 (hardcover : alk. paper) 1. Skull base–Cancer. 2. Skull base–Tumors–Surgery. I. Hanna, Ehab Y. II. DeMonte, Franco. [DNLM: 1. Skull Base Neoplasms. WE 707 C7374 2008] RD662.5.C66 2008 616.99 481–dc22 2008036404

For Corporate Sales and Reprint Permissions call 212-520-2700 or write to: Sales Department, 52 Vanderbilt Avenue, 16th floor, New York, NY 10017. Visit the Informa Web site at www.informa.com and the Informa Healthcare Web site at www.informahealthcare.com

This book is dedicated to My wife, Sylvie, for her grace, gentle spirit, and beauty; Our daughters, Gabrielle Grace “Gigi” Hanna and Camille Lauren Hanna, for the joy and blessing they bring to our lives; My parents, who encouraged me to follow my dreams; My Fellows, residents, and students, who continue to teach me; And my patients, Whose endurance, resilience, and faith continue to amaze me. Ehab Y. Hanna, MD

My parents, Dolinda and Giacomo, for their love, their sacrifices, and their dedication to my education. To my children, Evan and Madeline who fill me with pride. And, with limitless love, to Paula, my wife and best friend. Franco DeMonte, MD

Preface

Section two covers site-specific information regarding the various anatomic regions of the cranial base, including surgical anatomy, regional pathology, differential diagnosis, clinical assessment, diagnostic imaging, and surgical approaches. This regional classification includes the anterior cranial fossa, sinonasal region, nasopharynx, clivus, infratemporal fossa, parapharyngeal space, temporal bone, sella turcica, middle cranial fossa, petrous apex, cerebellopontine angle, jugular foramen, and craniovertebral junction. Section three covers comprehensive multidisciplinary discussion of tumor-specific topics such as tumor incidence and epidemiology, pathology, staging, treatment, outcome, and prognosis. This section covers the following tumors of the cranial base: squamous and nonsquamous cell carcinoma, olfactory neuroblastoma, melanoma, sarcomas, angiofibromas and other vascular tumors, chordomas and chondrosarcomas, meningiomas, schwannomas, paragangliomas, pituitary adenomas, craniopharyngiomas, epidermoid and dermoid cysts, fibro-osseous lesions, and metastatic tumors. This organizational schema is intended to provide a simple yet comprehensive way for readers to find the information they need. For example, the reader who wants to know about the latest advances in radiation therapy of skull base tumors is referred to the first section, a reader who has a patient with a tumor of the petrous apex is referred to the second section, and another who wants to know all the available treatment options and prognostic factors for esthesioneuroblastoma is referred to the third section. With such an organizational structure, some redundancy is unavoidable, but not, as you will see, detrimental. As with any textbook, some omissions are inevitable and we hope our readership will forgive any shortcomings of this work. We believe that the greatest value of this book is the incredible expertise of the contributing authors. They truly represent the world’s experts on their specific topics. We are honored by their contribution and humbled by their graciousness to join us in this work.

The management of patients with tumors of the skull base has evolved significantly in the last two decades. Major advances have been achieved in the surgical management of these patients, particularly in the areas of tumor resection and surgical reconstruction. These advances can be mainly attributed to the collaborative efforts of dedicated teams representing various surgical disciplines including neurosurgery, head and neck surgery, neuro-otology, oromaxillofacial surgery, ophthalmology, and plastic and reconstructive surgery. Meanwhile significant advances in radiation delivery using various methods of conformal therapy, including three-dimensional computerized tomography (3D-CT), intensity-modulated radiation therapy (IMRT), proton beam therapy, and stereotactic radiation, as well as advances in chemotherapy and targeted biologic therapy have added significantly to the menu of treatment options for patients with tumors of the skull base. Although there are many excellent references describing the surgical management of patients with tumors of the cranial base, this textbook is intended to be a comprehensive guide to help navigate the complexity of contemporary multidisciplinary management of these patients. In addition, we hope that this reference will also provide the reader with deeper understanding of the unique biologic behavior and the underlying molecular and genetic aberrations of the various tumor types originating from or involving the cranial base, and the potential for these molecular derangements to be putative targets for future development of more effective biologic therapy. To address these goals, we have organized the book in three sections: general principles, site-specific chapters, and tumor-specific chapters. Section one covers general topics pertinent to all patients with neoplasms of the skull base, regardless of specific location or tumor type. These topics include anatomy, pathology, genetics, clinical evaluation, diagnostic imaging, anesthesia, minimally invasive surgery, surgical reconstruction, prosthetic rehabilitation, radiation and radiobiology, chemotherapy, evaluation and rehabilitation of speech and swallowing, functional outcomes and quality of life issues, neurocognitive assessment, and cerebrovascular management.

Ehab Y. Hanna Franco DeMonte

v

Contents

Preface . . . . v Contributors . . . . ix

12. Rehabilitation of Speech and Swallowing of Patients with Tumors of the Skull Base 181 Gail L. Davie, Denise A. Barringer, and Jan S. Lewin

Section 1: General Principles

13. Quality of Life of Patients with Skull Base Tumors 189 Ziv Gil and Dan M. Fliss

1. Anatomy of the Cranial Base 3 Carolina Martins and Albert L. Rhoton, Jr.

14. Neurocognitive Assessment of Patients with Tumors of the Skull Base 201 Mariana Witgert and Tracy Veramonti

2. Pathology of Tumor and Tumor-like Lesions of the Skull Base 43 Michelle D. Williams and Adel K. El-Naggar

15. Cerebrovascular Management in Skull Base Tumors 207 Sabareesh Kumar Natarajan, Basavaraj Ghodke, and Laligam N. Sekhar

3. Genetic Abnormalities of Skull Base Tumors 63 Ziv Gil and Dan M. Fliss 4. Imaging of Skull Base Neoplasms 81 Lawrence E. Ginsberg

Section 2: Site-Specific Considerations

5. Head, Neck, and Neuro-otologic Assessment of Patients with Tumors of the Skull Base: Clinical Examination, Auditory Testing, Vestibular Testing, and Equilibrium 95 Paul W. Gidley

16. Surgical Management of Tumors of the Nasal Cavity, Paranasal Sinuses, Orbit, and Anterior Skull Base 227 Ehab Y. Hanna, Michael Kupferman, and Franco DeMonte

6. Anesthesia and Intraoperative Monitoring of Patients with Tumors of the Skull Base 119 Walter S. Jellish and Steven B. Edelstein

17. Tumors of the Nasopharynx 267 William Ignace Wei and Paul K. Y. Lam 18. Clival Tumors 277 Franco DeMonte, Mark J. Dannenbaum, and Ehab Y. Hanna

7. Minimally Invasive Techniques: Endonasal Endoscopic Skull Base Surgery 131 Allan D. Vescan, Ricardo L. Carrau, Carl H. Snyderman, Amin B. Kassam, Arlan Mintz, and Paul Gardner

19. Tumors of the Anterior Skull Base 293 Vijayakumar Javalkar, Bharat Guthikonda, Prasad Vannemreddy, and Anil Nanda

8. Reconstruction of Skull Base Defects 139 Peter C. Neligan, Christine B. Novak, and Patrick J. Gullane

20. Infratemporal/Middle Fossa Tumors 305 Paul J. Donald

9. Prosthetic Rehabilitation of Patients Undergoing Skull Base Surgery 149 Theresa M. Hofstede, Rhonda F. Jacob, Pattii C. Montgomery, Peggy J. Wesley, Jack W. Martin, and Mark S. Chambers

21. Tumors of the Parapharyngeal Space 331 Eric J. Moore and Kerry D. Olsen 22. Tumors of the Temporal Bone 345 Sam J. Marzo and John P. Leonetti

10. Radiobiology and Radiation Therapy of Skull Base Tumors 159 Simon S. Lo, John H. Suh, and Eric L. Chang

23. Evaluation and Management of Sellar Tumors 355 Jay Jagannathan, Edward R. Laws, and John A. Jane, Jr.

11. Chemotherapy for Skull Base Tumors 175 Bilal Ahmed, Ehab Y. Hanna, and Merrill S. Kies

24. Tumors of the Middle Cranial Fossa 367 Ali A. Baaj, Siviero Agazzi, and Harry R. van Loveren

vii

viii

Contents

25. Tumors of the Petrous Apex 375 Ricardo Ramina, Maur´ıcio Coelho Neto, Yvens Barbosa Fernandes, Erasmo Barros da Silva, Jr., and Kristofer Luiz Fingerle Ramina 26. Tumors of the Cerebellopontine Angle 389 Bryan C. Oh, Daniel J. Hoh, and Steven L. Giannotta 27. Tumors of the Jugular Foramen 403 Samer Ayoubi, Badih Adada, and Ossama Al-Mefty 28. Tumors of the Craniovertebral Junction 417 Douglas Fox, Scott Wait, Steve Chang, G. Vini Khurana, Curtis A. Dickman, Volker K. H. Sonntag, and Robert F. Spetzler

Section 3: Tumor-Specific Considerations 29. Squamous Cell Carcinoma of the Nasal Cavity and Paranasal Sinuses 429 Patrick Sheahan, Snehal G. Patel, and Jatin P. Shah 30. Nonsquamous Cell Carcinoma of the Nasal Cavity and Paranasal Sinuses 445 ` Carlo L. Solero, Stefano Riccio, and Giulio Cantu, Sarah Colombo 31. Esthesioneuroblastoma 453 Valerie J. Lund and David J. Howard 32. Melanoma of the Nasal Cavity and Paranasal Sinuses 459 Ziv Gil, Mark H. Bilsky, and Dennis H. Kraus 33. Sarcomas of the Skull Base 473 Katherine A. Thornton and Robert S. Benjamin

34. Angiofibromas and Vascular Tumors of the Skull Base 481 Andrew G. Sikora and Randal S. Weber 35. Chordoma and Chondrosarcoma of the Skull Base 495 Gordon T. Sakamoto and Griffith R. Harsh 36. Meningioma 503 Ashwin Viswanathan and Franco DeMonte 37. Schwannomas of the Skull Base 513 Daniel W. Nuss and Emily Lifsey Burke 38. Paragangliomas of the Head and Neck 539 David P. Goldstein, Mark G. Shrime, Bernard Cummings, and Patrick J. Gullane 39. Pituitary Adenomas 557 Mark Hornyak and William T. Couldwell 40. Craniopharyngioma: Neurosurgical Management 573 Douglas James Cook and James T. Rutka 41. Epidermoids, Dermoids, and Other Cysts of the Skull Base 583 Samuel P. Gubbels and Bruce J. Gantz 42. Fibro-Osseous Lesions of the Skull Base 597 Ian T. Jackson 43. Metastatic Skull Base Tumors 615 Krishna Satyan and Sujit S. Prabhu Index . . . . 619

Contributors

Maur´ıcio Coelho Neto Department of Neurosurgery, Neurological Institute of Curitiba, Curitiba, Brazil

Badih Adada Department of Neurosurgery, The University of Arkansas for Medical Sciences, Little Rock, Arkansas, U.S.A.

Sarah Colombo Cranio-Maxillo-Facial Department, National Cancer Institute, Milano, Italy

Siviero Agazzi Department of Neurosurgery, University of South Florida, Tampa, Florida, U.S.A.

Douglas James Cook Division of Neurosurgery, University of Toronto, Toronto, Ontario, Canada

Bilal Ahmed Departments of Thoracic/Head and Neck Medical Oncology, The University of Texas M.D. Anderson Cancer Center, Houston, Texas, U.S.A.

William T. Couldwell Department of Neurosurgery, University of Utah, Salt Lake City, Utah, U.S.A.

Ossama Al-Mefty Department of Neurosurgery, University of Arkansas for Medical Sciences, Little Rock, Arkansas, U.S.A.

Bernard Cummings Department of Radiation Oncology, Princess Margaret Hospital, University of Toronto, Toronto, Ontario, Canada

Samer Ayoubi Department of Neurosurgery, Abbassi Medical Center, Damascus, Syria

Mark J. Dannenbaum Department of Neurosurgey, The University of Texas M.D. Anderson Cancer Center, Houston, Texas, U.S.A.

Ali A. Baaj Department of Neurosurgery, University of South Florida, Tampa, Florida, U.S.A.

Gail L. Davie Department of Head and Neck Surgery, The University of Texas M.D. Anderson Cancer Center, Houston, Texas, U.S.A.

Denise A. Barringer Department of Head and Neck Surgery, The University of Texas M.D. Anderson Cancer Center, Houston, Texas, U.S.A.

Franco DeMonte Department of Neurosurgey, The University of Texas M.D. Anderson Cancer Center, Houston, Texas, U.S.A.

Erasmo Barros da Silva, Jr. Department of Neurosurgery, Neurological Institute of Curitiba, Curitiba, Brazil, Postgraduate Course in Surgery, Pontifical Catholic University of Parana, Curitiba, Brazil

Curtis A. Dickman Division of Neurological Surgery, Barrow Neurological Institute, St. Joseph’s Hospital and Medical Center, Phoenix, Arizona, U.S.A.

Robert S. Benjamin Department of Sarcoma Medical Oncology, The University of Texas M.D. Anderson Cancer Center, Houston, Texas, U.S.A.

Paul J. Donald Department of Otolaryngology-Head and Neck Surgery, University of California-Davis, Sacramento, California, U.S.A.

Mark H. Bilsky Department of Neurosurgery, Memorial Sloan-Kettering Cancer Center, Cornell University Medical College, New York, New York, U.S.A.

Steven B. Edelstein Loyola University Medical Center and Loyola University Stritch School of Medicine, Department of Anesthesiology, Maywood, Illinois, U.S.A.

Emily Lifsey Burke Department of Otolaryngology-Head & Neck Surgery, Louisiana State University Health Sciences Center, New Orleans and Baton Rouge, Louisiana, U.S.A.

Adel K. El-Naggar Department of Pathology, The University of Texas M.D. Anderson Cancer Center, Houston, Texas, U.S.A.

Giulio Cantu´ Cranio-Maxillo-Facial Department, National Cancer Institute, Milano, Italy

Yvens Barbosa Fernandes Department of Neurosurgery, Neurological Institute of Curitiba, Department of Neurosurgery, State University of Campinas (UNICAMP), Campinas, Brazil

Ricardo L. Carrau Department of Otolaryngology and Head & Neck Surgery, Department of Neurological Surgery, and Minimally Invasive Endoneurosurgery Center, University of Pittsburgh Medical Center, Pittsburgh, Pennsylvania, U.S.A.

Dan M. Fliss Department of Otolaryngology Head and Neck Surgery, Tel-Aviv Sourasky Medical Center, Sackler School of Medicine, Tel Aviv University, Tel Aviv, Israel

Mark S. Chambers Section of Oncologic Dentistry and Prosthodontics, Department of Head and Neck Surgery, The University of Texas M.D. Anderson Cancer Center, Houston, Texas, U.S.A.

Douglas Fox

Private practice, Austin, Texas, U.S.A.

Bruce J. Gantz Department of Otolaryngology/Head and Neck Surgery, University of Iowa Hospitals and Clinics, Iowa City, Iowa, U.S.A.

Eric L. Chang Department of Radiation Oncology, The University of Texas M.D. Anderson Cancer Center, Houston, Texas, U.S.A.

Paul Gardner Department of Neurological Surgery and Minimally Invasive Endoneurosurgery Center, University of Pittsburgh Medical Center, Pittsburgh, Pennsylvania, U.S.A.

Steve Chang Division of Neurological Surgery, Barrow Neurological Institute, St. Joseph’s Hospital and Medical Center, Phoenix, Arizona, U.S.A. ix

x

Contributors

Paul W. Gidley Department of Head and Neck Surgery, The University of Texas M.D. Anderson Cancer Center, Houston, Texas, U.S.A.

Vijayakumar Javalkar Department of Neurosurgery, Louisiana State University Health Sciences Center — Shreveport, Louisiana, U.S.A.

Basavaraj Ghodke Departments of Neurological Surgery, and Neuroradiology, University of Washington, Seattle, Washington, U.S.A.

Walter S. Jellish Department of Anesthesiology, Loyola University Hospital Medical Center, Maywood, Illinois, U.S.A.

Steven L. Giannotta Department of Neurological Surgery, Keck School of Medicine, University of Southern California, Los Angeles, California, U.S.A.

Amin B. Kassam Department of Neurological Surgery, Department of Otolaryngology and Head & Neck Surgery, and Minimally Invasive Endoneurosurgery Center, University of Pittsburgh Medical Center, Pittsburgh, Pennsylvania, U.S.A.

Ziv Gil Department of Otolaryngology Head and Neck Surgery, Tel-Aviv Sourasky Medical Center, Sackler School of Medicine, Tel Aviv University, Tel Aviv, Israel Lawrence E. Ginsberg Department of Radiology, The University of Texas M.D. Anderson Cancer Center, Houston, Texas, U.S.A. David P. Goldstein Department of Surgical Oncology, Princess Margaret Hospital, Toronto, Ontario, Department of Otolaryngology– Head and Neck Surgery, University of Toronto, Toronto, Ontario, Canada Samuel P. Gubbels Department of Surgery, Division of Otolaryngology/Head and Neck Surgery, University of Wisconsin, Madison, Wisconsin, U.S.A. Patrick J. Gullane Department of Otolaryngology– Head and Neck Surgery, University of Toronto, Department of Surgical Oncology, Princess Margaret Hospital, Toronto, Ontario, Canada Bharat Guthikonda Department of Neurosurgery, Louisiana State University, Health Sciences Center—Shreveport, Louisiana, U.S.A.

G. Vini Khurana The Canberra Hospital, Department of Neurosurgery, Canberra Medical School, Australian National University, Canberra, Australian Capital Territory, Australia Merrill S. Kies Departments of Thoracic/Head and Neck Medical Oncology, The University of Texas M.D. Anderson Cancer Center, Houston, Texas, U.S.A. Dennis H. Kraus Head and Neck Service, Department of Surgery, Memorial Sloan-Kettering Cancer Center, Cornell University Medical College, New York, New York, U.S.A. Michael Kupferman Department of Head and Neck Surgery, The University of Texas M.D. Anderson Cancer Center, Houston, Texas, U.S.A. Paul K. Y. Lam Division of Otorhinolaryngology, Head and Neck Surgery, University of Hong Kong Medical Centre, Queen Mary Hospital, Hong Kong SAR, P.R. China Edward R. Laws Department of Neurosurgery, Harvard University, Boston, Massachusetts, U.S.A.

Ehab Y. Hanna Department of Head and Neck Surgery, The University of Texas M.D. Anderson Cancer Center, Houston, Texas, U.S.A.

John P. Leonetti Department of Otolaryngology, Head and Neck Surgery, Loyola University Health System, Maywood, Illinois, U.S.A.

Griffith R. Harsh Department of Neurosurgery, Stanford University, Palo Alto, California, U.S.A.

Jan S. Lewin Department of Head and Neck Surgery, The University of Texas M.D. Anderson Cancer Center, Houston, Texas, U.S.A.

Theresa M. Hofstede Section of Oncologic Dentistry and Prosthodontics, Department of Head and Neck Surgery, The University of Texas M.D. Anderson Cancer Center, Houston, Texas, U.S.A. Daniel J. Hoh Department of Neurological Surgery, Keck School of Medicine, University of Southern California, Los Angeles, California, U.S.A. Mark Hornyak Department of Neurosurgery, University of Utah, Salt Lake City, Utah, U.S.A.

Simon S. Lo Department of Radiation Medicine, Arthur G. James Cancer Hospital and Richard J. Solove Research Institute, Ohio State University Medical Center, Columbus, Ohio, U.S.A. Valerie J. Lund Royal National Throat Nose and Ear Hospital, Ear Institute, University College London, London, U.K.

David J. Howard Royal National Throat Nose and Ear Hospital, Ear Institute, University College London, London, U.K.

Jack W. Martin Section of Oncologic Dentistry and Prosthodontics, Department of Head and Neck Surgery, The University of Texas M.D. Anderson Cancer Center, Houston, Texas, U.S.A.

Ian T. Jackson Craniofacial Institute, Southfield, Michigan, U.S.A.

Carolina Martins Medical School of Pernambuco IMIP & ´ Hospital Getulio Vargas, Recife, Brazil

Rhonda F. Jacob Section of Oncologic Dentistry and Prosthodontics, Department of Head and Neck Surgery, The University of Texas M.D. Anderson Cancer Center, Houston, Texas, U.S.A.

Sam J. Marzo Department of Otolaryngology, Head and Neck Surgery, Loyola University Health System, Maywood, Illinois, U.S.A.

Jay Jagannathan Department of Neurosurgery, University of Virginia Health Sciences Center, University of Virginia, Charlottesville, Virginia, U.S.A. John A. Jane, Jr. Department of Neurosurgery, University of Virginia Health Sciences Center, University of Virginia, Charlottesville, Virginia, U.S.A.

Arlan Mintz Department of Neurological Surgery and Minimally Invasive Endoneurosurgery Center, University of Pittsburgh Medical Center, Pittsburgh, Pennsylvania, U.S.A. Pattii C. Montgomery Section of Oncologic Dentistry and Prosthodontics, Department of Head and Neck Surgery, The University of Texas M.D. Anderson Cancer Center, Houston, Texas, U.S.A.

Contributors

Eric J. Moore Department of Otolaryngology, The Mayo Graduate College of Medicine, The Mayo Clinic, Rochester, Minnesota, U.S.A. Anil Nanda Department of Neurosurgery, Louisiana State University Health Sciences Center—Shreveport, Louisiana, U.S.A. Sabareesh Kumar Natarajan Department of Neurological Surgery, University of Washington, Seattle, Washington, U.S.A. Peter C. Neligan Division of Plastic Surgery, University of Washington, Seattle, Washington, U.S.A. Christine B. Novak Wharton Head and Neck Center, University Health Network, University of Toronto, Toronto, Ontario, Canada Daniel W. Nuss Department of Otolaryngology-Head & Neck Surgery, and Department of Neurological Surgery, Louisiana State University Health Sciences Center, New Orleans and Baton Rouge, Louisiana, U.S.A. Bryan C. Oh Department of Neurological Surgery, Keck School of Medicine, University of Southern California, Los Angeles, California, U.S.A. Kerry D. Olsen Department of Otolaryngology, The Mayo Graduate College of Medicine, The Mayo Clinic, Rochester, Minnesota, U.S.A. Snehal G. Patel Head and Neck Service, Department of Surgery, Memorial Sloan-Kettering Cancer Center, New York, New York, U.S.A. Sujit S. Prabhu Department of Neurosurgery, The University of Texas M.D. Anderson Cancer Center, Houston, Texas, U.S.A. Kristofer Luiz Fingerle Ramina Department of ¨ ¨ Neurosurgery, University of Tubingen, Tubingen, Germany Ricardo Ramina Department of Neurosurgery, Neurological Institute of Curitiba, Curitiba, Brazil, Postgraduate Course in Surgery, Pontifical Catholic University of Parana, Curitiba, Brazil Albert L. Rhoton, Jr. Department of Neurosurgery, University of Florida, Gainesville, Florida, U.S.A. Stefano Riccio Cranio-Maxillo-Facial Department, National Cancer Institute, Milano, Italy James T. Rutka Division of Neurosurgery, The Hospital for Sick Children, The University of Toronto, Toronto, Ontario, Canada Gordon T. Sakamoto Stanford University, Palo Alto, California, U.S.A. Krishna Satyan Department of Neurosurgery, The University of Texas M.D. Anderson Cancer Center, Houston Texas, U.S.A.

xi

Mark G. Shrime Department of Surgical Oncology, Princess Margaret Hospital, Toronto, Ontario, Department of Otolaryngology– Head and Neck Surgery, University of Toronto, Toronto, Ontario, Canada Andrew G. Sikora Department of Otolaryngology, Mount Sinai School of Medicine, New York, New York, U.S.A. Carl H. Snyderman Department of Otolaryngology and Head & Neck Surgery, Department of Neurological Surgery, and Minimally Invasive Endoneurosurgery Center, University of Pittsburgh Medical Center, Pittsburgh, Pennsylvania, U.S.A. Carlo L. Solero Second Neurosurgical Unit, Istituto Nazionale Neurologico “C. Besta”, Milano, Italy Volker K. H. Sonntag Division of Neurological Surgery, Barrow Neurological Institute, St. Joseph’s Hospital and Medical Center, Phoenix, Arizona, U.S.A. Robert F. Spetzler Division of Neurological Surgery, Barrow Neurological Institute, St. Joseph’s Hospital and Medical Center, Phoenix, Arizona, U.S.A. John H. Suh Department of Radiation Oncology, The Cleveland Clinic Foundation, Cleveland, Ohio, U.S.A. Katherine A. Thornton Johns Hopkins University School of Medicine, Sidney Kimmel Comprehensive Cancer Center, Baltimore, Maryland, U.S.A. Harry R. van Loveren Department of Neurosurgery, University of South Florida, Tampa, Florida, U.S.A. Prasad Vannemreddy Department of Neurosurgery, Louisiana State University Health Sciences Center — Shreveport, Louisiana, U.S.A. Tracy Veramonti Department of Neuro-Oncology, The University of Texas M.D. Anderson Cancer Center, Houston, Texas, U.S.A. Ashwin Viswanathan Department of Neurosurgery, The University of Texas M.D. Anderson Cancer Center, Houston, Texas, U.S.A. Allan D. Vescan Department of Otolaryngology and Head & Neck Surgery, and Minimally Invasive Endoneurosurgery Center, University of Pittsburgh Medical Center, Pittsburgh, Pennsylvania, U.S.A. Scott Wait Division of Neurological Surgery, Barrow Neurological Institute, St. Joseph’s Hospital and Medical Center, Phoenix, Arizona, U.S.A. Randal S. Weber Department of Head and Neck Surgery, The University of Texas M.D. Anderson Cancer Center, Houston, Texas, U.S.A. William Ignace Wei Division of Otorhinolaryngology, Head and Neck Surgery, University of Hong Kong Medical Centre, Queen Mary Hospital, Hong Kong SAR, P.R. China

Laligam N. Sekhar Department of Neurological Surgery, University of Washington, Seattle, Washington, U.S.A.

Peggy J. Wesley Section of Oncologic Dentistry and Prosthodontics, Department of Head and Neck Surgery, The University of Texas M.D. Anderson Cancer Center, Houston, Texas, U.S.A.

Jatin P. Shah Head and Neck Service, Department of Surgery, Memorial Sloan-Kettering Cancer Center, New York, New York, U.S.A.

Michelle D. Williams Department of Pathology, The University of Texas M.D. Anderson Cancer Center, Houston, Texas, U.S.A.

Patrick Sheahan Head and Neck Service, Department of Surgery, Memorial Sloan-Kettering Cancer Center, New York, New York, U.S.A.

Mariana Witgert Department of Neuro-Oncology, The University of Texas M.D. Anderson Cancer Center, Houston, Texas, U.S.A.

Section 1 General Principles

1 Anatomy of the Cranial Base Carolina Martins and Albert L. Rhoton, Jr.

OVERVIEW

galli and the cribriform plate of the ethmoid bone anteriorly and the planum of the sphenoid body posteriorly. The lateral part, which covers the orbit and the optic canal, is formed by the frontal bone and the lesser wing of the sphenoid bone, which blends medially into the anterior clinoid process (Figs. 3 and 4). The foramen caecum in the midline serves as the site of passage of an emissary vein and the cribriform plate is pierced by the filaments of the olfactory nerve. The optic canal transmits the optic nerve and the ophthalmic artery. The anterior cranial base faces the frontal lobes with the gyri recti medially and the orbital gyri laterally, along with the branches of the anterior cerebral arteries medially and middle cerebral arteries laterally.

No part of the cranial base is immune to surgical pathology or to its use as a pathway to access lesions in the intra- or extracranial spaces. Tumors and multiple other lesions can involve any of the intracranial fossae, and can appear in the paranasal sinuses, nasal cavity, infratemporal and pterygopalatine fossae, orbit, and in the retropharyngeal and craniocervical regions (Fig. 1). Managing these lesions requires an extensive knowledge of the anatomy of the cranial base and its intra- and extracranial relationships. This chapter provides a concise review of the cranial base. The skull is divided into the cranium and the facial skeleton. The cranium is divided into the calvarium and the cranial base. The cranial base has an endocranial surface, which faces the brain, and an exocranial surface, which faces the nasal cavity and sinuses, orbits, pharynx, infratemporal and pterygopalatine fossae, and the parapharyngeal and infrapetrosal spaces (Fig. 2). Both surfaces are connected by canals, foramina, and fissures through which numerous neural and vascular structures pass. Both the endocranial and exocranial cranial base surfaces are divided into anterior, middle, and posterior parts, each of which has a central and paired lateral portions. On the intracranial side, the three parts correspond to the anterior, middle, and posterior cranial fossae (Figs. 2 and 4) (1,2). On the endocranial side, the border between the anterior and middle cranial bases is the sphenoid ridge joined medially by the chiasmatic sulcus, and the border between the middle and posterior cranial bases is formed by the petrous ridges joined by the dorsum sellae and posterior clinoid processes. On the exocranium side, the anterior and middle cranial bases are divided at the level of a transverse line extending through the pterygomaxillary fissures and the pterygopalatine fossae at the upper level and the posterior edge of the alveolar process of the maxilla at a lower level. Medially, this corresponds to the anterior part of the attachment of the vomer to the sphenoid bone. The middle and posterior cranial bases are separated by a transverse line crossing at or near the posterior border of the vomer– sphenoid junction, the foramen lacerum, carotid canal, jugular foramen, styloid process, and the mastoid tip. The osseous structures, their foramina and fissures, canals and their muscular, and neural and vascular relationships are described in this chapter.

Exocranial Surface On the exocranial side, the anterior cranial base is divided in a medial part related to the ethmoid and the sphenoid sinuses with the nasal cavity below and a lateral part that corresponds to the orbit and maxilla (Figs. 2, 6, 7, and 8) (1). The ethmoid bone forms the anterior and middle third and the sphenoid body forms the posterior third of the medial part. The ethmoid is formed by the cribriform plate with the olfactory fila traveling through it, the perpendicular plate, which joins the vomer in forming the nasal septum, and two lateral plates located in the medial walls of the orbits. The lateral plates separate the lateral wall of the nasal cavity and the orbit. The superior turbinate, an appendage of the ethmoid bone, projects into the superior part of the nasal cavity. The body of the sphenoid bone harbors the sphenoid sinus just below the planum sphenoidale, with the anterior orifices located above the superior turbinate. The orbital roof is formed by the lesser sphenoid wing and by the orbital plate of the frontal bone; the lateral wall is formed by the greater sphenoid wing and the zygomatic bone; the inferior wall is formed by the zygomatic, maxillary, and palatine bones; and the medial wall is formed by maxillary, lacrimal, and ethmoid bones (3,4). The main foramina of the region are the anterior and posterior ethmoidal foramen located in the superomedial orbital wall, transmitting the anterior and posterior ethmoidal nerves and arteries; the supraorbital and supratrochlear notches or foramina, transmitting the arteries and nerves of the same name; and the optic canal, through which the optic nerve and ophthalmic artery pass (Figs. 4, 5, and 7). The superior orbital fissure is located between the lesser and greater wing of the sphenoid bone on the lateral side of the optic canal. The inferior orbital fissure, located between the greater sphenoid wing behind and the maxillary and palatine bones anteriorly, is closed by fibrous tissue and orbital muscle. Covered with periorbita and filled with a great amount of fat, the orbit is divided into an anterior space where the globe lies and a posterior space that shelters the nerves, vessels, and muscles behind the globe (5). The

ANTERIOR CRANIAL BASE Endocranial Surface The anterior endocranial surface, formed by the ethmoid, sphenoid, and frontal bones, is divided into medial and lateral portions (Figs. 3–5). The medial part, covering the upper nasal cavity and the sphenoid sinus, is formed by the crista 3

4

Martins and Rhoton

A

Car. A.

B

Orbit roof Sphen. sinus Eth. sinus

V3 V2

Pterygopal. fossa Infratemp. fossa

Car. A.

V2 V3

Sphen. sinus Nasal septum

Nasal cavity

Maxilla

Pterygopal. fossa Nasal cavity

Eth. sinus Maxilla

Orbit roof

C

D

Pit. stalk

Car. A.

Car. A.

Sella Orb. apex

Max. sinus Eust. tube Vomer

Sphen. sinus

V3 Front. N. V2 Max. A. CN II Eth. air cell Pterygopal. fossa Nasopharynx Max. sinus

Septal cart.

Infraorb. N. Max. sinus Nasolac. duct

Sphen. sinus

V3 Pterygopal. gang.

Pteryg. M. Max. A. Br. Vomer Nasal cavity

Maxilla Alv. part

Figure 1

Chapter 1: Anatomy of the Cranial Base

annular tendon of Zinn, a fibrous ring that surrounds the central part of the superior orbital fissure and the optic canal, gives attachment to the superior, medial, inferior, and lateral rectus muscles (Fig. 4). The superior oblique attaches above the annular tendon and the inferior oblique arises from the inferomedial orbital wall just behind the rim. The oculomotor foramen, located inside the annular tendon and through which the oculomotor nerve passes, is located between the upper and lower attachment of the lateral rectus muscle. Just before passing through the superior orbital fissure and the oculomotor foramen in the annular tendon, the oculomotor nerve divides into an upper division supplying the superior rectus and levator muscles and a lower division to the medial and inferior rectus and inferior oblique muscles. The oculomotor nerve gives rise to the parasympathetic motor root to the ciliary ganglion which lies lateral to the optic nerve. The abducens nerve passes through the oculomotor foramen and enters the medial surface of the lateral rectus muscle. The ophthalmic nerve divides just behind the annular tendon into lacrimal and frontal nerves which pass outside the annular tendon, and into the nasociliary nerve which passes through the annular tendon. The ophthalmic nerve gives rise to the long ciliary nerves and the sensory root to the ciliary ganglion; the former conveys the sympathetic pupillomotor fibers and the latter conveys corneal sensation. The trochlear nerve passes above and outside the superomedial edge of the annular tendon. The optic nerve passes superior and medial from the globe to reach the optic canal and divides the retro-orbital space in medial and lateral parts. The main arterial supply to the orbit is by the ophthalmic artery and its branches. This artery courses below the optic nerve in the optic canal, crosses to the lateral side of the nerve at the orbital apex, and then courses from lateral to medial above the optic nerve. The main branches are the central retinal artery and the lacrimal, ciliary, ethmoidal, supraorbital, and dorsal nasal arteries, plus numerous muscular branches. The main venous drainage of the orbit is through the superior and inferior ophthalmic veins that exit the orbit by passing outside the annular tendon and through the superior orbital fissure. The lacrimal gland, located in the superolateral part of the orbit, receives its sensory innervation from the lacrimal nerve, and its parasympathetic and sympathetic innervation from the greater and deep petrosal nerves. The petrosal nerves

5

join to form the vidian nerve that enters the pterygopalatine ganglion, which sends branches to the zygomatic nerve that anastomose with the lacrimal nerve to reach the gland.

MIDDLE CRANIAL BASE Endocranial Surface The endocranial surface of the middle portion of the middle cranial base, formed by the sphenoid and temporal bones, has medial and lateral parts (Figs. 2, 3, 5, and 9). The medial part is formed by the body of the sphenoid bone, the site of the tuberculum sellae, pituitary fossa, middle and posterior clinoid processes, the carotid sulcus, and the dorsum sellae (Fig. 8). The lateral part is formed by the lesser and greater sphenoid wings, with the superior orbital fissure between them (Figs. 3 and 5). The lesser wing is connected to the body of the sphenoid bone by an anterior root, which forms the roof of the optic canal, and by a posterior root, also called the optic strut, which forms the floor of the optic canal and separates the optic canal from the superior orbital fissure (Fig. 3). The greater wing forms the largest part of the endocranial surface of the middle fossa, with the squamosal and the petrosal parts of the temporal bone completing this surface. The superior orbital fissure transmits the oculomotor, trochlear, ophthalmic, and abducens nerves, a recurrent meningeal artery, and the superior and inferior ophthalmic veins (6). The maxillary and mandibular nerves pass through the foramen rotundum and ovale, both located in the greater wing of the sphenoid. The not infrequently occurring sphenoidal emissary foramen, located anteromedial to the foramen spinosum, gives passage to a vein connecting the cavernous sinus and the pterygoid venous plexus. The upper surface of the petrous bone is grooved along the course of the greater and lesser petrosal nerves (Fig. 5) (7). The carotid canal extends upward and medially and provides passage to the internal carotid artery and carotid sympathetic nerves in their course to the cavernous sinus. The posterior trigeminal root reaches the middle fossa and the impression on the upper surface of the petrous bone where Meckel’s cave and the semilunar ganglion sit. The roof of the carotid canal opens below the trigeminal ganglion near the distal end of the carotid canal (Figs. 5, 6, and 9). The arcuate eminence approximates

←−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−− Figure 1 Anterior and middle cranial base. (A) On the left side, the floor of the anterior fossa and the upper portion of the maxilla have been removed to expose the structures deep to the anterior and middle cranial fossa. The frontal, ethmoidal, and sphenoid sinuses and the nasal cavity lie below the medial part of the anterior cranial base. The orbit and maxilla are located below the lateral part of the anterior cranial base. The sphenoid sinus and sella are located in the medial part of the middle cranial base, and the infratemporal and pterygopalatine fossa are located below the lateral part of the middle cranial base. The carotid arteries pass upward on the medial part of the middle cranial base and are intimately related to the sphenoid and cavernous sinuses. The infratemporal fossa, which contains branches of the mandibular nerve, pterygoid muscles, pterygoid venous plexus, and maxillary artery, is located below the middle cranial base and greater sphenoid wing. The alveolar process of the maxilla, which encloses the roots of the upper teeth, has been preserved on the left side. The maxillary nerve enters the pterygopalatine fossa, which is located medial to the infratemporal fossa between the posterior wall of the maxilla and the pterygoid process of the sphenoid bone. (B) Superior view of the anterior and middle cranial base. The infratemporal fossa is located posterolateral to the maxilla. The right ethmoid air cells are exposed on the medial side of the right orbit. The nasal cavity extends upward between the ethmoid sinuses. (C) Oblique anterior view. The facial structures on the right side have been removed to expose the orbital apex located above the maxillary sinus. The wall of the right maxillary sinus forms the floor of the orbit, much of the lateral wall of the nasal cavity, and the anterior wall of the pterygopalatine and infratemporal fossa. On the left side, the mandibular nerve enters the infratemporal fossa. The maxillary nerve enters the pterygopalatine fossa, which is located in the lateral wall of the nasal cavity and contains the maxillary nerve, pterygopalatine ganglion, and terminal branches of the maxillary artery. (D) Anterior view. The orbital apex is located above the pterygopalatine fossa. The frontal branch of the ophthalmic nerve passes along the roof of the orbit, and the infraorbital branch of the maxillary nerve courses in the floor of the orbit. The posterior ethmoid air cells are located medial to the orbital apex. The vomer forms the posterior part of the nasal septum and attaches to the maxilla and palatine bones below and to the body of the sphenoid bone above. The sphenoid sinus is located in the middle cranial base below the sella turcica. The upper brain stem is seen in the posterior part of the exposure (1).) Abbreviations: A., artery; Alv., alveolar; Br., branch; Car., carotid; Cart., cartilage; CN, cranial nerve; Eth., ethmoid; Eust., eustachian; Front., frontal; Gang., ganglion; Infraorb., infraorbital; Infratemp., infratemporal; M., muscle; Max., maxillary; N., nerve; Nasolac., nasolacrimal; Orb., orbital; Pit., pituitary; Pteryg., pterygoid; Pterygopal., pterygopalatine; Sphen., Sphenoid.

6

Martins and Rhoton

A Semicirc. canals Trautman’s triangle

Temp. dura Lat. Rec. M. CN III CN II CN IV

Vestibule Cochlea Jug. bulb V1

Gr. Pet. N.

Inf. Rec. M.

Car. N.

V3

V2

Infraorb. N.

Vidian N. Car. A.

Pterygopal. fossa Max. sinus Eust. tube Max. A.

Vert. A. Int. Jug. V. Infratemp. fossa

B Front. lobe Temp. lobe Ped.

Lat. Rec. M.

CN V

V1 Sphen. sinus

Gr. Pet. N. Cochlea

V2 Vidian N.

CN IX-XI

V3

Pterygopal. fossa Zygoma Eust. tube Max. A.

Car. A. Infratemp. fossa Figure 2 Lateral view of the anterior, middle, and posterior cranial base. (A) The bone and structures lateral to the orbit, infratemporal, and pterygopalatine fossa, and the parapharyngeal space and petrous part of the temporal bone have been removed to expose the structures below the anterior, middle, and posterior cranial base. The orbit and maxillary sinus are located below the anterior cranial base. The infratemporal and pterygopalatine fossae and the parapharyngeal space are located below the middle cranial base, and the suboccipital area is located below the temporal and occipital bones. The first trigeminal division is related to the upper part of the orbit. The second trigeminal branch is related to the lower part of the orbit and maxilla. The mandibular nerve exits the cranium through the foramen ovale and enters the infratemporal fossa. The pterygoid and levator and tensor veli palatini muscles have been removed to expose the eustachian tube and its opening into the nasopharynx. The lateral part of the temporal bone has been removed to expose the cochlea, vestibule, and semicircular canals. The petrous carotid passes upward and turns medially below the cochlea. The sigmoid sinus turns downward under the semicircular canals and vestibule where the jugular bulb is located. The segment of the vertebral artery passing behind the atlanto-occipital joint is located below the posterior cranial base. (B) The dura has been opened to show the relationships of the frontal and temporal lobes and the cerebellum to the cranial base. The orbit is exposed below the frontal lobe. The pterygopalatine and infratemporal fossae and the temporal bone are located below the temporal lobe. The jugular bulb and internal jugular vein have been removed to show cranial nerves IX through XII exiting the jugular foramen (1). Abbreviations: A., artery; Car., carotid; CN, cranial nerve; Eust., eustachian; Front., frontal; Gr., greater; Inf., inferior; Infraorb., infraorbital; Infratemp., infratemporal; Int., internal; Jug., jugular; Lat., lateral; M., muscle; Max., maxillary; N., nerve; Ped., peduncle; Pet., petrosal; Pterygopal., pterygopalatine; Rec., rectus; Semicirc., semicircular; Sphen., sphenoid; Temp., temporal; V., vein; Vert., vertebral.


7

B

A

Crib. plate Front. bone

Eth. bone Sphen. ridge

Maxilla

Less. wing Vomer Optic canal Sphen. body For. lacerum For. ovale For. ovale Temp. bone Car. canal Temp. bone Jug. for. Pet. part

Gr. wing Ant. clin. For. spinosum Pet. Ridge

C

Palat. bone Horiz. plate Gr. wing Pteryg. proc. Mandib. fossa Clivus Occip. bone

D

Front. bone

Infraorb. for.

Supraorb. notch

Zygoma

Sup. Orb. Fiss. Maxilla Nasal bone Inf. Orb. Fiss.

Less. wing Gr. wing

Palat. bone Horiz. plate Vomer

Pteryg. proc. Mandib. fossa

Eth. Perp. plate Inf. concha

Vomer

Temp. bone Pet. part

Occip. bone

Jug. for.

Maxilla

Mandible

Figure 3 Osseous relationships of the anterior and middle cranial base. (A) On the endocranial surface, the anterior and middle cranial bases correspond to the anterior and middle fossae. The anterior part of the cranial base is separated from the middle fossa by the sphenoid ridge and the chiasmatic sulcus. The middle cranial base is separated from the posterior cranial base by the dorsum sellae and the petrous ridges. The upper surface of the anterior cranial base is formed by the frontal bone, which roofs the orbit; the ethmoid bone, which is interposed between the frontal bones and the site of the cribriform plate; and the lesser wing and anterior part of the body of the sphenoid, which forms the posterior part of the floor of the anterior fossa. The upper surface of the middle cranial base floor is formed by the greater sphenoid wing and posterior two-thirds of the sphenoid body anteriorly and the upper surface of the temporal bone posteriorly. The posterior part of the cranial base is formed by the temporal and occipital bones. The cribriform plate, sella, and clivus are located in the medial part of the cranial base. The lateral part of the cranial base is located above the orbits, pterygopalatine and infratemporal fossae, and the subtemporal and lateral part of the suboccipital areas. (B) Exocranial surface of the cranial base. This surface is more complicated than the endocranial surface. It is not demarcated into three well-defined fossae as is the endocranial surface. The exocranial surface is formed by the maxilla, zygomatic, palatine, sphenoid, temporal, and occipital bones, and the vomer. The maxilla, orbits, and nasal cavity are located below the anterior fossa. The anterior part of the hard palate is formed by the maxilla and the posterior part is formed by the palatine bone. The anterior part of the zygomatic arch is formed by the zygoma and the posterior part by the squamosal part of the temporal bone. The mandibular fossa on the lower surface of the temporal squama is located below the posterior part of the middle fossa. The vomer attaches to the lower part of the body of the sphenoid and forms the posterior part of the nasal septum. (C) Anterior view. The orbital rim is formed by the frontal bone, zygoma, and maxilla. The roof of the orbit is formed by the frontal and sphenoid bones; the lateral wall by the greater sphenoid wing and the zygomatic bone; the floor by the maxilla, except for a small part of the posterior floor formed by the palatine bone; and the medial wall of the orbit by the maxilla, lacrimal, and ethmoid bones. The nasal bone is interposed above the anterior nasal aperture between the maxillae. The nasal cavity is located between the ethmoid bones above and the maxillae and palatine bones, and sphenoid pterygoid process below. It is roofed by the frontal and ethmoid bones and the floor is formed by the maxillae and palatal bones. The osseous nasal septum is formed by the perpendicular ethmoid plate and the vomer. The inferior concha is a separate bone, and the middle and superior conchae are appendages of the ethmoid bone. The orbit opens through the superior orbital fissure into the middle fossa and through the inferior orbital fissure into the pterygopalatine and infratemporal fossae. (D) Anteroinferior view of the cranial base. The anterior part of the hard palate is formed by the maxillae and the posterior part is formed by the horizontal plate of the palatine bone. The vomer forms the posterior part of the nasal septum and divides the posterior nasal aperture in the midline. The infratemporal fossa is located below the greater sphenoid wing. The clivus is formed above by the body of the sphenoid bone and below by the basal part of the occipital bone. The petrous apex is interposed between the greater sphenoid wing and the clival part of the occipital bone. The mandibular condyles sit in the mandibular fossa located below the posterior part of the middle fossa on the inferior surface of the squamosal part of the temporal bone. (E) The cranial base is formed, in the lateral view, from anterior to posterior, by the maxilla and the frontal, zygomatic, sphenoid, temporal, and occipital bones. The zygomatic and frontal bones form the lateral part of the orbital rim. The pterion on the greater sphenoid wing marks the lateral end of the sphenoid ridge. The keyhole, a burr hole that exposes the dura of the anterior fossa and the periorbita in its depth, is located just above the frontozygomatic suture, behind the superior temporal line. The zygomatic arch is formed by the zygomatic bone and the squamosal part of the temporal bone. The condylar fossa, in which the mandibular condyle sits, is positioned above on the lower surface of the squamosal part of the temporal bone and posteriorly on the tympanic part of the temporal bone. The lower end of the pterygoid process unites with the posterior maxilla, but above, the process separates from the maxilla to create the pterygomaxillary fissure, which opens medially into the pterygopalatine fossa. (Continued).

8

Martins and Rhoton

E

F Keyhole Pterion

Temp. bone Squam. part Mandib. fossa Occip. bone

Zygoma Infraorb. for. Maxilla Pteryg. Proc. Mandib. cond. Max. sinus

Eth. perp. plate

Eth. Air Cells

Coronoid proc. Vomer Pterygopal. fossa

Pteryg. proc.

For. ovale

H

G Maxilla Front. bone

Supraorb. for.

Front. bone

Zygoma Eth. sinus Sup. orb. fiss. Less. sing

Eth. air cells Gr. wing Inf. orb. fiss. Vomer Vidian canal Pteryg. Proc. removed

Palat bone Gr. wing Horiz. plate Infratemp. fossa For. ovale Temp. bone For. spinosum

Clivus Occip. bone

For. rotundum Sphen. sinus Mandib. fossa

H

Figure 3 (Continued) (F) Inferior view of a cross section extending through the maxillae. The maxilla, which contains a large air-filled sinus, forms the anteromedial wall of the infratemporal fossa, the anterior wall of the pterygopalatine fossa, the lateral wall of the nasal cavity, the anterior portion of the hard palate, and much of the floor of the orbit. The pterygopalatine fossa is located between the pterygoid process and the posterior maxillary wall. The nasal septum is formed anteriorly and above by the perpendicular ethmoid plate and posteriorly and below by the vomer. (G) The right half of the maxilla and zygomatic arch has been removed. The inferior orbital fissure is located between the greater sphenoid wing and the maxilla. The right orbital roof and ethmoid air cells have been preserved. The right pterygoid process has been removed at its junction with the sphenoid body. The roof of the vidian canal, which extends through the base of the pterygoid process, has been preserved. (H) Anteroinferior view of the cranial base. The midline of the cranial base is formed, from anterior to posterior, by the frontal, ethmoid, sphenoid, and occipital bones. The roof of the orbit is formed by the frontal bone and lesser sphenoid wing. The ethmoidal sinuses are located anterior to the sphenoid sinus between the orbits. (I) Lateral view of the pterygomaxillary fissure. The pterygomaxillary fissure is located between the posterior maxillary wall and the pterygoid process. The pterygomaxillary fissure opens from the infratemporal fossa into the pterygopalatine fossa. The mandibular fossa is formed above by the squamosal part of the temporal bone and posteriorly by the tympanic part of the temporal bone, which also forms the anterior and lower wall of the external auditory meatus. (J) Anterior view through the maxillary sinus. The anterior and posterior walls of the maxillary sinus have been removed to expose the pterygoid process, which forms the posterior wall of the pterygopalatine fossa. The lower part of the superior orbital fissure is seen through the upper part of the maxillary sinus. The foramen rotundum opens into the pterygopalatine fossa and is separated from the superior orbital fissure by the maxillary strut. The vidian canal opens through the pterygoid process below and medial to the foramen rotundum. (K) Anterior view of a cranium sectioned through the posterior part of the ethmoid and maxillary sinuses. The ethmoidal sinuses are located anterior to the sphenoid body and sphenoid sinus. The part of the posterior wall of the maxilla forming the anterior wall of the pterygopalatine fossa has been preserved. The perpendicular plate of the palatine bone forms the medial wall of the pterygopalatine fossa. The ethmoidal sinus overlaps the lateral margin of the sphenoid ostia. The superior orbital fissure is located between the lesser and greater sphenoid wings and the sphenoid body. The infratemporal fossa is located below the greater wing of the sphenoid. The temporal fossa, which contains the temporalis muscle, is located between the greater wing and the zygomatic arch. (L) The posterior wall of the maxilla and ethmoidal sinuses have been removed to expose the sphenoid sinus and pterygopalatine fossa. The lateral wing of the sphenoid sinus extends laterally into the pterygoid process below the foramen rotundum. Septae divide the sphenoid sinus. The vidian canal opens through the base of the pterygoid process into the pterygopalatine fossa. (M) The osseous cross section has been extended posteriorly to just in front of the superior orbital fissure. The optic strut extends from the base of the anterior clinoid to the sphenoid body and separates the optic canal from the superior orbital fissure. The foramen rotundum is located below the medial part of the superior orbital fissure. The vidian canal opens into the pterygopalatine fossa below and medial to the foramen rotundum. (Continued).


Gr. wing

I

9

J

Temp. bone Squam. part Mandib. fossa

Inf. orb. fiss. Zygoma

Pterygomax. fiss. Infratemp. fossa

Sup. orb. fiss.

For. rotundum

Maxilla Vidian canal Pteryg. proc. Pteryg. proc.

K Sup. orb. fiss. Gr. wing

Less. wing Sphen. ostia

Max. sinus Post. wall Ant. wall

L Gr. wing

Sphen. septa Sphen. sinus For. rotundum

Eth. sinus Temp. bone Squam. part

Max. strut

Infratemp. fossa Palat. bone Perp. plate

Vidian canal

Pteryg. proc.

Max. sinus Post. wall

M

Gr. waing

Optic canal

N

Ant. clin. Less. wing Optic strut Sup. orb. fiss. Sup. orb. fiss. For. rotundum Gr. wing Vidian canal Pteryg. proc.

Zygoma

Ant. clin. Optic canal Optic strut Sphen. body Vidian canal Pteryg. proc. Maxilla Med. pteryg. plate

Lat. pteryg. plate Palat. bone Horiz. plate

Figure 3 (Continued) (N) Posterior view of the specimen in K showing the anterior part of the middle fossa from behind. The superior orbital fissure is positioned below the lesser sphenoid wing. The optic strut extends from the base of the anterior clinoid to the sphenoid body and separates the optic canal from the superior orbital fissure. The greater wing extends laterally to form part of the floor and anterior and lateral walls of the middle fossa. The medial and lateral pterygoid plates project backward from the pterygoid process. The horizontal plate of the palatine bone forms the posterior part of the hard palate. The posterior opening into the vidian canal is located above the medial pterygoid plate and extends forward through the pterygoid process at its junction with the sphenoid body (1). Abbreviations: Ant., anterior; Car., carotid; Clin., clinoid; Cond., condyle; Crib., cribriform; Eth., ethmoid; Fiss., fissure; For., foramen; Front., frontal; Gr., greater; Horiz., horizontal; Inf., inferior; Infraorb., infraorbital; Infratemp., infratemporal; Jug., jugular; Lat., lateral; Less., lesser; Mandib., mandibular; Max., maxillary; Med., medial; Orb., orbital; Occip., occipital; Palat., palatine; Perp., perpendicular; Pet., petrosal, petrous; Post., posterior; Proc., process; Pteryg., pterygoid; Pterygomax., pterygomaxillary; Pterygopal., pterygopalatine; Sphen., sphenoid; Squam., squamosal; Sup., superior; Supraorb., supraorbital; Temp., temporal.

10

Martins and Rhoton

A

B

Front. sius Front. bone Eth. sinus Front. N. CN IV Sphen. ridge

Crib. plate

Olf. bulb

Eth. bone Olf. Tr.

Planum Optic sheath

Sphen. bone

Less. wing MCA

ACA

D

C Front. N.

Trochlea Lac. N. CN IV

Sup. ophth. V. Crib. plate

Sup. Obl. M. Lev. M.

Olf. Tr.

Front. N. Nasocil. N. Sup. rec. M. Sup. ophth. V.

Eth. sinus

Sphen. sinus

Optic sheath MCA CN II Ant. Clin. removed Car. A.

Figure 4 Anterior fossa, orbit, and perinasal sinuses. (A) Superior view. The anterior cranial fossa is formed by the frontal, ethmoid, and sphenoid bones. The frontal bone splits anteriorly into two laminae, which enclose the frontal sinus. The ethmoid bones, which contain the ethmoid air cells and are the site of the crista galli and cribriform plate, are interposed between the frontal bones. Posteriorly, the frontal and ethmoid bones join the sphenoid bone, which encloses the sphenoid sinus and has the pituitary fossa on its upper surface. The olfactory bulbs and tracts have been preserved. (B) The roof of the right orbit has been removed to expose the periorbita. The right anterior clinoid process and roof of the optic canal have been removed to expose the optic nerve enclosed within the optic sheath as it passes through the optic canal to reach the orbital apex. (C) The frontal, trochlear, and lacrimal nerves can be seen through the periorbita. The trochlear nerve crosses above the orbital apex to reach the superior oblique muscle. (D) The orbital fat has been removed and the sphenoid sinus opened. The frontal branch of the ophthalmic nerve courses above the levator muscle. The ophthalmic artery, nasociliary nerve, and superior ophthalmic vein are located medially in the anterior part of the orbit and cross between the optic nerve and the superior rectus muscle and are thus situated on the lateral side of the optic nerve at the orbital apex. (E) Enlarged view. The superior oblique muscle has been retracted medially to expose the anterior and posterior ethmoidal branches of the ophthalmic artery and nasociliary nerve entering the anterior and posterior ethmoidal canal. The trochlea of the superior oblique muscle is attached to the superomedial margin of the orbit just behind the orbital rim. The frontal nerve divides into supraorbital and supratrochlear branches. (F) The levator and superior rectus muscle have been retracted posteriorly to expose the nasociliary nerve, ophthalmic artery, and superior ophthalmic vein passing above the optic nerve. (G) Superior view of the anterior fossa in another specimen. The nasal cavity, sphenoid sinus, and orbit have been unroofed. The dura has been removed from the roof and lateral wall of the cavernous sinus. The medial strip below the anterior cranial base is formed, from anterior to posterior, by the frontal, ethmoidal, and sphenoid sinuses. The orbital fat has been removed to expose the intraorbital structures. The frontal nerve courses above the levator muscle. The trochlear nerve passes above the annular tendon to reach the superior oblique muscle. The trochlea of the superior oblique muscle is attached in the superomedial part of the anterior orbit. The lacrimal nerve courses above the lateral rectus muscle. The ophthalmic artery and superior ophthalmic vein are seen in the interval between the levator and superior oblique muscle. The anterior and posterior ethmoidal branches of the ophthalmic artery course through the anterior and posterior ethmoidal canals. (Continued).


E

11

F

Supratroch. N.

Supraorb. N.

Trochlea Sup. Ophth. V. Ant. Eth. A. & N.

Ophth. A.

Nasocil. N.

Front. N. Sup. Obl. M.

Post. Eth. A. & N.

Med. Rec. M. Sup. Obl. M.

Sup. Rec. M. Ophth. A.

Lev. M.

Sup. Ophth. V. Nasocil. N.

Sup. Ophth. V. Lev. M.

Sup. Rec. M.

G

H

Nasocil. N.

Trochlea Lev. M. Sup. Obl. M. Lev. M. CN IV Sup. Rec. M. Sup. Rec. M.

Sup. Obl. M. Ophth. A. Ant. Eth. A.

Clin. Seg. Sup. Ophth. V.

CN II

CN III

Anular tendon

CN II Front. N. Lac. N. Ophth. A.

Post. Eth. A. CN IV Ophth. A.

Car. A.

Ophth. A.

V2 V1

V2 CN III Cav. Seg. CN IV

V1 V3

Figure 4 (Continued) (H) Enlarged view of cavernous sinus, superior orbital fissure, and orbital apex. The superior oblique, levator, and superior rectus muscles have been removed. The ophthalmic artery and nasociliary nerve enter the orbital apex on the lateral side of the optic nerve and cross between the optic nerve and superior rectus muscle to reach the medial part of the orbit. The optic nerve has been elevated to expose the ophthalmic artery, which courses through the optic canal on the lower side of the optic nerve and enters the orbital apex on the lateral side of the optic nerve. The ophthalmic artery then crosses medially between the optic nerve and the superior rectus muscle, as does the nasociliary nerve. The maxillary nerve exits the foramen rotundum to enter the pterygopalatine fossa, and the mandibular nerve exits the foramen ovale to enter the infratemporal fossa (1). Abbreviations: A., artery; ACA, anterior cerebral artery; Ant., anterior; Car., carotid; Cav., cavernous; Clin., clinoid; CN, cranial nerve; Crib., cribriform; Eth., ethmoid, ethmoidal; Front., frontal; Lac., lacrimal; Less., lesser; Lev., levator; M., muscle; Med., medial; MCA, middle cerebral artery; N., nerve; Nasocil., nasociliary; Obl., oblique; Olf., olfactory; Ophth., ophthalmic; Post., posterior; Rec., rectus; Seg., segment; Sphen., sphenoid; Sup., superior; Supraorb., supraorbital; Supratroch., supratrochlear; Tr., tract; V., Vein.

12

Martins and Rhoton

A

B Temp. M. Temp. M. Max. A.

CN II Ophth. A.

V2 Lat. Pteryg. M.

Gr. wing

V2 V1 CN III Cav. Seg. CN IV

CN III CN IV V3

Mandib. Cond.

V3

Pteryg. Plex.

CN VI

CN VI

Eust. Tube

Pet. Seg. Gr. Pet. N. CN VII Mast. antrum CN VIII CN IX-XI

CN VII CN VIII

C Lat. Pteryg. M. Mandib. Cond.

Max. A.

Ext. Ac. meatus V2

Chorda Tymp. N. Gr. Pet. N. Eust. tube Cochlea Semicirc. canal

V3 V1 Pet. Seg. CN III CN IV

Mast. antrum

D V1 V2 V3 Vidian N.

Petroling. Lig. Cav. Seg. CN VI CN IV CN III Petrosphen. Lig.

Chorda Tymp. N. Less. Pet. N. Deep Pet. N. Gr. Pet. N.

Pet. Seg. CN VII

CN VIII

Figure 5 Superior view of middle cranial base. (A) The floor of the middle fossa has been preserved. The anterior part of the floor of the middle fossa is formed by the greater sphenoid wing, which roofs the infratemporal fossa, and the posterior part of the floor is formed by the upper surface of the temporal bone. The internal acoustic meatus, mastoid antrum, and tympanic cavities have been unroofed. The dural roof and lateral wall of the cavernous sinus have been removed. The petrous segment of the internal carotid artery is exposed lateral to the trigeminal nerve. The temporalis muscle is exposed in the temporal fossa lateral to the greater sphenoid wing. (B) The floor of the middle fossa has been removed to show the relationship below the floor. The temporalis muscle descends medial to the zygomatic arch in the temporal fossa to insert on the coronoid process of the mandible. The infratemporal fossa is located medial to the temporal fossa, below the greater sphenoid wing, and contains the pterygoid muscles and venous plexus and branches of the mandibular nerve and maxillary artery. The mandibular condyle is located below the posterior part of the middle fossa floor, which is formed by the temporal bone. (C) Enlarged view of the posterior part of the area below the middle fossa floor. The roof of the temporal bone, which forms the posterior part of the floor of the middle fossa, has been opened to expose the mastoid antrum, eustachian tube, semicircular canals, cochlea, the nerves in the internal acoustic meatus, and the mandibular condyle. (D) The trigeminal nerve has been reflected forward. The abducens nerve passes below the petrosphenoid ligament and through Dorello’s canal. The petrous segment of the carotid passes below the petrolingual ligament to enter the cavernous sinus. The greater petrosal nerve is joined by the deep petrosal branch of the carotid sympathetic plexus to form the vidian nerve, which passes forward in the vidian canal, which has been unroofed. The lesser petrosal nerve arises from the tympanic branch of the glossopharyngeal nerve, which passes across the promontory in the tympanic nerve plexus and regroups to cross the floor of the middle fossa, exiting the cranium to provide parasympathetic innervation through the otic ganglion to the parotid gland. The tensor tympani muscle and eustachian are layered, with the former above the latter, along and separated from the anterior surface of the petrous carotid by a thin layer of bone (1). Abbreviations: A., artery; Ac., acoustic; Cav., cavernous; CN, cranial nerve; Cond., condyle; Eust., eustachian; Ext., external; Gr., greater; Lat., lateral; Less., lesser; Lig., ligament; M., muscle; Mandib., mandibular; Mast., mastoid; Max., maxillary; N., nerve; Ophth., ophthalmic; Petroling., petrolingual; Pet., petrosal, petrous; Petrosphen., petrosphenoid; Plex., plexus; Pteryg., pterygoid; Seg., segment; Semicirc., semicircular; Temp., temporalis; Tymp., tympani.


B

13

Crib. Plate

Nasolac. duct

A

Maxilla Max. sinus Palat. Bone

Inf. meatus

Eth. sinus

Vome r

Inf. Orb. Fiss. Vidian canal

Pteryg. Proc . Gr. Wing Gr. wing For. Oval e For. Lacerum Pterygopal. fossa Sulc. Tuba e

Pteryg. Proc. removed Car. canal

Vomer Palat. bone Perp. palate Pteryg. Proc.

Infratemp. fossa For. ovale

C

D

Max. A.

Max. sinus

Mid. concha Pteryg. Plex.

Infraorb. N. Temp. M.

Lat. Pteryg. M.

Max. sinus

Mandible Cond. Proc.

Pterygopal. fossa

Max. A.

Eust. Tube Tens. Vel. Pal. M. Lev. Vel. Pal. M.

Pteryg. Proc. Infratemp. fossa Mandible Max. A. Car. A. Parotid Gl. Int. Jug. V.

Car. Sheath Eust. tube Styloid Proc. Rosenmueller’s faossa Clivus Parotid Gl. Int. Jug. V.

Long. Cap. M.

Car. A.

CN IX-XII

Rec. Cap. Lat. M.

Figure 6 (A) Inferior view of cranial base. The right pterygoid process has been sectioned and removed at its junction with the greater wing and body of the sphenoid bone to expose the pterygopalatine fossa and the vidian canal. The vidian nerve, formed by the union of the superficial and deep petrosal nerves, courses in the vidian canal, which passes through the root of the pterygoid process. It opens posteriorly at the anterolateral margin of the foramen lacerum and anteriorly into the medial portion of the pterygopalatine fossa. The sulcus tubae, which is the attachment site of the cartilaginous part of the eustachian tube to the cranial base, is located on the extracranial surface of the sphenopetrosal fissure, anterolateral to the foramen lacerum and the carotid canal, and posteromedial to the foramina ovale and spinosum. The lateral part of the inferior orbital fissure opens into the infratemporal fossa, located below the greater sphenoid wing, and the medial part opens into the pterygopalatine fossa, located below the orbital apex between the maxilla and pterygoid process. The right zygomatic arch has been removed. (B) Inferior view of axial section of a cranium at the level of the maxillary sinus. The pterygopalatine fossa is located between the posterior wall of the maxillary sinus and the pterygoid process. The roof of the maxillary sinus forms the floor of the orbit. The infratemporal fossa is located below the greater wing of the sphenoid and opens medially into the pterygopalatine fossa. The medial wall of the pterygopalatine fossa is formed by the perpendicular plate of the palatine bone, which has an opening, the sphenopalatine foramen, through which branches of the maxillary artery and nerve reach the nasal cavity. The ethmoid air cells are located medial to the orbit. (Continued).

14

Martins and Rhoton

Max. sinus

E Max. A.

F

Max. sinus

Pterygopal. fossa Max. A.

Lat. Pteryg. M. Pteryg. Proc. Max. A. V3

Gr. wing

Eust. tube

Pterygopal. fossa Palat. bone Perp. plate Pteryg. Proc.

Infratemp. fossa Rosenmueller’s fossa Vomer

Mandible

V3

H

G Infraorb. N. Zygo. N.

Infraorb. N. V2

Pterygopal. Gang.

V2 V3

Lat. Pteryg. M.

Eust. tube

For. rotundum Vidian N. V3 For. ovale Eust. tube

Sphen. sinus Sella Pet. Seg. Clivus

Styloid Proc. Pet. Seg. CN VII For. lacerum Int. Jug. V.

CN IX-XII Sig. sinus

Figure 6 (Continued) (C) Inferior views of an axial section of the cranial base. The infratemporal fossa is surrounded by the maxillary sinus anteriorly, the mandible laterally, the pterygoid process anteromedially, and the parapharyngeal space posteromedially. It contains the mandibular nerve and maxillary artery and their branches, the medial and lateral pterygoid muscles, and the pterygoid venous plexus. The posterior nasopharyngeal wall is separated from the lower clivus by the longus capitis, and the nasopharyngeal roof rests against the upper clivus and floor of the sphenoid sinus. (D) Enlarged view with highlighting of the pre- (red) and poststyloid (yellow) compartments of the parapharyngeal space. The styloid diaphragm, formed by the anterior part of the carotid sheath, separates the parapharyngeal space into pre- and poststyloid parts. The prestyloid compartment, a narrow fat-containing space between the medial pterygoid and tensor veli palatini muscle, separates the infratemporal fossa from the medially located lateral nasopharyngeal region containing the tensor and levator veli palatini and the eustachian tube. The poststyloid compartment, located behind the prestyloid part, contains the internal carotid artery, internal jugular vein, and the cranial nerves IX through XII. (E) Some of the lateral pterygoid muscle has been removed to expose the branches of the mandibular nerve in the infratemporal fossa. The lower part of the pterygoid process has been removed to expose the maxillary artery in the pterygopalatine fossa. The pharyngeal recess (fossa of Rosenmüller) projects laterally from the posterolateral corner of the nasopharynx below the foramen lacerum. (F) Enlarged view. The pterygopalatine fossa is located between the posterior maxillary wall anteriorly, the sphenoid pterygoid process posteriorly, the perpendicular plate of the palatine bone medially, and the infratemporal fossa laterally. The medial part of the eustachian tube has been removed. (G) The pterygoid process has been removed to expose the maxillary nerve passing through the foramen rotundum to enter the pterygopalatine fossa where it gives rise to the infraorbital and zygomatic nerves and communicating rami to the pterygopalatine ganglion. The vidian nerve exits the vidian canal and joins the pterygopalatine ganglion. The terminal part of the petrous carotid is exposed above the foramen lacerum. (H) Enlarged view of the region of the carotid canal and jugular foramen. The bone below the carotid canal has been removed to expose the petrous carotid. The deep portion of the parotid gland has been removed to expose the facial nerve at the styloid foramen. The sigmoid sinus hooks downward from the posterior fossa and opens into the internal jugular vein. A portion of the occipital condyle has been removed to expose the hypoglossal nerve joining the nerves exiting the jugular foramen to pass downward in the carotid sheath. The styloid process and facial nerve at the stylomastoid foramen are located on the lateral side of the internal jugular vein. The right half of the floor of the sphenoid sinus has been removed to expose the sella (1). Abbreviations: A., artery; Cap., capitis; Car., carotid; CN, cranial nerve; Cond., condyle; Crib., cribriform; Eth., ethmoid; Eust., eustachian; Fiss., fissure; For., foramen; Gang., ganglion; Gl., gland; Gr., greater; Inf., inferior; Infraorb., infraorbital; Infratemp., infratemporal; Int., internal; Jug., jugular; Lat., lateral, lateralis; Lev., levator; Long., longus; M., muscle; .Max., maxillary; Mid., middle; N., nerve; Nasolac., nasolacrimal; Orb., orbital; Pal., palatini; Palat., palatine; Perp., perpendicular; Pet., petrosal, petrous; Plex., plexus; Proc., process; Pteryg., pterygoid; Pterygopal., pterygopalatine; Rec., rectus; Seg., segment; Sig., sigmoid; Sphen., sphenoid; Sulc., sulcus; Temp., temporalis; Tens., tensor; V., vein; Vel., veli; Zygo., Zygomatic.

the position of the superior semicircular canal. A thin lamina of bone, the tegmen tympani, roofs the area above the middle ear and auditory ossicles on the anterolateral side of the arcuate eminence. The internal auditory canal can be identified below the floor of the middle fossa by drilling along a line

approximately 60 degrees medial to the arcuate eminence, near the middle portion of the angle between the greater petrosal nerve and arcuate eminence (Fig. 5). The petrous apex, medial to the internal acoustic meatus, is free of important structures.


15

B

A

Front. N.

Olf. bulb Crib. plate

Sup. Ophth. V. Nasocil. N.

Lev. M. Front. N. Sup. Obl. M. Lac. N. CN IV

Ophth. A. Optic sheath CN IV Ant. Clin. removed

Optic canal

Ant. Clin. removed

Car. A.

CN II MCA ACA

C

D

Anular tendon Orb. apex

Max. Sinus Nasolac. duct Infraorb. N.

Optic Strut Falc. Lig.

Zygo. N.

Planum Optic sheath Clin. Seg.

Pit. Stalk

Orb. apex Optic sheath

Lam. Term.

Figure 7 Superior view of the anterior cranial base. (A) Both orbits have been unroofed to expose the periorbita. The optic canals have been unroofed and the anterior clinoids removed to expose the optic nerves, which are enclosed in the optic sheath within the optic canal. The frontal, trochlear, and lacrimal nerves can be seen through the periorbita. The roof of the ethmoidal sinuses and the olfactory bulbs sitting on the cribriform plate has been preserved. The anterior cerebral arteries course above the optic chiasm. (B) The intraorbital fat has been removed and the levator and superior rectus muscles have been retracted laterally to expose both globes, ophthalmic arteries, superior ophthalmic veins, and nasociliary nerves. (C) The orbital contents have been removed to expose the lateral wall and floor of the orbit. The maxillary sinuses are exposed below the orbital floors. The maxillary nerves give rise to the infraorbital nerve, which courses along the floor of the orbit to reach the cheek, and the zygomatic nerve, which courses along the lateral wall of the orbit to reach the malar eminence and temple. (D) Enlarged view. The optic nerves are enclosed within the optic sheath as they course through the optic canal. The annular tendon, from which the rectus muscles arise, surrounds the optic nerve and medial portion of the superior orbital fissure. Removal of the anterior clinoid exposes the clinoid segment of the carotid artery. The optic strut, which separates the optic canal and superior orbital fissure, has also been removed. The segment of anterior cerebral arteries passing above the chiasm has been removed to expose the lamina terminalis. The falciform dural fold extends across the optic nerve at the entrance into the optic canal (1). Abbreviations: A., artery; ACA, anterior cerebral artery; Ant., anterior; Car., carotid; Clin., clinoid; CN, cranial nerve; Crib., cribriform; Falc., falciform; Front., frontal; Infraorb., infraorbital; Lac., lacrimal; Lam., lamina; Lev., levator; Lig., ligament; M., muscle; Max., maxillary; MCA, middle cerebral artery; N., nerve; Nasocil., nasociliary; Nasolac., nasolacrimal; Obl., oblique; Olf., olfactory; Ophth., ophthalmic; Orb., orbital; Pit., pituitary; Seg., segment; Sup., superior; Term., terminalis; Zygo., zygomatic.

The middle cranial base can be divided into a lateral portion, containing the middle cranial fossa and the upper surface of the temporal bone, and a medial portion, the sellar and the parasellar region, where the pituitary gland and cavernous sinus are located (Figs. 3 and 8) (8). The basal temporal lobe, formed by the parahippocampal, occipitotemporal, infratemporal gyri, and uncus and supplied by branches of the anterior choroidal, posterior cerebral, and middle cerebral arteries, rests on the middle fossa floor. The cavernous sinus, situated between two layers of dura, is formed by an outer layer facing the brain, and inner or periosteum layer, covering the bone of the middle fossa (9). The inner layer splits into two parts when it reaches the cavernous sinus; one invests the

nerves and forms the inner layer of the lateral wall, and the medial layer faces the sphenoid body and forms the medial wall of the sinus. The same inner layer invests the oculomotor, trochlear, and ophthalmic nerves and the distal part of the abducens nerve in their course through the lateral wall of the cavernous sinus. The internal carotid artery with its vertical posterior bend, horizontal anterior bend, and clinoidal segments runs inside the cavernous sinus. The clinoidal segment of the internal carotid artery is between the distal and proximal dural rings and is covered by a layer of dura, which forms a collar, the carotid collar, around the artery (10). In a previous study, we found that the venous plexus, forming the cavernous sinus, extends through the lower

16

Martins and Rhoton

B

A

CN VI Crib. plate Sphenoeth. Rec. Sup. concha Mid. concha Inf. concha

Sup. Orb. Fiss. Sphen. sinus Sup. concha Int. Car. A. Mid. concha Eust. tube

V2 Cav. sinus

Inf. concha

D

C Front. sinus Crib. plate

Periorbita

Front. sinus ostium Max. sinus ostium

CN II Chiasm

Olf. bulb

Rosenmueller’s fossa

Infraorb. N. Max. Sinus Max. A.

Sup. Orb. Fiss. CN VI V2 Vidian N. Pterygopal. Fossa

F

E

CN VI Zygo. N. Infraorb. N.

Sup. Obl. M. Med. Rec. M.

Sphenopal. A. Sphenopal. Gang. Vidian N.

Anular tendon Max. A.

Inf. Rec. M. Gr. Palat. N.

Figure 8 Structures below the medial part of the anterior and middle cranial fossae. (A) Midsagittal section of the anterior and middle cranial base to the right of the nasal septum. The area below the medial part of the anterior cranial fossa is formed by the frontal and ethmoidal sinuses and the nasal cavity. The nasal cavity is divided into the inferior, middle, and superior meati and the sphenoethmoidal recess by the inferior, middle, and superior conchae. The inferior meatus is located below the inferior turbinate, and the sphenoethmoidal recess, into which the sphenoid sinus opens, is located above the superior turbinate. The central part of the middle cranial base is formed by the body of the sphenoid bone, which contains the sphenoid sinus and sella with the pituitary gland. The cribriform plate is located in the roof of the nasal cavity. The nasopharynx and the opening of the eustachian tube are located below the sphenoid sinus. (B) Some of the mucosa has been removed from the concha. The inferior concha is a separate bone attached to the maxilla. The middle and superior concha are appendages of the ethmoid bone. The carotid artery courses along the lateral margin of the sphenoid sinus. The prominence within the sphenoid sinus, formed by the superior orbital fissure, is located anterior to the intracavernous carotid, and the prominence overlying the maxillary nerve is located below the intracavernous carotid. (C) The middle and superior turbinates have been removed to expose the ostia of the maxillary and frontal sinuses. Both open into the middle meatus below the middle turbinate. The nasolacrimal duct opens below the inferior concha. Rosenmüller’s fossa is located behind the eustachian tube. (D) The medial wall of the maxillary sinus and the ethmoid air cells have been removed to expose the orbit. The optic nerve enters the orbit above the superior orbital fissure. The maxillary nerve exits the foramen rotundum to enter the pterygopalatine fossa. The vidian nerve passes through the vidian canal and enters the posterior margin of the sphenopalatine ganglion in the pterygopalatine fossa. The floor of the anterior cranial fossa forms much of the roof of the orbit and maxillary sinus forms most of the floor of the orbit. The abducens nerve is seen below the intracavernous segment of the internal carotid artery. The pterygopalatine fossa is located anterior to the sphenoid sinus and below the orbital apex. (E) The intraorbital fat has been removed to expose the superior oblique and medial and inferior rectus muscles. (F) Enlarged view of the pterygopalatine fossa. The maxillary nerve exits the foramen rotundum to enter the pterygopalatine fossa, where it gives rise to the infraorbital, zygomatic, and palatine nerves and communicating rami to the pterygopalatine ganglion. The vidian nerve exits the vidian canal to enter the pterygopalatine ganglion. The pterygopalatine fossa contains branches of the maxillary nerve, the junction of the vidian nerve with the pterygopalatine ganglion, and terminal branches of the maxillary artery (1). Abbreviations: A., artery; Car., carotid; Cav., cavernous; Crib., cribriform; CN, cranial nerve; Eust., eustachian; Fiss., fissure; Front., frontal; Gang., ganglion; Inf., inferior; Infraorb., infraorbital; Int., internal; M., muscle; Max., maxillary; Med., medial; Mid., middle; N., nerve; Obl., oblique; Olf., olfactory; Orb., orbital; Palat., palatine; Pterygopal., pterygopalatine; Rec., recess, rectus; Sphen., sphenoid; Sphenoeth., sphenoethmoidal; Sphenopal., sphenopalatine; Sup., superior; Zygo., zygomatic.


A

17

B Sup. Temp. A.

Orb. Oculi M. Orb. Oculi M. CN VII Frontotemp. Plex. Parotid duct Parotid gland Mass. M. Parotid duct

C

Front. M.

CN VII Frontotemp. Plex.

D

E Temp. M. Temp. M.

Orb. Oculi M.

CN VII to Front. M.

Sup. Temp. A.

Sup. Temp. A. Zygomatic M.

Orb. Oris M.

Zygo. arch Sup. Temp. A.

CN VII Brs.

TM joint

Parotid duct Mass. M. CN VII Brs. Parotid gland Bucc. M.

Zygoma Mandib. Cond.

CN VII

Coronoid Proc.

Mass. M. Bucc. M. Bucc. M.

Platysma M.

Figure 9 (A) The branches of the facial nerve, which form a fine plexus in the fat pad overlying the temporalis fascia and are directed to the orbicularis oculi and frontalis muscle, have been dissected free and a small piece of black material placed deep to their fine branches to highlight this neural network in the fat pad. (B) Enlarged view of the facial nerve plexus innervating the orbicularis oculi and frontalis muscle. (C) Lateral view of the structures superficial to the anterior and middle cranial base. The frontotemporal and zygomatic branches of the facial nerve are exposed anterior to the parotid gland. The orbicularis oculi surrounds the orbit, and the frontalis muscle extends upward from the superior orbital rim. The levators of the lip and zygomaticus muscles are located in front of the maxilla. The orbicularis oris surrounds the mouth and the buccinator muscle surrounds the oral cavity deep to the masseter muscle. The parotid duct crosses the masseter muscle. The superficial temporal artery divides into anterior and posterior branches. The parotid gland has been removed to show the branches of the facial nerve. (D) The parotid gland has been removed to expose the facial nerve exiting the stylomastoid foramen. The facial nerve branch to the frontalis muscle has been preserved in the dissection and has been laid back against the temporalis muscle to show it crossing the zygomatic arch in its course to the forehead. The superficial temporal artery passes deep to the facial nerve in front of the ear. (E) The masseter muscle has been removed to expose the temporalis muscle inserting on the coronoid process. The buccinator muscle, which surrounds the oral cavity, is situated on the deep side of the masseter muscle. (F) The coronoid process and lower part of the temporalis muscle have been removed to expose the deep temporal branches of both the maxillary artery and the mandibular nerve passing upward along the greater sphenoid wing and temporal squama to enter the deep side of the temporalis muscle. The lateral pterygoid muscles extend backward from the pterygoid process and greater wing of the sphenoid to insert along the mandibular condyle and temporomandibular joint. (G) A craniotomy has been done to expose the floor of the middle fossa, and the lateral wall of the orbit has been removed to expose the extraocular muscles. The mandibular condyle has been removed and the pterygoid muscles reflected to expose the mandibular nerve at the foramen ovale. The pterygopalatine fossa is located behind the maxilla. The floor of the orbit and the upper part of the maxilla has been removed to expose the nasal cavity. (Continued).

18

Martins and Rhoton

F

G

Temp. M.

Front. lobe

Sup. Temp. A. Deep Temp. A. & N. Temp. lobe Lat. Rec. M.

Lat. Pteryg. M. Inf. Obl. M. Mandib. Cond.

Infraorbital N. Pterygopal. fossa

Mid. fossa floor Mandib. fossa Max. A. V3

Maxilla Inf. concha

Med. Pteryg. M.

H

I V2 Orbit V1

Pet. Seg.

Tymp. Memb. Rec. Cap. Ant. M. Long. Cap. M.

V2 Chorda Tymp. N. Gr. Pet. N. V3

V3

CN VII

Car. A.

Vert. A.

Eust. tube

Int. Jug. V. CN XI

CN VII

J

K Sig. sinus

Sup. Pet. sinus Sup. Pet. Sinus V1 Car. canal

Sig. sinus Jug. Bulb

V2

CN V Post. root

V3 Pons CN IX-XII

Car. A.

Car. A. Vert. A.

Figure 9 (Continued) (H) Enlarged view following resection of the floor of the middle fossa and the external auditory canal to expose the tympanic membrane and the mandibular nerve below the foramen ovale. The mastoid segment of the facial nerve has been preserved. The greater petrosal nerve crosses above the petrous carotid. The tensor tympani muscle and eustachian tube are layered along the anterior margin of the petrous carotid. (I) The eustachian tube and tensor tympani have been resected to expose the upper cervical and petrous carotid. The nasopharyngeal mucosa has been opened to expose the longus capitus and rectus capitus anterior muscles. (J) The carotid artery has been reflected forward out of the carotid canal. This exposes the petrous apex in front of the jugular foramen on the medial side of the internal carotid artery. (K) The petrous apex has been drilled and the dura opened below the trigeminal nerve to expose the upper anterior part of the posterior cranial fossa. A segment of the internal jugular vein and jugular bulb has been resected to expose the IX through XII cranial nerves below the jugular foramen and hypoglossal canal (1). Abbreviations: A., artery; Ant., anterior; Brs., branches; Bucc., buccinator; Cap., capitis; Car., carotid; CN, cranial nerve; Cond., condyle; Eust., eustachian; Front., frontal, frontalis; Frontotemp., frontotemporal; Gr., greater; Inf., inferior; Int., internal; Jug., jugular; Lat., lateral; Long., longus; M., muscle; Mandib., mandibular; Mass., masseter; Max., maxillary; Med., medial; Memb., membrane; N., nerve; Obl., oblique; Orb., orbital; Pet., petrosal, petrous; Plex., plexus; Post., posterior; Proc., process; Pteryg., pterygoid; Pterygopal., pterygopalatine; Rec., rectus; Seg., segment; Sig., sigmoid; Sup., superior; Temp., temporal, temporalis; Tymp., tympani, tympanic; V., vein; TM, temporomandibular; Vert., vertebral; Zygo., zygomatic.


19

B

A

Ophth. A. Eth. sinus CN II

Orbit Med. Rec. M. Eth. sinus Sphen. ostia

Lat. Rec. M .

Septum Max. A.

Mid. concha Mid. meatus Mid. concha

Max. sinus

Inf. concha

Inf. meatus

Max. sinus

Inf. meatus

CN II Ophth. A.

D

C

Cav. Seg.

CN III CN II V1

Ophth. A. Cav. Seg.

CN II

Med. Rec. M.

CN VI

CN VI V2

Lat. Rec. M. Cav. sinus

Sphen. sinus

For. Rotundum

Inf. Rec. M. Pet. Seg. Infraorb. N.

Pet. Seg.

Pterygopal. fossa

Vidian N. Pterygopal. Gang. Max. A. Infratemp. fossa

Pterygopal. fossa Max. A.

Gr. Palat. N.

Max. sinus Pteryg. Proc. Eust. tube

Infratemp. fossa Lat. Pteryg. M.

Figure 10 (A) Anterior view of a coronal section, anterior to the sphenoid sinus, through the nasal cavity, orbits, and maxillary sinuses. The upper part of the nasal cavity is separated from the orbits by the ethmoidal sinuses. The lower part of the nasal cavity is bounded laterally by the maxillary sinuses. The middle concha projects medially from the lateral nasal wall at the junction of the roof of the maxillary and ethmoidal sinuses. The posterior ethmoid air cells are located in front of the lateral part of the sphenoid sinus. (B) The middle and inferior nasal conchae on the left side and the nasal septum and the posterior ethmoidal sinuses on both sides have been removed to expose the posterior nasopharyngeal wall, the anterior aspect of the sphenoid body, and the sphenoid ostia. The posterior ethmoid air cells overlap the lateral margin of the sphenoid ostia. (C) Enlarged view showing the relationships of the nasal cavity, pterygopalatine and infratemporal fossae, orbit, and sphenoid sinus. The nasopharynx is located below the sphenoid sinus. The pterygopalatine fossa is located in the lateral wall of the nasal cavity behind the upper part of the maxillary sinus and below the orbital apex. The posterior maxillary wall is so thin that the maxillary artery coursing in the pterygopalatine fossa can be seen through the bone. The sphenopalatine branch of the maxillary artery passes through the sphenopalatine foramen to reach the walls of the nasal cavity and the sphenoid face. (D) The posterior wall of the maxillary sinus has been removed to expose the pterygopalatine and infratemporal fossae and the internal carotid artery and nerves coursing through the cavernous sinus. The maxillary artery passes through the infratemporal fossa and enters the pterygopalatine fossa, where it gives rise to branches that follow the branches of the maxillary nerve. Some of these arteries course along the sphenoid face where careful hemostasis during transsphenoidal surgery reduces the need for nasal packing after transsphenoidal operations. The maxillary nerve exits the foramen rotundum to enter the pterygopalatine fossa where it gives rise to the infraorbital and greater palatine nerves and communicating rami to the pterygopalatine ganglion. The eustachian tube opens into the nasopharynx along the posterior edge of the medial pterygoid plate (1). Abbreviations: A., artery; Cav., cavernous; CN, cranial nerve; Eth., ethmoid; Eust., eustachian; For., foramen; Gang., ganglion; Gr., greater; Inf., inferior; Infraorb., infraorbital; Infratemp., infratemporal; Lat., lateral; M., muscle; Max., maxillary; Med., medial; Mid., middle; N., nerve; Ophth., ophthalmic; Palat., palatine; Pet., petrosal; Proc., process; Pteryg., pterygoid; Pterygopal., pterygopalatine; Rec., rectus; Seg., segment; Sphen., sphenoid.

20

Martins and Rhoton

A

B

Sphen. sinus

Sphen. sinus

Car. A. Cav. sinus

Infratemp. fossa Lat. Pteryg. M.

Clivus

Med. Pteryg. M.

V3 Max. A.

Nasopharynx Car. A. Palate

For. magnum

Mass. M. Oropharynx

D

C

Bas. A. Eust. tube V3

Lat. Pteryg. M.

V3 Car. A.

Vert. A. Alar Lig.

Max. A. Long. Cap. M. Car. A.

Atlas

Int. Jug. V. Car. A.

Mandib. Cond. Trans. Lig.

Odontoid

Styloid M. Vert. A. Long. Colli M.

E

F Eth. sinus

Infraorb. N. V2

Pterygopal. fossa

Infraorb. N.

V2 Pterygopal. Gang.

Sup. concha Comm. rami Max. A. Pterygopal. Gang.

Mid. concha

Gr. Palat. N. & A.

Inf. concha

Vidian N. Sphenopal. A.

Max. A.

Gr. Palat. N. & A.

Figure 11 Anterior view. Stepwise dissection of a cross section showing the relationships below the middle cranial base. (A) The soft palate, which has been preserved, is located at the level of the foramen magnum. The infratemporal fossa, located below the greater sphenoid wing and middle cranial fossa, contains the pterygoid muscles, maxillary artery, mandibular nerve branches, and the pterygoid venous plexus and opens posteriorly into the area around the carotid sheath as shown on the left side. (B) Enlarged view. The soft palate has been divided in the midline, and the leaves reflected laterally. The atlanto-occipital joints and the foramen magnum are located at approximately the level of the hard palate. The anterior arch of C1 and the dens are located behind the oropharynx and the clivus is located behind the nasopharynx and sphenoid sinus. The prominence over the longus capitus and the anterior arch of C1 are seen through the pharyngeal mucosa. (C) The mucosa lining the posterior pharyngeal wall has been reflected to the right, exposing the longus capitus, which attaches to the clivus, and the part of the longus colli that attaches to the anterior arch of C1. The left eustachian tube has been divided. (D) The clivus and anterior arch of C1 have been removed. The dura has been opened to expose the vertebral and basilar arteries. The dens has been preserved. The structures in the right infratemporal fossa and a segment of the right carotid artery and mandible have been removed to expose the right vertebral artery ascending between the C2 and C1 transverse processes. (E) Cross section through the ethmoidal and maxillary sinuses and the nasal cavity in front of the posterior maxillary wall. The posterior wall of the maxillary sinus has been removed to expose the pterygopalatine fossa and ganglia on both sides. The maxillary nerves enter the pterygopalatine fossa by passing through the foramen rotundum. The maxillary arteries enter the pterygopalatine fossa laterally by passing through the pterygomaxillary fissure and give rise to its terminal branches in the pterygopalatine fossa. Another branch enters the greater palatine canal with the greater palatine nerves. (F) Enlarged view of the pterygopalatine fossa. The vidian nerve exits the vidian canal to enter the pterygopalatine ganglion, which receives communicating rami from the maxillary nerve. The sphenopalatine branch passes through the sphenopalatine foramen to enter the lateral nasal cavity (1). Abbreviations: A., artery; Bas., basilar; Cap., capitis; Car., carotid; Cav., cavernous; Comm., communicating; Cond., condyle; Eth., ethmoid; Eust., eustachian; For., foramen; Gang., ganglion; Gr., greater; Inf., inferior; Infraorb., infraorbital; Infratemp., infratemporal; Int., internal; Jug., jugular; Lat., lateral; Lig., ligament; Long., longus; M., muscle; Mandib., mandibular; Mass., masseter; Max., maxillary; Med., medial; Mid., middle; N., nerve; Palat., palatine; Pteryg., pterygoid; Pterygopal., pterygopalatine; Sphen., sphenoid; Sphenopal., sphenopalatine; Sup., superior; Trans., transverse; V., vein; Vert., Vertebral.


A

21

Sphenoid bone

Temp. bone

Occip. bone

Int. Ac. meatus

Jug. For.

Occip. condyle Occip. bone

B

PCA CN III

SCA Bas. A.

CN IV CN V Tent. edge CN VI CN VII, VIII

AICA

CN IX, X

CN XII Vert. A.

P.I.C.A.

CN XI

Figure 12 (A) Superior view of the posterior cranial fossa. The osseous walls of the posterior fossa are formed by the occipital, temporal, and sphenoid bones. The fossa is bounded in front by the dorsum sellae and posterior part of the sphenoid bone and the clival part of the occipital bone; behind by the lower portion of the squamosal part of the occipital bone; and on each side by the petrous and mastoid parts of the temporal bone, and the lateral part of the occipital bone. One small part above the temporal bone is formed by the inferior angle of the parietal bone. (B) Nerves and arteries of the posterior fossa. Only two of the twelve pairs of cranial nerves course entirely outside the posterior fossa. The tentorium, which is attached along the petrous ridges, roofs the posterior fossa. The superior cerebellar artery courses below the oculomotor and trochlear nerves and above the trigeminal nerve; the anteroinferior cerebellar artery courses near the abducens, facial, and vestibulocochlear nerves; and the posteroinferior cerebellar artery courses near the glossopharyngeal, vagus, accessory, and hypoglossal nerves (12). Abbreviations: A., artery; Ac., acoustic; AICA, anteroinferior cerebellar artery; Bas., basilar; CN, cranial nerve; For., foramen; Int., internal; Jug., jugular; Occip., occipital; PCA, posterior cerebral artery; PICA, posteroinferior cerebellar artery; SCA, superior cerebellar artery; Temp., temporal; Tent., tentorial; Vert., vertebral.

22

Martins and Rhoton

A

B Zygomatic proc. Ant. Root Trig. Impress. Squamosal part Groove for Gr. Pet. N. Zygomatic Proc. Post. Root

Petrous part Squamosal part

Petrous part

Mandib. fossa

Arc. Emin. Car. canal

Tymp. Part Styloid proc.

Jug. fossa

Squamosal part Tegmen

Stylomast. For.

Mastoid part

Mastoid part

D

C

Vert. crest

Squamosal part

Facial canal Arc. Emin.

Trans. crest

Sup. Vest. area

Trig. Impress. Sig. sulcus Petrous part Int. Ac. Meatus

Cochlear area Mastoid part

Inf. Vest. area

Vest. Aqueduct

Figure 13 Temporal bone. (A) and (B) Inferior views of temporal bone. (A) The temporal bone has a squamosal part, which forms some of the floor and lateral wall of the middle cranial fossa. It is also the site of the mandibular fossa in which the mandibular condyle sits. The tympanic part forms the anterior, lower, and part of the posterior wall of the external canal, part of the wall of the tympanic cavity, the osseous portion of the eustachian tube, and the posterior wall of the mandibular fossa. The mastoid portion contains the mastoid air cells and mastoid antrum. The petrous part is the site of the auditory and vestibular labyrinth, the carotid canal, the internal acoustic meatus, and the facial canal. The petrous part also forms the anterior wall and the dome of the jugular fossa. The styloid part projects downward and serves as the site of attachment of three muscles. (B) Superior view. The medial part of the upper surface is the site of the trigeminal impression in which Meckel’s cave sits. Further laterally is the prominence of the arcuate eminence overlying the superior semicircular canal. Anterior and lateral to the arcuate eminences is the tegmen, a thin plate of bone overlying the mastoid antrum and epitympanic area. The temporal bone articulates anteriorly with the sphenoid bone, above with the parietal bone, and posteriorly with the occipital bone. The zygomatic process of the squamosal part has an anterior and a posterior root between which, on the lower surface, is located the mandibular condyle. (C) Posterior view of a right temporal bone. The sigmoid sulcus descends along the posterior surface of the mastoid portion. The internal acoustic meatus enters the central portion of the petrous part of the bone. The trigeminal impression and arcuate eminence are located on the upper surface of the petrous part. The vestibular aqueduct connects the vestibule in the petrous part with the endolymphatic sac, which sits on the posterior petrous surface inferolateral to the internal acoustic meatus. (D) Enlarged view of the right internal acoustic meatus. The transverse crest divides the meatal fundus into superior and inferior parts. The anterior part above the transverse crest is the site of the facial canal and the posterior part is the site of the superior vestibular area. Below the transverse crest the cochlear area is anterior and the inferior vestibular area is posterior. The vertical crest, also called “Bill’s Bar,” separates the facial and superior vestibular areas (7). Abbreviations: Ac., acoustic; Ant., anterior; Arc., arcuate; Car., carotid; Emin., eminence; For., foramen; Gr., greater; Impress., impression; Inf., inferior; Int., internal; Jug., jugular; Mandib., mandibular; N., nerve; Pet, petrosal; Post, posterior; Proc., process; Sig., sigmoid; Stylomast., stylomastoid; Sup., superior; Trans., transverse; Trig., trigeminal; Tymp., tympanic.; Vert., vertebral; Vest., vestibular.


23

A Basal Part (Clivus)

Car. canal

Styloid Proc.

B Clivus

Jug. For. Stylomast. For. Occip. Cond.

Condylar part

Petrocliv. Fiss. Jug. tubercle

Jug. For. Occip. Cond.

Digast. groove Sig. sulcus

Occip. A. groove

Vermian fossa Ext. Occip. crest Int. Occip. crest Sup. nuchal line Squamosal part Inion

Sup. sag. Sinus sulcus

Trans. sinus sulcus

D

C

For. ovale For. lacerum Clivus

Petrocliv. Fiss. Styloid Proc.

Pharyng. tubercle Jug. For.

Styloid Proc.

Stylomast. For.

Occip. Cond.

Car. canal

Clivus

Jug. For.

Condylar part

21 mm

Occip. Cond. 8 mm

Occip. bone Jug. Proc.

Squamosal Part

E

F For. Ovale Int. Ac. meatus

Petrocliv. Fiss.

Clivus

Coch. aqueduct Pet. part

Clivus

Vest. aqueduct

Petrocliv. Fiss. Car. canal

Coch. aqueduct Intrajug. Proc. Sig. Part

Hypogl. canal Occip. Cond.

Pet. part Sig. part Occip. bone Jug. Proc.

Figure 14 Occipital bone, foramen magnum, and jugular foramen. (A–D) Occipital bone and foramen magnum. (A) Inferior view. (B) Superior view. (C, D) Anteroinferior views. (A–C) The occipital bone surrounds the oval shaped foramen magnum, which is wider posteriorly than anteriorly. The narrower anterior part sits about the odontoid process and is encroached on from laterally by the occipital condyles. The wider posterior part transmits the medulla. The occipital bone is divided into a squamosal part located above and behind the foramen magnum; a basal (clival) part situated in front of the foramen magnum; and paired condylar parts located lateral to the foramen magnum. The basilar part of the occipital bone, which is also referred to as the clivus, is a thick quadrangular plate of bone, concave from side to side, that extends forward and upward to join the sphenoid bone just below the dorsum sellae. The clivus is separated on each side from the petrous part of the temporal bone by the petroclival fissure, which ends posteriorly at the jugular foramen. The condylar parts of the occipital bone, on which the occipital condyles are located, are situated lateral to the foramen magnum on the external surface. The hypoglossal canal is situated above the condyle. The jugular process of the occipital bone extends laterally from the posterior half of the condyle and articulates with the jugular surface of the temporal bone. The sulcus of the sigmoid sinus crosses the superior surface of the jugular process. The jugular foramen is bordered posteriorly by the jugular process of the occipital bone and anteriorly by the jugular fossa of the petrous temporal bone. (Continued).

24

Martins and Rhoton

ring, inside the collar of dura, and around the clinoid segment to the level of the upper ring. The meningohypophyseal trunk, with its tentorial, inferior hypophyseal, and dorsal meningeal branches, and the inferolateral trunk, also called the artery of the inferior cavernous sinus, arise from the intracavernous carotid artery. The proximal abducens nerve passes through Dorello’s canal, located below the petrosphenoid ligament, and receives sympathetic branches from the internal carotid nerve, which pass to the ophthalmic nerve to enter the orbit. The main venous afferents to the cavernous sinus are the superior and inferior ophthalmic veins and the sphenoparietal sinus (Figs. 4 and 7). Several venous compartments, named according to their relationship to the cavernous carotid artery, empty mainly in the basilar and superior and inferior petrosal sinuses, or, by way of the foramina in the middle fossa floor, into the pterygoid venous plexus (4). The sella houses the pituitary gland and is partially closed above by the diaphragma sellae. Anterolateral to the diaphragm, the carotid cave, a dural depression at the level of the distal dural ring, extends downward medial to the initial intradural segment of the internal carotid artery. The tensor tympani muscle and eustachian tube cross medial to the foramen spinosum, below the floor of the middle fossa, and anterior to the horizontal segment of the petrous carotid (Fig. 5). The greater petrosal nerve crosses the area above and parallel to the petrous carotid artery, laterally joins the geniculate ganglion, and medially joins the deep petrosal branch of the carotid sympathetic nerves to form the vidian nerve, which enters the pterygopalatine ganglion (Figs. 2, 5, and 6). The lesser petrosal nerve runs anterior to the greater petrosal nerve and exits the cranium, passing through the foramen spinosum to join the otic ganglion. The cochlea is situated below the floor of the middle cranial fossa, at the apex of the angle between the greater petrosal and labyrinthine segment of the facial nerve.

Exocranial Surface The exocranial surface of the middle cranial base is also divided into central and lateral parts (Figs. 2, 3, 6, and 9) (1). The central part encompasses the sphenoid body and the upper part of the basal (clival) part of the occipital bone and corresponds to the sphenoid sinus and the nasopharynx. The lateral part is formed by the greater sphenoid wing; the petrous, tympanic, and squamous parts of the temporal bone; the styloid process; and the zygomatic, palatine, and maxillary bones. The medial and lateral parts are separated by a parasagittal plane passing through the medial pterygoid plate. The foramen lacerum is located at the union of the sphenoid, occipital, and petrous bones and is enclosed on its lower side by fibrocartilaginous tissue to form the inferior wall of the carotid canal. Structures transversing the lateral part include the carotid artery in the carotid canal, the glossopharyngeal, vagus, and accessory nerves in the jugular foramen, the third trigeminal division in the foramen ovale, the middle meningeal artery in the foramen spinosum, and the facial nerve in the facial canal. The pterygomaxillary fissure is the lateral opening of the pterygopalatine fossa into the infratemporal fossa. The glenoid fossa harbors the mandibular condyle. The roof of the fossa is divided into anterior and posterior parts by the squamotympanic fissure, along which the chorda tympani passes. The area below the middle cranial base includes the infratemporal fossa, parapharyngeal space, infrapetrosal space, and pterygopalatine fossa (Figs. 6, 9, 10, and 11). The boundaries of the infratemporal fossa are the middle pterygoid muscle and the pterygoid process medially, the mandible laterally, the posterior wall of the maxillary sinus anteriorly, the greater wing of the sphenoid superiorly, and the medial pterygoid muscle joining the mandible and the pterygoid fascia posteriorly. The fossa opens into the neck below. The infratemporal fossa contains the branches of mandibular nerve, the maxillary artery, and the pterygoid muscles and venous plexus. The mandibular nerve, after exiting the

←−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−− Figure 14 (Continued) The jugular tubercle lies on the internal surface above the hypoglossal canal. The squamous part is internally concave. The convex external surface of the squamosal part has several prominences. The largest prominence, the external occipital protuberance (inion), is situated at the central part of the external surface. The superior nuchal line radiates laterally from the protuberance. A vertical ridge, the external occipital crest, descends from the external occipital protuberance to the midpoint of the posterior margin of the foramen magnum. The internal surface of the squamous part has a prominence, the internal occipital protuberance near its center. The internal occipital crest bifurcates above the foramen magnum to form a V-shaped ridge between the limbs of which is the vermian fossa. (D) Inferior view of the occipital condyles and foramen magnum. The occipital condyles are located along the lateral margin of the anterior half of the foramen magnum. Their articular surfaces are convex, face downward and laterally, and articulate with the superior facet of C1. A probe inserted through the hypoglossal canal passes forward approximately 45 degrees from the midsagittal plane in an anterolateral direction. The hypoglossal canal is located above the middle third of the occipital condyle and is directed from posterior to anterior and from medial to lateral. The intracranial end of the hypoglossal canal is located approximately 5 mm above the junction of the posterior and middle third of the occipital condyle and approximately 8 mm from the posterior edge of the condyle. The extracranial end of the canal is located approximately 5 mm above the junction of the anterior and middle third of the condyle. The average length of the longest axis of the condyle is 21 mm. The large arrow shows the direction of the transcondylar approach and the cross-hatched area shows the portion of the occipital condyle that can be removed without exposing the hypoglossal nerve in the hypoglossal canal. The stylomastoid foramen is situated lateral to the jugular foramen. The styloid process is located anterior and slightly medial to the stylomastoid foramen. (E, F) Jugular foramen. (E) Posterosuperior view. The jugular foramen is located between the temporal and occipital bones. The sigmoid groove descends along the mastoid and crosses the occipitomastoid suture where it turns forward on the upper surface of the jugular process of the occipital bone and enters the foramen by passing under the posterior part of the petrous temporal bone. The foramen has a larger lateral sigmoid part through which the sigmoid sinus empties and a smaller anteromedial petrosal part through which the inferior petrosal sinus empties. The two parts are separated by the intrajugular processes of the occipital and temporal bones. The glossopharyngeal, vagus, and accessory nerves pass through the intrajugular portion of the foramen located between the sigmoid and petrosal parts. The foramen is asymmetric from side to side, with the right side often being larger as shown. The cochlear aqueduct opens just above the anterior edge of the petrosal part. The vestibular aqueduct opens into the endolymphatic sac, which sits on the back of the temporal bone superolateral to the sigmoid part of the jugular foramen. (F) Anteroinferior view. The roof over the jugular foramen, formed by the jugular fossa of the temporal bone, is shaped to accommodate the jugular bulb. The posterior margin of the foramen is formed by the jugular process of the occipital bone, which connects the basal (clival) part of the occipital bone to the squamosal part. The petroclival fissure intersects the anteromedial margin of the petrosal part of the foramen. The entrance into the carotid canal is located directly in front of the medial half of the jugular foramen. (A–C) (13), (D) (14), (E–G) (18). Abbreviations: A., artery; Ac., acoustic; Car., carotid; Coch., Cochlear; Cond., condyle; Digast., digastric; Ext., external; Fiss., fissure; For., foramen; Hypogl., hypoglossal; Int., internal; Intrajug., intrajugular; Jug., jugular; Occip., occipital; Petrocliv., petroclival; Pet., petrous; Pharyng., pharyngeal; Proc., process; Sag., sagittal; Sig., sigmoid; Stylomast., stylomastoid; Sup., superior; Trans., transverse; Vest., vestibular.


A

25

B Sup. Pet. V. Subarc. A. CN VII

CN V

CN VIII

Sup. Vest. N. Nerv. Intermed. Inf. Vest. Coch. N.

Labyr. A. A.I.C.A.

Flocc.

Flocc. CN VI CN IX CN IX PICA

PICA CN X

CN X-XI

Chor. Plex.

CN XI

D

C

CN V CN V

CN VII Sup. Vest. N. Nerv. Intermed.

Coch. N.

Sup. Vest. N.

Inf. Vest. N. AICA

Subarc. A. CN VII Inf. Vest. N. Coch. N. Labyr. A.

Flocc.

Inf. Vest. N. Flocc.

CN VI

CN IX

CN VI

CN IX

Chor. Plex. PICA

AICA

Chor. Plex. PICA

Figure 15 Retrosigmoid exposure of the nerves in the right cerebellopontine angle. (A) The vestibulocochlear nerve enters the internal acoustic meatus with a labyrinthine branch of the AICA. The PICA courses around the glossopharyngeal, vagus, and accessory nerves. The abducens nerve ascends in front of the pons. The subarcuate branch of the AICA enters the subarcuate fossa superolateral to the porus of the meatus. Choroid plexus protrudes into the cerebellopontine angle behind the glossopharyngeal and vagus nerves. (B) The posterior wall of the internal acoustic meatus has been removed. The cleavage plane between the upper bundle, formed by the superior vestibular nerve, and the lower bundle, formed by the inferior vestibular and cochlear nerves, was begun laterally where the nerves normally separate near the meatal fundus and extended medially. The nervus intermedius arises on the anterior surface of the vestibulocochlear nerve, has a free segment in the cistern and/or meatus, and joins the facial nerve distally. The facial nerve is located anterior to the superior vestibular nerve and the cochlear nerve is anterior to the inferior vestibular nerve. (C) The cleavage plane between the cochlear and inferior vestibular nerves, which is well developed in the lateral end of the internal acoustic meatus, has been extended medially. Within the cerebellopontine angle, the superior vestibular nerve is posterior and superior, the facial nerve anterior and superior, the inferior vestibular nerve posterior and inferior, and the cochlear nerve anterior and inferior. (D) The superior and inferior vestibular nerves have been divided to expose the facial and cochlear nerves. A labyrinthine branch of the AICA enters the internal meatus (15). Abbreviations: A., artery; AICA, anteroinferior cerebellar artery; Chor. Plex., choroid plexus; CN, cranial nerve; Coch., cochlear; Flocc., flocculus; Inf., inferior; Intermed., intermedius; Labyr., labyrinth; N., nerve; Nerv., nervus; Pet., petrosal; PICA, posteroinferior cerebellar artery; Subarc., subarcuate; Sup., superior; V., vein; Vest., vestibular.

26

Martins and Rhoton

B

A Pon. Mes. Sulcus

CN VI Pon. Med. Sulcus CN IX, X, XI Olive Medulla

CN VII CN VII, VIII CN VIII Chor. Plex. Pyramids AICA Pet. Surface CN IX, X CN XII CN XII CN XI PICA Vert. A. PCA CN IV SCA

PCA CN III SCA

D

AICA

PICA

CN XII

Ant. Hem. V.

Ant. Sp. A.

PICA Vert. A.

Vert. A.

CN V CN VI AICA

PICA

Ped. V.

CN VI Trans. Pon. V. AICA Sup. Pet. V. CN IX -XI

CN VII, VIII

CN III SCA

CN V

Pons

C

CN V

CN IV

Midbrain

Med. Ant. Pon. Mes. V. V. Cer. Pon. Fiss.

Ant. Hem. V. V. Pon. Med. sulcus Trans. Med. V.

Med. Ant. Med. V.

Figure 16 Brain stem, anterior cerebellar surface, and posterior skull base. (A) The petrosal (anterior) surface of the cerebellum, called the petrosal surface, and front of the brain stem face the endocranial surface of the posterior fossa. The fourth ventricle is positioned behind the pons and medulla. The midbrain and pons are separated by the pontomesencephalic sulcus and the pons and medulla by the pontomedullary sulcus. The trigeminal nerves arise from the midpons. The abducens nerve arises in the medial part of the pontomedullary sulcus, rostral to the medullary pyramids. The facial and vestibulocochlear nerves arise at the lateral end of the pontomedullary sulcus immediately rostral to the foramen of Luschka. The hypoglossal nerves arise anterior to the olives and the glossopharyngeal, vagus, and accessory nerves arise posterior to the olives. Choroid plexus protrudes from the foramen of Luschka behind the glossopharyngeal and vagus nerves. (B) Anterior view of the brain stem with the arteries preserved. (C) Posterior view of the skull base with the cranial nerves and arteries preserved. (B, C) The SCA arises at the midbrain level and encircles the brain stem near the pontomesencephalic junction. The SCA courses below the oculomotor and trochlear nerves and above the trigeminal nerve. The SCA loops down closer to the trigeminal nerve in C than in B. The AICA arises at the pontine level and courses by the abducens, facial, and vestibulocochlear nerves. In B, both AICAs pass below the abducens nerves. In C, the left abducens nerve passes in front of the AICA and the right abducens nerve passes behind the AICA. The PICAs arise from the vertebral artery at the medullary level and course in relation to the glossopharyngeal, vagus, accessory, and hypoglossal nerves. The origin of the SCAs is quite symmetrical from side to side. There is slight asymmetry in the level of origin of the AICAs and marked asymmetry in the level of the origin of the PICAs, especially in B. (D) The veins on the anterior surface of the pons and medulla and the petrosal cerebellar surface drain predominantly into the superior petrosal veins which empty into the superior petrosal sinuses. The median anterior pontomesencephalic and median anterior medullary veins ascend on the front of the brain stem. The transverse pontine and transverse medullary veins run transversely across the pons and medulla surfaces. The anterior hemispheric veins drain the petrosal cerebellar surface and commonly empty into the vein of the cerebellopontine fissure, which ascends to join the superior petrosal veins. The vein of the pontomedullary sulcus passes across the pontomedullary junction. The peduncular vein crosses the cerebral peduncle. Abbreviations: A., artery; AICA, anteroinferior cerebellar artery; Ant., anterior; Cer. Pon., cerebellopontine; Chor., choroid; CN, cranial nerve; Fiss., fissure; Hem., hemispheric; Med., medial, medullary; Pet., petrosal; PCA, posterior cerebral artery; Ped., peduncular; PICA, posteroinferior cerebellar artery; Plex., plexus; Pon., pontine; Pon. Med., pontomedullary; Pon. Mes., pontomesencephalic; SCA, superior cerebellar artery; Sulc., sulcus; Sp., spinal; Sup., superior; Trans., transverse; V., vein; Vert., vertebral.

foramen ovale, lies anterolateral to the otic ganglion and divides immediately into its terminal branches: the pterygoid, buccal, masseteric, and temporal branches along the superior wall of the fossa; the inferior alveolar and the lingual branches, after being joined by the chorda tympani, descend between both pterygoid muscles; and the auriculotemporal branch with the maxillary artery course between the mandible and the sphenomandibular ligament. The auriculotemporal nerve carries the parasympathetic innervation of

the parotid gland, which travels through the tympanic branch of the glossopharyngeal nerve, which in turn forms the lesser petrosal nerve, to reach the otic ganglion before joining the auriculotemporal nerve. The maxillary artery, which arises as a terminal branch of the external carotid artery with the superficial temporal artery, is divided into three segments. The first, or mandibular segment, passes between the sphenomandibular ligament and the mandibular neck and gives rise to the deep auricular, anterior tympanic, middle meningeal,


A

27

B Post. Digast. M. Rec. Cap. Post. Maj. M.

Sup. Obl. M. Suboccip. triangle

Trans. Proc. C1

Inf. Obl. M.

X

Lev. Scap. M.

C

C

CN IX, X

D

th P.I.C.A. 4 Vent.

PICA

CN XI CN XI

Occip. Cond. Atl. Occip. Joint

CN XII C1 Cond. CN XII Vert. A. Dural cuff

Dent. Lig.

C1 Cond. Vert. A. C1 Trans. Proc. C1

Hypogl. canal

Vert. A. Vert. A.

Figure 17 (A–D) Far-lateral and transcondylar approach. (A) A suboccipital scalp flap is commonly selected for the far-lateral exposure. The medial limb extends downward in the midline so that a wide upper cervical laminectomy can be completed if needed. The lateral limb extends below the level of the C1 transverse process (×), which can be palpated between the mastoid tip and the angle of the jaw, to access the vertebral artery as it ascends through the C1 transverse process. The muscles superficial to the suboccipital triangle can be reflected from the suboccipital area in a single layer with the scalp flap, leaving a cuff of suboccipital muscle and fascia attached along the superior nuchal line to aid in closure. (B) The scalp and muscles are reflected in a single layer to expose the suboccipital triangle in the depths of which the vertebral artery courses behind the atlanto-occipital joint and across the posterior arch of C1. The triangle is located between the superior and inferior oblique and the rectus capitis posterior major. (C) A suboccipital craniectomy has been completed, the posterior arch of C1 has been removed, the posterior root of the transverse foramen of the C1 has been removed, the area above the occipital condyle has been drilled to expose the hypoglossal canal, and the dura has been opened. The dural incision completely encircles the vertebral artery, leaving a narrow dural cuff on the artery, so that the artery can be mobilized. The drilling in the supracondylar area can be extended extradurally to the level of the jugular tubercle to increase access to the front of the brain stem. (D) Comparison of the exposure with the far-lateral and transcondylar approaches. On the right side, the far-lateral exposure has been extended to the posterior margins of the atlantal and occipital condyles and the atlanto-occipital joint. The prominence of the condyles limits the exposure along the anterolateral margin of the foramen magnum. On the left side, a transcondylar exposure has been completed by removing the posterior part of the condyles. The dura can be reflected further laterally with the transcondylar approach than with the far-lateral approach. The condylar drilling provides an increased angle of view of the clivus and front of the brain stem. The dentate ligament and accessory nerve ascend from the region of the foramen magnum (18). Abbreviations: A., artery; Atl., atlanto; Cap., capitis; CN, cranial nerve; Cond., condyle; Dent., dentate; Digast., digastric; Hypogl., hypoglossal; Inf., inferior; Lev., levator; Lig., ligament; M., muscle; Maj., major; Obl., oblique; Occip., occipital; PICA, posteroinferior cerebellar artery; Post., posterior; Proc., process; Rec., rectus; Scap., scapula; Suboccip., suboccipital; Sup., superior; Trans., transverse; Vent., ventricle; Vert., vertebral.

accessory middle meningeal (enters through the foramen ovale), and the inferior alveolar artery. The second, or pterygoid segment, courses through the middle of the infratemporal fossa and gives rise to the posterosuperior alveolar, infraorbital, masseteric, pterygoid, temporal, and buccal branches. The third, or pterygopalatine segment, courses in

the fossa of the same name. The pterygoid venous plexus connects through the middle fossa foramina and inferior orbital fissure with the cavernous sinus and empties into the retromandibular and facial veins. The pterygopalatine fossa is located between the maxillary sinus in the front, the pterygoid process behind, the

28

Martins and Rhoton

Ant. tubercle Ant. arch

A

B

Ant.tubercle

Ant. arch

Trans. proc.

Sup. Art. Facet Trans. Proc.

Inf. Art. Facet

Lat. mass. Trans. For. Trans. For. Post. arch

Groove for Vert. A.

Post. arch

Post. tubercle

C

Post. tubercle

D

Sup. Art. Facet

Sup. Art. Facet

Trans. Proc. Groove for Vert. A. Trans. Proc.

Post. arch

Ant. arch

Ant. tubercle

Inf. Art. Facet

E

Odontoid proc.

Post. tubercle

Inf. Art. Facet

F Odontoid proc.

Art. facet Sup. Art. Facet

Spinous proc.

Sup. Art. Facet

Sup. Art. Facet Trans. For.

Trans. proc.

Lamina

Trans. for.

Trans. proc. Body

Body Trans. proc.

Inf. Art. Facet Inf. Art. Facet

Inf. Art. Facet

G

Odontoid proc. Sup. Art. Facet Trans. proc.

Sup. Art. Facet Trans. proc.

Body

H Trans. for Trans. ror.

Body

Trans. proc.

Trans. for. Inf. Art. Facet

Pedicle

Inf. Art. Facet

Inf. Art. Facet Inf. Art. Facet Lamina

Lamina

Spinous proc. Spinous proc.

Figure 18 Atlas and Axis. (A–D) The atlas. (A) Superior view. (B) Inferior view. (C) Anterior view. (D) Posterior view. The atlas consists of two thick lateral masses situated at the anteromedial part of the ring, and which are connected in front by a short anterior arch and posteriorly by a longer curved posterior arch. The anterior and posterior tubercles are at the anterior and posterior midline. The superior articular facet is an oval, concave facet that faces upward and medially to articulate with the occipital condyle. The inferior articular facet is a circular, flat, or slightly concave facet that faces downward, medially, and slightly backward and articulates with the superior articular facet of the axis. The medial aspect of each lateral mass has a small tubercle for the attachment of the transverse ligament of the atlas. The transverse process projects from the lateral masses. The transverse foramina transmit the vertebral arteries. The upper surface of the posterior arch adjacent to the lateral masses has paired grooves in which the vertebral arteries course. (E–H) The axis. (E) Anterior view. (F) Lateral view. (G) Superior view. (H) Inferior view. The axis is distinguished by the odontoid process (dens). On the front of the dens is an articular facet that forms a joint with the facet on the back of the anterior arch of the atlas. The dens is grooved at the base of its posterior surface where the transverse ligament of the atlas passes. The oval superior articular facets articulate with the inferior facets of the atlas. The superior facets are anterior to the inferior facets. The pedicles and laminae are thicker than on the other cervical vertebra and the lamina fuses behind to form a large spinous process. The transverse foramina are directed superolaterally, thus permitting the lateral deviation of the vertebral arteries as they pass up to the more widely separated transverse foramina in the atlas. The inferior articular facets face downward and forward (13). Abbreviations: A., artery; Ant., anterior; Art., articular; For., foramen; Inf., inferior; Lat., lateral; Mass., masses; Post., posterior; Proc., process; Sup., superior; Trans., transverse; Vert., vertebral.


29

Nasal bone B

A Zygoma

Nasal bone

Maxilla Med. Canthal Lig.

Front. bone

Nasolac. duct Lac. bone Periorbita

D

C Maxilla

Nasal cavity Med. Canthal Lig.

Nasolac. duct

Sup. concha

Lac. bone Nasal bone Eth. Perp. plate Ant. Eth. A. Eth. sinus Crib. plate

Frontonasal suture

Periorbita

Dura

E

F Sphenopal. A. Clivus

Vomer Ant. Eth. A.

Septa Car. A.

Sphen. sinus

Sphen. ostia Sphen. sinus Crib. plate Crib. plate

Figure 19 (A–F). Relationships in the transbasal and extended frontal approaches. (A) The inset shows the bicoronal scalp incision. A large bifrontal craniotomy and a fronto-orbitozygomatic osteotomy have been completed. The osteotomized segment may extend through the nasal bone and from one to the other lateral orbital rims, as shown. However, for most lesions, a more limited bone flap and osteotomy will suffice and can be tailored as needed to deal with the involvement of the cranial base, nasal cavity, paranasal sinuses, or orbit. For an orbital lesion, an orbitofrontal craniotomy, elevating only the superior orbital rim (yellow arrows) and orbital roof, is all that is needed. For a cavernous sinus or unilateral lesions of the anterior or middle fossa, an orbitozygomatic osteotomy will usually suffice (blue arrow). For a clival lesion, a more limited bifrontal approach (red arrow) will suffice. (B) The periorbita has been separated from the walls of the orbit in preparation for the osteotomies. Division of the medial canthal ligament is not necessary for most lesions, but may be required for lesions extending into the lower nasal cavity or orbit. The ligaments should be re-approximated at the end of the operation. (C) The right medial canthal ligament has been divided and the orbital contents retracted laterally to expose the nasolacrimal duct and the anterior ethmoidal branch of the ophthalmic artery at the anterior ethmoidal foramen. (D) The osteotomies have been completed and the frontal dura elevated. The dura remains attached at the cribriform plate. The upper part of both orbits are exposed. (E) An osteotomy around the cribriform plate leaves it attached to the dura and olfactory bulbs, a maneuver that has been attempted in order to preserve olfaction but has been uncommonly successful. The anterior face of the sphenoid sinus and both sphenoid ostia are exposed between the orbits. (F) The sphenoid sinus has been opened to expose the septa within the sinus. The sphenopalatine arteries cross the anterior face of the sphenoid. (Continued).

30

Martins and Rhoton

G

H

Vomer

Vomer Clival dura

Car. A. Cav. Seg.

Clivus

Car. A. Optic canal

Bas. sinus Pit. gland

Pit. gland Optic canal

I

J Med. Pteryg. plate Olf. Tr.

Eust. tube Pteryg. Proc. Max. A. Pterygopal. Gang. Infraorb. N. Sphen. Sinus

CN II Chiasm MCA

MCA

V2 ACA

Car. A. Figure 19 (Continued) (G) The septa within the sphenoid sinus, the sellar floor, and the lateral sinus wall have been removed to expose the intracavernous carotid, pituitary gland, and optic canals. (H) The clivus has been opened to expose the dura facing the brain stem. The basilar sinus, which interconnects the posterior parts of the cavernous sinus, is situated between the layers of dura on the upper clivus. (I) The exposure has been extended laterally by opening the medial and posterior wall of the maxillary sinus to expose the branches of the maxillary nerve and artery in the pterygopalatine fossa, located behind the posterior maxillary wall. The posterior wall of the pterygopalatine fossa is formed by the pterygoid process. The maxillary nerve enters the pterygopalatine fossa where it gives rise to the infraorbital nerve, which courses along the floor of the orbit and to the palatine nerves, which descend to the palatal area. The eustachian tube opens into the nasopharynx by passing along the posterior edge of the medial pterygoid plate. The lateral wing of the sphenoid sinus extends laterally below the maxillary nerve. (J) The frontal dura has been opened and the frontal lobes elevated to expose the olfactory and optic nerves and the internal carotid and anterior and middle cerebral arteries (1). Abbreviations: A., artery; ACA, anterior cerebral artery; Ant., anterior; Bas., basilar; Car., carotid; Cav., cavernous; CN cranial nerve; Crib., cribriform; Eth., ethmoid, ethmoidal; Eust., eustachian; Front., frontal; Gang., ganglion; Infraorb., infraorbital; Lac., lacrimal; Lig., ligament; Max., maxillary; Med., medial; MCA, middle cerebral artery; N., nerve; Nasolac., nasolacrimal; Olf., olfactory; Perp., perpendicular; Pit., pituitary; Proc., process; Pteryg., pterygoid; Pterygopal., pterygopalatine; Seg., segment; Sphen., sphenoid; Sphenopal., sphenopalatine; Sup., superior; Tr., Tract.

palatine bone medially and the body of the sphenoid bone above (Figs. 3, 6, 10, and 11). The fossa opens laterally through the pterygomaxillary fissure into the infratemporal fossa and medially through the sphenopalatine foramen to the nasal cavity. Both the foramen rotundum for the maxillary nerve and the pterygoid canal for the vidian nerve open through the posterior wall of the fossa formed by the sphenoid pterygoid process. The palatovaginal canal carrying the pharyngeal nerve and artery and the greater and lesser palatine canals conveying the greater and lesser palatine arteries open into the pterygopalatine fossa. The inferior orbital fissure, across which the orbital muscle stretches, lies in front of the pterygopalatine fossa. This fossa contains branches of the maxillary nerve, vidian nerve, the pterygopalatine ganglion, and the pterygopalatine segment of the maxillary artery. The

maxillary nerve passes through the foramen rotundum to enter the fossa and, after giving communicating rami to the pterygopalatine ganglion, divides into the posterosuperior alveolar, infraorbital, and zygomatic nerves. The zygomatic nerve, in addition to its sensory fibers, carries the parasympathetic fibers from the pterygopalatine ganglion to the lacrimal gland. The vidian (nerve of the pterygoid canal) ends in the pterygopalatine ganglion, which sends rami to the maxillary nerve and gives rise to the greater and lesser palatine, pharyngeal nerves, and nasal branches. The third part of the maxillary artery enters the fossa and divides into its terminal lesser and greater palatine, sphenopalatine, vidian, and pharyngeal branches. The parapharyngeal space lies between the structures in the pharynx wall medially, the medial pterygoid muscle

31


A

C

B

Hard Palate Clivus

Clivus

Soft palate reflected Long. Cap. M.

Soft palate Uvula For. magnum

Ant. Arch C1

Ant. Arch C1

Dens Long. Colli M.

D

E Nasal septum

Osteotomy

Vert. A.

Max. sinus

E

Maxilla Max. sinus Gr. Palat. A. & N.

Nasal septum Max. sinus

Nasal floor

Figure 20 (A) Anterior view through the open mouth. The soft palate, which extends backward from the hard palate, will block the view of the upper clivus. An incision has been outlined in the midline of the soft palate. (B) The soft palate has been divided to expose the mucosa lining the lower clivus. (C) The pharyngeal mucosa has been opened in the midline and the left longus capitus and longus coli have been reflected laterally. (D) The transverse maxillary (Le Fort I) osteotomy extends through the maxillary sinus above the apex of the teeth and below the infraorbital canals. (E) The lower maxilla has been displaced downward. A clival window and vertebral arteries are seen through the exposure (1). Abbreviations: A., artery; Ant., anterior; Cap., capitis; For., foramen; Gr., greater; Long., longus; Max., maxillary; M., muscle; N., nerve; Palat., palatine; Vert., vertebral.

and the parotid fascia laterally, and the styloid fascia investing the styloglossus, stylopharyngeal, and the stylohyoid muscles posteriorly (Fig. 6). In its upper medial wall, the eustachian tube, covered below by the tensor and levator veli palatine muscles, runs from the tympanic cavity to the pharyngeal wall. This is predominantly a fat-filled space, but also contains pharyngeal branches of the ascending pharyngeal and facial arteries and branches from the glossopharyngeal nerve. The last of the four spaces below the middle fossa is the infrapetrosal space, also referred to as the poststyloid part of the parapharyngeal space. It is located behind the styloid fascia, below the petrous bone, and medial to the mastoid process (Figs. 2, 6, and 9). Among the foramina in the area connecting the intra- and extracranial spaces is the jugular foramen containing the jugular bulb and lower end of the inferior petrosal sinus. It also contains branches of the ascending pharyngeal artery, the glossopharyngeal, vagus, and accessory nerves, and the opening of the carotid canal through which the carotid artery and the carotid sympathetic

nerves pass. Two tiny foramina located between the jugular foramen and carotid canal carry the tympanic branch of the glossopharyngeal nerve and the auricular branch of the vagus nerve. The stylomastoid foramen, conveying the facial nerve and the stylomastoid artery, opens between the mastoid tip and the styloid processes. The main fissure in the area is the petroclival fissure on the upper and lower side of which courses the inferior petrosal sinus and the inferior petroclival vein, respectively. The main nerves of the area are the glossopharyngeal nerve coursing below the styloglossus muscle, the vagus nerve descending between the internal carotid artery and the jugular vein, and the accessory nerve passing lateral to the jugular vein on its way to the sternocleidomastoid muscle. The facial nerve runs to the parotid gland where it divides into cervicofacial and temporofacial trunks. The hypoglossal nerve, after exiting the hypoglossal canal, descends between the carotid artery and the jugular vein, turning anteriorly across the lateral wall of the artery below the level of the digastric muscle. The main arteries in the area are the internal carotid artery with its cervical

32

Martins and Rhoton

A Front. bone

Orb. Oculi M.

Nasal bone Eth. bone

C

B

Lac. bone Med. canthal Lig.

Lac. Sac

Lac. Canalic.

Maxilla Max. sinus

Zygoma

Nasolac. duct Nasal cavity

D

Lac. sac Med. canthal Lig.

Lac. Canalic.

Max. sinus

F

E

Pteryg. Gang.

V2 Pterygopal. fossa Nasolac. duct Nasal cavity Max. A.

Pons

Bas. A.

Pteryg. Proc.

Inf. concha Inf. meatus

Pit. gland

Long. Cap. M.

Gr. Palat. N.

Figure 21 Relationships of the medial orbit. (A) The medial part of the orbital rim is formed by the frontal bone and maxilla. The anterior part of the nasolacrimal canal is formed by the maxilla and the posterior part by the lacrimal bone, which joins the ethmoid bone posteriorly and the frontal bone above. (B) The medial part of the orbicularis oculi muscle has been exposed. The anterior band of the medial canthal ligament, which crosses in front of the lacrimal sac, is attached to the frontal process of the maxilla medially and to the superior and inferior tarsi laterally. (C) The medial part of the orbicularis oculi muscle and some of the maxilla have been removed to expose the lacrimal sac, nasolacrimal duct, and a small part of the nasal cavity and maxillary sinus. (D) The anterior band of the medial canthal ligament has been reflected laterally to expose the superior and inferior lacrimal canaliculi joining the lacrimal sac. Additional maxilla has been removed to expose the nasal cavity and inferior turbinate medially and the maxillary sinus laterally. The nasolacrimal duct opens into the inferior nasal meatus. (E) Some of the posterior and medial wall of the maxillary sinus has been removed to expose the pterygopalatine fossa, which contains the maxillary nerve and artery and their branches and the pterygopalatine ganglion. (F) The approach has been directed through the nasal cavity medial to the pterygopalatine ganglion and fossa to the clivus, which has been opened to expose the basilar artery (1). Abbreviations: A., artery; Bas., basilar; Canalic., canaliculi; Cap., capitis; Eth., ethmoid; Front., frontal; Gang., ganglion; Gr., greater; Inf., inferior; Lac., lacrimal; Lig., ligament; Long., longus; M., muscle; Max., maxillary; Med., medial; N., nerve; Nasolac., nasolacrimal; Orb., orbital; Palat., palatine; Pit., pituitary; Proc., process; Pteryg., pterygoid; Pterygopal., pterygopalatine.

and petrous segments. The branches of the petrous segment are the caroticotympanic and vidian arteries. The ascending pharyngeal artery ascends medial to the carotid artery, giving meningeal branches which pass through the hypoglossal canal and jugular foramen as well as pharyngeal branches. The occipital artery passes posteriorly on the medial side of the posterior belly of the digastric muscle. The veins in the area are the internal jugular vein, which receives drainage from the inferior petrosal sinus, and the venous plexus of the hypoglossal canal outside the jugular foramen. The main structures in the area are the styloglossus, stylopharyngeal, and stylohyoid, the digastric nerve, and the stylomandibular ligament. The medial part of the temporal bone is constituted mainly by the internal auditory canal, the carotid canal, and the petrous apex (7,11). Laterally, within the petrous part of the temporal bone on the medial side of the mastoid antrum,

lies the semicircular canals and vestibule enclosed within the otic capsule (Fig. 5). The tympanic segment of the facial nerve passes below the lateral semicircular canal, and the mastoid segment descends to the stylomastoid foramen. The vestibule (vestibular cavity), which communicates with both ends of the semicircular canals, is situated medial to the lateral semicircular canal and below the superior semicircular canal. The aditus of the mastoid antrum opens into the tympanic cavity, which contains the malleus, incus, and stapes; the chorda tympani and tympanic nerve; the tensor tympani; and stapedius muscles. The tympanic cavity is limited laterally by the tympanic membrane, medially by the bone over the cochlea, and opens anteriorly into the eustachian tube. The arteries feeding the area arise from the stylomastoid, anterior tympanic, petrosal, and caroticotympanic arteries. Posterolateral to the otic capsule, anterior to the sigmoid sinus, and inferior to the


A

33

B

Infraorb. N.

Infraorb. N.

Zygoma Infratemp. fossa Max. sinus Infratemp. fossa

Maxilla

C

D Nasolac. duct

Infraorb. canal Pteryg. Ven. Plex. Max. A. Max. sinus Post. wall

V2

Nasal septum

nasal Septum Pterygopal. fossa

Mid. concha Inf. concha Vomer

Lat. Pteryg. M.

Infratemp. fossa

Maxilla

Clivus

V3 Brs.

E

F

Sella Cav. Seg.

V2 Sphen. Septum Pterygopal. Fossa Pteryg. Ven. Plex.

Pterygopal. fossa

Cav. sinus

Max. A. V2 Pet. Seg. CN VI Bas. A.

Vert. A. V3 Brs. For. magnum

Figure 22 (A–C) Transmaxillary exposure of the cranial base. (A) In this dissection, a midfacial soft tissue flap has been reflected laterally to expose the anterior surface of the right maxilla. The operative approach to the maxillary sinus is more commonly performed using a sublabial incision in the gingivobuccal margin rather than through an incision on the face. The approach can be completed without dividing the infraorbital nerve, but in this dissection, it was divided below the infraorbital foramen. The nerve, if divided, can be resutured at the time of closing. The infratemporal fossa, which is situated below the greater sphenoid wing, has been exposed by removing the coronoid process of the mandible and a narrow wedge of zygoma. (B) The anterior wall of the maxillary sinus has been removed. The roof of the maxillary sinus forms the majority of the floor of the orbit. The infratemporal fossa contains the pterygoid muscles, mandibular nerve, maxillary artery, and the pterygoid venous plexus. (C) The medial and lateral walls of the maxillary sinus have been opened, but the posterior part of the sinus wall, which forms the anterior wall of the pterygopalatine fossa, has been preserved. Removing the medial wall of the sinus exposes the nasal cavity, turbinates, and nasal septum. The maxillary artery crosses the lateral pterygoid muscle to reach the pterygopalatine fossa, which is located behind the upper part of the posterior wall of the maxillary sinus and below the orbital apex. (D) The posterior wall of the maxillary sinus has been removed to expose the pterygopalatine fossa and orbital floor. The pterygopalatine fossa is located below the orbital apex and the posteromedial part of the inferior orbital fissure. The maxillary nerve enters the pterygopalatine fossa by passing through the foramen rotundum. The maxillary nerve gives rise to the infraorbital nerve, which passes forward in the infraorbital canal in the sinus roof and orbital floor. (E) Enlarged view of infratemporal and pterygopalatine fossae. Distally, the maxillary artery enters the pterygopalatine fossa, which is located in the lateral wall of the nasal cavity below the orbital apex. (F) The exposure has been directed medially through the nasal cavity to the clivus, which has been opened to expose the vertebral and basilar arteries and the front of the brain stem. The exposure has been extended upward by opening the sphenoid sinus and exposing the left intracavernous carotid. The margin of the foramen magnum has been preserved (1). Abbreviations: A., artery; Bas., basilar; Brs., branches; Cav., cavernous; CN, cranial nerve; For., foramen; Inf., inferior; Infraorb., infraorbital; Infratemp., infratemporal; Lat., lateral; M., muscle; Max., maxillary; Mid., middle; N., nerve; Nasolac., nasolacrimal; Pet., petrosal; Plex., plexus; Post., posterior; Pteryg., pterygoid; Pterygopal., pterygopalatine; Seg., segment; Sphen., sphenoid; Ven., venous; Vert., vertebral.

34

Martins and Rhoton

A

B Temp. M. CN VII Front. Br. Sup. Temp. A.

Lat. canthal Lig. Inf. Obl. M.

Zygoma

Infraorb. N. CN VII

Nasal cavity Max. sinus

Coronoid Proc.

Mass. M.

C

Pterion

D

Cav. sinus V1 Pterygopal. Fossa

Lat. Pteryg. M. Inf. Obl. M. Infraorb. N.

Gr. Palat. N. & A.

V2 Mid. Men. A. Eust. tube V3

Max. A.

Nasal septum

Nasal septum

E

F Front. N.

Lac. gland Lac. gland

Lat. Rec. M. CN IV Inf. Obl. M. Inf. Rec. M.

Sup. Obl. M. Lac. N.

CN III Cav. Sinus V1 V2 Vidian N. Inf. Obl. M.

Infraorb. N.

Lat. Rec. M.

Inf. Rec. M. CN III to Inf. Obl. M.

V3

Figure 23 Upper subtotal maxillotomy. Exposure obtained with mobilization of the upper part of the maxilla. (A) This approach uses paranasal, lower conjunctival, transverse temporal, and preauricular incisions. In the usual approach, the cheek flap is elevated as a single layer using subperiosteal dissection. In this dissection, the layers of the cheek flap were dissected separately to illustrate the structures in the flap. The facial muscles and branches of the facial nerve are exposed. The parotid gland has been removed. The frontal branch of the facial nerve crosses the mid portion of the zygomatic arch. If facial nerve branches are transected in the approach, they are tagged in preparation for re-approximation at closure. (B) A hemicoronal scalp incision and reflection of the temporalis muscle expose the lateral orbital rim. The cheek flap containing the facial muscles, branches of the facial nerve, parotid gland, and masseter muscle has been reflected inferiorly to the level of the maxillary attachment of the buccinator muscle. The orbital, maxillary, and zygomatic osteotomies have been completed and the lower half of the orbital rim; the anterior, medial, and lateral walls of the maxillary sinus; and the zygomatic arch have been reflected. The lower horizontal cut, located at Le Fort I level, extends above the apical dental roots and hard palate and along the inferior nasal meatus medially. The maxillotomy, at this stage, does not include the posterior maxillary wall or cross the greater and lesser palatine canals. The lateral nasal wall was included with the maxillotomy to expose the nasal cavity. The infraorbital nerve, which crosses the orbital floor, may be preserved for reconstruction. (C) The posterior wall of the maxillary sinus has been removed to expose the pterygopalatine fossa and the palatine nerves and arteries. The base of the coronoid process was divided, and the temporalis reflected downward to expose the lateral pterygoid muscle and maxillary artery in the infratemporal fossa. (D) A frontotemporal bone flap has been elevated, and the dura covering the frontal and temporal lobes and lateral wall of the cavernous sinus has been opened, and the temporal lobe has been elevated. The pterygoid muscles, the pterygoid process and plates, and the part of the middle fossa floor formed by the greater sphenoid wing have been removed to expose the nerves passing through the foramina rotundum and ovale. The eustachian tube is exposed behind the mandibular nerve and the middle meningeal artery. (E) Magnified view of the cavernous sinus, superior orbital fissure, and orbit. The oculomotor, trochlear, and ophthalmic nerves course through the lateral wall of the cavernous sinus. The ophthalmic nerve sends its branches along the upper part of the orbit. The maxillary nerve exits the foramen rotundum and passes through the pterygopalatine fossa, where it gives rise to the infraorbital nerve that courses along the floor of the orbit. The mandibular nerve passes through the foramen ovale and sends its branches through the infratemporal fossa. The vidian nerve passes forward in the vidian canal below the maxillary nerve to join the pterygopalatine ganglion in the pterygopalatine fossa. (F) Enlarged view of the orbital exposure. The lacrimal gland sits on the superolateral margin of the globe. The lacrimal nerve courses above the lateral rectus muscle. The inferior oblique muscle passes below the attachment of the inferior rectus muscle and upward between the globe and lateral rectus muscle to insert on the globe near the tendon of insertion of the superior oblique muscle (1). Abbreviations: A., artery; Br., branch; Cav., cavernous; CN, cranial nerve; Eust., eustachian; Front., frontal; Gr., greater; Inf., inferior; Infraorb., infraorbital; Lac., lacrimal; Lat., lateral; Lig., ligament; M., muscle; Mass., masseter; Max., maxillary; Men., meningeal; Mid., middle; N., nerve; Obl., oblique; Palat., palatine; Proc., process; Pteryg., pterygoid; Pterygopal., pterygopalatine; Rec., rectus; Sup., superior; Temp., temporal, temporalis.


35

B

A

Mid. Men. A. V3 V2

Arc. Emin.

Gr. Pet. N. CN V Post. root

C

D Sup. canal

CN VII Laby. Seg. Vert. crest

.

Tens. Tymp. M.

Genic. Gang.

Pet. Car. A. Sup. Vest. N.

Modiolus

Inf. Vest. N.

Coch. N.

Clivus CN V

Inf. Pet. sinus Int. Ac. meatus

CN VII Meat. Seg.

E CN V CN III Bas. A. Post. Comm. A. CN IV Car. A. SCA Ant. Chor. A.

PCA

Figure 24 Middle fossa approach to the internal acoustic meatus. (A) The vertical line shows the site of the scalp incision and the stippled area outlines the bone flap bordering the middle fossa floor. (B) The dura has been elevated to expose the middle meningeal artery, the greater petrosal nerve, and the arcuate eminence. (C) The roof of the meatus has been opened to expose the superior and inferior vestibular, facial, and cochlear nerves. The vestibule and semicircular canals are located posterolateral and the cochlea is located anteromedial to the meatal fundus. In the middle fossa approach, for an acoustic neuroma, the cochlea and semicircular canal are not opened, as seen in this dissection illustrating the important structures which are to be avoided in opening the meatus. The vertical crest (Bill’s Bar) separates the facial and superior vestibular nerves at the meatal fundus. The superior and inferior vestibular nerves are located posteriorly and the facial and cochlear nerves anteriorly in the meatus with the cochlear nerve passing below the facial nerve to enter the modiolus. The labyrinthine segment of the facial nerve courses superolateral to the cochlea. (D) The bone of the petrous apex between the trigeminal nerve and the internal acoustic meatus has been removed to complete an anterior petrosectomy and to expose the inferior petrosal sinus and the lateral edge of the clivus. (E) The dura, exposed in the anterior petrosectomy and facing the posterior fossa and the tentorium, has been opened to expose the upper brain stem, oculomotor, trochlear, and trigeminal nerves and the basilar artery (7). Abbreviations: A., artery; Ac., acoustic; Ant., anterior; Arc., arcuate; Bas., basilar; Car., carotid; Chor., choroidal; CN, cranial nerve; Coch., cochlear; Comm., communicating; Emin., eminence; Gang., ganglion; Genic., geniculate; Gr., greater; Inf., inferior; Int., internal; Laby., labyrinthine; M., muscle; Meat., meatal; Mid., middle; Men., meningeal; N., nerve; PCA, posterior cerebral artery; Pet., petrous, petrosal; Post., posterior; SCA, superior cerebellar artery; Seg., segment; Sup., superior; Tens., tensor; Tymp., tympani; Vert., vertebral; Vest., vestibular.

36

Martins and Rhoton

A

B Sternocleidomast. M.

Sp. Henle Incus Mast. antrum deep Mastoid Tip Sup. Semicirc. canal Digastric M. Lat. Semicirc. canal Post. Semicirc. canal

Suprameat. Triang.

Suprameat. crest

CN VII Tymp. Seg. CN VII Mast. Seg. Digast. Groove Jug. Bulb Trautman’s Triang.

Mid. fossa dura Asc. Pharyg. A. Br. Longiss. Cap. M. Sinodural angle

D

C CN V motor root CN V AICA Nerv. Intermed. CN VII

Incus Chor. Tymp. N. CN IX CN VII Tymp. Seg. CN VII CN VII Mast. Seg. Stapes Laby. Seg. CN X CN VII Meat. Seg. Jug. bulb Sup. Vest. N. Inf. Vest. N. CN IX CN X CN V CN VIII Flocc. AICA Sig. sinus

CN VIII

Flocc.

F

E Malleus Cochlea Modiolus Scala Tympani Scala Vestibuli Chor. Tymp. N. CN V AICA CN VIII Coch. N. CN VII

Sig. sinus

Clivus Spiral Crest

Inf. Pet. Sinus AICA

CN VII Bas. A.

CN IX

CN VI

Jug. Bulb CN V

Flocc. Pons

CN V Motor Root Figure 25 Mastoidectomy, retrolabyrinthine, partial labyrinthine, translabyrinthine, and transcochlear approaches. (A) Right mastoid. The retroauricular flap and the sternocleidomastoid muscle have been reflected forward and the trapezius and underlying splenius capitus have been reflected backward to expose the mastoid and attachment of the longissimus capitus muscle. The posterior belly of the digastric muscle originates medial to the mastoid tip along the digastric groove. The spine of Henle is positioned at the posterosuperior margin of the external meatus, superficial to the deep site of the lateral semicircular canal and junction of the tympanic and mastoid segments of the facial nerve. The supramastoid crest, a continuation of the superior temporal line, is positioned at approximately the level of the upper margin of the transverse and sigmoid sinuses. The area below the anterior part of the supramastoid crest and behind the spine of Henle, called the suprameatal triangle, is positioned superficial to the mastoid antrum. The semicircular canals are positioned deep to the mastoid antrum. (B) The drilling has been extended to expose the middle fossa dura above, the sigmoid sinus posteriorly and the jugular bulb below. The superior, lateral, and posterior semicircular canals are located deep to the mastoid antrum and suprameatal triangle. The superior canal projects upward below the arcuate eminence. The posterior canal faces the posterior fossa dura. The lateral canal is positioned above the tympanic segment of the facial nerve. The facial nerve passes below the lateral canal and turns downward to form the mastoid segment. The dura between the sigmoid sinus and the semicircular canals, named Trautman’s triangle, faces the anterior surface of the cerebellum and cerebellopontine angle. A meningeal branch of the ascending pharyngeal artery passes through the jugular foramen and ascends in the dura of Trautman’s triangle. The jugular bulb is positioned medial to the cortical bone overlying the digastric groove. The sinodural angle is positioned at the junction of the sigmoid, transverse, and superior petrosal sinuses, and where the sigmoid sinus intersects the middle fossa dura. The short process of the incus points toward the tympanic segment of the facial nerve passing between the lateral semicircular canal and the stapes sitting in the oval window. The endolymphatic sac sits beneath the dura on the posterior surface of the temporal bone above and medial to the lower part of the sigmoid sinus. (Continued).


superior petrosal sinus lies the presigmoid dura, referred to as Trautman’s triangle, under which the endolymphatic sac sits.

POSTERIOR CRANIAL BASE Endocranial Surface The posterior cranial base corresponds to the floor of the posterior fossa, an area around the foramen magnum. It is formed by the sphenoid, temporal, and occipital bones (Figs. 12, 13, and 14) (7,12–14). Medially, it is formed by the dorsum sellae, basilar (clival) portion of the occipital bone, and the foramen magnum. Laterally, the endocranial surface is formed by the posterior surface of the temporal and the occipital bones, with the petro-occipital fissure and the jugular foramen lying between the occipital and temporal bones. The endolymphatic sac, which sits beneath the dura, inferolateral to the internal acoustic meatus, is connected through the endolymphatic duct with the vestibule. The facial–vestibulocochlear nerve complex courses through the internal auditory canal (Fig. 15) (15). The arrangement of the nerves inside the meatus is as follows: the facial nerve, anterior and superior; the superior vestibular nerve, superior and posterior; the inferior vestibular nerve, inferior and posterior; and the cochlear nerve, anterior and inferior (Figs. 5 and 15) (7,11). The intermediate nerve courses with the eighth nerve adjacent to the brain stem and jumps to the seventh nerve at some point along the cisternal or the meatal segments of the facial nerve. The trigeminal nerve exits the posterior fossa by passing through the porus trigeminus, a dural ostium that is located between the superior petrosal sinus and the petrous apex along the posterior margin of the trigeminal impression. The subarcuate fossa, a depression lateral to the internal auditory canal, is pierced by the subarcuate artery, which ends in bone in the region of the superior semicircular canal (Fig. 15). The jugular foramen lies below the internal auditory canal between the petrous part of the temporal bone and the condylar part of the occipital bone (Figs. 12 and 14) (16). It is divided into a medially situated petrosal part, through which the inferior petrosal sinus passes, a laterally situated sigmoid part through which the sigmoid sinus passes, and an intermediately positioned intrajugular part through which the nerves pass. The hypoglossal canal is located below and medial to the jugular tubercle and above the middle third of the occipital condyles, which project downward along the anterior

37

half of the foramen magnum. The posterior condylar canal, located behind the condyle, conveys the posterior condylar vein, which connects the vertebral venous plexus with the sigmoid sinus. The inferior petrosal sinus courses along the petroclival fissure, connecting the posterior cavernous sinus and jugular bulb. The abducens nerve ascends and pierces the dura to course extradural through Dorello’s canal located between the petrosphenoid ligament and the upper edge of the fissure between the dorsum sella and the petrous apex, to enter the cavernous sinus. Below the foramen magnum, at the level of the atlanto-occipital joint, and anterior to the tectorial membrane, the cruciform, apical, and alar ligaments maintain the stability of the odontoid and craniocervical junction. The petrosal cerebellar surface, which faces the posterior surface of the temporal bone, the anterior surface of the brain stem, and the cerebellar peduncles, faces the endocranial surface of the posterior skull base (Fig. 16). The medulla, pons, and mesencephalon face the clivus. The surface of the medulla is divided longitudinally by the preolivary and postolivary sulci, with the pyramid in front and the inferior cerebellar peduncle behind the olive. The hypoglossal nerve arises along the preolivary sulcus and the glossopharyngeal, vagus, and accessory nerves arise near the retroolivary sulcus. The vestibulocochlear and facial nerves arise a few millimeters above the retro-olivary sulcus in the lateral part of the pontomedullary sulcus. The abducens nerve arises in the medial part of the pontomedullary sulcus and ascends behind the clivus. The trigeminal nerve arises from the anterolateral surface of the midpons with the rootlets which join to form the motor root arising around the superior third of the sensory root. The trochlear nerve arises below the inferior colliculus and passes forward along the pontomesencephalic sulcus in the quadrigeminal and ambient cisterns to enter the tentorial edge just behind the cavernous sinus. The oculomotor nerve crosses the interpeduncular cistern and pierces the roof of the cavernous sinus. Both vertebral arteries enter the cranium at the posterior edge of the occipital condyles and ascend to join in the midline, thus forming the basilar artery (Figs. 12 and 16). The main branches of the intradural vertebral artery are the posterolateral spinal artery, which supplies the posterior third of the medulla and upper spinal cord; the anterior spinal artery, which joins its mate from the opposite side on the anterior surface of the cord to supply the anterior two-thirds of the medulla and upper spinal cord; and the

←−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−− Figure 25 (Continued) (C) The retrolabyrinthine exposure has been completed and the dura has been opened to expose cranial nerves V to X in the cerebellopontine angle. The vestibulocochlear nerve has been depressed to expose the facial nerve and the nervus intermedius. The motor root of the trigeminal nerve is exposed superomedial to the main sensory root. The glossopharyngeal and vagus nerves are at the lower margin of the exposure. The flocculus protrudes from the foramen of Luschka behind the vestibulocochlear nerve. The anteroinferior cerebellar artery loops laterally between the facial and vestibulocochlear nerves. (D) The translabyrinthine approach has been completed to expose the vestibulocochlear and facial nerves in the internal acoustic meatus. The meatal and labyrinthine segments of the facial nerve are exposed proximal and the tympanic and mastoid segments are exposed distal to the geniculate ganglion. The dura of Trautman’s triangle has been opened to expose the trigeminal, glossopharyngeal, and vagus nerves in the cerebellopontine angle. The anteroinferior cerebellar artery loops laterally into the meatus before turning back toward the brain stem. The facial and superior and inferior vestibular nerves are exposed at the fundus of the meatus. The cochlear nerve is hidden anterior to the inferior vestibular nerve. (E) The greater petrosal nerve has been sectioned just distal to the apex of the geniculate ganglion and the facial nerve has been displaced posteriorly for removal of the cochlea in the transcochlear approach. The semicircular canals and vestibule, the end organs of the superior and vestibular nerves, have been removed. The incus has been removed but the malleus remains attached to the tympanic membrane. Drilling has been extended forward into the cochlea. The cochlear nerve enters the modiolus in the center of the spiral turns of the cochlea. (F) Removal of the cochlea opens the channel for removing the remainder of the petrous apex. The exposure extends to the lateral edge of the clivus and the inferior petrosal sinus. The basilar artery and anterior surface of the pons are at the deep end of the exposure. The abducens nerve passes behind the anteroinferior cerebellar artery and lateral to the basilar artery. Abbreviations: A., artery; AICA, anteroinferior cerebellar artery; Asc., ascending; Bas., basilar; Br., branch; Cap., capitis; Chor., chorda; CN, cranial nerve; Coch., cochlear; Digast., digastric; Flocc., flocculus; Inf., inferior; Intermed., intermedius; Jug., jugular; Laby., labyrinthine; Lat., lateral; Longiss., longissimus; M., muscle; Mast., mastoid; Meat., meatal; Mid., middle; N., nerve; Nerv., nervus; Pet., petrosal; Pharyng., pharyngeal; Post., posterior; Seg., segment; Semicirc., semicircular; Sig., sigmoid; Sp., spine; Sternocleidomast., sternocleidomastoid; Sup., superior; Suprameat., suprameatal; Triang., triangle; Tymp., tympani, tympanic; Vest., vestibular.

38

Martins and Rhoton

A

B

CN III

Sup. Pet. sinus V. of Labbe’

CN IV

Trautman’s triangle Otic capsule

S.C.A. Sup. Pet. V.

Sp. Henle CN V

Jug. bulb

Bas. A. CN VI CN VII CN VIII

AICA

PICA Vert. A. CN X

C

CN IX

D SCA

CN IV

SCA Marg. Br. CN V V. of Labbe’

Int. Ac. meatus Sup. canal

CN VII CN VIII

CN V Lat. canal CN VII CN VIII

Post. canal

Chor. Tymp. N. CN IX CN X

CN VII

CN VII CN IX CN X

Figure 26 Presigmoid approach. (A) The insert shows the temporo-occipital craniotomy and the mastoid exposure. The mastoidectomy has been completed and the dense cortical bone around the labyrinth has been exposed. The tympanic segment of the facial nerve and the lateral canal are situated deep to the spine of Henle. Trautman’s triangle, the patch of dura in front of the sigmoid sinus, faces the cerebellopontine angle. (B) Retrolabyrinthine exposure. The presigmoid dura has been opened and the superior petrosal sinus and tentorium divided, taking care to preserve the vein of Labbé, which joins the transverse sinus, and the trochlear nerve, which enters the anterior edge of the tentorium. The abducens and facial nerves are exposed medial to the vestibulocochlear nerve. The PICA courses in the lower margin of the exposure with the glossopharyngeal and vagus nerves. The SCA passes below the oculomotor and trochlear nerves and above the trigeminal nerve. (C) The semicircular canals have been opened. The superior canal is located under the middle fossa’s arcuate eminence and the posterior canal is located immediately lateral to the posterior wall of the internal acoustic meatus. (D) The labyrinthectomy has been completed to expose the internal acoustic meatus (7). Abbreviations: A., artery; Ac., acoustic; AICA, anteroinferior cerebellar artery; Bas., basilar; Br., branch; Chor., chorda; CN, cranial nerve; Int., internal; Lat., lateral; Marg., margin; N., nerve; Pet., petrosal; PICA, posteroinferior cerebellar artery; Post., posterior; SCA, superior cerebellar artery; Sp., spine; Sup., superior; Tymp., tympani; V., vein; Vert., vertebral.


A

39

B Occip. A.

Int. Car. A.

Post. Aur. N. Sup. nuchal line

Parotid Gl. Gr. Aur. N. Sternocleidomast. M.

Stylomast. A. CN VII

C1 Trans. Proc. Int. Jug. V.

Digastric M. Jug. bulb

Inf. Obl. M. Sup. Obl. M.

Semicirc. canals Sig. sinus

C

D CN VII

CN VII Incus

Chor. Tymp. N. CN VII Jug. Bulb

Lat. canal Sup. canal

Sig. Sinus

Pet. Car. A.

Mid. fossa Int. Jug. V.

Stapes Round window Lat. canal

Tympanic N.

Post. canal Jug. Bulb

F

E Symp. Tr. Sup. Laryn. N. CN XII CN X

Post. canal

Int. Car. A. CN VII

CN XII CN IX Int. Car. A. CN X, XI CN IX Intrajug. ridge C1 Trans. Proc. Intrajug. Proc. Inf. Pet. sinus

CN XI

Jug. bulb Med. wall

CN IX CN X, XI

Occip. A. Figure 27 (A–D) Postauricular exposure of the jugular foramen. (A) The C-shaped retroauricular incision (insert) provides access for the mastoidectomy, neck dissection, and parotid gland displacement. The scalp flap has been reflected forward to expose the sternocleidomastoid muscle and the posterior part of the parotid gland. (B) The more superficial muscles and the posterior belly of the digastric have been reflected to expose the internal jugular vein and the attachment of the superior and inferior oblique muscles to the transverse process of C1. A mastoidectomy has been completed to expose the facial nerve, sigmoid sinus, and the semicircular canals. (C) Enlarged view of the mastoidectomy. The jugular bulb is exposed below the semicircular canals. The chorda tympani arises from the mastoid segment of the facial nerve and passes upward and forward. The tympanic segment of the facial nerve courses below the lateral canal. (D) The external auditory canal has been transected and the middle ear structures have been removed, except the stapes, which remains in the oval window. The lateral edge of the jugular foramen has been exposed by completing the mastoidectomy, transposing the facial nerve anteriorly, and fracturing the styloid process across its base and reflecting it caudally. The petrous carotid is surrounded in the carotid canal by a venous plexus. (E) A segment of the sigmoid sinus, jugular bulb, and internal jugular vein has been removed. The lateral wall of the jugular bulb has been removed while preserving the medial wall and exposing the opening of the inferior petrosal sinus into the jugular bulb. Removing the medial venous wall exposes the portion of the glossopharyngeal, vagus, accessory, and hypoglossal nerves that are hidden deep to the vein. The main inflow from the petrosal confluens is directed between the glossopharyngeal and vagus nerves. (F) The medial venous wall of the jugular bulb has been removed. The intrajugular ridge extends forward from the intrajugular process and divides the jugular foramen between the sigmoid and petrosal parts. The glossopharyngeal, vagus, and accessory nerves enter the dura on the medial side of the intrajugular process, but only the glossopharyngeal nerve courses through the foramen entirely on the medial side of the intrajugular ridge (14). Abbreviations: A., artery; Aur., auricular; Car., carotid; Chor. Tymp., chorda tympani; CN, cranial nerve; Gl., gland; Gr., greater; Inf., inferior; Int., internal; Intrajug., intrajugular; Jug., jugular; Laryn., laryngeal; Lat., lateral; M., muscle; Med., medial; Mid., middle; N., nerve; Obl., oblique; Occip., occipital; Pet., petrosal, petrous; Post., posterior; Proc., process; Semicirc., semicircular; Sig., sigmoid; Sternocleidomast., sternocleidomastoid; Stylomast., stylomastoid; Sup., superior; Symp., sympathetic; Tr., trunk; Trans., transverse; V., vein.

40

Martins and Rhoton

posteroinferior cerebellar artery (PICA), which courses around the medulla and supplies the suboccipital cerebellar surface. The main branches of the basilar artery are the anteroinferior cerebellar artery (AICA), which passes around the pons and supplies the petrosal cerebellar surface and the nerves entering the internal auditory canal; the superior cerebellar artery (SCA), which encircles the midbrain and upper pons and supplies the tentorial cerebellar surface and dentate nucleus; and the posterior cerebral artery (PCA), which passes to the supratentorial area. The main venous drainage is by way of the petrosal veins, which join the superior and inferior petrosal veins emptying into the superior and inferior petrosal sinuses. The vascular and neural structures of the posterior fossa can be divided into upper, middle, and lower groups related to the three cerebellar arteries: the upper group is related to the SCA, which encircles the midbrain, courses on the superior cerebellar peduncle and within the cerebellomesencephalic fissure, passes below the oculomotor and trochlear nerves and above the trigeminal nerve, and supplies the tentorial cerebellar surface; the middle group is related to the AICA, which encircles the pons, courses on the middle cerebellar peduncle, dips in the cerebellopontine fissure passing by and sending branches to the facial and vestibulocochlear nerves, and supplies the petrosal cerebellar surface; and the lower group is related to the PICA, which encircles the medulla, passes near or between the rootlets of the lower four cranial nerves to course on the inferior cerebellar peduncle, dips into the cerebellomedullary fissure, and supplies the suboccipital cerebellar surface (Figs. 12 and 16) (17).

Exocranial Surface This portion is divided in central and lateral portions (Fig. 14) (14,18,19). The center portion is formed by the basal (clival) part of the occipital bone, which slopes upward from the foramen magnum and has the pharyngeal tubercle for the attachment of the superior pharyngeal constrictor on its lower surface, and the occipital condyles lateral to its lower portion at the anterolateral margin of the foramen magnum. The hypoglossal foramen, conveying the hypoglossal nerve, crosses above the middle one-third of the long axis of the condyle. The posterior condylar canal carries the posterior condylar vein interconnecting the vertebral venous plexus with the sigmoid sinus. Lateral to the condyle lies the jugular process of the occipital bone, which forms the posterior edge of the jugular foramen, connects the squamosal and basal parts of the occipital bone, and receives the attachment of the rectus capitus lateralis muscle posterior to the jugular foramen and jugular bulb. Two grooves lateral to the jugular process, a medial one for the occipital artery and a lateral one, the digastric groove, for the origin of the posterior belly of the digastric muscle, separate the jugular process from the mastoid process. The remaining muscles are the longus capitis attached to the lower clivus and the rectus capitis anterior attached in front of the occipital condyle. The nerves of the area are the hypoglossal and C1 nerves. The vertebral artery ascends through the C1 transverse process and crosses medially behind the superior articular pillar of the atlas or the atlanto-occipital joint to enter the dura (Figs. 17 and 18). The first segment of the vertebral artery ascends from the subclavian artery running up to the transverse foramen of C6. The second segment runs from C6 to C2, where the artery changes to a more lateral direction. The third segment ascends laterally to reach the C1 transverse foramen and turns medially and horizontally behind the atlanto-occipital joint

to course in the depths of the suboccipital triangle delimited by three muscles: the superior oblique extending from the occipital bone to the C1 transverse process, the inferior oblique extending from the C1 transverse process to the C2 spine, and the rectus capitus posterior major extending from the C2 transverse process to the occipital bone (Figs. 17 and 18). The fourth segment of the vertebral artery extends from the dural entrance to the vertebrobasilar junction. The third segment gives rise to the posterior meningeal and muscular arteries and occasionally the PICA. The other vascular structures in the area are the occipital and ascending pharyngeal arteries, the vertebral venous plexus, and the posterior and anterior condylar veins.

DISCUSSION With the development of microsurgical techniques and skull base surgical principles, it has become possible to access all parts of the cranial base. Lesions involving the central part of the anterior two-thirds of the cranial base and clivus can be accessed through intracranial routes, such as the orbitozygomatic, transcranial–transbasal, or extended frontal approaches, or by subcranial routes utilizing the various modifications of the transnasal, transoral, transsphenoidal, transmandibular, transmaxillary, transcervical, or facial translocation approaches or by a combination of the intracranial and subcranial routes (Figs. 19–23). The approaches can also be extended to the middle skull base by using the orbitozygomatic, preauricular infratemporal fossa, subtemporal anterior petrosectomy, or other extensions of the middle fossa routes (Fig. 24). Further posteriorly, approaches directed through the temporal bone, such as the retrolabyrinthine, translabyrinthine, transcochlear, and combined supra-infratentorial presigmoid approaches, or a combination of preauricular and postauricular transtemporal approaches, may be considered (Figs. 25 and 26). Lesions in the posterior fossa and posterior skull base may be reached through the retrosigmoid or suboccipital routes or the farlateral approach and its transcondylar, supracondylar, and paracondylar modifications (Fig. 24) (19). The jugular foramen is most often accessed by a postauricular transtemporal approach (Fig. 27). Cranial base tumors frequently invade intracranial and subcranial spaces and require innovative combinations of these transcranial, subcranial, and combined approaches. Thoughtful consideration of skull base anatomy is essential to successful surgery for these tumors. REFERENCES 1. Rhoton AL Jr. The anterior and middle cranial base. Neurosurgery. 2002;51(1 Suppl):S273–S302. 2. Rhoton AL Jr., Seoane E. Surgical anatomy of the skull base. In: Harsh, G., ed. Chordomas and Chondrosarcomas of the Skull Base and Spine. New York: Thieme Medical Publishers Inc, 2003:57–79. 3. Rhoton AL Jr. The orbit. Neurosurgery. 2002;51(1 Suppl):S303– S334. 4. Rhoton AL Jr., Natori Y. The Orbit and Sellar Region: Microsurgical Anatomy and Operative Approaches. New York, NY: Thieme Medical Publishers Inc, 1996:3–25. 5. Natori Y, Rhoton AL Jr. Transcranial approach to the orbit: Microsurgical anatomy. J Neurosurg. 1994;81:78–86. 6. Natori Y, Rhoton AL Jr. Microsurgical anatomy of the superior orbital fissure. Neurosurgery. 1995;36:762–775. 7. Rhoton AL Jr. The temporal bone and transtemporal approaches. Neurosurgery. 2000;47(3 Suppl):S211–S265.

Chapter 1: Anatomy of the Cranial Base 8. Rhoton AL Jr. The sellar region. Neurosurgery. 2002;51(1 Suppl): S335–S374. 9. Rhoton AL Jr. The cavernous sinus, the cavernous venous plexus, and the carotid collar. Neurosurgery. 2002;51(1 Suppl): S375– S410. 10. Seoane E, Rhoton AL Jr., de Oliveira E. Microsurgical anatomy of the dural collar (carotid collar) and rings around the clinoid segment of the internal carotid artery. Neurosurgery. 1998;42:869– 886. 11. Pait TG, Zeal AA, Harris FS, et al. Microsurgical anatomy and dissection of the temporal bone. Surg Neurol. 1977;8:363– 391. 12. Rhoton AL Jr. Cerebellum and fourth ventricle. Neurosurgery. 2000;47(3 Suppl): S7–S27. 13. Rhoton AL Jr. The foramen magnum. Neurosurgery. 2000;47(3 Suppl):S155–S193.

41

14. Rhoton AL Jr. Jugular foramen. Neurosurgery. 2000;47(3 Suppl): S267–S285. 15. Rhoton AL Jr. The cerebellopontine angle and posterior fossa cranial nerves by the retrosigmoid approach. Neurosurgery. 2000;47(3 Suppl):S93–S129. 16. Katsuta T, Rhoton AL Jr., Matsushima T. The jugular foramen: Microsurgical anatomy and operative approaches. Neurosurgery. 1997;41:149–202. 17. Rhoton AL Jr. The cerebellar arteries. Neurosurgery. 2000;47(3 Suppl):S29–S68. 18. Rhoton AL Jr. The far-lateral approach and its transcondylar, supracondylar, and paracondylar extensions. Neurosurgery. 2000;47(3 Suppl):S195–S209. 19. Wen HT, Rhoton AL Jr., Katsuta T, et al. Microsurgical anatomy of the transcondylar, supracondylar, and paracondylar extensions of the far-lateral approach. J Neurosurg. 1997;87:555–585.

2 Pathology of Tumor and Tumor-like Lesions of the Skull Base Michelle D. Williams and Adel K. El-Naggar

INTRODUCTION

to avoid complications such as CSF leak. Generally, these lesions present as smooth, homogenous, tan soft tissue mimicking brain parenchyma. Histologically, they are typically composed of neural tissue with fibrosis and astrocytic and gametocytic cell proliferation.

The skull base regions are composed of complex tissue structures that give rise to histogenetically and biologically heterogeneous neoplasms of ectodermal, endodermal, and mesodermal origins. The morphologic and histogenetic differences are complicated by overlapping features and frequently pose diagnostic difficulties. The vast majority of tumors at these locations are malignant with a small percentage being benign or tumor-like lesions (Table 1). Accurate diagnosis and understanding of the clinical and pathologic presentations of the varied tumor entities in this region are essential for proper management. Although the majority of tumors at this location are of primary origin, metastasis can be encountered and will be discussed. Nonmetastatic tumors are either primary or an extension from neighboring structures.

Differential Diagnosis This entity can be differentiated from encephalocele, which frequently shows meninges.

Respiratory Epithelial Adenomatoid Hamartoma This is a benign proliferation of minor seromucinous glands of the sinonasal tract, occurring more commonly in men than women in their sixth decade of life. The major symptoms include nasal obstruction, epistaxis, and recurrent sinusitis. These lesions appear normally as polypoid tan to reddishbrown and rubbery tissue nodules (4–6). Histologically, they are formed of numerous glandular structures lined by ciliated respiratory epithelium with thickened basement membrane and intervening fibrotic and/or edematous stroma (Fig. 2).

Biopsies and Frozen Sections The evaluation of sinonasal pathology typically requires a tissue biopsy that may be limited dependent upon the accessibility of the target region. Obtaining adequate and representative materials is essential for accurate diagnosis and better planning of patient management. The initial assessment of these tumors is commonly conducted intraoperatively for either provisional or definitive diagnosis and/or verifying adequacy for representative tissue. Communication with the pathologist at the time of frozen section is key to coordinating patient care. At this stage, non-neoplastic processes, lymphoma, and metastatic neoplasms can be determined. For primary tumors, the frozen tissue biopsy may be adequate for diagnosis and planning ancillary tests but efforts to secure additional tissue for permanent processing is strongly recommended for optimal morphologic assessment and biomarker characterization.

Differential Diagnosis These lesions may be confused with Schneiderian inverted papilloma and sinonasal adenocarcinomas. The benign glandular structures lined by columnar cells that form these lesions are key to differentiating it from both of these entities.

Ectopic Pituitary Adenoma Pituitary adenomas may occur in the sphenoid bone and sinuses either as a separate lesion or as an extension from a primary adenoma arising in the sella (7,8). Embryonic residue along the Rathke pouch formation is the presumed derivation. Females are more affected than males (2:1 ratio). Patients may present with nasal obstruction, headache, or epistaxis. Approximately half of patients manifest hormonal abnormalities. Histologically, an ectopic pituitary adenoma is identical to that of a conventional pituitary adenoma with monotonous round cells (Fig. 3).

NON-NEOPLASTIC AND CONGENITAL LESIONS Encephalocele Based on the age of the patient and the location of the lesion, the diagnosis of an encephalocele may be made by imaging prior to submitting histology. Frequently, encephaloceles extending into the nasal cavity or sinus include meninges and glial tissue associated with fibrosis (Fig. 1) (1,2).

Differential Diagnosis This lesion should be differentiated from carcinoid tumor, neuroblastoma, and other small undifferentiated tumors at these locations. Immunohistochemical staining for hormonal receptors, especially for ACTH and prolactin, is helpful.

Nasal Glial Heterotopia This is a congenital malformation in which ectopic glial tissue is found without connection to intracranial structures (3). Nasal glial heterotopia may present as an extranasal, intranasal, or mixed intranasal and subcutaneous mass. Patients may also present with symptoms and findings of nasal polyp, chronic sinusitis, and otitis media. Radiological confirmation of the lack of intracranial communication is stressed

Inflammatory Pseudotumor This is a benign reactive process where a spindle cell tumor– like proliferation with inflammatory component is the cardinal feature (9–11). They may arise at any site in the skull base regions. 43

44

Williams and El-Naggar Table 1

Incidence of Malignant Sinonasal Tumors∗

Feature

%

Sites • Maxillary • Nasal cavity • Ethmoid • Frontal & sphenoid

60% 22% 15% 3%

Derivation • Epithelial • Mesenchymal • Neuroectodermal • Others

55% 30% 15% 5%

∗ ∗

0.2 msec, (ii) a wave I–V interpeak latency (I–V IPL) >4.4 msec, and/or (iii) poor waveform morphology with either absent wave or no response. Absolute latencies are not considered as useful as interpeak latencies for the diagnosis of ANs because the absolute latency is affected by many factors, such as click intensity, hearing loss, and age (117). Some authors have not used interpeak latencies since wave I or II is difficult to identify even in normal listeners (112). Additionally, waveform amplitude is not used as a criterion since it is highly variable (112). A compilation of criteria used by several authors is presented in Table 9. Many different studies have examined the sensitivity of ABR by tumor size (Table 10), and a statistically significant positive correlation between tumor size and wave V latency has been reported (92). Additionally, when ABR waveforms from the contralateral ear are abnormal (e.g., a delayed wave V or prolonged wave I–V interval), a tumor larger than 2 cm should be suspected (118). Thus, it is concluded that ABR is nearly 100% sensitive for tumors larger than 2 cm (62,116,119). For this reason, ABR is a desirable screening test in the elderly and poor surgical risk patients for whom surgery may be indicated only for a symptomatic, large tumor (61,111). However, to have a reasonable chance of hearing preservation, tumors should be diagnosed as early (small) as possible, where small is defined as 2 cm or smaller (99,120). ABR sensitivity rates for tumors smaller than 1 cm range from 63% to 93% (62,97,113–116). Healthy patients with unilateral symptoms (hearing loss, poor discrimination, and tinnitus) should have MRI with gadolinium enhancement to find an AN. ABR still provides excellent insight into the physiology of the acoustic nerve, and this may have important implications for hearing preservation. Matthies and Samii (121) found that preoperative ABR was more important than the preoperative hearing quality for the chances of hearing preservation. They found that the presence of wave III correlated with better postoperative results, especially SDS (121). Robinette et al. (99) found similar results regarding the presence of waves I, III, and V when looking at preoperative predictors of hearing preservation. They reported that when these three waves are present, 61% of patients had hearing preservation, while only 27% of patients with one or more waves absent had hearing preserved (99). One can find papers that dispute any relationship between ABR waveforms and hearing preservation as well (120,122,123). It should be noted that poor ABR waveforms should not be used as a criteria to exclude the possibility of hearing preservation. Stidham and Roberson (124) reported a series of 30 patients undergoing middle fossa craniotomy for hearing preservation. They described seven patients with hearing improvement, classified as an increase in PTA ≥ 5 dB and/or an improvement in SDS by ≥ 12%. Interestingly, no patient with normal preoperative ABR experienced a hearing

Chapter 5: Head, Neck, and Neuro-otologic Assessment of Patients with Tumors of the Skull Base

111

Table 9 Definitions of Abnormal ABR Interaural wave V latency difference, msec (ILD-V) (a.k.a. IT5)

Authors

Absolute wave V latency, msec

Interaural latency difference of I–V, msec (ILD I–V)

I–V interpeak latency, msec (I–V IPL)

Waveform morphology

House and Brackmann, 1979 (125) Bauch, 1982 (112)

>0.2

>6

Absence of wave V

>0.2

>6.1

Josey, 1988 (169)

>0.4

No response or poor overall waveform at high intensities Absence of V despite good hearing

Weiss, 1990 (170) Wilson, 1992 (113)

≥0.4 ≥0.4

>6.03

Dornhoffer, 1994 (114) Chandrasekhar, 1995 (115)

≥0.4 >0.2

>5.9

Gordon and Cohen, 1995 (62) Berrettini 1996 (83)

Ferguson, 1996 (106) Ruckenstein, 1996 (97)

>0.2

Saleh 1996 (95) Zappia, 1997 (116)

>0.3 >0.2

Godey, 1998 (84)

>0.2

Noguchi, 1999 (171) El Kashlan, 2000 (61)

≥0.3 >0.4

Haapaniemi, 2000 (144) Marangos, 2001 (110) Rupa, 2003 (119)

≥0.4 >0.3 ≥0.3

Cueva, 2004 (111)

>0.2

>4.4 ≥0.4

“abnormally prolonged” (>2 SD above normal limit for patient’s age and gender) >6.10 (male) >5.97 (female) Abnormal absolute wave V latency >6

≥4.45 ≥4.4

Abnormal ipsilateral or contralateral waveforms Abnormal or absent waveform morphology

>0.2 >0.3

>4.3

>0.3

>4.58 (male) >4.34 (female) ≥4.4

>0.2

>4.4 ≥4.4 >4.4

>7.75 >=0.4 >0.2

≥4.4 >4.4 ≥4.4

Abnormal absolute wave V latency

improvement. Of course, ABR is the most common technique used to monitor hearing intraoperatively (Chap. 6).

Other Tumors ABR results for posterior fossa meningiomas have similar rates of sensitivity as is found for ANs. In the pre-MRI era, House and Brackmann (125) found that only 75% of patients

Poor morphology in spite of adequate hearing

Absent or poor waveform morphology Absent waves, if adequate PTA Absent or abnormal waveform morphology Absent or abnormal waveforms Complete absence of waves if adequate PTA or absence of waves beyond wave I Abnormal or absent Absence of one or more waves, poor waveform morphology Absent or distorted waveform morphology

with non-AN pathology had abnormal ABRs. In their paper, 3 out of 10 meningiomas had normal ABR (125). Laird et al. (126) and Granick et al. (127) each found six out of six posterior fossa meningiomas had abnormal ABR. Aiba et al. (128) reported abnormal ABR in 8/10 cases; and Hart and Lillehie (129) reported abnormal ABR in 5/7 cases. Baguley contributed another 25 cases of CPA meningiomas, and found

Table 10 ABR Sensitivity with Respect to Tumor Size Study

No. of patients

Gordon and Cohen, 1995 (62) Chandrasekhar, 1995 (115) Bauch, 1996 Zappia, 1997 (116)

105 197

Marangos, 2001 (110)

309

Wilson, 1992 (113) Godey, 1998 (84)

111

40 89

≤10 mm

11–20 mm

>21 mm

69% 83% 82% 89% 25 mm 96.7%

Intracanalicular 66% 77%

Extracanalicular 96% 94%

112

Gidley

abnormal ABR in 100% of their tumors (100). Marangos et al. (110) found that 23.5% of meningiomas had normal ABR. Clearly, MRI is required to make the diagnosis of this tumor as well. Epidermoid tumors generally present at an advanced stage with multiple cranial nerve deficits and cerebellar signs (102,103). In the series by Quaranta (102), tumors ranged from 3.5 to 7 cm in maximum diameter. They found that ABR was normal in just one case. Absent or delayed waves were present in five cases, ipsilateral to the tumor. Four cases had bilateral abnormalities on ABR. Thus, 90% of their patients had abnormalities on ABR (102).

Stacked ABR Don et al. (130) described a new technique of ABR they called “stacked ABR.” In this test, ABR is obtained using 63 dB normal hearing level clicks in a high-pass noise-making procedure. The wave V amplitude is constructed by temporally aligning wave V of each derived-band ABR and summing the time-shifted responses. Using this technique, they found significantly lower wave V amplitudes in five AN patients who were missed by conventional ABR technique. These five tumors were all less than 1 cm in greatest dimension. They propose this technique as a cost-effective approach for AN screening. In a further study of stacked ABR, Philibert et al. (131) noticed that stacked ABR required a masking technique that might not be readily available. Additionally, they found that the relatively high intensity of the test might be annoying to the patient. Instead, they propose using tone burst to obtain a frequency-specific ABR.

Otoacoustic Emissions The phenomenon of sound being produced by the ear was first described in 1948 (132), and the definitive paper on otoacoustic emissions (OAEs) was published in 1978 (133). However, it was not until the 1990s that OAE testing became clinically widespread (134). OAE testing has enjoyed a significant increase in usage as part of a neonatal hearing screening strategy, monitoring ototoxicity or noise-induced hearing loss, and in suspected cases of functional hearing loss (134). OAEs are generated by outer hair cells of the cochlea (135). While OAEs are not useful as a screening test for ANs or other skull base tumors, they are measures of “cochlear reserve” and have been examined as possible predictors of hearing preservation (77,99,136). OAEs are divided into two groups: spontaneous and evoked. Spontaneous emissions are present in roughly 60% of normal ears (137). Evoked emissions are present in virtually 100% of normal ears (134). Evoked emissions are divided into distortion product (DPOAEs) and transient evoked (TEOAEs) emissions [Fig. 11(A) and 11(B)]. In a literature review, Robinette et al. (99) examined five studies describing 236 AN patients and reported that TEOAEs were present in at least one frequency in 47% of tumor ears. Brackmann et al. (77) described 333 AN patients considered for hearing preservation, 56 of these patients had DPOAEs measured. Normal DPOAEs were found in 91% (77). Ferber-Viart et al. (138) examined 168 AN patients with TEOAEs; (21%) had normal preoperative TEOAEs. They did not find an association with tumor size, functional symptoms, PTA, ABR, or electronystagmography (ENG) response with TEOAEs. Patients with TEOAEs tended to be younger, on average six years younger than those without TEOAEs.

In the subpopulation of 63 patients who underwent hearing conservation surgery, TEOAEs were present in 28% and absent in 72%. Sixty-six percent of those with TEOAEs present had hearing preserved, while only 44% with absent TEOAEs had hearing preserved. This difference was not statistically significant (138). In a more recent study, Kim et al. (136) examined 93 patients with AN that were candidates for hearing preservation. Fifty-one patients had hearing preserved. Eleven (22%) of these 51 patients had TEOAEs present in all five frequencies tested (1–4 kHz), while 40 (78%) had TEOAE responses anywhere from 0 to 4 of the frequencies tested. In the 42 patients who did not have hearing preserved, only three (7%) had positive TEOAEs in all five frequency bands, and 39 (93%) had TEOAEs from 0 to 4 frequency bands (p < 0.05). Other positive factors for hearing preservation in their series were small tumor size, tumor within the IAC, better hearing, and shorter latencies on ABR. Their conclusion was that a robust preoperative TEOAE pattern may be used as a favorable indicator for hearing preservation, especially when combined with the other positive factors listed above (136).

Electronystagmography Since its introduction in the 1960s, ENG (or more commonly now videonystagmography) has established itself as the most common test performed in evaluating patients with complaints of dizziness and vertigo (139). This test combines positional testing, optokinetic testing, random saccades and visual pursuit tests, and caloric stimulation to evaluate the vestibular ocular reflex and visual tracking centers of the brain. Findings on ENG for peripheral lesions are well described (140). ENG is an extremely valuable tool for examining the anatomic and functional integrity of the central and peripheral vestibular systems (141). The sensitivity of caloric testing for acoustic tumors ranges from 44% to 95% (83,140,142–144). Reduced or absent caloric response is the most frequent finding in AN patients (140,145,146). The amount of caloric weakness is proportional to the size of the tumor (140,145,146), although this has been disputed by others (92). Different authors use various criteria to describe a significant weakness; this can range from 20% (84) to 25% (77,147) and have a significant impact on the sensitivity and specificity of the test. The incidence of diminished caloric response by ENG for AN patients is presented in Table 11. More recently, head-shaking nystagmus (HSN) has been studied as a possible screening test for ANs. Humphriss et al. (27) studied 102 AN patients seen preoperatively. They used a passive head-shaking maneuver (1–2 Hz) and recorded eye movements with an ENG system. A significant response was five or more beats of nystagmus with a slow phase of at least 3 degrees/sec. In their study, significant caloric paresis was ≥ 25%. All patients had tumors confirmed by MRI and surgery. They found HSN in only 22 patients (i.e., sensitivity 22%). HSN was contralaterally beating in 19 patients (86%), ipsilaterally beating in three patients (14%), but absent in the remaining 80%. HSN was found more often in patients with either a greater canal paresis or have central vestibular signs than patients without HSN; however, the sensitivity still remains low [only 36% sensitivity even with severe (75–100%) canal paresis] (27). A low sensitivity rate (47.6%) for HSN was also reported by Asawavichiangianda et al. (148) for ANs. The range of reported sensitivity and specificity of HSN has been tabulated by Humphriss et al. (27). Sensitivity ranges from 22% to 95%; and specificity for a unilateral vestibular


(A)

30

EAR: TONE PAIR:

SPL (dB)

20

113

Left Sequential

10 0 −10 −20

500

1K

8K

Score = (None)

Test result = (n/a) (B)

4K

2K F2 (Hz)

30

EAR: TONE PAIR:

SPL (dB)

20

Right: Sequential

10 0 −10 −20

500

1K

2K

4K

8K

F2 (Hz) Test result = (n/a)

Score = (None)

Figure 11 (A) Distortion product otoacoustic emissions (marked with Xs) from left ear of same patient with ipsilateral intracanalicular acoustic neuroma. Emissions are diminished but present from 1 to 3 kHz, and absent for higher frequencies, consistent with the audiometric findings. (B) Distortion product otoacoustic emissions (marked by circles) from right ear from same patient with left intracanalicular acoustic neuroma and are normal from 1 to 4 kHz.

disorder ranges from 53% to 92%. Across the 10 studies reviewed, many different criteria are used for the “gold standard” with ENG canal paresis definitions ranging from greater than 13% to greater than 30% difference. Additionally, the methods (active vs. passive) and nature of HSN (e.g., >3 beats, >5 beats, >2.5 degrees/sec, >6 degrees/sec) differed

across these different studies, which points out the variable nature of definitions used for HSN. Optokinetic and smooth pursuit abnormalities, when present, are reliable signs of brainstem compression (140). Berrettini (83) found a higher frequency of central findings in tumors greater than 3 cm.

Table 11 Incidence of Diminished Caloric Response by ENG Study Linthicum, 1979 (142) Haapaniemi, 2000 (144) Berrettini, 1996 (83) Naessens, 1996 (143) Godey, 1998 (84)

Small 43% Intracanalicular 55% Small tumors (3 cm) 16/16

114

Gidley

Caloric testing has been examined as a predictor of hearing preservation. The horizontal semicircular canal is stimulated through caloric testing, and thus an insight is gained regarding the superior vestibular nerve (144). Small superior vestibular nerve tumors have a more favorable prognosis for hearing preservation (77,113,149). Thus, it is reasoned that patients with reduced or absent caloric responses have a better chance of hearing preservation, since the superior vestibular nerve is involved. However, in practice, caloric results are not so clear cut. Linthicum (142) showed that 97.2% of superior nerve tumors had a caloric weakness, while only 60% of inferior vestibular nerve tumors had a caloric weakness. Holsinger et al. (150) examined 47 AN cases with planned hearing preservation. Their overall rate of measurable hearing postoperatively was 60%. ENG was obtained on 36 patients. Twenty-five patients demonstrated a significant unilateral weakness and measurable hearing was preserved in 14 (56%). Eleven patients had no caloric weakness and five had hearing preservation (45%). Brackmann et al. (77) published their series of 333 patients with tumors less than 2 cm considered candidates for hearing preservation. ENG was performed in 261 patients: 49% had “normal reduced response” (i.e., ≤25% weakness) and 51% had a reduced vestibular response (i.e., >25% weakness). They found similar rates of reduced responses across all hearing categories and that no significant difference existed between the preserved hearing groups and the no measurable hearing group (77). Despite its lack of sensitivity and its inability to discern superior from inferior nerve of origin, ENG might be helpful in identifying patients preoperatively that will have prolonged imbalance postoperatively. Driscoll et al. (151) found that central signs seen on ENG in AN patients portended a higher incidence of persistent (>3 mo) disequilibrium than those without central signs. Age greater than 55.5 years, female gender, constant preoperative disequilibrium present for >3.5 months were also associated with prolonged postoperative disequilibrium in their study.

Other Tumors The literature regarding ENG findings in meningioma is scant compared with that for ANs. Baguley et al. (100) compared the results of 18 of their patients with the tabulated results of caloric testing performed in four previous studies. Overall, abnormal caloric results were found in 55/67 (82%) patients; and the incidence of abnormal ABR ranged from 66% to 95% among the five studies. Marangos et al. (110) examined 309 CPA tumors, including 17 meningiomas, found on either CT or MRI. In the four meningiomas with normal ABR, ENG was normal in three. The article does not reveal the ENG results of the other 13 cases of meningiomas.

Rotatory Chair Testing Rotatory chair testing uses a computer controlled rotational stimulus of the horizontal VOR (usually from 0.01–0.64 Hz). Rotational stimuli produce a reflex, slow eye movement in the opposite direction of rotation with a rapid corrective saccade contralaterally (147). Gain, phase, asymmetry, and failure of visual fixation can be calculated at each test frequency and compared to age-specific normative data. The definition of abnormal findings varies from center to center, but a generally accepted rule is an abnormality of gain, phase, or symmetry seen in two frequencies (147). As with most balance tests, a “gold standard” is lacking with which to compare sensitivity and specificity. Most

Table 12 Test Conditions in Computerized Dynamic Posturography Sensory Organization Test Test 1 2 3 4 5 6

Condition Eyes open, fixed support Eyes closed, fixed support Visual surround referenced to sway, fixed support Eyes open, force plate referenced to sway Eyes closed, force plate referenced to sway Force plate and visual surround referenced to sway

reports on dizziness evaluation rely on history and physical findings to indicate “normal” and “abnormal”. Since the examining physicians are also the interpreting physicians of the balance test, there are no blinded comparison reviews. Rotary chair has been compared with ENG for sensitivity and specificity in identifying patients with vestibulopathy. In this regard, rotary chair has a higher sensitivity for peripheral vestibular pathology than ENG, but the specificity of ENG is higher than rotary chair (147).

Computerized Dynamic Posturography Computerized dynamic posturography (CDP) has been clinically available since 1986 (152). This test has two distinct parts: motor control test (MCT) and somatosensory organization test (SOT). MCT is a technique used to measure the functional ability of the subject to create adequate motor responses to changes in the pitch plane (58). Three trials are made of small, medium, large forward and backward movement and the latency, amplitude and symmetry of neuromuscular response to this movement (152). Electromyography (EMG) and biomechanical measures can be used to track the patient’s responses to each balance perturbation. The commercially available Neurocom MCT uses a strain gauge in the support surface, and not EMG, to track ankle-torque corrections during balance-correction. Normal latencies range between 130 and 160 msec for medium and large perturbations. The SOT measures the relative contributions from the vestibular, proprioceptive, and visual systems to maintain balance. This test uses a force platform, which can be stable or referenced to sway (move in a horizontal plane or pitch back and forth), and a visual surround, which can be stationary or referenced to sway, to determine the relative importance of visual, somatosensory, and vestibular input (153,154). Six different testing situations are created by combinations of stationary or moving force plate and stationary or moving visual surrounds (Table 12). The computer can change the position of the force plate or visual surround so that it remains in the same position relative to the patient’s sway (“sway referenced”). The interested reader is encouraged to read Allum and Sheperd’s (58) excellent review of this topic. The most commonly recognized pattern with vestibular lesions is abnormalities in situations five and six. Although CDP does not help in localizing a lesion, it does provide a functional measure of a patient’s ability to use properly the various input systems to maintain balance (153–155). Sensitivity of CDP to pick up vestibular pathology, in comparison to ENG, has been evaluated (141); however, these two modalities provide different types of information. Since not all imbalances are due to a vestibular lesion, CDP might certainly be abnormal in a patient with a normal ENG. Nonetheless, CDP gives information that is helpful regarding the functional status of the patient and might help to tailor a rehabilitation program for that patient (152).


Levine et al. (156) used preoperative CDP to determine the nerve of origin for ANs less than 1.5 cm. In a small series, they found that patients with an inferior vestibular nerve tumor had abnormalities on SOT in conditions five and six, while patients whose tumors were from the superior vestibular nerve had normal CDP findings. Bergson and Sataloff (157) examined 21 patients with AN using CDP. They found abnormal test results in 81%, usually in conditions five and six. They found no correlation between the presence or severity of preoperative CDP results and postoperative balance function. Similar findings were reported by El-Kashlan et al. (155). Collins et al. (158) examined changes in balance following AN resection using balance posturography. They found patterns of abnormal sway and prolonged recovery times both pre- and postoperative, and these were most marked one month postoperatively. The limitations of CDP are that it does not provide lateralizing information or any information regarding cause (153). However, CDP does provide insight into how well patients can use their balance and how imbalance affects their activities of daily living.

Vestibular Evoked Potentials Vestibular evoked myogenic potentials (VEMPs) are shortlatency potentials recorded from surface electrodes over the tonically contracted SCM muscle evoked by high-level acoustic stimuli (159). The test subject is seated upright and asked to turn the head to the opposite side of the tested muscle. Surface electrodes are placed over the upper half of the SCM, while a ground electrode is placed on the forehead or sternum. Click stimuli sounds are delivered to the ipsilateral ear at intensities of 85 to 100 dB, and EMG is measured. The source of these responses is thought to be saccule. Matsuzaki et al. (160) described two patients with AN that had normal ABR results but abnormal VEMP. The tumors were 8 and 10 mm in size. They concluded that VEMP might be useful in the early diagnosis of ANs in patients with normal ABR. In a follow-up study, they review their experience with 87 AN patients, 79% of whom had decreased or absent VEMPs (161). Murofishi et al. (162) examined 21 patients with AN. They found abnormal or diminished VEMP ipsilateral to the tumor in 80% of patients, while all contralateral VEMPs were normal. Takeichi et al. (163) studied 18 patients with AN. They found diminished VEMP on the affected side in 13 patients (72%). They did not find any correlation with disequilibrium, spontaneous nystagmus, canal paresis, or pure-tone hearing. Tsutsumi et al. (164) examined 28 patients with AN and VEMP. They found no correlation between VEMP and caloric response, nerve of origin, audiometric threshold, or size of tumor. In a larger study of 170 patients with AN, Patko et al. (165), found abnormally low or absent VEMPs in 78.8% of patients. They did not find any correlation with horizontal canal weakness. Rauch has shown that VEMP has usefulness in the diagnosis of superior semicircular canal dehiscence syndrome and Meniere disease (166). Additionally, VEMP might provide some insight into brainstem pathology from stroke (167) or multiple sclerosis (168); however, at the time of this writing, their utility with AN or other skull base tumors is limited.

115

CONCLUSION The evaluation of patients with skull base tumors requires a rigorous history and physical examination. Careful evaluation of cranial nerve function is demanded. The examining physician should be aware of nontumor conditions that can mimic the findings of a skull base tumor. Judicious use of ancillary tests of neuro-otologic function helps in determining the extent of disease and plays an important role in deciding treatment methods. Some of these tests can be used to predict postoperative hearing and balance function.

REFERENCES 1. Bott L, Thumerelle C, Cuvellier JC, et al. Ataxia-telangiectasia: A review. Arch Pediatr. 2006;13(3):293–298. 2. Taylor AM, Byrd PJ. Molecular pathology of ataxia telangiectasia. J Clin Pathol. 2005;58(10):1009–1015. 3. Cross JH. Neurocutaneous syndromes and epilepsy—Issues in diagnosis and management. Epilepsia. 2005;46 Suppl 10:17–23. 4. Kreusel KM. Ophthalmological manifestations in VHL and NF 1: Pathological and diagnostic implications. Fam Cancer. 2005;4(1):43–47. 5. Patel NP, Mhatre AN, Lalwani AK. Molecular pathogenesis of skull base tumors. Otol Neurotol. 2004;25(4):636–643. 6. Xiao GH, Chernoff J, Testa JR. NF2: The wizardry of merlin. Genes Chromosomes Cancer. 2003;38(4):389–399. 7. Smith DS, Lee KK, Milczuk HA. Otolaryngologic manifestations of PHACE syndrome. Int J Pediatr Otorhinolaryngol. 2004;68(11):1445–1450. 8. Comi AM. Pathophysiology of Sturge-Weber syndrome. J Child Neurol. 2003;18(8):509–516. 9. Freedman SF, Amedee RG, Molony T. Neurotologic manifestations of von Hippel Lindau disease. Ear Nose Throat J. 1992;71(12):655–658. 10. Brandt T, Strupp M. General vestibular testing. Clin Neurophysiol. 2005;116(2):406–426. 11. Goldberg M. The control of gaze. In: Kandell E, Schwartz J, Jessell T, eds. Principles of Neural Science. New York: McGrawHill, 2000 782–800. 12. Glanze W, ed. Mosby’s Medical Dictionary. 3rd ed. St. Louis: C.V. Mosby, 1990. 13. Ross R. How to Examine: The Nervous System. New York: Medical Publishing Co., 1985. 14. Traccis S et al. Evaluating patients with vertigo: Bedside examination. Neurol Sci. 2004;25 Suppl 1:S16–S19. 15. Leigh R, Zee D. The Neurology of Eye Movements, 4th ed. New York: Oxford University Press, 2006. 16. Lee K. The vestibular system and its disorders. In: Lee K, ed. Essential Otolaryngology. New York: Medical Examination Publishing, 1995 93–124. 17. Nedzelski JM. Cerebellopontine angle tumors: Bilateral flocculus compression as cause of associated oculomotor abnormalities. Laryngoscope. 1983;93(10):1251–1260. 18. Longridge NS, Mallinson AI. The dynamic illegible E (DIE) test: A simple technique for assessing the ability of the vestibuloocular reflex to overcome vestibular pathology. J Otolaryngol. 1987;16(2):97–103. 19. Herdman SJ et al. Computerized dynamic visual acuity test in the assessment of vestibular deficits. Am J Otol. 1998;19(6):790– 796. 20. Dix MR, Hallpike CS. The pathology symptomatology and diagnosis of certain common disorders of the vestibular system. Proc R Soc Med. 1952;45(6):341–354. 21. Halmagyi GM, Curthoys IS. A clinical sign of canal paresis. Arch Neurol. 1988;45(7):737–739. 22. Schubert MC et al. Optimizing the sensitivity of the head thrust test for identifying vestibular hypofunction. Phys Ther. 2004;84(2):151–158.

116

Gidley

23. Foster CA et al. Functional loss of the horizontal doll’s eye reflex following unilateral vestibular lesions. Laryngoscope. 1994;104(4):473–478. 24. Harvey SA, Wood DJ. The oculocephalic response in the evaluation of the dizzy patient. Laryngoscope. 1996;106(1 Pt 1):6–9. 25. Harvey SA, Wood DJ, Feroah TR. Relationship of the head impulse test and head-shake nystagmus in reference to caloric testing. Am J Otol. 1997;18(2):207–213. 26. Beynon GJ, Jani P, Baguley DM. A clinical evaluation of head impulse testing. Clin Otolaryngol Allied Sci. 1998;23(2):117–122. 27. Humphriss RL, Baguley DM, Moffat DA. Head-shaking nystagmus in patients with a vestibular schwannoma. Clin Otolaryngol Allied Sci. 2003;28(6):514–519. 28. Shockley WW, Stucker FJ Jr. Squamous cell carcinoma of the external ear: A review of 75 cases. Otolaryngol Head Neck Surg. 1987;97(3):308–312. 29. Ross DA, Sasaki CT. Cancer of the Ear and Temporal Bone. In: Myers E., Suen J, ed. Cancer of the Head and Neck. Philadelphia, PA: W.B. Saunders, 1996; 586–597. 30. Clayman GL et al. Mortality risk from squamous cell skin cancer. J Clin Oncol. 2005;23(4):759–765. 31. Burkey JM et al. Clinical utility of the 512-Hz Rinne tuning fork test. Am J Otol. 1998;19(1):59–62. 32. Vikram KB, Naseeruddin K. Combined tuning fork tests in hearing loss: Explorative clinical study of the patterns. J Otolaryngol. 2004;33(4):227–234. 33. Poe DS, Abou-Halawa A, Abdel-Razek O. Analysis of the dysfunctional eustachian tube by video endoscopy. Otol Neurotol. 2001;22(5):590–595. 34. Meyerhoff WL, Anderson RG. Non-auditory presentations of cerebellopontine angle lesions. Laryngoscope. 1984;94(7):904– 906. 35. Nemzek WR. The trigeminal nerve. Top Magn Reson Imaging. 1996;8(3):132–154. 36. Saper C. Brain stem, reflexive behavior, and the cranial nerves. In: Kandell E, Schwartz J, Jessell T, eds. Principles of Neural Science. New York: McGraw-Hill, 2000 873–888. 37. Carpenter M. Core Text of Neuroanatomy. 3rd ed. Baltimore: Williams and Wilkins, 1985. 38. Doty RL, Mishra A. Olfaction and its alteration by nasal obstruction, rhinitis, and rhinosinusitis. Laryngoscope. 2001;111(3):409–423. 39. Rombaux P et al. Smell disorders in ENT clinic. B-Ent. 2005;Suppl 1:97–107; quiz 108–109. 40. Hummel T, Welge-Luessen A. Assessment of olfactory function. Adv Otorhinolaryngol. 2006;63:84–98. 41. Wrobel BB, Leopold DA. Clinical assessment of patients with smell and taste disorders. Otolaryngol Clin North Am. 2004;37(6):1127–1142. 42. Schiffman SS. Taste and smell in disease (first of two parts). N Engl J Med. 1983;308(21):1275–1279. 43. Doty RL et al. University of Pennsylvania Smell Identification Test: A rapid quantitative olfactory function test for the clinic. Laryngoscope. 1984;94(2 Pt 1):176–178. 44. Jackson CG et al. Facial paralysis of neoplastic origin: Diagnosis and management. Laryngoscope. 1980;90(10 Pt 1):1581–1595. 45. House JW, Brackmann DE. Facial nerve grading system. Otolaryngol Head Neck Surg. 1985;93(2):146–147. 46. Yeh S, Foroozan R. Orbital apex syndrome. Curr Opin Ophthalmol. 2004;15(6):490–498. 47. Zaffaroni M, Baldini SM, Ghezzi A. Cranial nerve, brainstem and cerebellar syndromes in the differential diagnosis of multiple sclerosis. Neurol Sci. 2001;22 Suppl 2:S74–S78. 48. Krasnianski M, Neudecker S, Zierz S. The Schmidt and Vernet classical syndrome. Alternating brain stem syndromes that do not exist?. Nervenarzt. 2003;74(12):1150–1154. 49. Johnston JL. Parasellar syndromes. Curr Neurol Neurosci Rep. 2002;2(5):423–431. 50. Schweinfurth JM, Johnson JT, Weissman J. Jugular foramen syndrome as a complication of metastatic melanoma. Am J Otolaryngol. 1993;14(3):168–174. 51. Paparounas K et al. Collet-Sicard syndrome disclosing glomus tumor of the skull base. Eur Neurol. 2003;49(2):103–105.

52. Chacon G, Alexandraki I, Palacio C. Collet-sicard syndrome: An uncommon manifestation of metastatic prostate cancer. South Med J. 2006;99(8):898–899. 53. Romberg M. Lehrbuch Der Nervenkrankheiten Des Menschen. Berlin: Duncker, 1846. 54. Weber PC, Cass SP. Clinical assessment of postural stability. Am J Otol. 1993;14(6):566–569. 55. Fukuda T. Statokinetic Reflexes in Equilibrium and Movement. Tokyo, Japan: University of Tokyo Press, 1983. 56. Hickey SA et al. Unterberger stepping test: A useful indicator of peripheral vestibular dysfunction? J Laryngol Otol. 1990;104(8):599–602. 57. Bonanni M, Newton R. Test-retest reliability of the Fukuda Stepping Test. Physiother Res Int. 1998;3(1):58–68. 58. Allum JH, Shepard NT. An overview of the clinical use of dynamic posturography in the differential diagnosis of balance disorders. J Vestib Res. 1999;9(4):223–252. 59. Kuipers-Upmeijer J, Oosterhuis HJ. Unterberger’s test not useful in testing of vestibular function. Ned Tijdschr Geneeskd. 1994;138(3):136–139. 60. Horak FB. Clinical measurement of postural control in adults. Phys Ther. 1987;67(12):1881–1885. 61. El-Kashlan HK, Eisenmann D, Kileny PR. Auditory brain stem response in small acoustic neuromas. Ear Hear. 2000;21(3):257– 262. 62. Gordon ML, Cohen NL. Efficacy of auditory brainstem response as a screening test for small acoustic neuromas. Am J Otol. 1995;16(2):136–139. 63. Cushing H. Tumors of the Nervus Acusticus and the Syndrome of the Cerebellopontine Angle. Philadelphia, PA: WB Saunders, 1917. 64. Johnson EW. Auditory test results in 110 surgically confirmed retrocochlear lesions. J Speech Hear Disord. 1965;30(4):307–317. 65. Glasscock ME 3rd, Levine SC, McKennan KX. The changing characteristics of acoustic neuroma patients over the last 10 years. Laryngoscope. 1987;97(10):1164–1167. 66. Jerger J. Clinical experience with impedance audiometry. Arch Otolaryngol. 1970;92(4):311–324. 67. Jerger J. Impedance terminology. Arch Otolaryngol. 1975;101(10):589–590. 68. Jerger J, Hayes D. Diagnostic speech audiometry. Arch Otolaryngol. 1977;103(4):216–222. 69. Hayes D, Jerger J. Impedance audiometry in otologic diagnosis. Otolaryngol Clin North Am. 1978;11(3):759–767. 70. Clemis JD, Mc Gee T. Brain stem electric response audiometry in the differential diagnosis of acoustic tumors. Laryngoscope. 1979;89(1):31–42. 71. Dix MR, Hallpike CS. The otoneurological diagnosis of tumours of the VIII nerve. Proc R Soc Med. 1958;51(11):889–896. 72. Walsh TE, Goodman A. Speech discrimination in central auditory lesions. Laryngoscope. 1955;65(1):1–8. 73. Hirsch A, Anderson H. Audiologic test results in 96 patients with tumours affecting the eighth nerve. A clinical study with emphasis on the early audiological diagnosis. Acta Otolaryngol Suppl. 1980;369:1–26. 74. Telischi F. An objective method of analyzing cochlear versus noncochlear patterns of distortion-product otoacoustic emissions in patients with acoustic neuromas. Laryngoscope. 2000;110(4):553–562. 75. Selesnick SH, Jackler RK. Clinical manifestations and audiologic diagnosis of acoustic neuromas. Otolaryngol Clin North Am. 1992;25(3):521–551. 76. Committee on Hearing and Equilibrium guidelines for the evaluation of hearing preservation in acoustic neuroma (vestibular schwannoma). American Academy of Otolaryngology-Head and Neck Surgery Foundation, INC. Otolaryngol Head Neck Surg. 1995;113(3):179–180. 77. Brackmann DE et al. Prognostic factors for hearing preservation in vestibular schwannoma surgery. Am J Otol. 2000;21(3):417– 24. 78. Anderson H, Barr B, Wedenberg E. The early detection of acoustic tumours by the stapedius reflex test. In: Sensorineural Hearing Loss. Ciba Foundation Symposium. 1970: p. 275–294.

Chapter 5: Head, Neck, and Neuro-otologic Assessment of Patients with Tumors of the Skull Base 79. Harner SG, Laws ER Jr. Clinical findings in patients with acoustic neurinoma. Mayo Clin Proc. 1983;58(11):721–728. 80. Moffat DA, Hardy DG. Early diagnosis and surgical management of acoustic neuroma: Is it cost effective? J R Soc Med. 1989;82(6):329–332. 81. Thomsen J, Tos M. Acoustic neuroma: Clinical aspects, audiovestibular assessment, diagnostic delay, and growth rate. Am J Otol. 1990;11(1):12–19. 82. Kanzaki J et al. Audiological findings in acoustic neuroma. Acta Otolaryngol Suppl. 1991;487:125–132. 83. Berrettini S et al. Acoustic neuroma: Correlations between morphology and otoneurological manifestations. J Neurol Sci. 1996;144(1-2):24–33. 84. Godey B et al. Sensitivity of auditory brainstem response in acoustic neuroma screening. Acta Otolaryngol. 1998;118(4):501– 504. 85. Moffat DA et al. Contralateral auditory brainstem response abnormalities in acoustic neuroma. J Laryngol Otol. 1989;103(9):835–838. 86. Beck HJ et al. Acoustic neuromas with normal pure tone hearing levels. Otolaryngol Head Neck Surg. 1986;94(1):96–103. 87. Roland PS et al. Normal hearing in patients with acoustic neuroma. South Med J. 1987;80(2):166–169. 88. Selesnick SH, Jackler RK, Pitts LW. The changing clinical presentation of acoustic tumors in the MRI era. Laryngoscope. 1993;103(4 Pt 1):431–436. 89. Acoustic Neuroma: NIH Consensus Statement. NIH, 1991:1–24. 90. Morrison A. Acoustic Neuroma. In: Management of Sensorineural Hearing Loss. London, England: Butterworths, 1975:46–79. 91. Telian SA, Kileny PR. Pitfalls in neurotologic diagnosis. Ear Hear. 1988;9(2):86–91. 92. Rosenberg SI. Natural history of acoustic neuromas. Laryngoscope. 2000;110(4):497–508. 93. Schuknecht H. Pathology of the Ear. Cambridge, MA: Harvard University Press, 1974. 94. Graamans K, Van Dijk JE, Janssen LW. Hearing deterioration in patients with a non-growing vestibular schwannoma. Acta Otolaryngol. 2003;123(1):51–54. 95. Saleh EA et al. Normal hearing in acoustic neuroma patients: A critical evaluation. Am J Otol. 1996;17(1):127–132. 96. Welling DB et al. Acoustic neuroma: A cost-effective approach. Otolaryngol Head Neck Surg. 1990;103(3):364–370. 97. Ruckenstein MJ et al. A prospective study of ABR and MRI in the screening for vestibular schwannomas. Am J Otol. 1996;17(2):317–320. 98. Obholzer RJ, Rea PA, Harcourt JP. Magnetic resonance imaging screening for vestibular schwannoma: Analysis of published protocols. J Laryngol Otol. 2004;118(5):329–332. 99. Robinette MS et al. Nonsurgical factors predictive of postoperative hearing for patients with vestibular schwannoma. Am J Otol. 1997;18(6):738-745. 100. Baguley DM et al. Audio-vestibular findings in meningioma of the cerebello-pontine angle: A retrospective review. J Laryngol Otol. 1997;111(11):1022–1026. 101. Doyle K, De la Cruz A. Cerebellopontine angle epidermoids: Results of surgical treatment. Skull Base Surgery. 1996;6:27– 33. 102. Quaranta N et al. Audiologic presentation of cerebellopontine angle cholesteatoma. J Otolaryngol. 2003;32(4):217–221. 103. Kaylie DM et al. Neurotologic management of intracranial epidermoid tumors. Laryngoscope. 2005;115(6):1082–1086. 104. Jewett DL, Williston JS. Auditory-evoked far fields averaged from the scalp of humans. Brain. 1971;94(4):681–696. 105. Brackmann DE, Selters WA. Brainstem electric audiometry: Acoustic neurinoma detection. Rev Laryngol Otol Rhinol (Bord). 1979;100(1-2):49–51. 106. Ferguson MA et al. Efficiency of tests used to screen for cerebello-pontine angle tumours: A prospective study. Br J Audiol. 1996;30(3):159–176. 107. Bu-Saba NY et al. Significance of false-positive auditory brainstem response: A clinical study. Am J Otol. 1994;15(2):233–236. 108. Hall J. Handbook of Auditory Evoked Responses. Boston, MA: Allyn and Bacon, 1992.

117

109. Josey AF, Jackson CG, Glasscock ME 3rd. Brainstem evoked response audiometry in confirmed eighth nerve tumors. Am J Otolaryngol. 1980;1(4):285–290. 110. Marangos N et al. Brainstem response in cerebellopontine angle tumors. Otol Neurotol. 2001;22(1):95–99. 111. Cueva RA. Auditory brainstem response versus magnetic resonance imaging for the evaluation of asymmetric sensorineural hearing loss. Laryngoscope. 2004;114(10):1686–1692. 112. Bauch CD, Rose DE, Harner SG. Auditory brain stem response results from 255 patients with suspected retrocochlear involvement. Ear Hear. 1982;3(2):83–86. 113. Wilson DF et al. The sensitivity of auditory brainstem response testing in small acoustic neuromas. Laryngoscope. 1992;102(9):961–964. 114. Dornhoffer JL, Helms J, Hoehmann DH. Presentation and diagnosis of small acoustic tumors. Otolaryngol Head Neck Surg. 1994;111(3 Pt 1):232–235. 115. Chandrasekhar SS, Brackmann DE, Devgan KK. Utility of auditory brainstem response audiometry in diagnosis of acoustic neuromas. Am J Otol. 1995;16(1):63–67. 116. Zappia JJ et al. Rethinking the use of auditory brainstem response in acoustic neuroma screening. Laryngoscope. 1997;107(10):1388–1392. 117. Burkey JM et al. Acoustic reflexes, auditory brainstem response, and MRI in the evaluation of acoustic neuromas. Laryngoscope. 1996;106(7):839–841. 118. Musiek FE, Kibbe K. Auditory brain stem response wave IVV abnormalities from the ear opposite large cerebellopontine lesions. Am J Otol. 1986;7(4):253–257. 119. Rupa V et al. Cost-effective initial screening for vestibular schwannoma: Auditory brainstem response or magnetic resonance imaging? Otolaryngol Head Neck Surg. 2003;128(6):823– 828. 120. Nadol JB Jr et al. Preservation of hearing and facial nerve function in resection of acoustic neuroma. Laryngoscope. 1992;102(10):1153–1158. 121. Matthies C, Samii M. Management of vestibular schwannomas (acoustic neuromas): The value of neurophysiology for intraoperative monitoring of auditory function in 200 cases. Neurosurgery. 1997;40(3):459–466; discussion 466–468. 122. Nadol JB Jr et al. Preservation of hearing in surgical removal of acoustic neuromas of the internal auditory canal and cerebellar pontine angle. Laryngoscope. 1987;97(11):1287–1294. 123. Kemink JL et al. Hearing preservation following suboccipital removal of acoustic neuromas. Laryngoscope. 1990;100(6):597– 602. 124. Stidham KR, Roberson JB Jr. Hearing improvement after middle fossa resection of vestibular schwannoma. Otol Neurotol. 2001;22(6):917–921. 125. House JW, Brackmann DE. Brainstem audiometry in neurotologic diagnosis. Arch Otolaryngol. 1979;105(6):305–309. 126. Laird FJ et al. Meningiomas of the cerebellopontine angle. Otolaryngol Head Neck Surg. 1985;93(2):163–167. 127. Granick MS et al. Cerebellopontine angle meningiomas: Clinical manifestations and diagnosis. Ann Otol Rhinol Laryngol. 1985;94(1 Pt 1):34–38. 128. Aiba T et al. Clinical characteristics of rare cerebellopontine angle tumours: Comparison with acoustic tumors. In: Proceedings of the First International Conference on Acoustic Neuroma. New York, NY: Kugler, 1992. 129. Hart MJ, Lillehei KO. Management of posterior cranial fossa meningiomas. Ann Otol Rhinol Laryngol. 1995;104(2):105– 116. 130. Don M et al. Successful detection of small acoustic tumors using the stacked derived-band auditory brain stem response amplitude. Am J Otol. 1997;18(5):608–621; discussion 682–685. 131. Philibert B et al. Stacked tone-burst-evoked auditory brainstem response (ABR): Preliminary findings. Int J Audiol. 2003;42(2):71–81. 132. Gold T. Hearing II. The physical basis of the action of the cochlea. Proc Royal Soc Britain. 1948;135:492–498. 133. Kemp DT. Stimulated acoustic emissions from within the human auditory system. J Acoust Soc Am. 1978;64(5):1386–1391.

118

Gidley

134. Hall J. Handbook of Otoacoustic Emissions. San Diego, CA: Singular, 2000. 135. Brownell WE et al. Evoked mechanical responses of isolated cochlear outer hair cells. Science. 1985;227(4683):194–196. 136. Kim AH et al. Transient evoked otoacoustic emissions pattern as a prognostic indicator for hearing preservation in acoustic neuroma surgery. Otol Neurotol. 2006;27(3):372–379. 137. Burns EM, Arehart KH, Campbell SL. Prevalence of spontaneous otoacoustic emissions in neonates. J Acoust Soc Am. 1992;91(3):1571–1575. 138. Ferber-Viart C et al. Is the presence of transient evoked otoacoustic emissions in ears with acoustic neuroma significant? Laryngoscope. 1998;108(4 Pt 1):605–609. 139. Jongkees LB, Philipszoon AJ. Electronystagmography. Acta Otolaryngol Suppl. 1964;189. 140. McGee ML. Electronystagmography in peripheral lesions. Ear Hear. 1986;7(3):167–175. 141. Amin M et al. A comparison of electronystagmography results with posturography findings from the BalanceTrak 500. Otol Neurotol. 2002;23(4):488–493. 142. Linthicum F, Khalessi M, Churchill D. Electronystagmographic caloric bithermal vestibular test (ENG): results in acoustic tumor cases. In: House and Luetje C eds. Acoustic Tumors. Baltimore: University Park Press, 1979 pp 237–240. 143. Naessens B et al. Re-evaluation of the ABR in the diagnosis of CPA tumors in the MRI-era. Acta Otorhinolaryngol Belg. 1996;50(2):99–102. 144. Haapaniemi JJ et al. Audiovestibular findings and location of an acoustic neuroma. Eur Arch Otorhinolaryngol. 2000;257(5):237– 241. 145. Bergenius J, Magnusson M. The relationship between caloric response, oculomotor dysfunction and size of cerebello-pontine angle tumours. Acta Otolaryngol. 1988;106(5-6):361–367. 146. Guyot JP et al. Diagnosis of cerebellopontine angle tumors. ORL J Otorhinolaryngol Relat Spec. 1992;54(3):139–143. 147. Arriaga MA, Chen DA, Cenci KA. Rotational chair (ROTO) instead of electronystagmography (ENG) as the primary vestibular test. Otolaryngol Head Neck Surg. 2005;133(3):329–333. 148. Asawavichiangianda S et al. Significance of head-shaking nystagmus in the evaluation of the dizzy patient. Acta Otolaryngol Suppl. 1999;540:27–33. 149. Shelton C et al. Acoustic tumor surgery. Prognostic factors in hearing conversation. Arch Otolaryngol Head Neck Surg. 1989;115(10):1213–1216. 150. Holsinger FC, Coker NJ, Jenkins HA. Hearing preservation in conservation surgery for vestibular schwannoma. Am J Otol. 2000;21(5):695–700. 151. Driscoll CL et al. Preoperative identification of patients at risk of developing persistent dysequilibrium after acoustic neuroma removal. Am J Otol. 1998;19(4):491–495. 152. Voorhees RL. The role of dynamic posturography in neurotologic diagnosis. Laryngoscope. 1989;99(10 Pt 1):995–1001. 153. Furman JM. Role of posturography in the management of vestibular patients. Otolaryngol Head Neck Surg. 1995;112(1):8–15.

154. Monsell EM et al. Computerized dynamic platform posturography. Otolaryngol Head Neck Surg. 1997;117(4):394– 398. 155. El-Kashlan HK et al. Disability from vestibular symptoms after acoustic neuroma resection. Am J Otol. 1998;19(1):104– 111. 156. Levine SC, Muckle RP, Anderson JH. Evaluation of patients with acoustic neuroma with dynamic posturography. Otolaryngol Head Neck Surg. 1993;109(3 Pt 1):392–398. 157. Bergson E, Sataloff RT. Preoperative computerized dynamic posturography as a prognostic indicator of balance function in patients with acoustic neuroma. Ear Nose Throat J. 2005;84(3):154–156. 158. Collins MM et al. Dynamic assessment of imbalance in acoustic neuroma patients by sway magnetometry. Clin Otolaryngol. 2000;25(6):570–576. 159. Akin FW, Murnane OD. Vestibular evoked myogenic potentials: Preliminary report. J Am Acad Audiol. 2001;12(9):445-452; quiz 491. 160. Matsuzaki M, Murofushi T, Mizuno M. Vestibular evoked myogenic potentials in acoustic tumor patients with normal auditory brainstem responses. Eur Arch Otorhinolaryngol. 1999;256(1):1– 4. 161. Ushio M et al. Click- and short tone burst-evoked myogenic potentials in cerebellopontine angle tumors. Acta Otolaryngol Suppl. 2001;545:133–135. 162. Murofushi T, Matsuzaki M, Mizuno M. Vestibular evoked myogenic potentials in patients with acoustic neuromas. Arch Otolaryngol Head Neck Surg. 1998;124(5):509–512. 163. Takeichi N et al. Vestibular evoked myogenic potential (VEMP) in patients with acoustic neuromas. Auris Nasus Larynx. 200128 Suppl:S39–S41. 164. Tsutsumi T et al. Prediction of the nerves of origin of vestibular schwannomas with vestibular evoked myogenic potentials. Am J Otol. 2000;21(5):712–715. 165. Patko T et al. Vestibular evoked myogenic potentials in patients suffering from an unilateral acoustic neuroma: A study of 170 patients. Clin Neurophysiol. 2003;114(7):1344–1350. 166. Rauch SD. Vestibular evoked myogenic potentials. Curr Opin Otolaryngol Head Neck Surg. 200614(5):299–304. 167. Chen CH, Young YH. Vestibular evoked myogenic potentials in brainstem stroke. Laryngoscope. 2003;113(6):990–993. 168. Alpini D et al. Vestibular evoked myogenic potentials in multiple sclerosis: Clinical and imaging correlations. Mult Scler. 2004;10(3):316–321. 169. Josey AF, Glasscock ME 3rd, Musiek FE. Correlation of ABR and medical imaging in patients with cerebellopontine angle tumors. Am J Otol. 1988;9 Suppl:12–16. 170. Weiss MH, Kisiel DL, Bhatia P. Predictive value of brainstem evoked response in the diagnosis of acoustic neuroma. Otolaryngol Head Neck Surg. 1990;103(4):583–585. 171. Noguchi Y, Komatsuzaki A, Nishida H. Cochlear microphonics for hearing preservation in vestibular schwannoma surgery. Laryngoscope. 1999;109(12):1982–1987.

6 Anesthesia and Intraoperative Monitoring of Patients with Tumors of the Skull Base Walter S. Jellish and Steven B. Edelstein

INTRODUCTION

for posterior fossa surgeries, with some modifications. The prone position is particularly useful for accessing lesions at or near midline and the fourth ventricle (3). Prone positioning is associated with numerous hemodynamic and respiratory changes that must be monitored closely. Significant V/Q mismatching is present in the prone position and access to the airway is compromised. Vigilance by the anesthesiologist is required when positioning the patient since dislodging of the endotracheal tube will require emergency re-intubation in suboptimal conditions. When transferring patients from supine to prone, significant hemodynamic changes can occur, secondarily to acute changes in preload from either abdominal compression or thoracic impedance. Cardiac dysrhythmias may occur from changes in preload as a result of this compression, thus electrocardiographic monitoring is essential (4). Several items, such as chest rolls and orthopedic frames, have been developed to help decrease these side effects. Some of the frames include: the Jackson spine table, simple chest and abdominal rolls composed of fabric or gel, and padded square frames that have a large opening for the abdominal contents. Each frame or support device is associated with its own series of risks and benefits that is beyond the scope of this article. Other than hemodynamic and respiratory problems associated with the prone position, if a slight head-up position is used, the patient is at risk for venous air embolism (VAE). VAE is a significant concern during skull base procedures and this will be discussed in further detail later in this chapter. Prolonged prone positioning has also been associated with significant facial edema, orbital/facial swelling, central venous retinal thrombosis, and posterior ischemic optic neuropathy—a disastrous condition that can result in permanent blindness.

Many issues surround the administration of anesthesia for patients undergoing surgery of the skull base. Not only are there the difficulties surrounding exposure of deeply seated anatomic structures, but there are also issues regarding positioning, prevention of iatrogenic nerve injury, and blood loss. This chapter describes many of the issues surrounding the delivery of anesthesia for skull base surgery and will touch on some of the neurophysiological monitors that can be used to improve the outcome. The review of this subjective matter is by no means exhaustive, but is meant to give some insight into the multiple and complex problems faced by the operative team during these procedures.

POSITIONING One of the challenges of skull base surgery has to do with positioning. The exact approach will depend on patient’s anatomy, clinical status, and tumor size. These approaches relate to the bone routes the surgeon will take in order to reach the neoplasm. These approaches include craniofacial, orbitocranial, infratemporal, and suboccipital (transcondylar). Lateral approaches include retrosigmoid, translabyrinthine, and orbitocranial zygomatic (1). Each of these approaches is associated with particular positions that have specific morbidities. Some of these morbidities and concerns will be discussed in further detail.

Supine Position Anterior skull base lesions can be quite difficult to surgically approach, though the physical position of the patient is that of supine. Supine position is, for the anesthesiologist, the easiest one to manage. Access to the airway is simple, though the patient may be rotated 180 degrees. There are limited changes in the patient’s hemodynamic profile, but there are significant changes in the pulmonary system, especially those related to diaphragmatic elevation. This altered diaphragmatic elevation ultimately promotes atelectasis and ventilation-to-perfusion (V/Q) mismatching. Posterior fossa lesions can also be accessed via the supine position, but require the head be laterally rotated and flexed. This flexion is sometimes impossible, especially in the elderly population and may be associated with venous obstruction (2).

Park-Bench Position The park-bench position (lateral oblique position) is commonly used in skull base procedures involving the posterior fossa. It is a semiprone lateral position with the head flexed and slightly elevated—5 to 10 degrees. It is quicker than the true prone and allows both lateral and midline approaches to the posterior fossa (3). The flexion of the neck may impinge on the venous circulation and an obstruction may not always be recognized by examination of the external position (5). This position has less hemodynamic and respiratory effects than the full prone position, but the neck flexion may be associated with brachial plexus injuries if care is not taken when final position is achieved. Extreme flexion may compromise spinal cord perfusion and has been associated with quadriplegia (5).

Prone Position The prone position is one of the most frequently used positions for spine surgery, though it is sometimes used 119

120

Jellish and Edelstein

Lateral Position This position is frequently used for intracerebellar procedures involving lateral or cerebellar hemispheric lesions, lesions of the clivus, petrous ridge, anterior and lateral foramen magnum. Typically the retromastoid, transtentorial, and transcondylar surgical approaches require this position. Again, there are some significant hemodynamic and pulmonary implications to the position. Significant V/Q mismatching takes place, though to a much less degree than the prone position. Some of the disadvantages of this position include the potential for lateral popliteal nerve palsy in the dependent leg, compression of the inferior shoulder and axillary structures, and the superior shoulder being in the line of sight of the surgeon (3). As such, it is important that the anesthesiologist confirms that the final position of the patient is appropriate. Devices such as axillary rolls (designed to decompress dependent axillary neurovascular structures) must be correctly placed and leg positions be padded.

Sitting Position Of all positions used by surgeons involved with skull base procedures, the sitting position has undergone the most scrutiny. The theoretical advantage of the sitting position is that it allows for improved cerebral relaxation and promotes gravity drainage of blood and cerebral spinal fluid (7). However, the complications are numerous and include hemodynamic instability, VAE with the possibility of paradoxical air embolism, pneumocephalus, quadriplegia (especially in the presence of extreme neck flexion), and compressive neuropathy (8,9). Another potential complication includes inadequate cerebral perfusion after assuming the sitting position, but may be balanced by a reduction in cerebral rate of metabolic oxygen consumption (10). Early physiological studies reveal significant changes when a patient assumes the upright position. A change in cardiac and systemic vascular resistance is known to increase 19% and 10%, respectively. In addition, stroke volume and cardiac index may decrease as much as 21% and 10%, respectively (11). In the presence of inhalational anesthetic agents, arterial hypotension may be profound, especially since these agents may cause vasodilation, venous pooling in the lower extremities, and dose-dependent cardiovascular depression (12). It is essential that the anesthesiologist carefully monitor systemic blood pressure as the patient assumes the sitting position. Interventions to maintain blood pressure may include fluid administration, decreasing inhalational agent concentration, compression of distal extremities, and short-term infusions of phenylephrine or other vasoactive drugs. There is a perceived advantage of the sitting position from the point of view of the respiratory system. In this position, access to the chest wall and the airway is obtained, while ventilation is unimpeded because diaphragmatic excursion is greater than in the horizontal position and consequently airway pressure is lower (13). It is important to keep in mind these potential complications when performing skull base procedures in sitting positions. A vital role for the anesthesiologist who is taking care of the sitting patient is to assure that the entire care team participates in the careful positioning of the patient. Goals are to avoid large changes in blood pressure, impairment of ventilation, and avoidance of potential nerve compression and entrapment. Recent reviews have gone as far as to list contraindications for assuming this position, including advanced age, hypertonia, chronic obstructive lung disease, and diagnosed patent foramen ovale (13).

Table 1 Four Grades of Intensity of Venous Air Embolus (VAE) Grade I: characteristic changes in Doppler sounds Grade II: changes in the Doppler sound plus fall of end-expiratory CO2 concentration by more than 0.4% Grade III: changes in Doppler sounds, fall in end-expiratory CO2 concentration, plus aspiration of air through the atrial catheter Grade IV: combination of above signs with arterial hypotension over 20% and/or arrhythmia or other pathological ECG changes Source: From Ref. 14

COMPLICATIONS DURING SKULL BASE PROCEDURES Venous Air Embolism VAE is one of the complications that concern most physicians during skull base surgery. Since skull base procedures commonly deal with venous structures that do not collapse and typically are held open by a bone, it is easy to see that venous air can be entrained via one or several of these open venous plexuses. The occurrence of VAE is especially concerning in patients with patent foramen ovale (present in 10–30% of the population), in which the potential for paradoxical venous air embolus is high. The concern is that changes in hemodynamics may result in right atrial pressures rising higher than left atrial pressures, leading to significant systemic complications such as cerebral infarction. Matjasko and colleagues have described four grades of VAE seen during sitting craniotomies. These grades (I–IV) relate to changes in precordial Doppler sounds, changes in end-tidal carbon dioxide (ETCO2 ) levels, the ability to aspirate air, and the presence of hemodynamic instability (Table 1) (14). Factors that play a role in the development of VAE include the pressure gradient between the surgical field and the right atrium, the surgical technique, and the amount of air entrained. In addition, many neurosurgical patients are hypovolemic, which reduces central venous pressure and further increases the risk of a VAE. Although frequently detected [there was a 72% incidence of VAE detected by transesophageal echocardiography (TEE) in one study (15)], the overall morbidity of VAE is currently low at less than 0.36%. Monitoring for VAE has been extensively described and usually consists of one or more of the following devices (in descending order of sensitivity): TEE, precordial Doppler (in the right third to sixth intercostal space), pulmonary artery pressure, ETCO2 /end-tidal nitrogen, right atrial pressure, electrocardiography (ECG), and esophageal stethoscope (13). Although many clinicians have used a transesophageal echo for detection of VAE because of increased sensitivity, this increase in sensitivity produces too many false positives reducing the specificity as a monitor to detect air entrainment (16). In addition, TEE is expensive and requires trained personnel to be available for interpretation. TEE has also been associated with vocal cord paralysis from recurrent laryngeal nerve palsy after prolonged use. Recommended therapeutic measures in the event of air entrainment include bilateral compression of the jugular veins, having the surgeon flood the wound with saline, discontinuation of nitrous oxide, aspiration of air from a central catheter, and downward tilt of the surgical table (17). The purpose of the atrial catheter is to allow efficient aspiration of trapped air. As such, it is essential that the right atrial catheter be multiorificed in nature. The catheters can be difficult to place, since it is necessary that the catheter tip be in the right atrium, a position that is hard to confirm unless one uses radiographic imaging or ECG tracing (biphasic P-wave morphology). A recent prospective trial regarding the placement

Chapter 6: Patients with Tumors of the Skull Base

of right atrial air aspiration catheters noted that the intravenous ECG P-wave morphology that correlated with the right atrial superior vena cava junction, identified by TEE, was the largest monophasic negative P wave without any biphasic component (18). We prefer the use of an angiographic catheter (Swan-Ganz Angiographic Catheter, Baxter Healthcare Corp, Irvine, CA). This multiorifice catheter is placed in a similar fashion to that of a Swan-Ganz and requires the use of a small pilot balloon at the tip to float the catheter into the right ventricle. Once an right ventricular (RV) waveform is observed, the catheter is withdrawn 1 to 2 cm until the trace disappears. This places the tip in the proper position in the right atrium. Another common practice is changing the position of the patient to left lateral recumbent; however, this has been refuted by Geissler and colleagues (19), who noted that body position had no effect on hemodynamics, nor did the occurrence of right heart failure appear to be related to outflow obstruction. Hypotension and a decrease in coronary perfusion pressure appeared to play more of a role in explaining the cardiovascular effects of air emboli. The role of positive end-expiratory pressure (PEEP) has also come into question. Schmitt and colleagues (15) noted that when PEEP was released, there was a significant occurrence of air emboli as documented by TEE. The group postulates that a sudden decrease of moderate PEEP might decrease right atrial pressure and subsequently increase venous return from cerebral veins. This would result in an increase in air entrainment and possibly an increase in the detectable number of VAE.

Pneumocephalus Another complication associated with skull base surgery is pneumocephalus, defined as the presence of intracranial air. In a retrospective review that compared sitting, park-bench, and prone positioning, 100% of the patients in sitting, 73% of the patients in park-bench, and 57% of the patients in prone positioning had evidence of intraventricular air (20). This has been attributed to the large amount of cerebrospinal fluid (CSF) drained due to gravitational effect. A patient in the sitting position is subject to the effects of gravity more than other positions. Ultimately, more CSF is drained, leading to the high incidence of pneumocephalus seen in the sitting patient. Not every case of intracranial air results in tension pneumocephalus. Historically, approximately 3% of sitting posterior fossa cases are noted to have developed tension pneumocephalus (8). Postoperative care and length of surgery play a role in the development of tension pneumocephalus; however, single contributing factors such as preexisting ventriculo-peritoneal shunts, the utilization of nitrous oxide, or intraoperative diuretics do not appear to play a solitary role (20).

Macroglossia/Facial Swelling A rare, but potentially catastrophic complication of skull base surgeries includes the development of macroglossia. Several case reports describe the occurrence of macroglossia in the immediate postoperative period, but the overall incidence is around 1% (21). Venous drainage of the face, tongue, larynx, and orbits enters the internal jugular system, which, when the neck is maximally flexed, may kink and lead to partial or complete obstruction of the system. In the worse case scenario, this may lead to thrombosis of the internal jugular vein (22). Other theories regarding the etiology of macroglossia also exist, including arterial compression, a neurogenic event,

121

reperfusion injury, and impaired lymphatic drainage (5,23). Lam and colleagues (5) believe that there is a significant role for reperfusion injury since many of the cases are not associated with cerebral swelling and edema, a condition one would expect if obstruction of venous drainage of the internal jugular system was present. The risk factors of obesity, neck flexion, local compression, and long surgical duration have been identified and should be kept in mind when patients present for skull base procedures. Regardless of the specific etiology of macroglossia, careful positioning of the head and neck is essential. As a rule of thumb, the authors assure a space of approximately 2 fingerbreadths between the mandible and the clavicle to prevent venous occlusion.

Cerebrovascular Complications Fortunately, cerebrovascular accidents and complications are rare during skull base procedures. Injury to the carotid artery is one of the most feared complications and may lead to stroke and other brain injuries. Repair of the carotid may be required and may include: saphenous vein bypass graft from the extracranial carotid artery to the petrous carotid artery and superficial temporal to middle cerebral artery bypass (26). Usually, preoperatively, when the carotid artery is affected by tumor, a balloon-occlusion test will be performed. This will help to identify those patients who would tolerate the sacrifice of the carotid artery if it became necessary during the procedure. “Blowout injury” is another carotid vascular complication that may occur. This situation is caused when the carotid artery is inadvertently lacerated as a bone spur is being removed. Blood loss may be brisk at the time of the injury and interventions include packing and urgent angiography with balloon occlusion (27). Delayed blowout injury can also occur if the carotid artery is exposed in the nasopharynx; however, this potentially can be prevented with muscle flap coverage (26). Vasospasm has also been reported and may result in stroke. Vasospasm tends to be seen in younger patients and felt to be due to a myogenic reaction in the vessel wall. It may result from arterial contact with fresh blood or arterial traction. Treatment usually consists of topical vasodilators, such as papaverine, though systemic drugs may have a role (28).

Arrhythmias Since many skull base procedures are in the area of the trigeminal and vagus nerves, as well as the brain stem, arrhythmias during the surgical procedure are common. Direct stimulation of the vagus may lead to negative chronotropy and inotropy manifested as sinus bradycardia, bradycardia terminating in asystole, asystole with no bradycardia, and arterial hypotension. When the trigeminal nerve is involved, sensory nerve endings of the trigeminal send signals to the sensory nuclei at the Gasserian ganglion. These signals ultimately continue along the short internuncial nerve fibers to connect with the efferent pathway in the motor nuclei of the vagus nerve. Again, the effects of stimulation of the cardioinhibitory efferent fibers of the vagus are seen (29). In this situation, the anesthesiologist should alert the surgeon and may request that the surgeon release traction or may choose pharmacologic intervention with a vagolytic substance, such as glycopyrrolate or atropine. Atropine, due to its quick onset, may be the drug of choice; however, the duration of action is shorter than that of glycopyrrolate. Over time, the reflex tends to decrease in its intensity. If the skull base surgery involves resection of a glomus jugulare tumor, one needs to be aware if the tumor is

122

Jellish and Edelstein

catecholamine secreting. Glomus jugulare tumors may be considered arteriovenous malformations, may be giant in size, and may be associated with multiple paragangliomas or adrenal tumors. The incidence of catecholamine secretion is around 4% and can be associated with tachyarrhythmias, hypertension, sweating, myocardial infarction, and cardiovascular collapse, which reflects catecholamine excess. It is important for the anesthesiologist to be prepared to treat any hypertensive crises arising from manipulation of a catecholamine secreting glomus jugulare tumor. Appropriate invasive arterial monitoring is essential and utilization of fast- and short-acting vasodilators, such as nitroglycerin, nitroprusside, and to a lesser extent phentolamine. As seen in the surgical cases involving resection of catecholaminesecreting pheochromocytomas, significant hypotension may occur once the output of norepinephrine has been interrupted. As such, short-term utilization of direct arterial vasopressors (such as phenylephrine or norepinephrine) may be needed to maintain adequate cerebral and coronary perfusion pressures. The resection of glomus jugulare tumors may also result in new cranial nerve injuries of up to 7.1%, with the highest being cranial nerve VII and IX (30).

Blood Loss Blood loss during resection of skull base tumors may be significant and the appropriate preoperative measures for such an event are essential. Many of the tumors are highly vascular in nature and preoperative assessment is necessary to delineate involvement of the cavernous sinus and jugular bulb. Of particular note, meningiomas have been shown to produce tissue plasminogen activator and may lead to increased fibrinolysis during resection (31). Other skull base lesions, known for their vascularity, include jugular paragangliomas, vagal paragangliomas, and nasopharyngeal angiofibromas. Preoperative embolization of feeding vessels may lead to decrease blood loss via a decrease in blood flow and pressure through the tumor (26). Maintenance of normovolemia is essential in skull base operations. When to transfuse will depend on the patient’s coexisting disease processes and is always indicated when there are signs of inadequate oxygen delivery to the tissues. Numerous blood conservation techniques are available including utilization of cell salvage devices in the presence of benign tumors, induced hypotension, acute normovolemic hemodilution, and antifibrinolytic therapy. Risks and benefits of each of the conservation techniques must be assessed prior to institution of therapy.

Peripheral Nerve/Cranial Nerve Injuries Protection from peripheral nerve injuries is of major concern during skull base procedures. The peripheral nerves are at particular risk given the multitude of positions assumed during surgery. All positions from supine to seated have been at one time or another associated with nerve injuries. Varying degrees of injury have been described and the Seddon Classification describes three broad classifications of nerve injury. These include neurapraxia, axonotmesis, and neurotmesis. Neurapraxia is a mild insult that results in conduction failure across an affected segment. This is reversible and tends to be the type of injury most seen during surgical procedures. Axonotmesis is when the axon is physically disrupted but the epineurium and perineurium are preserved. Recovery depends on the speed of neural regeneration. The worst injuries are those in which neurotmesis has taken place. In this situation, there is a complete disruption of the nerve and sup-

port structures and the prognosis for recovery is exceedingly poor (32). Although brachial plexus injuries have been described and felt to be secondary to cervical contralateral flexion rotation and lateral rotation of the shoulder and fixation of the shoulder girdle in a neutral position (33), other injuries have been noted including common peroneal nerve injuries leading to footdrop (34). Ulnar nerve injuries may also occur, but the complete etiology of their occurrence has yet to be determined. The role of extension, rotation, obesity, preexisting disease states, such as diabetes mellitus, have been mentioned as possible contributing factors for ulnar nerve injury. Whatever the etiology, meticulous documentation of appropriate intraoperative padding and the awareness of the existence of preexisting neurological defects is essential when taking care of these patients. Peripheral nerves are not the only nerves at risk during skull base procedures. It is well known that cranial nerve injuries are possible, especially when the operation is close to nerve origin. As mentioned earlier, anosmia is a complication of anterior craniofacial resections and may also lead to changes in taste. Transphenoidal approaches for tumor resection may lead to ocular and oculomotor nerve injuries. This may result in the loss of vision or the development of diplopia. Another major cranial nerve injury encountered during skull base procedures involves the vagus. Vagal injuries result in problems with dysphagia and the potential for aspiration. High vagal nerve paralysis can manifest as true vocal cord paralysis, palate paralysis, loss of pharyngeal muscle function, loss of pharyngeal sensation, and failure of cricopharyngeus to relax (26). Postoperative swallowing evaluation and laryngoscopy is indicated to assess these patients prior to allowing oral intake.

MONITORING AND ANESTHESIA Monitoring for skull base procedures depends on the type of procedure to be performed, the vascular and nerve structures that are involved, and the position in which the patient will be placed for the surgery. In all instances, the patient will have standard routine monitoring, such as ECG, noninvasive blood pressure, pulse oximetry, capnography, and temperature. Other monitors are added as the complexity, blood loss, surgical trauma, and comorbidities of the patient are also factored. Monitoring for skull base procedures must assure for adequate central nervous system perfusion, maintenance of cardiovascular stability and the integrity of the neurologic pathways that are being manipulated. Arterial line placement is the standard for most intracranial and extracranial procedures of the skull base. Invasive blood pressure monitoring provides for closer control of blood pressure and better titration of hyperventilation and blood pH. In addition, hemoglobin levels and electrolyte abnormalities can be easily detected by following serial arterial blood samples obtained from this catheter. Central access with either large-bore single lumen or double lumen catheters is contingent on the length of the surgery, anticipated blood loss, need for estimation of central vascular volume, and position of the patient. Depending on tumor type and location, neurophysiologic monitoring may also be employed to detect disruption of neural tracts or trauma to cranial nerves that may be near the site of surgery. Cranial nerve monitoring has markedly decreased postoperative morbidity after skull base surgery. Electromyography (EMG) of the facial nerve, vagus

Chapter 6: Patients with Tumors of the Skull Base

or trigeminal nerve is used during surgical resection to identify the nerve and preserve neurologic integrity, especially if the nerve is surrounded by a tumor (35). EMG of cranial nerves provides early recognition of surgical trauma, facilitates tumor excision, identifies nerve dysfunction, and confirms nerve function after the tumor is removed. The use of muscle relaxants in conjunction with facial nerve monitoring may be problematic. Muscle relaxants prevent movement and reduce the amount of anesthetic agents needed. At any given level of neuromuscular blockade, a facial muscle response is more resistant to neuromuscular blocking agents than a peripheral muscle (36). This is due to larger motor unit size and the increased number of neuromuscular junctions in the facial muscles. Several studies have demonstrated that neuromuscular blockade, titrated to a T1 of 25%, still allows adequate response from compound motor action potentials of facial muscles to adequately monitor nerve function (37). Nerve irritation and tumor infiltration, however, can lead to reduced or blocked conduction. Preoperative partial facial paralysis may make the monitoring difficult. In addition, external or mechanical noise artifacts could mask muscular contraction and if high-dose inhalational agents are used, muscle activity with nerve stimulation can be further reduced. If muscle relaxants must be used to facilitate surgical resection, alternative methods to monitor the facial nerve may be used. The first method stimulates the nerve at the stylomastoid foramen (38). Antidromic responses are recorded in the operative field. However, this method is awkward to use and does not provide the information obtained by continuous recording. Nerve action potentials can also be recorded at the stylomastoid junction; however, this technique is an evoked response. There is no audible feedback compared to EMG and no information concerning this method’s sensitivity in detecting injury. The final method that may be used to determine facial nerve integrity in the presence of muscle relaxants is the brain stem facial evoked response (39); this nerve monitoring method is based on cross auricular responses to sound that controls ear movement. The facial nerve response is recorded at the mastoid after sound stimulation of the contralateral ear. This technique is technically challenging because of the need for digital computer filtering. The best conditions for monitoring cranial nerve function would be with an anesthetic technique that avoids muscle relaxants. Newer inhalational agents such as sevoflurane and desflurane have low blood gas partition coefficients and reduced fat solubility. These agents produce a rapid induction and emergence with minimal accumulation of anesthetics even after prolonged anesthesia. In addition, short-acting opioids such as remifentanil are extremely potent and have minimal accumulation when given as an infusion. The half life is three to eight minutes and metabolism is produced by nonspecific esterases. The anesthetic technique that produces the best operating conditions for surgery and cranial nerve monitoring uses an anesthetic induction with propofol and fentanyl coupled with an intermediate acting muscle relaxant to facilitate intubation. Maintenance of anesthesia is provided with low-dose desflurane or sevoflurane administered in a 50:50 air/O2 mixture with a background infusion of either fentanyl 2 µg/kg/hr or remifentanil 0.25–0.35 µg/kg/min. No further muscle relaxants are administered as the initial dose is reversed by the time surgery begins. Brain stem auditory evoked potentials (BAEPs) and somatosensory evoked potential (SSEP) monitoring may also be used during resection of skull base tumors, especially if there is possible vascular compromise or ischemia due to temporary occlusion of vascular structures or manipulation

123

of the brain stem. Recording short-latency auditory responses by electrocochleography determines the status of the cochlea and nerves peripheral to the tumor while BAEPs monitor activity central to the tumor (40). Headphone stimulators are unacceptable in the operating room because of their size. Small electrodes are used, the most important of which is placed near the mastoid, on or in the ear. The reference electrode is placed at the top of the head and the ground electrode is placed on the forehead. In general, anesthetic affects on the brain stem auditory evoked response are not dramatic. Slow shifts may be seen as the concentration of inhalational agents increase. Since these recordings are of small amplitude, thousands of responses must be recorded to acquire an adequate average. Frequently, the responses are abnormal and smaller than normal due to the effects of the tumor. Also the time interval required to acquire sufficient responses may reduce the sensitivity of this technique in determining neural injury during tumor removal. Of primary importance for intraoperative monitoring are waves I, III, and V. The interpeak latency I–III provides information regarding the integrity of the peripheral component of the auditory pathway. Shortlatency BAEPs are usually resistant to both intravenous and inhalational agents. Increasing blood levels of barbiturates and ketamine will increase interpeak latency. Propofol however, given at 2 mg/kg/bolus followed by an infusion will increase the latency of I, III, and V waves by 2.5 cm Poor hearing: > 50 dB on pure tone audiogram, < 50% speech discrimination SRS: Sterotactic radiosurgery Retrosig: Retrosigmoid craniectomy Translab: Translabyrinthine craniectomy MRI: Magnectic resonanace imaging

Figure 7

Treatment algorithm for patients with acoustic neuromas.

394

Oh et al.

with younger patients. If total surgical excision cannot be achieved, patients are then offered stereotactic radiosurgery. In older patients (>70 years), small tumors are followed with serial MRI scans. For patients with small acoustic neuromas that demonstrate growth, stereotactic radiosurgery is offered. Older patients with large tumors are treated with surgery for tumor debulking followed by stereotactic radiosurgery. The management paradigm for patients with CPA meningiomas is similar. In general, patients with meningiomas in the CPA also undergo a preoperative preparation process similar to patients with acoustic neuromas. In several instances, large cranial base meningiomas that also happen to involve the CPA require staged surgical resection or debulking followed by stereotactic radiosurgery. Finally, CPA epidermoids are usually treated with a retrosigmoid approach for surgical excision. There has been no demonstrated role for radiosurgery or radiotherapy in the treatment of epidermoids. In many instances, epidermoids are difficult to remove because of their adherence to adjacent nerves and vessels. The use of constant electromyographic facial nerve monitoring is now accepted as standard practice for surgery of acoustic neuromas and other CPA tumors. Electrodes are placed in the ipsilateral orbicularis oculi and orbicularis oris for the detection of muscle action potentials in response to surgical manipulation or monopolar or bipolar electrical stimulation of the facial nerve (2). For maximal benefit from continuous facial nerve electromyographic monitoring, it is preferable that a nonmuscle relaxant anesthetic technique is used. Perioperatively, a Foley catheter and arterial line are inserted. Antibiotics are given prior to skin incision. Dexamethasone, furosemide, and mannitol are also given intravenously prior to exposure and opening of the dura.

sen for right-sided tumors to prevent the patient’s shoulder from getting in the way of the surgeon’s right hand. The left retrosigmoid approach is performed with the patient in the supine position and head turned towards the right. In this case, the patient does not need to be in the lateral position because patient’s shoulder will not be in the way of the operating surgeon’s right hand. For an approach from either side, the patient’s head is held in place with a Mayfield threepoint fixation device. The abdomen is prepared for possible harvest of a fat graft. However, an abdominal fat graft is generally not needed for closure unless the patient has had a previous surgical approach to the CPA.

Incision and Soft Tissue Dissection A curvilinear incision centered 1 to 2 cm medial to the mastoid process is made. Scalp flaps are then developed with monopolar electrocautery and then elevated with fishhooks and rubber bands. The suboccipital fascia and muscles are then incised in a hockey-stick fashion and carefully separated from their attachments to the bone using subperiosteal dissection and monopolar electrocautery (Fig. 8). The fascia and muscles are also reflected with fishhooks. This two-layer opening facilitates better closure and may have a role in decreasing incidence of postoperative CSF leaks. An emissary vein is usually exposed in the region just medial to the mastoid process. Bleeding from this vein is stopped with bone wax.

SURGICAL TECHNIQUE Tumors of the CPA are generally approached by either a suboccipital retrosigmoid or translabyrinthine approach. Since an acoustic neuroma is by far the most common tumor of the CPA, further discussion of surgical technique will focus on this tumor. The microsurgical removal of an acoustic neuroma can be achieved via a suboccipital retrosigmoid, translabyrinthine, or middle fossa approach. Both the suboccipital and translabyrinthine approaches have been used to remove acoustic neuromas of all sizes (32). The middle fossa approach is used by some for small tumors in the IAC when an attempt is being made to preserve hearing (32). When hearing preservation is an issue, the suboccipital retrosigmoid approach is generally preferred. General considerations for selecting an operative approach have already been reviewed in the “Preoperative Preparation” section of this chapter. The retrosigmoid and translabyrinthine approaches for microsurgical excision of acoustic neuromas are now presented.

Retrosigmoid Approach Positioning The semi-sitting, prone, supine-oblique, lateral decubitus, and lateral oblique positions have all been used for suboccipital retrosigmoid removal of acoustic neuromas (32,33). At our center, the right retrosigmoid approach is performed with the patient in a left lateral decubitus position. The legs, hips, and arms are carefully padded. This approach is cho-

Figure 8 Illustration of the hockey-stick incision and the two-layer soft tissue dissection for a retrosigmoid or translabyrinthine craniectomy.

395

Chapter 26: Tumors of the Cerebellopontine Angle

Trigeminal nerve

Vestibulocochlear nerve

Tumor Facial nerve

Superior vestibular nerve

Tumor

Cerebellum

Figure 9 Extended retrosigmoid approach with dura opened. By drilling the bony constraints overlying the sigmoid sinus, the sigmoid sinus can be mobilized and the dura reflected more anteriorly allowing better exposure of more ventrolaterally located structures.

Inferior vestibular nerve Facial nerve

Bony Dissection The bony dissection for this approach involves a standard suboccipital craniectomy. A burr hole is placed behind the mastoid. Using a combination of Leksell and Kerrison rongeurs, the burr hole is enlarged. In areas where the bone is too thick for the rongeurs, a high-speed air drill with a diamond burr is used to thin the bone. The craniectomy is taken as high as the lateral sinus and as anterior as the sigmoid sinus. For an extended retrosigmoid approach, the bone is carefully drilled off the sigmoid sinus, exposing only a small amount of mastoid air cells. By decompressing the bony constraints of the sinus in this manner, better exposure is obtained ventrolaterally (Fig. 9). Exposed mastoid air cells, such as in an extended retrosigmoid approach or a partial mastoidectomy, should be thoroughly sealed off with bone wax. At this point, the operating microscope is brought into the field.

Tumor Resection If the dura remains tense at this point, additional measures are taken to relieve the tension. This includes lowering the patient’s pCO2 , temporarily raising the head of the operating table, or administering medications to induce additional diuresis. Once the dura is adequately relaxed, it is opened under high magnification in a curvilinear fashion and reflected anteriorly. A retractor is then gently placed on the cerebellum until the cisterna magna is identified. The cisterna magna is then sharply opened, which leads to the copious escape of CSF and, invariably, a recession of the cerebellum away from the posterior surface of the petrous bone (34). At this point, the tumor should be easily visible. The operative field should now be inspected while the tumor is still covered with its arachnoid investments, as inspection may reveal a branch of the AICA or AICA itself lying in a looplike manner across the cisternal portion of tumor (Fig. 10). By sharply cutting the arachnoid investments, it is possible to deliver the AICA out of the way of tumor dissection. In most cases, the vestibulocochlear nerve should now be easily identified splaying out into the medial aspect of the tumor while the facial nerve courses over the rostral aspect of the tumor. There is a difference of opinion as to whether an acoustic neuroma should first be decompressed internally or

Branch of anterior inferior cerenbellar artery

Figure 10 A branch of the AICA or AICA itself lie in a loop-like manner across the cisternal portion of an acoustic neuroma, as demonstrated by this intraoperative photograph. It needs to be delivered out of the field for surgical resection to proceed safely.

whether the IAC should be opened first so as to identify the cochlear and facial nerves in the canal. Our position as well as the position of Ciric et al (34) is to decompress a larger tumor first to deflect the medial edge of the tumor away from the bifurcation of the eighth cranial nerve into its cochlear and vestibular components. The decompression of the tumor is accomplished with pituitary forceps. The cochlear nerve can be significantly attenuated in patients with large tumors and may be barely recognizable even under high magnification. Once it is clear that the tumor is originating from the superior vestibular nerve, this nerve can be divided. As the tumor is decompressed, the identity of the cochlear nerve becomes progressively clearer to a point at where a plane may be visualized between the cochlear nerve and the tumor capsule (Fig. 11). Planes can then be developed between the tumor and the facial and inferior vestibular nerves. In all cases, it is important to avoid stretching or putting tension on the cochlear and facial nerves to prevent avulsion of the fibers. Once the cochlear nerve is identified, it is followed towards the porus acousticus. After the dura overlying the porus acousticus is stripped off, a 3-mm diamond burr is used to drill off the porus acousticus for approximately 3 to 4 mm to expose the dura of the IAC and thus the lateral extent of tumor. Excessive drilling may result in inadvertent opening of the posterior semicircular canal or endolymphatic duct and compromise hearing. The cochlear and facial nerves are once

396

Oh et al.

Translabyrinthine Approach Positioning

Inferior vestibular Superior vestibular Facial nerve

Figure 11 As an acoustic neuroma is internally decompressed, the identity of the cochlear nerve becomes progressively clearer to a point at where a plane may be visualized between the cochlear nerve and the tumor capsule as demonstrated by this intraoperative photograph.

The patient is placed supine on the operating table. After successful induction of general anesthesia, the patient’s head is placed on a donut turned so that the side of the tumor is facing up. The abdomen, ear, and postauricular area are prepped and draped in a sterile fashion. Patient position for a left translabyrinthine approach is shown in Figure 14.

Incision and Soft Tissue Dissection A curvilinear incision is marked out that starts from just below the tip of the mastoid process, runs upward over the lateral surface of the mastoid and just behind the root of the pinna, and courses to a point about 2 cm above the tip of the pinna (Fig. 15). Following incision, scalp flaps are developed using monopolar electrocautery and elevated with fishhooks and rubber bands. A hockey-stick incision is then made with the monopolar cautery in the temporalis fascia and muscle. The muscle and fascia are then separated off the

again identified within the IAC (Fig. 12). Under high magnification, tumor is then sharply dissected off of the nerves in the IAC and removed piecemeal. Bipolar forceps with lowcurrent settings may be used within the tumor capsule for coagulation. Outside the tumor capsule, it should be used very sparingly or not at all so as not to interrupt blood supply to the facial and cochlear nerves.

Closure Once hemostasis is achieved, the dura is closed over a piece of R suturable DuraGen (Integra LifeSciences, Plainsboro, NJ) in a watertight fashion. This closure is then reinforced with TISSEEL fibrin glue (Baxter Healthcare Corp., Deerfield, IL). Dural closure is demonstrated in Figure 13(A) to 13(D). A second piece of DuraGen is then placed on top of the construct and R once again reinforced with fibrin glue. A piece of Gelfoam (Pharmacia and Upjohn, Kalamazoo, MI) is then placed in the epidural space, and the cranial defect is repaired using a titanium mesh screen. The wound is then thoroughly irrigated with antibiotic irrigation. The fascia and muscle layer is then closed in an interrupted manner using absorbable suture. The scalp flaps are then reapproximated with interrupted absorbable suture. Finally, the skin is closed with a running, interlocking nylon suture.

(A)

(B)

Figure 12 Retrosigmoid approach after drilling of the internal acoustic canal, demonstrating the facial and vestibulocochlear nerve complex.

Figure 13 This series of pictures demonstrates dural closure of a suboccipital retrosigmoid approach for acoustic neuroma removal. (A) A piece of DuraGen is placed beneath the dural flap. (B) Dura on both sides of the durotomy is sutured to the underlying DuraGen. This maneuver helps to create a better seal. (C) A completed dural closure over the piece of DuraGen is demonstrated. (D) A layer of TISSEEL (fibrin glue) is placed over the construct.


397

Figure 15 The incision to be made for a left translabyrinthine approach is marked out in this photograph. An incision is marked out that starts from just below the tip of the mastoid process, runs upward over the lateral surface of the mastoid and just behind the root of the pinna, and courses to a point about 2 cm above the tip of the pinna.

(C)

bone using subperiosteal dissection using monopolar electrocautery. The fascia and muscle are then elevated with fishhooks and rubber bands. A marking pen is then used to outline the approximate position of the sigmoid sinus (Fig. 16). This can be facilitated with the use of image guidance.

Bony Dissection A high-speed air drill with a cutting burr is then used to begin the bony dissection (Fig. 17). Anteriorly, the posterior wall of the external meatus is thinned. Superiorly, the dissection should expose the edge of the middle fossa dura and superior petrous sinus. The dissection is carried anteriorly above the external meatus as far as possible. The sigmoid (D)

Figure 13 (Continued.)

Asterion

Figure 14 Patient positioning for a left translabyrinthine approach is demonstrated.

Figure 16 In a translabyrinthine approach, the muscle and fascia are then separated off the bone using subperiosteal dissection using monopolar electrocautery. The fascia and muscle are then elevated with fishhooks and rubber bands. A marking pen is then used to outline the approximate position of the sigmoid sinus.

398

Oh et al.

Lateral canal

Jugular bulb Superior canal Posterior semicircular canal

Figure 17 A high-speed air drill with a cutting burr is then used to begin the bony dissection for a left translabyrinthine approach. (A)

sinus and roughly 1 cm of dura behind the sinus are exposed posteriorly. Inferiorly, the mastoid process is hollowed out. It is important to remove sufficient bone posteriorly and superiorly to improve access by retraction of the dura (2). With the aid of the operating microscope, the dissection is then deepened in the space between the middle fossa dura and superior margin of the meatus to open the mastoid antrum and atticus (Fig. 18). Following exposure of the incus and the head of the malleus, the incus is removed. The lateral semicircular canal is then identified on the medial wall of the epitympanic recess [Fig. 19(A) and 19(B)]. Dense bone marks the otic capsule surrounding the semicircular canals. The lateral semicircular canal is removed first. The superior canal is then drilled until the identification of the common crus, which can then be used to find the posterior canal. After completion of the labyrinthectomy, the bone covering the IAC is completely skeletonized over 270 degrees. The bone over the porus acousticus is then removed. The cavity resulting from this dissection is roughly pyramidal in shape, with its base at the cortical opening of the mastoid. The dura of the posterior fossa lies posteriorly. Dura of the middle fossa lies above, and the petrous bone, middle ear cavity, and descending facial nerve lie anteriorly.

Facial nerve

Canals opened

(B)

Figure 19 (A) Transmastoidal bone work prior to beginning of labyrinthectomy demonstrating the lateral, superior, and posterior semicircular canals. (B) Opening of the canals and the relationship of these structures to the facial nerve.

Tumor Resection

Sigmoid sinus

Antrum

Figure 18 Bony dissection of the sigmoid sinus and mastoid antrum are demonstrated. The sigmoid sinus is labeled in the photograph. The tip of the sucker is in the mastoid antrum.

To begin the next phase of the operation, the dura of the meatus and posterior fossa is opened. The extent of the incision in the posterior fossa is dependent on tumor size. Figure 20 shows a view of the surgical field immediately after the dura is opened. At this point, it is important to first identify the facial nerve in the internal auditory meatus. Once this is achieved, the tumor is then freed from its attachments to the arachnoid and dura at the porus, thus increasing the mobility of the entire mass. The arachnoid is then opened and a plane between the membrane and the surface of the tumor is established. Both the seventh and eighth nerves are then stimulated with a nerve stimulator to confirm their identities. The tumor is then gradually debulked. In nearly every case, it is not prudent to attempt tumor removal in one piece as the body of the tumor often obscures attachment to the facial nerve or internal auditory artery. By using the technique of internal tumor debulking, collapse of the tumor capsule is facilitated.


Anterior/earlobe Acoustic neuroma Dura

Cranial

Caudal

399

communication. Eustachian tube obliteration is performed by drilling out the facial recess and subsequently packing muscle into the facial recess, out the tubotympanum into the Eustachian tube. Thin strips of harvested abdominal fat graft are placed into the dural defect in the region of the IAC under direct facial nerve monitoring. A larger piece of fat is placed on top of this construct. A piece of titanium mesh is then fashioned to cover the cranial defect. After the fascia and muscle, scalp flaps, and skin are then each closed in separate layers, a sterile dressing is applied.

POSTOPERATIVE CARE

Cerebellum

Posterior

Figure 20 This is the surgical field from a left translabyrinthine approach right after the opening of the dura. The picture is labeled for purposes of orientation.

The collapse allows the dissection to be extended under direct vision around the far side of the tumor, where arteries, cranial nerves, and the brainstem lie. As the tumor is reduced in size, dissection proceeds along its upper, lower, anterior, and posterior poles. Figure 21 displays a view of the surgical field after some tumor debulking and dissection have occurred. After tumor removal is completed, hemostasis is obtained. However, bipolar electrocautery is kept to a minimum during this process.

The patient is then extubated in the operating room and immediately taken to the intensive care unit. If the facial nerve is anatomically intact at the end of the operation, the facial musculature should be examined as soon as the patient recovers from anesthesia. If there are no complications, invasive monitoring devices are removed on the first postoperative day and the patient is transferred out to the ward. Oral feedings are also started on this day. Over the next few days, the patient is mobilized. Once adequately mobile and no complications are detected, the patient is sent home.

COMPLICATIONS AND COMPLICATION AVOIDANCE Postoperative Hematoma Hematoma in the postoperative cavity is exceedingly rare at our institution as well as at most other high-volume centers. If the patient does not recover promptly from anesthesia, or if there is an unexpected neurologic deficit or delayed mental status deterioration, a CT scan must be immediately performed to rule-out a cerebellar hematoma. Prompt removal of a hematoma can lead to a dramatic recovery and can be lifesaving.

Hydrocephalus

Closure Once hemostasis is achieved, the edges of the dural opening into the posterior fossa are approximated. The most important point of the closure is to seal off the middle ear from CSF

Hydrocephalus is another rare complication of surgery of the CPA. At our center, we have had one case of postoperative hydrocephalus after removal of an acoustic neuroma. If the patient develops signs and symptoms of hydrocephalus or if a tense subgaleal fluid collection begins to accumulate, a CT scan is warranted. In most cases, hydrocephalus resolves spontaneously. Should hydrocephalus persist, a ventriculoperitoneal shunt is placed.

Porous acousticus

CSF Leak Trigeminal nerve Jugular bulb Temporal line

VIII

Tumor

Sigmoid

Figure 21 This is the surgical field from a left translabyrinthine approach after some tumor has been debulked. The fifth and eighth cranial nerves are clearly visible. The porus acousticus has been drilled away.

CSF leak is a complication of CPA surgery, regardless of the approach. It has been reported with a frequency of between 2% and 25% (35–37). Depending on the case, CSF leak may present as rhinorrhea or leakage from the wound. Although it is most commonly evident within a few days of surgery, it may be a delayed complication and may occur after the patient is discharged from the hospital. Upon discharge from the hospital, patients should be given specific instructions about recognizing and reporting CSF leaks. In most cases, placement of a lumbar drain is the primary treatment for CSF leak. Should the problem persist after the lumbar drain is placed, the patient may need to be taken back to surgery for reexploration and graft repacking.

Meningitis When there is postoperative fever along with headache and/or nuchal rigidity, the possibility of either bacterial or aseptic meningitis should be explored. In cases of epidermoid

400

Oh et al.

tumor excision, aseptic meningitis is more common. A CT with contrast or an MRI scan is performed to look for an area that might be harboring an infection. A lumbar puncture is then performed and the CSF sent for appropriate studies. The patient is then started on broad-spectrum, intravenous antibiotics. Subsequent patient management is guided by results from the CSF studies.

Vertigo Postoperatively, patients may experience symptoms of vertigo, however, this complication is usually transient. In the case of chronic or permanent vestibular nerve dysfunction, patients generally eventually accommodate with central compensation (37).

Facial Nerve Dysfunction As mentioned above, facial musculature should be examined immediately in patients with anatomically intact facial nerves as soon as the patient recovers from anesthesia. Should facial paralysis be noted, eye care is of great importance, especially if facial analgesia is present and the corneal reflex is absent. The patient should have ointment and artificial tears applied to the affected eye on a regular basis. Additionally, the patient should wear an eye chamber while sleeping for further eye protection. During follow-up, the patient is specifically examined for corneal ulcerations. If there is suggestion of any problems, immediate referral to an ophthalmologist is warranted. A severely damaged cornea may lead to incapacitating pain and even visual loss. Sampath et al (38) have summarized the results of facial nerve function from several large series of acoustic neuroma patients. The House–Brackmann facial nerve grading system is the standard system used to record facial nerve function (18). When facial paralysis does not recover, the patient may need to undergo a hypoglossal–facial anastomosis. Additionally, tarsorrhaphy and/or gold weight placement in the upper eyelid may need to be performed. REFERENCES 1. Rhoton AL Jr. The cerebellopontine angle and posterior fossa cranial nerves by the retrosigmoid approach. Neurosurgery. 2000;47(3)(Suppl):S93–129. 2. Kaye AH, Briggs RJS. Acoustic neurinoma (vestibular schwannoma). In: Kaye AH, Laws ER Jr, eds. Brain Tumors. 2nd ed. London, England: Churchill Livingstone, 2001:619–669. 3. Osborn AG, Rauschning W. Brain tumors and tumorlike masses: Classification and differential diagnosis. In: Osborn AG, ed. Diagnostic Neuroradiology. 1st ed. St. Louis, MO: Mosby,1994:401– 528. 4. Matsuno H, Rhoton AL Jr, Peace D. Microsurgical anatomy of the posterior fossa cisterns. Neurosurgery. 1988;23(1):58–80. 5. House WF. Translabyrinthine approach. In: House WF, Luetje CM, eds. Acoustic Tumors: II-Management. 1st ed. Baltimore, MD: University Park Press, 1979:43–87. 6. Rhoton AL Jr. Microsurgical anatomy of the brainstem surface facing an acoustic neuroma. Surg Neurol. 1986;25(4):326–639. 7. Matsushima T, Rhoton AL Jr, de Oliveira E, et al. Microsurgical anatomy of the veins of the posterior fossa. J Neurosurg. 1983;59(1):63–105. 8. Nakamura M, Roser F, Dormiani M, et al. Facial and cochlear nerve function after surgery of cerebellopontine angle meningiomas. Neurosurgery. 2005;57(1):77–90; discussion 77–90. 9. Voss NF, Vrionis FD, Heilman CB, et al. Meningiomas of the cerebellopontine angle. Surg Neurol. 2000;53(5):439–446; discussion 46–47.

10. Yuh WT, Mayr-Yuh NA, Koci TM, et al. Metastatic lesions involving the cerebellopontine angle. Am J Neuroradiol. 1993;14(1):99– 106. 11. Deb P, Sharma MC, Gaikwad S, et al. Cerebellopontine angle paraganglioma—Report of a case and review of literature. J Neurooncol. 2005;74(1):65–69. 12. Sade B, Mohr G, Dufour JJ. Cerebellopontine angle lipoma presenting with hemifacial spasm: Case report and review of the literature. J Otolaryngol. 2005;34(4):270–273. 13. Harner SG, Laws ER Jr. Clinical findings in patients with acoustic neurinoma. Mayo Clin Proc. 1983;58(11):721–728. 14. Nakamura M, Roser F, Mirzai S, et al. Meningiomas of the internal auditory canal. Neurosurgery. 2004;55(1):119–127; discussion 27– 28. 15. Kobata H, Kondo A, Iwasaki K. Cerebellopontine angle epidermoids presenting with cranial nerve hyperactive dysfunction: Pathogenesis and long-term surgical results in 30 patients. Neurosurgery. 2002;50(2):276–285; discussion 85–86. 16. Meng L, Yuguang L, Feng L, et al. Cerebellopontine angle epidermoids presenting with trigeminal neuralgia. J Clin Neurosci. 2005;12(7):784–786. 17. Kavar B, Kaye AH. Dermoid, epidermoid, and neurenteric cysts. In: Kaye AH, Laws ER Jr, eds. Brain Tumors. 2nd ed. London, England: Churchill Livingstone, 2001:965–981. 18. House JW, Brackmann DE. Facial nerve grading system. Otolaryngol Head Neck Surg. 1985;93(2):146–147. 19. Johnson EW. Auditory test results in 500 cases of acoustic neuroma. Arch Otolaryngol. 1977;103(3):152–158. 20. Harner SG, Reese DF. Roentgenographic diagnosis of acoustic neurinoma. Laryngoscope. 1984;94(3):306–309. 21. Wu EH, Tang YS, Zhang YT, et al. CT in diagnosis of acoustic neuromas. Am J Neuroradiol. 1986;7(4):645–650. 22. Aoki S, Sasaki Y, Machida T, et al. Contrast-enhanced MR images in patients with meningioma: Importance of enhancement of the dura adjacent to the tumor. Am J Neuroradiol. 1990;11(5):935– 938. 23. Moller A, Hatam A, Olivecrona H. Diagnosis of acoustic neuroma with computed tomography. Neuroradiology. 1978;17(1):25– 30. 24. Thomsen J, Klinken L, Tos M. Calcified acoustic neurinoma. J Laryngol Otol. 1984;98(7):727–732. 25. Rosenberg SI. Natural history of acoustic neuromas. Laryngoscope. 2000;110(4):497–508. 26. Liu P, Saida Y, Yoshioka H, et al. MR imaging of epidermoids at the cerebellopontine angle. Magn Reson Med Sci. 2003;2(3):109– 115. 27. Rutherford SA, King AT. Vestibular schwannoma management: What is the ‘best’ option? Br J Neurosurg. 2005;19(4):309– 316. 28. Kondziolka D, Lunsford LD, Flickinger JC. Acoustic tumors: Operation versus radiation–making sense of opposing viewpoints. Part II. Acoustic neuromas: Sorting out management options. Clin Neurosurg. 2003;50:313–328. 29. Mangham CA Jr. Retrosigmoid versus middle fossa surgery for small vestibular schwannomas. Laryngoscope. 2004;114(8):1455– 1461. 30. Rhoton AL Jr. Meningiomas of the cerebellopontine angle and foramen magnum. Neurosurg Clin N Am. 1994;5(2):349– 377. 31. Nakamura M, Roser F, Dormiani M, et al. Surgical treatment of cerebellopontine angle meningiomas in elderly patients. Acta Neurochir (Wien). 2005;147(6):603–609; discussion 9–10. 32. Ojemann RG. Suboccipital transmeatal approaches to vestibular schwannomas. In: Schmidek HH, Sweet WH, eds. Operative Neurosurgical Techniques.Vol.1.3rd ed. Philadelphia,PA: W.B. Saunders Company, 1995:829–841. 33. Tew JM Jr, Scodary DJ. Neoplastic disoders-surgical positioning. In: Apuzzo MLJ, ed. Brain Surgery Complication Avoidance and Management. 1st ed. New York, NY: Churchill Livingstone, 1993:1609–1620. 34. Ciric I, Zhao JC, Rosenblatt S, et al. Suboccipital retrosigmoid approach for removal of vestibular schwannomas:

Chapter 26: Tumors of the Cerebellopontine Angle Facial nerve function and hearing preservation. Neurosurgery. 2005;56(3):560–570; discussion 70. 35. Gjuric M, Wigand ME, Wolf SR. Enlarged middle fossa vestibular schwannoma surgery: Experience with 735 cases. Otol Neurotol. 2001;22(2):223–230; discussion 30–31. 36. Gormley WB, Sekhar LN, Wright DC, et al. Acoustic neuromas: Results of current surgical management. Neurosurgery. 1997;41(1):50–58; discussion 58–60.

401

37. Ojemann RG. Acoustic neuroma (vestibular schwannoma). In: Youmans JR, ed. Neurological Surgery. Vol. 4, 4th ed. Philadelphia, PA: W.B. Saunders Company, 1996:2841–2861. 38. Sampath P, Rini D, Long DM. Microanatomical variations in the cerebellopontine angle associated with vestibular schwannomas (acoustic neuromas): A retrospective study of 1006 consecutive cases. J Neurosurg. 2000;92(1):70– 78.

27 Tumors of the Jugular Foramen Samer Ayoubi, Badih Adada, and Ossama Al-Mefty

preservation during tumor removal. By contrast, medially positioned tumors (some meningiomas, some schwannomas, and some glomus tumors) displace the cranial nerves onto the lateral tumor surface, where they interpose between the surgeon and the tumor—an unfavorable location (3). The carotid artery passes anteromedial to the internal jugular vein to reach the carotid canal. At the level of the skull base, this artery runs anterior to the vein and is separated from it by the carotid ridge. It ascends a short distance in the canal (the vertical segment), and then turns at a right angle anteromedially toward the petrous apex (the horizontal segment). Three branches of the external carotid artery—the ascending pharyngeal, the occipital artery, and the posterior auricular artery—can contribute significant blood supply to lesions of the jugular fossa. The sigmoid sinus (Fig. 3) courses down the sigmoid sulcus, turning anteriorly toward the jugular foramen, and crossing it into the jugular bulb. It then flows downward behind the carotid canal into the internal jugular vein. The inferior petrosal sinus courses on the surface of the petroclival fissure, forming a plexiform confluence as it enters the petrosal part of the jugular fossa. The position of the lower cranial nerves with respect to the inferior petrosal sinus varies; therefore, overpacking or cautery over the sinus can injure these nerves (6).

INTRODUCTION The jugular fossa is an area of complex anatomy. It is also an area of variant pathologies, each warranting special surgical considerations. Because lesions in this area involve the lower cranial nerves and major venous channels, each patient needs an individual approach that takes into account the location, size, and pathology of the lesion, as well as the patient’s general and neurological condition.

SURGICAL ANATOMY The jugular foramen is an opening or gap in the skull that connects the posterior cranial fossa and the jugular fossa (1). It is configured around the sigmoid and inferior petrosal sinuses between the temporal and occipital bones, and extends in a posterolateral-to-anteromedial direction. The jugular foramen hosts two venous compartments: the sigmoid part, which receives flow from the sigmoid sinus and the petrosal part, which receives drainage from the inferior petrosal sinus. A fibro-osseous diaphragm separates these two vascular channels, and the lower cranial nerves lie on either side of this partition at the site of the intrajugular processes of the temporal and occipital bones (2–4) [Fig. 1(A)]. The jugular fossa is a deep depression located at the inferior surface of the petrous portion of the temporal bone, and it communicates with the posterior cranial fossa via the jugular foramen. It houses the jugular bulb, which continues as the jugular vein inferiorly [Fig. 1(B)]. The 9th, 10th and 11th cranial nerves enter the dura on the medial side of the intrajugular process. The entrance porus of the glossopharyngeal nerve is separated from the entrance of the vagus and accessory nerves by a dural crest in the jugular fossa (5). The glossopharyngeal nerve (Fig. 2) passes forward, coursing through the jugular fossa, and exiting on the lateral surface of the internal carotid artery deep to the styloid process. The vagus nerve exits the fossa vertically, in intimate relation with the accessory nerve behind the glossopharyngeal nerve on the posteromedial wall of the internal jugular vein (4). The accessory nerve descends laterally between the carotid artery and the internal jugular vein and then backward across the lateral surface of the vein to the muscle. The hypoglossal nerve passes through the hypoglossal canal and does not traverse the jugular foramen. It passes adjacent to the vagus nerve and descends between the internal carotid artery and the jugular vein. Then, it turns abruptly forward toward the tongue. Lesions located lateral to the fibro-osseous diaphragm, which divides the two vascular channels, include glomus tumors, some meningiomas, and some schwannomas. These displace the nerves medially, a position favorable for nerve

REGIONAL PATHOLOGY AND DIFFERENTIAL DIAGNOSIS A variety of lesions can arise from the structures normally found within the jugular foramen and fossa or from contiguous structures. But the surgeon must be able to recognize anatomical variations of the jugular bulb, particularly a high jugular bulb or turbulent flow, so as not to misdiagnose them as a jugular foramen tumor (Fig. 4). Although rare, the three most common mass lesions within the jugular foramen are paragangliomas (including glomus jugulare tumors), schwannomas, and meningiomas (3,7–10) (Fig. 5). A firm preoperative diagnosis of these lesions is crucial because each has different surgical considerations, such as preoperative embolization for glomus jugulare tumors or the amount of bone to remove for meningiomas. Magnetic resonance imaging (MRI) complemented by computed tomography (CT) studies allow differentiation between these tumors (Table 1). Differential diagnoses to be considered are acoustic schwannomas and lesions such as chordomas and chondrosarcomas, malignant tumors (carcinomas), metastases, peripheral primitive neuroectodermal tumors, cholesteatomas, chondromas, lymphangiomas, choroid plexus papillomas, salivary gland tumors, lipomas, aneurysmal bone cysts, hemangiopericytomas, plasmacytomas and inflammatory granulomas, pseudomasses such as normal 403

404

Ayoubi et al.

Figure 2 Anatomical dissection demonstrating the relationship of the jugular bulb to the carotid artery, facial nerve, and lower cranial nerves.

Figure 1 Photographs of dry anatomical specimens from the right side. (A) Anterior perspective, in which the view extends from the outside to the inside, delineating the jugular foramen (white arrows) and jugular fossa (black arrows). (B) Posterior perspective, in which the view extends from the inside to the outside, delineating the jugular foramen (black arrows) and jugular fossa (white arrows). Note the difference between the two perspectives. Source: From Ref. 1.

vascular asymmetry, a high jugular bulb or jugular diverticulum, and aneurysms of the petrous carotid artery (7,9,10–15). A jugular foramen abscess has also been reported (16).

GLOMUS JUGULARE TUMORS Pathology Glomus jugulare tumors arise from glomus bodies located in the dome of the jugular bulb (17). These tumors grow along the path of least resistance and can gain access to the subarachnoid space by penetrating the dura of the posterior fossa, growing along cranial nerves, or, less commonly, penetrating the dura of the middle fossa (18). Glomus jugulare tumors are uncommon tumors of the head and neck, accounting for only 0.03% of all neoplasms and 0.6% of head and neck tumors. Nonetheless, they are the most common neoplasms of the middle ear and second to vestibular schwannomas as the most common tumor involving the temporal bone. These tumors appear in patients in their second decade (or earlier) to the ninth decade, although most tumors

Figure 3 The course of the sigmoid sinus inside the temporal bone and down to the neck.

Chapter 27: Tumors of the Jugular Foramen

405

Table 2 Classifications of Glomus Tumors Type

Physical Findings

Glasscock-Jackson Classification of Glomus Tumors I Small tumor involving jugular bulb, middle ear, and mastoid. II Tumor extending under internal auditory canal may have intracranial extension. III Tumor extending into petrous apex may have intracranial extension. IV Tumor extending beyond petrous apex into clivus or intratemporal fossa may have intracranial extension. Fisch Classification of Glomus Tumors A Tumors limited to the middle ear cleft. B Tumors limited to the tympanomastoid area with no infralabyrinthine compartment involvement. C Tumors involving the infralabyrinthine compartment of the temporal bone and extending into the petrous apex. Tumors with an intracranial extension less than 2 cm in diameter. D1 Tumors with an intracranial extension greater than 2 cm in D2 diameter. Figure 4 MRI of a high jugular bulb that may be mistaken for a glomus jugulare tumor. (A) Axial view. (B) Coronal view. (C) Magnetic resonance venography.

Figure 5 The most common pathological lesions in the jugular foramen. (A) Paraganglioma. (B) Schwannoma. (C) Meningioma.

manifest in the fourth decade of life. There is no clear racial predilection, but glomus tumors seem to be more common among Caucasians. There is a marked predominance among females; women are affected three to six times more commonly than men, with a peak incidence during the fifth decade of life. Multiple paragangliomas are reported in more than 10% of cases. Familial cases, most of which involve

Source: Adapted from Refs. 20, 21.

fathers and daughters, have a much higher rate of multicentricity, up to 55%. Evidence supports an autosomal dominant inheritance pattern consistent with genomic imprinting and an association with the haplotype at chromosome band 11q23. Most multicentric tumors are carotid body tumors. Only a few cases of bilateral glomus jugulare tumors associated with carotid body tumors have been reported (19). With rare exceptions, tumors of the glomus jugulare are benign, slow-growing paragangliomas. The two established classifications of these tumors, those of Fisch (20) and Glasscock and Jackson (21), are based mainly on tumor size, with special emphasis on intracranial extension as a decisive factor for resectability (Table 2). A subgroup of glomus jugulare tumors is rarely encountered but presents a formidable challenge for treatment. Al-Mefty and colleagues (22) used the following criteria to describe this group.

r r r r r

Giant size (Fig. 6) Multiple paragangliomas (bilateral or ipsilateral) (Fig. 7) Malignancy (Fig. 8) Catecholamine secretion Association with other lesions such as a dural arteriovenous malformation or an adrenal tumor, or previous treatment with an adverse outcome that makes surgical intervention a much higher risk, such as sacrifice of the carotid artery, radiation therapy, or postoperative deficits or adverse effects from embolization.

Table 1 Imaging Differences According to Type of Tumor Tumor

CT

MRI

Glomus jugulare

Erosion and destruction of the jugular spine and carotid crest (carotico-jugular spine, the bone separating the petrous carotid from the jugulare bulb). Moth-eaten pattern Enlarged fossa with smooth, distinct, sclerotic margin.

Salt-and-pepper appearance. Nonhomogeneous enhancement

Schwannoma

Meningioma

Frequently involves bone (including the jugular spine and particularly the jugular tubercle), producing hyperostosis and bone thickening without bone erosion.

Low to isointense on T1-weighted image, high signal on T2. May be cystic. Often has dumbbell shape. Moderate-to-marked enhancement. Tail sign. Extensive enhancement.

406

Ayoubi et al.

Figure 9 Deviation of the soft palate with atrophy and deviation of the tongue as presenting symptoms in a patient with a jugular foramen lesion.

Figure 6 MRI.

Examples of a giant glomus jugulare tumor. (A) CT scan. (B)

Clinical Presentation Symptoms of glomus tumors usually include pulsatile tinnitus, hearing loss, and lower cranial nerve palsies (18). Conductive hearing loss is the result of mechanical obstruction of the ossicular mechanism by the tumor, while sensorineural hearing loss is a result of the involvement of the labyrinth. Lower cranial nerve deficits are the prominent feature in patients with symptomatic jugular tumors (22) (Fig. 9). Symptoms of catecholamine production include palpitations, excessive sweating, and headache. These symptoms should be considered in any patient with a glomus jugulare tumor. Large tumors can cause obstructive hydrocephalus with symptoms and signs of increased intracranial pressure.

Appearance on Diagnostic Imaging

Figure 7 Examples of multiple paragangliomas. (A) Angiogram. (B) MRI. Abbreviations: RCCA, right common carotid artery; LCCA, left common carotid artery.

Figure 8 Examples of a malignant paraganglioma. (A, B) MRI after previous treatment with surgery and radiotherapy. (C) MRI showing marked growth a year later. (D) MRI after resection. (E) MRI showing rapid growth after 4 months. (F) Immunochemical staining for chromogranin. Note the brown granules in the cytoplasm. Source: From Ref. 22.

Bone-window CT scans show skull-base infiltration with erosion and enlargement of the jugular foramen (23,24). The extensive bone destruction is characterized by an irregular “moth-eaten” pattern of erosion (Fig. 10). MRIs show the enhanced tumor with flow voids and a “salt-and-pepper” appearance (Fig. 11), and also disclose the presence of multiple tumors. MRI of the neck is done to exclude associated paragangliomas. The arteriographic findings of glomus jugulare tumors are typically a hypervascular mass with an intense, characteristic tumor “blush.” Large feeding vessels and early draining veins are commonly encountered, indicative of early arteriovenous shunting. Similar to other tumors in this region,

Figure 10 Bone destruction by a glomus jugulare tumor, as seen on the bone window of a CT scan.


407

Figure 13 Angiographic depiction of a new blood supply through the opposite internal carotid artery and the vertebrobasilar system in a patient who had prior treatment with embolization and carotid occlusion. (A) Opposite carotid injection. (B) Vertebral injection.

Figure 11 Contrast-enhanced MRI demonstrating the typical salt-andpepper enhancement.

speech evaluations and swallowing studies are conducted before surgery.

Hormonal Studies glomus jugulare tumors are predominantly supplied by the external carotid artery system, mainly the ascending pharyngeal artery (23) (Fig. 12). Angiographic studies are critical for assessing the appropriateness of preoperative embolization after the tumor’s blood supply has been delineated. The most critical aspect of the angiographic evaluation in patients who have undergone previous embolization or carotid occlusion is the identification of new feeding vessels from the internal carotid artery and the vertebrobasilar circulation (22) (Fig. 13).

Preoperative Preparation With some modifications, the same preoperative protocol is used for all lesions of the jugular foramen. Extensive audiological and otolaryngological evaluations are carried out, and

Paragangliomas have the potential to secrete a wide variety of neuropeptide hormones, including adrenocortical hormones, serotonin, catecholamines, and dopamine (22). Patients with hypersecreting tumors (catecholamine levels four times higher than normal) require preparation with combined alpha- and beta-blocker medication before surgery, angiography, or embolization. Beta-blockers should not be given before or without alpha-blockers. Screening for excess catecholamines is necessary in all patients with a glomus jugulare tumor, and the actual treatment and duration of prophylaxis depends on the level of catecholamine secretion and its source. Patients with these tumors might also harbor adrenal norepinephrine-secreting tumors. Therefore, adrenal imaging is part of our workup and is particularly important for patients with hypersecreting tumors.

Diagnostic Imaging MRI scans with and without contrast enhancement, angiography, or arteriovenous magnetic resonance angiography, and very thin slices of CT scans constitute the radiological workup needed to explore the anatomy of each patient’s jugular fossa, temporal bone, and condyles. The nature of the tumor (cystic or solid), extensions (intracranial, extracranial, or dumbbell), and the characteristics of bone involvement (the presence of sclerosis and enlargement of the canal) are studied with MRI images and CT scans with the aid of a bone algorithm. The dominant vertebral artery and the characteristics of the vertebrobasilar system are also studied. Special attention is paid to the venous phase to determine the size, dominance, and tributaries (superior petrosal, inferior petrosal, and vein of Labbé) of the transverse and sigmoid sinuses, and the position and size of the jugular bulb. Figure 12 The typical appearance of a glomus jugulare tumor on arteriograms. (A) Internal carotid. (B) The external carotid image shows the tumor’s high vascularity, venous shunting, and blood supply through the ascending pharyngeal artery.

Embolization Embolization is indicated for patients with paragangliomas and other highly vascular lesions (7,25,26). Surgery is more challenging when the patient has undergone prior

408

Ayoubi et al.

Figure 14 Preoperative embolization is of great value in decreasing the tumor’s blood supply, particularly from the external carotid artery. (A) Preembolization angiogram. (B) Postembolization angiogram.

embolization or carotid artery occlusion, after which new feeding vessels from the internal carotid and vertebrobasilar circulation developed (22). Current techniques are successful for embolizing the tumor bed and reducing blood loss (Fig. 14). The thoroughness of embolization is critical; however, partial embolization of the external carotid feeder augments the internal carotid feeders. Furthermore, embolization has accompanying risks and complications, including reflux cerebral emboli in the internal carotid artery, cranial nerve deficits from a “dangerous anastomosis,” and tumor hemorrhage. Carrier and colleagues reported that preoperative embolization of the inferior petrosal sinus, the anterior condylar vein complex, and the posterior condylar vein reduced preoperative bleeding considerably (27). Because of the absence of shift at the rigid structures of the jugular foramen, intraoperative image-guided frameless navigation is particularly useful during the surgical procedure (Fig. 15).

Intraoperative Neurophysiological Monitoring The 10th cranial nerve is monitored intraoperatively with an electromyographic endotracheal tube. Electromyographic needles are inserted into the facial musculature, the sternocleidomastoid muscle, and the tongue to monitor the 7th, 11th, and 12th cranial nerves, respectively. Auditory evoked potentials are obtained if hearing is present and no plan is made for closing the ear canal or resecting tumor from the middle ear. The tube is placed under sterile conditions after the ear is prepared. Multiple paragangliomas present the greatest challenge to treating complex paragangliomas because the treatment decision is not based on a single tumor, but on the quality and length of the patient’s life. Whether to treat, when to treat, which tumors to treat, with which modality (surgery or radiation) and in which sequence are all questions that must be addressed at the first evaluation and thoroughly considered throughout the patient’s follow-up. The surgeon must try to prevent the consequences of multiple bilateral cranial nerve deficits (22).

Figure 15 Intraoperative navigation is a useful orientation tool for surgery in the foramen magnum. (A) MRI. (B) Intraoperative CT scan indicating the extension of the tumor to the top of the petrous apex.

data, the surgeon can then decide the most appropriate approach (28). The approaches include the infratemporal, combined infratemporal–posterior fossa, and combined approaches with a total petrosectomy (22).

Patient Position The patient is placed supine with the head elevated, turned away from the side of the lesion, and fixed in the three-point head frame. The abdomen and thigh are prepared for removal of fat and fascia lata grafts (Fig. 16).

Incision and Soft-Tissue Dissection An open C-shaped incision is made behind the ear and extended up to the temporal area and down transversely along the natural skin crease in the neck. In selected patients, in whom the middle ear is involved, the external ear canal is transected at the bony cartilaginous junction. The skin of the external ear canal is everted and closed as a blind sac. A small periosteal flap is kept attached to the skin flap and closed over the ear canal (Fig. 17). The skin flap, including the auricle, is

Surgical Approach The surgical approach is tailored in each patient according to the findings of preoperative imaging, the local anatomy, and the tumor’s characteristics and extension. Jugular foramen tumors with an intracranial extension should be carefully evaluated with regard to size, position, infiltrative potential, and vascularization. On the basis of this

Figure 16 Skin incision and patient positioning for glomus tumor surgery.


409

Figure 17 Skin dissection and closure of the ear canal. (Insert) Operative photograph. Source: From Ref. 49.

reflected anteriorly, and the sternomastoid muscle is detached from its insertion in the mastoid process. The neurovascular structures in the neck are dissected and exposed; these include the common carotid artery, internal carotid artery, external carotid artery, the jugular vein, and the cranial nerves (9th through 12th).

Figure 19 Skeletonizing of the facial nerve from the stylomastoid foramen to the geniculate ganglion.

Bone Removal The mastoidectomy is done with a high-speed drill, and bleeding is anticipated during drilling of the bone. The eardrum is removed and the tumor is located in the middle ear. The semicircular canals are exposed (Fig. 18), and the facial canal is located caudal to the lateral semicircular canal. The facial nerve is skeletonized from the stylomastoid foramen to the geniculate ganglion (Fig. 19). We have abandoned the practice of routine transposition of the facial nerve; after its skeletonization, we keep it in a tiny protective bony canal. If the nerve needs to be transposed, it is moved out of the fallopian canal and secured anteriorly. A radical mastoidectomy exposes the sigmoid sinus down to the jugular bulb and, in

Figure 18 The mastoidectomy and opening of the middle ear.

cases of intradural extension, is followed by a lateral and low posterior fossa craniectomy.

Tumor Isolation To isolate the tumor, the internal carotid artery and the jugular vein are followed upward toward the base of the skull. To expose the tumor, the posterior belly of the digastric muscle and the stylohyoid muscles is transected and the styloid process is removed. The ascending mandibular ramus is dislocated anteriorly, if necessary. The sigmoid sinus is then ligated distal to the tumor’s extension and proximal to the mastoid emissary vein. If the tumor extends into the middle ear or along the petrous carotid artery, the remnant of skin in the external ear canal is removed with the tympanic membrane. The internal carotid artery is exposed in the petrous canal through drilling of the bone over the carotid canal if the tumor has not already destroyed this bone. The eustachian tube is obliterated with a piece of muscle. The anterior pole of the tumor is then dissected from the internal carotid artery, and the small feeding arteries are coagulated with bipolar electrocautery. At this stage, the extradural tumor is completely exposed (Fig. 20). If the tumor does not extend into the middle ear, the approach should be modified to preserve both the middle and the inner ear. Thus, the tumor is exposed in the infralabyrinthine space. The superior pole of the tumor is freed from the infratemporal fossa. The inferior pole is removed by dissecting and elevating the jugular vein after it is ligated to prevent early venous drainage. The lower cranial nerves are preserved as they emerge from the jugular foramen. Intrabulbar dissection, a maneuver described by AlMefty and Teixeira (22), helps preserve the lower cranial nerves. This maneuver can be used for any tumor, as long as the tumor itself has not penetrated the wall of the jugular bulb or actually infiltrated the cranial nerves. The outer wall

410

Ayoubi et al.

Intradural Tumor Removal

Figure 20 Isolating the tumor. Source: From Ref. 22.

of the lower sigmoid sinus is incised along the jugular bulb into the jugular vein. The tumor is then removed from inside the jugular bulb and the sigmoid sinus, and the tail end of the jugular vein is separated from the lower cranial nerves. The innermost venous wall separating the tumor from the nerves is left in situ to minimize dissection, trauma, manipulation, or devascularization of the lower cranial nerves (Fig. 21). Using this technique helps preserve the immediate postoperative function of the lower cranial nerves for patients in whom the tumor does not transgress the venous wall at the jugular foramen. When the tumor does transgress the venous wall, the cranial nerves can be infiltrated on a microanatomical level despite having normal function (29). In these situations, total resection may not be possible without sacrificing these nerves (29). Profuse bleeding from the inferior petrosal artery is controlled through packing with appropriate hemostatic materials.

To remove the intradural portion of the tumor, the dura mater is excised posterior to the sigmoid sinus and carried forward, and the intradural extension of the tumor is exposed. The cranial nerves (8th through 12th) are meticulously dissected from the tumor and kept intact. Tumor encroachment on the medulla is removed through microdissection, and the basilar artery, the anteroinferior cerebellar artery (AICA), and the posteroinferior cerebellar artery (PICA) are dissected from the tumor and preserved. Any tumor extension into the foramen magnum is followed and removed after it is freed from the lateral and anterior surfaces of the medulla and the vertebrobasilar junction. When it is giant, the tumor should be isolated for safe surgical removal. This is best done in one stage through the combined posterior fossa and infratemporal approach described earlier (30). This approach allows the tumor to be devascularized from the intrapetrous carotid artery. It is also used to separate the tumor from the posterior fossa and to dissect the lower portion from the nerves with minimal blood loss while preserving the vessels. Resecting tumors of the glomus jugulare requires special techniques in the handling of both arterial and venous dissection. Although paragangliomas engulf, adhere to, and receive blood from the internal carotid artery, with the aid of the operating microscope, a plane of dissection can be identified to separate the tumor from the carotid. Thus, the carotid artery does not need to be sacrificed or reconstructed. Because exposing tumors of the glomus jugulare requires neck dissection, associated tumors of the carotid body can be removed at the same time without additional morbidity or undue lengthening of the operating time. Glomus jugulare tumors often shunt blood with high venous outflow. Accordingly, they should be handled as arteriovenous malformations. Therefore, venous drainage from the tumor should be preserved and the proximal end of the jugular vein should not be ligated until the tumor is isolated and its arterial supply is devascularized.

Closure Once total removal of the tumor is assured, the eustachian tube (if exposed) is covered with a small piece of muscle and fascia. The dura mater is repaired with a graft of fascia lata, and the cavity is obliterated with fat. The temporal muscle is swung inferiorly and sutured to the sternomastoid muscle, and the skin is closed in two layers (Fig. 22).

Postoperative Care The patient’s ability to swallow is tested postoperatively, and oral intake is allowed only if these studies show satisfactory results, which assure that the function of the lower cranial nerves can protect the airway. Otherwise, the patient is kept on the transpyloric feeding by Dobhoff tube until swallowing recovers. Adaptations are made for satisfactory airway protection and other appropriate precautions are taken. A CT scan is obtained during the early postoperative period to check for hemorrhage, hydrocephalus, edema, or infarction. Any residual tumor is better assessed on later contrast-enhanced and fat-suppression MRIs.

Reconstruction Figure 21 Intrabulbar dissection preserves the wall of the jugular bulb, if it is not involved by the tumor, and protects the lower cranial nerves.

Large surgical defects must be repaired with vascularized flaps (temporalis fascia, cervical fascia, sternocleidomastoid muscle, and temporalis muscle) (26). Reconstruction of the carotid artery as a graft bypass to the middle cerebral artery


411

Deficits of the lower cranial nerves are the main surgical complications. However, the success in maintaining function has alleviated many of these concerns, and vigilant percussion postoperatively has minimized pulmonary complications from aspiration until the patient adapts or vocal cord medialization is done. Thus, total resection is indicated and successful in treating complex glomus jugulare tumors despite the challenge encountered (22). The rarely encountered malignant type, however, carries a poor prognosis (22,33).

Other Treatments

Figure 22 Closing the dura and packing the space with a fat graft.

may be necessary if the carotid is injured beyond repair (26,31). Facial reconstruction with the greater auricular nerve, the sural nerve, or a 12/7 anastomosis might be needed if the facial nerve was transected (26).

Results The results of surgical treatment of glomus jugulare tumors have improved drastically with the advances in skull-base surgery. Surgical resection with long-term follow-up shows the effectiveness of total removal in achieving a cure (19,21,32) (Fig. 23). Even the most formidable tumors have been successfully resected with no mortality and low morbidity (22).

Radiation therapy has long been used to treat glomus tumors, particularly those that are only partially removed or have recurred. But glomus tumors are known to be radioresistant, and the effect of radiation is often the induction of fibrosis, mainly along the vessels supplying the tumor. Furthermore, persistent viable tumors are often present long after the patient undergoes radiation therapy. Radiation therapy has also been associated with long-term side effects that include osteonecrosis of the temporal bone, the development of a new malignancy, and demyelination. Early reports of stereotactic radiosurgery for glomus tumors are encouraging, and this modality may be useful in controlling symptoms. Radiosurgery appears to provide control if the target size is within the optimal size for treatment, and the preliminary results of this treatment suggest a symptomatic improvement of cranial nerve function. If it is proved effective in long-term control with few complications from cranial nerve deficits, radiosurgery will be a great complement to the current treatment of bilateral glomus jugulare tumors and residual lesions from the resection of giant tumors (17). In a review of papers published from 1994 to 2004, Gottfried and colleagues found that death and recurrence after treatment with either radiosurgery or surgery were infrequent, and concluded that both treatments could be considered safe and efficacious. Although associated with higher morbidity rates, surgery immediately and totally eliminates the tumor. The results of radiosurgery were promising, but the long-term recurrence rate is still unknown (34).

Complications In patients with glomus jugulare tumors, mortality is around 1%. Leaks of cerebrospinal fluid (CSF) occur in 8% of patients. Other complications include aspiration, wound infection, pneumonia, and meningitis (34). Postoperative CSF leakage is one of the most important complications in the surgery of jugular fossa tumors (35). A CSF leak can occur either from the skin or, more often, in the form of rhinorrhea. Meticulous closure is very important in preventing the leak. If a leak does occur, spinal drainage is carried out for 72 hours. If the leak continues, the wound is reexplored. If meningitis occurs as a result of a CSF leak, it is treated with antibiotics and spinal drainage.

Follow-up and Rehabilitation

Figure 23 Complete surgical resection is successful in the overwhelming number of patients with large or giant glomus tumors and provides the prospect of cure. (A) Preoperative CT scan of a giant glomus tumor. (B) Postoperative CT scan after total removal.

After surgery, each patient is followed up with MRI images at 3 months, 6 months, and then annually to detect any recurrence. Dysphonic symptoms are treated with medialization of the vocal cords (34,36). Most patients adapt to unilateral deficits of the lower cranial nerves; however, the older the patient, the longer the recovery. Hence, aspiration must be avoided and tube feeding is needed in the early stages, or a jejunostomy is done if aspiration persists for prolonged periods of time. Eye weights or tarsorrhaphy may be necessary if facial weakness is present (34).

412

Ayoubi et al.

SCHWANNOMAS Pathology Jugular foramen schwannomas are a rare pathological condition comprising 2% to 4% of intracranial schwannomas (7), with approximately 100 cases reported in the literature (37). Schwannomas located at the jugular foramen may arise from the glossopharyngeal, vagus, or accessory nerves (7,37,38). The proximity and clinical manifestation of schwannomas originating from the hypoglossal nerve are the reason some authors classify these tumors with jugular foramen schwannomas. These tumors can originate from the cisternal portion of the nerve and present with major intracranial growth, or from the foraminal portion expanding the bone, or from the distal portion and present with extracranial growth. Some appear with both extra- and intracranial growth through an enlarged jugular foramen (7). Pellet and colleagues (39) have classified jugular foramen schwannomas into four types.

r Type A: primarily intracranial, minimal extension into the jugular foramen

r Type B: primarily intraosseous, with or without an intracra-

Figure 24 Imaging studies of a jugular foramen schwannoma. (A) Preoperative axial MRI showing the dumbbell shape and enhancement. (B) Coronal T2-weighted image showing a hyperintense signal and a cystic component. (C) Preoperative sagittal MRI showing the extension into the upper trunk. (D) CT scan showing expansion and scalloping of the jugular foramen, which is typical of schwannomas. (E, F) Postoperative MRIs showing total removal.

nial extension

r Type C: primarily extracranial, minimal extension into the jugular foramen

r Type D: saddlebag- or dumbbell-shaped intra- and extracranial extensions

Clinical Presentation The primary symptoms of a jugular foramen schwannoma include dizziness, hearing loss, dysphagia, diplopia, tongue paresis, and hoarseness (38). Preoperative findings include mainly audiovestibular (hearing loss, tinnitus, and dizziness) and lower cranial nerve signs (dysphagia, hoarseness, weakness of the shoulder, and tongue paresis) (36–38). Almost all patients with dumbbell-shaped schwannomas have glossopharyngeal and vagal deficits. Hypoglossal and accessory nerve deficits appear in most patients, while hearing loss is less common (7). With large tumors, the abducens and facial nerves can be affected and patients may develop cerebellar signs or hydrocephalus. Careful evaluation of the lower cranial nerves is particularly important in jugular foramen schwannomas because they are the source of most life-threatening postoperative complications.

Appearance on Diagnostic Imaging An enlarged jugular foramen with well-delineated sclerotic margins is seen on thin-cut CT scans with bone algorithms (7,23,40). Schwannomas usually expand the foramen without eroding it (41), and have a low to isointense signal on T1-weighted images and a high signal on T2-weighted images. The tumor shows moderate-to-marked enhancement after the injection of gadolinium (7,23,38,40,42). Cystic degeneration can occur, and is well delineated on MRIs (38,43) (Fig. 24). On conventional or magnetic resonance angiography, schwannomas are avascular (7).

Preoperative Preparation Diagnostic imaging and intraoperative neurophysiological monitoring are done in the same way as for patients with glomus tumors. Hormonal studies and embolization are not needed.

Surgical Approach The real challenge for neurosurgical treatment of these tumors is to preserve the function of the lower cranial nerves

while achieving radical resection and decreasing the risk of recurrence. A repeated operation drastically increases the chance of injury to the lower cranial nerves. These challenges are found especially in patients with dumbbell-shaped tumors, in which the nerves are at risk during resection of the intracranial, intraforaminal, and extracranial sections. Schwannomas differ from glomus tumors and meningiomas located within the jugular foramen and fossa because they compress rather than invade the jugular bulb, and the nerves of origin are positioned anterior to this structure. The labyrinth is also preserved as hearing might occasionally improve after removal of the tumor. Because schwannomas within the jugular foramen tend to displace the jugular bulb posteriorly, the suprajugular approach allows the surgeon to remove the tumor without opening the wall of the bulb. Even if preoperative studies reveal an absence of flow into the jugular bulb, the transjugular approach is not used because the sinus usually recovers its patency after decompression. The suprajugular approach is essentially a presigmoid infralabyrinthine route (Fig. 25). It is used if the tumor extends anteriorly to the jugular bulb. A postauricular incision is made, and the internal carotid artery; external carotid artery; internal jugular vein; and 9th, 10th, 11th, and 12th cranial nerves are identified in the cervical region. The sternomastoid muscle is dissected, mobilized, and reflected inferiorly. A mastoidectomy is followed by complete skeletonization of the sigmoid sinus, jugular bulb, and jugular vein. The presigmoid, infralabyrinthine space is exposed and the dura mater is identified superior to the patent jugular bulb and inferior to the labyrinth. After the cerebellomedullary cistern is opened, releasing CSF, the tumor is exposed and debulked. The lower cranial nerves (9th through 12th), the PICA, the AICA, and the vertebral artery are dissected away from the tumor through the arachnoid plane, and the lesion is radically removed. The removal of a schwannoma is accomplished without sectioning the ear canal, entering the middle ear, or transposing the facial nerve (7). Some surgeons transpose the facial nerve for selective circumstances, such as cases in which scar tissue from a previous operation impedes control of the carotid artery and safe removal, the tumor has a large extension anteriorly to the petrous apex, or the middle ear


413

Figure 26 (A) MRI of a small intraforaminal schwannoma. (B) Magnetic resonance venography image showing a single and dominant drainage through the corresponding jugular bulb. Radiosurgery might be preferable because of the increased risks presented by the venous configuration. Figure 25 The suprajugular approach usually used for jugular foramen schwannomas. Abbreviations: JB, jugular bulb; JV, jugular vein; SS, sigmoid sinus; SPS, superior petrosal sinus. Source: From Ref. 1.

is extensively involved (37,38,44). Since schwannomas are avascular and largely soft and suckable, we have not found a need to transpose the facial nerve. Dumbbell-shaped jugular foramen schwannomas present a special challenge, with the risk of injury to the lower cranial nerves intracranially, intraforaminally, and extracranially (7,31). However, these tumors can be removed without creating additional neurological deficits. Furthermore, the patient can be expected to recover function in the affected cranial nerves (7).

Postoperative Course In these patients, the acute development of postoperative deficits before the development of compensatory mechanisms requires careful attention. Speech pathology and otolaryngological evaluations with pre- and postoperative swallowing studies are obtained. Oral intake is withheld and parenteral nutrition is administered. Swallowing exercises and soft mechanical diets with swallowing precautions are prescribed if the patient exhibits a risk of aspiration. Vocal cord medialization is done if there is persistent dysphasia or aspiration (7).

Results Complete excision is achieved in the majority of patients with a schwannoma (7,26,38,43). Preoperative palsy of the 5th, 6th, 7th, 9th, 10th, and 12th nerves can improve after the removal of a jugular foramen schwannoma (7,38). Hearing can also improve (7,43).

Other Treatments Gamma-knife radiosurgery can be offered to patients who have small tumors or intact lower cranial nerve function, and those who have declined surgery. It is also considered for patients who have residual or recurrent tumors after microsurgical resection (45). Experience with radiosurgery is limited due to the rarity of these tumors. We prefer to reserve radiosurgery treatment for the rare patient in whom the venous anatomy presents a considerable risk, such as when the patient has a single functioning ipsilateral sigmoid sinus and jugular bulb (Fig. 26).

MENINGIOMAS Pathology Primary jugular fossa meningiomas are one of the rarest subgroups of meningioma, with fewer than 40 cases reported in the literature (1,46). They constitute 9% of jugular fossa tumors (9,36). These meningiomas presumably arise from arachnoid cells lining the jugular bulb in the jugular fossa (1). Women are affected more than men (1). The transitional type of meningioma is most commonly found, closely followed by the meningotheliomatous type and the less common psammomatous meningioma (47). Their intimate relationship with the lower cranial nerves and jugular bulb, their involvement of the temporal bone, and their tendency to extend intracranially and extracranially have traditionally made their removal fraught with difficulty (1). Thus, the surgeon needs to tailor the approach to the local anatomy (the tumor–neurovascular relationship).

Clinical Presentation The symptoms of a jugular foramen meningioma are similar to those of a schwannoma, with signs of lower cranial nerve deficits and altered hearing (1).

Appearance on Diagnostic Imaging A meningioma of the jugular foramen permeates and shows sclerotic changes on bone algorithm CT studies (Fig. 27). These lesions most often demonstrate isointense or low signal on T1-weighted MRIs and intermediate signal intensity on T2-weighted images. They also show avid, homogeneous enhancement on contrast MRIs and a dural tail (Fig. 28). Meningiomas of the jugular foramen may demonstrate a relatively more aggressive appearance than similar intracranial lesions, but most often retain their cerebrospinal fluid/vascular cleft with the brain parenchyma intracranially. Angiography shows avid arterial blushing and prolonged contrast retention into the venous phase (38). MRAs have mostly replaced conventional angiography as a preoperative test.

Preoperative Preparation The steps to prepare the patient for surgery are similar to those for patients with glomus tumors and schwannomas. Particular emphasis is given to the position, patency, and size of the jugular bulb as seen on magnetic resonance venography performed during both arterial and venous phases.

414

Ayoubi et al.

Figure 29 The three approaches used to remove jugular foramen meningiomas. Source: From Ref. 1. Figure 27 CT of a jugular foramen meningioma showing bone invasion with hyperostotic features, which is characteristic of meningiomas.

Surgical Approach For patients with meningiomas, the involved dura mater and the bone of the jugular fossa should be resected to minimize the chance of tumor recurrence (1). The surgical approach is tailored to the local anatomy of the tumor and its relation to the neurovascular structures. Three different routes can be used (Fig. 29).

r The suprajugular approach, a presigmoid, infralabyrinthine route, is chosen if the jugular bulb is patent and the tumor extends primarily anteriorly. r The retrojugular approach, a transcondylar, transtubercle, retrosigmoid route, is chosen if the jugular bulb is patent and the tumor extends primarily behind. r The transjugular approach, an infratemporal route, is chosen for patients in whom the jugular bulb is totally occluded by the tumor. The position, incision, and soft-tissue dissection for each approach are similar to those used for schwannomas.

The Suprajugular Approach In the suprajugular approach, a total mastoidectomy is done with complete skeletonization of the sigmoid sinus,

jugular bulb, and jugular vein. The jugular fossa is accessed in the presigmoid infralabyrinthine space. The dura mater located superior to the patent jugular bulb and inferior to the labyrinth is opened. CSF is released from the cerebellomedullary cistern, and the tumor is dissected away from the lower cranial nerves (9th through 11th), the PICA, the AICA, and the vertebral artery, while the arachnoidal surgical dissection planes are preserved. The tumor is debulked with suction and bipolar coagulation or with an ultrasonic aspirator. The procedure is completed with microsurgical radical resection of the tumor.

The Retrojugular Approach In the retrojugular approach, the suboccipital bone is exposed and a small, inferior, lateral suboccipital craniotomy is performed, followed by a mastoidectomy and complete skeletonization of the sigmoid sinus, jugular bulb, and jugular vein. Drilling approximately one-third of the condyle usually suffices for the exposure, and postoperative stabilization is not necessary. Attention is then turned to the jugular tubercle, which is completely drilled away to facilitate opening of the jugular fossa, which lodges the jugular bulb. With the aid of an operating microscope, the dura mater is incised along the posterior border of the sigmoid sinus (Fig. 30). The tumor is carefully separated from the medulla oblongata, lower cranial nerves, vertebral artery, and PICA along arachnoidal planes, and is followed toward the jugular fossa. Careful, meticulous dissection of the tumor from the jugular bulb and the wall of the jugular vein is important. Ultrasonic aspiration or suction and bipolar coagulation are used to debulk the tumor.

The Transjugular Approach

Figure 28 Contrast-enhanced MRI of a jugular foramen meningioma showing intense homogeneous enhancement and the dural tail characteristic of a meningioma. (A) Preoperative image. (B) Postoperative image.

For patients in which the meningioma has invaded the sinus and occupied the jugular bulb, a transjugular approach similar to that for a glomus tumor is used (Fig. 31). The neurovascular structures in the neck (9th through 12th cranial nerves, jugular vein, and carotid artery) are dissected and exposed. A radical mastoidectomy exposes the sigmoid sinus down to the jugular bulb and is followed by a posterior fossa craniotomy. The jugular vein is followed superiorly to the jugular bulb. To enlarge the exposure, the posterior belly of the digastric muscle and the stylohyoid muscle are transected


415

Figure 30 The retrojugular approach, which includes skeletonization of the sigmoid sinus and drilling of the jugular tuberculum. Source: From Ref. 1.

Figure 32 Illustration of the transjugular removal of a jugular foramen meningioma that invades and occupies the jugular bulb.

and the styloid process is removed. The sigmoid sinus and jugular vein are ligated at a location proximal to the mastoid emissary vein and distal to the tumor obstruction. The inferior pole of the tumor is then dissected off the internal carotid artery and the jugular vein. The extradural tumor is thus completely exposed (Fig. 32). Bleeding from the inferior petrosal sinus may be profuse and is controlled by packing with Gelfoam. With the aid of the microscope, the dura mater is opened posterior to the sigmoid sinus and carried forward. The intradural tumor extension is then exposed. Meticulous intradural dissection of the tumor, performed while maintaining the arachnoidal dissection planes, helps preserve the function of the lower cranial nerves and the vertebral artery, the PICA, and the AICA at the anterolateral surface of the medulla oblongata.

of a good outcome, provided extensive evaluation and appropriate tailoring of the operative approach is done.

Results Radical tumor removal can be achieved in 83% to 100% of patients with jugular fossa meningiomas (1,36,46). The most common complications are transient deficits of the lower cranial nerves, which resolve or are compensated for in all patients within 1 month (1,36). Therefore, jugular fossa meningiomas can be radically resected with the expectation

Other Treatments As in the case of schwannomas, experience with radiosurgery is limited due to the rarity of these tumors. The results, however, are expected to be the same as those of radiosurgery for basal meningiomas, when it is used for residual or recurrent tumors or as the primary treatment. Recent literature is expanding the data on the efficacy, technique, control rates, risks, and complications of this treatment. The goal of radiosurgery, however, is “control” and long-term results are not yet available. The average follow-up for the usual reported radiosurgery series is too short to ascertain significant control of this slow-growing tumor. The risks and complications of radiosurgery are not negligible and include seizures, brain edema, neurological deficits, cranial nerve deficits, and the potential for radiation-induced tumors or a new malignant progression. Some authors have reported a pattern of aggressive recurrence after years of control with radiosurgery (48). For tumors of the jugular foramen, we prefer to reserve radiosurgery for the rare patient in whom the venous anatomy presents a considerable risk, such as a single functioning ipsilateral sigmoid sinus and jugular bulb or recurrent tumor, or for patients who are unsuitable for or decline surgery.

REFERENCES

Figure 31 MRI of a patient with a jugular foramen meningioma that invaded and occupied the jugular bulb. This lesion was targeted through the transjugular approach. (A) Preoperative image. (B) Postoperative image.

1. Arnautović KI, Al-Mefty O. Primary meningiomas of the jugular fossa. J Neurosurg. 2002;97:12–20. 2. Katsuta T, Rhoton AL Jr, Matsushima T. The jugular foramen. Microsurgical anatomy and operative approaches. Neurosurgery. 1997;41:149–201. 3. Lustig LR, Jackler RK. The variable relationship between the lower cranial nerves and jugular foramen tumors: Implications for neural preservation. Am J Otol. 1996;17:658–668. 4. Rhoton AL Jr. Jugular foramen. Neurosurgery. 2000;47(Suppl 3):S267–S285. 5. Ozveren MF, Ture U. The microsurgical anatomy of the glossopharyngeal nerve with respect to the jugular foramen lesions. Neurosurg Focus. 2004;17(E3):12–16. 6. Inserra MM, Pfiser M, Jackler RK. Anatomy involved in the jugular foramen approach for jugulotympanic paraganglioma resection. Neurosurg Focus. 2004;17(E6):41–44.

416

Ayoubi et al.

7. Kadri PAS, Al-Mefty O. Surgical treatment of dumbbellshaped jugular foramen schwannomas. Neurosurg Focus. 2004;17(E9):56–62. 8. Lee YB, Kim SH, Kim HT, et al. Jugular foramen neurilemmoma mimicking an intra-axial brainstem tumor. A case report. J Korean Med Sci. 1996;11:282–284. 9. Ramina R, Maniglia JJ, Fernandes YB, et al. Tumors of the jugular foramen. Diagnosis and management. Neurosurgery. 2005;57(1 Suppl): 59–68; discussion 59–68. 10. Tekdemir I, Tuccar E, Aslan A, et al. Comprehensive microsurgical anatomy of the jugular foramen and review of terminology. J Clin Neurosci. 2001;8:351–356. 11. Chao CK, Sheen TS, Lien HC, et al. Metastatic carcinoma to the jugular foramen. Otolaryngol Head Neck Surg. 2000;122:922– 923. 12. Harvey SA, Wiet RJ, Kazan R. Chondrosarcoma of the jugular foramen. Am J Otol. 1994;15:257–263. 13. Iwasaki S, Ito K, Takai Y, et al. Chondroid chordoma at the jugular foramen causing retrolabyrinthine lesions in both the cochlear and vestibular branches of the eighth cranial nerve. Ann Otol Rhinol Laryngol. 2004;113:82–86. 14. Prasanna AV, Muzumdar DP, Goel A. Lipoma in the region of the jugular foramen. Neurol India. 2003;51(1):77–78. 15. Yamazaki T, Kuroki T, Katsume M, et al. Peripheral primitive neuroectodermal tumor of the jugular foramen. Case report. Neurosurgery. 2002;51(5):1286–1289; discussion 1289. 16. Mirza S, Dutt SN, Irving RM. Jugular foramen abscess. Otol Neurotol. 2001;22(6):973–974. 17. Teixeira A, Al-Mefty O, Husain MM. Paragangliomas of the skull base. In: Berger MS, Prados MD, eds. Textbook of NeuroOncology. Philadelphia, PA: Elsevier, 2004:366–370. 18. Jackson CG, Kaylie DM, Coppit G, et al. Glomus jugulare tumors with intracranial extension. Neurosurg Focus. 2004;17(E7):45–50. 19. Borba LAB, Al-Mefty O. Paragangliomas of the skull base. Neurosurg Quart. 1995;5(4):256–277. 20. Fisch U, Fagan P, Valavanis A. The infratemporal fossa approach for the lateral skull base. Otolaryngol Clin North Am. 1984;17:513–552. 21. Jackson CG. Basic surgical principles of neurotologic skull base surgery. Laryngoscope. 1993;103:29–44. 22. Al-Mefty O, Teixeira A. Complex tumors of the glomus jugulare. Criteria, treatment, and outcome. J Neurosurg. 2002;97:1356– 1366. 23. Lowenheim H, Koerbel A, Ebner FH, et al. Differentiating imaging findings in primary and secondary tumors of the jugular foramen. Neurosurg Rev. 2006;29:1–11. 24. Oghalai JS, Leung MK, Jackler RK, et al. Transjugular craniotomy for the management of jugular foramen tumors with intracranial extension. Otol Neurotol. 2004;52:570–579. 25. Kinney SE. Glomus jugulare tumors with intracranial extension. Am J Otol. 1979;1:67–71. 26. Ramina R, Maniglia JJ, Fernandes YB, et al. Jugular foramen tumors. Diagnosis and management. Neurosurg Focus. 2004;17(2):E5. 27. Carrier DA, Arriaga MA, Gorum MJ, et al. Preoperative embolization of anastomoses of the jugular bulb. An adjuvant in jugular foramen surgery. AJNR. 1997;18:1252–1256. 28. Fisch U. Intracranial extension of jugular foramen tumors [Letter]. Otol Neurotol. 2004:25(6):1041; author reply 1041–1042.

29. Sen C, Hague K, Kacchara R, et al. Jugular foramen. Microscopic anatomic features and implications for neural preservation with reference to glomus tumors involving the temporal bone. Neurosurgery. 2001;48:838–847. 30. Al-Mefty O, Fox JL, Rifai A, et al. A combined infratemporal and posterior fossa approach for the removal of giant glomus tumors and chondrosarcomas. Surg Neurol. 1987;28:423–431. 31. Kawahara N, Sasaki T, Nibu K, et al. Dumbbell type jugular foramen meningioma extending both into the posterior cranial fossa and into the parapharyngeal space. Report of 2 cases with vascular reconstruction. Acta Neurochir. 1998;140(4):323–330. 32. Michael LM II, Robertson JH. Glomus jugulare tumors. Historical overview of the management of this disease. Neurosurg Focus. 2004;17(E1):1–5. 33. Bojrab DI, Bhansali SA, Glasscock ME III. Metastatic glomus jugulare. Long-term follow-up. Otolaryngol Head Neck Surg. 1991;104:261–264. 34. Gottfried ON, Liu JK, Couldwell WT. Comparison of radiosurgery and conventional surgery for the treatment of glomus jugulare tumors. Neurosurg Focus. 2004;17(E4):22–30. 35. Ramina R, Maniglia JJ, Paschoal JR, et al. Reconstruction of the cranial base in surgery for jugular foramen tumors. Neurosurgery. 2005;56:337–343. 36. Ramina R, Neto MC, Fernandes YB, et al. Meningiomas of the jugular foramen. Neurosurg Rev. 2006;29:55–60. 37. C ¸ okkeser Y, Brackmann DE, Fayad JN. Conservative facial nerve management in jugular foramen schwannomas. Am J Otol. 2000;21:270–274. 38. Wilson MA, Hillman TA, Wiggins RH, et al. Jugular foramen schwannomas. Diagnosis, management, and outcomes. Laryngoscope. 2005;115:1486–1492. 39. Pellet W, Cannoni M, Pech A. The widened transcochlear approach to jugular foramen tumors. J Neurosurg. 1988;69:887–894. 40. Eldevik OP, Gabrielsen TO, Jacobsen EA. Imaging findings in schwannomas of the jugular foramen. AJNR. 2000;21:1139–1144. 41. Flint D, Fagan P, Sheehy J. An intracranial vagal schwannoma without jugular foramen erosion or vagal dysfunction. Otolaryngol Head Neck Surg. 2005;132:507–508. 42. Valvassori G, Palacios E. Schwannoma of the jugular foramen. Ear Nose Throat. 1998;77:732. 43. Carvalho GA, Tatagiba M, Samii M. Cystic schwannomas of the jugular foramen. Clinical and surgical remarks [see comment]. Neurosurgery. 2000;46(3):560–566. 44. Sanna M, Falcioni M. Conservative facial nerve management in jugular foramen schwannomas [Comment]. Am J Otol. 2000;21(6):892. 45. Muthukumar N, Kondziolka D, Lunsford LD, et al. Stereotactic radiosurgery for jugular foramen schwannomas. Surg Neurol. 1999;52(2):172–179. 46. Gilbert ME, Shelton C, McDonald A, et al. Meningioma of the jugular foramen. Glomus jugulare mimic and surgical challenge. Laryngoscope. 2004;114(1):25–32. 47. Maloney TB, Brackmann DE, Lo WW. Meningiomas of the jugular foramen. Otolaryngol Head Neck Surg. 1992;106:128–136. 48. Couldwell WT, Cole CD, Al-Mefty O. Patterns of skull base meningioma progression after failed radiosurgery. J Neurosurg. 2007;106:30–35. 49. Al-Mefty O. Atlas of Meningiomas. New York:Lippincott-Raven Press, 1997.

28 Tumors of the Craniovertebral Junction Douglas Fox, Scott Wait, Steve Chang, G. Vini Khurana, Curtis A. Dickman, Volker K. H. Sonntag, and Robert F. Spetzler

arch of the atlas with the laminae of the axis. The occiput is connected to the atlas anteriorly and posteriorly by the atlanto-occipital membrane. The lateral border of the posterior membrane passes posteriorly to the vertebral artery and first cervical nerve root and may be ossified in this area (2). The tectorial membrane, alar ligaments, and apical ligament of the dens also help attach the occiput to the axis. The CVJ contains the upper spinal cord, caudal brain stem, cerebellum, and lower cranial and upper spinal nerves. The spinal cord blends into the medulla of the brainstem where the ventral rootlets emerge to form the first cervical nerve. Dorsal nerves cannot always be identified, and the sensory component of the first cervical nerve is not always present. Thus, the medulla occupies the foramen magnum. The upper cervical region is notable for the dentate ligaments, which are fibrous attachments between the spinal cord and spinal dura that are present midway between the ventral and dorsal cervical nerve rootlets. The accessory nerve has a cervical component that forms from rootlets that emerge from the spinal cord anterior to the dorsal cervical rootlets. The cerebellum surrounds the foramen magnum but normally does not occupy space or pass through the opening. The cerebellar tonsils although superior to the lateral edge of the foramen can herniate through the foramen in numerous conditions. The lower four cranial nerves are also in the region of the foramen magnum. The glossopharyngeal, vagus, and accessory nerve exit the jugular foramen and can often be identified due to their separation as they pierce the dura entering the jugular fossa. The spinal component of the accessory nerve is the only nerve that passes through the foramen magnum, arising from rootlets of the cervical spine. The spinal and cranial portions of the accessory nerve combine as it exits the jugular foramen. The hypoglossal nerve exits through a canal bearing its name lateral to the tubercle of the foramen magnum. The hypoglossal nerve often passes posteriorly to the vertebral artery and may be intimately associated with the posteroinferior cerebellar artery. The vertebral arteries and their branches constitute the major vascular structures in the region of the CVJ. The posterior spinal arteries arise from the vertebral artery, often from its extradural portion, and enter through the dural perforation with the parent artery. The posteroinferior cerebellar artery typically arises intradurally from the vertebral artery but may do so extradurally as well. The spinal arteries arise from both vertebral arteries and combine to form the anterior spinal artery, which descends through the foramen magnum. The dura of the foramen magnum is supplied by the meningeal branches of the vertebral artery. These branches are located extradurally. They also arise from the internal carotid artery circulation via the ascending pharyngeal artery, which enters through the hypoglossal canal.

INTRODUCTION The craniovertebral junction (CVJ) refers to the region of the occipital bone, which constitutes the foramen magnum and the atlas and axis vertebrae. This region encompasses the brainstem, upper cervical spinal cord, cranial and spinal nerves, and the vertebral arteries and their branches. These vital structures are closed together within a bony enclosure and are thus susceptible to lesions involving the area. Given the numerous structures involved, clinical manifestations are diverse. Physicians must be aware of the many symptoms that can relate to the region of the foramen magnum. Lesions of the CVJ can involve multiple structures, which create special problems that demand intense planning and focus from surgeons.

SURGICAL ANATOMY The bony structures of the CVJ include the occipital bone of the skull and the atlas and axis vertebra of the spine. This bony structure protects the brain stem, upper spinal cord, cranial and spinal nerves, and vertebral arteries (Fig. 1) (1). The occipital bone creates the foramen magnum through which the medulla passes. The occipital bone consists of a clival and a squamous portion. The latter includes the occipital crest, which connects with the falx cerebelli. The lateral edge of the foramen is created by the occipital condyles, which articulate the connection between skull and atlas. On the foramen, small anterior tubercles are the site of attachment of the alar ligament of the dens. The hypoglossal canal, which transmits the hypoglossal nerve, lies just lateral to this tubercle in the foramen. The atlas is the first cervical vertebra and has no vertebral body or spinous process. The atlas consists of two lateral masses connected by an anterior and posterior arch. The appearance of the axis is more typical of the rest of the vertebrae, except for the dens, which projects upward to articulate with the atlas. Synovial joints are present between the articulating surfaces of the occiput, atlas, and axis. The dens also has a synovial joint that articulates with the atlas anteriorly and with the transverse ligament posteriorly. The atlas and axis are connected by the anterior and posterior longitudinal ligaments, the cruciform ligament, and the articular capsules. The portion of the posterior longitudinal ligament that covers the cruciate ligament and dens as it extends upward to the clivus is referred to as the tectorial membrane (Fig. 2). The cruciate ligament has vertical and transverse segments that form a cross behind the dens. The transverse ligament attaches to the tubercles of the atlas and articulates with the dens. The ligamentum flavum also connects the posterior 417

418

Fox et al.

Figure 1 Cadaveric view of the craniovertebral junction. Source: Image courtesy of Mauro Ferreira, MD.

The venous anatomy in this area is also critical. The basilar plexus and occipital and marginal sinuses encircle the foramen magnum. An epidural venous plexus is located laterally in the CVJ and provides a medial border for the vertebral artery.

REGIONAL PATHOLOGY AND DIFFERENTIAL DIAGNOSES Tumors can arise from many different structures that occupy the CVJ. The rate of growth and location of the tumor determine the clinical presentation. The source of the tumor, bone or soft tissue, determines the imaging modality that will best detail the lesion. Extramedullary tumors of the CVJ are difficult to diagnose, often manifesting with symptoms that are difficult to localize. Most of these tumors are meningiomas, which are three times as common as neuromas. Meningiomas are most common in females, and this predilection holds for lesions in

Figure 2 Bony and ligamentous considerations of the axis and atlas. With permission from Barrow Neurological Institute.

the foramen magnum. Other lesions—including dermoids, teratomas, neurenteric cysts, arachnoid cysts, fatty tumors, and lesions of infectious etiology such as tuberculomas—can rarely be found at the CVJ. Neurofibromas of the CVJ are also rare except in the presence of neurofibromatosis, which can be associated with multiple lesions. Meningiomas can have many radiographic findings: hyperostosis, bony erosion, enlarged vascular channels in the bone, meningeal thickening, calcification, and hemorrhage. CT scanning is necessary to define the bony involvement of the lesion. Enlarged foramen and evidence of bony erosion may help in the determination between meningioma and neurilemmoma. MRI with and without gadolinium will normally show diffuse and intense contrast enhancement in meningiomas, and this will often be less prevalent with neurilemomas. Arterial and venous structures can also be detailed with MRI and MRA/MRV. Arterial encasement and displacement are important preoperative details, which may prevent complications during surgical resection. Extramedullary tumors are usually symptomatic at a younger age and affect males and females in equal number. The slow-growing nature of most of these lesions means that symptoms usually occur late when the lesion has reached a considerable size. Symptoms can include headaches, neck pain, cranial neuropathies, swallowing problems, balance disturbances, hydrocephalus, breathing problems, hoarseness, and hyperreflexia. Symptoms are often progressive but can have a remitting course that can confuse the evaluation. Malignant tumors in the region of the foramen magnum also must be considered. Chondrosarcomas are malignant tumors that can arise primarily from the sphenopetroclival junction or from malignant transformation of enchondromas (3). Most chondrosarcomas are low-grade lesions, but a more aggressive subtype, including dedifferentiated and mesenchymal varieties, portend a poor prognosis. Their rate of recurrence is high even if they appear histologically benign. Thus, surgeons must make every effort to achieve gross total resection. Repeat resection is warranted for residual or recurrent tumor. Recent reports suggest that the recurrencefree survival rate at 10 years is 32% (3). Radiosurgery is effective in the treatment of these lesions. Local control after proton radiotherapy is 70% at 5 years (4). Radiotherapy may increase the interval between resection

Chapter 28: Tumors of the Craniovertebral Junction

and recurrence (5). Radiosurgery combined with fractionated radiation for the treatment of chondrosarcomas is associated with high complication rates (6). Chemotherapy has demonstrated no efficacy in the treatment of these lesions. Chordomas are thought to arise from remnants of the notochord. Clival chordomas were first described by Virchow and Luschka in 1856. These tumors can occur anywhere along the neuraxis. About 50% are located in the sacrococcygeal region, 37% in the clivus/cranium, and 13% in the vertebral bodies. Half of the latter are in the cervical spine. Chordomas account for 3% to 4% of primary malignant bone tumors. They are the fourth most common primary malignancy in bone after osteosarcomas, chondrosarcomas, and Ewing sarcomas. Chordomas have a small peak in incidence that occurs during the first and second decades. Their incidence increases slowly and again peaks in the fifth to seventh decades. The mean age at diagnosis of cranial chordomas is 38 years. Cranial lesions occur equally in males and females. Three subtypes of chordomas have been described. First, conventional chordomas are characterized by a lobular architecture, vacuolated (physaliphorous) cytoplasm, and mucoid matrix. Pleomorphism and mitoses are uncommon. Second, chondroid chordomas are characterized by the Heffelfinger morphologic criteria, which consist of the presence of chondroid differentiation in near zones of more conventional chordoma (7). In a morphologically heterogeneous tumor, diffuse cytokeratin staining is necessary to diagnose chondroid chordoma (8). It occurs almost exclusively in the cranial region and at one-third the rate of conventional chordomas. Third, dedifferentiated chordomas are similar to chordomas or chondroid chordomas but with the addition of a secondary high-grade sarcomatous component. The sarcomas resemble malignant fibrous histiocytoma, but elements of fibrosarcoma, osteosarcoma, and high-grade chondrosarcoma have also been described. They compose 1% to 8% of all chordomas and may occur spontaneously or after radiation to a conventional chordoma. Their prognosis is exceedingly poor. Most patients die of tumor-related complications within one year (9). CT is the best modality to evaluate bony erosion and invasion (10). Chordomas typically lyse surrounding bone. T1-weighted MRIs with contrast show low-to-moderate signal intensity while T2-weighted MRIs show heterogeneous hyperintensity (11). Local recurrence is the rule and metastasis may occur. As many as 20% of patients have clinically evident metastasis, and as many as 40% have autopsy-evident metastasis (12). The tumors recur a mean of 2 to 3 years after treatment but can recur as late as 10 years thereafter (13). Complication and death from chordomas are primarily related to local mass effect and recurrence. As with most central neuraxis lesions, symptoms depend on the size and location of the tumor at diagnosis. In one report, visual difficulties (blurring or cranial nerve III, IV, or VI palsy) and headache were the most common symptoms occurring in 62% and 18% of patients with cranial chordomas, respectively. Sixth nerve palsy was the most common sign with an incidence of 40% (14). With the exception of patients younger than 5, who do poorly, prognosis is thought to be better with younger patients than with older. Young children typically have malignant pathologicappearing tumors and tend to die quickly (15,16). When 40 years is used as the cutoff, 5- and 10-year survival rates are 75% and 63%, respectively, for the younger group and 30% and 11%, respectively, for the older group (17). After aggressive surgical resection, proton-beam radiation is

419

the preferred treatment for these lesions. Control or shrinkage rates as high as 80% have been reported (18). Given their slow growth, chordomas respond minimally if at all to chemotherapy. Both osseous and soft-tissue tumors can be benign or malignant. Metastatic tumors are more common in this area than primary osseous tumors (19,20). Osteogenic, or bone-producing tumors, include osteoid osteomas and osteoblastomas. Osteoid osteomas are found five times more frequently in males than females and manifest with localized back pain (21). Aspirin often relieves the pain. Pain can be severe and operative excision can be curative. As many as 10% of osteoid osteomas can localize to the spine (22), usually involving the posterior elements and manifesting with intense pain. Because they are less than 2 cm, these lesions do not cause neurologic compromise. Tumors arising from the cartilaginous growth plates are called osteochondromas or enchondromas. Osteochondromas rarely involve the spine and rarely need resection. However, they can become large enough to cause enough pain and disfigurement to require surgical attention. When the vertebral column is involved, almost half of the cases involve the cervical spine with a predilection for C1 and C2. Spinal enchondromas are extremely rare, and only the rare case that undergoes malignant degeneration needs treatment. This situation occurs with multiple enchondromatosis and Maffuci syndrome, defined by enchondromatosis associated with soft-tissue hemangiomas. Eosinophilic granulomas result when the growth of reticuloendothelial cells is uncontrolled. Eosinophilic granulomas can affect the region of the CVJ. When symptomatic, it can be treated with curettage and possibly with chemotherapy and radiation (23). Multifocal eosinophilic granulomas can be treated with chemotherapy. Spontaneous regression is possible, but the lesion can also be reactivated. Conservative management with immobilization should be tried for patients presenting with pain although biopsy and curettage of the lesion are often performed to confirm the diagnosis of eosinophilic granulomas (24). Plasmacytomas can occur in the CVJ and usually present with pain and cranial nerve deficits (25). Plasmacytomas are B-cell lymphocytic tumors and precursors to the systemic disease of multiple myeloma. These tumors must be treated aggressively with multimodality treatments that include surgical resection, chemotherapy, and radiotherapy. Depending on the extent of the tumor and the completeness of resection, surgical stabilization may be needed. The success rates of painful pathological fractures treated with vertebroplasty or kyphoplasty are high (26,27). Outcome is based on the later occurrence of multiple myeloma, which has a poor prognosis. It is most common in skull base and spinal plasmacytomas (28). Intramedullary tumors involving the foramen magnum are most commonly glial in origin, with an astrocytic predominance in the pediatric population and an ependymal predominance in the adult population. Other less common tumors include cavernous malformations, hemangioblastomas, and melanocytomas. Before surgery is planned, lesions with an infectious etiology that constitute the diagnosis of myelitis must be excluded in the initial evaluation. Many tumors originating from the jugular foramen may extend to the CVJ. Most of these are schwannomas, glomus tumors, or paragangliomas. For the purposes of this review, paragangliomas and glomus tumors are considered under the term glomus tumors. Glomus tumors arise from the paraganglia cells situated near the jugular foramen. Pulsatile

420

Fox et al.

tinnitus and hearing loss are the most common presenting symptoms. Their close association with lower cranial nerves, important vascular structures, and the middle ear makes their surgical management complex, and the resultant morbidity may cause significant functional limitations. Glomus tumors have an incidence of 1 case per 30,000 individuals (29). Glomus jugulare and tympanicum tumors are the second most common temporal bone tumor. Females are affected two to five times more often than males (30–33). Familial syndromes account for 20% of glomus tumors. In those patients, multiple lesions occur 35% to 78% of the time compared to only 10% of sporadic cases (30,33–37). Patients with glomus tumors may have hypertension related to endocrinologically active tumors. One to 3% of tumors secrete a symptomatic amount of catecholamines (36,38). Norepinephrine is the most frequently produced vasoactive amine, although vasoactive substances such as dopamine have been reported (34,38–41). Catecholamine breakdown products greater than four to five times normal may be clinically apparent, thus requiring alpha- and betaadrenergic blockade before surgery to avoid labile blood pressure (38). Radiation and open surgical excision are both excellent management options, depending on an individual’s specific clinical situation and anatomy. A meta-analysis revealed a total resection rate of 88.2% at first surgery, a surgical control rate of 92.1%, a recurrence rate of 3.1%, and a mortality rate of 1.3%. Among patients who received radiosurgery as their primary treatment, 36.5% of the tumors shrank and 61.3% stabilized. Symptoms improved in 39% of patients, and 2.1% had a recurrence. The morbidity rate directly attributable to treatment was 8.5% (42).

CLINICAL ASSESSMENT Variability in presentation is the norm for lesions of the CVJ. Pain is the most common complaint for lesions, and sensory disturbances are also common. Diffuse symptoms and signs include pain and dysesthesias, cranial neuropathies, balance problems, swallowing dysfunction, gait abnormalities, nystagmus, atrophy in skeletal musculature, hyperreflexia, discoordination, and sphincter disturbance. Progressive pain that occurs at night and that worsens with motion may signify a bony lesion. Local pain may be related to compression of a specific nerve root or to the pathologic collapse of surrounding vertebral structures. Extramedullary tumors compress and often involve surrounding structures, leading to cranial neuropathies, myelopathy, hyperreflexia, atrophy of limb muscles, hydrocephalus, and pain. Intrinsic lesions often are reported by the patient as a tightness in the affected area, and reports of pain and sensory problems are common. When lesions occur in the medulla, breathing problems, nausea, and vomiting are frequent findings. Lesions with extension to the jugular fossa usually cause unilateral lower cranial nerve dysfunction. Large tumors can involve cranial nerves VII, VIII, and XII. Pulsatile tinnitus may differentiate a glomus tumor from a jugular foramen schwannoma. Other symptoms of jugular fossa lesions can include bleeding from the ear, pain in the mastoid region, and cerebellar and brainstem signs from external compression. The diagnosis of many lesions is delayed because the evaluation is focused on other organ systems that appear involved. Thus, the varied presentation of lesions in the area of the CVJ must be considered during the evaluation.

DIAGNOSTIC STUDIES Imaging of the CVJ must be able to define the soft tissue, bone, central nervous tissue, and vascular structures as well as the relationships among structures. Osseous anatomy is best identified with CT with sagittal and coronal reconstructions. Lesions causing erosion, remodeling, fracture, or even ossification can be identified with CT. The relationships among the occiput, atlas, and axis can also be evaluated with sagittal and coronal reconstructions. Given the ease of CT scanning, diagnostic radiography has mostly been abandoned. Dynamic flexion and extension xrays, however, are still used routinely to determine instability. Discussion of the stability of the CVJ is beyond the scope of this chapter. Nonetheless, understanding the relationships among the respective components of the bony CVJ is essential when pursuing tumor resections that may cause immediate or eventual instability. MRI of the CVJ is important in both diagnosis and in determination of an operative goal and plan. Relationships between the offending lesion and the critical nervous and vascular structures are well delineated with MRI. Angiography is reserved for special cases where embolization may be a preferred route prior to further treatment or there is a need to better define both the arterial and venous anatomy. Computed tomography angiography (CTA) provides excellent three-dimensional reconstruction of vascular and bony anatomy and their relationships. Soft tissue detail is not as apparent on CTA compared to MRI, but CTA may be helpful in surgical planning.

SURGICAL TECHNIQUE Transoral Approach and Its Extensions The most appropriate indication for this procedure is an extradural mass lesion in the anterior skull base (Figs. 3 and 4). The transoral approach allows access to the lower clivus, atlas, and axis. MRI and CT with sagittal and coronal reconstructions are performed preoperatively. A vascular study is often obtained to determine the vertebral anatomy, and CTA with coronal and sagittal reconstructions is preferred. Broad-spectrum antibiotics are administered before surgery and continued for 24 hours after surgery. The patient is positioned supine in a three-pin fixation system with the head slightly extended (Fig. 5). Preoperative tracheostomy is an option in patients with severe bulbar or respiratory symptoms, but it usually is unnecessary with newer-generation retraction devices. Retractors elevate the palate and uvula, retract the tongue and endotracheal tube caudally, and allow lateral exposure of the soft tissue (Fig.6). A microscope is used for visualization, and the surgeon positions himself or herself at the head of the patient. Once the patient is positioned, intraoperative fluoroscopy is used to determine spinal alignment and to confirm the cranial and caudal extent of the exposure created with the retraction system. The C1 tubercle should be palpated to identify the midline. A midline incision in the median raphe of the pharyngeal wall sharply continues through the mucosa, muscle, and anterior longitudinal ligament. Subperiosteal dissection is then used to expose the area of interest from the clivus through C2. Self-retaining retractor blades are used to hold the soft tissue laterally to obtain direct exposure of the lower clivus, C1, and C2. Thereafter, the direction of surgery depends on the pathology and its location. The inferior aspect of the C1 arch can be partially resected to identify the lateral aspect of the


421

Figure 3 Axial trajectory of surgical approaches to the craniovertebral junction and skull base (superior view). With permission from Barrow Neurological Institute.

odontoid process. The apex of the dens can be identified between the clivus and the superior aspect of the anterior arch of C1. The base of the dens is drilled posteriorly to the cortex, and a small Kerrison rongeur is used to complete the transection from the body of C2. Once the ligamentous attachments

Figure 4 Sagittal view of the transoral exposure from the lower clivus to the dens. With permission from Barrow Neurological Institute.

Figure 5 Lateral view with transoral retractor in place. With permission from Barrow Neurological Institute.

Figure 6 Sagittal view depicting transoral retractor (inset: surgeon’s view of operative field). With permission from Barrow Neurological Institute.

422

Fox et al.

are cut at the apex and the dens is free, it can be pulled caudally and ventrally. The C1 anterior arch can also be removed to expose the entire dens, which can then be removed piecemeal with a high-speed drill. Its removal must start at the apex as it becomes difficult if the dens is disconnected from C2. The transverse ligament and tectorial membrane may need to be removed to resect an extradural mass lesion. The use of standard microsurgical dissection techniques avoids the complication of a cerebrospinal fluid (CSF) leak. Closure is a single step encompassing all layers with a running 2–0 Vicryl suture. If the procedure is likely to be intradural or if an intraoperative CSF leak develops, lumbar drainage should be performed. If the dura must be closed or repaired, a fascia graft and fibrin glue are used to strengthen the closure. Once surgery is complete, an enteral feeding tube is placed under direct visualization. After one week of enteral feeding, patients are advanced to an oral diet, starting with clear liquids. The endotracheal tube remains until tongue and pharyngeal swelling decreases. The postoperative stability of the spine also must be determined. The patient must be in a rigid orthosis. If required, stabilization and fusion are performed posteriorly. When possible, fusion is limited to C1 and C2. Anterior bone grafts are associated with significant risks of displacement and infection; thus, they are rarely used. Extensions of the transoral approach can increase rostral exposure. A transoral-extended maxillotomy allows access to lesions extending from the sella to C2.

Transfacial Approaches These approaches provide a downward angle of approach to the CVJ (Fig. 7). The approaches can be divided into transnasal, transmaxillary, and transpalatal and are done with

Figure 8 Sagittal view of the suboccipital exposure. With permission from Barrow Neurological Institute.

the assistance of an experienced craniofacial surgeon. The angle of approach depends on the area of disease and its relation to the clivus, C1, and C2. Variations of the transfacial approach are based on extensions of the supraorbital bar. Retaining the medial orbits and midline nasal structures on the supraorbital bar improves access to the inferior clivus. Including the lateral orbital walls allows lateral displacement of the orbits to increase the lateral extent of the exposure. A cribriform osteotomy can be used to preserve olfaction and to help prevent a CSF leak. Transmaxillary and transpalatal approaches require a Weber–Ferguson incision with LeFort II and LeFort I osteotomies, respectively. The widened exposure of the clivus provides access to lesions that extend rostrally behind the sella and inferiorly to the upper cervical vertebrae. Transnasal approaches provide access to lesions of the anterior cranial fossa, nasopharynx, and clivus. The transnasal approach is best for clival lesions with predominant anterior growth. Large clival lesions that extend in multiple directions including posteriorly and inferiorly need a transmaxillary approach. A transpalatal approach will allow for access to the entire clivus without violating the nasal sinuses and is preferred for smaller lesions (43).

Retrosigmoid and Far Lateral Approaches

Figure 7 Approaches to the skull base. (A) Transfacial, (B) transmaxillofacial, (C) transoral/transpalatal. With permission from Barrow Neurological Institute.

The suboccipital approach is used to access lesions involving the foramen magnum (Fig. 8). The posterior aspect of the occipital bone is removed to provide access to the posterior cervicomedullary junction. Based on the extent of bony removal, access can extend from the cerebellar hemispheres to the CVJ. The lateral suboccipital craniotomy, more often described as a retrosigmoid craniotomy, allows access to the CPA. When a retrosigmoid craniotomy is extended to the foramen magnum, the posterolateral CVJ region can be accessed. The retrosigmoid approach is performed with the patient in a park-bench position or supine with the head turned. Lumbar drainage is used routinely to relax the brain and to prevent postoperative CSF leaks. Intraoperative


Figure 9

423

Intraoperative far lateral exposure showing tumor at foramen magnum displacing the lower cranial nerves.

monitoring is used during all lateral suboccipital exposures. It includes the monitoring of somatosensory-evoked potentials, auditory-evoked responses, the facial nerve, and often cranial nerves IX through XII. The ninth and tenth cranial nerves are monitored via electrodes along the endotracheal tube; this procedure should be discussed with the patient before intubation. Motor-evoked potentials are monitored when intraparenchymal tumors are resected. When large, tumors in the anterolateral region of the CVJ displace and rotate the spinal cord (Fig. 9). Spinal nerve rootlets and the spinal accessory nerve can be draped over the posterior aspect of a tumor and often must be separated from the lesion. For exposure of lateral and anterior lesions at the CVJ, far lateral and extreme lateral approaches are preferred. The far lateral approach is usually sufficient for exposure of lesions at the CVJ. The far lateral exposure requires less bony removal and may lessen the risk of future instability of the spine when compared to the extreme lateral approach. The vertebral artery requires less manipulation in the case of the far lateral exposure and thus limits the risk of vascular injury. For a far lateral exposure (Fig. 10), the patient is placed in the park-bench position with the head rotated 45 degrees and flexed anteroposteriorly and laterally (Fig. 11). The arm is held in a sling under the headholder (Fig. 12). A lumbar drain is routinely placed unless the patients have an obstructive lesion. This aids in relaxation of the brain as well as potential reduction in postoperative complication of CSF fistula. A paramedian incision is used, and midline is identified at the distal end of the incision. The muscle layers are incised with cautery. Subperiosteal dissection proceeds until the foramen magnum and C1 and C2 lamina are exposed. Fish hooks are then used to retract the soft tissue from the field of view. A Leyla bar is often used to provide retraction and is moved once the tissue has relaxed. The vertebral artery seldom needs to be dissected free. Rather, its location should be respected and care taken to avoid injury. Bony exposure is the most critical step in obtaining adequate exposure. The bone is removed with a high-speed drill and can be completed as a craniotomy or craniectomy. The opening extends from the asterion and exposes the boundaries of the sigmoid and transverse sinuses. The craniotomy continues to the foramen magnum. Rongeurs or a high-speed drill is then used to increase the lateral exposure. The lateral two-thirds of the occipital condyle can be safely resected

because the hypoglossal foramen resides in the anterior third of the condyle. Bony exposure should be as complete as needed to provide a trajectory that requires minimal, if any, retraction. Venous bleeding can be controlled by packing with hemostatic agents. The dura is opened in a curvilinear fashion and reflected laterally to allow dissection of the vertebral artery, lower cranial nerves, upper cervical rootlets, and superior dentate ligament. Sharp section of the arachnoid layer and superior dentate ligament allows gentle retraction of the upper spinal cord and lower brainstem. Use of the far lateral approach minimizes retraction during resection of lesions in this area. It also affords proximal and distal control of the vertebral artery if needed for complex vascular lesions. Removal of a lesion depends on various factors including its size, vascularity, and firmness. Microsurgical techniques are combined with ultrasonic aspiration to resect most tumors.

Figure 10 Sagittal view of the (A) transpetrosal, (B) retrosigmoid, and (C) far lateral exposure of the craniocervical junction. With permission from Barrow Neurological Institute.

424

Fox et al.

for large tumors that are difficult to manage through standard approaches. In transpetrosal and transcochlear approaches, the defect must be reconstructed with fat to help prevent CSF leaks. Tumors of the jugular fossa require a combined approach with a skull base otolaryngologist to perform the mastoidectomy and petrosectomy. When added to a retrosigmoid craniotomy, these key components provide access to lesions.

COMPLICATIONS Tumors of the skull base are inherently difficult to resect because vital structures are often distorted by or involved with the lesions. The varied complications can include CSF leaks, lower cranial nerve palsies, vertebral and spinal artery injuries, brainstem injury, hydrocephalus, and infection. Postoperative medical complications may be high in this population because these patients typically have significant disabilities before they undergo definitive procedures. Instability in the spinal axis must also be addressed and can lead to neurologic and vascular compromise that affects outcomes. Depending on the extent of resection of the CVJ, stabilization techniques may be necessary.

CONCLUSION

Figure 11 Lateral view of the patient position for the far lateral craniotomy. With permission from Barrow Neurological Institute.

When possible, tumor capsules and arachnoid planes are preserved to prevent injury to surrounding structures. Surgical adjuncts that increase exposure for tumors extending above the CVJ include the combined middle fossa approach, petrosectomy, and transcochlear approach (Fig. 3). These approaches increase exposure and may be warranted

Tumors of the CVJ are complex lesions by their location and varied pathologies. Treatment paradigms range from biopsy with adjuvant therapies through to aggressive resection with adjuvant therapies. The lesions can require various approaches depending on the size and location of the lesion, which must be tailored on a case-by-case basis. We have provided a general overview of the lesions that are present and their respective treatments. Surgical approaches have been based on the workhorses of retrosigmoid, far lateral, suboccipital, and transoral, with the variations and extensions that are possible but less frequently used. Given the complex anatomy of the region, tumors of the CVJ are difficult to treat. This chapter details the presentation; diagnosis; treatment, including approaches for surgical access; and outcomes of the tumors that can involve the CVJ. Overall, the treatment of lesions involving the CVJ remains complicated and a challenge to all skull base surgeons.

Acknowledgment We thank Mauro Ferreira, MD, for contributing the dissection photograph for Figure 1. REFERENCES

Figure 12 Anterior view of the patient position for the far lateral craniotomy. With permission from Barrow Neurological Institute.

1. De Oliveira E, Rhoton AL Jr, Peace D. Microsurgical anatomy of the region of the foramen magnum. Surg Neurol. 1985;24:293– 352. 2. Rhoton AL Jr, De Oliveira E. Anatomical basis of surgical approaches to the foramen magnum. In: Dickman CA, Spetzler RF, Sonntag VKH, eds. Surgery of the Craniovertebral Junction. New York, NY: Thieme, 1998; Ch 2. 3. Tzortzidis F, Elahi F, Wright DC, et al. Patient outcome at longterm follow-up after aggressive microsurgical resection of cranial base chondrosarcomas. Neurosurgery. 2006;58:1090–1098. 4. Hug EB, Slater JD. Proton radiation therapy for chordomas and chondrosarcomas of the skull base. Neurosurg Clin N Am. 2000;11:627–638. 5. Noel G, Feuvret L, Ferrand R, et al. Radiotherapeutic factors in the management of cervical-basal chordomas and chondrosarcomas. Neurosurgery. 2004;55:1252–1260.

Chapter 28: Tumors of the Craniovertebral Junction 6. Krishnan S, Foote RL, Brown PD, et al. Radiosurgery for cranial base chordomas and chondrosarcomas. Neurosurgery. 2005;56:777–784. 7. Heffelfinger MJ, Dahlin DC, Maccarty CS, et al. Chordomas and cartilaginous tumors at the skull base. Cancer. 1973;32:410– 420. 8. Radner H, Katenkamp D, Reifenberger G, et al. New developments in the pathology of skull base tumors. Virchows Arch. 2001;438:321–335. 9. Barnes EL, Kapadia SB, Nemzek WR, et al. Biology of selected skull base tumors. In: Janecka IP, Tiedmann K, eds. Skull Base Surgery: Anatomy, Biology, and Technology. Philadelphia, PA: Lippincott-Raven, 1997:263–292. 10. Larson TC III, Houser OW, Laws ER Jr. Imaging of cranial chordomas. Mayo Clin Proc. 1987;62:886–893. 11. Meyers SP, Hirsch WL Jr, Curtin HD, et al. Chordomas of the skull base: MR features. Am J Neuroradiol. 1992;13:1627– 1636. 12. Laws E, Thapar K. Parasellar lesions other than pituitary adenomas. In: Powell M, Lightman SL, eds. Management of Pituitary Tumors: A Handbook. New York, NY: Churchill-Livingstone, 1996:175–222. 13. Amendola BE, Amendola MA, Oliver E, et al. Chordoma: Role of radiation therapy. Radiology. 1986;158:839–843. 14. Colli BO, Al Mefty O. Chordomas of the skull base: Follow-up review and prognostic factors. Neurosurg Focus. 2001;10:E1. 15. Coffin CM, Swanson PE, Wick MR, et al. Chordoma in childhood and adolescence. A clinicopathologic analysis of 12 cases. Arch Pathol Lab Med. 1993;117:927–933. 16. Borba LA, Al Mefty O, Mrak RE, et al. Cranial chordomas in children and adolescents. J Neurosurg. 1996;84:584–591. 17. Forsyth PA, Cascino TL, Shaw EG, et al. Intracranial chordomas: A clinicopathological and prognostic study of 51 cases. J Neurosurg.1993;78:741–747. 18. Chang SD, Martin DP, Lee E, et al. Stereotactic radiosurgery and hypofractionated stereotactic radiotherapy for residual or recurrent cranial base and cervical chordomas. Neurosurg Focus. 2001;10:E5. 19. Black P. Spinal metastasis: Current status and recommended guidelines for management. Neurosurgery. 1979;5:726–746. 20. Boland PJ, Lane JM, Sundaresan N. Metastatic disease of the spine. Clin Orthop Relat Res. 1982;169:95–102. 21. Goldstein GS, Dawson EG, Batzdorf U. Cervical osteoid osteoma: A cause of chronic upper back pain. Clin Orthop Relat Res. 1977;129:177–180. 22. Raskas DS, Graziano GP, Herzenberg JE, et al. Osteoid osteoma and osteoblastoma of the spine. J Spinal Disord. 1992;5:204– 211. 23. Fernando UL, Cabezudo JM, Porras LF, et al. Solitary eosinophilic granuloma of the cervicothoracic junction causing neurological deficit. Br J Neurosurg. 2003;17:178–181. 24. Bertram C, Madert J, Eggers C. Eosinophilic granuloma of the cervical spine. Spine. 2002;27:1408–1413.

425

25. Menezes AH, Traynelis VC, Fenoy AJ, et al. Honored guest presentation: Surgery at the crossroads: Craniocervical neoplasms. Clin Neurosurg. 2005;52:218–228. 26. Fourney DR, Schomer DF, Nader R, et al. Percutaneous vertebroplasty and kyphoplasty for painful vertebral body fractures in cancer patients. J Neurosurg. 2003;98:21–30. 27. Hentschel SJ, Burton AW, Fourney DR, et al. Percutaneous vertebroplasty and kyphoplasty performed at a cancer center: Refuting proposed contraindications. J Neurosurg Spine. 2005;2:436– 440. 28. Schwartz TH, Rhiew R, Isaacson SR, et al. Association between intracranial plasmacytoma and multiple myeloma: Clinicopathological outcome study. Neurosurgery. 2001;49:1039–1044. 29. Mariman EC, Van Beersum SE, Cremers CW, et al. Analysis of a second family with hereditary non-chromaffin paragangliomas locates the underlying gene at the proximal region of chromosome 11q. Hum Genet. 1993;91:357–361. 30. Spector GJ, Gado M, Ciralsky R, et al. Neurologic implications of glomus tumors in the head and neck. Laryngoscope. 1975;85:1387–1395. 31. Alford BR, Guilford FR. A comprehensive study of tumors of the glomus jugulare. Laryngoscope. 1962;72:765–805. 32. Greer JA, Cody TR, Weiland LH. Neoplasms of the temporal bone. J Otolaryngol. 1976;5:391–398. 33. Spector GJ, Ciralsky R, Maisel RH, et al. Multiple glomus tumors in the head and neck. Laryngoscope. 1975;85:1066–1075. 34. Blumenfeld J, Cohen N, Anwar M, et al. Hypertension and a tumor of the glomus jugulare region. Evidence for epinephrine biosynthesis. Am J Hypertens. 1993;6:382–387. 35. Hodge KM, Byers RM, Peters LJ. Paragangliomas of the head and neck. Arch Otolaryngol Head Neck Surg. 1988;114:872–877. 36. Netterville JL, Jackson CG, Miller FR, et al. Vagal paraganglioma: A review of 46 patients treated during a 20-year period. Arch Otolaryngol Head Neck Surg. 1998;124:1133–1140. 37. Van Der Mey AG, Maaswinkel-Mooy PD, Cornelisse CJ, et al. Genomic imprinting in hereditary glomus tumours: Evidence for new genetic theory. Lancet. 1989;2:1291–1294. 38. Schwaber MK, Glasscock ME, Nissen AJ, et al. Diagnosis and management of catecholamine secreting glomus tumors. Laryngoscope. 1984;94:1008–1015. 39. Azzarelli B, Felten S, Muller J, et al. Dopamine in paragangliomas of the glomus jugulare. Laryngoscope. 1988;98:573–578. 40. Blumenfeld JD, Cohen N, Laragh JH, et al. Hypertension and catecholamine biosynthesis associated with a glomus jugulare tumor. N Engl J Med. 1992;327:894–895. 41. Troughton RW, Fry D, Allison RS, et al. Depression, palpitations, and unilateral pulsatile tinnitus due to a dopamine-secreting glomus jugulare tumor. Am J Med. 1998;104:310–311. 42. Gottfried ON, Liu JK, Couldwell WT. Comparison of radiosurgery and conventional surgery for the treatment of glomus jugulare tumors. Neurosurg Focus. 2004;17:E4. 43. Beals SP, Joganic EF, Hamilton MG, et al. Posterior skull base transfacial approaches. Clin Plast Surg. 1995;22:491–511.

Section 3 Tumor-Specific Considerations

29 Squamous Cell Carcinoma of the Nasal Cavity and Paranasal Sinuses Patrick Sheahan, Snehal G. Patel, and Jatin P. Shah

INTRODUCTION

to rise with the length of exposure and in inverse proportion to the age at first exposure. The risk is thought to be multiplicative in nickel refinery workers who also smoke (15). Exposure to wood dust is well-established as a risk factor for adenocarcinoma of the ethmoid sinuses (16). The association between wood dust and SCC is less strong, however, Luce noted an eightfold increased risk of SCC in carpenters and joiners who had worked in the wood manufacturing industry for more than 15 years (17). Other authors have also reported a moderately increased risk for SCC among patients with long histories of wood dust exposure (18,19). There is some evidence that while exposure to hardwood dust is strongly associated with adenocarcinoma, exposure to softwood dust alone may be a risk factor for SCC (20). Other occupations associated with increased risk of nasal or paranasal sinus SCC include electrical workers (16), bakers and pastry cooks (17), grain millers (17), textile workers (21), farm workers (17), construction workers (17). Occupational exposure to asbestos (22) and formaldehyde (22,23) has also been implicated. Unlike SCC of the rest of the upper aerodigestive tract, tobacco and alcohol are traditionally not considered to be major risk factors for sinonasal cancer. However, there is good evidence from several studies that heavy smokers as well as snuff users do have an increased risk of SCC of the nose and paranasal sinuses (7,19,24,25). The risk would appear to be greatest in recent smokers (26). Radiation exposure would appear to be another factor capable of causing sinonasal SCC. Thorotrast (thorium dioxide) is a radiopaque material that was used as a contrast agent as recently as 1954. Upon injection into the maxillary sinus, it decays to mesothorium with concomitant release of alpha, beta, and gamma rays. Thorotrast appears to be retained in the maxillary sinus for life and radioactivity reaches a peak after 15 years. The link between thorotrast and sinonasal SCC is based on a series of case reports documenting maxillary sinus SCC in patients who had undergone thorotrast injection 10 to 21 years prior (7). In addition, workers in the erstwhile radium dial painting industry had been reported to have an increased risk of sinonasal SCC. The radium was believed to have been absorbed via the oral mucosa in workers who used to lick paintbrushes for painting the dials of watches. Sinonasal tumors were reported to arise after a median latent period of 34 years (7,27). The link between SCC and sinonasal papillomas is now well established. Sinonasal papillomas are benign epithelial neoplasms composed of well-differentiated, ciliated columnar or respiratory epithelium, with variable squamous metaplasia. Three types are identified: inverted papilloma, columnar cell (oncocytic Schneiderian) papilloma, and exophytic papilloma (28). Inverted papilloma is characterized by its endophytic or “inverted” growth pattern into the underlying

Squamous cell carcinoma (SCC) of the nose and paranasal sinuses is rare. Large-scale experience in management of this tumor is not available from a single institution. The relevance of most reported outcome data is further diluted by inclusion of tumors of various histologic types. Meaningful interpretation of results of treatment of sinonasal cancers is thus difficult because many published reports include other tumors such as adenocarcinoma and esthesioneuroblastoma, which have different biologic behavior and prognosis compared to SCC (1–4). A clear understanding of the biologic behavior of sinonasal SCC is therefore essential to treatment selection. Cancers arising in this anatomic area tend to remain relatively asymptomatic until late in their course, and thus tend to present at an advanced stage, by which time they have frequently extended to involve vital adjacent structures such as the orbit, skull base, or brain. Treatment of these tumors is therefore difficult and outcomes remain suboptimal in spite of advances in imaging and therapeutic techniques. The objective of this chapter is to review the epidemiology, etiology, pathology, natural course, and outcome of treatment of SCC arising in the nose and paranasal sinuses.

EPIDEMIOLOGY Malignant neoplasms of the nasal cavity and paranasal sinuses account for only 0.2% to 0.8% of all carcinomas, and only 2% to 3% of those in the upper aerodigestive tract (5,6). The incidence of cancers of the sinonasal tract in the United States and Europe is estimated at 0.3–1/100,000 per year (7,8). A higher incidence has been reported in Japan, Indonesia (9), and parts of Africa (10), with incidence rates of up to 2.6/100,000 per year reported in Japan (10). SCC is the most common histologic type (4,8,11,12) but precise incidence statistics for sinonasal SCC in the United States are not available. In Denmark, the annual incidence of SCC of the nose and paranasal sinuses has been estimated at 2.5 cases per million (11). The peak age incidence for SCC is in the sixth and seventh decades, however, adults of all ages may be affected (11). Sinonasal SCC occurring in a child is exceptionally rare. Males are more commonly affected than females by a factor of 1.5 to 2 (7,8).

ETIOLOGY A significantly increased risk of sinonasal SCC has been reported in nickel-refining workers (7,13,14). This risk appears 429

430

Sheahan et al.

stroma but with an intact basement membrane, and is the type that is usually associated with SCC, however, the columnar cell variety may also be associated with malignant transformation (29). The risk with exophytic papillomas is much smaller (28). The etiology of inverted papillomas is not clear. Human papillomavirus does appear to have a causal relationship with exophytic papilloma, but has been demonstrated in only a minority of cases of inverted papilloma (28,30,31). SCC of the nasal cavity and paranasal sinuses is more common in patients with a history of sinonasal papilloma. In fact, a recent study from Denmark suggested that 15% of cases of sinonasal SCC are associated with papillomas (32). The reported incidence of carcinoma associated with inverted papilloma has varied from as low as 2% (33) to as high as 53% (34), however, it is likely that incidences at the extremes of this range are due to either referral bias, papilloma misclassification, or failure to recognize small areas of SCC within inverted papillomas. More recent series report an SCC incidence of 5% to 26% (35–40). The carcinoma may be either synchronous (i.e., the diagnosis of SCC is established at the same time as that of the inverted papilloma), or metachronous (i.e., appearing at an area from where an inverted papilloma had previously been removed). Metachronous lesions usually appear at the same site as the inverted papilloma, and the mean time to their appearance is 52 months (41). On the basis of a review of personal series of 63 cases of inverted papilloma, along with a review of a further 3058 cases reported in the literature, Mirza estimated the incidence of synchronous SCC to be 7.1%, and of metachronous SCC to be 3.6% (41). While it is possible that some cases of metachronous SCC are due to failure to diagnose SCC, which was present at the time of the original resection, malignant transformation of inverted papillomas to SCC has been demonstrated histologically (42). Several other substances and occupations have been associated with sinonasal adenocarcinoma, but not with SCC. These include chromium, boot and shoe manufacturers in the leather tanning industry, textile workers, and isopropyl alcohol manufacturers (7).

PATHOLOGY AND NATURAL HISTORY SCC is the most commonly encountered epithelial malignancy of the sinonasal region and comprises 55% to 60% of all sinonasal tumors (8,10,43). It most commonly arises from the maxillary sinus (50–71%), followed by the nasal cavity (20–32%) and ethmoid sinus (10–15%) (8,10,43–45). Tumors of the sphenoid (2%) and frontal (≤1%) sinuses are exceedingly rare. Within the maxillary and ethmoid sinuses, males are more commonly affected than females, however, in the nasal cavity, females predominate over males (8). Most are poorly differentiated and keratinizing but about 20% are of the nonkeratinizing type (46). The histologic grade of these tumors does not predict outcome as reliably as their anatomic extent although poorly differentiated tumors are generally more aggressive in their course. Between 10% and 20% of paranasal sinus SCCs are very poorly differentiated (47). These so-called undifferentiated or anaplastic carcinomas are rapidly growing and produce early metastases. They are more evenly distributed throughout the maxillary sinus, nasal cavity, and ethmoid sinus, and consequently comprise a higher proportion of nasal and ethmoid carcinomas (8,10). They are said to be more common in females (8) and have a poorer prognosis than SCC (8,48). These tumors should be distinguished from other poorly differen-

tiated carcinomas, such as sinonasal undifferentiated carcinoma and malignant melanoma. A well-differentiated but nonkeratinizing variant is occasionally seen (2–11%). Names used to describe this variety include transitional cell carcinoma, cylindrical cell carcinoma, and Ringertz carcinoma (46). It is characterized by a plexiform or ribbon-like growth pattern and invades the underlying tissue with a smooth, well-delineated border. It is evenly distributed in all sinonasal locations (8,49). This variant is more common in males, in whom it occurs at a younger age than in females (8) and has a better prognosis than SCC (8). Rare variants of nasal and paranasal sinus SCC include papillary SCC, which is an exophytic carcinoma with a papillary configuration composed of thin fingers of tumor surrounding fibrovascular cores (46); verrucous carcinoma, which is an extremely well-differentiated lesion associated with minimal invasiveness (50); basaloid SCC, which consists of predominantly pleomorphic, basaloid-appearing cells, and is associated with a dismal prognosis (51); spindle cell carcinoma, an aggressive tumor with carcinomatous and spindle cell components (35), and adenosquamous carcinoma, a tumor containing areas of squamous carcinoma and areas of glandular differentiation, which is generally also considered to be highly aggressive (52). Small tumors contained within the sinus of origin are generally asymptomatic, thus presentation with early stage cancers is unusual. As the tumor enlarges, it comes into contact with the bony walls of the sinus and quickly destroys the bone, spreading into adjacent sinuses and other nearby structures, thence giving rise to symptoms and signs. Carcinomas of the maxillary sinus spread superiorly into the orbit, inferiorly into the hard palate and lower alveolus, medially into the nasal cavity, and anteriorly into the subcutaneous tissues of the cheek. More advanced tumors demonstrate posterior extension into the pterygomaxillary fissure, pterygoid plates, and infratemporal fossa, as well as superior extension up to the skull base and anterior spread through the cheek skin. Ethmoid tumors usually show early spread laterally into the orbit, superiorly through the cribriform plate into the anterior cranial fossa, and inferomedially to the nasal cavity and nasal septum. The lamina papyracea and cribriform plates are quickly destroyed; however, the periorbita and dura are more resistant to tumor invasion. Spread of tumor into the maxillary antrum is also commonly seen. Posterior spread into the sphenoid sinus and nasopharynx occurs later but spread to the frontal sinus is less common. Cancers of the nasal cavity most commonly (85%) arise from the lateral nasal wall. Less commonly, they may arise from the nasal septum, the nasal floor, or nasal vestibule (53). Tumors arising within the nasal cavity quickly expand to fill the nasal cavity. From there, extension occurs laterally into the ethmoid and maxillary sinuses, posteriorly into the nasopharynx, and superiorly to the skull base. Carcinomas arising from the nasal vestibule commonly involve the anterior septum and columella. From there, tumors have a propensity for aggressive spread along the periosteum of the premaxilla and maxilla. SCC of the vestibule is unusual in that it has been reported to demonstrate a more indolent natural course (54,55). Ohngren’s line (Fig. 1) represents an imaginary plane running from the medial canthus of the eye to the angle of the mandible, which has traditionally been used to separate sinonasal tumors into those with good and bad prognosis based on their location. Tumors situated below, medial, and anterior to this plane cause early symptoms, thus leading

Chapter 29: Squamous Cell Carcinoma of the Nasal Cavity and Paranasal Sinuses

431

Figure 1 Ohngren’s line.

to diagnosis at an early stage, are easier to control surgically and are thus associated with a better prognosis. On the other hand, tumors situated above, lateral, and posterior to the plane present late due to a lack of early symptoms, have a higher incidence of invasion of the orbit, intracranial compartment, and infratemporal fossa, are more difficult to remove en bloc, and are therefore associated with a worse prognosis (56). Ohngren’s line formed the basis for the initial version of the UICC staging system for maxillary sinus carcinoma. Lymphatic drainage from the anterior part of the nose and paranasal sinuses is by means of facial lymphatic vessels into cervical lymph nodes at levels I and II. Drainage from the posterior part of the nose and paranasal sinuses is to retropharyngeal nodes, and thence to upper deep cervical nodes (57–59). Regional lymph node metastasis at the time of diagnosis is generally less common with SCC of the nose and paranasal sinuses than with SCC at other head and neck primary sites, with reported incidences ranging from 4% to 16% (60–68). There are little data on the incidence of occult disease in the clinically N0 neck since elective neck dissection is not routinely practiced for these tumors. Distant metastases at the time of presentation are rare and generally only seen in patients with regional disease (66).

with proptosis secondary to invasion into the orbit. A mass at the medial canthus, diplopia, and orbital pain may also be present. Paralysis of medial gaze secondary to invasion of the medial rectus muscle occurs later. Presenting symptoms of maxillary tumors include a facial mass, a palatal or upper alveolus mass, loosening of teeth or denture problems, facial or dental pain, or proptosis. Epiphora is caused by tumor invading or obstructing the nasolacrimal duct. Anesthesia of the cheek, upper lip, and upper teeth signifies involvement of the infraorbital nerve. Tumors of the frontal sinus usually present with a mass above the eye and symptoms of obstructive sinusitis. Signs indicative of more advanced tumors of the paranasal sinuses include trismus (invasion of the masticator muscles), numbness of the chin (mandibular nerve invasion at the foramen ovale), visual loss (secondary to optic nerve invasion), and cervical lymphadenopathy. The new appearance of any unilateral nasal symptoms, especially in an adult patient should always prompt careful endoscopic examination of the nasal cavity to rule out a tumor. Proptosis; external swelling of the face, periorbital region, nose, palate, or upper gum; unilateral epiphora; and facial numbness are all signs that should be regarded as highly suspicious of a tumor (49,69), and should always be promptly investigated by a CT scan.

SYMPTOMS AND SIGNS

WORKUP AND ASSESSMENT

Squamous carcinomas of the nose and paranasal sinuses generally do not produce any symptoms or signs until they have expanded to a significant size and/or have extended outside the bony confines of the sinus cavity. Nasal tumors most commonly present with unilateral nasal obstruction, rhinorrhea, or epistaxis. Bleeding may take the form of blood-stained nasal secretions. As the tumor enlarges, it may present into the nostril and/or cause external nasal deformity. Occasionally, patients may present with clear rhinorrhea secondary to cerebrospinal fluid leakage. Ethmoid tumors usually present

Workup of patients with sinonasal SCC should include a full history and physical examination, relevant laboratory studies, imaging studies, tissue diagnosis, and appropriate consultations from other specialists.

Imaging Radiologic imaging studies are an essential component in the diagnosis, staging, and follow-up of sinonasal malignancies. Computed tomography (CT) scan gives a good initial overview of the tumor’s location with excellent bone detail.

432

Sheahan et al.

Because the paranasal sinuses and nasal cavity are mucosal lined bony chambers, CT is helpful in determining whether a tumor remains confined within these natural boundaries or has eroded through the surrounding bone. CT provides details of the extent of local bone invasion, and is particularly useful in assessing the lamina papyracea, orbital floor, ethmoid roof, cribriform plate, pterygoid plates, hard palate, and skull base (70). The clinician should be aware that the brain, meninges, orbit, and facial soft tissues are inadequately evaluated when a bone algorithm and bone windows are used; when evaluating cancers in this region, soft tissue windows are also essential. However, magnetic resonance imaging (MRI) is preferred for accurate soft tissue details, so that most patients benefit from both CT and MRI. In comparison to CT, MRI allows a better distinction of tumor from adjacent soft tissue. MRI is particularly useful for determining invasion of the orbital contents, dura, brain, cavernous sinus, and infratemporal fossa (70). MRI may also be better for assessing carotid artery invasion, and newer techniques such as MRA permit intraluminal carotid assessment without the associated risks of direct angiography. MRI also differentiates fluid collection secondary to an obstructed sinus from tumor. CT and MRI therefore complement one another in the assessment of sinonasal tumors. CT provides excellent bone detail, while MRI offers better soft tissue imaging. In assessing CT and MRI scans, both coronal and axial views should be studied. Coronal views are particularly helpful in assessing invasion of the cribriform plate, lamina papyracea, and palate; while axial views are helpful in delineating posterior extension into the sphenoid sinus, orbital apex, pterygoid plates, pterygomaxillary fissure, or infratemporal fossa, and anterior extension into the cheek. In cases where tumor extension is present into the infratemporal fossa, coronal views allow assessment of the foramen ovale. The hallmark of sinonasal carcinomas on CT scans is bone destruction, which is seen in approximately 80% of scans at initial presentation (71). The tumor itself is usually of soft tissue density. Intravenous contrast causes tumor enhancement, however, inflamed mucosa may enhance similarly. Because of this, differentiation between tumor, mucosal thickening, and obstructed secretions may be difficult on CT scan. This distinction is greatly facilitated by MRI. Malignant tumors tend to be of intermediate signal on T2weighted image (72). In contrast, inflamed mucosa, retained secretions, and benign polyps generally have a high signal on T2-weighted images. In addition, even in cases where the secretions become increasingly inspissated and the signal intensity on T2-weighted scans decreases, the tumor can usually still be distinguished by its typical heterogeneity, in contrast to the smooth homogenous appearance of secretions (72). It should be noted that some schwannomas, minor salivary gland tumors, and inverted papillomas may also be bright on T2-weighted imaging. With gadolinium injection on T1-weighted images, tumors enhance less intensely than inflamed mucosa, while secretions do not enhance. Benign tumors extending intracranially tend to be more heterogeneous on MRI than malignant tumors (73). Mucoceles may be distinguished from tumors by peripheral enhancement (74). In addition to the above, imaging of the neck, using either CT or MRI, should be performed to assess for regional metastases. Chest CT and/or positron emission tomography (PET scan) may be performed to rule out distant metastases, however, in the absence of cervical metastases, distant metastases are very unlikely.

Tissue Diagnosis Biopsy Once the site and extent of the tumor has been identified, tissue diagnosis is required. A fundamental principle should be to obtain representative tissue by the least invasive method possible. Avoiding an open procedure is advantageous in preventing (i) the disturbance of intact anatomic structures and boundaries, (ii) possible tumor contamination of normal tissues, and (iii) disturbance of the tumor’s location and obscuration of its margins, making future localization and surgical treatment significantly more difficult. An optimal procedure for biopsy of sinonasal malignancies is through an endoscopic approach through the nares. This approach offers several advantages, including excellent visualization, low morbidity, and minimal alteration of the tumor and its surrounding structures. Even small, lateral tumors within the maxillary sinus may be accessible with the creation of a middle meatal antrostomy, visualization with a 45 degree or 70 degree endoscope, and biopsy using a long curved giraffe instrument. If the tumor presents itself at the nasal vestibule, biopsy in the office may be considered, however, in general, biopsy in the operating room is preferred as this allows the surgeon to deal with any bleeding that may arise. It is important to ensure by clinical and radiologic examination that the mass is neither contiguous with the cerebrospinal fluid space nor highly vascular. If the mass compresses easily or appears vascular then further imaging should be obtained prior to biopsy. In rare cases where a maxillary sinus tumor is not accessible transnasally with the endoscope, a canine fossa puncture can be combined with endoscopic visualization and biopsy. With the availability of endoscopic techniques, Caldwell– Luc antrostomy is rarely necessary. Open biopsy should be avoided at all costs as this violates tissue planes in the subcutaneous tissue and skin, which will then have to be sacrificed with a margin of normal skin at the time of definitive surgical resection.

Review of Histology Review of the biopsy slides by an experienced histopathologist is strongly advised in patients with sinonasal carcinoma. As discussed previously, the histologic type of the tumor is a powerful predictor of outcome and may also influence treatment selection. Accurate typing of tumor histology is therefore crucial prior to commencing treatment, especially since it has been recognized that up to one in five cases have the diagnosis modified after expert review (75).

Multidisciplinary Consultation The care of the patient with sinonasal SCC requires a multidisciplinary approach. In addition to the head and neck surgeon, consultations may be necessary from (i) neurosurgery, if the patient is going to require a craniofacial resection; (ii) plastic/reconstructive surgery, if the patient is going to require a free flap to reconstruct the surgical defect; (iii) prosthodontics, if the patient is going to require a dental obturator or other prostheses; (iv) radiation oncology; (v) medical oncology, if chemoradiation therapy is a consideration; (vi) internal medicine, if the patient has medical conditions that need to be optimized prior to surgery; (vii) ophthalmology, if the tumor has affected the eye or if treatment has the potential to impact visual function.

Chapter 29: Squamous Cell Carcinoma of the Nasal Cavity and Paranasal Sinuses Table 1 AJCC/UICC Classification of Tumors of Maxillary Sinus

Table 3

Tx T0 Tis T1

N0 N1

T2

T3

T4a

T4b

Primary tumor cannot be assessed No evidence of primary tumor Carcinoma in situ Tumor limited to the antral mucosa with no erosion or destruction of bone Tumor causing bone erosion or destruction, including extension into the hard palate and/or middle meatus, but excluding extension to posterior wall of maxillary sinus or pterygoid plates Tumor invades any of the following: bone of posterior wall of maxillary sinus, subcutaneous tissues, floor or medial wall of orbit, pterygoid fossa, or ethmoid sinuses Tumor invades anterior orbital contents, skin of cheek, pterygoid plates, infratemporal fossa, cribriform plate, frontal or sphenoid sinuses Tumor invades any of the following: orbital apex, dura, brain, middle cranial fossa, cranial nerves other than maxillary division of the trigeminal nerve, nasopharynx and/or clivus

STAGING Tumors of the maxillary sinus, nasal cavity, and ethmoid sinus are staged according to the TNM system of the UICC/AJCC. The latest edition of this was published in 2002 (76). The staging system for maxillary sinus tumors is given inTable 1, and the staging system for tumors of the nasal cavity and ethmoid sinus is shown in Table 2. Currently, no staging system exists for tumors arising primarily in the frontal or sphenoid sinus. Neck staging is the same as for other head and neck sites (Table 3). Stage grouping is given in Table 4. Of note, for maxillary sinus carcinomas, extension into the infratemporal fossa, pterygoid plates, or skin of cheek, is now classified as T4a, and not as T3, as had been the case with the 1997 classification.

PROGNOSTIC FACTORS The following factors have consistently been shown to be significant prognostic factors for local control and survival in patients with sinonasal SCC: T stage (2,77–79), N stage (65,66,68,79,80), intracranial extension (2,80,81), dural invasion, orbital invasion (2,79), adverse tumor histology (80). Other adverse factors include advanced age (68,79). SCC asTable 2 AJCC/UICC Classification of Tumors of Nasal Cavity or Ethmoid Sinus Tx T0 Tis T1 T2

T3 T4a

T4b

Primary tumor cannot be assessed No evidence of primary tumor Carcinoma in situ Tumor restricted to any one subsitea , with or without bony invasion Tumor invading two subsites in a single region or extending to involve an adjacent region within the nasoethmoidal complex, with or without bony invasion Tumor extends to invade the medial wall or floor of the orbit, maxillary sinus, palate, or cribriform plate Tumor invades any of the following: anterior orbital contents, skin of nose or cheek, minimal extension to anterior cranial fossa, pterygoid plates, frontal or sphenoid sinuses Tumor invades any of the following: orbital apex, dura, brain, middle cranial fossa, cranial nerves other than maxillary division of the trigeminal nerve, nasopharynx and/or clivus

a Subsites are: right ethmoid sinus, left ethmoid sinus, nasal septum, nasal floor, nasal lateral wall, nasal vestibule.

N2a N2b N2c N3

433

AJCC/UICC Classification of the Neck Nodes

No regional lymph node metastasis Metastasis in single ipsilateral lymph node, 3 cm or less in greatest dimension Metastasis in single ipsilateral lymph node, greater than 3 cm but no more than 6 cm in greatest dimension Metastasis in multiple ipsilateral lymph nodes, none more than 6 cm in greatest dimension Metastasis in multiple contralateral or bilateral lymph nodes, none more than 6 cm in greatest dimension Metastasis in lymph node greater than 6 cm in greatest dimension

sociated with inverted papilloma has been associated with a better prognosis (82). Cervical metastasis is associated with a particularly poor prognosis, with reported 5-year survivals of no better than 10% to 15% (63,65,68). The most common cause of treatment failure in sinonasal SCC is local recurrence. Patients with local recurrence will usually die of disease. Hence, optimizing local control is of paramount importance in sinonasal SCC. Regional failure in cervical lymph nodes, even without elective neck treatment, is uncommon. Neck failure has been shown to be significantly associated with nodal stage (2,81). Among series containing only patients with SCC, distant metastases are reported in 11% to 13% (77,83,84). Factors associated with increased risk of developing distant metastases include high T stage (2), nodal metastases (81), intracranial invasion (2,81), and orbital invasion (81).

TREATMENT A wide variety of management approaches to sinonasal SCC exists. Most of these approaches involve some combination of surgery, radiotherapy, and/or chemotherapy. It is notable that there are no randomized trials comparing outcomes between the different treatment types. The treatment options for sinonasal SCC can be summarized as follows: (i) primary surgery, usually combined with postoperative radiotherapy; (ii) primary radiotherapy, reserving surgery for salvage of persistent disease; (iii) neoadjuvant chemotherapy, followed by various combinations of further chemotherapy, radiotherapy and/or surgery; (iv) radiotherapy, with or without neoadjuvant or concurrent chemotherapy, combined with “conservative surgery.” Comparing outcomes of the various treatment modalities for sinonasal SCC is problematic for the following reasons: (i) Sinonasal SCC is rare, so few centers have contemporary experience with a large volume of cases. (ii) Owing to the complex anatomy of the paranasal sinuses and surrounding structures, sinonasal SCCs comprise a heterogenous group of tumors with differing locations and extents of invasion. (iii) Most of the published series include a substantial number of tumors with histologies other than SCC. These other histologies, including adenocarcinoma and esthesioneuroblastoma, have very different biologic behavior and different Table 4 Stage 1 Stage 2 Stage 3 Stage 4

AJCC/UICC Stage Grouping for Sinonasal Carcinoma T1N0M0 T2N0M0 T3N0M0, or T1–T3, N1, M0 T4 or T1–T3, N2–N3 or M1

434

Sheahan et al.

implications for prognosis (1–4). (iv) In many series, patients with large advanced tumors, considered to be unresectable, are over-represented in the radiotherapy alone group, thus introducing a considerable bias when making comparisons between patients who did and did not undergo surgery (66). Prior to making a decision regarding treatment, a thorough assessment of the patient and tumor is essential. Particular points to consider include the following: 1. Tumor involvement of the cribriform plate, +/– intracranial extension. If this is present, then a craniofacial resection will be required to ensure adequate clearance. 2. Tumor invasion of the orbit. If this is present, then orbital exenteration may be required (see below). 3. Tumor invasion of the palate. If the palate or upper alveolus is involved, then infrastructure maxillectomy will be required. The patient will need to see a prosthodontist prior to surgery in order to have a dental obturator prefabricated for immediate fitting following tumor resection in order to close the palatal defect, unless the surgical defect is to be reconstructed using a free flap. 4. Extension into the sphenoid sinus. Tumor in the sphenoid sinus may also invade the internal carotid artery and cavernous sinus, as well as the optic nerve. These findings may render the tumor inoperable (see below). 5. Posterior extension into the masticator space, +/– involvement of the foramen ovale and middle cranial fossa. 6. Involvement of the skin. If skin is resected, the resulting defect may be very difficult to repair using local flaps. Tenuous repairs are at high risk of breaking down during postoperative radiotherapy, leading to a facial fistula. This is a particular problem near the medial infraorbital rim where the skin is usually very thin. Therefore, a free flap will be necessary in most patients that require resection of the skin. 7. Presence of cervical metastases. Patients with cervical metastases have a dismal prognosis; this should be borne in mind prior to proceeding with surgery if resection of the primary tumor is likely to entail significant morbidity. 8. General medical condition of the patient, and ability to withstand a prolonged surgery, including possible craniotomy and free flap reconstruction.

illectomy with complete ethmoidectomy +/– sphenoid exenteration; suprastructure maxillectomy +/– orbital exenteration; infrastructure maxillectomy; or total maxillectomy +/– orbital exenteration, resection of the pterygoid plates, and masticator space dissection. Resection of the skin of the cheek, nasal septum, and contralateral ethmoid and sphenoid sinuses may also be required. Our preferred approach is via a modified Weber–Ferguson incision respecting the nasal subunits, with or without subciliary (Fig. 2) or Lynch extension (Fig. 3). This may be combined with a bicoronal incision for craniofacial resection as required. For more limited infrastructure maxillectomies, a peroral or midfacial degloving approach (Fig. 4) may also be used. Medial maxillectomy combined with ethmoidectomy (Fig. 5) is indicated for tumors primarily arising in the ethmoid sinuses or nasal cavity with minimal invasion of the maxillary antrum and no invasion of the floor of the nose. The extent of surgery will vary with the location of the disease; however, in general this involves en bloc resection of the medial part of the maxilla from the infraorbital rim to the lower part of the piriform aperture, the inferior turbinate, the lacrimal bone, the lamina papyracea, and the anterior ethmoid air cells. In most cases, it should be possible to preserve the infraorbital nerve, as well as the roots of the upper teeth. The resection usually includes a variable proportion of the frontal process of the maxilla and the nasal bone. Tumors of the ethmoid sinuses that have suspected or obvious orbital invasion require discussion regarding orbital exenteration (see below). Access for medial maxillectomy is usually by means of a lateral rhinotomy incision with Lynch extension. Upper lip split should not be necessary. The anterior ethmoid artery is an important surgical landmark, which enters the orbit through or just below the frontoethmoidal suture line

In the discussion that follows, emphasis has been placed on studies reporting only on patients with SCC rather than those that included patients of various other histologies.

Surgery Surgery with postoperative radiotherapy continues to be the mainstay of treatment for sinonasal SCC in most centers throughout the world. Careful preoperative workup is essential to assess the resectability of these tumors, as well as the surgical approach. Surgical extirpation of tumors of the nose and paranasal sinuses can be challenging not just from the technical standpoint but also because the tumor and its treatment is likely to result in significant functional and cosmetic morbidity. Preoperative assessment of these tumors should thus take into consideration the site of origin and extension of the tumor, and the performance status of the patient, as the extent of resection in many patients may mandate a craniotomy, and/or reconstruction using a free flap. The ideal surgical procedure would be an en bloc resection of the entire tumor with negative microscopic margins. However, in practice, this is sometimes difficult to achieve. The type of surgery will depend on the location and extent of invasion by the tumor. Surgery may involve medial max-

Figure 2

Weber–Ferguson incision with subciliary extension.


Figure 5

Figure 3 Weber–Fergusion incision with Lynch extension

roughly 14 to 18 mm posterior to the anterior lacrimal crest. This marks the junction between the lamina papyracea and skull base and so guides the upper limit of resection within the orbit. The posterior ethmoid artery is roughly 10 mm posterior to the anterior ethmoid artery and is a useful landmark for the optic nerve, which enters the orbit 4 to 7 mm posterior, but may be as close as 2 mm lateral (Fig. 6). The inferior turbinate is typically resected en bloc with the rest of the specimen; however, the middle turbinate is usually attached to the skull base and so has to be removed separately.

435

Medial maxillectomy.

For tumors that reach the level of the cribriform plate, craniofacial resection should be performed. This allows en bloc resection of the roof of the ethmoid and cribriform plate as well as the middle turbinate with the specimen. For tumors that arise in the nasal cavity, resection of the nasal septum is often necessary. In such cases, the anterior part of the septum should be preserved if possible in order to maintain support for the tip of the nose. Depending on the posterior extent of the tumor, exenteration of the posterior ethmoid and sphenoid sinus may be required. This should be performed to the level of the skull base, which is flat and relatively easy to identify in this region, however, owing to the proximity of such vital structures as the optic nerve and internal carotid

Anterior ethmoidal a. Ethmoid sinuses

Posterior ethmoidal a.

Optic nerve

Figure 4 Midfacial degloving incision.

Figure 6

Orbit with AEA, PEA, and optic nerve.

436

Sheahan et al.

artery, it is usually difficult to achieve monobloc resection of tumor involving these areas. For tumors that invade intracranially but remain extradural, craniofacial resection should be performed. Craniofacial resection also facilitates satisfactory tumor removal from the posterior ethmoid and sphenoid sinuses. Of note, the dura provides a good barrier against tumor invasion. In cases where the tumor erodes the cribriform plate and extends intracranially, then the dura should be resected and reconstructed in order to ensure clear margins in this area. Gross dural invasion may also be resected and reconstructed. Brain invasion with SCC is generally considered a contraindication to tumor resection since outcomes are dismal even in selected patients who are amenable to craniofacial resection (1). Suprastructure maxillectomy (Fig. 7) involves resection of the superior part of the maxilla, but preserving the hard palate and lower alveolus. It is indicated for ethmoid tumors with more extensive invasion of the maxillary antrum, or for maxillary tumors that do not involve the floor of the maxillary sinus or roots of the upper teeth. Typically, the infraorbital nerve and most of the orbital floor are removed. Depending on the extent of orbital invasion, orbital exenteration may also be performed. Suprastructure maxillectomy may also be combined with ethmoidectomy with or without craniofacial resection, removal of the posterior wall of the maxillary antrum, as well as variable removal of the pterygoid plates and contents of the pterygomaxillary fissure and pterygopalatine fossa. Because most of the orbital floor is removed, one of the main problems with suprastructure maxillectomy is postoperative support of orbital contents in cases where the globe has been preserved. In cases where the orbital periosteum and medial canthal ligament are fully preserved, then these structures may provide adequate support to suspend the or-

Figure 7

Suprastructure maxillectomy.

Figure 8 Infrastracture maxillectomy.

bital contents postoperatively. However, partial or complete removal of either or both of these may necessitate further measures to provide support. Possible reconstructive options in this situation include free fascia lata grafts, vascularized temporalis myofascial flaps, vascularized calvarial bone flaps (85), and free flaps. However, in practice, it is often difficult for pedicled flaps to fully bridge the defect, while access for free flap vessels is also difficult (86). Free bone grafts are not always the best option, as most of these patients will require postoperative radiotherapy, which will prevent neovascularization of the bone graft. Thus, such patients are at high risk of subsequently developing osteoradionecrosis. Similarly, in cases where the orbital floor is reconstructed using prosthetic materials, radiotherapy leads to a high incidence of implant exposure and infection (87). Infrastructure maxillectomy (Fig. 8) is indicated for tumors arising on the upper alveolus, hard palate, or lower part of the maxillary sinus that do not involve the ethmoid or the roof of the maxillary sinus. The resection includes part or all of the upper alveolus and hard palate. The floor of the orbit is spared. Depending on the extent of resection, the operation may be performed via a peroral, midfacial degloving, or upper lip-split approach. Reconstruction of the hard palate and upper alveolus is most easily accomplished using a prosthetic dental obturator. Although split-thickness skin grafts are frequently used on the inner surface of the cheek flap, our practice is to avoid using these as superior granulation and mucosalization of the cavity is achieved without them. Total maxillectomy (Fig. 9) is indicated for larger tumors, which involve most of or fill the maxillary sinus. Resection may also involve ethmoidectomy, orbital exenteration, and/or removal of the pterygoid plates and contents of the pterygomaxillary fissure and masticator space. The usual approach is via a modified Weber–Ferguson incision with subciliary extension. Care must be taken raising the


437

chiasm involvement, bilateral cavernous sinus involvement, or internal carotid artery involvement (1,63). The results for surgery alone for sinonasal carcinoma are generally disappointing, with typical 5-year local control rates of 40% (66) and 5-year overall survival rates of 20% (65). Most centers now offer postoperative radiotherapy to all but the earliest stage tumors. Using combined treatment, typical 5-year local control rates range from 49% to 67% (65,81,88), with 5-year disease-specific survival of 64% reported (60), and reported 5-year overall survival ranging from 42% to 66% (81,89).

Radiotherapy

Figure 9

Total maxillectomy.

cheek flap as tumor may easily erode through the thin bone of the canine fossa to involve the subcutaneous tissues. Medially, the incisor teeth arising from the premaxilla may be preserved if this area is not involved by tumor. Laterally, the maxilla may be separated from the zygoma at the zygomatomaxillary suture, preserving the prominence of the cheek. However, for more extensive tumors, it may be necessary to divide the frontal process of the zygoma and the zygomatic arch in order to allow removal of the entire infraorbital rim. For tumors invading through the posterior wall of the maxillary sinus, the pterygoid plates should also be removed. This may be achieved by insinuating an osteotome behind the hard palate into the pterygoid fossa to fracture the top of the pterygoid plates from the skull base, and using a Mayo scissor to divide the pterygoid muscles from the lateral pterygoid plate. Alternatively, improved exposure of the region of the pterygoids may be achieved by either dividing the insertion of the temporalis muscle from the coronoid process of the mandible and then removing the coronoid process or by performance of a mandibular swing (60). Failure to mobilize the pterygoid plates will usually result in a fracture across the back wall of the maxillary sinus when the specimen is removed, necessitating piecemeal removal of tumor in the pterygomaxillary fissure and masticator space. Total maxillectomy usually results in a large defect, which requires reconstruction for both functional and cosmetic reasons. Reconstruction may be either with a prosthetic obturator or a free flap. Advantages of a free flap include the provision of support for the orbital contents, as well as avoidance of a large cavity, which will require fastidious irrigation and cleaning. In this situation, the best option is usually a rectus abdominis free flap, which provides a large volume, as well as epithelial surfaces, which may be used intraorally and/or intranasally (86). Contraindications to surgery are not universally accepted; however, they generally include the following: presence of distant metastases, presence of brain invasion or extensive intracranial invasion, bilateral optic nerve or optic

Substantial uncertainty surrounds the optimal radiation volumes and techniques for paranasal SCC. However, the results for radiotherapy alone in these tumors have been disappointing, with reported 5-year local control rates ranging from 14% to 53% (53,65,68,88,90–93) and 5-year survival rates ranging from 0–16% (43,65,80,88,92–93). The drawbacks of radiotherapy alone as treatment for paranasal sinus SCC are not only due to lack of efficacy, but also to the considerable potential for adverse effects to nearby structures such as the retina, optic nerves, optic chiasm, and frontal lobes. The incidence of serious visual complications with conventional radiotherapy has been reported to range between 16% to 66% (79,94–96). Visual complications include severe keratitis, cataracts, chronic tearing, retinopathy, optic neuropathy, optic atrophy, and blindness. Unilateral blindness has been reported to occur in up to 20% to 35% of patients with sinonasal cancers undergoing radiotherapy, with bilateral blindness occurring in up to 6% to 10% (97–100). The risk to the eye would appear to be particularly high for ethmoid carcinomas (94,97,101), although high incidences of ipsilateral blindness (30%) have also been reported in patients with maxillary carcinomas (102,103). The most common scenario for radiation-induced blindness is radiation retinopathy of an eye irradiated to a high dose; however, contralateral blindness secondary to optic neuropathy has been reported to occur in 8% of cases (99). The risk of ocular complications appears to be related to the total dose of radiation, as well as to the treatment ports and treatment schedule. Parsons found retinopathy to occur in 100% of eyes receiving radiation doses of greater than 65 Gy, in 50% of eyes exposed to doses between 45 Gy and 55 Gy, while the risk was very low in eyes receiving less than 45 Gy (104). The risk of optic neuropathy appears to be increased in optic nerves receiving doses of 60 Gy or greater, in fractions of 1.9 Gy or larger (105). With modern radiotherapy techniques, acceptable maximum doses to vital optic structures are generally accepted to be 45 Gy to the retina, and 54 Gy to the optic nerve and chiasm (81,106). It has been suggested that hyperfractionation may reduce the risk of ocular complications (101). Other complications of radical radiotherapy for sinonasal carcinomas include brain necrosis (0–12%) (48,95,99,103), bone necrosis (0–8%) (2,80,92,99,103), hearing loss (2.5–8.5%) (2,3), hypopituitarism (4–5%) (99,103), trismus (5–12%) (3,92,103), facial fistula (2–3%) (92,103), saddle-nose deformities (95), diplopia (95), meningitis (48), and radiationinduced tumors (99). In addition, a study by Meyers et al. suggested that memory impairment may occur in up to 80% of patients treated with radiotherapy to the skull base by conventional techniques (107). One of the limiting factors of radiotherapy alone for sinonasal carcinoma is the high dose of radiation that is necessary, leading to an increased risk of bilateral blindness. The

438

Sheahan et al.

advantage of using combined surgery and radiotherapy is that a lower dose of radiation may be used, thus reducing the risk of bilateral blindness. In the University of Florida experience, bilateral blindness did not develop in any patient undergoing combined treatment; thus, the treatment policy at that institution was changed from one of irradiation alone, to one of surgical resection followed by radiotherapy (99). Recent evolutions in radiotherapy techniques may allow for improved efficacy and reduced adverse effects. Modern techniques such as intensity-modulated radiotherapy (IMRT) allow for very homogenous dose distributions with a sharp dose fall-off gradient, and thus an increased therapeutic ratio. Using these techniques, it is possible to irradiate the tumor with doses of up to 70 Gy, while keeping the maximum doses to the surrounding ocular structures to acceptable levels (95,106,108). In a review of 127 patients with sinonasal carcinomas of various histologies, treated over five decades, with conventional radiotherapy, 3-D conformal radiotherapy, and IMRT, Chen reported that while there was no difference in local control or survival according to the decade of treatment, a significant decrease in complications was seen. The incidence of late grade 3/4 ocular toxicity among patients treated with conventional radiotherapy, 3-D conformal radiotherapy, and IMRT was 20%, 9%, and 0%, respectively (95). Other authors also reported a low incidence of serious complications with 3-D conformal (2,109,110) and IMRT techniques (81,106), as well as with the use of proton beam irradiation (111). Duthoy prospectively recorded toxicity data of 39 patients with sinonasal carcinoma who were treated postoperatively with IMRT to a median dose of 70 Gy. Visual impairment developed in five (15%), two (6%) of whom had grade 3 impairment. There were two cases of brain necrosis (108). It should be noted that most patients in these studies underwent radiotherapy in a postoperative setting, and there are little data on the efficacy of IMRT as a primary treatment modality. Furthermore, radiation-induced optic neuropathy and retinopathy typically take 2 to 5 years to develop, thus the follow-up periods of some of these studies are relatively short (108,109). In the future, further dose reductions to normal tissue may be possible with the use of proton beam radiotherapy (112). When radiotherapy is combined with surgery, the optimal sequencing of these modalities has been debated. Dirix reporting on patients with various histologies, found local control and disease-free survival to be better in patients receiving postoperative compared to preoperative radiotherapy (2). Other authors have reported an improved survival in patients who received preoperative radiotherapy compared with those who received postoperative radiotherapy (113,114), however, a higher complication rate is reported among patients undergoing surgery after radiotherapy (1,113). In cases where patients with sinonasal carcinoma fail initial treatment with radiotherapy, the results of salvage surgery are disappointing. The present treatment policy at the University of Toronto consists of primary high-dose radiotherapy with curative intent to the equivalent of 70 Gy (63). Curran reported on 95 patients with sinonasal cancer of various histologies who failed initial treatment at the University of Toronto. Of these, 17 had distant metastases, 20 were deemed to have unresectable disease, and 24 were medically unfit or refused surgery at the time of the recurrence. Thus, 34 of the 95 patients proceeded to undergo salvage surgery. Within this group, the disease-specific 5-year survival was 47%, and the overall 5-year survival was 35%. The 5-year survival for patients with SCC was only 25%. Patients undergoing primary radiotherapy for sinonasal cancer in that

institution now undergo repeat imaging, examination under anesthesia, and biopsy of any suspicious areas 6 to 8 weeks after completion of radiotherapy, in an effort to detect residual or recurrent disease at an earlier stage (63,115). Of note, radiotherapy alone has been reported to be an effective treatment with good cosmetic outcomes for carcinomas of the nasal vestibule (54,116,117).

Chemotherapy Combined with Surgery and Radiotherapy In 1970, Sato et al. reported promising treatment outcomes in sinus carcinoma with a combination of necrotomy and radiotherapy concurrent with intra-arterial chemotherapy, for preventing functional and cosmetic loss caused by radical surgery. Complete response was achieved in 67% (118). Since then, several authors, mostly from Japan, have reported on the use of neoadjuvant chemotherapy prior to definitive treatment with surgery, radiotherapy, and/or chemoradiotherapy. The most common agents used are intra-arterial cisplatin (83,119) or 5-fluorouracil (5-FU) (61,67,83). These are administered by selectively introducing an intra-arterial catheter into the maxillary artery via the femoral artery or the superficial temporal artery (61,67). Tumor staining may be examined by CT angiography, and if insufficient, angiography of other arteries, including the facial (119) or ascending pharyngeal (120) may be added. The usual dose of cisplatin varies from 100 to 150 mg/m2 (119,120), given for 2 to 4 cycles (77,119,120). Disadvantages of intra-arterial chemotherapy include possible irregular distribution of the infused drugs due to the existence of alternate feeding vessels, and possible damage to branches of the external carotid artery, which may be required at a later date for free flap reconstruction of skull base defects. Thus, the use of systemic chemotherapy has been favored by other authors (121,122). Various combinations of chemotherapy, radiotherapy, and surgery have been reported. Samant reported on the use of concurrent intra-arterial chemotherapy combined with 50 Gy of radiotherapy, followed by surgery on 19 patients with sinonasal malignancy (14 with SCC). Five-year disease-free and overall survivals were both 53% (120). Lee reported overall survival of 72.7% on a subgroup analysis of 19 patients with stage III and IV paranasal sinus carcinoma (11 with SCC) who were treated with preoperative systemic chemotherapy with cisplatin and 5-FU, followed by radical surgery, followed by postoperative chemoradiotherapy using hydroxyurea and 5-FU, with a median radiation dose of 60 Gy (121). Konno reported a 5-year survival of 75% among 32 patients with maxillary SCC treated by concurrent radiotherapy (60 Gy) and intra-arterial 5-FU and cisplatin followed by radical surgery (67). There are little data regarding the outcome of patients treated with primary chemoradiotherapy without surgery. High progression-free rates have been reported even in patients with skull base invasion (123), although further validation of this initial observation is required in larger series. Among most reported series of patients undergoing treatment with primary chemoradiotherapy, surgery has been reported to be a significant predictor of improved outcome, although such series are biased by the inclusion of patients with unresectable disease in the nonsurgical group (77).

Combinations of Chemotherapy, Radiotherapy, and “Debulking” Surgery The combination of conservative “debulking” surgery with radical radiotherapy in order to avoid the morbidity of radical surgery has been reported. Debulking surgery has been


defined as removal of all macroscopic tumor without additional removal of noninvolved bony structures. This removal is usually accomplished piecemeal, and may be performed via a lateral rhinotomy or, more commonly, by a sublabial approach (79,124). During the initial debulking surgery, a cavity may be established, which permits subsequent further debulking in the clinic (124). Alternatively, patients may undergo repeat debulking via an open approach (84,124). Using this approach, 5-year local control rates of 59% to 67% have been reported (79,124). Of note, Jansen reported that the combination of debulking surgery with radiotherapy offered a significant survival advantage when compared to radiotherapy alone (79). The absence of gross residual disease as judged by the surgeon after debulking has also been associated with a better outcome (124). The addition of neoadjuvant or concurrent chemotherapy to radiotherapy and debulking surgery has also been reported. Using these approaches, 5-year disease-free and overall survival rates of 88% and 72% to 76% respectively, have been reported (83,84). An interesting study conducted by Knegt et al. in 1985 described the use of surgical debulking and low-dose radiotherapy followed by repeated topical chemotherapy using 5-FU and necrotomy in patients with paranasal sinus cancer. The 5-year survival rate for SCC and undifferentiated carcinoma of the maxillary sinus was 52%. This protocol would appear to be particularly suited to adenocarcinomas. In 2001, the same group reported on this treatment for 62 patients with adenocarcinoma of the ethmoid sinus complex. Forty-nine patients had T3/T4 tumors, and 40% had anterior skull base involvement. The use of low-dose radiotherapy was omitted in the latter part of the study. Impressive 5- and 10-year disease-specific survival rates of 87% and 74% were reported (125). However, there have been little further data regarding the use of this treatment protocol for sinonasal SCC from other institutions. The obvious advantage of debulking surgery over conventional surgery is that it avoids the functional and cosmetic losses associated with more radical surgery, and avoids orbital exenteration. In addition, tumor volume is well known to be an important predictor of the radiocurability of a tumor. Reduction of the tumor volume by debulking surgery may thus enhance the likelihood of cure by radiotherapy (61). Kawashima found gross tumor volume after debulking surgery to be a more important predictor of local control than T-classification (61). Furthermore, radiotherapy dose of greater than 60 Gy was found to be an important predictor of local control in patients with small volumes of gross residual tumor. On the other hand, recent advances in free flap reconstructive surgery can now provide satisfactory restoration of masticatory and phonatory function after radical surgery, as well as good cosmetic results (86). Nibu suggested that the improved survival in patients treated in latter years at their institution may be in part due to their undergoing en bloc tumor resection rather than the piecemeal resections from earlier years (83). It should also be borne in mind that in studies reporting on the efficacy of debulking surgery, the presence of gross residual disease (122,124) was found to be a significant adverse prognostic indicator.

Comparison of Various Treatment Protocols Despite the numerous diverse combined treatment protocols, there are little data comparing the outcomes of any of these. Tiwari reported on 38 patients treated with curative intent for maxillary sinus SCC. Twenty-nine were treated by surgery

439

followed by radiotherapy, and had a 5-year survival rate of 64%. Nine were treated by sequential chemoradiotherapy, and had a 37% 2-year survival. The authors suggested that surgery followed by radiotherapy remains the treatment of choice for maxillary SCC (60). Shiga reported on 50 patients with maxillary sinus SCC. All received neoadjuvant intraarterial cisplatin. In 25 patients, this was followed by radical surgery with postoperative radiotherapy, while in the other 25, this was followed by concurrent chemoradiotherapy. There was no significant survival difference between the two groups, however, grade 3/4 toxicity was higher in the concurrent chemoradiotherapy group (119). Isobe found no significant difference between three different treatment protocols (77). The authors practice en bloc surgical resection of the tumor with appropriate reconstruction followed by adjuvant radiation with or without chemotherapy as the standard of care for surgically resectable tumors of the paranasal sinuses at their institution.

Management of the Orbit The specific indications for orbital preservation and exenteration have evolved over the past forty years and remain a controversial subject. Although in the 1950s, the orbit was almost routinely exenterated for any extension of maxillary sinus carcinoma toward the orbital floor, the emerging consensus is that the orbit can often be preserved without compromising overall survival or local control of disease. This approach has been complicated, however, by differing criteria that have been used to determine the indications for orbital preservation. Carrau examined 58 patients with bony orbital invasion by SCC of the sinonasal tract and found that 3-year survival was not affected by orbital preservation in the absence of orbital soft tissue invasion. The authors concluded that the orbit may be spared if the full thickness of the periorbita is not breached by tumor (126). McCary and Perry concluded that periorbital invasion does not necessarily indicate a need for orbital exenteration. They found that preoperative radiation therapy followed by intraoperative frozen section and selective resection of the involved periorbita may conserve the eye without a compromised outcome (127,128). Tumor extension through the periorbita does not necessarily condemn the eye to exenteration. Tiwari has noted that a thin fascial layer exists around the periorbital fat that is distinct from the periorbita and believes that invasion of this layer should determine the need for exenteration (129). Quatela has taken an even more aggressive approach by resecting intraorbital tumor with involved orbital fat and extraocular muscles off Tenon’s fascia surrounding the globe, and then preserving or “banking” the residual, nonfunctional globe in vivo (130). Care must be taken to avoid attempting orbital preservation at the potential cost of decreased local disease control and survival. Our approach is to resect involved periorbita and preserve the orbital contents in cases of periorbital involvement. Indications for orbital exenteration include the following: (i) invasion of orbital fat, (ii) invasion of extraocular muscles, (iii) invasion of bulbar conjunctiva or sclera, (iv) involvement of the orbital apex. Besides oncologic outcome, the other main point of contention with orbital preservation in sinonasal malignancy is functional outcome. Stern reported that only one-sixth of patients undergoing resection of the orbital floor retained useful function of the ipsilateral eye (131). Notably, no reconstructive efforts were made to repair lost orbital support. When efforts are made to reconstruct orbital floor defects,

440

Sheahan et al.

improved functional outcomes are seen (132). Among patients with sinonasal malignancy undergoing orbital preservation, between 86% and 91% of patients are reported to retain a useful functioning eye, however, 41% to 50% of these develop one or more ocular sequelae (132,133). The most common problem reported is abnormal globe position (enophthalmos or hypophthalmos). Abnormalities of globe position occur more frequently and are of greater severity in patients undergoing subtotal or total orbital floor resection, particularly if reconstructive efforts to support the globe are not employed. Only a minority of patients with abnormal globe position develop diplopia, and in most of these, diplopia is transient (132). Other ocular sequelae include epiphora and lid malposition (ectropion, canthal dystopia). The incidence of epiphora would appear to be decreased in patients undergoing lacrimal stenting (132). Of note, eye problems are more common when postoperative radiation therapy is administered (131,133).

Management of the Neck Clinically positive metastatic disease in the neck is generally managed with neck dissection, the type depending on the extent and location of the nodal metastases. Careful examination of parotid and facial lymph nodes should be performed prior to surgery as these nodes may occasionally also be involved. In addition, imaging should be performed to investigate the status of retropharyngeal nodes. Lymph nodes in level I, even if not clinically palpable, should always be included in the specimen in patients who have metastatic disease at other levels. Postoperative radiation therapy to the neck is indicated for multiple positive nodes, any single node >3 cm in size, or extracapsular spread. The management of the clinically negative neck remains controversial. Among patients with SCC who do not undergo elective neck treatment, the incidence of subsequent neck failure in the clinically N0 neck is generally reported to be between 5% to 14% (64,67,68,79,84). In patients who do develop regional failure, the results of salvage treatment are generally disappointing (64). Neck failure is also reported to be strongly associated with the development of distant metastases and decreased survival (64,68). Two widely quoted studies have reported incidences of neck failure as high as 29% to 33% (92,103), however, it should be noted that both of these studies included tumors of various histologies. Furthermore, the incidence of isolated neck failure in patients who were clinically N0 at the time of initial presentation in these studies was only 14% and 18% respectively. Of note, a recent systematic review and meta-analysis reported a weightedaverage incidence of neck failure for nasal cavity SCC of 18.1%; however, of the 23 studies reviewed, 20 reported exclusively on patients with SCC of the nasal vestibule or septum (134). Some studies have supported the use of elective neck irradiation for SCC of the maxillary sinus (64,89,103,135) and nasal cavity (134). Le reported that neck failure in his series was not predicted by primary tumor control, but was effectively prevented by elective neck irradiation (64). Jiang also found radiotherapy to effectively prevent neck failure (103). Despite this, given the low risk of neck conversion reported in most series, most institutions do not electively treat the clinically N0 neck. Exceptions may apply in cases where the sinonasal tumor encroaches upon areas of increased risk for lymphatic spread such as the nasopharynx or soft palate.

Management of Skin or Cartilage Invasion Cancers of the nasal vestibule and maxillary sinus and ethmoid sinus may commonly involve the skin of the cheek

or the nose. Even more common is for carcinomas of the maxillary sinus to break through the front wall of the maxillary sinus and involve the subcutaneous tissues, but not the skin itself. In such cases, care should be taken to raise a thin cheek flap, in order to leave an adequate margin of soft tissue over the tumor. When this is possible, skin resection is not necessary. On the other hand, when the skin is involved, then resection of the involved portion of skin is essential. An adequate margin of normal skin will need to be taken around the involved area. This usually creates a considerable soft tissue defect, which will require a free flap for reconstruction. In cases where there is also a large volume surgical defect, this is usually best accomplished using a rectus abdominis free flap. In cases where the volume of the defect is small, then a radial forearm free flap is better suited. Involvement of the skin and cartilages of the external nose is more problematic, as surgical resection will create a considerable cosmetic defect. On occasion, total rhinectomy, along with resection of the upper lip will be necessary. If possible, some or all of the nasal bones should be preserved in order to facilitate future placement of osseointegrated implants. Subsequent surgical reconstruction of the nose is extremely difficult, and the most satisfactory results are oftentimes obtained using a nasal prosthesis. If the upper lip is resected, this is reconstructed using advancement flaps, while a dental obturator is used to reconstruct the premaxilla. The surgeons should be wary of attempting to undertake immediate surgical reconstruction of the external nose unless they can be certain that the margins are negative. In many instances, delayed reconstruction may be a better option and this may need to be deferred until after adjuvant treatment is complete.

Role of Endoscopic Surgery Recent advances in endoscopic sinus surgery has led to increasing interest in the application of this technique to benign sinonasal and skull base tumors. More recently, the use of endoscopic surgery for malignant tumors has also been reported. Shipchandler reported on 11 patients with SCC of the nose or paranasal sinuses who were treated with endoscopic resection, combined with craniotomy in four cases. Eight patients received adjuvant chemotherapy, radiotherapy, or both. Two patients developed local recurrence and underwent repeat endoscopic resection. At a median followup of 31 months, 10 patients (91%) were alive with no evidence of disease (136). Buchmann et al. reported on 63 patients who underwent surgery for nasal/paranasal sinus cancer (26 with SCC). Of these, 27 were operated using a classic open approach, whereas 36 were operated using endoscopic techniques, either alone or in combination with a midfacial degloving approach or subfrontal craniotomy. There was no significant difference in survival outcomes. However, it is unclear which histologic tumor types underwent what type of surgery, and which cases received salvage treatment with surgery, radiotherapy, and chemotherapy (137). At the current time, caution should be exercised in treating patients with SCC of the paranasal sinuses with endoscopic resection since incomplete excision in inexperienced hands is liable to adversely impact outcomes even if the patient is subsequently subjected to more radical treatment.

OUTCOMES AND PROGNOSIS The prognosis of nasal cavity and paranasal sinus SCC continues to be suboptimal. Tumors generally present at an advanced stage. Many reported 5-year local control and


survival rates are misleading as studies include patients with histologies other than SCC. In addition, most series include patients with tumors of all stages and all sites; however, by and large, the bulk of patients with SCC in any given series have advanced (T3/T4) primary tumors, with clinically negative (N0) necks. Among series reporting results only for patients with SCC after surgery with postoperative radiotherapy, typical 5year local control, disease-specific, and overall survival rates are 49% to 67% (65,81,88), 64% (60), and 42% to 66% (81,89) respectively. After combined treatment with chemotherapy, radiotherapy, and surgery, reported 5-year local control rates range from 59% to 88% (77,84,119,121,124), with 5-year disease-specific survival rates ranging from 52% to 75% (77,84,119,120,121), and 5-year overall survival rates ranging from 52% to 75% (67,77,83,119). Clearly, despite the varied treatment protocols, the results of treatment continue to be disappointing and novel approaches are needed to improve both oncologic and functional outcomes in these patients.

REFERENCES 1. Ganly I, Patel SG, Singh B, et al. Craniofacial resection for malignant paranasal sinus tumors: Report of an international collaborative study. Head Neck. 2005;27:575–584. 2. Dirix P, Nuyts S, Geussens Y, et al. Malignancies of the nasal cavity and paranasal sinuses: Long-term outcome with conventional or three-dimensional conformal radiotherapy. Int J Radiat Oncol Biol Phys. 2007;69:1042–1050. 3. Ozsaran Z, Yalman D, Baltalarli B, et al. Radiotherapy in maxillary sinus carcinomas: Evaluation of 79 cases. Rhinology. 2003;41:44–48. 4. Bhattacharyya N. Factors affecting survival in maxillary sinus cancer. J Oral Maxillofac Surg. 2003;9:1016–1021. 5. Osguthorpe JD. Sinus neoplasia. Arch Otolaryngol Head Neck Surg. 1994;120;19–25. 6. Ali S, Tiwari RM, Snow GB, et al. Incidence of squamous cell carcinoma of the head and neck: Report of 1000 cases. J Laryngol Otol. 1986;100:315–327. 7. Rousch GC. Epidemiology of cancer of the nose and paranasal sinuses: Current concepts. Head Neck Surg. 1979;2:3–11. 8. Robin PE, Powell DJ, Stansbie JM. Carcinoma of the nasal cavity and paranasal sinuses: Incidence and presentation of different histological types. Clin Otolaryngol. 1979;4:431–456. 9. Rifki N. Problems of paranasal sinus malignancy in Indonesia. ORL Indonesia. 1985;16:175–180. 10. Muir CS, Nectoux J. Descriptive epidemiology of malignant neoplasms of nose, nasal cavity, middle ear and accessory sinuses. Clin Otolaryngol. 1980;5:195–211. 11. Grau C, Jakobsen MH, Harbo G, et al. Sino-nasal carcinoma in Denmark 1982–1991— A nationwide survey. Acta Oncol. 2001;40:19–23. 12. Myers LL, Nussenbaum B, Bradford CR, et al. Paranasal sinus malignancies: An 18-year single institution experience. Laryngoscope. 2002;112:1964–1969. 13. Andrews P. Carcinomas of the nose and paranasal sinuses in former employees of a sinter plant at Copper Cliff, Ontario. J Otolaryngol. 1983;12:255–256. 14. Sunderman FW Jr, Morgan LG, Andersen A, et al. Histopathology of sinonasal and lung cancers in nickel refinery workers. Ann Clin Lab Sci. 1989;19:44–50. 15. Andersen A, Berge SR, Engeland A, et al. Exposure to nickel compounds and smoking in relation to incidence of lung and nasal cancer among nickel refinery workers. Occup Environ Med. 1996;53:708–713. 16. Hemelt M, Granstrom C, Hemminki K. Occupational risks for nasal cancer in Sweden. J Occup Environ Med. 2004;10:1033– 1040. 17. Luce D, Leclerc A, Morcet JF, et al. Occupational risk factors for sino-nasal cancer: A case-control study in France. Am J Ind Med. 1992;21:163–175.

441

18. Demers PA, Kogevinas M, Boffetta P, et al. Wood dust and sinonasal cancer: Pooled reanalysis of twelve case-control studies. Am J Ind Med. 1995;28:151–166. 19. Fukuda K, Shibata A. Exposure-response relationships between woodworking, smoking, or passive smoking, and squamous cell neoplasms of the maxillary sinus. Cancer Causes Control. 1990;1:165–168. 20. Hernberg S, Westerholm P, Schultz-Larsen K, et al. Nasal and sinonasal cancer. Connection with occupational exposures in Denmark, Finland, and Sweden. Scand J Work Enviro Health. 1983;9:315–326. 21. Luce D, Gerin M, Morcet JF, et al. Sinonasal cancer and occupational exposure to textile dust. Am J Ind Med. 1997;32:205–210. 22. Luce D, Leclerc A, Begin D, et al. Sinonasal cancer and occupational exposures: A pooled analysis of 12 case-control studies. Cancer Causes Control. 2002;13:147–157. 23. Hayes RB, Raatgever JW, de Bruyn A, et al. Cancer of the nasal cavity and paranasal sinuses, and formaldehyde exposure. Int J Cancer. 1986;37:487–492. 24. Brinton LA, Blot WJ, Becker JA, et al. A case-control study of cancers of the nasal cavity and paranasal sinuses. Am J Epidemiol. 1984;119:896–906. 25. ‘t Mannetje A, Kogevinas M, Luce D, et al. Sinonasal cancer, occupation, and tobacco smoking in European women and men. Am J Ind Med. 1999;36:101–107. 26. Hayes RB, Kardaun JW, de Bruyn A. Tobacco use and sinonasal cancer: A case-control study. Br J Cancer. 1987;56:843–846. 27. Littman MS, Kirsh IE, Keane AT. Radium-induced malignant tumors of the nose and paranasal sinuses. Am J Roentgenol. 1978;131:773–785. 28. von Buchwald C, Bradley PJ. Risks of malignancy in inverted papilloma of the nose and paranasal sinuses. Curr Opin Otolaryngol Head Neck Surg. 2007;15:95–98. 29. Kaufman MR, Brandwein MS, Lawson W. Sinonasal papillomas: Clinicopathologic review of 40 patients with inverted and oncocytic schneiderian papillomas. Laryngoscope. 2002;112:1372– 1377. 30. Buchwald C, Norrild B. Human papillomavirus in sinonasal papillomas and in normal nasal mucosa. Papillomavirus Report. 1997;8:99–104. 31. Weiner JS, Sherris D, Kasperbauer J. Relationship of human papillomavirus to Schneiderian papillomas. Laryngoscope. 1999;109:21–26. 32. Buchwald C, Lindeberg H, Pedersen BL, et al. Human papilloma virus and p53 expression in carcinomas associated with sinonasal papillomas: A Danish epidemiological study 1980– 1998. Laryngoscope. 2001;111:1104–1110. 33. Batsakis JG. The pathology of head and neck tumors: Nasal cavity and paranasal sinuses. Head Neck Surg. 1980;2:410– 419. 34. Yamaguchi KT, Shapshay SM, Incze JS, et al. Inverted papilloma and squamous cell carcinoma. J Otolaryngol. 1979;8:171– 178. 35. Lund VJ. Granulomatous diseases and tumors of the nose and paranasal sinuses. In: Kennedy DW, Bolger WE, Zinreich SJ, eds. Diseases of the Sinuses. Diagnosis and Management. London: BC Decker, 2001:85–106. 36. Philips PP, Gustafson RO, Facer GW. The clinical behavior of inverting papilloma of the nose and paranasal sinuses: Report of 112 cases with review of the literature. Laryngoscope. 1990;100:463–469. 37. Segal K, Atar E, Mor C, et al. Inverting papilloma of the nose and paranasal sinuses. Laryngoscope. 1986;4:394–398. 38. Smith O, Gullane PJ. Inverting papilloma of the nose: Analysis of 48 patients. J Otolaryngol. 1987;16:154–156. 39. Benninger MS, Roberts JK, Sebek BA. Inverted papillomas and associated squamous cell carcinomas. Otolaryngol Head Neck Surg. 1990;103:457–461. 40. Schlosser RJ, Mason JC, Gross CW. Aggressive endoscopic resection of inverted papilloma: An update. Otolaryngol Head Neck Surg. 2001;25:49–53. 41. Mirza S, Bradley PJ, Acharya A, et al. Sinonasal inverted papilloma: Recurrence, and synchronous and metachronous malignancy. J Laryngol Otol. 2007;121:857–864.

442

Sheahan et al.

42. Lawson W, Kaufman MR, Biller HF. Treatment outcomes in the management of inverted papilloma: An analysis of 160 cases. Laryngoscope. 2003;113:1548–1556. 43. Lewis JS, Castro EB. Cancer of the nose and paranasal sinuses. J Laryngol Otol. 1972;86:255–262. 44. Weinstein GS, Harvey R. Open surgical resection of sinus neoplasms. In: Kennedy DW, Bolger WE, Zinreich SJ, eds. Diseases of the Sinuses. Diagnosis and Management. London: BC Decker, 2001:405–422. 45. Spiro JD, Soo KC, Spiro RH. Squamous cell carcinoma of the nasal cavity and paranasal sinuses. Am J Surg. 1989;158:328– 332. 46. Pilch BZ, Bouquot J, Thompson LD. Squamous cell carcinoma. In: Barnes L, Eveson JW, Reichart P, Sidransky D, eds. World Health Organization Classification of Tumours. Lyon: IARC Press, 2000:15–17. 47. Weber AL, Stanton AC. Malignant tumors of the paranasal sinuses: Radiologic, clinical, and histopathologic evaluation of 200 cases. Head Neck Surg. 1984;3:761–776. 48. Uchida D, Shirato H, Onimaru R, et al. Long-term results of ethmoid squamous cell or undifferentiated carcinoma treated with radiotherapy with or without surgery. Cancer J. 2005;11:152– 156. 49. Panchal L, Vaideeswar P, Kathpal D, et al. Sino-nasal epithelial tumours: A pathological study of 69 cases. J Postgrad Med. 2005;51:30–35. 50. Daoud A, Lannigan FJ, McGlashan JA, et al. Verrucous carcinoma of the maxillary antrum. J Laryngol Otol. 1991;105:696– 699. 51. Wieneke JA, Thompson LD, Wenig BM. Basaloid squamous cell carcinoma of the sinonasal tract. Cancer. 1999;85:841–854. 52. Alos L, Castillo M, Nadal A, et al. Adenosquamous carcinoma of the head and neck: Criteria for diagnosis in a study of 12 cases. Histopathology. 2004;44:570–579. 53. Batsakis JG. Pathology of tumors of the nasal cavity and paranasal sinuses. In: Thawley SE, Panje WR, Batsakis JG, Lindberg RD, eds. Comprehensive Management of Head and Neck Tumors. Philadelphia: W.B. Saunders Co, 1999:522–539. 54. McCollough WM, Mendenhall NP, Parsons JT, et al. Radiotherapy alone for squamous cell carcinoma of the nasal vestibule: Management of the primary site and regional lymphatics. Int J Radiat Oncol Biol Phys. 1993;26:73–79. 55. Batasakis JG, Rice DH, Solomon AR. The pathology of head and neck tumors. Squamous and mucus gland carcinomas of the nasal cavity, paranasal sinuses, and larynx, part 6. Head Neck Surg. 1980;2:497–508. 56. Ohngren LG. Malignant tumours of the maxillo-ethmoidal region. Acta Otolaryngol Suppl. 1933;19:101–106. 57. Rouviere H. Anatomie Des Lymphatiques De L’homme. Paris: Masson et cie, 1932. 58. Werner JA, Dunne AA, Myers JN. Functional anatomy of the lymphatic drainage system of the upper aerodigestive tract and its role in metastasis of squamous cell carcinoma. Head Neck. 2003;25:322–332. 59. Fernandez JM, Santaolalla F, Del Rey AS, et al. Preliminary study of the lymphatic drainage system of the nose and paranasal sinuses and its role in the detection of sentinel metastatic nodes. Acta Otolaryngol. 2005;125:566–570. 60. Tiwari R, Hardillo JA, Mehta D, et al. Squamous cell carcinoma of maxillary sinus. Head Neck. 2000;22:164–169. 61. Kawashima M, Ogino T, Hayashi R, et al. Influence of postsurgical residual tumor volume on local control in radiotherapy for maxillary sinus cancer. Jap J Clin Oncol. 2001;31:195– 202. 62. Qureshi SS, Chaukar DA, Talole SD, et al. Squamous cell carinoma of the maxillary sinus: A Tata Memorial Hospital experience. Indian J Cancer. 2006;43:26–29. 63. Waldron J, Witterick I. Paranasal sinus cancer: Caveats and controversies. World J Surg. 2003;27:849–855. 64. Le QT, Fu KK, Kaplan MJ, et al. Lymph node metastases in maxillary sinus carcinoma. Int J Radiat Oncol Biol Phys. 2000;46:541– 549.

65. St. Pierre S, Baker SR. Squamous cell carcinoma of the maxillary sinus: Analysis of 66 cases. Head Neck Surg. 1983;5:508–513. 66. Stern SJ, Goepfert H, Clayman G, et al. Squamous cell carcinoma of the maxillary sinus. Arch Otolaryngol Head Neck Surg. 1993;119:964–969. 67. Konno A, Ishikawa K, Terada N, et al. Analysis of long-term results of our combination therapy for squamous cell carcinoma of the maxillary sinus. Acta Otolaryngol Suppl. 1998;537:57–66. 68. Giri SP, Reddy EK, Gemer LS, et al. Management of advanced squamous cell carcinomas of the maxillary sinus. Cancer. 1992;69:657–661. 69. Biswas D, Crank S. Aetiopathology of maxillary swelling— A 3-year prospective study. Eur Arch Otorhinolaryngol. 2007;11:1295–1299. 70. Kraus DH, Lanzieri CF, Wanamaker JR, et al. Complementary use of computed tomography and magnetic resonance imaging in assessing skull base lesions. Laryngoscope. 1992;102;623– 629. 71. Yousem DM. Imaging in sinus disease. In: Kennedy DW, Bolger WE, Zinreich SJ, eds. Diseases of the Sinuses. Diagnosis and Management. London: BC Decker, 2001:129–148. 72. Som PM, Shapiro MD, Biller HF, et al. Sinonasal tumors and inflammatory tissues: Differentiation with MR imaging. Radiology. 1988;167;803–808. 73. Som PM, Dillon WP, Sze G, et al. Benign and malignant sinonasal lesions with intracranial extension: Differentiation with MRI imaging. Radiology. 1989;172;763–766. 74. Lanzieri CF, Shah M, Kraus D, et al. Use of gadoliniumenhanced MR imaging for differentiating mucoceles from neoplasms in the paranasal sinuses. Radiology. 1991;178;425–428. 75. Cox G. Nose and sinuses. In: Wilson JA, ed. Effective Head and Neck Cancer Management. Third Consensus Document, 2002. London: British Association of Otorhinolaryngologists Head and Neck Surgeons, 2002:165–170. 76. Sobin LH, Wittekind C. International Union Against Cancer: TNM Classification of Malignant Tumors. New York, NY: John Wiley & Sons, 2002. 77. Isobe K, Uno T, Hanazawa T, et al. Preoperative chemotherapy and radiation therapy for squamous cell carcinoma of the maxillary sinus. Jpn J Clin Oncol. 2005;35:633–638. 78. Anniko M, Franzen L, Lofroth PO. Long-term survival of patients with paranasal sinus carcinoma. ORL J Otorhinolaryngol Relat Spec. 1990;52:187–193. 79. Jansen EP, Keus RB, Hilgers FJ, et al. Does the combination of radiotherapy and debulking surgery favor survival in paranasal sinus carcinoma. Int J Radiat Oncol Biol Phys. 2000;48:27–35. 80. Blanco AI, Chao KS, Ozyigit G, et al. Carcinoma of the paranasal sinuses: Long-term outcomes with radiotherapy. Int J Radiat Oncol Biol Phys. 2004;59:51–58. 81. Hoppe BS, Stegman LD, Zelefsky MJ, et al. Treatment of nasal cavity and paranasal sinus cancer with modern radiotherapy techniques in the postoperative setting—the MSKCC experience. Int J Radiat Oncol Biol Phys. 2007;67:691–702. 82. Lavertu P, Roberts JK, Kraus DH, et al. Squamous cell carcinoma of the paranasal sinuses: The Cleveland Clinic experience 1977– 1986. Laryngoscope. 1989;99:1130–1136. 83. Nibu K, Sugasawa M, Asai M, et al. Results of multimodality therapy for squamous cell carcinoma of maxillary sinus. Cancer. 2002;94:1476–1482. 84. Nishino H, Miyata M, Morita M, et al. Combined therapy with conservative surgery, radiotherapy, and regional chemotherapy for maxillary sinus carcinoma. Cancer. 2000;89:1925–1932. 85. Lee HB, Hong JP, Kim KT, et al. Orbital floor and infraorbital rim reconstruction after total maxillectomy using a vascularized calvarial bone flap. Plast Reconstr Surg. 1999;104:646–653. 86. Cordeiro PG, Santamaria E. A classification system and algorithm for reconstruction of maxillectomy and midfacial defects. Plast Reconstr Surg. 2000;105:2331–2346. 87. Nakayama B, Hasegawa Y, Hyodo I, et al. Reconstruction using three-dimensional orbitozygomatic skeletal model of titanium mesh plate and soft-tissue free flap transfer following total maxillectomy. Plast Reconstr Surg. 2004;114:631–639.

Chapter 29: Squamous Cell Carcinoma of the Nasal Cavity and Paranasal Sinuses 88. Ahmad K, Cordoba RB, Fayos JV. Squamous cell carcinoma of the maxillary sinus. Arch Otolaryngol. 1981;107:48–51. 89. Jeremic B, Shibamoto Y, Milicic B, et al. Elective ipsilateral neck irradiation of patients with locally advanced maxillary sinus carcinoma. Cancer. 2000;88:2246–2251. 90. Amendola BE, Eisert D, Hazra TA, et al. Carcinoma of the maxillary antrum: Surgery or radiation therapy? Int J Radiat Oncol Biol Phys. 1981;7:743–746. 91. Lee F, Ogura JH. Maxillary sinus carcinoma. Laryngoscope. 1981;91:133–139. 92. Paulino AC, Marks JE, Bricker P, et al. Results of treatment of patients with maxillary sinus carcinoma. Cancer. 1998;83:457– 465. 93. Guntinas-Lichius O, Kreppel MP, Stuetzer H, et al. Single modality and multimodality treatment of nasal and paranasal sinuses cancer: A single institution experience of 229 patients. Eur J Surg Oncol. 2007;33:222–228. 94. Waldron JN, O’Sullivan B, Warde P, et al. Ethmoid sinus cancer: Twenty-nine cases managed with primary radiation therapy. Int J Radiat Oncol Biol Phys. 1998;41:361–369. 95. Chen AM, Daly ME, Bucci MK, et al. Carcinomas of the paranasal sinuses and nasal cavity treated with radiotherapy at a single institution over five decades: Are we making improvement? Int J Radiat Oncol Biol Phys. 2007;69:141– 147. 96. Karim AB, Kralendonk JH, Njo KH, et al. Ethmoid and upper nasal cavity carcinoma: Treatment, results, and complications. Radiother Oncol. 1990;19:109–120. 97. Parsons JT, Kimsey FC, Mendenhall WM, et al. Radiation therapy for sinus malignancies. Otolaryngol Clin North Am. 1995;28:1259–1268. 98. Parsons JT, Mendenhall WM, Mancuso AA, et al. Malignant tumors of the nasal cavity and ethmoid and sphenoid sinuses. Int J Radiat Oncol Biol Phys. 1988;14:11–22. 99. Katz TS, Mendenhall WM, Morris CG, et al, Malignant tumors of the nasal cavity and paranasal sinuses. Head Neck. 2002;24:821– 829. 100. Dulguerov P, Allal AS. Nasal and paranasal sinus carcinoma: How can we continue to make progress? Curr Opin Otolaryngol Head Neck Surg. 2006;14:67–72. 101. Monroe AT, Bhandare N, Morris CG, et al. Preventing radiation retinopathy with hyperfractionation. Int J Radiat Oncol Biol Phys. 2003;55:776–784. 102. Isaacs JH Jr, Mooney S, Mendenhall WM, et al. Cancer of the maxillary sinus treated with surgery and/or radiation therapy. Am J Surg. 1990;56:327–330. 103. Jiang GL, Ang KK, Peters LJ, et al. Maxillary sinus carcinoma: Natural history and results of postoperative radiotherapy. Radiother Oncol. 1991;21:193–200. 104. Parsons JT, Bova FJ, Fitzgerald CR, et al. Radiation retinopathy after external beam irradiation: Analysis of time-dose factors. Int J Radiat Biol Phys. 1994;30:765–773. 105. Parsons JT, Bova FJ, Fitzgerald CR, et al. Radiation optic neuropathy after megavoltage external beam irradiation: Analysis of time-dose factors. Int J Radiat Biol Phys. 1994;30:755–763. 106. Daly ME, Chen AM, Bucci MK, et al. Intensity-modulated radiation therapy for malignancies of the nasal cavity and paranasal sinuses. Int J Radiat Oncol Biol Phys. 2007;67:151–157. 107. Meyers CA, Geara F, Wong PF, et al. Neurocognitive effects of therapeutic irradiation for skull base tumors. Int J Radiat Oncol Biol Phys. 2000;46:51–55. 108. Duthoy W, Boterberg T, Claus F, et al. Postoperative intensitymodulated radiotherapy in sinonasal carcinoma. Clinical results in 39 patients. Cancer. 2005;104:71–82. 109. Pommier P, Ginester C, Sunyach M, et al. Conformal radiotherapy for paranasal sinus and nasal cavity tumors: Three dimensional treatment planning and preliminary results in 40 patients. Int J Radiat Oncol Biol Phys. 2000;48:485–493. 110. Roa WH, Hazuka MB, Sandler HM, et al. Results of primary and adjuvant CT-based 3-dimensional radiotherapy for malignant tumors of the paranasal sinuses. Int J Radiat Oncol Biol Phys. 1994;28:857–865.

443

111. Weber DC, Chan AW, Lessell S, et al. Visual outcome of accelerated fractionated radiation for advanced sinonasal malignancies employing photons/protons. Radiother Oncol. 2006;81:243– 249. 112. Lomax AJ, Goitein M, Adams J. Intensity modulation of radiotherapy: Photons versus protons in the paranasal sinus. Radiother Oncol. 2003;66:11–18. 113. Hu YH, Tu GY, Qi YQ, et al. Comparison of pre- and postoperative radiation in the combined treatment of carcinoma of the maxillary sinus. Int J Radiat Oncol Biol Phys. 1982;8:1045–1049. 114. Backhouse TW. The treatment of malignant tumors of the maxillary antrum. Clin Radiol. 1986;37:179–182. 115. Curran AJ, Gullane PJ, Waldron J, et al. Surgical salvage after failed radiation for paranasal sinus malignancy. Laryngoscope. 1998;108:1618–1622. 116. Levendag PC, Nijdam WM, von Moolenburgh SE, et al. Interstitial radiation therapy for early stage nasal vestibule cancer: A continuing quest for optimal tumor control and cosmesis. Int J Radiat Oncol Biol Phys. 2006;66:160–169. 117. Mazeron JJ, Chassagne D, Crook J, et al. Radiation therapy of carcinomas of the skin of nose and nasal vestibule: A report of 1676 cases by the Groupe Europeen de Curietherapie. Radiother Oncol. 1988;13:165–173. 118. Sato Y, Morita M, Takahashi H, et al. Combined surgery, radiotherapy, and regional chemotherapy in carcinoma of the paranasal sinuses. Cancer. 1970;25:571–579. 119. Shiga K, Yokoyama J, Hashimoto S, et al. Combined therapy after superselective arterial cisplatin infusion to treat maxillary squamous cell carcinoma. Otolaryngol Head Neck Surg. 2007;136:1003–1009. 120. Samant S, Robbins T, Vang M, et al. Intra-arterial cisplatin and concomitant radiation therapy followed by surgery for advanced paranasal sinus cancer. Arch Otolaryngol Head Neck Surg. 2004;130:948–955. 121. Lee MM, Vokes EE, Rosen A, et al. Multimodality therapy in advanced paranasal sinus carcinoma: Superior long-term results. Cancer J Sci Am. 1999;5:219–223. 122. Rosen A, Vokes EE, Scher N, et al. Locoregionally advance paranasal sinus carcinoma. Favorable survival with multimodality therapy. Arch Otolaryngol Head Neck Surg. 1993;119:743–746. 123. Harrison LB, Pfister DG, Kraus D, et al. Management of unresectable malignant tumors at the skull base using concomitant chemotherapy and radiotherapy with accelerated fractionation. Skull Base Surg. 1994;4:127–131. 124. Itami J, Uno T, Aruga M, et al. Squamous cell carcinoma of the maxillary sinus treated with radiation therapy and conservative surgery. Cancer. 1998;82:104–107. 125. Knegt PP, Ah-See KW, vd Velden LA, et al. Adenocarcinoma of the ethmoidal sinus complex. Surgical debulking and topical fluorouracil may be the optimal treatment. Arch Otolaryngol Head Neck Surg. 2001;127:141–146. 126. Carrau RL, Segas J, Nuss DW, et al. Squamous cell carcinoma of the sinonasal tract invading the orbit. Laryngoscope. 1999;109:230–235. 127. McCary WS, Levine PA, Cantrell RW. Preservation of the eye in the treatment of sinonasal malignant neoplasms with orbital involvement. Arch Otolaryngol Head Neck Surg. 1996;122;657– 659. 128. Perry C, Levine PA, Williamson BR, et al. Preservation of the eye in paranasal sinus cancer surgery. Arch Otolaryngol Head Neck Surg. 1988;114;632–634. 129. Tiwari R, van der Wal J, van der Waal I, et al. Studies of the anatomy and pathology of the orbit in carcinoma of the maxillary sinus and their impact on preservation of the eye in maxillectomy. Head Neck. 1998;20:193–196. 130. Quatela VC, Futran ND, Boynton JR. Eye banking: Techniques for eye preservation in selected neoplasms encroaching on the globe. Otolaryngol Head Neck Surg. 1993;108;662–670. 131. Stern SJ, Goepfert H, Clayman G, et al. Orbital preservation in maxillectomy. Otolaryngol Head Neck Surg. 1993;109:111– 115.

444

Sheahan et al.

132. Imola MJ, Schramm VL. Orbital preservation in surgical management of sinonasal malignancy. Laryngoscope. 2002;112:1357–1365. 133. Andersen PE, Kraus DH, Arbit D, et al. Management of the orbit during anterior fossa craniofacial resection. Arch Otolaryngol Head Neck Surg. 1996;122:1305–1307. 134. Scurry WC, Goldenberg D, Chee MY, et al. Regional recurrence of squamous cell carcinoma of the nasal cavity. A systematic review and meta-analysis. Arch Otolaryngol Head Neck Surg. 2007;133:796–800.

135. Penzer RD, Moss WT, Tong D, et al. Cervical lymph node metastasis in patients with squamous cell carcinoma of the maxillary antrum: The role of elective irradiation of the clinically negative neck. Int J Radiat Biol. 1979;5:1977–1980. 136. Shipchandler TZ, Batra PS, Citardi MJ, et al. Outcomes for endoscopic resection of sinonasal squamous cell carcinoma. Laryngoscope. 2005;115:1983–1987. 137. Buchman L, Larsen C, Pollack A, et al. Endoscopic techniques in resection of anterior skull base/paranasal sinus malignancies. Laryngoscope. 2006;116:1749–1754.

30 Nonsquamous Cell Carcinoma of the Nasal Cavity and Paranasal Sinuses ` Carlo L. Solero, Stefano Riccio, and Sarah Colombo Giulio Cantu,

In this chapter, we will deal with adenoid cystic carcinoma and, in particular, with adenocarcinoma.

INCIDENCE AND EPIDEMIOLOGY Malignant tumors of the nasal cavity and paranasal sinuses are relatively rare, accounting for 3% of head and neck carcinomas and about 0.5% of all malignancies (1). Despite the low incidence rate, a great variety of histologic types exist. Therefore, the published series from each institution are generally small and heterogeneous, preventing definitive conclusions. Cancers of the paranasal sinuses occur more frequently during the fifth and sixth decades of life, with a male to female ratio of about 2:1 (2). The most common histologic type is squamous cell carcinoma or one of its variants (e.g., transitional, verrucous); adenocarcinoma, adenoid cystic carcinoma, mucoepidermoid carcinoma, and undifferentiated carcinoma are less common. The maxillary sinus is the most frequent site of origin for paranasal sinus tumors (50–70%), particularly squamous cell, adenoid cystic, and mucoepidermoid carcinoma. In the ethmoid sinus, there is a higher incidence rate of undifferentiated carcinoma and adenocarcinoma. The etiology of sinonasal cancers was first hypothesized in 1890, when a tumor was detected in a worker exposed to chrome (3). Since then, many etiologic studies on workers exposed to different materials have been performed. The carcinogenicity of some of these agents in humans has been clearly demonstrated (wood and leather dusts, nickel, chrome, isopropyl alcohol, and arsenic). The role of other work environments in tumor development is more questionable (textile and building industries). For squamous cell carcinoma, smoking is an important risk factor (4). However, the most interesting tumor for which there is an indisputable occupational etiology is intestinal type adenocarcinoma (ITAC); we will present the details in the paragraph dealing with this histologic type.

CLINICAL FEATURES The natural history and clinical features of nonsquamous cell carcinomas obviously depend on the histologic type. The details of each tumor are presented herein. As a rule, the signs and symptoms of nonsquamous cell carcinoma differ from those of squamous cell carcinoma. Many of these are slow-growing tumors, and the first symptoms may date back many months or even years. The history of a submucosal and painless mass may extend over 10 years. Because tumors of the paranasal sinuses arise in air-filled cavities, they can infiltrate the bony walls before signs and symptoms develop. Given this early clinical silence, most patients at presentation have advanced disease extensively involving surrounding structures. We may, indeed, say that malignant tumors of the sinuses do not display clear evidence of their presence until they have broken out of the sinus of origin. Tumors arising in the upper part of the nasal cavity and in the ethmoid may invade the orbit and the anterior cranial fossa through the cribriform plate. They also may destroy the septum and nasal bone, thus infiltrating the skin. The sphenoid sinus and the nasopharynx are invaded in advanced tumors. The possible extension of tumors of the maxillary sinus varies by the site of origin. Lesions arising in the anteroinferior wall often present in the oral cavity as submucosal swelling causing dental pain, loosening of teeth, or improper seating of a denture. Tumors arising on the medial wall may easily invade the nasal cavity through this thin bone. Posterior lesions are the most dangerous for their long clinical silence. Pain is a late-occurring event and often indicates infiltration of the second and/or third branch of trigeminal nerve. The tumor destroys the pterygoid plates and invades the pterygoid and infratemporal fossa. It may approach and erode the greater wing of the sphenoid, spreading to the middle cranial fossa. The incidence of lymph node metastases is low, both at presentation and during follow-up. A higher rate of lymphatic spread occurs only with tumors invading the oral cavity and/or nasal cavity mucosa. This was recognized as far back as 1937 by del Regato (7) and was later confirmed by other authors (8,9). For nonsquamous cell carcinomas, the rate of regional metastases is even lower.

PATHOLOGY Although squamous cell carcinoma is the most common histologic subtype among paranasal sinus malignancies, it is less predominant in this anatomic location than in any other site within the upper aerodigestive tract. Actually, a large variety of histopathologically different tumors occur in this region. The World Health Organization classification divides nasal cavity and paranasal sinus primary malignancies into malignant epithelial tumors, malignant soft tissue tumors, malignant tumors of bone and cartilage, hematolymphoid tumors, and neuroectodermal tumors (5). With regard to epithelial tumors, there are two basic types: those originating from the epithelium and those originating from mucous glands. Nonsquamous cell carcinomas take their origin from the mucous membranes, minor salivary glands, and seromucinous glands (6).

STAGING The difficulties establishing an indisputable prognostic staging for each extension of paranasal sinus carcinomas are demonstrated in the long list of different classifications that existed in times past. These classifications (10–16) considered 445

446

Cantu` et al.

Table 1 T1 T2 T3 T4

INT Staging of Ethmoid Malignant Tumors

Tumor involving the ethmoid and nasal cavity sparing the most superior ethmoid cells Tumor with extension to or erosion of the cribriform plate, with or without erosion of the lamina papyracea and without extension into the orbit Tumor extending into the anterior cranial fossa extradurally and/or into the anterior two-thirds of the orbit, with or without erosion of the anteroinferior wall of the sphenoid sinus, and/or involvement of the maxillary and frontal sinus Tumor with intradural extension, or involving the orbital apex, the sphenoid sinus, the pterygoid plate, the infratemporal fossa, or the skin

only the maxillary sinus, and almost all assigned a higher stage for tumors involving the posterosuperior part of ¨ maxillary sinus. Ohngren line (10) (a plane connecting the inner canthus of the eye to the mandibular angle) divides the maxillary sinus into an anteroinferior portion (infrastructure) and a superoposterior portion (suprastructure). It was used in all versions of the American Joint Committee on Cancer (AJCC) classification of maxillary tumors to distinguish tumors with good prognosis (infrastructure) from tumors with poor prognosis (suprastructure). The International Union Against Cancer (UICC) began to stage maxillary sinus tumors in its fifth edition and used the same criterion. The poorer outcome of superoposterior tumors is the consequence of their early invasion of critical structures such as the orbit, pterygoid, infratemporal fossa, and skull base. Some of these extensions have been considered unresectable for many years, or resectable without wide, clear margins. The introduction of craniofacial resections in routine surgical procedures has enabled skilled surgeons to attain clean margins also in these cases. The AJCC–UICC fifth classification of maxillary sinus carcinomas has been tested in several (including small) series. Results showed the classification to be rather prognostic, with a progressive worsening of the prognosis from T1 to T4. Dulguerov et al. (9), in their meta-analysis of publications on nasal and paranasal sinus carcinoma from 1960 to 2000, found a clear correlation between T stage and survival. Moreover, they demonstrated a nearly unchanged prognosis for T1–T2 tumors during the study period, whereas there was a progressive improvement in outcome for T3–T4 tumors. Even if this result is the logical consequence of the evolution in surgical procedures, one must be aware that the widespread application of modern imaging methods may result in a possible shift in classification from lower to upper stages. This process results in an apparent improvement in outcome by moving the worst cases of a lower stage to the even worse cases of the higher stage, which is the basis of the well-known Will Rogers phenomenon. An example is the division of the stage T4 of the AJCC–UICC-1997 classification (17,18) into stages T4a and T4b of the last 2002 classification (19,20). Thus, the results from upcoming studies should be carefully interpreted, especially when assessing therapeutic improvements. Apart from tumors of the frontal and sphenoid sinuses, where primary tumors are exceptionally rare, the AJCC and UICC had not provided staging guidelines for the more common ethmoid sinus and nasal cavity tumors before 1997. This led to an obvious lack of disease staging in the reported literature on ethmoid cancer. Sisson et al. (21) wrote, “The ethmoid cancers were not staged because there is no generally accepted staging system for this site.” Spiro et al. (22), after having staged tumors of the maxillary sinuses, wrote, “As there is no widely accepted staging system for the remaining sinuses or the nasal cavity, no attempt was made to stage tumors arising in these sites.” Nevertheless, others have attempted to stage nasoethmoid tumors. Kadish et al. (23), Biller et al. (24), and Dulguerov and Calcaterra (25) proposed a classification for

esthesioneuroblastomas. Ellingwood and Million (26) published a classification for cancers of the nasal cavity and ethmoid/sphenoid sinuses in 1979. Finally, Roux et al. (27) adopted a staging system they called “modified TNM.” Moreover, some of the classifications, despite their historical significance as first attempts at staging, were never tested on large series of patients to verify their prognostic value. Ethmoid carcinomas were finally staged in the fifth edition of both the AJCC Cancer Staging Manual (17) and the UICC’s TNM Classification of Malignant Tumours (18). In the absence of a universally accepted staging system and on the basis of our experience with anterior craniofacial resections, we developed in 1993 and presented in 1997 an original classification for malignant ethmoid tumors (28) based on the most commonly accepted unfavorable prognostic factors (involvement of dura mater; intradural extension; involvement of the orbit and, in particular, its apex; invasion of maxillary, frontal, and/or sphenoid sinus; and invasion of the infratemporal fossa and skin) (Table 1). We applied this classification to all consecutive malignant nasoethmoid tumors that were treated at our institution (29). In 1999, we successfully validated our original INT (Istituto Nazionale Tumori) classification for ethmoid cancers (30) against the fifth edition of the AJCC–UICC classification (17,18). On the basis of these encouraging results, we tested the sixth AJCC–UICC (19,20) 2002 classification in terms of prognostic performance versus the 1997 AJCC–UICC and the INT classifications (31). Both the 1997 and 2002 AJCC–UICC classifications seemed to have limited prognostic value. By contrast, the INT classification satisfied one of the main goals of tumor staging, demonstrating the progressive worsening of prognosis with different tumor classes. The validity of the INT classification was confirmed on our large series of 241 patients, and it was shown to achieve the best prognostic discrimination among T classifications not only for the overall series but also when applied separately to untreated patients, recurring cases, and adenocarcinomas, the most frequent histologic type in our series. Dulguerov et al., in their recent review (32), stated, “While the evolution of TNM staging is a work in continuous progress, the T staging of ethmoid and nasal primaries needs an urgent revision.” We agree with this statement. After providing these general comments on paranasal sinuses malignancies, we now discuss two histologic types among nonsquamous cell carcinomas.

ADENOID CYSTIC CARCINOMA Incidence and Epidemiology Adenoid cystic carcinoma is the most frequent malignant tumor of the minor salivary glands, constituting more than one-third of cases. According Harrison and Lund (33), it represents about 1.5% of all tumors of the paranasal sinuses; the maxilla and hard palate are the most frequent sites of origin. In our series of 704 cases of malignant tumors of the paranasal sinuses, we found 115 cases of adenoid cystic carcinoma (16.4%): 24/305 cases (7.9%) in the ethmoid sinus and

Chapter 30: Nonsquamous Cell Carcinoma of the Nasal Cavity and Paranasal Sinuses

91/399 (22.8%) in the maxillary sinus. In a series of 334 patients who underwent a craniofacial resection collected from 17 institutions, Ganly et al. (34) reported 32 cases of minor salivary glands carcinoma (9.6%). There is no known etiologic factor for this tumor. The male-to-female ratio is different in each published series and is likely equal. In our series, there was a small prevalence of females (65/50). The age range is quite broad, but there are peaks in the fifth and sixth decades of life.

Pathology Adenoid cystic carcinoma has no distinguishing gross features apart from its infiltrative growth, which makes it difficult to demarcate the tumor from surrounding normal tissues (6). Three histologic types have been described: tubular, cribriform, and solid. Batsakis et al. (35,36) suggested that the solid type is the most aggressive form but other authors have not found an indisputable correlation between histology and outcome (37,38). Perineural spread is the distinctive feature of adenoid cystic carcinoma; it may extend great distance from the primary tumor along nerve pathways.

Clinical Features The clinical features of adenoid cystic carcinoma depend on the site of origin. Because minor salivary glands are rare in the anterior part of the hard palate, the tumor often appears as a submucosal mass involving the posterior hard palate and/or soft palate. Many of these lesions are indolent and painless until ulceration appears, so the history may go back many months or even years. Tumors that present in the maxillary sinus may be associated with facial swelling and pain. Pain, tingling, or paresthesias indicate neural involvement. Bearing in mind the aforesaid capacity for perineural spread, it is easy to understand how extension of maxillary tumors along the second trigeminal branch (greater palatine and infraorbital nerves) can reach intracranial spaces, in particular the Gasserrian ganglion and cavernous sinus (Fig. 1). Ethmoid tumors can easily involve the anterior cranial fossa along olfactory nerves. Traversing the dura, the tumor may extend into the brain. In case of orbital invasion, in particular of the apex, the tumor may follow the first branch of trigeminal nerve; the third, fourth, and sixth cranial nerves; and the optic nerve.

447

Lymph node metastases are rare, and lymphatic spread plays a modest role in the treatment of adenoid cystic carcinoma. Only one patient in our series died from nodal metastases per se. On the contrary, systemic metastases (lung, brain, bone, and liver) are frequent. Many patients manifest metastatic disease concurrent with or after local recurrence within 5 years, and about 5% of patients have lung metastases at presentation. However, distant metastases without local recurrence may appear 15 to 20 years later (33).

Treatment Surgery and postoperative radiotherapy is usually considered the treatment of choice (39). Only small palatal lesions can be resected with partial or subtotal maxillectomy. Unfortunately, the majority of patients present with large maxillary tumors are either approaching or involving the infratemporal fossa, the orbit, and the middle and/or anterior skull base. Obviously, the goal of surgery must be complete resection with negative margins. Technical advances in anterior, lateral, and anterolateral craniofacial resections with free flap reconstruction allow the resection of tumors that were considered inoperable in the past. However, oncologic resection of paranasal sinuses adenoid cystic carcinoma often has an uncertain outcome for the aforesaid factors. The surgeon must be aware that even aggressive surgery typically does not result in cure. Local relapse and/or distant metastases are nearly always the rule. Thus, a balance between radical surgery and low morbidity must be sought. In our opinion, the patient’s clinical condition before treatment must be the main factor in making a decision. Acute pain from trigeminal infiltration, or the presence of an ulcerated, necrotic, and bleeding tumor, may justify a wide resection and reconstruction to give to the patient a better quality of life, sometimes for many years. The role of radiotherapy alone in the treatment of sinonasal adenoid cystic carcinoma is controversial. Current opinion is that this tumor is not radiocurable with conventional megavoltage photon and/or electron beams (33). Some reports indicate that neutron radiotherapy might be more efficacious than conventional radiotherapy (40,41). However, these articles underline unsatisfactory outcomes in patients with skull base involvement, as this extension is dose limiting because of the sensitivity of the central nervous system structures to neutron radiotherapy. Chemotherapy is generally reserved for palliative treatment of metastatic disease or locoregional recurrence for which further surgery or radiation is not possible (42).

Outcome and Prognosis

Figure 1 Maxillary adenoid cystic carcinoma involving the third branch of trigeminal nerve.

Because of the long natural history of this tumor, a 5-year follow-up period is inadequate to convey the ultimate outcome of adenoid cystic carcinoma. Patients may have frequent local recurrences and/or hematogenous dissemination, sometimes 10 or more years later. Because few papers report a large number of patients with a long follow-up, it is difficult to assess the real outcome for patients presenting with adenoid cystic carcinoma of the sinonasal tract involving the skull base. Despite aggressive surgery and postoperative radiotherapy, about 70% of patients will experience a tumor recurrence, and the cure rate is even lower for patients treated for local recurrence after previous surgery (43). In our series of 115 patients with adenoid cystic carcinoma of the paranasal sinuses, 52 patients presented with a T4a-b tumor involving the skull base. While 7 of 24 patients who underwent an anterior craniofacial resection for a tumor localized in the ethmoid sinus are free

448

Cantu` et al.

of disease (29%), none of the 28 patients requiring a lateral craniofacial resection for involvement of the middle cranial fossa is alive without disease. We believe that although surgery appears to be palliative for many patients with advanced adenoid cystic carcinoma, a tumor that is bleeding and causing pain may justify a wide craniofacial resection and reconstruction to give to the patient a better quality of residual life.

ADENOCARCINOMA Pathology, Incidence, and Epidemiology The true incidence of sinonasal adenocarcinomas is unknown because of the lack of consensus among pathologists regarding its classification. While almost all pathologists clearly distinguish between salivary types and true sinonasal adenocarcinomas, most of published clinical articles on paranasal sinus tumors use the general term adenocarcinoma without subclassification (9,44–46). Batsakis (47) categorized adenocarcinomas into three clinicopathologic forms: papillary, sessile, and alveolarmucoid, the last of which includes tumors that closely simulate colonic carcinomas. Abecasis et al. (48) used different clinicopathologic and immunohistochemical classifications: high- and low-grade adenocarcinoma, papillary, and ITAC. In that classification, ITACs were given a poor prognosis. Heffner et al. (49) divided adenocarcinomas of the sinonasal tract into low grade and high grade, and included ITACs in the high-grade group. In contrast, Bashir et al. (50) classified ITACs as well- or moderately differentiated tumors. The last paper is paradigmatic about the difficulties of histologic classification. These authors classified 11 sinonasal adenocarcinomas into three groups: intestinal type, glandular type, and solid type. However, after this statement, the authors wrote: “Judging from other articles, some of our tumors that we have classified as SNA-G (glandular) and/or SNA-S (solid) may be considered intestinal type by others.” Barnes (51) stated that ITACs of the nasal cavity and paranasal sinuses might occur sporadically or as an occupational disease, in cases of wood dust exposure. Histologically, he recognized five variants: papillary, colonic, solid, mucinous, and mixed. Comparing 17 cases of sporadic-type ITAC to those among woodworkers, he found some important differences. In the former group, there were nine men and eight women, with eight tumors originating in the maxillary sinus, seven in the nasal cavity, and only two in the ethmoid sinus. In contrast, ITACs in woodworkers occurred primarily in men and originated almost exclusively in the nasal cavity or ethmoid sinus. Immunohistochemical marking and expression of oncoproteins did not completely provide resolution. Bashir et al. (50) wrote, “The study of expression of CK7 and CK20 in sinonasal adenocarcinoma is not useful in making the accurate differential diagnosis between primary or metastatic intestinal tumors. The CK7/CK20 profile was successful in distinguishing adenocarcinoma from transitional carcinoma, pointing to its utility in the differential diagnosis of these two entities.” We may find similar conclusions in the paper by Abecasis et al. (48). According to Choi et al. (52), “All primary enteric-type carcinomas and the 2 colonic metastases were reactive to CK20, but all non-enteric-type tumors were negative for CK20 and positive for CK7.” These authors conclude, “Non-enteric-type (seromucinous) adenocarcinoma may originate directly from surface respiratory-type epithelium or from seromucous glands, metaplastic transformation of surface respiratory to enteric-type epithelium pre-

cedes the development of enteric adenocarcinoma, and coordinate analyses of CK7 and CK20 reactivity may aid the differential diagnosis of adenocarcinoma in the sinonasal tract.” Kennedy et al. (53) stained 12 sinonasal adenocarcinomas with monoclonal antibodies to CK7, CK20, CDX-2, and villin. The authors’ conclusion was, “Sinonasal ITACs have a distinctive phenotype, with all cases expressing CK20, CDX-2, and villin. Most ITACs also express CK7, although a proportion of tumors are CK7 negative. ITAC seems to be preceded by intestinal metaplasia of the respiratory mucosa, which is accompanied by a switch to an intestinal phenotype.” As a consequence of these difficulties in pathologic classification, the literature contains a wide range of incidence rates for adenocarcinomas among sinonasal malignant tumors [e.g., 4–9% for Harrison and Lund (33), 10–20% for Bashir (50)]. With regard to epidemiology and etiology, almost all authors report the role of wood dust. This role was recognized in 1970 by Hadfield (54), who analyzed 35 cases of sinonasal adenocarcinoma in woodworkers in the furniture industries. In all 35 patients, the tumor appeared to originate in the ethmoid sinuses. Since then, several papers have been published on this matter, and the leather and shoe industries have also been associated with an increased risk of adenocarcinoma (55–57). The different roles that hardwoods and softwoods play in the development of these tumors present a vexing question. Some authors (58,59) in northern Europe (where furniture industries use softwoods) underlined a minor and different carcinogenic quality of softwoods compared with hardwoods that are more often used in southern Europe. General consensus does exist, however, that wood dust levels above 5 mg/m3 in the work environment present higher risk. Interestingly, the rates of adenocarcinoma in series of patients with sinonasal tumors treated with anterior craniofacial resection vary greatly between Europe and North America. The rates of adenocarcinoma in European series are very high: Roux (60) (France), 74%; Suarez (61) (Spain), 53%; Cantu (31) (Italy), 49%; and Cheesman (62) (United Kingdom), 27%. In contrast, the rates in American series are much lower: McCutcheon (63) (United States), 17%; Bentz (64) (United States), 12%; Donald (65) (United States), 6%; and Irish (66) (Canada), 5%. Bridger (67) (Australia) reports a rate of 37% of adenocarcinoma, similar to the rates in Europe. It is difficult to find an unambiguous and exhaustive explanation for these discrepancies. We may advance some hypotheses: The high-risk threshold commonly accepted for wood dust level in the air is 5 mg/m3 ; it is likely that this threshold was exceeded in many European artisan furniture factories and joineries in the past. Hardwoods, which present higher risk than softwoods, are probably more widespread in Europe than in America. Safety measures such as the use of masks and aspiration devices became widespread in the United States before Europe. Considering that the latency period from the beginning of exposure to clinical evidence of tumor is about 40 years (68) and that factory conditions have improved in Europe, we may predict a possible reduction in the incidence of the disease in Europe in coming decades. We investigated the exposure to wood dust or leather dust in 499 patients with sinonasal tumors treated at the National Cancer Institute of Milan between 1987 and 2001 (69). Of 249 patients with ethmoid tumors, 124 had adenocarcinomas (115 males and 9 females), and 107 of these patients (86.3%) had previously been exposed to wood or


leather dusts. Only 2 of the 125 patients with ethmoid tumors other than adenocarcinomas had been exposed to these dusts. Adenocarcinomas were 17 among 250 nonethmoid sinonasal tumors; no exposure to wood or leather dusts was reported in any of these patients. The first remarkable outcome of our study was the confirmation, on a larger series, of Hadfield’s remark; (54) the ethmoid sinus is the only paranasal site where a relationship between adenocarcinoma and wood or leather exposure was evident. Actually, in many epidemiologic papers, we may find general terms like “sinonasal adenocarcinoma,” “nasal cancers,” or sinonasal cancers.” Our study confirms a very week correlation between wood or leather dust exposure and maxillary sinus tumors or ethmoid neoplasms other than ITAC. The second noteworthy result was the duration of exposure. Among males, 87 of 104 patients had been exposed to wood or leather dust for many years during their work life (25–55 years); on the contrary, 17 patients had been exposed for just a short period (less than 10 years), and many years before the clinical evidence of the tumor (23–47 years). This result was unexpected because the analysis of 12 casecontrol studies demonstrated that the risk for ethmoid adenocarcinomas is proportional to the duration of exposure to the oncogenic agents (70). By contrast, our data suggest that even a short period of exposure followed by a long latency may be sufficient for gene deregulation, which leads to the onset of the disease. Previous work from our institution (71) demonstrated a high percentage of TP53, p14ARF and p16aINK4 deregulation, and H-ras mutations in patients with ethmoid adenocarcinoma exposed to wood or leather dusts, thus supporting the epidemiologic observation of a genotoxic origin of this tumor.

Clinical Features Sinonasal adenocarcinoma may present differently based on the site of origin. In the rare cases originating in the maxillary sinus, the symptoms are similar to those of other histotypes. However, because most adenocarcinomas originate in the ethmoid sinus, the symptoms are often nonspecific and innocuous: unilateral nasal obstruction, rhinorrhea, and epistaxis (49,51,56). Pain, epiphora, and other orbital symptoms are less common and often occur late in development of the tumor. Because these symptoms frequently occur in many sinonasal diseases, ITACs are often not detected until the tumor is very large. Most patients in published series had a T3–T4 tumor. Rarely the tumor may present with a glabellar mass involving the frontal bones. The first symptom reported by our 167 patients with ethmoid adenocarcinoma who underwent an anterior craniofacial resection was in decreasing order of occurrence: nasal obstruction, 94 patients (56%); epistaxis, 48 patients (29%); and rhinorrhea, 10 patients (6%). Pain, anosmia, exophthalmos, glabellar swelling, and visual disturbances were rare. Anosmia is a very interesting symptom; even if only 3 patients reported it as the first symptom that led the patient to seek medical care, most patients remembered a partial or total anosmia some years before the beginning of nasal obstruction or epistaxis. As reported by other authors (49,56), duration of symptoms in our patients ranged from 1 to 30 months. ITAC presents as an exophytic pink mass bulging into the nasal cavity, often with a gray, necrotic, and friable appearance. Coronal CT and/or MRI with contrast enhancement are mandatory to define the extent of the disease, in particular for orbital and intracranial involvement.

449

Cervical node metastases are rare (33). Only two patients in our series had nodal involvement at presentation (1.2%). Five patients developed node metastases during follow-up, together with a relapse of the primary; however, none of them died from cervical metastases per se. The situation is different for maxillary sinus. Le et al. (72) report a 25% rate of nodal involvement for maxillary sinus adenocarcinomas, in agreement with our results (22%). Probably in some series of maxillary sinus tumors there are also nonintestinal type cases, and these tumors act like adenocarcinomas of salivary glands. Distant metastases are infrequent (33,56). Only three patients in our series had developed lung metastases (two of them together with a relapse of the primary).

Treatment Because of the rarity of ethmoid adenocarcinoma, it is nearly impossible to compare the various treatment options in a clinical trial. However, surgical radical resection is the most frequent primary treatment for patients with this tumor (9,32,73). Because most of these tumors approach or involve the cribriform plate, anterior craniofacial resection is the established “gold standard” (34,45,46,60–67,74). Some authors have demonstrated an improvement in disease-free survival with craniofacial resection compared with transfacial resection alone (33,45,74). Interest in endoscopic surgery for malignant tumors of the anterior skull base is increasing. Although most articles on this surgical approach involve esthesioneuroblastoma (75–77), some authors advocate completely endoscopic resection for ethmoid adenocarcinomas also (78,79). We believe that the resection of the sinonasal component of the tumor, however it is done, must be radical. Because ITAC is often an occupational disease with likely wide field mucosal changes, at least a total ethmoidectomy must be performed. The metaplastic transformation of ethmoid mucosa to enteric-type epithelium precedes the development of enteric adenocarcinoma (52,53), and possible preneoplastic foci may be present in macroscopically uninvolved sites of ethmoid. Moreover, ITAC is a locally aggressive tumor that easily infiltrates the underlying bone (47). As the main characteristic of endoscopic approach is the subperiosteal resection (77), we wonder if such resection may be suitable for this tumor. We have already seen three cases of ITAC arising in unresected parts of the ethmoid after an endoscopic resection performed elsewhere (Fig. 2). Other authors stress a combined endoscopic and intracranial approach to avoid any external facial incision (80,81). Because of the infrequency of nodal involvement, a prophylactic neck dissection in patients with N0 disease is not indicated (82). It is difficult to establish the results of radiotherapy alone in the treatment of patients with sinonasal adenocarcinoma. Most published series probably have some selection bias, because favorable lesions are treated with surgery, leaving larger tumors to radiotherapy. However, except as seen in a few publications, the results of radiation alone are poorer than those with treatments that include surgery (32). On the contrary, there is a general consensus about the use of radiotherapy in combination with surgery (9,32,73). The most effective sequence for surgery and radiotherapy has not been definitively determined. Although most authors prefer primary surgery (29,32,44,61,63,67), some (83) continue to choose primary radiotherapy with surgery for salvage. However, these authors report a very high rate of visual complications, with 20 of 29 patients (69%) developing blindness or impaired vision in at least one eye.

450

Cantu` et al.

resection; and postoperative radiotherapy. Twelve patients achieved a pCR, and 18 did not (overall response rate, 40%). In patients with wild-type TP53 or functional p53 protein, the pCR rates were 83% and 80%, respectively; in patients with mutated TP53 or impaired p53 protein, the pCR rates were 11% and 0%, respectively. At a median 55-month follow-up, all pCR patients were disease free; 44% of nonresponding patients had a relapse. These results are indicative of the probable existence of two genetic ITAC subgroups. The differences in TP53 mutation status or protein functionality strongly influence pCR after chemotherapy and prognosis. In spite of these encouraging results, we must remember that only 30% to 40% of ITAC patients proved to be chemoresponders; moreover, because the goal of this study was possible avoidance of surgery in chemoresponders, in our experience it was impossible to predict pCRs before surgery, even with the use of CT scan, MRI, and PET imaging. Knegt et al. (88) proposed a very unusual treatment protocol: surgical debulking via an extended anterior maxillary antrostomy followed by a combination of repeated topical chemotherapy (5-fluorouracil emulsion) and necrotomy. However, their good results have never been reproduced elsewhere. Figure 2 Relapse of ethmoid adenocarcinoma after endoscopic resection. The tumor involves unresected parts of the ethmoid.

The role of chemotherapy (intravenous or intraarterial) remains unclear. Many reports do not distinguish paranasal sinus carcinomas by specific subsites and histology. In cases in which a distinction has been made, there are few cases of adenocarcinoma (84). The only published article involving a number of ethmoid sinus adenocarcinomas suggested that chemotherapy associated with craniofacial resection and radiotherapy could improve the overall treatment outcome, and patients achieving a good clinical, and especially a pathologic, response appeared to obtain the greatest benefit (85). Based on these indications from the literature, a prospective phase II study in 49 patients with paranasal sinus cancer (47/49 with ITAC) was conducted in our institution to investigate the role of primary chemotherapy within the multidisciplinary approach to these tumors (86). After 3 to 5 cycles of chemotherapy (leucovorin, 5-fluorouracil, and cisplatin), 42 patients underwent an anterior craniofacial resection and postoperative radiotherapy. All gross specimens were carefully evaluated with at least 20 to 25 tumor sections. A complete pathologic remission (pCR) was found in eight cases (16%), and all patients achieving pCR were free of disease at a median follow-up of 26 months. In light of these results, we tried to find biologic markers able to predict the response to primary chemotherapy, to maximize treatment benefit and to avoid unnecessary toxicity in patients treated with potentially ineffective drugs (87). Because it was demonstrated (71) that the presence of TP53 mutations is one of the main genetic hallmarks of ITACs, considering the correlation between ITAC and professional exposure to wood or leather dusts, the link between genotoxic exposition and loss of p53 function, and the relationship between TP53 functional status and response to DNA-damaging treatment, we investigated 30 patients with ITAC, assessing the TP53 gene mutation profile on pretreatment biopsy. All patients underwent primary chemotherapy with cisplatin, 5-fluorouracil, and leucovorin; craniofacial

Outcome and Prognosis It is not easy to report indisputable data on outcome and prognosis for patients with ethmoid adenocarcinoma. Unfortunately, many published reports describing treatment and outcome do not distinguish paranasal sinus carcinomas in specific subsites and histology. We may find ethmoid cancers with nasal cavity and maxillary sinus tumors (44, 45,48, 49). Those few papers dealing with ethmoid tumors often contain a small number of cases and/or different histologies (46,50,52,53,56,83). Heffner et al. (49) classified their cases as low- and high-grade adenocarcinoma, reporting a good prognosis for the former and a poor outcome for the later. Barnes (51) stated 40% overall survival for patients with professional ITAC. Choi et al. (52) noted a higher local recurrence in patients with the enteric-type adenocarcinoma. In the series of Abecasis et al. (48), intestinal-type tumors were associated with a worse prognosis than were transitional tumors. Overall, in precraniofacial resection period, the reported outcomes were disappointing (33,56). Because the most important prognostic factor is local extension, the introduction of anterior craniofacial resection improved the cure rates (33,45,74). Considering that the lack of a widespread clinical prognostic classification does not allow a clear comparison among published series, we may find very different cure rates: Bridger (67), 70%; Bentz (64), 68%; Orvidas (45) and Howard (74), 58%; and Suarez (61), 31%. Ganly et al. (34) report a 5-year overall survival of 44.8% for patients with adenocarcinoma of paranasal sinuses, and who are treated with anterior craniofacial resection in 17 different institutions. However, we must remember that 5-year survival figures may fail to give the real picture of ITACs, because some patients develop a recurrence after more than 5 years (33). In Howard’s series (74), the 58% 5-year overall survival dropped to 40% at 10 years. Also Roux et al. (60) report an overall survival of 51% and 23% at 5 and 10 years, respectively. Knegt’s results (88) are astonishing (disease-free survival of 87% at 5 years and 74% at 10 years). Few papers have classified patients with ITAC in clinical stages. Roux (60) used a modified TNM staging; overall 10-year survival rates for his patients were 75% for T3, 38% for T4a, and 0% for T4b.


Five-year overall survival rate for our patients was 43.7%. Categorizing the cases according to the INT staging system (31), the corresponding rates were as follows: 61% for T2, 49% for T3, and 10% for T4. Stage by stage, we found lower cure rates for patients who underwent craniofacial resection after previous surgery performed elsewhere, in comparison with untreated patients. In conclusion, sinonasal adenocarcinomas encompass a variety of histologic types. It is difficult to declare an indisputable association between tumor histology and patient survival. Instead, survival is clearly related to intracranial extension. Anterior craniofacial resection and postoperative radiotherapy remain the standard treatment, and previous inadequate resection may jeopardize the results. The use of endoscopic resection remains controversial. As relapses may occur after many years, a long-term follow-up is crucial for patient’s survival.

REFERENCES 1. Muir CS, Nectoux J. Descriptive epidemiology of malignant neoplasms of nose, nasal cavities, middle ear and accessory sinuses. Clin Otolaryngol Allied Sci. 1980;5(3):195–211. 2. Rousc GC. Epidemiology of cancer of the nose and paranasal sinuses: Current concepts. Head Neck Surg. 1979;2(1):3–11. 3. Newman DA. A case of adeno-carcinoma of the left inferior turbinated body, and perforation of the nasal septum, in the person of a worker in chrome pigments. Glasgow Med J. 1890;33:469– 470. 4. Mannetje A, Kogevinas M, Luce D, et al. Sinonasal cancer, occupation, and tobacco smoking in European women and men. Am J Ind Med. 1999;36(1):101–107. 5. Barnes L, Eveson J, Reichart P, et al. World Health Organization Classification of Tumours: Pathology and Genetics–Head and Neck Tumours. Lyon, France: IARC Press, 2005. 6. Batsakis JG, Rice DH, Solomon AR. The pathology of head and neck tumors. Squamous and mucous gland carcinomas of the nasal cavity, paranasal sinuses and larynx. Part 7. Head Neck Surg. 1980;2(6):497–508. 7. del Regato JA. Roentgentherapy in epitheliomas of the maxillary sinus. Surg Gynecol Obstet. 1937;65:657–665. 8. Kim GE, Chung EJ, Lim JJ, et al. Clinical significance of neck node metastasis in squamous cell carcinoma of the maxillary antrum. Am J Otolaryngol. 1999;20(6):383–390. 9. Dulguerov P, Jacobsen MS, Allal AS, et al. Nasal and paranasal sinus carcinoma: Are we making progress? Cancer. 2001;92(12):3012–3129. ¨ 10. Ohngren LG. Malignant tumors of the maxillo-ethmoidal region: A clinical study with special reference to the treatment with electrosurgery and irradiation. Acta Otolaryngol (Stockolm). 1933;19:1–476. 11. Sisson GA, Johnson NE, Amiri CS. Cancer of the maxillary sinus. Clinical classification and management. Ann Otol Rhinol Laryngol. 1963;72(12):1050–1059. 12. Sakai S, Hamasaki Y. Proposal for the classification of carcinoma of the paranasal sinuses. Acta Otolaryngol (Stockolm). 1967;63(1):42–48. 13. Lederman M. Cancer of the upper jaw and nasal chambers. Proc R Soc Med. 1969; 62(1):65–72. 14. Rubin P. Cancer of the head and neck: Nose, paranasal sinuses. JAMA. 1972;219(3):336–338. 15. Chandler JR, Guillamondegui OM, Sisson GA, et al. Clinical staging of cancer of the head and neck: A new “new” system. Am J Surg. 1976;132(4):525–528. 16. Harrison DF. Critical look at the classification of maxillary sinus carcinomata. Ann Otol Rhinol Laryngol. 1978;87(1 Pt 1):3–9. 17. American Joint Committee on Cancer. AJCC Cancer Staging Manual, 5th ed. Philadelphia, PA: Lippincot-Raven, 1997. 18. Sobin LH, Wettekind C, eds. TNM Classification of Malignant Tumors, 5th ed. New York, NY: John Wiley & Sons Inc, 1997.

451

19. American Joint Committee on Cancer. AJCC Cancer Staging Manual, 6th ed. New York, NY: Springer-Verlag, 2002. 20. UICC. TNM Classification of Malignant Tumours, 6th ed. New York, NY: Wiley-Liss, 2002. 21. Sisson GA, Toriumi DM, Atiyah RA. Paranasal sinus malignancy: A comprehensive update. Laryngoscope. 1989;99(2):143– 150. 22. Spiro JD, Soo KC, Spiro RH. Nonsquamous cell malignant neoplasms of the nasal cavities and paranasal sinuses. Head Neck. 1995;17(2):114–118. 23. Kadish S, Goodman M, Wang CC. Olfactory neuroblastoma. A clinical analysis of 17 cases. Cancer. 1976;37(3):1571–1576. 24. Biller HF, Lawson W, Sachdev VP, et al. Esthesioneuroblastoma: Surgical treatment without radiation. Laryngoscope. 1990;100(11):1199–1201. 25. Dulguerov P, Calcaterra T. Esthesioneuroblastoma: The UCLA experience 1970–1990. Laryngoscope. 1992;102(8):843–849. 26. Ellingwood KE, Million RR. Cancer of the nasal cavity and ethmoid/sphenoid sinuses. Cancer. 1979;43(4):1517–1526. 27. Roux FX, Brasnu D, Menard M, et al. Les abords combinés des tumeurs malignes de l’ethmo¨ıde et autres sinus paranasaux. Principes et résultats. Ann Otolaryngol Chir Cervicofac. 1991;108(5):292–297. 28. Cantu G, Solero CL, Salvatori P, et al. A new classification of malignant ethmoid tumors. 3rd European Skull Base Congress. London, 9–11 April 1997. Skull Base Surg. 1997;7(Supp2): 33. 29. Cantu` G, Solero CL, Mariani L, et al. Anterior craniofacial resection for malignant ethmoid tumors–A series of 91 patients. Head Neck. 1999;21(3):185–191. 30. Cantu G, Solero CL, Mariani L, et al. A new classification for malignant tumors involving the anterior skull base. Arch Otolaryngol Head Neck Surg. 1999;125(11):1252–1257. 31. Cantu G, Solero CL, Miceli R, et al. Which classification for ethmoid malignant tumors involving the anterior skull base? Head Neck. 2005;27(3):224–231. 32. Dulguerov P, Allal AS. Nasal and paranasal sinus carcinoma: How can we continue to make progress? Curr Opin Otolaryngol Head Neck Surg. 2006;14(2):67–72. 33. Harrison D, Lund VJ. Non epidermoid epithelial neoplasms. In: Harrison D, Lund VJ, eds. Tumors of the Upper Jaw. Edinburgh, UK: Churchill Livingstone, 1993:105–134. 34. Ganly I, Patel SG, Sing B, et al. Craniofacial resection for malignant paranasal sinus tumors: Report of an international collaborative study. Head Neck. 2005;27(7):575–584. 35. Batsakis JG, Luna MA, el-Naggar A. Histopathologic grading of salivary gland neoplasms: III. Adenoid cystic carcinomas. Ann Otol Rhinol Laryngol. 1990;99(12):1007–1009. 36. Matsuba HM, Spector GJ, Thawley SE, et al. Adenoid cystic salivary gland carcinoma. A histopathologic review of treatment failure patterns. Cancer. 1986;57(3):519–524. 37. Spiro RH, Huvos AG, Strong EW. Adenoid cystic carcinoma of salivary origin. A clinicopathologic study of 242 cases. Am J Surg. 1974;128(4):512–520. 38. Pitman KT, Prokopakis EP, Aydogan B, et al. The role of skull base surgery for the treatment of adenoid cystic carcinoma of the sinonasal tract. Head Neck. 1999;21(5):402–407. 39. Mendenhall WM, Morris CG, Amdur RJ, et al. Radiotherapy alone or combined with surgery for adenoid cystic carcinoma of the head and neck. Head Neck. 2004;26(2):154–162. 40. Douglas JG, Laramore GA, Austin-Seymour M, et al. Treatment of locally advanced adenoid cystic carcinoma of the head and neck with neutron radiotherapy. Int J Radiat Oncol Biol Phys. 2000;46(3):551–557. 41. Douglas JG, Koh W, Ausin-Seymour M, et al. Treatment of salivary gland neoplasms with fast neutron radiotherapy. Arch Otolaryngol Head Neck Surg. 2003;129(9):944–948. 42. Laurie SA, Licitra L. Systemic therapy in the palliative management of advanced salivary gland cancers. J Clin Oncol. 2006;24(17):2673–2678. 43. Pitman KT, Prokopakis EP, Aydogan B, et al. The role of skull base surgery for the treatment of adenoid cystic carcinoma of the sinonasal tract. Head Neck. 1999;21(5):402–407.

452

Cantu` et al.

44. Katz TS, Mendenhall WM, Morris CG, et al. Malignant tumors of the nasal cavity and paranasal sinuses. Head Neck. 2002;24(9):821–829. 45. Orvidas LJ, Lewis JE, Weaver AL, et al. Adenocarcinoma of the nose and paranasal sinuses: A retrospective study of diagnosis, histologic characteristics, and outcomes in 24 patients. Head Neck. 2005;27(5):370–375. 46. Wax MK, Yun KJ, Wetmore SJ, et al. Adenocarcinoma of the ethmoid sinus. Head Neck. 1995;17(4):303–311. 47. Batsakis JG. Mucous gland tumors of the nose and paranasal sinuses. Ann Otol Rhinol Laryngol. 1970;79(3):557–562. 48. Abecasis J, Viana AG, Pissarra C, et al. Adenocarcinomas of the nasal cavity and paranasal sinuses: A clinicopathological and immunohistochemical study of 14 cases. Histopathology. 2004;45(3):254–259. 49. Heffner DK, Hyams VJ, Hauck KW, et al. Low-grade adenocarcinoma of the nasal cavity and paranasal sinuses. Cancer. 1982;50(2):312–322. 50. Bashir AA, Robinson RA, Benda JA, et al. Sinonasal adenocarcinoma: immunohistochemical marking and expression of oncoproteins. Head Neck. 2003;25(9):763–771. 51. Barnes L. Intestinal-type adenocarcinoma of the nasal cavity and paranasal sinuses. Am J Surg Pathol. 1986;10(3):192–202. 52. Choi HR, Sturgis EM, Rashid A, et al. Sinonasal adenocarcinoma: evidence for histogenetic divergence of the enteric and nonenteric phenotypes. Hum Pathol. 2003;34(11):1101–1107. 53. Kennedy MT, Jordan RCK, Berean KW, et al. Expression pattern of CK7, CK20, CDX-2, and villin in intestinal-type sinonasal adenocarcinoma. J Clin Pathol. 2004;57(9):932–937. 54. Hadfield EH. A study of adenocarcinoma of the paranasal sinuses in woodworkers in the furniture industry. Ann R Coll Surg Engl. 1970;46(6):301–319. 55. Acheson ED. Nasal cancer in the furniture and boot and shoe manufacturing industries. Prev Med. 1976;5(2):295–315. 56. Klintenberg C, Olofsson J, Hellquist H, et al. Adenocarcinoma of the ethmoid sinuses. A review of 28 cases with special reference to wood dust exposure. Cancer. 1984;54(3):482–488. 57. Luce D, Leclerc A, Morcet JF, et al. Occupational risk factors for sinonasal cancer: A case-control study in France. Am J Ind Med. 1992;21(2):163–175. 58. Hernberg S, Westerholm P, Shultz-Larsen K, et al. Nasal and sinonasal cancer. Connection with occupational exposures in Denmark, Finland and Sweden. Scand J Work Environ Health. 1983;9(4):315–326. 59. Voss R, Stenersen T, Roald Hoppedal B, et al. Sinonasal cancer and exposure to softwood. Acta Otolaryngol. 1985;99(1–2):172– 188. 60. Roux FX, Pages JC, Nataf F, et al. Malignant ethmoidsphenoidal tumors. 130 cases. Retrospective study. Neurochirurgie. 1997;43(2):100–110. 61. Suarez C, Llorente JL, Fernandez DeLeon R, et al. Prognostic factors in sinonasal tumors involving the anterior skull base. Head Neck. 2004;26(2):136–144. 62. Cheesman AD, Craniofacial resection for tumors of the nasal cavity and paranasal sinuses. Head Neck Surg. 1986;8(6):429– 435. 63. McCutcheon IE, Blacklock JB, Weber RS, et al. Anterior transcranial (craniofacial) resection of tumors of the paranasal sinuses: surgical technique and results. Neurosurgery. 1996;38(3):471–479; discussion 479–480. 64. Bentz BG, Bilsky MH, Shah JP, et al. Anterior skull base surgery for malignant tumors: A multivariate analysis of 27 years of experience. Head Neck. 2003;25(7):515–520. 65. Donald PJ. Complications in skull base surgery for malignancy. Laryngoscope. 1999;109(12):1959–1966. 66. Irish JC, Gullane PJ, Gentili F, et al. Tumors of the skull base: Outcome and survival analysis of 77 cases. Head Neck. 1994;16(1):3– 10.

67. Bridger GP, Kvok B, Baldwin M, et al. Craniofacial resection for paranasal sinus cancers. Head Neck. 2000;22(8):772–780. 68. Nylander LA, Dement JD. Carcinogenetics effects of wood dust: Review and discussion. Am J Ind Med. 1993;24(5):619–647. 69. Bimbi G, Saraceno MS, Riccio S, et al. Adenocarcinoma of ethmoid sinus: An occupational disease. Acta Otorhinolaryngol Ital. 2004;24(4):199–203. 70. Demers PA, Kogevinas M, Boffetta P, et al. Wood dust and sinonasal cancer: pooled reanalysis of twelve case-control studies. Am J Ind Med. 1995;28(2):151–166. 71. Perrone F, Oggionni M, Birindelli S, et al. TP53, p14ARF, p16INK4a and H-ras gene molecular analysis in intestinal-type adenocarcinoma of the nasal cavity and paranasal sinuses. Int J Cancer. 2003;105(2):196–203. 72. Le QT, Fu KK, Kaplan MJ, et al. Lymph node metastasis in maxillary sinus carcinoma. Int J Radiat Oncol Biol Phys. 2000;46(3):541– 549. 73. Waldron J, Witterick I. Paranasal sinus cancer: Caveats and controversies. World J Surg. 2003;27(7):849–855. 74. Howard DJ, Lund VJ, Wei WI. Craniofacial resection for tumors of the nasal cavity and paranasal sinuses: A 25-year experience. Head Neck. 2006;28(10):867–873. 75. Casiano RR, Numa WA, Falquez AM. Endoscopic resection of esthesioneuroblastoma. Am J Rhinol. 2001;15(4):271–279. 76. Unger F, Haselsberger K, Walch C, et al. Combined endoscopic surgery and radiosurgery as treatment modality for olfactory neuroblastoma (esthesioneuroblastoma). Acta Neurochir (Wien). 2005;147(6):595–601. 77. Castelnuovo PG, Delu G, Sberze F, et al. Esthesioneuroblastoma: Endonasal endoscopic treatment. Skull Base. 2006;16(1):25–30. 78. Haberman W, Zanarotti U, Groell R, et al. Combination of surgery and gamma knife radiosurgery–a therapeutic option for patients with tumors of nasal cavity or paranasal sinuses infiltrating the skull base. Acta Otorhinolaryngol Ital. 2002;22(2):74–79. 79. Bockmuhl U, Minovi A, Kratzsch B, et al. Endonasal microendoscopic tumor surgery: State of the art. Laryngorhinootologie. 2005;84(12):884–891. 80. Thaler ER, Kotapka M, Lanza DC, et al. Endoscopically assisted anterior cranial skull base resection of sinonasal tumors. Am J Rhinol. 1999;13(4):303–310. 81. Castelnuovo PG, Belli E, Bignami M, et al. Endoscopic nasal and anterior craniotomy resection for malignant nasoethmoid tumors involving the anterior skull base. Skull Base. 2006;16(1):15–18. 82. Jiang GL, Morrison WH, Garden AS, et al. Ethmoid sinus carcinomas: Natural history and treatment results. Radiother Oncol. 1998;49(1):21–27. 83. Waldron JN, O’Sullivan B, Warde P, et al. Ethmoid sinus cancer: Twenty-nine cases managed with primary radiation therapy. Int J Radiat Oncol Biol Phys. 1998;41(2):361–369. 84. Samant S, Robbins KT, Vang M, et al. Intra-arterial cisplatin and concomitant radiation therapy followed by surgery for advanced paranasal sinus cancer. Arch Otolaryngol Head Neck Surg. 2004;130(8):948–955. 85. Brasnu D, Laccourreye O, Bassot V, et al. Cisplatin-based neoadjuvant chemotherapy and combined resection for ethmoid sinus adenocarcinoma reaching and/or invading the skull base. Arch Otolaryngol Head Neck Surg. 1996;122(7):765–768. 86. Licitra L, Locati LD, Cavina R, et al. Primary chemotherapy followed by anterior craniofacial resection and radiotherapy for paranasal cancer. Ann Oncol. 2003;14(3):367–372. 87. Licitra L, Suardi S, Bossi P, et al. Prediction of TP53 status for primary cisplatin, fluorouracil, and leucovorin chemotherapy in ethmoid sinus intestinal-type adenocarcinoma. J Clin Oncol. 2004;22(24):4901–4906. 88. Knegt PP, Ah-See KW, vd Velden LA, et al. Adenocarcinoma of the ethmoidal sinus complex: Surgical debulking and topical fluorouracil may be the optimal treatment. Arch Otolaryngol Head Neck Surg. 2001;127(2):141–146.

31 Esthesioneuroblastoma Valerie J. Lund and David J. Howard

preponderance is more pronounced with a 2:1 ratio and the age range is 12 to 70 years (mean 46 years), though children under 10 have occasionally been reported (16)(Fig. 1).

INCIDENCE AND EPIDEMIOLOGY In common with all tumors of the anterior skull base, esthesioneuroblastoma (ENB) is comparatively rare. This malignant neuroendocrine neoplasm arose from the olfactory mucosa and was first recognized by Berger et al. in 1924 (1), who coined the term “esthesioneuroepitheliome olfactif.” However, a wide range of other terms has been used including esthesioneurocytoma, esthesioneuroma, intranasal neuroblastoma, olfactory neuroepithelial tumor, and olfactory neuroblastoma. In 1966, Skolnik et al. (2) found only 97 cases reported in 42 papers in the English literature with most authors only treating 2 or 3 cases, and by 1989 O’Connor estimated that ≤300 cases had been published which represented 1% to 5% of all malignant tumors of the nasal cavity (3). However, this number had risen to 945 by 1997 (4), a report which did not include a large series from the Armed Forces Institute of Pathology (5) nor from the Institut Gustave-Roussy (6). In 2000, the National Cancer Data Base included 664 cases from over 500 U.S. hospitals over a 10-year period (1985–1995). In recent years, increasing numbers of this tumor are being described, almost certainly due to an increasing awareness and improved histological techniques for diagnosis. Despite this, it remains difficult to accrue large individual series, compromising statistical analysis of outcome. The authors, working in a tertiary referral center, have had the opportunity of managing 78 cases since 1970. Hitherto, unlike adenocarcinoma, occupational factors in the development of olfactory neuroblastoma have not been identified in men other than a single case report in a woodworker (7). However, in rodents the administration of N-nitroso compounds has been reported to produce esthesioneuroepitheliomas (8–10) when administered parenterally, orally, or topically, as has administration of bischloromethyl ether (11). To evaluate possible etiological factors, the occupational history and exposure to possible carcinogens was examined in a series of 54 patients with olfactory neuroblastoma using questionnaires and/or structured interviews. This revealed four individuals (8 %) who were dental practitioners (2) or dental nurses (2), whereas no other members of the dental profession have been found in a cohort of over 700 other sinonasal malignancies (12). It is neither clear what chemical, if any, is implicated nor does this constitute sufficient evidence to determine a definite link but is worthy of continued observation. No hereditary patterns have been described in this tumor and there is no apparent racial predilection. In the literature, there appears to be a slight male preponderance and the tumor may occur over a wide age range (3–90 years) with a reported bimodal peak in the second/third decades and sixth/seventh decades (3,13–15). In our own series, this male

PATHOLOGY Olfactory neuroblastoma generally arises in the nasal roof, corresponding to the anatomical distribution of the olfactory epithelium, which extends from the olfactory niche onto the upper nasal septum and superior turbinates on the lateral wall. Evidence for the origin of olfactory neuroblastoma from specialized olfactory epithelium, however, is somewhat circumstantial (5), though tumors occasionally found outside this distribution have been ascribed to ectopic olfactory epithelium. Tumors arising in the cribriform niche can easily spread superiorly along olfactory fibers into the anterior cranial fossa to affect the olfactory bulb and tracts. The superior septum is often involved and from thence the tumor may spread to the contralateral side and into the ethmoids and adjacent orbit. Histological studies suggest that there is microscopic intracranial involvement in the majority of patients even when not suggested by imaging and the macroscopic appearances at surgery (17). Macroscopically, the tumor is characteristically a polypoid reddish gray mass that bleeds readily. Microscopically, the tumor typically forms clusters of cells arranged in patterns, which vary from small nests surrounded by a fibrillary stroma to diffuse areas separated by fibrovascular septa. The cells may palisade around blood vessels and occasionally true rosettes form. It had been suggested that ENB was part of the Ewing sarcoma/peripheral neuroectodermal group of tumors but this has not been supported by immunohistochemical studies (18). However, it continues to present some difficulties in diagnosis even with modern techniques and can be confused with a host of other small-cell tumors such as lymphoma, malignant melanoma, and undifferentiated sinonasal carcinoma by those unfamiliar with sinonasal malignancy. This prompted Ogura and Schenck to describe ENB as the “great imposter” (19). Immunohistochemistry using a broad panel of antibodies is usually employed to confirm the diagnosis. These include general neuroendocrine markers such as neuronspecific enolase, synaptophysin, chromogranin, and protein gene product–9.5, which are usually positive (20). S-100 positivity can be demonstrated at the periphery of the tumor nests and some tumors are also positive using MNF 116 and CAM 5.2, both stains for certain cytokeratins. Conversely LP 34, a high molecular weight cytokeratin stain, epithelial membrane antigen, carcinoembryonic antigen, and glial fibrillary acidic protein are generally negative. 453

454

Lund and Howard

The age distribution of the olfactory tumors 8 7

Number of cases

6 5 4 3 2 1 0

11–15 16–20 21–25 26–30 31–35 36–40 41–45 46–50 51–55 56–60 61–65 66–70 71–76 76–80 Age (years)

Figure 1 Histogram showing distribution of patients by age in authors’ cohort.

Other features such as microfilaments, microtubules, and neurosecretory granules may be seen on electron microscopy. Indeed, this technique was used in the days before immunohistochemistry to confirm diagnosis. Histological features have been used by some authors to determine prognosis, with limited success (5,21,22).

STAGING A number of staging systems have been proposed. Kadish et al. (23) is a somewhat crude system dividing the tumors into three stages. Stage A: lesions confined to the nasal cavity Stage B: involvement of nasal cavity + one or more of the paranasal sinuses Stage C: involvement beyond the nasal cavity including the orbit, skull base, intracranial cavity, cervical lymph nodes, or systemic metastases. This can now be considerably refined by the use of modern imaging protocols (24) validated by craniofacial resection. However, neither the Kadish system nor subsequent modifications (5,22,25) have proved entirely successful, largely due to the late presentation of most patients, despite efforts to refine advanced disease by creating a Stage D for metastases.

MANAGEMENT Clinical Features The usual site of origin results in fairly innocuous symptoms initially, only remarkable by their sudden onset and unilaterality. As a consequence, there is often considerable delay in diagnosis, with some patients waiting for more than a year [24% of 40 cases reported by Schwabb et al. (6)]. There is little specific to this particular tumor, the symptoms being common to all nasal cavity lesions, that is, blockage, discharge, some epistaxis due to the tumor’s vascularity, and hyposmia. In a series of 42 patients, unilateral obstruction occurred in 93%, epistaxis in 55%, and rhinorrhea in 30% (15). Anosmia was reported rather rarely (5%) in our patients and invasion of the anterior cranial fossa is otherwise generally silent. As the tumor spreads to the orbit, patients may get epiphora, displacement of the eye, diplopia, and eventually visual loss,

though this is usually a late phenomenon. Ocular symptoms occurred in 11% of our patients (15). Curiously, in this series the left side was more often affected (62%) compared to the right (29%), with both sides affected at presentation in 9%. Occasionally, involvement of the Eustachian orifice may result in otalgia and a conductive hearing loss. The incidence of cervical metastases varies considerably from report to report, compounded by the generally small numbers in each series (14). A retrospective review (26) suggested an incidence of 27% based on 10 series, which included 207 cases. However, if only Kadish Stage C was considered, the rate increased to 44% but subsequent review suggests that some of the data in this study was incorrect. In 1993, Harrison and Lund (27) found only one case out of 20 in contrast to Morita et al. (25), who reported 20.4% (10/49) in a series from the Mayo Clinic. Reviewing 320 cases, reported in 15 series, Rinaldo et al. (14) reported a lymph node metastatic frequency of 23.4% (range 5–100%), though only eight studies validated the diagnosis of ENB with immunohistochemistry. As ENB is a neuroendocrine tumor, it can be associated with syndromes caused by inappropriate hormone production such as Cushing or antidiuretic secretion (6,28–31).

Imaging All patients are ideally submitted to a preoperative imaging protocol, which employs a combination of high-resolution contrast-enhanced computer tomography (coronal and axial planes) combined with multiplanar magnetic resonance imaging (MRI) enhanced with gadolinium diethylenetriamine pentaacetic acid (DTPA) (24). Following surgery, all patients should also undergo a rigorous follow-up protocol of MRI, ideally combined with sinonasal endoscopy (with biopsy of any suspicious lesions), every 3 to 4 months for the first 2 years and then 6 monthly thereafter (32). Patients may develop recurrence many years after initial treatment and this can be anywhere in relation to the surgical field, orbit, or intracranial cavity suggesting some local embolic phenomenon. Appropriate imaging using a combination of computed tomography (CT) and magnetic resonance imaging (MRI) will often suggest the diagnosis but more importantly indicate extent (24,33). There are no features which are specific to ENB, though the position of the mass and associated bone erosion indicates a malignant nasal tumor (Fig. 2). Coronal CT remains the most accurate method of demonstrating early anterior skull base erosion, while the addition of contrast enhancement and MRI shows extent of intracranial and orbital spread. Typical features are an intense signal on precontrast T2-weighted spin echo sequences and strong enhancement after gadolinium on T1-weighted sequences. However, even the most sophisticated imaging cannot be absolutely relied upon to demonstrate involvement of the dura and orbital periosteum, which can only be determined by surgery with histological confirmation. Ultrasound of the neck combined with fine-needle aspiration cytology is an important screening technique with a high degree of accuracy (34). Although metastatic spread can occur outside the head and neck, it is rare at presentation, and, therefore, it is not the authors’ practice to undertake more extensive body imaging at this stage.

Treatment The advent of craniofacial resection has revolutionized the treatment of ENB, doubling survival figures and is now regarded as the “gold standard.” In a meta-analysis, Dulgerov et al. confirmed that this procedure combined with radiotherapy was the treatment of choice (13).

Chapter 31: Esthesioneuroblastoma

(A)

455

(C)

(B)

(D)

Figure 2 Imaging of olfactory neuroblastoma. (A) Coronal CT showing small tumor confined to nasal cavity. (B) Coronal MRI in same patient (T1 with gadolinium) confirming extent of lesion and suitable for endoscopic resection. (C) Coronal CT showing larger tumor apparently confined to nasal cavity in another patient. (D) Contemporaneous coronal MRI in same patient (T1 with gadolinium) showing obvious extension into anterior cranial fossa requiring craniofacial resection. (C) Coronal CT showing larger tumor apparently confined to nasal cavity in another patient. (D) Contemporaneous coronal MRI in same patient (T1 with gadolinium) showing obvious extension into anterior cranial fossa requiring craniofacial resection.

Surgery Craniofacial resection was introduced in the 1970s by Ketcham and others, providing the combination of an en bloc oncologic resection with low morbidity and excellent cosmesis (35–37). By approaching the tumor from the nose and anterior cranial fossa, the operation directly addresses the origin and local spread of this tumor, allowing resection of dura and the olfactory system including the olfactory epithelium, cribriform plate, olfactory bulb, and tracts. This directly deals with macro- and microscopic spread of disease, reducing local recurrence (38,39). There are many variations on the technique but essentially all involve some form of craniotomy, together with a nasal approach using various incisions and forms of repair.

Using a coronal incision in the scalp and a sublabial incision for a midfacial degloving, these scars can be hidden, though the use of an extended lateral rhinotomy or a supraorbital spectacle incision heals well. The skull base repair may be affected with a pericranial flap or fascia lata and split skin. In our series which extended over 27 years, using this technique of craniofacial resection in 308 patients, postoperative hospital stay has been on average 14 days and major complications low (39) (Table 1). Prior to craniofacial resection, conventional wisdom dictated that the orbit should be sacrificed if tumor was either adjacent or had transgressed the periosteum. However, it is clear that it is possible to salvage a proportion of these eyes without compromising survival. If the tumor has not

456

Lund and Howard Table 1 Complications Associated with Craniofaciala and Endoscopic Resection Complication Immediate DVT CSF leak Long term Glaucoma Epilepsy Epiphora Serous otitis media Frontal bone necrosis None Total cohort

Craniofacial n

Endoscopic n

Chemotherapy

1 1 1 1 1 1 3 37 42

advent of intensity modulation radiotherapy may of some advantage in this respect. Generally, radiotherapy has been used as an adjunct to surgery (15,49) and postoperative delivery is preferred.

1

10 11

a Complications multiple in some cases. Abbreviations: DVT, deep-vein thrombosis; CSF, cerebrospinal fluid.

penetrated through the full thickness of orbital periosteum on frozen section, it is possible to resect it widely and skin graft the area. Nonetheless, the orbit should be cleared if there is full thickness periosteal penetration or frank infiltration of orbital contents. More recently, endoscopic resection has been used for selected cases, usually for those who are without significant skull base erosion or for patients who are a poor anaesthetic risk. The endoscopic approach should not be regarded as a limited procedure, rather as the nasal component of the craniofacial resection (40) that includes a wide-field resection of tissue. The main difference is that this is undertaken in a more piecemeal fashion but is performed under excellent direct visualization. The ability to perform skull base resection and repair via an endoscopic approach has facilitated this (41), and, indeed, some surgeons have advocated undertaking extensive intracranial resection from below in teams which involve neurosurgical input (42). Alternatively, the endoscopic approach can be combined with an external craniotomy (43). Patients should be apprised that both might be required or that an endoscopic approach may need to be extended to include a formal craniotomy. In a recent series of 49 patients undergoing endoscopic resection for malignant sinonasal tumors, 11 were olfactory neuroblastomas but follow-up was a mean of only 36 months (40). While all are presently alive, one has been converted to craniofacial resection at 1 year. In other publications, cohorts of malignant tumors including olfactory neuroblastoma have been described (44–48) but, as in this series, follow-up was relatively short. However, complication rates, if any, are very low, hospital stay on average 5 days and postoperative radiotherapy may be started very promptly (Table 1). It should also be noted that endoscopic surgery could have a role following conventional craniofacial resection in the management of localized recurrence. The neck is not traditionally treated prophylactically, although a selective neck dissection is undertaken in the presence of disease.

Radiotherapy Radiation in this area must be carefully administered to deliver the maximum dose while preserving the adjacent brain and optic nerves. An external megavoltage beam and threefield technique has generally been used. An anterior port combined with wedged lateral fields delivers a dose of 55 to 65 Gy. Because of the proximity of the optic chiasm, the dose given must remain below normal tissue tolerance and the

The use of concomitant chemotherapy has not been fully evaluated, though chemosensitivity has been found in retrospective series. ENB has been shown to respond to platinumbased regimes (50–54). In 8 of 18 of our patients between 1999 and 2005 who received radiotherapy and cisplatin, results suggested a reduced recurrence rate, though the number of patients precludes any reliable statistical conclusions (55).

OUTCOME AND PROGNOSIS Prior to craniofacial resection, the use of lateral rhinotomy and radiotherapy provided poor results (56–58) of ≤40% at 5 years, largely due to the inability to deal with intracranial spread. Craniofacial resection specifically addresses this area and allows removal of the olfactory bulbs and tracts where microscopic disease may be residing undetected. Thus when large series with long-term follow-up after craniofacial are considered, the 5-year survival is seen to have improved significantly (59–63) or even doubled as in our own series to 77% (15) or to 89% in that of Diaz et al. (59). However, there is continued loss over time and local recurrence can occur up to 12 years after treatment (range 12–144 months, mean 37 months). In our series of 42 cases, disease-free survival drops from 77% at 5 years to 53% at 10 years. In a further study of this cohort enlarged to 56 individuals, 15-year survival fell to 40%, which emphasizes the importance of long-term follow-up. The most frequent recurrence is local and occurred in 17% of our series in keeping with other published series using craniofacial resection and radiotherapy. Local recurrence has been shown to be decreased by the addition of radiotherapy, but it does not seem to matter whether this is given before or after surgery, even when the therapeutic interval differs between pre- and postoperative administration. This applies to both survival (p = 0.515) and complications (p = 0.07). Interestingly, previous treatment did not seem to affect 5-year actuarial survival either, but in the patients who developed local recurrence, 5-year survival after further salvage treatment was 54%. However, it should again be noted that the site of “local” recurrence can be anywhere and on either side of the nose, sinuses, orbit, or intracranial cavity, so follow-up must be especially vigilant if this disease is to be detected early. Multivariate analysis of the craniofacial series shows that involvement of the brain and orbit are independent factors affecting outcome (15). When survival was considered according to orbital involvement, 5-year actuarial survival was 97% when the eye was not affected and 49% when the periosteum was affected, but when the eye was frankly infiltrated even when the eye was sacrificed, there were no 5-year survivors (p = 0.0067) (Fig. 3). When skull base and intracranial involvement were considered, there was a difference between those whose tumor was confined to the nasal cavity, when compared with those in whom the skull base was affected; those in whom the olfactory tracts were involved; those in whom the dura was additionally infiltrated; and those in whom the brain was affected (p = 0.035). When patients with tumor in the

Chapter 31: Esthesioneuroblastoma

457

cial resection specifically addresses the site of origin and local spread of the tumor and when combined with radiotherapy offers the optimum treatment, which has doubled survival as compared with previous treatments. However, endoscopic resection may offer an alternative in selected cases, though comparable numbers and follow-up are lacking as yet. Thus, the results of endoscopic surgery must be compared with the large craniofacial series with 10 or more years of follow-up as the natural history of this tumor extends over a lifetime. Whichever surgical approach is used, when the orbital periosteum is affected without penetration into the orbital structures themselves, the eye can be preserved without an adverse effect on survival.

REFERENCES

Figure 3 Kaplan Meier survival for orbital involvement in olfactory neuroblastoma. Source: From Ref. 15.

nasal cavity and/or skull base were compared with the other groups, as might be expected there was a statistical difference between these patients and those with dura and/or olfactory tract involvement (p = 0.006) and those with brain involvement (p = 0.039) (Fig. 4). Those patients treated by endoscopic approaches are too limited in number and follow-up to make any meaningful comparison at present, but it is clear that in carefully selected individuals, early results are at least as good as craniofacial resection (40,44–47) and the complication rate and morbidity commensurately less (64). However, given the case mix, this is to be expected. Cervical lymphadenopathy constitutes an important prognostic factor. Koka et al. (31) showed a 29% survival with nodes versus 64% without nodes, and this was supported by two subsequent meta-analyses (13,14). Distant metastases with locoregional control are relatively rare [≤10% (15,54)] and carry a poor prognosis.

CONCLUSION ENB is a rare nasal tumor with a unique natural history. Its management requires experience in histopathology, radiology, sinonasal surgery, and medical oncology. Craniofa-

Figure 4 Kaplan Meier survival for skull base and intracranial involvement in olfactory neuroblastoma. Source: From Ref. 15.

1. Berger L, Luc R, Richard D. L’estheioneuroepitheliome olfactif. Bull Assoc Fr Etude Cancer. 1924;13:410–421. 2. Skolnik EM, Massari FS, Tenta LT. Olfactory neuroepithelioma. Arch Otolaryngol. 1966;84:644–653. 3. O’Connor TA, McLean P, Juillard GS. Olfactory neuroblastoma. Cancer. 1989;63:2426–2428. 4. Broich G, Pagliaria, Ottaviani F. Esthesioneuroblastoma: A general review of the cases published since the discovery of the tumour in 1924. Anticancer Res. 1927;17:2683–2706. 5. Hyams VJ. Olfactory neuroblastoma. In: Hyams VJ, Baksakis JG, Michaels L, eds. Tumours of the Upper Respiratory Tract and Ear. Washington, DC: Armed Forces Institute of Pathology, 1998:240– 248. 6. Schwaab G, Michaeu C, Le Guillou C, et al. Olfactory esthesioneuroma: A report of 40 cases. Laryngoscope. 1988;98:872–876. 7. Magnavita N, Sacco A, Bevilacqua L, et al. Aesthesioneuroblastoma in a woodworker. Occup Med. 2003;53:231–234. 8. Magee PN, Montesano R, Preussmann R. N-nitroso compounds and related carcinogens. In: Searle CE, ed. Chemical carcinogens. Washington, DC: USA American Chemical Society, 1976:491–625. 9. Herrold K. Induction of olfactory neuroepithelia tumours in Syrian hamsters by diethlynitrosamine. Cancer. 1964;17:114–121. 10. Vollrath M, Altmannsberger M, Weber K, et al. Chemically induced tumours of rat olfactory epithelium: A model for human esthesioneuroepithelioma. J Natl Cancer Inst. 1986;76:1205–1169. 11. Leong BK, Kociba RJ, Jersey GC. A lifetime study of rats and mice exposed to vapour of bis(chloromethyl)ether. Toxicol Appl Pharmacol. 1981;58(2):269–281. 12. Personal communication with VJ Lund. 13. Dulguerov P, Abdelkarim SA, Calcaterra TC. Esthesioneuroblastoma: A meta-analysis and review. Lancet Oncol. 2001;2:683–688. 14. Rinaldo A, Ferlito A, Shaha AR, et al. Esthesioneuroblastoma and cervical lymph node metastases: Clinical and therapeutic implications. Acta Otolaryngol. 2002;122:125–221. 15. Lund VJ, Howard D, Wei W, et al. Olfactory neuroblastoma. Laryngoscope. 2003;113:502–507. 16. Kumar M, Fallon R, Hill J, et al. Esthesioneuroblastoma in children. J Pediatr Hematol Oncol. 2002;24:482–487. 17. Harrison D. Surgical pathology of olfactory neuroblastoma. Arch Otolaryngol. 1984;7:60–64. 18. Argani P, Perez-Ordonez B, Xiao H, et al. Olfactory neuroblastoma is not related to the Ewing family of tumors: Absence of EWS/FL11 gene fusion and MIC2 expression. Am J Surg Pathol. 1998;22:391–398. 19. Ogura J, Schenck N. Unusual nasal tumors. Problems in diagnosis and treatment. Otolaryngol Clin N Am. 1973;6:813–837. 20. Lund VJ, Milroy C. Olfactory neuroblastoma: Clinical and pathological aspects. Rhinology. 1993;31:1–6. 21. Hirose T, Scheithauer BW, Lopes MB, et al. Olfactory neuroblastoma: An immunohistochemical, ultrastructural, and flow cytometric study. Cancer. 1995;76:4–19. 22. Papadaki H, Kounelis S, Kapadia S, et al. Relationship of p53 gene alterations with tumor progression and recurrence in olfactory neuroblastoma. Am J Surg Pathol. 1996;20:715–721.

458

Lund and Howard

23. Kadish S, Goodman M, Wang C. Olfactory neuroblastoma: A clinical analysis of 17 cases. Cancer. 1976;37:1571–1576. 24. Lloyd GAS, Lund VJ, Howard DJ, et al. Optimum imaging for sinonasal malignancy. J Laryngol Otology. 2000;114;557–562. 25. Morita A, Ebersold M, Olsen K, et al. Esthesioneuroblastoma: Prognosis and management. Neurosurgery. 1993;32:706–715. 26. Davis R, Weissler M. Esthesioneuroblastoma and neck metastasis. Head Neck. 1992;14:477–482. 27. Harrison DNF, Lund VJ. Neuroectodermal lesions. In: Tumours of the Upper Jaw, London, Edinburgh, UK: Churchill Livingstone. 1993:295–328. 28. Singh W, Ramage C, Best P, et al. Nasal neuroblastoma secreting vasopressin. Cancer. 1980;45:961–966. 29. Srigley J, Dayal V, Gregor R, et al. Hyponatremia secondary to olfactory neuroblastoma. Arch Otolaryngol. 1983;109:559– 562. 30. Myers S, Hardy D, Wiebe C, et al. Olfcatory neuroblastoma invading the oral cavity in a patient with inappropriate antidiuretic hormone secretion. Oral Surg Oral Med Oral Pathol. 1994;77:645– 650. 31. Koka V, Julieron M, Bourhis J, et al. Aesthesioneuroblastoma. J Laryngol Otol. 1998;112:628–633. 32. Lund VJ, Howard DJ, Wei WI, et al. Craniofacial resection for tumors of the nasal cavity and paranasal sinuses—a 17-year experience. Head Neck. 1998;20;97–105. 33. Pickuth D, Heywang-Kobrunner S, Spielman R. Computed tomography and magnetic resonance imaging features of olfactory neuroblastoma: An analysis of 22 cases. Clin Otolaryngol. 1999;24;457–461. 34. Collins B, Cramer H, Hearn S. Fine needle aspiration cytology of metastatic neuroblastoma. Acta Cytol. 1997;41:802–810. 35. Ketcham AS, Van Buren JM. Tumors of the paranasal sinuses: A therapeutic challenge. Am J Surg. 1985;150:406–413. 36. Clifford P. Transcranial approach to cancer of the antro-ethmoidal area. Clin Otolaryngol. 1977;2:115–130. 37. Terz JJ, Young HF, Lawrence W Jr. Combined craniofacial resection for locally advanced carcinoma of the head and neck. Am J Surg. 1980;140:613–624. 38. Shah JP, Kraus DH, Bilsky MH, et al. Craniofacial resection for malignant tumors involving the anterior skull base. Arch Otolaryngol Head Neck Surg. 1997;123:1312–1317. 39. Howard DJ, Lund VJ, Wei WI. Craniofacial resection for sinonasal neoplasia—a twenty-five year experience. Head Neck. 2006;28:867–873. 40. Lund VJ. Endoscopic resection of malignant tumours of the nose and sinuses. Am J Rhinol Fast Track. 2007;21:89–94. 41. Lund VJ. Endoscopic management of CSF leaks. Am J Rhinol. 2002;16:17–23. 42. Kassam A, Horwitz M, Welch W, et al. The role of endoscopic assisted microneurosurgery (image fusion technology) in the performance of neurosurgical procedures. Minim Invasive Neurosurg. 2005;48:191–196. 43. Thaler ER, Kotapka M, Lanza D, et al. Endoscopically assisted anterior cranial skull base resection of sinonasal tumors. Am J Rhinol. 1999;13:303–310. 44. Draf W, Schick B, Weber R, et al. Endoscopic micro-endoscopic surgery of nasal and paranasal sinus tumours. In: Stamm AC,

45.

46.

47. 48.

49.

50.

51.

52. 53.

54.

55.

56. 57.

58. 59. 60. 61. 62.

63.

64.

Draf W, eds. Micro-Endoscopic Surgery of the Paranasal Sinuses and the Skull Base. Berlin, Germany: Springer, 2000:481–488. Goffart Y, Jorissen M, Daele J, et al. Minimally invasive endoscopic management of malignant sinonasal tumours. Acta Otorhinolaryngol Belg. 2000;54:221–232. Walch C, Stammberger H, Anderhuber W, et al. The minimally invasive approach to olfactory neuroblastoma: Combined endoscopic and stereotactic treatment. Laryngoscope. 2000;110;635– 640. Casiano RR, Numa WA, Falquez AM. Endoscopic resection of esthesioneuroblastoma. Am J Rhinol. 2001;15;271–279. Batra PS, Citardi MJ, Worley S, et al. Resection of anterior skull base tumors: Comparison of combined traditional an endoscopic techniques. Am J Rhinol. 2005;19;521–528. Foote RL, Morita A, Ebersold M, et al. Esthesioneuroblastoma: The role of adjuvant radiation therapy. Int J Radiat Oncol Biol Phy. 1993;27:835–842. McElroy E, Buckner J, Lewis J. Chemotherapy for advanced esthesioneuroblastoma: The Mayo Clinic experience. Neurosurgery. 1998;42:1023–1028. Kim D, Jo Y, Heo D, et al. Neoadjuvant etoposide, ifosafamide and cisplatin for the treatment of olfactory neuroblastoma. Cancer. 2004;101:2257–2260. Dulguerov P, Calcaterra T. Esthesioneuroblastoma: The UCLA experience 1970–1990. Laryngoscope. 1992;102:843–849. Eden BV, Debo RF, Larner JM, et al. Esthesioneuroblastoma: Long-term outcome and patterns of failure—the University of Virginia experience. Cancer. 1994;73:2556–2562. Levine PA, Gallagher R, Cantrell RW. Esthesioneuroblastoma: Reflections of a 21-year experience. Laryngoscope. 1999;109:1539– 1543. Nikapota A, Sevitt T, Lund VJ, et al. Outcomes of radical conformal radiotherapy and concomitant cisplatin chemotherapy for olfactory neuroblastoma–a review of a single centre experience. Abstr Am Soc Clin Oncol. 2006. Shah J, Feghali J. Esthesioneuroblastoma. Am J Surg. 1981;142:456–458. Appelblatt N, McClatchey K. Olfactory neuroblastoma: A retrospective clinicopathological study. Head Neck Surg. 1982;5:108– 113. Bailey B, Barton S. Olfactory neuroblastoma. Arch Otolaryngol. 1975;101:1–5. Eriksen J, Bastholt L, Krogdahl A, et al. Esthesioneuroblastoma. What is the optimal treatment? Acta Oncol. 2000;39:231–235. Resto V, Eisele D, Forastiere A, et al. Esthesioneuroblastoma: The Johns Hopkins experience. Head Neck. 2000;55:550–558. Patel SG, Singh B, Polluri A, et al. Craniofacial surgery for malignant skull base tumors. Cancer. 2003;98:1179–1187. Constantinidis J, Steinhart H, Koch M, et al. Olfactory neuroblastoma: The University of Erlangen-Nuremberg experience 1975– 2000. Otolaryngol Head Neck Surg. 2004;130:567–574. Diaz E, Johnigan R, Pero C, et al. Olfactory neuroblastoma. The 22 year experience at one comprehensive cancer center. Head Neck. 2005;27:138–149. Ganly I, Patel SG, Singh B, et al. Complications of craniofacial resection for malignant tumours of the skull base: Report of an international collaborative study. Head Neck. 2005;27;445–451.

32 Melanoma of the Nasal Cavity and Paranasal Sinuses Ziv Gil, Mark H. Bilsky, and Dennis H. Kraus

that among 84,836 cases of malignant melanoma only 1074 cases involved the mucosal membrane, half of them arising in the head and neck (1). According to a report by the Armed Forces Institute of Pathology, sinonasal mucosal melanomas account for only 1% of all melanomas and for 0.6% to 4% of all tumors of the nasal cavity and paranasal sinuses (9). Conley reported in his series that melanoma involved up to 6.7% of all sinonasal malignancies (10). Ganly et al. have studied 1307 patients with malignant skull base tumors and found that 4% of them had sinonasal melanoma (11). Lewis and Martin have studied the incidence of malignant melanoma of the nasal cavity in Ugandan Africans and found that this disease accounted for 2.6% of all cases of melanoma (12). In Denmark, melanoma of the upper aerodigestive tract accounts for 0.8% of all melanoma cases and for 8% of all head and neck melanomas (13). Japanese have the highest rate of mucosal melanomas compared to Caucasians (4:1 relative rate), most commonly in the oral cavity (14). Interestingly, the incidence of cutaneous melanoma among Japanese is lower than Caucasians. The mean age of presentation of sinonasal melanoma is 65 with a range of 50 to 80 years (15–17). The gender distribution of sinonasal melanomas shows a slight male predominance (18–22). Ganly et al. have reported that of a total of 53 patients with anterior skull base melanoma, 70% were males and 30% females (11). Manolidis and Donald reviewed 172 cases of nasal melanomas and reported that 57% of the cases were males and 43% females (23). Female patients with cutaneous melanoma tend to have a better prognosis than men. However, similar outcomes occurred for males and females with sinonasal melanoma in regard to long-term survival (24–26). The main factor involved in the development of cutaneous melanoma is exposure to sunlight and ultraviolate radiation, whereas the etiology of the ultraviolate lightprotected mucosal melanoma remains unknown. Holmstrom and Lund have suggested that prolonged occupational exposure to formaldehyde may cause significant mucosal irritation eventually causing paranasal malignant melanoma (27). Similarly, Thompson et al. reported formaldehyde exposure in 9 of 115 patients (7.8%) in whom a work history could be elicited (9). The majority of these patients were painters, furniture makers, construction workers, and laundry workers. A joint Danish–Finnish–Swedish case-referent investigation initiated in 1977 studied the connection between nasal and sinonasal cancer and various occupational exposures. The authors found that formaldehyde was evenly distributed among cases with different tumors of paranasal origin (28). The nasal cavity is the most common origin of head and neck mucosal melanomas (55–79%), followed by the oral cavity (29–32). The lateral wall of the nasal cavity and the inferior turbinate are the most common origin of melanoma

Melanoma is a malignancy of ectodermal origin that involves the skin in the vast majority of the cases. The disease is classically divided according to the site of origin of the primary tumor, i.e., cutaneous, noncutaneous, and unknown primary melanoma. Noncutaneous melanomas are infrequent and may be found in the retina, genitourinary tract, anus, and upper aerodigestive tract. The most frequent origin of noncutaneous melanoma is the eye (5.3%) followed by melanomas of an unknown origin (2.2%) and mucosal melanomas (1.3%) (1). Mucosal melanomas of head and neck origin can arise in the oral cavity, oropharynx, nasal cavity, and paranasal sinuses. Melanoma of the sinonasal cavities is a rare neoplasm that can involve various compartments of the respiratory mucosa. The first description of mucosal melanoma was by Weber in 1856 (2). Thirteen years later, Lucke reported the first resection of a nasal mucosal melanoma in a 52-year-old man with “melanotic sarcoma” (3). One of the first reports of sinonasal melanoma is attributed to Viennois, who described the surgical extirpation of “polype melanique du nez” infiltrating the globe (4). In 1885, Lincoln made the first report in the English literature of a “melanosarcoma” arising in the nasal antrum and treated with galvanocauterization (5). Since then more than 1200 cases of mucosal melanoma have been reported in the English literature, one-third of them originating in the sinonasal area. Significant progress has been made during the last decade in the understanding of the biology of melanomas, and its sensitivity to chemotherapy, radiotherapy, and immunotherapy (6). However, due to the low prevalence of sinonasal melanoma, the pathophysiology of the disease as well as the role of radiotherapy and immunotherapy in its treatment still remains controversial. In this chapter, we review the current literature on paranasal and nasal mucosal melanomas and present the results of the latest studies on the epidemiology, pathology, staging, and treatment of patients with this disease.

INCIDENCE AND EPIDEMIOLOGY The incidence of melanoma has been increasing 4% to 6% each year since 1973, a greater rate than any other human cancer in the United States (7). It was estimated that nearly one in 75 persons will develop melanoma during their lifetime (8). The main factor believed to be involved in the significant rise in the incidence of cutaneous melanoma is exposure to sunlight and ultraviolate radiation. Unlike melanoma of the skin, mucosal melanoma did not show any increase in incidence during the last decade, suggesting that a distinct pathophysiologic mechanism is associated with this tumor. An analysis of the National Cancer Data Base performed throughout a period of 19 years (1985–1994) showed 459

460

Gil et al. Table 1

Site of Origin of Sinonasal Melanomas

Site Nasal cavity Nasal NOS Septum Lateral wall Inferior turbinate Middle turbinate Turbinate NOS Floor of nose Subtotal Paranasal sinuses Maxillary sinus Ethmoid sinus Frontal sinus Sinus NOS Subtotal Total a

Manolidis and Donald (23) Number (%) 159 (48) 44 (13) 44 (13) 12 (3.7) 7 (2.1) 12 (3.7) 5 (1.5) 283 (88) 32 (9.8) 8 (2.4) 2 (0.6) 21 (6.4) 63 (18.2) 328 (100)

Nandapalan et al. (34) Number (%)

61 (37) 10 (6.7) 34 (6.1) 43 (26) 2 (1.2) 150 (92) 5 (3) 8 (4.6)

13 (7.9) 163 (100)

a

Thompson et al. (9) Number (%) 34 (44) 20 (26) 10 (13)

64 (84) 3 (4)

9 (12.5) 12 (16) 72 (100)

In other 39 patients the specific compartment was not specified. Abbreviation: NOS, not otherwise specific.

of the nasal mucosa. In the sinuses, the exact site of origin of melanoma is difficult to identify since most tumors are diagnosed at advanced stage and frequently infiltrate multiple compartments. The most common site of origin for melanoma of the sinuses is the maxillary sinus, followed by the ethmoid and sphenoid sinuses (23,26). Melanomas of the frontal sinus, as well as the middle turbinate, superior turbinate, and cribriform plate are very rare (33). Table 1 shows the sites of origin of sinonasal melanomas among 563 patients reviewed in three different studies (9,23,34). Advance stage tumors most frequently involve multiple compartments. For example, in a recent multicenter study, Ganly et al. have reported that 50% of the patients with skull base melanomas had tumors infiltrating the cribriform plate, 34% had periorbital invasion, 26% orbital invasion, and 17% dural invasion (11). The majority of patients with sinonasal melanomas present without regional or distant metastases, at the time of diagnosis. Positive lymph nodes are found in 4% to 18% of the patients at the time of diagnosis (9,19,20,22). The incidence of lymph node involvement in patients with mucosal melanoma of the oral cavity and oropharynx is 4.7 times higher than in patients with sinonasal melanoma, due to the dense lymphatic drainage of the oral cavity and oropharyngeal mucosa (29). Distant metastases of sinonasal melanoma are rare at presentation. For example, Harrison et al. reported that none of the 40 patients in their series had distant metastases at presentation (19). In those patients who develop distant metastases, the most common sites are the lungs, bone, liver, brain, and skin (29).

PATHOLOGY The primary cell of origin of melanoma is the melanocyte, which can be found in the nasal and paranasal mucosa. These melanocytes have migrated as neuroectodermal derivatives and embedded in the endodermally derived respiratory mucosa (35). It is estimated that the distribution of melanocytes in the respiratory mucosa is 1500 cells/mm2 , which is less than two-thirds of that found in the skin (36). Mucosal melanosis is defined as a benign pigmented lesion characterized by pigmentation of basal keratinocytes

with normal or slightly increased number of melanocytes (37). Association between mucosal melanosis and increased incidents of nasal or paranasal melanoma was previously suggested by several authors (12,38,39). A preexisting pigmentation of the sinonasal mucosa is seen in less than 10% of the patients with mucosal malignant melanoma (40). In accordance with this finding, Thompson et al. have found preexisting melanosis in only 8% of the patients with sinonasal melanoma (9). The low rate of melanosis in patients with mucosal melanoma indicates that most sinonasal melanomas arise de novo. The majority of melanomas of the respiratory mucosa are large and polypoid with a median thickness of 9 mm, significantly thicker than melanoma of the oral cavity (41). Similarly, Thompson et al. reported a mean thickness of 7.2 mm (range 2–19 mm) and size of 24 mm (range 5–65 mm) (9). Infrequently, mucosal melanoma in situ can be identified after the biopsy of a suspected lesion (Fig. 1). The authors found no correlation between survival and tumor thickness. Lee et al. have found that depth of invasion > 7 mm is an independent factor for a poor prognosis in patients with mucosal melanomas of the head and neck (42). Melanoma has a notorious tendency to mimic other tumors, and in the sinonasal mucosa it can be easily confused with other tumors, which occur relatively more commonly in this region. The challenge in making the diagnosis of sinonasal melanoma is more significant in case of amelanotic melanoma and in the presence of ulceration. A study by the Armed Forces Institute of Pathology has previously shown that more than two-thirds of the cases of sinonasal melanoma are initially misclassified as other neoplasm on initial pathologic evaluation (9). Histologically, mucosal melanoma cells may have different characteristics including small cells, spindle cells, epithelioid cells, or rarely pleomorphic cells (21,43). Spindle cell melanoma appears sarcomatous, and is composed of cells with eosinophilic cytoplasm and nuclei that may vary in shape and number. Epithelioid-type melanoma is characterized by large cells with eosinophilic cytoplasm and acentric nucleus. Although melanomas display morphologic diversity, undifferentiated small round cells or polygonal cells are the most prominent cells found in sinonasal melanomas

Chapter 32: Melanoma of the Nasal Cavity and Paranasal Sinuses

461

(A)

Figure 1 Mucosal melanoma in situ. The photomicrograph shows mucosal melanoma confined to the sinonasal respiratory epithelium. The neoplastic melanocytes are three to four times larger than the benign surrounding cells. Convoluted nuclear membranes, as well as prominent nucleoli with increased nuclear to cytoplasmic ratio, can also be identified. 100× H&E staining.

(41,44). A pseudopapillary growth pattern is found in up to 25% of patients with sinonasal melanomas, but not in melanoma of the oral cavity (Fig. 2) (21,41). The most significant factor in establishing the diagnosis of melanoma is melanin production and the appearance of junctional changes (Fig. 3). Melanin pigment is found in two-thirds of the cases of sinonasal melanoma and should be considered when confronted by a sinonasal myxoid tumor with melanin (9). Such an example is melanoma botryoides, a polypoid tumor that contains small amounts of melanin with a botryoid or myxoid pattern. Amelanotic tumors show similar biologic behavior and prognosis as melanotic nasal melanomas, but are far more difficult to diagnose compared with conventional melanomas (45). A variant of melanoma, which does not contain melanin, is desmoplastic melanoma (46). Rarely found in sinonasal melanoma, these cells are comprised of amelanotic, poorly circumscribed fascicles and bundles of spindle cells with hyperchromatic nuclei (47). Desmoplastic melanoma is a neurotrophic tumor, which frequently expresses aberrant p53 protein (46). This variant of amelanotic malignant melanoma is difficult to differentiate from other soft tissue tumors of the nasal cavity such as esthesioneuroblastoma, sarcoma, spindle cell carcinoma, and malignant peripheral nerve sheath tumors (48). Melanoma of the sinonasal mucosa often demonstrates deep invasion, necrosis, and vascular invasion. These characteristics are well-established predictors of reduced survival in cutaneous melanomas and melanomas of the head and neck (41). Prasad et al. have shown that 60% of paranasal melanomas are detected in an advanced stage, presenting with significant infiltration of skeletal muscles, cartilage, and bone at the time of surgery (41). Rarely nasal mucosal melanomas may show bone and osteoid formation (9,49). Osteoid and metaplastic bone formation may be caused by repeated trauma, mesenchymal metaplasia, and secondary reaction to bone invasion and introduction of bone formation from the surrounding tissues (9).

(B)

(C)

Figure 2 Nonmelanotic nasopharyngeal mucosal melanoma. (A) Small round blue cells grow radially outward from central vessels demonstrating the pseudopapillary architecture one may frequently see in mucosal melanoma. (B) A photomicrograph of the specimen showing malignant cells arranged in a pavement-like sheet, with no evidence of pigmentation. Large cherry red macronucleoli in the center of the cell nucleus are very characteristic, but not pathognomonic for melanoma. 600× H&E staining. (C) Immunohistochemical staining for S-100, in both the nucleus and cytoplasm are indicative of malignant melanoma.

462

Gil et al.

or paranasal origin. Thompson et al. studied 115 cases of sinonasal melanoma and found positive staining for S-100 in 91% of the cases and HMB-45 in 76% of the cases (9). Tyrosinase antigens were expressed in 77.7% of the cases. Table 2 shows the different antigens expressed in various tumors of the paranasal sinuses including malignant melanoma. Electron microscopy can be utilized as an adjunct for the diagnosis of melanoma in borderline cases and is very specific in identifying premelanosomes, a subcellular organelle present in melanomas. Wright and Heenan identified a subclass of premelanosomes with a high propensity to metastasize (56), however, these premelanosomes are not found in all cases of mucosal melanoma (19,57). Since melanoma of the paranasal sinuses accounts for less than 7% of neoplasms involving this anatomic compartment, a proper different diagnosis workup is crucial for determination of the management of the disease. The microscopic differential diagnosis of sinonasal melanoma can be divided into three groups: small round blue cell tumors, pleomorphic cell tumors, and spindle cell tumors (9). The first group includes tumors such as olfactory neuroblastoma, primitive neuroectodermal tumor, Ewing sarcoma, melanocytic neuroectodermal tumor of infancy, pituitary adenoma, lymphoma, plasmacytoma, small cell neuroendocrine carcinoma, and mesenchymal chondrosarcoma. The pleomorphic cells group includes sinonasal undifferentiated carcinoma, anaplastic large cell lymphoma, and rhabdomyosarcoma. The group of tumors characterized by spindle cells includes malignant peripheral nerve sheath tumor, fibrosarcoma, malignant fibrous histiocytoma, and synovial sarcoma (9,44,58–62). Olfactory esthesioneuroblastoma can show similar morphology to melanoma; however, this tumor frequently shows neurofibrillary background, Homer–Wright rosettes, and focal sustentacular cell pattern even in high-grade tumors (Hyman grade II–IV). Furthermore, most esthesioneuroblastoma shows immunocytochemical reactions to neuronspecific enolase and chromogranin and negative staining to vimentin. As described earlier, mucosal melanoma frequently lack melanin pigment and therefore can be indistinguishable from other high-grade tumors such as sinonasal undifferentiated carcinoma, undifferentiated nasopharyngeal carcinoma, poorly differentiated nonkeratinizing squamous cell carcinoma, small cell carcinoma, and anaplastic large cell lymphoma (63). Fortunately, most epithelial tumors show strong cytokeratin immunostaining and fail to express S-100 protein. Anaplastic large cell lymphoma does not stain for cytokeratin

Figure 3 Melanin pigmented cells in malignant melanoma of the sinonasal mucosa. The most important factor in establishing the diagnosis of melanoma is the appearance of malignant cells containing melanin. Unlike carcinoma, these cells have visible apparent spaces between their cytoplasmic boarders.

Due to the complicated differential diagnosis of this tumor, immunocytochemical staining is frequently required to establish the diagnosis of melanoma of the paranasal sinuses, particularly in cases of amelanotic variants. Melanoma stains positive for vimentin, HMB-45, and S-100 protein. In contrast to cutaneous melanomas, paranasal melanoma is negatively stained to synaptophysin and actin leukocytic common antigen (50,51). Interestingly, staining for P-97, which is frequently positive in melanoma of the esophageal mucosa, is not found in melanomas originating in the paranasal sinuses or nasal mucosa (52,53). Other antigens, which are specifically stained in paranasal melanomas, are KC-2, SK46, and KBA-62 (54). Other melanoma-associated antigens found in the nasal cavity are tyrosinase (T311), D5, A103 (anti-melan-A/MART-1), and TRP-1 (39). In a study of 44 sinonasal melanomas, Prasad et al. have found that all tumors were positive for tyrosinase, 98% for HMB-45, 95% for S-100 protein, and 91% for D5 (55). They concluded that tyrosinase is the most sensitive marker for melanomas of nasal

Table 2

SCC SNUC ONB SCNUC MMM T/NK ML RMS PNET/EWS ∗

Histologic and Antigenic Characteristics of Malignant Melanoma and Other Undifferentiated Tumors Originating in the Nasal and Paranasal Sinuses CK

NSE

CG

SYN

S100

HMB

LCA

CD56

CD99

VIM

DES

Myf4

+ + − + − − − R+

− V + + − − − V

− − V + − − − −

− − V + − − − V

− − +∗ + + − − V

− − − − + − − −

− − − − − V − −

− − − − − + − −

− − − − − − − +

− − − − + V + +

− − − − − − + −

− − − − − − + −

Positive in the peripherally situated sustentacular cells. Abbreviations: SCC, squamous cell carcinoma; SNUC, sinonasal undifferentiated carcinoma; SCUNC, small-cell undifferentiated neuroendocrine carcinoma; MMM, mucosal malignant melanoma; T/NK ML, nasal-type T/natural killer-cell lymphoma; RMS, rhabdomyosarcoma; PNET/EWS, primitive (peripheral) neuroectodermal tumor/extraosseous Ewing sarcoma; CK, cytokeratin; NSE, neuron-specific enolase; CG, chromogranin; SYN, synaptophysin; HMB, HMB 45 (as well as other melanocytic markers [melan A]); CD99, Ewing marker; VIM, vimentin; DES, desmin; (+), positive; (−) negative; v, variably positive; R+, rarely positive. Source: From Ref. 63.


463

but express CD30 and anaplastic large cell lymphoma kinase proteins. Malignant peripheral nerve sheath tumors may also express S-100 protein, complicating the pathologic diagnosis of melanoma. Malignant peripheral nerve sheath tumor does not express other antigens, such as HMB-45, which are frequently found in sinonasal melanoma.

CLINICAL PRESENTATION AND FINDINGS The clinical symptoms in patients with paranasal sinuses melanoma are frequently nonspecific and include pain, malaise, and weight loss. Other symptoms may be directly associated with the location of the primary tumor. Melanomas originating in the nasal septum may cause irritation and be visible to the patients (64). This may explain the observation that 75% of the patients with nasal mucosal melanoma are diagnosed early with a clinically localized disease, in comparison to those with paranasal sinuses melanoma. Most patients with paranasal melanoma suffer from nasal discharge, recurrent epistaxis, or nasal obstruction (85–90% of cases). Thompson et al. reported that 52 of the 115 patients with melanoma of the sinuses had epistaxis, 42 had a visible mass, and 32 had obstructive symptoms (9). Pain (20%) and a visible facial mass (9%) occur in more advanced disease stage (20). Signs representing orbital involvement are proptosis, ophthalmoplegia, decreased visual fields, and monocular blindness and are associated with poor prognosis (11,42). Most skin or oral cavity melanomas are more likely to be discovered by the patient or by the primary health care physician upon routine examination, whereas sinonasal melanomas are inaccessible to self-examination and are routinely diagnosed at an advanced stage. The duration of the symptoms depends on the biologic behavior of the disease. In case of a slowly growing tumor of paranasal origin, airway obstruction may develop slowly and the disease, which is obscured from the patient and the physician, may develop months or years before the diagnosis is established (19). Most authors report mean symptoms duration of 2 to 8 months (13,20,31,42,65). Holdcraft and Gallagher have reported that 50% of their patients had suffered from symptoms for 1 to 4 months before diagnosis, and 30% had symptoms for 6 to 24 months (18). Thompson et al. reported a mean duration of 8.2 months ranging between 2 weeks and 8 years (9). Evaluation of patients with suspected sinonasal melanoma should include complete history and physical evaluation with emphasis on the head and neck. Fiberoptic evaluation of the paranasal sinuses and upper aerodigestive tracts is indicated in all patients, in order to evaluate the tumor extent and potential for resection (Fig. 4). Radiologic evaluation should always include both computed tomography (CT) and magnetic resonance imaging (MRI) of the head, neck, and paranasal sinuses for evaluation of bony and soft tissue involvement, respectively (66). Patients should be evaluated for involvement of cranial nerve, orbit, skull base, dural or brain infiltration using both CT and MRI. Although perineural spread of disease occurs more commonly with squamous cell carcinoma and adenoid cystic carcinoma of the paranasal sinuses, malignant melanoma must also be included in this differential diagnosis, particularly if the patient’s pathology is known to be desmoplastic melanoma. Imaging in patient with malignant melanoma of the paranasal sinuses should focus on the likely potential for perineural spread. In a recent study of nine patients with melanomas of the facial skin and paranasal sinuses (including five desmoplastic melanomas) with symptomatic cranial

Figure 4 Endoscopic photograph of a mucosal melanoma of the paranasal sinuses. Fiberoptic examination of the paranasal sinuses and upper aerodigestive tract is indicated in all patients prior to surgery, in order to evaluate the tumor extent and the presence of a second primary. The photograph shows mucosal melanoma arising from the sinonasal mucosa (arrow).

neuropathy, MR imaging demonstrated post gadolinium enhancement of the trigeminal nerve in all nine cases and of other cranial nerves in five cases (67). Other findings included abnormal contrast enhancement and soft tissue thickening in the cavernous sinus, Meckel cave, and/or the cisternal segment of the trigeminal nerve. Suspicious neck metastases can be evaluated with CT, MRI, or ultrasound. Ultrasound-guided fine-needle biopsy can be utilized if indicated. Sentinel lymph node biopsy, which is frequently used for lymphatic mapping of cutaneous melanomas, is not a common practice for sinonasal disease (68). Chest radiograph should be performed for evaluation of lung metastases. Chest CT, liver ultrasound, and bone scan should be performed if metastatic disease is suspected. Patients with mucosal melanomas can be evaluated for the presence of metastases using positron emission tomography (PET). Goerres et al. have found that all mucosal melanomas of the head and neck can be visualized using FDG PET (66). Large lesions with a nodular growth are better demonstrated than lesions with a superficial mucosal spread. Similarly, lesions originating in the nasal vestibule are more challenging to identify than those in the posterior sinonasal complex, due to nonspecific uptake in the skin and muscles of the mouth. Due to the high yield in staging metastatic disease, utilizing metabolic PET imaging can replace staging techniques employing multiple imaging modalities (i.e., chest x-ray, neck and liver ultrasound, total body CT, and bone mapping) (69). Metastatic cutaneous melanoma to the paranasal sinuses is very rare and account for 1% of the patients with cutaneous melanoma (44,70). Nevertheless, despite its rarity, full workup should be performed to exclude isolated metastasis of melanoma of the skin to the paranasal sinuses.

STAGING There is no formal staging system for mucosal or sinonasal melanoma. The TNM classification of melanomas is only

464

Gil et al.

available for skin and ocular lesions. The staging of cutaneous melanoma is based on the staging system revised by the American Joint Committee on Cancer (AJCC) (71). In this system, the prognosis of patients with localized disease (T1: tumors < 1.0 mm or T2: 1–2 mm in thickness) is good, whereas for patients with melanomas >2.0 mm in thickness, a worse survival rate is expected (T3, T4). Patients with localized disease and no regional or distant metastases are classified as having stage I or II disease. In patients with regional metastases (stage III), tumor burden is expressed as the number of positive nodes (N1 for a single node, N2 for 2–3 nodes and N3 if ≥4 lymph nodes are involved). In patients with distant metastatic disease (stage IV), the sites of metastases determine outcome and M classification is graded from a–c accordingly (i.e., skin, lung, and visceral, respectively). As with cutaneous melanoma, the outcome of mucosal melanoma initially depends on the stage at presentation (42). Unfortunately, because of an absence of histologic landmarks identifiable as a papillary and reticular dermis in the respiratory mucosa, the AJCC cutaneous classification system cannot be applied for this disease. Furthermore, sinonasal melanomas are frequently polypoid rather than being deeply invasive and tumor thickness cannot accurately predict the prognosis (9). An alternative system for classification of sinonasal melanoma is the AJCC staging criteria for nasal and paranasal epithelial tumors (72). As described in detail in chapter 29 “Squamous Cell Carcinoma of the Nasal Cavity and Paranasal Sinuses,” this staging system considers the extent of the primary lesion (stage I–IV) and the presence of regional metastases (stage III–IV) or distant metastases (stage IV) as main prognostic indicators for survival. In the absence of a conventional classification for sinonasal melanomas, few authors have adopted the AJCC staging system for nasal and paranasal epithelial tumors for sinonasal mucosal melanoma (20,29,73). One must take into consideration that the AJCC TNM classification applies only to histologically confirmed carcinomas, and it was not validated for melanoma. The most common system used for staging of mucosal melanoma of the head and neck was first suggested by Ballantyne (74). In this classification, a stage I disease represents tumors confined to the primary site, stage II denotes the existence of positive regional lymph nodes, and stage III indicates a distant metastatic disease. This system does not take into consideration the size and extent of the disease process, but has been used repeatedly for classification of sinonasal melanomas and melanomas of the upper aerodigestive tract (26,30,75). Since regional and metastatic disease is not common during initial diagnosis, most patients are characterized in the stage I group. Clearly, the main drawback of this staging system is its lack of ability to differentiate between patients with localized disease, and those with advanced tumors and poor prognosis (i.e., orbital infiltration or intracranial extension). Thompson et al. have studied a group of 115 patients with sinonasal melanomas in an attempt to develop a validated staging system by incorporating features of size, site, regional and distant metastases into a single staging system (9). The T classification of this staging system separates between tumors localized to a single anatomic compartment (T1) and those involving more than one anatomic level (T2). The N classification accounts for the absence (N0) or presence (N1) of lymph node metastases (whether ipsilateral, bilateral or contralateral). Patients with T1 and T2 disease in the absence of regional or distant metastases are grouped into disease stage I and II, respectively. A stage III disease accounts for patients with any T, N1, and M0, whereas pa-

Table 3 Proposed Staging for Sinonasal Tract and Nasopharynx Mucosal Malignant Melanoma Nasal cavity, paranasal sinuses, and nasopharynx histopathology staging Primary tumor T1 T2 Regional lymph node N1 Distant metastasis M1 Stage grouping Stage I Stage II Stage III Stage IV

Single anatomic site Two or more anatomic sites Any lymph node metastasis Distant metastasis T1, N0 M0 T2, N0 M0 Any T, any N, M1 Any T, any N, M1

T, primary tumor. TX, primary tumor cannot be assessed. T0, no evidence of primary tumor. T1, tumor limited to a single anatomic site. A single anatomic site is defined as one of the following: nasal cavity, maxillary sinus, frontal sinus, ethmoid sinus, sphenoid sinus, nasopharynx. Subsites, such as septum, lateral wall, turbinate, nasal floor, or nasal vestibule are not separately considered. T2, tumor involving more than one anatomic site. More than one anatomic site is defined by tumor involvement of more than one anatomic site (although not subsite) as cited above, including any extension into subcutaneous tissues, skin, palate, pterygoid plate, floor, wall, or apex of the orbit, cribriform plate, infratemporal fossa, dura, brain, middle cranial fossa, cranial nerves, clivus. N, regional lymph nodes (cervical lymph nodes). NX, regional lymph nodes cannot be assessed. N0, no regional lymph node metastasis. N1, metastasis in regional lymph node(s) of any size, whether ipsilateral, bilateral, or contralateral (midline nodes are considered ipsilateral nodes). M, distant metastasis. MX, distant metastasis cannot be assessed. M0, no distant metastasis. M1, distant metastasis. pTNM, pathological classification. The pT, pN, and pM categories correspond to the T, N, and M categories. From a practical standpoint, documentation of metastatic disease (lymph node or distant) is based on findings within 90 days peri-diagnosis (ie, a lymph node is the initial presentation and a mucosal primary is documented within 3 months; a STMMM is diagnosed and then CT, MR or other studies are performed over the ensuing 6 weeks and identify metastatic disease). pN0, histologic examination of a selective neck dissection specimen will ordinarily include 6 or more lymph nodes. Histologic examination of a radical or modified radical neck dissection specimen will ordinarily include 10 or more lymph nodes. If the lymph nodes are negative, but the number ordinarily examined is not met, classify as pN0. Source: Adapted from Ref. 9.

tients with distant metastases are classified as having a stage IV disease. The TNM classification suggested by Thompson et al. predicted patients’ outcome based on the anatomic site of involvement and metastatic disease (Table 3). Prasad et al. at Memorial Sloan–Kettering Cancer Center suggested further microscopic classification of stage I (lymph node-negative) sinonasal melanoma (76). Their microstaging system was performed according to disease invasion into three compartments: level I, melanoma in situ; level II, invasion into the lamina propria; and level III, invasion into deep tissue (i.e., skeletal muscles, bone, or cartilage). Kaplan– Meier analysis showed significant difference in 5-year disease


specific survival of patients with level I (75%), level II (52%), and level III (23%).

TREATMENT There is a general consensus that surgery remains the treatment of choice for mucosal melanoma of sinonasal origin (75). A recent report of the National Cancer Institute and the Center for Disease Control has demonstrated an absolute gain in overall survival of melanoma patients during the last decade (77). The improvement in survival of cutaneous melanoma patients, despite the increase in incidence, is attributed to early detection and improvements in therapy. The mode of therapy for sinonasal melanoma awaits further evaluation due to lack of prospective studies and objective data for the benefit of adjuvant modalities such as postoperative radiotherapy, immunotherapy, and chemotherapy.

Surgery In the absence of distant metastases, complete tumor extirpation is the mainstay of treatment for malignant melanoma of the paranasal sinuses and nasal mucosa. Although negative margins after surgical resection is reported in over 85% of patients, it is frequently impossible to achieve en bloc tumor resection, and 75% of the patients will eventually develop local recurrence (11). Two possible explanations for the high recurrence rate are (i) presence of a multifocal disease and (ii) submucosal lymphatic spread of melanoma cells in the respiratory mucosa, which is not clinically or radiographically apparent (26). Freedman et al. suggested multicentricity as a main factor predicting local recurrence after surgery (20). Thus wide surgical resection, without unnecessary compromise of function and cosmesis is essential.

(A)

(B)

465

The route of spread of tumors originating in the anterior skull base and paranasal sinuses is determined by the complex anatomy of the craniofacial compartments. A tumor arising in the ethmoid sinus or paranasal cavity may invade laterally to the orbit, inferiorly to the maxillary antrum and palate, posteriorly to the nasopharynx and pterygopalatine fossa (PPF), and superiorly to the dura, brain, or cavernous sinus. The recent improvement of endoscopic technology now allows for the resection of benign neoplasms or early malignant neoplasms (78). However, for sinonasal melanoma, an open approach is more suitable to allow extirpation of tumors in an en bloc fashion and with wide margins, in the opinion of the authors. The type of surgery is planned according to the extent of the tumor. For small tumors involving the nasal septum, resection of the tumor along with the perichondrium and septal cartilage may be performed via lateral rhinotomy incision (Fig. 5). However, frequently these tumors also infiltrate adjacent structures such as the hard palate, ethmoid sinuses, and medial maxillary walls. In these cases a unilateral or bilateral medial maxillectomy is performed via a lateral rhinotomy incision. Conventional exposure of the infra- and suprastructure of the maxilla is achieved via a lateral rhinotomy with lip split or subciliary extension as indicated (Fig. 6). Tumors infiltrating the cribriform plate and fovea ethmoidalis are safely accessed via the craniofacial or subcranial approach (11). These approaches offer wide exposure of the tumor, allowing resection of the intracranial and extracranial extensions of the tumor. Massive orbital involvement or orbital apex infiltration requires orbital exenteration. In this case, orbital exenteration is performed with radical maxillectomy or craniofacial resection, as determined by the tumors extension. Combinations of the craniofacial approach with transorbital and middle fossa

(C)

Figure 5 Nonmelanotic melanoma of the nasal septum. (A) Endoscopic photograph of the lesion shows a nonmelanotic nasal septal lesion (asterisk, nasal septum; arrow, lateral nasal wall). (B) A preoperative coronal CT scan showing a left nasal septal lesion. (C) An intraoperative picture demonstrating exposure of the lesion. The type of surgery is planned according to the extent of the tumor. For small tumors involving the nasal septum, resection of the tumor along with the perichondrium and septal cartilage can be performed via lateral rhinotomy incision.

466

(A)

Gil et al.

(B)

(C)

Figure 6 Mucosal melanoma of the left maxillary sinus. (A) A preoperative coronal CT scan showing a hyperdense mass in the right maxillary antrum. (B) An intraoperative picture demonstrating exposure of the lesion. En bloc resection of the tumor along with a radical maxillectomy was achieved via the transfacial approach (lateral rhinotomy with lip split). (C) The excised specimen showing the melanotic lesion extending into the maxillary antrum.

approaches were described in detail by Shah et al. for malignant tumors of the anterior skull base and paranasal sinuses (79). A combined facial translocation approach was also described and safely used by Hao et al. for malignant tumors of the paranasal cavity (80,81). Following resection of the primary tumor, the surgical margins should be evaluated for residual disease using multiple permanent sections at the periphery of resection, sampling bone, mucosa, soft tissue and other tissue as indicated. Although complete tumor resection should be the goal of care for sinonasal melanoma, a recent analysis of 53 patients with skull base melanoma has shown no survival benefit of negative margins in this anatomic area (11). Dural and anterior skull base reconstruction is required after craniofacial resection. Dural reconstruction is performed principally with pericranial, galeal, temporalis fascia, or fascia lata grafts. Bovine pericardium can also be utilized for reconstruction. Fibrin glue is used in order to provide additional protection against cerebrospinal fluid leak. Reconstruction of the medio-orbital walls is not typically performed. If indicated, a split calvarial bone graft, a fascia lata sling, or 3dimensional titanium mesh covered by pericranium are used for reconstruction of the orbit. A temporalis muscle flap and a split-thickness skin graft to cover the orbital socket can be used after orbital exenteration. In cases of a radical maxillectomy with or without orbital exenteration, a lateral thigh free flap or a rectus abdominis musculocutaneous free flap may be utilized to obliterate this large defect and to support the obturator. Since neck lymph nodes are rarely encountered in cases of sinonasal melanoma, neck dissection should only be performed if regional metastases are identified, based on clinical or radiologic evaluation. The postoperative complication rate following surgical resection of malignant skull base tumors is 36% (11).

Among patients treated with craniofacial resection for excision of sinonasal melanomas, the postoperative mortality is 6% and major postoperative complications occur in 26% of the patients.

Radiation The use of radiation therapy for treatment of melanoma is controversial. Despite the long-standing debate regarding the radiosensitivity of melanoma reported in the past, there has been a significant increase in the use of adjuvant radiation therapy for the treatment of this disease (1). Both clinical and basic science data indicate that melanoma cells have the ability to repair cellular damage, providing resistance to conventional fractionated radiation therapy (82). It was therefore speculated that hypofractionation or high-dose-perfractionation (HDPF) therapy will give more effective radiation treatment to these patients. Several nonrandomized, retrospective studies have reported improved locoregional control rates of patients with high-risk cutaneous melanoma of the head and neck using conventional or hypofractionation adjuvant radiation compared with surgery alone (83,84). Moreover, Raben et al. have reported 70% local control rate in 10 patients treated with the high-dose-per-fractionation regimen following surgical resection of head and neck malignant melanomas, with minimal morbidity (85). However, no change in overall survival was found in this study following hypofractionation adjuvant radiotherapy due to the development of disseminated disease. In patients with mucosal melanoma, Patel et al. reported no advantage of postoperative radiotherapy compared to surgery alone (26). In contrast, Ganly et al. reported that postoperative radiation therapy was an independent positive predictor of overall, disease-specific, and recurrence-free survival on a multivariate analysis of patients with skull base melanoma (11). Patients treated with surgery and postoperative radiotherapy

Chapter 32: Melanoma of the Nasal Cavity and Paranasal Sinuses Table 4

Melanoma of the Nasal and Paranasal Sinuses and Survival

Study

Year

N

5-year survival

Impact of radiotherapy on survival

Holdcraft & Gallagher (18) Freedman et al. (20) Eneroth & Lundberg (91) Harrison (19) Trapp et al. (16) ∗ Gilligan et al. (22) Kingdom and Kaplan (15) Harbo et al. (92) Brandwein et al. (58) Lund VJ et al. (77) Owens et al. (85) Thompson et al. (9) Patel et al. (26) Nakaya et al. (93). Bridger et al. (72) Ganly et al. (11)

1968 1973 1975 1976 1987 1991 1995 1997 1997 1999 2003 2003 2002 2004 2005 2006

39 56 24 40 17 28 17 25 36 72 11 115 35 16 27 53

10% 30% 17% 27.5% 25 18% 20% 24% 36% 28% 33% 42.6% 47% 31.8% 43% 24%

– NS – – – – Prolonged – – NS NS NS NS NS – >2-fold increase

∗

467

Primary radiotherapy. Abbreviation: NS, nonsignificant.

had 39% 3-year overall survival, compared with 18% in patients treated with surgery alone. Freedman et al. have reported no survival benefit for patients with paranasal and nasal cavity melanoma treated with surgery and adjuvant radiotherapy compared with those treated with surgery alone (20). In the same study, it was reported that none of the 18 patients treated with radiation alone survived after 5 years. Owens et al. recently reported a retrospective evaluation of patients with mucosal melanoma (23% of them with sinonasal disease) treated with surgery alone, surgery and adjuvant radiotherapy, or surgery and biochemotherapy, with or without adjuvant radiotherapy (86). Radiation therapy was generally used as an adjuvant therapy for patients with extensive disease. The patients with sinonasal tumors received 6000 cGy in 30 fractions, while those with oral lesions received 3000 cGy in 5 fractions. Biochemotherapy (cisplatin, vinblastine, and dacarbazine, with or without the addition of interferon alfa-2b and interleukin 2) was used almost exclusively in patients who had recurrent disease or distant metastases. The addition of radiotherapy tended to decrease the rate of local failure, but did not prevent distant metastases or improve overall survival. Biochemotherapy regimens used for metastatic or recurrent disease had no significant impact on survival. An evaluation of the impact of postoperative radiotherapy on local control and survival of patients with head and neck mucosal melanoma was reported by Temam et al. at the Institut Gustave–Roussy (Villejuif) (29). Sixty-nine patients with local disease were managed by surgery without postoperative radiotherapy, two-thirds of whom had sinonasal disease. The study suggested that postoperative radiotherapy increased local control in patients with small tumors, but did not impact survival. Most reports of definitive radiation therapy for mucosal melanomas involve small series of patients. Gilligan and Slevin reported one of the largest series of melanomas of the paranasal sinuses and nasal cavity treated with radiation alone (22). Complete response was achieved in 79% of the 28 patients included in the study, with relatively low treatment morbidity. In this study, the overall 3- and 5-year survival was 49% and 18%, respectively (Table 4). Stern and Guillamondegui at MD Anderson Cancer Center reported two of the

five patients alive and disease free, 5 years after radiotherapy alone (31). At the Princess Margaret Hospital in Toronto, Canada, Harwood and Cummings reported 50% local control rate at 6 months to 4.2 years after primary radiotherapy (n = 10 patients) (87). Trotti and Peters reviewed a series of reports using radiotherapy alone for mucosal head and neck melanoma and concluded that 50% to 75% of the patients had documented complete response, with long-term control in 50% to 66% (88). They concluded that in view of the poor results of radical surgery, radiation should be seriously considered as the initial treatment of choice for primary mucosal melanomas of the head and neck. Albertsson et al. reported the result of hyperfractionation radiation in combination with cisplatinum (89). Three of four patients treated for local recurrence achieved local control. Radiation may be also appropriate as a primary treatment for patients with unresectable disease, elderly patients, patients with high surgical risk or palliative therapy (90). External beam radiation may achieve local control, relieve pain, and decrease tumor mass compression on vital structures including cranial nerves, orbit, airway or brain. Radiation therapy has the potential for complications, especially if applied to the cranial base. Severe morbidity associated with radiation of the anterior skull base includes osteoradionecrosis, encephalomalacia, optic neuropathy and retinopathy (94). Radiation therapy has also been associated with a decreasing quality of life in patients with skull base malignancies (95). Using heavy particle radiation sources (i.e., proton or carbon ions), as well as accurate delivery using intensity-modulated radiation therapy, may be beneficial in enhancing therapeutic outcomes and reducing complication rates. Fast neutron therapy was used to treat primary, recurrent, or metastatic cutaneous and mucosal melanoma in 48 patients, showing complete regression in 71%, with a 9% incidence of local recurrence (96). The median survival was 14.5 months and complications, including fibrosis and necrosis, occurred in 22% of patients. In another series, Linstadt et al. reported local control in two of the six patients treated with neon ions for melanoma located in a variety of sites including paranasal sinus (97). Promising results were found in a dose escalation study using carbon ions, reporting 100% 5-year

Gil et al.

local control rate in five patients with mucosal melanoma (98).

Chemotherapy and Immunotherapy Chemotherapy is regarded today mainly as palliative treatment for recurrent, metastatic, or inoperable disease. A variety of chemotherapeutic agents, both as adjuvant or neoadjuvant therapy, have been used for treatment of distant disease based on treatment of patients with cutaneous melanomas. Cisplatinum, dacarbazine, and vinca alkaloids are some of the chemotherapeutic agents used for systemic treatment. Unfortunately, most authors reported no advantage for chemotherapy in patients with distant metastatic melanoma originating from a head and neck mucosal primary (13,25,30,99). An emergent therapeutic strategy for treatment of systemic disease is the use of immunotherapy or biologic response modifiers with or without chemotherapy. The most frequently used biologic response modifiers are bacillus Calmette–Guerin (BCG) vaccine, interleukin-2, and interferon alpha-2b (IFN alpha-2b). In 2002, the National Comprehensive Cancer Network committee had recommended for high-risk patients with localized cutaneous melanomas greater than 4.0 mm in thickness to participate in adjuvant therapy clinical trials, including treatment with high-dose adjuvant IFN alpha-2b versus observation (100). This recommendation was based on two studies performed by the Eastern Cooperative Oncology Group (trial 1684 and 1690); these were randomized controlled studies of IFN alpha-2b administered at maximum tolerated doses versus observation (101). Another study showed a relapse-free survival advantage for high-dose interferon with no difference in overall survival (102). Legha et al. evaluated the use of concurrent biochemotherapy including cisplatin, vinblastine, and dacarbazine in combination with IFN-alpha and interleukin-2 in patients with metastatic melanoma (103). Among the 53 patients treated in this study, 21% achieved a complete response and 43% achieved a partial response. The median survival was 11.8 months. The toxicity reported was severe myelosuppression, nausea, vomiting, and hypotension that required inpatient care and support. There were no treatmentrelated deaths. An anecdotal report that evaluated the utility of hormonal therapy with tamoxifen for palliative treatment of patients with sinonasal melanoma reported good response in all three patients treated with the drug (91). Since there are no trials for adjuvant immunotherapy treatment of sinonasal mucosal melanoma, the decisions regarding the use of adjuvant therapy for these patients should be made on an individual basis, extrapolating from available data from the cutaneous adjuvant trials, and after discussion with the patient, including an explanation of the adjuvant treatment clinical results and anticipated morbidity.

OUTCOME AND PROGNOSIS Of all patients with mucosal melanomas of the head and neck, those with disease involving the paranasal sinuses have the poorest outcome. Notably, tumors isolated to the nasal cavity are associated with the best prognosis of head and neck mucosal melanoma. Patients with sinonasal melanomas have an interposed course of disease with multiple local or regional recurrences, followed by distant metastases and ultimately death from disseminated disease. Head and neck mucosal melanomas have the lowest 5-year survival rate compared to cutaneous and ocular melanomas (32%, 75%, and 80%, respectively) (1). The estimated 10-year survival rate of these

patients is 7% (104). In spite of the development of new surgical approaches and novel therapeutic agents, there was no significant increase in the overall survival of patients with sinonasal and nasal cavity melanomas. A large cohort study performed by the Armed Forces Institute of Pathology reported no influence of treatment modality on overall survival (9). Among the 115 patients participated in this study, there was no statistical significant difference between patients managed by surgery alone, surgery with chemotherapy, surgery with radiotherapy or surgery with combined therapy. After a mean follow-up of almost 14 years, 35% of the patients were alive or have died without evidence of disease. Owens et al. reported 14.3 months’ average interval to failure in patients with sinonasal tumors (range, 3–36 months), compared to 31.6 months in patients with oral or oropharyngeal lesions (range, 3–147 months) (86). For patients with sinonasal disease, the overall survival rates at 3 and 5 years were 50% and 33%, respectively. A recent report of a large cohort of patients with sinonasal melanoma undergoing craniofacial resection accumulated from multiple international institutions reported a 3-year disease-free and overall survival of 28% and 29%, respectively (Fig. 7) (11). Orbital involvement was found to be an independent predictor of overall and disease specific survival. For comparison, the overall 5-year survival of patients with all malignant skull base tumors treated by the same group of surgeons was 54%, including esthesioneuroblastoma (64%) and squamous cell carcinoma (50%) (105). Ganly et al. demonstrated a threefold increase in overall and disease-free survival following adjuvant conventional radiation therapy (11). The risk of recurrence was fourfold in patients not receiving radiotherapy. As a result of incomplete data regarding precise staging, dose or delivery of radiation, and previous treatment, such retrospective studies must be interpreted with caution. Harrison et al. reported 3-, 5-, and 10-year disease-specific survival of 47%, 28% and 8%, respectively with surgery alone (19). Similar results were reported by Freedman and colleagues who found 46% and 31% survival at 3 and 5 years among their group of patients treated with surgery and postoperative radiotherapy (20).

1.0 3-Year OS (28.2%) 3-Year DSS (29.7%) 3-Year RFS (25.5%)

0.9 0.8 Proportion surviving

468

0.7 0.6 0.5 0.4 0.3 0.2 0.1 0.0

0

10

20

30 40 Follow-up time, mo

50

60

Figure 7 Survival analysis of patients with malignant melanoma of the paranasal sinuses. Three-year overall survival (OS), disease-specific survival (DSS), and recurrence-free survival (RFS) for craniofacial resection for malignant melanoma invading the skull base. Source: From Ref. 11.


LOCAL

NECK

17%

6%

3%

8% 12%

3%

14%

DISTANT Figure 8 Patterns of recurrence in sinonasal melanomas. No recurrence was found in 11 of the 35 patients (31.3%), isolated locoregional recurrence in 11 of 35 (31.3%), isolated distant failure in 4 of 35 (11.4%), and local and/or regional recurrence with distant failure in 9 of 35 patients (25.7%). Source: From Ref. 26.

Local failure is a significant cause of mortality in patients with sinonasal melanoma. Disease recurrence is common within the first 2 years after treatment and late recurrence may be seen even after 5 years of follow-up. Freedman et al. reported a series of 56 patients with sinonasal melanoma, 34 of them (60%) had recurrence in the primary site and 29% underwent salvage resection of the tumor (20). Patel et al. reported local failure rate of 50%, nodal failure rate of 20%, and distant failure rate of 40% (Fig. 8) (26). Only 6% of the patients in their study were eligible for salvage therapy. Patients with cutaneous melanoma and negative lymph nodes at presentation have a 5-year survival rate of 80% compared with 30% for those with positive nodes. In contrast, patients with mucosal melanoma of the head and neck have 27% and 19% 5-year survival for N0 and N+ disease, respectively (26,92). Similar results were found by Yii et al. who reported 26% 5-year survival rate for patients with localized disease, compared to 0% 5-year survival for patients with regional or distant metastases (32). Temam et al. reported that pathologic neck stage did not influence the overall survival of patients with mucosal melanomas of the head and neck (29). In most patients, local recurrence is an ominous sign for ongoing distant disease. Although rare at presentation, 37% of the patients will ultimately fail at distant sites, more than two-thirds of whom will also develop local or regional disease. Stern and Guillamondegui reported that 89% of patients with local recurrence also develop disseminated disease (31). The most common site of distant metastases is the lung and brain (33% and 14%, respectively). The median survival period from the time of detection of distant metastases to death is 7.1 months (26).

469

Although there are reports of anecdotal cases in which the disease remained dormant for long period of time, most patients with sinonasal melanoma die of distant metastases with or without local recurrence. The unpredictability of the clinical behavior of this tumor is commonly characterized by cases with a prolonged clinical course despite repeated local recurrences and regional lymph node metastases (104). Thus, there is a lifelong risk of recurrence for all patients with sinonasal mucosal melanoma. For this reason, lifelong surveillance of these patients is required. The value of salvage surgery for local recurrences and regional lymph node metastases cannot be predicted based on clinical and pathologic parameters alone. It is plausible that the course of disease is based on interactions between the primary tumor and the host’s immune system, but the exact mechanism of such interaction is currently not evident.

CONCLUSIONS Melanoma of the nasal cavity and paranasal sinuses is an uncommon malignancy arising in the respiratory sinonasal mucosa. The majority of cases originate in the nasal cavity followed by the paranasal sinuses. Whereas most head and neck melanomas are more likely to be discovered by the patient or during routine physical examination, sinonasal melanomas are not accessible to self-examination and are often diagnosed late, resulting in poor survival. Melanoma has the tendency to mimic other tumors pathologically and can lead to initial misdiagnosis. Due to the complicated differential diagnosis of this tumor, immunocytochemical staining is frequently required to establish the diagnosis of melanoma, particularly in cases of amelanotic variants. Surgery is considered the treatment of choice for primary mucosal melanoma of the sinonasal cavities. Due to the nature of this disease, it is challenging to achieve complete tumor resection and the majority of those patients will eventually develop local recurrence. The high recurrence rate of sinonasal melanoma is secondary to a high incidence of multifocal disease and the presence of submucosal lymphatic spread of tumor cells. Thus, all lesions require wide surgical resection, attempting to minimize unnecessary compromise of function and cosmesis. Of the patients with mucosal melanomas of the head and neck, those with disease involving the paranasal sinuses have the poorest outcome, whereas tumors isolated to the nasal cavity are associated with a better prognosis. Local failure is a significant cause of mortality in these patients and is clearly associated with high rate of nodal recurrence and distant metastases. Most patients with local recurrence are not amended to curative salvage therapy. The use of radiation therapy for treatment of melanoma is controversial. Recent studies indicate that the addition of adjuvant radiotherapy tends to decrease the rate of local failure, but has no significant impact on overall survival. Primary radiotherapy or chemotherapy is currently employed as palliative treatment of recurrent, inoperable or metastatic disease, or for patients with unacceptable surgical risk. Unfortunately, most reports showed no survival advantage for chemotherapy in patients with disseminated disease. The utility of other treatment modalities such as immunotherapy or biochemotherapy, as well as heavy particle radiation sources and intensity-modulated radiation therapy , awaits further evaluation. Further studies are required to determine the advantage of postoperative radiotherapy for treatment of patients with sinonasal melanoma. Nevertheless, it is advisable to add hypofractionation radiotherapy after surgery,

470

Gil et al.

especially for treatment of sinonasal melanomas, which historically has been associated with a high incidence of local recurrence after surgery alone.

ACKNOWLEDGMENT We thank Diana L. Carlson, MD, Department of Pathology, Memorial Sloan–Kettering Cancer Center, for providing the original pathological photographs for this chapter.

REFERENCES 1. Chang AE, Karnell LH, Menck HR. The National Cancer Data Base report on cutaneous and noncutaneous melanoma: A summary of 84,836 cases from the past decade. The American College of Surgeons Commission on Cancer and the American Cancer Society. Cancer. 1998;83(8):1664–1678. 2. Weber CO. Chirurgische Ehrfahrungen und Untersuchungen, nebst zahlreichen Beobachtungen aus der chirurgischen Klinik und dem evangelischen Krahkenhause zu Bonn. Berlin: G. Reimer, 1859. 3. Lucke A. Die Lehre von den Geschwulsten in anatomischer und klinischer Beziehung in Handbuch d. allg. u. spec. chir. Erlangen. 1869:244. 4. Viennois L. Osteotomie du nez (Obs. II - Polype melanique du nez (melanosarcome) ayant detruit la paroi interne de l’orbite et refoulant le globe de l’oeil en avant: abaissement du nez comme operation preliminaire; ablation du polyp, pyohemie; mort huit jours apres). Lyon Med. 1872;11:8–12. 5. Lincoln RP. A case of melano-sarcoma of the nose cured by galvanocauterization. N Y Med J. 1885;42:406–407. 6. Chin L. The genetics of malignant melanoma: Lessons from mouse and man. Nat Rev Cancer. 2003;3(8):559–570. 7. Wingo PA, Ries LA, Rosenberg HM, et al. Cancer incidence and mortality, 1973–1995: A report card for the U.S. Cancer. 1998;82(6):1197–1207. 8. Rigel DS. Malignant melanoma: Incidence issues and their effect on diagnosis and treatment in the 1990s. Mayo Clin Proc. 1997;72(4):367–371. 9. Thompson LD, Wieneke JA, Miettinen M, et al. Sinonasal tract and nasopharyngeal melanomas: A clinicopathologic study of 115 cases with a proposed staging system. Am J Surg Pathol. 2003;27(5):594–611. 10. Conley JJ. Melanoma of the Head and Neck. 1st ed. New York, NY: Georg Thiem Verlag, 1990:154–178. 11. Ganly I, Patel SG, Singh B, et al. Craniofacial resection for malignant melanoma of the skull base: Report of an international collaborative study. Arch Otolaryngol Head Neck Surg. 2006;132(1):73–78. 12. Lewis MG, Martin JA. Malignant melanoma of the nasal cavity in Ugandan Africans. Relationship of ectopic pigmentation. Cancer. 1967;20(10):1699–1705. 13. Andersen LJ, Berthelsen A, Hansen HS. Malignant melanoma of the upper respiratory tract and the oral cavity. J Otolaryngol. 1992;21(3):180–185. 14. Takagi M, Ishikawa G, Mori W. Primary malignant melanoma of the oral cavity in Japan. With special reference to mucosal melanosis. Cancer. 1974;34(2):358–370. 15. Kingdom TT, Kaplan MJ. Mucosal melanoma of the nasal cavity and paranasal sinuses. Head Neck. 1995;17(3):184–189. 16. Trapp TK, Fu YS, Calcaterra TC. Melanoma of the nasal and paranasal sinus mucosa. Arch Otolaryngol Head Neck Surg. 1987;113(10):1086–1089. 17. Blatchford SJ, Koopmann CF Jr, Coulthard SW. Mucosal melanoma of the head and neck. Laryngoscope. 1986;96(9 Pt 1):929–934. 18. Holdcraft J, Gallagher JC. Malignant melanomas of the nasal and paranasal sinus mucosa. Ann Otol Rhinol Laryngol. 1969;78(1):5–20.

19. Harrison DF. Malignant melanomata arising in the nasal mucous membrane. J Laryngol Otol. 1976;90(11):993–1005. 20. Freedman HM, DeSanto LW, Devine KD, et al. Malignant melanoma of the nasal cavity and paranasal sinuses. Arch Otolaryngol. 1973;97(4):322–325. 21. Matias C, Corde J, Soares J. Primary malignant melanoma of the nasal cavity: A clinicopathologic study of nine cases. J Surg Oncol. 1988;39(1):29–32. 22. Gilligan D, Slevin NJ. Radical radiotherapy for 28 cases of mucosal melanoma in the nasal cavity and sinuses. Br J Radiol. 1991;64(768):1147–1150. 23. Manolidis S, Donald PJ. Malignant mucosal melanoma of the head and neck: Review of the literature and report of 14 patients. Cancer. 1997;80(8):1373–1386. 24. Moore ES, Martin H. Melanoma of the upper respiratory tract and oral cavity. Cancer. 1955;8(6):1167–1176. 25. Panje WR, Moran WJ. Melanoma of the upper aerodigestive tract: A review of 21 cases. Head Neck Surg. 1986;8(4):309– 312. 26. Patel SG, Prasad ML, Escrig M, et al. Primary mucosal malignant melanoma of the head and neck. Head Neck. 2002;24(3):247– 257. 27. Holmstrom M, Lund VJ. Malignant melanomas of the nasal cavity after occupational exposure to formaldehyde. Br J Ind Med. 1991;48(1):9–11. 28. Hernberg S, Westerholm P, Schultz-Larsen K, et al. Nasal and sinonasal cancer. Connection with occupational exposures in Denmark, Finland and Sweden. Scand J Work Environ Health. 1983;9(4):315–326. 29. Temam S, Mamelle G, Marandas P, et al. Postoperative radiotherapy for primary mucosal melanoma of the head and neck. Cancer. 2005;103(2):313–319. 30. Guzzo M, Grandi C, Licitra L, et al. Mucosal malignant melanoma of head and neck: Forty-eight cases treated at Istituto Nazionale Tumori of Milan. Eur J Surg Oncol. 1993;19(4):316– 319. 31. Stern SJ, Guillamondegui OM. Mucosal melanoma of the head and neck. Head Neck. 1991;13(1):22–27. 32. Yii NW, Eisen T, Nicolson M, et al. Mucosal malignant melanoma of the head and neck: The Marsden experience over half a century. Clin Oncol (R Coll Radiol). 2003;15(4):199–204. 33. Ravid JM, Esteves JA. Malignant melanoma of the nose and paranasal sinuses and juvenile melanoma of the nose. Arch Otolaryngol. 1960;72:431–444. 34. Nandapalan V, Roland NJ, Helliwell TR, et al. Mucosal melanoma of the head and neck. Clin Otolaryngol Allied Sci. 1998;23(2):107–116. 35. Zak FG, Lawson W. The presence of melanocytes in the nasal cavity. Ann Otol Rhinol Laryngol. 1974;83(4):515–519. 36. Szabo G. The number of melanocytes in human epidermis. Br Med J. 1954;4869:1016–1017. 37. Mannone F, De Giorgi V, Cattaneo A, et al. Dermoscopic features of mucosal melanosis. Dermatol Surg. 2004;30(8):1118–1123. 38. Cove H. Melanosis, melanocytic hyperplasia, and primary malignant melanoma of the nasal cavity. Cancer. 1979;44(4):1424– 1433. 39. Hofbauer GF, Boni R, Simmen D, et al. Histological, immunological and molecular features of a nasal mucosa primary melanoma associated with nasal melanosis. Melanoma Res. 2002;12(1):77–82. 40. Bradley PJ. Primary malignant mucosal melanoma of the head and neck. Curr Opin Otolaryngol Head Neck Surg. 2006;14(2):100–104. 41. Prasad ML, Busam KJ, Patel SG, et al. Clinicopathologic differences in malignant melanoma arising in oral squamous and sinonasal respiratory mucosa of the upper aerodigestive tract. Arch Pathol Lab Med. 2003;127(8):997–1002. 42. Lee SP, Shimizu KT, Tran LM, et al. Mucosal melanoma of the head and neck: The impact of local control on survival. Laryngoscope. 1994;104(2):121–126. 43. Crawford RI, Tron VA, Ma R, et al. Sinonasal malignant melanoma—A clinicopathologic analysis of 18 cases. Melanoma Res. 1995;5(4):261–265.

Chapter 32: Melanoma of the Nasal Cavity and Paranasal Sinuses 44. Franquemont DW, Mills SE. Sinonasal malignant melanoma. A clinicopathologic and immunohistochemical study of 14 cases. Am J Clin Pathol. 1991;96(6):689–697. 45. Chetty R, Slavin JL, Pitson GA, et al. Melanoma botryoides: A distinctive myxoid pattern of sino-nasal malignant melanoma. Histopathology. 1994;24(4):377–379. 46. Prasad ML, Patel SG, Busam KJ. Primary mucosal desmoplastic melanoma of the head and neck. Head Neck. 2004;26(4):373– 377. 47. Batsakis JG, Regezi JA, Solomon AR, et al. The pathology of head and neck tumors: Mucosal melanomas, part 13. Head Neck Surg. 1982;4(5):404–418. 48. Kilpatrick SE, White WL, Browne JD. Desmoplastic malignant melanoma of the oral mucosa. An underrecognized diagnostic pitfall. Cancer. 1996;78(3):383–389. 49. Hoorweg JJ, Loftus BM, Hilgers FJ. Osteoid and bone formation in a nasal mucosal melanoma and its metastasis. Histopathology. 1997;31(5):465–468. 50. Ordonez NG, Ji XL, Hickey RC. Comparison of HMB-45 monoclonal antibody and S-100 protein in the immunohistochemical diagnosis of melanoma. Am J Clin Pathol. 1988;90(4):385–390. 51. Gown AM, Vogel AM, Hoak D, et al. Monoclonal antibodies specific for melanocytic tumors distinguish subpopulations of melanocytes. Am J Pathol. 1986;123(2):195–203. 52. Chello M, Marchese AR, Panza A, et al. Primary malignant melanoma of the oesophagus with a left atrial metastasis. Thorax. 1993;48(2):185–186. 53. Mohandas KM, Santhi Swaroop V, Dhir V, et al. Primary malignant melanoma of the esophagus palliated with endoscopic laser therapy. Endoscopy. 1993;25(8):551. 54. Uno Y, Saito R, Kanatani M, et al. Malignant melanoma of the nasal cavity. Auris Nasus Larynx. 1994;21(4):248–252. 55. Prasad ML, Jungbluth AA, Iversen K, et al. Expression of melanocytic differentiation markers in malignant melanomas of the oral and sinonasal mucosa. Am J Surg Pathol. 2001;25(6):782–787. 56. Wright JL, Heenan PJ. Electron microscopy findings in malignant melanoma of nose. ORL J Otorhinolaryngol Relat Spec. 1975;37(4):233–236. 57. Henzen-Logmans SC, Meijer CJ, Ruiter DJ, et al. Diagnostic application of panels of antibodies in mucosal melanomas of the head and neck. Cancer. 1988;61(4):702–711. 58. Brandwein MS, Rothstein A, Lawson W, et al. Sinonasal melanoma. A clinicopathologic study of 25 cases and literature meta-analysis. Arch Otolaryngol Head Neck Surg. 1997;123(3):290–296. 59. Lentsch EJ, Myers JN. Melanoma of the head and neck: Current concepts in diagnosis and management. Laryngoscope. 2001;111(7):1209–1222. 60. Mills SE, Fechner RE. “Undifferentiated” neoplasms of the sinonasal region: Differential diagnosis based on clinical, light microscopic, immunohistochemical, and ultrastructural features. Semin Diagn Pathol. 1989;6(4):316–328. 61. Nakhleh RE, Wick MR, Rocamora A, et al. Morphologic diversity in malignant melanomas. Am J Clin Pathol. 1990;93(6):731– 740. 62. Wick MR, Stanley SJ, Swanson PE. Immunohistochemical diagnosis of sinonasal melanoma, carcinoma, and neuroblastoma with monoclonal antibodies HMB-45 and anti-synaptophysin. Arch Pathol Lab Med. 1988;112(6):616–620. 63. Ejaz A, Wenig BM. Sinonasal undifferentiated carcinoma: Clinical and pathologic features and a discussion on classification, cellular differentiation, and differential diagnosis. Adv Anat Pathol. 2005;12(3):134–143. 64. Kuwabara H, Takeda J. Malignant melanoma of the lacrimal sac with surrounding melanosis. Arch Pathol Lab Med. 1997;121(5):517–519. 65. Hoyt DJ, Jordan T, Fisher SR. Mucosal melanoma of the head and neck. Arch Otolaryngol Head Neck Surg. 1989;115(9):1096– 1099. 66. Hammersmith SM, Terk MR, Jeffrey PB, et al. Magnetic resonance imaging of nasopharyngeal and paranasal sinus melanoma. Magn Reson Imaging. 1990;8(3):245–253.

471

67. Chang PC, Fischbein NJ, McCalmont TH, et al. Perineural spread of malignant melanoma of the head and neck: Clinical and imaging features. Am J Neuroradiol. 2004;25(1):5–11. 68. Civantos FJ, Moffat FL, Goodwin WJ. Lymphatic mapping and sentinel lymphadenectomy for 106 head and neck lesions: Contrasts between oral cavity and cutaneous malignancy. Laryngoscope. 2006;112(3)(Pt 2)(Suppl 109):1–15. 69. Boni R, Boni RA, Steinert H, et al. Staging of metastatic melanoma by whole-body positron emission tomography using 2-fluorine-18-fluoro-2-deoxy-D-glucose. Br J Dermatol. 1995;132(4):556–562. 70. Billings KR, Wang MB, Sercarz JA, et al. Clinical and pathologic distinction between primary and metastatic mucosal melanoma of the head and neck. Otolaryngol Head Neck Surg. 1995;112(6):700–706. 71. Balch CM, Buzaid AC, Soong SJ, et al. Final version of the American Joint Committee on Cancer staging system for cutaneous melanoma. J Clin Oncol. 2001;19(16):3635–3648. 72. American Joint Committee on Cancer. Head and Neck Sites. 6th ed. New York, NY: Springer, 2002:17–87. 73. Bridger AG, Smee D, Baldwin MA, et al. Experience with mucosal melanoma of the nose and paranasal sinuses. ANZ J Surg. 2005;75(4):192–197. 74. Ballantyne AJ. Malignant melanoma of the skin of the head and neck. An analysis of 405 cases. Am J Surg. 1970;120(4):425– 431. 75. Mendenhall WM, Amdur RJ, Hinerman RW, et al. Head and neck mucosal melanoma. Am J Clin Oncol. 2005;28(6):626–630. 76. Prasad ML, Patel SG, Huvos AG, et al. Primary mucosal melanoma of the head and neck: A proposal for microstaging localized, Stage I (lymph node-negative) tumors. Cancer. 2004;100(8):1657–1664. 77. Jemal A, Clegg LX, Ward E, et al. Annual report to the nation on the status of cancer, 1975–2001, with a special feature regarding survival. Cancer. 2004;101(1):3–27. 78. Lund VJ, Howard DJ, Harding L, et al. Management options and survival in malignant melanoma of the sinonasal mucosa. Laryngoscope. 1999;109(2 Pt 1):208–211. 79. Shah J. Nasal cavity and paranasal sinuses. In: Mosbey, ed. Head and Neck Surgery and Oncology. New York, NY: Elsevier, 2003:57–93. 80. Hao SP, Pan WL, Chang CN, et al. The use of the facial translocation technique in the management of tumors of the paranasal sinuses and skull base. Otolaryngol Head Neck Surg. 2003;128(4):571–575. 81. Bilsky MH, Bentz B, Vitaz T, et al. Craniofacial resection for cranial base malignancies involving the infratemporal fossa. Neurosurgery. 2005;57:339–347. 82. Bentzen SM, Overgaard J, Thames HD, et al. Clinical radiobiology of malignant melanoma. Radiother Oncol. 1989;16(3):169– 182. 83. Ang KK, Byers RM, Peters LJ, et al. Regional radiotherapy as adjuvant treatment for head and neck malignant melanoma. Preliminary results. Arch Otolaryngol Head Neck Surg. 1990;116(2):169–172. 84. Stevens G, Thompson JF, Firth I, et al. Locally advanced melanoma: Results of postoperative hypofractionated radiation therapy. Cancer. 2000;88(1):88–94. 85. Raben A, Zelefsky M, Harrison LB. High dose per fractionation, short course irradiation for mucosal melanoma of the head and neck. Presented at the 4th International Head and Neck Society Meeting. Toronto, Canada, June 15, 1996. 86. Owens JM, Roberts DB, Myers JN. The role of postoperative adjuvant radiation therapy in the treatment of mucosal melanomas of the head and neck region. Arch Otolaryngol Head Neck Surg. 2003;129(8):864–868. 87. Harwood AR, Cummings BJ. Radiotherapy for mucosal melanomas. Int J Radiat Oncol Biol Phys. 1982;8(7):1121–1126. 88. Trotti A, Peters LJ. Role of radiotherapy in the primary management of mucosal melanoma of the head and neck. Semin Surg Oncol. 1993;9(3):246–250. 89. Albertsson M, Tennvall J, Andersson T, et al. Malignant melanoma of the nasal cavity and nasopharynx treated with

472

90.

91.

92. 93.

94.

95.

96. 97.

98.

Gil et al. cisplatin and accelerated hyperfractionated radiation. Melanoma Res. 1992;2(2):101–104. Wada H, Nemoto K, Ogawa Y, et al. A multi-institutional retrospective analysis of external radiotherapy for mucosal melanoma of the head and neck in Northern Japan. Int J Radiat Oncol Biol Phys. 2004;59(2):495–500. Seo W, Ogasawara H, Sakagami M. Chemohormonal therapy for malignant melanomas of the nasal and paranasal mucosa. Rhinology. 1997;35(1):19-21. Shah JP, Huvos AG, Strong EW. Mucosal melanomas of the head and neck. Am J Surg. 1977;134(4):531–535. Harbo G, Grau C, Bundgaard T, et al. Cancer of the nasal cavity and paranasal sinuses. A clinico-pathological study of 277 patients. Acta Oncol. 1997;36(1):45–50. Garrott H, O’Day J. Optic neuropathy secondary to radiotherapy for nasal melanoma. Clin Experiment Ophthalmol. 2004;32(3):330–333. Gil Z, Abergel A, Spektor S, et al. Quality of life following surgery for anterior skull base tumors. Arch Otolaryngol Head Neck Surg. 2003;129(12):1303–1309. Blake PR, Catterall M, Errington RD. Treatment of malignant melanoma by fast neutrons. Br J Surg. 1985;72(7):517–519. Linstadt DE, Castro JR, Phillips TL. Neon ion radiotherapy: Results of the phase I/II clinical trial. Int J Radiat Oncol Biol Phys. 1991;20(4):761–769. Mizoe JE, Tsujii H, Kamada T, et al. Organizing Committee for the Working Group for Head-And-Neck Cancer. Dose escalation

99.

100. 101.

102.

103.

104.

105.

study of carbon ion radiotherapy for locally advanced headand-neck cancer. Int J Radiat Oncol Biol Phys. 2004;60(2):358– 364. Jacquillat C, Khayat D, Banzet P, et al. Chemotherapy by fotemustine in cerebral metastases of disseminated malignant melanoma. Cancer Chemother Pharmacol. 1990;25(4):263–266. National Comprehensive Cancer Network, Inc Version 1.2005, 2005. www.nccn.org Kirkwood JM, Strawderman MH, Ernstoff MS, et al. Interferon alfa-2b adjuvant therapy of high-risk resected cutaneous melanoma: The Eastern Cooperative Oncology Group Trial EST 1684. J Clin Oncol. 1996;14(1):7–17. Kirkwood JM, Ibrahim JG, Sondak VK, et al. High- and low-dose interferon alfa-2b in high-risk melanoma: First analysis of intergroup trial E1690/S9111/C9190. J Clin Oncol. 2000;18(12):2444– 2458. Legha SS, Ring S, Eton O, et al. Development of a biochemotherapy regimen with concurrent administration of cisplatin, vinblastine, dacarbazine, interferon alfa, and interleukin2 for patients with metastatic melanoma. J Clin Oncol. 1998;16(5):1752–1759. Eneroth CM, Lundberg C. Mucosal malignant melanomas of the head and neck with special reference to cases having a prolonged clinical course. Acta Otolaryngol. 1975;80(5–6):452–458. Patel SG, Singh B, Polluri A, et al. Craniofacial surgery for malignant skull base tumors: Report of an international collaborative study. Cancer. 2003;98(6):1179–1187.

33 Sarcomas of the Skull Base Katherine A. Thornton and Robert S. Benjamin

INCIDENCE AND EPIDEMIOLOGY

ment and are treated at specialty centers (4). The most comprehensive data addressing this issue in soft tissue sarcoma come from Sweden, where Gustafson and colleagues analyzed the quality of treatment in a population-based series of 375 patients with primary soft tissue sarcomas arising in the extremities (n = 329) or the trunk (n = 46) (5). Comparisons were made between patients referred to a specialty soft tissue tumor center before surgery (n = 195), those referred after surgery (n = 102), and those not referred for treatment of the primary tumor (n = 78). The total number of operations for the primary tumor was 1.4 times higher in the patients not referred and 1.7 times higher in the patients referred after surgery than in patients referred before surgery. Of greatest significance, however, was the finding that the local recurrence rate was 2.4 times higher in the patients not referred and 1.3 times higher in the patients referred after surgery than in patients referred to a specialty soft tissue tumor center before any manipulation of their tumor. These findings support the principle of centralizing treatment of these rare tumors, which frequently require complex multimodality therapy. While these data were collected on patients with sarcomas of the extremities and trunk, they should apply with even more certainty to patients with skull base sarcomas, where the expertise is even more limited.

Sarcomas comprise a group of relatively rare, anatomically and histologically diverse neoplasms. These tumors share a common embryologic origin, arising primarily from mesenchymal tissue. The notable exceptions are malignant peripheral nerve sheath tumors and primitive neuroectodermal tumors, also known as Ewing sarcomas, which are believed to arise from neuroectodermal tissue. Despite the fact that the somatic soft tissues account for as much as 75% of total body weight, neoplasms of the soft tissues are comparatively rare, accounting for less than 1% of adult malignancies and 15% of pediatric malignancies. The annual incidence of soft tissue sarcomas in the United States is about 15,000 if those arising in organs are included, according to SEER estimates. For bone sarcomas, the incidence is about 2500 cases. For sarcomas arising in somatic soft tissue, there are only about 8000 new cases, but only about 4% of these or about 300 cases arise in the head and neck. Nonetheless, the overall mortality rate of sarcomas is almost 50% at all sites, and tumors of the skull base are particularly hard to cure due to the inability to perform adequate oncologic surgery due to anatomic constraints. Head and neck sarcomas are uncommon, accounting for less than 1% of head and neck malignancies in adults. In recent large series of 176 (1), 188 (2), and 254 (3) patients with head and neck sarcomas, the most common anatomic sites were the neck (23–38%) and paranasal sinuses (14–30%). Thus, skull base sarcomas represent a small subset of a rare group of tumors. While almost any histologic type of sarcoma can occur in the head and neck, certain tumors are found more commonly at the skull base: hemangiopericytomas of the dura and chordomas. These will be discussed in detail following brief description of various soft tissue and bone sarcomas, any of which can involve the soft tissues or bones of the skull base.

SURGERY Surgical management of sarcomas is no different from that of other tumors at a similar location except that lymph node resection is rarely required. Wide excision of the tumor with a margin of normal tissue is preferred, but anatomic constraints often preclude more than a gross total resection. Details of surgical management are discussed elsewhere in this book. Wide surgical excision with microscopically negative margins is the therapeutic mainstay for all sarcomas, but is rarely achievable at the base of skull. Local recurrence remains a significant problem, with overall rates of local recurrence for all head and neck sarcomas ranging from 14% to 48% (1,2,6), making it especially important to consider appropriately applied principles of sarcoma management with combined modality approaches where appropriate. As with sarcomas elsewhere in the body, biologic behavior is a function of histologic grade, with local recurrence rates ranging from 22% for low-grade head and neck sarcomas to 48% for high-grade lesions (1). Systemic recurrence develops in 12% to 31% of patients despite complete resection (1,6). Overall 5-year survival rates are 45% to 68% (1–3,7).

DIAGNOSIS Methods of diagnosis, imaging, and biopsy for head and neck sarcomas do not differ substantially from those for other skull base tumors. CT and MRI are complementary studies, with CT providing sharper images of cortical bone and MRI showing details of intracranial extension or bone marrow involvement. While fine-needle aspiration may occasionally be definitive, most centers require at least a core biopsy to establish the diagnosis.

TREATMENT Treatment of Sarcoma Patients at Specialty Centers

RADIATION TREATMENT

Recent data on other tumor types have demonstrated improved outcomes for patients who required complex treat-

Based on experience gained in treating extremity sarcomas, adjuvant radiotherapy should be utilized routinely since 473

474

Thornton and Benjamin

there is always doubt as to the adequacy of surgical margins, and, in many cases, the location of the tumor precludes complete excision. Evidence for the benefit of adjuvant radiotherapy for head and neck sarcomas is less plentiful than for extremity lesions; however, Tran and colleagues from the University of California, Los Angeles have shown that local control was 52% with surgery alone versus 90% in head and neck patients treated with combined radiotherapy and surgery (8). Additional evidence from the Princess Margaret Hospital reveals that head and neck STS patients with clear surgical margins or microscopic residuum had similar local failure rates (26% and 30% failure, respectively), provided radiotherapy was administered (9). Indeed, these outcomes for radiotherapy after R0/R1 resections approach those achieved in extremity sarcoma. Data on skull base tumors are less clear. Gil et al. noted that postoperative radiotherapy in their cohort of patients with anterior skull base sarcomas did not affect disease-specific survival, but only 34% of their patient population was irradiated (10). Similarly, Prabhu et al. could not identify benefit in either progression-free survival or overall survival in their patients with skull base sarcomas, 62% of whom received postoperative radiation therapy (11). One strategy to improve outcome in head and neck sarcomas is through the use of preoperative radiotherapy. This approach may have particular advantages in this site because of the smaller volumes of radiotherapy and the lower doses that can be used compared to the postoperative treatment in difficult surgical access locations, especially in the base of skull. Obvious advantages provided relate to the ability to spare critical anatomy such as the optic structures (globes, optic nerves, and the optic chiasm), as well as the brain stem and spinal cord. If for no other reason, the preoperative approach promotes collaboration between the surgical and radiation oncologist, facilitates a complete management plan to be fashioned before any surgical intervention, and maximizes the opportunity to achieve local control even when disease may be resected with a small but planned positive margin against critical unexpendable anatomy, as discussed earlier (12). That said, many head and neck surgeons, including those at MD Anderson, prefer to use radiation only postoperatively. A prospective series of 40 patients with head and neck sarcomas (excluding rhabdomyosarcoma) with adverse selection criteria was managed with preoperative radiotherapy between 1989 and 1999 at the Princess Margaret Hospital. If the four patients with angiosarcoma are excluded (almost always involve the skin, and thus are not relevant to skull base tumors) there were three local relapses in the 36 patients (overall control rate of 92%) (13). This population of head and neck patients included five patients with intracranial extension, one with spinal cord compression, more than half were greater than 5 cm in size (a formidable problem for lesions in this anatomical location), and 85% were deep to the investing fascia. The metastatic relapse-free rate also exceeded 80% in this series, potentially related in part to the smaller overall dimension of sarcomas in this location compared to sarcomas elsewhere. Also the improved local control compared to a previous series of patients treated at the same institution may have contributed to this amelioration because the local control rate in the earlier series was substantially lower and death from concurrent local and metastatic disease was evident (9). Wound complications, assessed by the Canadian trial criteria (14), were also seen with less frequency in this prospective study of head neck lesions (overall rate of 8 of 40 or 20%) (13) than were noted earlier with preoperative

radiotherapy in extremity lesions. This may relate to the greater use of flaps for head and neck reconstruction. At present, useful guidelines for using preoperative radiotherapy in the head and neck are: (i) the need to maximally restrict radiotherapy volumes in some anatomic sites (e.g., close to critical anatomy); (ii) the desire to minimize radiation dose in some situations (e.g., where critical neurological tissues are in close proximity, as in the optic structures); and (iii) a desire not to irradiate new tissues, especially vascular reconstructions vulnerable to the effects of high-dose postoperative radiotherapy. Newer techniques such as IMRT or proton therapy should permit better targeting where appropriate.

CHEMOTHERAPY The activity of chemotherapy for sarcomas was established by treating patients with metastatic disease at various primary sites and with diverse histologic diagnoses. Indeed, considering the diversity of diseases lumped as sarcomas, it is amazing that any drugs have been found to have activity. Lessons from those studies can be applied to patients with skull base sarcomas, but with a number of histology-directed exceptions.

Anthracyclines The single-agent activity of doxorubicin against metastatic soft tissue sarcoma is well established as being in the range of 20% to 40% (15–18). A randomized study in the Southwest Oncology Group demonstrated a steep dose–response curve for doxorubicin in sarcomas, in contrast to other tumors (19). The response rate increased significantly from 17% at 45 mg/m2 to 37% at 75 mg/m2 . Doxorubicin administration is limited acutely by mucositis, especially when given over several days. It is limited chronically by cardiomyopathy that increases after cumulative doses exceeding 400 mg/m2 when given by rapid infusion. With cardioprotective strategies, continuous infusion over 48 to 96 hours or premedication with dexrazoxane, cumulative doses can be doubled, but at the cost of increased mucositis (infusion, particularly >48 hours) or myelosuppression (dexrazoxane). Epirubicin, developed as an active but minimally cardiotoxic analogue of doxorubicin, produced an objective response rate not significantly inferior than that of doxorubicin (18% versus 25%, P = 0.33) in an EORTC RCT of 167 patients receiving equimolar doses (75 mg/m2 ) of the drugs (20). There is some evidence of a dose–response relationship with epirubicin; a dose-escalation study showed response rates of 17%, 44%, and 100% for 140 mg/m2 , 160 mg/m2 , and 180 mg/m2 of epirubicin, respectively (21). Only three patients were entered at the maximum tolerated dose of 180 mg/m2 , and 160 mg/m2 was recommended for routine clinical use; however, the EORTC, in a three-arm randomized study of 334 patients, was unable to demonstrate any benefit from either of two schedules of epirubicin (150 mg/m2 ) compared with doxorubicin (75 mg/m2 ); all regimens produced response rates of 14% to 15% (22). Furthermore, there was considerably more myelosuppression in the two epirubicin arms, with two toxic deaths. Nevertheless, in some areas of Europe, epirubicin is frequently substituted for doxorubicin in high-dose regimens. A number of studies of liposomal anthracyclines have suggested that these agents have lower rates of cardiotoxicity but variable activity (23–27). An EORTC phase II RCT 320 demonstrated low activity for both doxorubicin and liposomal doxorubicin (Doxil), 9% versus 10%, but different

Chapter 33: Sarcomas of the Skull Base

spectrums of toxicity: less myelosuppression but palmer– plantar erythrodysesthesia (grade 3, 20%) as the dose-limiting toxicity with Doxil.

Ifosfamide After the reports of several studies documenting activity ranging from 24% to 67% (28–30), Bramwell and colleagues performed a randomized trial comparing ifosfamide (5 g/m2 by 24-hour infusion) with cyclophosphamide (1.5 g/m2 ) (31). Respective response rates were 18% and 8%, and although the difference was not statistically significant (P = 0.13), responses were seen only in patients failing cyclophosphamide and crossing over to ifosfamide and not the other direction. Further indirect data from several other studies in which ifosfamide and/or cyclophosphamide were added to doxorubicin have provided additional evidence that ifosfamide is a more active analogue than cyclophosphamide, and all sarcoma oncologists agree that ifosfamide is more active (15). Questions about the optimal scheduling of ifosfamide (multiple daily bolus doses versus continuous infusion) have never been satisfactorily resolved and are confounded by dose differences in many studies. Two consecutive phase II studies by investigators at MD Anderson (30) at 8 g/m2 and in Boston (28) at 10 g/m2 evaluated ifosfamide given as a continuous infusion or a 2-hour infusion daily for 4 to 5 consecutive days. Response rates in both groups were higher when intermittent short infusions were used.

High-Dose Ifosfamide Early studies of ifosfamide suggested that there was a dose– response relationship (30), and several groups have documented responses to high-dose ifosfamide in patients not responding to lower doses of the drug (32–35). Nevertheless dose-escalation studies of ifosfamide have produced conflicting results. Doses of 12 g/m2 without and 14 to 18 g/m2 with growth factor support seem achievable and have produced response rates of 33% to 45%, but nephrotoxicity and neurotoxicity are considerable (35–37). Frustaci and colleagues found high-dose ifosfamide to be well tolerated when infused at 1 g/m2 /d over 21 days (38). In 36 patients, they were able to administer up to three cycles of median duration 15 days, producing a response rate of 24%. Myelosuppression was dose limiting, but there was no significant nephrotoxicity or neurotoxicity. Pharmacokinetic data, reported by Cerny and colleagues demonstrated that ifosfamide doses greater than 14 to 16 g/m2 given over 5 days resulted in a relative decrease of the active metabolite phosphoramide mustard, suggesting dose-dependent saturation or inhibition of ifosfamide metabolism (39). Two consecutive phase II studies by investigators at MD Anderson evaluated ifosfamide (14 g/m2 ), given as a 72-hour continuous infusion or a 2-hour infusion for three consecutive days. Respective response rates were 19% and 42% (36).

Combination Chemotherapy In the 1970s and early 1980s before the widespread availability of ifosfamide, most combination chemotherapy regimens were based on doxorubicin and dacarbazine. These are still appropriate for patients over age 65 and those with impaired renal function. The addition of cyclophosphamide and vincristine, which are active against childhood sarcomas, created a regimen called CyVADIC, for which the Southwest Oncology Group reported response rates as high as 59% in patients with metastatic disease (40). Later investigators were unable to reproduce such high response rates with the same regimen, however, and summary data on variants of the CyVADIC

475

regimen revealed an overall response rate of 35% in 2092 patients (41). Most regimens now used for first-line chemotherapy are based on the combination of doxorubicin and ifosfamide. A recent systematic search of the literature (42) found three phase III studies and 16 phase II trials (excluding phase I studies and those recruiting less than 25 patients) in adult STS that used combination regimens including an anthracycline and ifosfamide. Although the response rate varied widely from 25% to 56% in the phase II studies, they were at the lower end of this range in the three phase III studies (43–45). In the Eastern Cooperative Oncology Group (ECOG) study (43) of 178 patients, the response rate was significantly higher for doxorubicin/ifosfamide at 60 mg/m2 and 6 g/m2 than for doxorubicin alone (34% versus 20%, P = 0.03) although median survivals were similar. In an EORTC study of 471 patients, however, there were no significant differences in response rate (28% versus 23%) or median survival for doxorubicin/ifosfamide at 50 mg/m2 and 5 g/m2 versus doxorubicin alone at 75 mg/m2 (44). In contrast, the highest response rates were seen in phase II studies from MD Anderson that exploited the dose–response relationship for the two agents (46). At doxorubicin doses of 75 to 90 mg/m2 and an ifosfamide dose of 10 g/m2 , the response rate for patients with metastatic disease was 62% (46).

Second-Line Chemotherapy Objective response rates have also been very low (3% to 5%) in three phase II studies of gemcitabine (47–49), although the MD Anderson group described a response rate of 18% in 39 patients, if GISTs were excluded (50). It is the only agent commercially available in the United States with significant single-agent activity in previously treated patients. Temozolomide (51–53), raltitrexed (54), irinotecan (55), sargramostim (56), topotecan (57,58), vinorelbine (59), and the taxanes (60–64) seem to have minimal activity in STS, despite their proven value in other tumor types. Of drugs currently in development, trabectidin (Yondelis, ET743), a DNA guanine-specific minor groove-binding agent, seems to have the most potential across the spectrum of sarcomas. Hints of activity in bone and STSs were observed in phase I trials (65,66) and appeared to be confirmed in phase II trials of this agent. Demetri and colleagues reported a response rate of 18% in 34 chemonaive sarcoma patients and 9% in 34 who had received prior chemotherapy. George and colleagues reported a lower progression rate (5%) but a substantial proportion (19%) of patients with minor responses or stable disease (67). Two European trials described response rates of 11% to 12% in previously treated sarcoma patients (68,69). The drug is particularly active against myxoid liposarcoma, where about 80% of patients show benefit (70). It is commercially available in Europe. Occasional severe toxicities, sometimes lethal, seemed to be related to elevated baseline liver function tests. Despite poor levels of activity as single agents, some drugs have been incorporated into nonanthracycline-based salvage regimens. Based on encouraging data in pediatric sarcomas, etoposide has been combined with ifosfamide, although with variable results (71–74). All but one such study produced response rates in the range of 38% to 46%; because ifosfamide given alone has produced up to 67% in phase II studies, however, the contribution of etoposide to these results is difficult to discern. Several investigators, the authors included, do not consider the contribution of etoposide significant in typical adult sarcomas. Combinations of docetaxel with gemcitabine are being evaluated, with preliminary

476


reports of encouraging synergistic activity (75–78). In a randomized study conducted by SARC (the Sarcoma Alliance for Research through Collaboration) of equitoxic doses of gemcitabine by timed infusion as a single agent versus the combination of timed infusion gemcitabine plus docetaxel, the combination produced a superior response rate, time to progression, and survival (78).

CHEMOTHERAPY CONSIDERATIONS FOR SPECIFIC HISTOLOGIC SUBTYPES Synovial Sarcoma Rosen and colleagues (79) were the first to suggest that synovial sarcomas were particularly responsive to ifosfamide. They documented three complete responses and nine partial responses in 13 patients (nine of whom had received prior doxorubicin-based chemotherapy) with metastatic synovial sarcomas. These investigators also reported on 14 patients with localized synovial sarcomas who received adjuvant doxorubicin/ifosfamide/cisplatin chemotherapy (80). There was one patient with local recurrence, but the remaining 13 patients (93%) remained disease free at a median follow-up period of 37 months (range, 6–85 months). In a large EORTC phase II trial of 124 patients with advanced STS receiving ifosfamide (12 g/m2 ), the overall response rate for all histological subtypes was 16% (81), but 8 (44%) of 18 patients with synovial sarcoma responded. Edmonson and colleagues described a higher response rate in synovial sarcomas for doxorubicin/ifosfamide than for doxorubicin alone (88% vs. 20%, P = 0.02) in the setting of the ECOG randomized phase III study of multiple histologic subtypes of STS (43). A subsequent ECOG phase II study of doxorubicin/ifosfamide in synovial sarcomas showed five partial responses (42%) in 12 patients; however, the median survival for the whole group was only 11 months, and the trial was closed because of poor accrual (82). In many studies evaluating ifosfamide, including some of the randomized studies, the question of response by histologic subtype has not been specifically addressed. In several studies (28,30,36,81), however, leiomyosarcoma stands out as particularly unresponsive, even when patients with GIST (30,36) are excluded. As many of these patients are young and fit, inclusion of ifosfamide in first-line chemotherapy for metastatic disease seems reasonable. If the circumstances merit adjuvant chemotherapy, an anthracycline/ifosfamide combination is the logical choice for fit patients younger than 65 with good renal function.

Liposarcoma Activation of the PPAR-gamma nuclear receptor stimulates terminal differentiation in preadipocytes. Thioglitazones, used in the treatment of diabetes mellitus, are activating ligands for PPAR-gamma. Biopsies, pre- and post-troglitazone therapy, were obtained in 34 of 49 patients with different types of liposarcomas entering a phase II trial (83). Five of seven (71%) evaluable patients with myxoid/round cell disease exhibited histologic evidence of lineage-appropriate differentiation of liposarcoma cells, whereas only one of three (33%) patients with high-grade pleomorphic disease showed such changes (84). Although this study provides proof-of-concept data, the clinical significance is uncertain as responses to troglitazone were not documented. Trabectidin, as mentioned above, is highly active against myxoid liposarcoma (70). It is significantly less active against other subtypes of liposarcomas and other sarcomas.

Pediatric Sarcomas (Rhabdomyosarcoma and Ewing Sarcoma) in Adults Embryonal rhabdomyosarcomas and the Ewing family of tumors, all seem to be chemosensitive when they occur in the adult age group. Adult patients with these tumors should receive aggressive combination chemotherapy. We prefer to use the basic doxorubicin/ifosfamide regimen, but to add vincristine. Others add vincristine and etoposide, and many prefer alternating regimens similar to those offered to children with the same disease (85,86). Nevertheless, the outcome is likely to be poorer for adults with “pediatric sarcomas” than for pediatric sarcoma patients. Actinomycin-D, inactive in most adult sarcomas, can be effective in both of these tumors, especially rhabdomyosarcomas. Topoisomerase I inhibitors, topotecan and irinotecan, also have definite activity, and are often combined with alkylating agents. These tumors are also more sensitive to radiation than most other sarcomas, and radiation is often an effective substitute for surgery if its use will decrease morbidity substantially.

Osteosarcoma Chemotherapy is usually employed in the neoadjuvant situation, and its value preoperatively has been conclusively demonstrated in patients with osteosarcoma of the extremities. Patients with osteosarcoma of the jaw have a better prognosis with surgery alone, but those with osteosarcoma arising elsewhere in the skull have a poor prognosis and should be treated with chemotherapy as initial therapy. There are four active agents: doxorubicin, ifosfamide (as for soft tissue sarcomas), cisplatin, and high-dose methotrexate. Patients with osteosarcoma of the extremities who show a complete response to preoperative chemotherapy with tumor destruction of at least 90% have significantly improved survival. Insufficient data exist about skull base osteosarcomas, but they are treated in similar fashion. Although a number of regimens have been used, we prefer the combination of doxorubicin 75 mg/m2 IV by 72-hour continuous infusion through a central venous catheter, and cisplatin 120 mg/m2 IV over 4 hours on day 1. If there is more than 90% necrosis, continue the same regimen until cisplatin neurotoxicity and then substitute ifosfamide. If there is less than 90% tumor necrosis at surgery, we would switch to an alternating regimen of high-dose methotrexate, high-dose ifosfamide, and doxorubicin/ifosfamide.

Chondrosarcoma There is no effective chemotherapy regimen for conventional chondrosarcoma of bone. Surgery remains the optimal treatment modality. Investigators at the Massachusetts General Hospital in Boston have championed the use of proton beam radiation for these tumors. The dose can be focused on the target, while achieving significant sparing of the brain, brain stem, cervical cord, and optic nerves and chiasm. For skull base chondrosarcomas, 10-year local control rates with combined proton–photon therapy are 94% (87).

Chordoma Chordomas are rare tumors of bone that originate from the embryonic remnants of the notochord, which is the tissue of derivation for the nucleus pulposus of the intervertebral disks in normal humans. The median age of presentation is around 60 years, with skull base tumors affecting a younger age population. They are typically slow growing tumors, however, they can be locally aggressive and invasive. They most often occur in the axial skeleton and typically arise from the spheno-occipital region of the skull base or the sacrum


(88–93). In adults, chordomas involve the sacrococcygeal region approximately 50% of the time, the base of the skull at the spheno-occipital region approximately 35% of the time, and 15% are found in the vertebral column (88). Craniocervical chordomas often involve the dorsum sella, clivus, and nasopharynx, and therefore, despite the tumors’ propensity to not initially metastasize, they can be extremely locally destructive and debilitating. Three distinct subtypes of chordoma are described: conventional, chondroid, and dedifferentiated. Conventional is the most commonly encountered subtype, and are identified by a lack of cartilaginous or mixed mesenchymal cell types. Chondroid chordomas contain both chordomatous and chondromatous features and can account for up to 33% of cranial chordomas (88). They occur in a younger population compared to conventional chordomas, and generally are less aggressive with a longer median survival. Dedifferentiated chordomas are those that have had a sarcomatous transformation, and make up 2% to 8% of all chordomas (94,95). They can develop at the onset of disease, or later, as the chordoma transforms. Most aggressive chordomas are aneuploid on DNA analysis, as compared to 27% of conventional chordomas, identified in a small DNA flow cytometric study (95). Surgery, with wide margins, can improve overall outcome (96,97). Given the anatomic constraints of chordomas with close proximity to vital structures, this is often impossible, and radiation is commonly used as adjunctive therapy. Utilizing radiation therapy in the management of chordomas is not without its inherent challenges. The tolerance dose of tissues like the brainstem, optic pathway, and spinal cord is much lower than the dose required for cure, approximately 70 Gy (98). Charged particle irradiation, like proton-beam irradiation, has typically been used in an adjuvant manner, enabling the delivery of higher radiation doses to the tumor, while limiting the dose to the surrounding structures, i.e., the eyes or spinal cord (96,99–101). This form of therapy has led to significant improvement in local control for patients with intracranial chordomas. One relatively large study treated 195 patients with chordomas of the base of the skull or cervical spine. At a median follow-up time of 54 months, 69% of patients were relapse free; 5-year and 10-year progression free survival rates of 70% and 45% respectively, were reported (101). There are few studies reporting the use of chemotherapy in chordomas. Although there have been anecdotal responses reported to anthracyclines, cisplatin, and alkylating agents, chemotherapy is generally thought to be inactive, with the exception of dedifferentiated chordomas, which behave more similarly to sarcomas with a propensity to be more responsive to cytotoxic therapy (102). A phase II study of 9-nitro-camptothecin in patients with advanced chordoma or soft tissue sarcoma showed modest activity in delaying progression in patients with unresectable or metastatic chordoma, whereas it had little activity in other soft tissue sarcomas and gastrointestinal stromal tumors (103). Recently, there have been a few small studies evaluating imatinib in the treatment of chordoma. Six cases were described in 2004, with a follow-up compassionate series reported shortly thereafter (104,105). Most patients responded to 800 mg daily of imatinib. Typically, responses were marked by hypodensity and decreased contrast uptake on CT scans. A tumor volume decrease was generally not encountered; however, despite lack of dimensional reduction, there was symptomatic improvement indicating response to therapy. The length of tumor response in a very advanced tumor population was generally 1 year (104,105).

477

Hemangiopericytoma Hemangiopericytoma (HPC) is a rare vascular tumor derived from the pericytes that can be found surrounding all capillaries (also referred to as pericytes of Zimmerman) and are thought to be immature smooth muscle cells of mesenchymal origin (106,107). Pericytes are thought to add to the structural support of blood vessels and play an active role in blood flow regulation. By virtue of their tissue of origin, they can be found virtually anywhere within the body, however, 15% to 25% present in the head and neck, especially associated with the dura. They can have an aggressive clinical course, therefore early diagnosis and implementation of local therapy and in some cases systemic therapy are important in effective management (108,109). The initial management for HPC is surgical resection with attempts at negative margins, if possible. This is oftentimes difficult in tumors situated in the head and neck. In a study published by Soyeur et al., 5-year local control rates after surgery were 84% for patients with gross total resections and 38% for partial resections (110). Postoperative radiation therapy has been shown to play a role in effective local control (111,112). In a study of 37 patients with HPC at a single institution, patients were treated with high-precision radiotherapy or intensity-modulated radiotherapy. Overall survival rates were 100% and 64% at 5 and 10 years, respectively (112). The role of chemotherapy in HPC is less certain. Various chemotherapeutic agents have been used with variable success rates, and given the overall rarity and heterogeneity of tumors, it has been difficult to perform adequate clinical trials. Cyclophosphamide, vincristine, methotrexate, dacarbazine, ifosfamide, and doxorubicin have all been used (113–114) in case series, with doxorubicin generally believed to be the most effective agent. These reports seem overly positive, and one wonders if cases of synovial sarcoma or myxoid liposarcoma, tumors with a prominent hemangiopercytic pattern, may have been included by pathologists with insufficient expertise in sarcomas. One case series utilized interferon alpha, given its antiangiogenesis properties. Tumor responses were seen and continuous freedom from disease progression of 18 to 24 months was reported (115). Other, more targeted antiangiogenic agents, like bevacizumab are currently being evaluated, with data anticipated. Our current approach utilizes a combination of temozolomide and bevacizumab. A preliminary report submitted to ASCO shows benefit in about 80% of patients in this small, 14 patient series (116).

REFERENCES 1. Farhood AI, Hajdu SI, Shiu MH, et al. Soft tissue sarcomas of the head and neck in adults. Am J Surg. 1990;160:365–369. 2. Weber RS, Benjamin RS, Peters LJ, et al. Soft tissue sarcomas of the head and neck in adolescents and adults. Am J Surg. 1986;152:386–392. 3. Freedman AM, Reiman HM, Woods JE. Soft-tissue sarcomas of the head and neck. Am J Surg. 1989;158:367–372. 4. Birkmeyer JD, Siewers AE, Finlayson EV, et al. Hospital volume and surgical mortality in the United States. N Engl J Med. 2002;346:1128–1137. 5. Gustafson P, Dreinhofer KE, Rydholm A. Soft tissue sarcoma should be treated at a tumor center: A comparison of quality of surgery in 375 patients. Acta Orthop Scand. 1994;65:47–50. 6. McKenna WG, Barnes MM, Kinsella TJ, et al. Combined modality treatment of adult soft tissue sarcomas of the head and neck. Int J Radiat Oncol Biol Phys. 1987;13:1127–1133. 7. Figueiredo MT, Marques LA, Campos Filho N. Soft-tissue sarcomas of the head and neck in adults and children: Experience

478

8.

9.

10.

11.

12.

13.

14.

15.

16.

17.

18.

19. 20.

21.

22.

23.

24.

25.

26.

Thornton and Benjamin at a single institution with a review of literature. Int J Cancer. 1988;41:198–200. Tran LM, Mark R, Meier R, et al. Sarcomas of the head and neck: Prognostic factors and treatment strategies. Cancer. 1992;70:169–177. Le Vay J, O’Sullivan B, Catton C, et al. An assessment of prognostic factors in soft-tissue sarcoma of the head and neck. Arch Otolaryngol Head Neck Surg. 1994;120:981–986. Gil Z, Patel SG, Singh B, et al. For the International Collaborative Study Group. Analysis of prognostic factors in 146 patients with anterior skull base sarcomas: An international collaborative study. Cancer. 2007;110:1033–1041. Prabhu SS, Diaz E, Sturgis EM, et al. Primary sarcomas of the skull base: An analysis of 63 cases. Clin Neurosurg. 2004;51:340– 342. Gerrand CH, Wunder JS, Kandel RA, et al. Classification of positive margins after resection of soft-tissue sarcoma of the limb predicts the risk of local recurrence. J Bone Joint Surg Br. 2001;83:1149. O’Sullivan B, Gullane P, Irish J, et al. Preoperative radiotherapy for adult head and neck soft tissue sarcoma (STS): Assessment of wound complication rates and cancer outcome in a prospective series. World J Surg. 2003;27:875–883. O’Sullivan B, Davis AM, Turcotte R, et al. Preoperative versus postoperative radiotherapy in soft-tissue sarcoma of the limbs: A randomised trial. Lancet. 2002;359:2235–2241. Demetri GD, Elias AD. Results of single-agent and combination chemotherapy for advanced soft tissue sarcomas: Implications for decision making in the clinic. Hematol Oncol Clin North Am. 1995;9:765–785. Bramwell VH, Anderson D, Charette ML. Doxorubicin-based chemotherapy for the palliative treatment of adult patients with locally advanced or metastatic soft-tissue sarcoma: A metaanalysis and clinical practice guideline. Sarcoma. 2000;4:103– 112. Van Glabbeke M, van Oosterom AT, Oosterhuis JW, et al. Prognostic factors for the outcome of chemotherapy in advanced soft tissue sarcoma: An analysis of 2185 patients treated with anthracycline-containing first-line regimens: A European Organization for Research and Treatment of Cancer Soft Tissue and Bone Sarcoma Group study. J Clin Oncol. 1999;17:150– 157. Benjamin R, Wiernik P, et al. Adriamycin: A new effective agent in the therapy of disseminated sarcomas. Med Pediatr Oncol. 1975;1:63–76. O’Bryan R, Baker L, et al. Dose response evaluation of adriamycin in human neoplasia. Cancer. 1977;39:1940–1948. Mouridsen HT, Bastholt L, Somers R, et al. Adriamycin versus epirubicin in advanced soft tissue sarcomas: A randomized phase II/phase III study of the EORTC Soft Tissue and Bone Sarcoma Group. Eur J Cancer Clin Oncol. 1987;23:1477– 1483. Lopez M, Vici P, Lauro L, et al. Increasing single epirubicin doses in advanced soft tissue sarcomas. J Clin Oncol. 2002;20:1329– 1334. Nielsen OS, Dombernowsky P, Mouridsen H, et al. High-dose epirubicin is not an alternative to standard-dose doxorubicin in the treatment of advanced soft tissue sarcomas: A study of the EORTC Soft Tissue and Bone Sarcoma Group. Br J Cancer. 1998;78:1634–1639. Casper ES, Schwartz GK, Sugarman A, et al. Phase I trial of doseintense liposome-encapsulated doxorubicin in patients with advanced sarcoma. J Clin Oncol. 1997;15:2111–2117. Garcia A, Kempf R, Rogers M, et al. A phase II study of Doxil (liposomal doxorubicin): Lack of activity in poor prognosis soft tissue sarcomas. Ann Oncol. 1998;10:1131–1133. Toma S, Tucci A, Villani G, et al. Liposomal doxorubicin (Caelyx) in advanced pretreated soft tissue sarcomas: A phase II study of the Italian Sarcoma Group (ISG). Anticancer Res. 2000;20:485– 491. Skubitz KM. Phase II trial of pegylated-liposomal doxorubicin (Doxil) in sarcoma. Cancer Invest. 2003;21:167–176.

27. Chidiac T, Budd GT, Pelley R, et al. Phase II trial of liposomal doxorubicin (Doxil) in advanced soft tissue sarcomas. Invest New Drugs. 2000;18:253–259. 28. Antman KH, Montella D, Rosenbaum C, et al. Phase II trial of ifosfamide with mesna in previously treated metastatic sarcoma. Cancer Treat Rep. 1985;69:499–504. 29. Stuart-Harris R, Harper PG, Kaye SB, et al. High-dose ifosfamide by infusion with mesna in advanced soft tissue sarcoma. Cancer Treat Rev. 1983;10(Suppl A):163–164. 30. Benjamin RS, Legha SS, Patel SR, et al. Single-agent ifosfamide studies in sarcomas of soft tissue and bone: The M. D. Anderson experience. Cancer Chemother Pharmacol. 1993;31(Suppl 2):S174–S179. 31. Bramwell VH, Mouridsen HT, Santoro A, et al. Cyclophosphamide versus ifosfamide: Final report of a randomized phase II trial in adult soft tissue sarcomas. Eur J Cancer Clin Oncol. 1987;23:311–321. 32. Elias AD, Eder JP, Shea T, et al. High-dose ifosfamide with mesna uroprotection: A phase I study. J Clin Oncol. 1990;8:170–178. 33. Cinat G, Mickiewicz E, Cabalar M, et al. High dose ifosfamide (HDIFO): A regimen that deserves to be taken into account in previously treated sarcoma bearing patients (Pts). Proc Am Soc Clin Oncol. 2000;19:576a. 34. Deligny N, Demaille M, Bui NB, et al. Phase II trial of high dose of ifosfamide (HDI) in adult patients with advanced soft tissue sarcomas (ASTS). Proc Am Soc Clin Oncol. 2001;20:292b. 35. Le Cesne A, Antoine E, Spielmann M, et al. High-dose ifosfamide: Circumvention of resistance to standard-dose ifosfamide in advanced soft tissue sarcomas. J Clin Oncol. 1995;13:1600–1608. 36. Patel SR, Vadhan-Raj S, Papadopoulos NJ, et al. High-dose ifosfamide in bone and soft tissue sarcomas: Results of phase II and pilot studies—dose-response and schedule dependence. J Clin Oncol. 1997;15:2378–2384. 37. Bui NB, Demaille M, Chevreau C, et al. qMAID vs MAID + 25% with G-CSF in adults with advanced soft tissue sarcoma (STS). First results of a randomized study of the FNCLCC Sarcoma Group. Proc Am Soc Clin Oncol. 1998;17:517a. 38. Frustaci S, Comandone A, Bearz A, et al. Efficacy and tolerability of an ifosfamide continuous infusion (IFO-c.i.) soft tissue sarcoma (STS) patients (pts). Proc Am Soc Clin Oncol. 1998;17:518a. 39. Cerny T, Leyvraz S, von Briel T, et al. Saturable metabolism of continuous high-dose ifosfamide with mesna and GMCSF: A pharmacokinetic study in advanced sarcoma patients. Swiss Group for Clinical Cancer Research (SAKK). Ann Oncol. 1999;10:1087–1094. 40. Gottlieb JA, Baker LH, O’Bryan RM, et al. Adriamycin (NSC123,127) used alone and in combination for soft tissue and bony sarcoma. Cancer Chemother Rep. 1975;6:271–280. 41. Crawford SM, Jerwood D. An assessment of the relative importance of the components of CYVADIC in the treatment of soft-tissue sarcomas using regression meta-analysis. Med Inform (Lond). 1994;19:311–321. 42. Verma S, Bramwell V. Dose-intensive chemotherapy in advanced adult soft tissue sarcoma. Expert Rev Anticancer Ther. 2002;2:201–215. 43. Edmonson JH, Ryan LM, Blum RH, et al. Randomized comparison of doxorubicin alone versus ifosfamide plus doxorubicin or mitomycin, doxorubicin, and cisplatin against advanced soft tissue sarcomas. J Clin Oncol. 1993;11:1269–1275. 44. Santoro A, Tursz T, Mouridsen HT, et al. Doxorubicin versus CYVADIC versus doxorubicin plus ifosfamide in first-line treatment of advanced soft tissue sarcomas: A randomized study of the European Organization for Research and Treatment of Cancer Soft Tissue and Bone Sarcoma Group. J Clin Oncol. 1995;13:1537–1545. 45. Antman K, Crowley J, Balcerzak SP, et al. An Intergroup phase III randomized study of doxorubicin and dacarbazine with or without ifosfamide and mesna in advanced soft tissue and bone sarcomas. J Clin Oncol. 1993;11:1276–1285. 46. Patel S, Vadhan-Raj S, et al. Results of two consecutive trials of dose-intensive chemotherapy with doxorubicin and


47.

48.

49.

50.

51.

52.

53.

54.

55.

56.

57.

58.

59.

60.

61.

62.

63.

64.

65.

ifosfamide in patients with sarcomas. Am J Clin Oncol. 1998;21:317–321. Merimsky O, Meller I, Flusser G, et al. Gemcitabine in soft tissue or bone sarcoma resistant to standard chemotherapy: A phase II study. Cancer Chemother Pharmacol. 2000;45:177–181. Svancarova L, Blay JY, Judson IR, et al. Gemcitabine in advanced adult soft-tissue sarcomas. A phase II study of the EORTC Soft Tissue and Bone Sarcoma Group. Eur J Cancer. 2002;38:556–559. Okuno S, Ryan LM, Edmonson JH, et al. Phase II trial of gemcitabine in patients with advanced sarcomas (E1797): A trial of the Eastern Cooperative Oncology Group. Cancer. 2003;97:1969– 1973. Patel SR, Gandhi V, Jenkins J, et al. Phase II clinical investigation of gemcitabine in advanced soft tissue sarcomas and window evaluation of dose rate on gemcitabine triphosphate accumulation. J Clin Oncol. 2001;19:3483–3489. Talbot SM, Keohan, ML, Hesdorffer, et al. A phase II trial of temozolomide in patients with unresectable or metastatic soft tissue sarcoma. Cancer. 2003;98:1942–1946. Woll PJ, Judson I, Lee SM, et al. Temozolomide in adult patients with advanced soft tissue sarcoma: A phase II study of the EORTC Soft Tissue and Bone Sarcoma Group. Eur J Cancer. 1999;35:410–412. Trent JC, Beach J, Burgess MA, et al. A two-arm phase II study of temozolomide in patients with advance gastrointestinal stromal tumors and other soft tissue sarcomas. Cancer. 2003;98:2693– 2699. Blay JY, Judson I, Rodenhuis S, et al. Phase II study of raltitrexed (Tomudex) for patients with advanced soft tissue sarcomas refractory to doxorubicin-containing regimens. Anticancer Drugs. 1999;10:873–877. De Angelo D, Naujoks R, Manola JB, et al. Phase II study of irinotecan (CPT-11) in relapsed or refractory soft tissue sarcoma (STS). Proc Am Soc Clin Oncol. 2000;19:555a. Carson E, Zalupski M, Redman B, et al. Phase II trial of sargramostim (yeast-derived recombinant human GM-CSF) as monotherapy for advanced sarcomas. Proc Am Soc Clin Oncol. 2000;19:563a. Bramwell VH, Eisenhauer EA, Blackstein M, et al. Phase II study of topotecan (NSC 609 699) in patients with recurrent or metastatic soft tissue sarcoma. Ann Oncol. 1995;6:847–849. Budd GT, Rankin C, Hutchins LF, et al. Phase II trial of topotecan by continuous infusion in patients with advanced soft tissue sarcomas, a SWOG study. Southwest Oncology Group. Invest New Drugs. 2002;20:129–132. Fidias P, Demetri GD, Harmon DC. Navelbine shows activity in previously treated sarcoma patients: Phase II results from MGH/Dana-Farber/Partners Cancer Care Study. Proc Am Soc Clin Oncol. 1998;17:1977. Verweij J, Lee SM, Ruka W, et al. Randomized phase II study of docetaxel versus doxorubicin in first- and second-line chemotherapy for locally advanced or metastatic soft tissue sarcomas in adults: A study of the European Organization for Research and Treatment of Cancer soft tissue and bone sarcoma group. J Clin Oncol. 2000;18:2081–2086. Kostler WJ, Brodowicz T, Attems Y, et al. Docetaxel as rescue medication in anthracycline- and ifosfamide-resistant locally advanced or metastatic soft tissue sarcoma: Results of a phase II trial. Ann Oncol. 2001;12:1281–1288. Edmonson JH, Ebbert LP, Nascimento AG, et al. Phase II study of docetaxel in advanced soft tissue sarcomas. Am J Clin Oncol. 1996;19:574–576. Gian VG, Johnson TJ, Marsh RW, et al. A phase II trial of paclitaxel in the treatment of recurrent or metastatic soft tissue sarcomas or bone sarcomas. J Exp Ther Oncol. 1996;1:186–190. Casper ES, Waltzman RJ, Schwartz GK, et al. Phase II trial of paclitaxel in patients with soft-tissue sarcoma. Cancer Invest. 1998;16:442–446. Delaloge S, Yovine A, Taamma A, et al. Ecteinascidin-743: A marine-derived compound in advanced, pretreated sarcoma patients—preliminary evidence of activity. J Clin Oncol. 2001;19:1248–1255.

479

66. Taamma A, Misset, JL, Riofrio M, et al. Phase I and pharmacokinetic study of ecteinascidin-743, a new marine compound, administered as a 24-hour continuous infusion in patients with solid tumors. J Clin Oncol. 2001;19:1256–1265. 67. George S, Maki RG, Harmon DC, et al. Phase II study of ecteinascidin-743 (ET-743) given by 3-hour IV infusion in patients (pts) with soft tissue sarcomas (STS) failing prior chemotherapies. Proc Am Soc Clin Oncol. 2002;21:408a. 68. Le Cesne A, Blay J-Y, Judson I, et al. Phase II study of ET-743 in advanced soft tissue sarcomas: A European Organisation for the Research and Treatment of Cancer (EORTC) soft tissue and bone sarcoma group trial. J Clin Oncol. 2005;23:576–584. Erratum in: J Clin Oncol. 2005;23:5276. 69. Dileo P, Casali P, Bacci G, et al. Phase II evaluation of 3-hour infusion ET-743 in patients with recurrent sarcomas. Proc Am Soc Clin Oncol. 2002;21:408a. 70. Grosso F, Jones RL, Demetri GD, et al. Efficacy of trabectedin (ecteinascidin-743) in advanced pretreated myxoid liposarcomas: A retrospective study. Lancet Oncol. 2007;8:595– 602. 71. Skubitz KM, Hamdan H, Thompson RC Jr. Ambulatory continuous infusion ifosfamide with oral etoposide in advanced sarcomas. Cancer. 1993;72:2963–2969. 72. Yalcin S, Gullu I, Barista I, et al. Treatment of advanced refractory sarcomas with ifosfamide and etoposide combination chemotherapy. Cancer Invest. 1998;16:297–302. 73. Saeter G, Alvegard TA, Monge OR, et al. Ifosfamide and continuous infusion etoposide in advanced adult soft tissue sarcoma. A Scandinavian Sarcoma Group Phase II Study. Eur J Cancer. 1997;33:1551–1558. 74. Papai Z, Bodoky G, Szanto J, et al. The efficacy of a combination of etoposide, ifosfamide, and cisplatin in the treatment of patients with soft tissue sarcoma. Cancer. 2000;89:177– 180. 75. Hensley ML, Maki R, Venkatraman E, et al. Gemcitabine and docetaxel in patients with unresectable leiomyosarcoma: Results of a phase II trial. J Clin Oncol. 2002;20:2824–2831. 76. Leu K, Ostruszka LJ, Shewach D, et al. Laboratory and clinical evidence of synergistic cytotoxicity of sequential treatment with gemcitabine followed by docetaxel in the treatment of sarcoma. J Clin Oncol. 2004;22:1706–1712. 77. Thall P. Wathen J. Covariate-adjusted adaptive randomization in a sarcoma trial with multi-stage treatments. Stat Med. 2005;24:1947–1964. 78. Maki R, Wathen J, Patel SR, et al. Randomized phase II study of gemcitabine and docetaxel compared with gemcitabine alone in patients with metastatic soft tissue sarcomas: Results of sarcoma alliance for research through collaboration study 002 [corrected]. J Clin Oncol. 2007;25:2755–2763. 79. Rosen G, Forscher C, Lowenbraun S, et al. Synovial sarcoma: Uniform response of metastases to high dose ifosfamide. Cancer. 1994;73:2506–2511. 80. Kampe CE, Rosen G, Eilber F, et al. Synovial sarcoma: A study of intensive chemotherapy in 14 patients with localized disease. Cancer. 1993;72:2161–2169. 81. Nielsen OS, Judson I, van Hoesel Q, et al. Effect of high-dose ifosfamide in advanced soft tissue sarcomas. A multicentre phase II study of the EORTC Soft Tissue and Bone Sarcoma Group. Eur J Cancer. 2000;36:61–67. 82. Edmonson JH, Ryan L, Blum RH. Phase II study of ifosfamide + doxorubicin in patients with advanced synovial sarcomas: (E1793): A Trial of the Eastern Cooperative Oncology Group. Sarcoma. 2003;7:9–11. 83. Demetri GD, Spiegelman BM, Fletcher CDM, et al. Differentiation of liposarcomas in patients treated with the PPAR-g ligand troglitazone: Documentation of biologic activity in myxoid/round cell and pleomorphic subtypes. Proc Am Soc Clin Oncol. 1999;18:535a. 84. Demetri GD, Fletcher CD, Mueller E, et al. Induction of solid tumor differentiation by the peroxisome proliferator-activated receptor-gamma ligand troglitazone in patients with liposarcoma. Proc Natl Acad Sci USA. 1999;96:3951–3956.

480


85. Ferrari A, Dileo P, Casanova M, et al. Rhabdomyosarcoma in adults: A retrospective analysis of 171 patients treated at a single institution. Cancer. 2003;98:571–580. 86. Little DJ, Ballo MT, Zagars GK, et al. Adult rhabdomyosarcoma: Outcome following multimodality treatment. Cancer. 2002;95:377–388. 87. Munzenrider JE, Liebsch NJ. Proton therapy for tumors of the skull base. Strahlenther Onkol. 1999;175(Suppl 2):57–63. 88. Heffelfinger MJ, Dahlin DC, MacCarty CS, et al Chordomas and cartilaginous tumors at the skull base. Cancer. 1973;32:410–420. 89. Perzin KH and Pushparaj N. Nonepithelial tumors of the nasal cavity, paranasal sinuses and nasopharynx. A clinicopathological study. XIV: Chordomas. Cancer. 1986;57:784–796. 90. Harbour JW, Lawton MT, Criscuolo GR, et al. Clivus chordoma: A report of 12 recent cases and review of the literature. Skull base Surg. 1991;1:200–206. 91. Volpe R, Mazabraud A. A clinicopathologic review of 25 cases of chordoma (a pleomorphic and metastasizing neoplasm). Am J Surg Path. 1983;7:161–170. 92. Rich TA, Schiller A, Suit HD, et al. Clinical and pathologic review of 48 cases of chordoma. Cancer. 1985;56:182–187. 93. Mirra J, et al. Chordoma. In: Fletcher CD, Unni K, Mertens F, eds. World Health Organization of Tumours: Pathology and Genetics of Tumours of Soft Tissue and Bone. Lyon, France: IARC Press, 2002:316–317. 94. Meis JM, Raymond AK, Evans HL, et al. Dedifferentiated chordoma. A clinicopathologic and immunohistochemical study of three cases. AM J Surg Path. 1987;11:516–525. 95. Hruban RH, Traganos F, Reuter VE, et al. Chordomas with malignant spindle cell components. A DNA flow cytometric and immunohistochemical study with histogenetic implications. Am J Pathology. 1990;137:435–447. 96. Carpentier A, Polivka M, Blanquet A, et al. Suboccipital and cervical chordomas: The value of aggressive treatment at first presentation of disease. J Neurosurge. 2002;97:1070–1077. 97. Tzortzidis F, Elahi F, Wright D, et al. Patient outcome at longterm follow-up after aggressive microsurgical resection of cranial based chordomas. Neurosurg Online. 2006;59:230–237. 98. Casali PG, Stacchiotti S, Sangalli C, et al. Chordoma. Curr Opin Oncol. 2007;19(4):367–370. 99. Hug EB, Loredo LN, Slater JD, et al. Proton radiation therapy for chordomas and chondrosarcomas of the skull base. J Neurosurg. 1999;91:432–439. 100. Fagundes MA, Hug EB, Liebsch NJ, et al. Radiation therapy for chordomas of the base of the skull and cervical spine: Patterns of failure and outcome after relapse. Int J Radiat Oncol Biol Phys. 1995;33:579–584.

101. Debus J, Hug EB, Liebsch NJ, et al. Brainstem tolerance to conformal radiotherapy of skull base tumors. Int J Radiat Oncol Biol Phys. 1997;39:967–975. 102. Fleming GF, Heimann PS, Stephens JK, et al. Dedifferentiated chordoma. Response to aggressive chemotherapy in two cases. Cancer. 1993;72:714–718. 103. Chugh R, Dunn R, Zalupski MM, et al. Phase II study of 9-nitrocamptothecin in patients with advanced chordoma or soft-tissue sarcoma. JCO. 2005;23;3597–3604. 104. Casali PG, Messina A, Stacchiotti S, et al. Imatinib mesylate in chordoma. Cancer. 2004;101;2086–2097. 105. Casali PG. Imatinib mesylate in 18 advanced chordoma patients [abstract]. J Clin Onc. 2005;23(suppl):9012. 106. Stout AP, Murray MR. Hemangiopericytoma: A vascular tumor featuring Zimmerman’s pericytes. Ann Surg. 1942;116:26– 33. 107. Stout AP. Hemangiopericytoma. A study of 25 new cases. Cancer. 1949;2:1027–1035. 108. Batsakis JG, Rice DH. The pathology of head and neck tumors: Vasoformative tumors, part 9B. Head Neck Surg. 1981;3:326– 340. 109. McMaster MJ, Soule EH, Ivins JC. Hemangiopericytoma. A clinicopathologic study and long-term follow-up of 60 patients. Cancer. 1975;36:2232–2244. 110. Soyuer S, Chang EL, Selek U, et al. Intracranial meningeal hemangiopericytoma: The role of radiotherapy: Report of 29 cases and review of the literature. Cancer. 2004;100;1491–1497. 111. Dufour H, Métellus P, Fuentes S, et al. Meningeal hemangiopericytoma: A retrospective study of 21 patients with special review of postoperative external radiotherapy. Neurosurgery. 2001;48:756–762. 112. Combs S, et al. Precision radiotherapy for hemangiopericytomas of the central nervous system. Cancer. 2005;104:2457– 2465. 113. Beadle GF, Hillcoat BL. Treatment of advanced malignant hemangiopericytoma with combination adriamycin and DTIC: A report of four cases. J Surg Onc. 1983;22;167–170. 114. Wong PP, Yagoda A. Chemotherapy of malignant hemangiopericytoma. Cancer. 1978;41:1256–1260. 115. Kirn DH, Kramer A. Long-term freedom from disease progression with interferon alpha therapy in two patients with malignant hemangiopericytoma. J Natl Cancer Inst. 1996;88:764– 765. 116. Park MS, Patel SR, Ludwig JA, et al. Combination therapy with temozolomide and bevacizumab in the treatment of hemangiopericytoma and malignant solitary fibrous tumor. J Clin Oncol. 26: 2008 (May 20 suppl; abstr 10512).

34 Angiofibromas and Vascular Tumors of the Skull Base Andrew G. Sikora and Randal S. Weber

that the majority of putative JNA occurring in nonadolescent males, and sporadic reports of JNA in females, is misdiagnosed. JNA is thought to be more common in Egypt and India, and possibly in other areas of South Asia, as well as in Kenya (3). Of note, this geographic distribution does not correspond to the distribution of areas where nasopharyngeal carcinoma is endemic, and does not seem to provide clues to genetic or environmental factors predisposing to JNA. Unlike JNA, HPC affects males and females with equal frequency and usually presents after the second decade (4,5). Hemangiomas in general have a female-to-male ratio of 2– 4:1 and usually present in the first year of life (6). Vascular malformations have a slight female predominance. The rarity of each of these lesions makes it unclear whether these trends are as true for lesions limited to the skull base as at other body sites.

INTRODUCTION Vascular tumors of the skull base are both rare and diverse. Despite the heterogeneity of lesions presenting in this area, they share common principles of management and surgical approach, including reliance on radiology to establish diagnosis, surgery as the primary treatment, and a vital role for angiography as both diagnostic and adjunctive therapeutic modality. With the exception of the malignant subtype of hemangiopericytoma (HPC), these tumors are generally benign but locally aggressive due to their proximity to vital structures and potential for intracranial spread. The anatomic and functional complexity of this region mandates a collaborative, multidisciplinary approach, which may require the participation of otolaryngology/head and neck surgery, neurosurgery, neuroradiology, interventional radiology, reconstructive surgery, ophthalmology, speech and swallowing therapeutics, and other teams. In the present chapter, we seek to provide an overview of the multidisciplinary management of vascular tumors of the base of skull, written from the skull base surgeon’s perspective. To provide a paradigm for management of these lesions, we describe in detail the evaluation and management of juvenile nasopharyngeal angiofibroma (JNA), the most common vascular tumor presenting in this region. Where appropriate we discuss other tumor types separately including hemangiomas, vascular malformations, HPC, and malignant HPC.

PATHOLOGY Juvenile Nasopharyngeal Angiofibroma Etiology and Pathogenesis Although most patients diagnosed with JNA undergo treatment, its natural history is thought to be androgen-driven growth during puberty followed by regression. Support for the concept that JNA can involute after puberty is provided by several case reports of lesions which underwent radiographically documented regression during observation or after incomplete excision (7,8). Despite the tantalizing clue provided by the nearexclusive occurrence of JNA in adolescent boys, and numerous theories proposed to explain its pathogenesis, the etiology of JNA is still unclear. Earlier histologists believed that JNA was primarily a fibrous tumor, which led to theories that JNA arose from hypertrophy of skull base periosteum or fascia, or inappropriate persistence of fibrocartilage rests. Later, appreciation of the predominantly vascular nature of the tumor, and similarity to erectile tissue, caused researchers to speculate that JNA results from misplaced inferior turbinate tissue, or other vascular tissue. The nature of the lesion— neoplasm or hamartoma—has also been a source of controversy, although a recent study that demonstrated frequent beta-catenin mutations in JNA specimens lends support to the idea that JNA is a clonal, neoplastic process (9). The only clue to genetic predisposition of JNA comes from patients with familial adenomatous polyposis (FAP), who have up to a 25-fold increase in the frequency of JNA (10–13). Since genes in the adenomatous polyposis coli (APC) gene pathway are commonly mutated in FAP, and APC regulates binding and degradation of beta-catenin, this lends additional support to the involvement of beta-catenin in pathogenesis of JNA. Of note, the study (9) which found increased

EPIDEMIOLOGY AND INCIDENCE Incidence Vascular lesions of the skull base are rare; the most common lesion, JNA, makes up less than 0.05% of all head and neck tumors (1). HPC is thought to account for less than 1% of all vascular tumors at all body sites (1). From 9% to 28% of HPCs have been estimated to occur in the head and neck, with sinonasal presentation being most common (1). The rarity of other vascular skull base lesions makes it difficult to estimate their incidence in this location. Paragangliomas are described in detail in a separate chapter of this book.

Epidemiology The epidemiology of JNA has proven fascinating to clinicians and researchers, who seek clues to its etiology in its demographic characteristics. JNA almost exclusively affects adolescent boys, and is seldom observed in young adults—this has led to speculation that the tumor develops in response to the hormonal milieu occurring during male puberty (see below) and regresses upon reaching adulthood. The average age of patients with JNA is 14 to 18 years, although the reported range is much broader (7–29 years) (2). It is suspected 481

482

Sikora and Weber

mutations of beta-catenin in JNA specimens examined tumors from patients with sporadic, not FAP-associated disease. Beta-catenin is a cytoplasmic protein which is involved in cell–cell adhesion, and also plays a role in oncogenic signal transduction—however, while the association between beta-catenin pathway mutations and JNA deserves further study, it does not yet provide a coherent picture of JNA pathogenesis. The peculiar association of JNA and adolescent males leaves little doubt that whatever the genetic/developmental events leading to susceptibility to JNA, its progression is under hormonal control. This has led to substantial interest in the staining of JNA specimens for receptors for androgenic and other hormones. The clinical behavior of JNA is quite consistent with androgen-dependent growth, and androgen receptors are frequently found on JNA specimens, localized to both stromal and endothelial cells (14,15). Staining of other hormonal receptors is less consistent, with positive staining for progesterone receptors observed sometimes, and staining for estrogen receptors observed rarely if at all (16,17). The geographical variation in JNA incidence and occurrence of JNA in some—but not all—FAP patients has led to interest in environmental factors which may contribute to JNA development. Proposed environmental factors, such as environmental toxins, trauma, dry air, infectious agents, allergy, and other causes of inflammation, are supported by little or no published data.

(A)

Anatomy Although established JNA may have broadly based or multiple attachments to the nasal cavity, it is thought to originate on the posterolateral wall of the nasal cavity, where the root of the pterygoid process, horizontal ala of the vomer, and sphenoidal process of the palatine bone meet (superior aspect of the sphenopalatine foramen) (18,19). This location has great significance because it allows JNA to grow into the nasal cavity, nasopharynx, or pterygopalatine fossa (PPF), allowing further spread to vital areas such as the infratemporal fossa, cranial cavity, and orbit. A slightly different origin has been suggested by Lloyd and colleagues, who propose that JNA originates “in the pterygopalatine fossa in the recess behind the sphenopalatine ganglion, at the anterior aperture of the pterygoid canal” based on imaging characteristics in a series of 72 patients (20). Despite the number of anatomical regions potentially accessible to JNA, it tends to spread in an orderly fashion via one of several pathways (Fig. 1). 1. Anteriorly into the nasal cavity, causing nasal obstruction. 2. Posteriorly into the nasopharynx, where it can access the sphenoid sinus, and may displace the palate downward or even protrude into the oral cavity. 3. Laterally via the sphenopalatine foramen or erosion of the posterior maxillary sinus wall into the pterygopalatine fossa; from there it may spread through the pterygomaxillary fissure to involve the infratemporal and even temporal fossae. 4. Superiorly into the orbit via the inferior orbital fissure. JNA can choose any of these pathways, and can spread in multiple directions simultaneously. The pathways of spread involved determine the presenting symptoms, and have great implications for choice of surgical approach. Continued spread can lead to the most challenging management aspects of JNA, intracranial involvement, via one of several well-defined pathways.

(B)

Figure 1 Routes of invasion of angiofibroma. (A) Sagittal view demonstrating spread of angiofibroma anteriorly into the nasal cavity (1), posteriorly into the nasopharynx (2), and intracranially via the sphenoid sinus (3). (B) Axial view, demonstrating spread of angiofibroma anteriorly into the nasal cavity (1), posteriorly into the nasopharynx (2), and laterally (3) toward the infratemporal and temporal fossae (3a).

1. Via the pterygopalatine fossa/infratemporal fossa, by eroding the bone of the anterior face of the greater wing of the sphenoid. Entry occurs through a region demarcated by the foramen rotundum, foramen ovale, and foramen lacerum. 2. Via the superior orbital fissure. 3. Superiorly, through the roof of the sphenoid sinus. Pathways #1 and #2 are lateral pathways, which result in extension of JNA lateral to the carotid artery and cavernous sinus. Pathway #3, the medial pathway, is less common but results in tumor spread to a much less favorable location medial to the carotid and cavernous sinus. This allows tumor to infiltrate the pituitary and optic chiasm, and makes surgical removal much more difficult. Direct spread superiorly through the cribriform plate is possible, but is extremely rare. The blood supply to the tumor is variable, but in most cases the main blood supply originates from the ipsilateral internal maxillary artery (IMA) (21). As the tumor grows, other vessels can become parasitized, including other branches of the external carotid system such as the sphenopalatine and ascending pharyngeal arteries, as well as branches of the vertebral and internal carotid arteries, such

Chapter 34: Angiofibromas and Vascular Tumors of the Skull Base

as the mandibulovidian artery which comes off the internal carotid artery (ICA). In each case, the tumor can recruit vessels from either the ipsilateral or contralateral side, mandating bilateral angiographic evaluation (see below).

Pathology The gross appearance of JNA is of a firm/spongy lesion, which is pinkish where covered by mucosa and gray or whitish where not. While the tissue lacks a true capsule, it usually has a well-defined pseudocapsule and is sharply demarcated from the surrounding tissue (22). Microscopically, JNA is composed of dense fibrous stroma, punctuated by numerous blood vessels of varying size and shape, which may have a “staghorn” appearance or be slit-like (Fig. 2) (22,23). The fibrous stroma is composed of plump/polygonal fibroblasts with round or vesicular nuclei and abundant connective tissue. The vessels are delicate and consist of a single

Table 1

483

Mulliken and Glowacki Classification of Vascular Lesions

Hemangioma Capillary Cavernous Combined

Vascular malformation “Low flow” Venous Capillary Lymphatic

“High flow” Arteriovenous Arterial A-V fistula

Source: Adapted from Ref. 6.

layer of endothelial cells without an elastic or smooth muscle layer—hence the propensity for these lesions toward massive, uncontrollable bleeding (22). Other fibrovascular neoplasms from which JNA must be differentiated include solitary fibrous tumor and HPC. In the case of solitary fibrous tumor, although the vessels may have a similar appearance, the stromal cells will appear more spindle-shaped with elongated nuclei (24,25). In HPC, the stromal cells may appear similar, but the vessels tend to be round and staghorn-shaped vessels are less common (26).

Hemangiopericytoma

(A)

HPCs are thought to arise from the vascular pericytes of Zimmerman, mesenchymal cells which line blood vessels and regulate blood flow and vessel contraction (27). Little is known about their etiology and pathogenesis, and some pathologists question the validity of HPC as a distinct pathologic entity (28). HPC can present as a spectrum of benign, malignant, and borderline phenotypes, although it has been suggested that HPC of the sinonasal area is more likely to be benign than at other sites (29). Lymphatic dissemination rarely occurs, but hematologic metastasis to the lungs, liver, and bone can occur. The natural history of these tumors is variable, with even histologically benign-appearing tumors sometimes becoming locally aggressive and metastasizing (30). Grossly, HPC can have a variable appearance: soft, firm, rubbery, or polypoid, and it can be tan, gray, or offwhite in color. Unlike JNA, HPC is sometimes mistaken for nasal polyps. The histological appearance is also variable, and can feature small, tightly packed cells with sparse cytoplasm and vesicular nuclei, although cells can also be spindle shaped (31). As is the case with JNA, the stroma is punctuated with thin, fragile vessels, which may occasionally take on the staghorn shape associated with that lesion. Immunostaining for smooth muscle actin is often observed and supports a pericystic origin for these tumors (31). Although solitary fibrous tumor is more likely to have a benign course than HPC, it can be difficult to differentiate these tumors (28); CD34 has been used for this purpose because solitary fibrous tumors tend to stain intensely, whereas staining of HPC tends to be weaker and more focal (24).

Hemangiomas and Vascular Malformations (B)

Figure 2 Gross and microscopic histological appearance of angiofibroma. (A) Gross specimen. Note lobulation and color variation from light tan to hemorrhagic appearance. (B) Characteristic histopathological appearance (original magnification 100X) of angiofibroma demonstrating thin, lymphaticlike endothelial channels without muscular or connective tissue layers. Inset is a high power view (original magnification 400X) demonstrating stromal cells with plump, ovoid nuclei. Source: Image courtesy of Dr. Michelle Williams, UT MD Anderson Cancer Center.

While classification of vascular anomalies can be controversial, the system of Mulikin and Glowaki is commonly accepted Table 1 (6). This classification distinguishes between hemangiomas and vascular malformations, which are further subdivided into high-flow (arteriovenous malformation) and low-flow (various lymphatic, venous, and capillary malformations) lesions. These distinctions are important because the natural histories of hemangioma and vascular malformation differ—hemangiomas undergo intense proliferation during the first months/years of life, followed by gradual involution, which may be partial or complete. They rarely involve bone. Vascular malformations develop in utero, and subsequently

484

Sikora and Weber

Table 2 Radkowski Modification of the Sessions Classification of JNA

Table 3

Tumor stage

Classification

IA

Tumor limited to posterior nares and/or nasopharyngeal vault Tumor involving the posterior nares and/or nasopharyngeal vault with involvement of at least 1 paranasal sinus Minimal lateral extension into the pterygomaxillary fossa Full occupation of the pterygomaxillary fossa with or without superior erosion of the orbital bones Extension into the infratemporal fossa or extension posterior to the pterygoid plates Erosion of the base of skull (middle cranial fossa/base of pterygoids)—minimal intracranial extension Extensive intracranial extension with or without extension into the cavernous sinus

A. Sessions IA Tumor limited to nose and/or nasopharyngeal vault IA Extension into 1 or more paranasal sinuses IIA Minimal extension into pterygomaxillary fissure IIB Full occupation of pterygomaxillary fissure with or without erosion of orbital bones IIC Extension to infratemporal fossa +/− involvement of cheek III Intracranial extension

IB

IIA IIB IIC IIIA IIIB

Source: Adapted from Ref. 36.

Other Classification Systems

B. Fisch I Tumor limited to nasal cavity and/or nasopharynx with no bone destruction II Tumor invading pterygomaxillary fissure or paranasal sinuses with bony destruction III Tumor invading infratemporal fossa, orbit, and/or parasellar region remaining lateral to cavernous sinus IV Tumor invading cavernous sinus, optic chiasm, and/or pituitary fossa Source: Adapted from Refs. 37 and 82.

grow in proportion to the growth of the child, and do not involute. They are more likely to involve bone. Hemangiomas of the skull base are rare, with the orbital apex being the most common site; lesions in this area can cause loss of vision by compression of the optic nerve or affect other structures passing through the superior orbital fissure. Conversely, true arteriovenous malformations are rarely limited to the orbit, and commonly arise intracranially. The etiology of hemangiomas and vascular malformations is not well understood. Hemangiomas are thought to be neoplasms which undergo a postnatal proliferation in response to abnormally regulated mediators of growth, angiogenesis, or inflammation, such as vascular endothelial growth factor, basic fibroblast growth factor, and transforming growth factor–beta (TGF-β) (32). Although some hemangiomas seem to arise in response to trauma, the majority are congenital. Vascular malformations are thought to be anomalies of vascular development, possibly due to defects in the ability of fetal mesenchyme to form endothelium and supporting structures beginning at 8 weeks of gestation. Most are sporadic, although some syndromes are associated with vascular malformations; Sturge-Weber syndrome (33) and von Hippel Lindau (34) syndrome have the potential for skull base involvement due to the predisposition toward vascular malformations in the distribution of the trigeminal and optic nerve or retina, respectively. Histopathological appearance of hemangiomas is diverse, with capillary, cavernous, and mixed types recognized (6). The vascular wall is similar to that of normal vessels and consists of mature endothelium. Mast cell infiltration may be prominent, especially during the involution phase. In vascular malformations, the vessels are dilated and ectatic, and consist of a single endothelial cell layer surrounded by a markedly attenuated muscular layer (35).

STAGING Numerous staging systems have been developed for JNA Tables 2 and 3. The most recently developed system in wide usage is the modification, by Radkowski and colleagues (36), of the staging system developed by Sessions and colleagues Table 2 (37). The Sessions classification reflects the proclivity of JNA to extend into the pterygopalatine fissure and infratemporal fossa, and it also reflects the prognostic impact of invasion of these areas and intracranial extension.

The Radkowski modification reflects the principle that progressively higher levels of involvement of the skull base lead to a greater chance of recurrence, and that extension posterior to the pterygoid plates into the medial and lateral pterygoid muscles complicates surgical excision. The Radkowski modification also differentiates between patients with skull base erosion or minimal intracranial spread and patients with extensive intracranial involvement, since this has implications for both management and prognosis.

CLINICAL ASPECTS Symptoms Symptoms of vascular skull base lesions are determined primarily by the anatomical location and extent of the lesion. In general, presentation at early stages is due to symptoms of nasal obstruction and episodic epistaxis. Since these symptoms are very nonspecific, particularly in the pediatric age group, diagnosis is often delayed and most patients have been symptomatic for months before referral to an otolaryngologist or skull base surgeon. While epistaxis is most commonly periodic and self-limited, the thin, fragile vessels of JNA lack a muscular layer, and thus cannot vasoconstrict in response to hemorrhage. Thus massive, uncontrollable nosebleeds can occur spontaneously or in response to seemingly minor trauma. Extension of JNA (and other lesions) within the nasopharynx and nasal cavity can lead to hyponasal speech; obstruction of the Eustachian tube with otalgia, effusion, and conductive hearing loss; mouth breathing, snoring, and obstructive sleep apnea; and contralateral nasal obstruction from bowing of the nasal septum to the opposite side. Prolonged obstruction can lead to anosmia, mucopurulent nasal discharge, and obstructive sinusitis. As these lesions progress, they gradually erode the skull base leading to invasion of critical regions such as the pterygopalatine and infratemporal fossae, the orbit, the optic chiasm, and sella turcica, leading to upper cranial nerve deficits and threatening vision (see description of progressive spread of JNA above). In this case, symptoms correspond to the area of invasion. Invasion of the orbit can lead to exophthalmos, visual compromise via compression of the optic nerve, or diplopia, which may be due to direct muscle impingement or involvement of the extraocular muscles. Characteristic deficits are produced by involvement of the superior


485

tion to perform a transnasal biopsy in the office, since severe bleeding may result. The diagnosis of these lesions is usually suggested by the clinical history and physical examination, and is confirmed radiographically.

Radiology

Figure 3 Typical appearance of adolescent male with JNA, which has spread laterally to involve the cheek.

orbital fissure (III, IV, V1, and VI); these deficits plus involvement of the optic nerve suggest spread to the orbital apex, and deficit of V2 suggests involvement of the cavernous sinus (38). Involvement of the pterygopalatine and infratemporal fossae may produce midfacial/dental anesthesia, headache, and excessive lacrimation. Involvement of the temporal fossa can result in trismus, or difficulty in chewing. Spread of tumor outside the confines of the skull base can lead to a visible or palpable swelling of the cheek, temple, or intraorally (Fig. 3). It should be emphasized that considerable infiltration of these anatomic regions can occur before becoming symptomatic, so clinicians must have a high index of suspicion when evaluating patients with nasal obstruction, epistaxis, and cranial nerve deficits. Continued erosion of the skull base leads to intracranial extension, with the potential for symptoms related to involvement of the cavernous sinus (see above), dura, and brain parenchyma including headache, seizures, central neurological deficits, and cranial nerve deficits caused by traction and tenting of the nerve by tumor.

Physical Examination A thorough head and neck evaluation, including meticulous documentation of cranial nerve status and nasopharyngoscopy, should be performed. Splaying of the nasal bones or facial contour deformity may be observed in long-standing cases. Anterior rhinoscopy may reveal bowing of the septum to the opposite side, nasal obstruction, and nasal discharge, which may be mucoid or mucopurulent. Examination of the posterior nasal cavity and nasopharynx with a rigid or flexible scope may display a pink or reddish mass filling these areas. If a vascular mass is suspected (e.g., in an adolescent male with nasal obstruction), one should avoid the tempta-

Imaging is of vital importance in the workup of vascular skull base lesions to confirm the suspected diagnosis, to determine the extent of disease and structures involved, and to serve as a road map for surgical planning (Figs. 4 and 5). CT is the cornerstone of imaging these lesions because it provides excellent resolution of the structures involved and extent of bony invasion. MRI is useful for evaluating extension to soft tissue, and particularly for evaluating cavernous sinus involvement and the extent of intracranial disease. MRI is also excellent for distinguishing invasion of paranasal sinuses by tumor from postobstructive opacification. The anatomic origin of JNA at the sphenopalatine foramen and pterygopalatine fossa leads to two consistent diagnostic features on axial CT scan: a mass involving the nasal cavity and pterygopalatine fossa, and bony erosion behind the sphenopalatine foramen at the root of the medial pterygoid plate (38). These features were seen in every patient in the series of Lloyd and colleagues (38); other areas of the skull base (e.g., nasopharynx, infratemporal fossa) are sometimes, but not always, involved. Another consistent feature is expansion of the pterygoid (vidian) canal, associated with bony erosion of the pterygoid process. Other findings include expansion of the orbital fissure and erosion of the maxillary sinus or basisphenoid. The antral bowing sign described by Holman and Miller (originally described for plain skull films, but better seen on axial CT), in which the posterior wall of the maxillary antrum is displaced forward, can be seen in slow-growing lesions other than angiofibroma, and is not pathognomonic. On CT, angiofibromas avidly enhance after contrast bolus. On MRI, the lesion may be hypo- to isointense on T1 sequences, hyperintense on T2, and enhances on T1 sequence with gadolinium contrast; signal voids consistent with hypervascularity may be seen (39). Imaging plays a vital role in the staging and preoperative planning of JNA. It is important to identify findings which predict a greater risk of recurrence and more difficult surgical resection, including involvement of the anterior fossa, sphenoid, pterygoid muscles or base of the pterygoid plates, or foramen lacerum (39). After treatment, imaging is used to enhance surveillance for recurrence, since tumor regrowth can be quite extensive before becoming clinically detectable. Imaging is also important in the evaluation of HPC, hemangioma and vascular malformations, to confirm the extent of disease, to allow planning for surgery, and to follow for recurrence. These lesions can have similar imaging characteristics (40). They tend to be isointense to muscle on CT with avid contrast enhancement. On MRI, they are iso- to hyperintense to muscle on T1-weighted sequences, hyperintense on T2-weighted sequences, and enhance markedly with gadolinium contrast. Hemangiomas often show a serpentine pattern of vascular structure, and may have either infiltrative or well-demarcated margins (41).

Angiography Angiography plays a role in both the diagnosis and the treatment of JNA, as preoperative embolization is widely recommended to reduce intraoperative hemorrhage and has even been proposed as definitive therapy. Angiography confirms the vascular nature of the tumor and allows delineation of

486

Sikora and Weber

A

B

Figure 4 Axial contrast-enhanced CT scan of the paranasal sinuses windowed for soft tissue (A) and (B) bone. The angiofibroma is seen as a contrast-avid mass, which occupies the right posterior nasal vault with minimal lateral extension into the pterygopalatine fossa, and is thus staged as a Radkowski IIA. Note invasion of the right sphenoid bone with widening of the right vidian canal in panel B (arrow). This lesion would be amenable to a maxillotomy approach supplemented with endoscopic visualization, with use of the high-speed drill to extirpate disease from the area of the vidian canal. Source: Images courtesy of Laurence Ginsberg, UT MDACC Cancer Center.

the vascular anatomy and blood supply. Generally, the tumor manifests as an intense and homogenous blush on angiography (Fig. 6). As described above, a branch of the external carotid system (usually the ipsilateral IMA) is the primary blood supply, but as the tumor grows, it can establish connections with ispsilateral or contralateral branches of both the external and internal carotid systems or vertebral arteries (39). Therefore, bilateral examination of the internal and external carotids and vertebral arteries should be performed in all cases. Embolization of feeding vessels has been recommended to reduce intraoperative blood loss and found to be effective in a number of studies (42–44). Embolization is typically performed transarterially, although direct puncture embolization of the lesion has also been described, and may be useful in cases where intravascular access to the tumor is difficult or likely to result in spillage of the emoblization particles into the internal carotid system with threat to the central nervous system or ophthalmic artery (45,46). The optimal time of embolization is thought to be 24 to 48 hours prior to surgery to allow for maximal devascularization without formation of collateral vessels. Some authors question the effectiveness of embolization since not all series show a reduction in intraoperative blood loss (47,48), and it has been proposed that postembolization tumor shrinkage can increase the risk of incomplete resection, especially where there is deep invasion of the sphenoid (20). However, we feel that for larger lesions, and in cases where a difficult or lengthy resection is anticipated, the benefits of embolization outweigh the potential disadvantages. Although embolization has been proposed as single therapeutic modality for selected JNA, most surgeons feel that it is better used in combination with definitive surgery. Balloon occlusion testing should be considered for patients with lesions involving the cavernous sinus where sacrifice of or injury to an internal carotid artery is a significant possibility. Angiography can also be helpful in the workup and management of other vascular lesions. In the case of hemangioma, angiography allows discrimination and vascu-

lar mapping of high-flow lesions (which may benefit from embolization) from low-flow lesions (which usually do not) (49,50). Angiography can also provide valuable pretreatment information about vascular malformations and HPCs; here too, preoperative embolization may be indicated for larger lesions. For lesions where reconstruction with a local flap (such as the temporalis muscle flap) is required, this fact must be communicated to the interventional radiologist to ensure that vascular compromise is avoided.

Other Considerations The issue of pretreatment biopsy for these vascular lesions is controversial. In the majority of cases, the clinical history and imaging findings provide adequate information for diagnosis and planning, and biopsy can be deferred to the time of definitive surgery. Even in the case of HPC, which has both benign and malignant presentations, histopathology does not readily distinguish between the two, and the diagnosis of malignancy is made on primarily clinical grounds. Thus, most clinicians feel that the risk of severe hemorrhage outweighs the potential benefits of pretreatment biopsy. However, others argue that the risk of uncontrollable bleeding has been overstated and that biopsy may provide valuable planning information, particularly in cases where the diagnosis is in question and other entities on the differential diagnosis may benefit from a change of treatment plan (51,52). In those cases, biopsy may be most prudently performed as a separate operative procedure, with definitive treatment deferred until permanent histopathology becomes available, or biopsy with intraoperative pathological consultation obtained at the time of resection. Since the possibility of considerable intraoperative hemorrhage can be anticipated for all these lesions, particular attention must be given to preoperative hematologic workup including hemoglobin/hematocrit, coagulation labs, and bleeding time. Strict instructions should be given to avoid non-steroidal anti-inflammatory drugs (NSAIDS) and other blood-thinners for 2 weeks prior to resection, including vitamin E, and herbal preparations containing ginko, ginseng, or


A

B

C

D

487

Figure 5 Advanced JNA with intracranial extension. Contrast-enhanced CT scan (A–C) and axial T1-weighted MRI with contrast (D). (A) Coronal view. Note invasion of sphenoid sinus, and intracranial penetration (arrow). Although tumor has entered the cranial cavity, it remains extradural (B–C). Axial CT scan viewed in soft tissue (B) and bone (C) windows shows massive invasion of the pterygopalatine fossa with lateral extension to the infratemporal fossa, extensive erosion of the skull base, and pronounced widening of the sphenopalatine foramen (arrows). (D) MRI at approximately the same level highlights the extent of lateral invasion, and pushing, rather than infiltrative, borders. This tumor would be staged as a Radkowski IIIB, and would require a combined head and neck/neurosurgical procedure, including facial translocation, infratemporal fossa, and subtemporal craniectomy approaches. Source: Images courtesy of Laurence Ginsberg, UT MDACC Cancer Center.

garlic. Type-and-cross should be obtained before surgery, and the availability of adequate units of blood should be verified; hypotensive anesthesia is commonly used to minimize blood loss. Autologous blood banking or cell-saver autotransfusion can be considered for large lesions or patients who are likely to poorly tolerate blood loss.

TREATMENT Choice of Treatment Selection of treatment and pretreatment plannings is best accomplished by the surgeon in consultation with a multidisciplinary team experienced in the management of these lesions. In the vast majority of cases, surgery is the preferred management of JNA and other vascular lesions of the skull base. While several series have suggested that primary radiation therapy can achieve control rates comparable to surgery

for JNA, involution of the lesion occurs over the course of months or even years. Radiation therapy also raises the concerns of secondary malignancy (53,54) and alterations of craniofacial growth (55) in this relatively young population. Although extensive intracranial involvement has been argued as an indication for radiation therapy, JNA generally respects the dura and can usually be separated from the intracranial contents without great difficulty. We recommend considering radiation as a primary modality only for patients with tumors that are deemed unresectable or resectable only with great morbidity due to involvement of the cavernous sinus or pituitary, internal carotid, optic chiasm, or optic nerve; or patients who are medically unfit for surgery. Other indications include treatment of persistent or recurrent disease in selected patients. Other therapies, including chemotherapy, hormonal therapy, and definitive embolization, are for the most part unsuitable as frontline therapy, and should be considered

488

Sikora and Weber

A

B

Figure 6 Angiogram of angiofibroma pre- and postembolization. (A) Pre-embolization: the angiofibroma’s primary blood supply is from the internal maxillary artery. (B) Postembolization: there is considerable reduction of tumor vascularity.

experimental. Observation is generally reserved for those unlikely to withstand treatment, since these lesions are unlikely to regress, or (in the case of JNA) unlikely to regress before causing complications.

Surgery Surgical Approaches The goal of surgery is complete removal of the lesion without injury to brain, optic nerve, and other important structures and with minimal hemorrhage. The tendency for these lesions to infiltrate complex and relatively inaccessible anatomical spaces traversed by important neurovascular structures means that failure to meticulously review imaging and to develop an individualized surgical plan virtually guarantees residual/recurrent disease and increased risk for complications. Therefore, there is no single best approach to these lesions; rather, the surgical approach or combination of approaches should be tailored to the anatomical location of the lesion, and surgeons must be comfortable with a wide range of endoscopic and open approaches. An outline for choosing the route of surgical access based on location of the lesion is presented in Table 4. Further information about each approach as it relates to vascular lesions of the skull base is given below; technical details of theses techniques are provided elsewhere in this volume.

Endoscopic Approach Although the appropriateness of endoscopic surgery for JNA has been considered controversial in the past, numerous case series (56–58) as well as growing endoscopic experience with other benign and malignant tumors have suggested that for selected lesions this is a valid approach. The controversy now lies in the extent of disease amenable to endoscopic surgery, with some advocating its use only in lesions limited to the nasal cavity, nasopharynx, and paranasal sinuses, and oth-

ers suggesting that it is appropriate for various degrees of invasion of the pterygopalatine and infratemporal fossae, or even limited intracranial disease (59,60). Advocates of a more aggressive approach to endoscopic resection of JNA cite the ability to closely inspect the resected bed, the generally noninfiltrative nature of the lesion, opportunity to ligate the ipsilateral internal maxillary and sphenopalatine arteries early in the procedure, avoidance of the hemorrhage incurred by approaches requiring osteotomy and extensive soft tissue dissection, and decreased total blood loss as justifications for this approach. Endoscopic approaches also have the obvious advantages of avoiding a facial scar, and avoiding disruption of facial bones and soft tissues, particularly in adolescents who are still undergoing facial growth. We feel that exclusively endoscopic approaches are appropriate for lesions involving the nasal cavity and nasopharynx with limited extension into the pterygomaxillary fissure, so long as the surgeon is already experienced in the techniques of endoscopic sinonasal surgery. Endoscopic approaches may also be used in conjunction with more laterally based open approaches when a lesion requires both anterior and lateral approaches for complete resection.

Transpalatal, Maxillectomy, and Facial Translocation Approaches The transpalatal approach provides potentially good access to medial structures such as the nasopharynx, nasal cavity, and sphenoid; however, it is difficult to get lateral exposure. This procedure also carries the risk of palatal fistula. Although the transpalatal approach was once commonly used for limited JNA, endoscopic techniques provide excellent access to the same areas and in many institutions have replaced the transpalatal approach. Medial maxillectomy, performed with a lateral rhinotomy and with Weber-Ferguson or midfacial degloving


489

Table 4 Choice of Surgical Access Based on Location of the Lesion Nasal Cavity Transpalatal Nasopharynx | Sphenoid Ethmoid Limited Pterygopalatine Foramen Extensive Pterygopalatine Foramen Limited Infratemporal Fossa Medial Cavernous Sinus Extensive Infratemporal Fossa Lateral Cavernous Sinus Middle Cranial Fossa Anterior Cranial Fossa

| Endoscopic | |

| Infratemporal Fossa Approach | Frontal Craniotomy

incision, provides better access to lateral structures with the potential to reach as far as the infratemporal fossa (ITF). Facial translocation approaches, which allow lateral extension of the exposure afforded by maxillectomy, allow improved access to the ITF and the lateral cavernous sinus (Fig. 7). In general, maxillectomy/transfacial approaches will provide adequate access to the majority of tumors without intracranial involvement. When tumor extends intracranially, the majority of tumors can be resected by combining a transfacial approach with craniotomy (61). Although there are numerous potential complications (such as hemorrhage, injury to the nasolacrimal duct, cheek numbness, and injury to the orbital contents), these procedures are generally well tolerated and significant complications are infrequent. Transfacial approaches share the theoretical concern that disruption of facial development may lead to facial asymmetry and retardation of facial growth in children and adolescents. However, several recent series of craniofacial procedures in children did not find evidence that this was the case (62,63), observations consistent with the experience of Randal S. Weber, the senior author of this article. Thus, while disruption of the facial skeleton should be avoided in young patients where possible, adequate exposure should not be compromised.

Infratemporal Fossa Approach The ITF approaches have been proposed to be versatile procedures for the treatment of extensive JNA, since they allow wide exposure of disease in the lateral reaches of the PPF and ITF, wide exposure of the ICA and cavernous sinus, and exposure of the middle cranial fossa, as well as adequate access to the nasal cavity, nasopharynx, and paranasal sinuses. However, tumor medial to the abducens nerve (VI) is not accessible by this technique, and is usually addressed by postoperative radiation. The type D approach is used for tumors which are not in proximity to the internal carotid artery, whereas the type C approach is used for tumors with medial extension into the ICA or cavernous sinus [reviewed in (64)]. The ITF approaches are extremely morbid, and have the potential for severe complications including hemorrhage, death, CSF leak, brain injury, stroke, malocclusion and Temporomandibular joint (TMJ) dysfunction (due to violation of the TMJ capsule during the approach), and facial nerve (VII) injury (64,65). The type C approach also requires subtotal petrosectomy with ablation of the middle ear cleft and Eustachian tube, leading to severe conductive hearing loss on the operated side (66). Even for advanced JNA, a transfacial approach tailored to the tumor’s location, supplemented with frontal craniotomy when tumors involve the anterior cranial fossa, can address nearly all resectable JNA with less morbidity than the ITF approaches.

| | | Maxillectomy/LeFort | | |

| | | | Transfacial | | |

Complications The potential for complications is proportional to the size, anatomical location of the lesion, and the extent of surgery. For JNA, the most common major complication (besides recurrence) is intraoperative hemorrhage, which can be severe enough to require transfusion, or be life-threatening (67). While steps may be taken to minimize this risk, including preoperative embolization of feeding vessels and intraoperative ligation of the IMA, resection of JNA is an inherently bloody procedure, and the possibility of massive hemorrhage must be planned for. Delayed bleeding is fortunately rare. Intracranial JNA carries the obvious risks of injury to the dura, CSF leak, injury to brain parenchyma, stroke, and other neurological sequelae. Resection of lesions involving, or in close proximity to, the orbit, optic chiasm, or cavernous sinus, carry risks of visual impairment and diplopia. Both infratemporal and transfacial approaches can cause facial numbness due to traction on, or sacrifice of, branches of V. This often improves with time. Virtually all patients should expect some nasal dryness and crusting in the postoperative period. Other complications include sinusitis, serous otitis media, and unfavorable facial scarring. The recurrence rate of JNA has traditionally been much higher than typical for benign disease, and has been estimated in some series to range between 20% and over 50%, with more advanced lesions carrying a significantly higher risk for recurrence (20). Risk factors for recurrence include intracranial disease (especially involvement of the anterior cranial fossa), extension into the basisphenoid through the pterygoid canal, erosion into the walls of the sphenoid sinus, involvement of the medial cavernous sinus, and infiltration into the pterygoid muscles and into the pterygoid plates (20,36). To avoid recurrence, the surgical approach should be carefully tailored to the extent of disease, and meticulous attention to removal of tumor extending into the basisphenoid is recommended.

Radiation Therapy Selected series have found that radiation therapy provides local control rates similar to that of surgery in patients with advanced JNA, and it was formerly considered the treatment of choice for patients with “unresectable” intracranial or cavernous sinus extension. As surgical approaches to JNA have improved, allowing for more aggressive resection of advanced disease with greatly decreased mortality and morbidity, surgery has become the standard of care even of tumors with extensive intracranial spread. However, radiation therapy may still play a role in tumors for which resection would pose unacceptable risk to the optic nerve, optic chiasm, internal carotid artery, and cavernous sinus, particularly when disease is found medial to the abducens (IV) nerve.

490

Sikora and Weber

A

B

C

D

Figure 7 Combined infratemporal fossa and facial translocation approach for advanced angiofibroma with intracranial extension. (A) Hemicoronal and WeberFergusson incisions. (B) Bone cuts for facial translocation, zygomatic osteotomy, and subtemporal craniectomy. (C) Vascularized anterior maxillary osteoplastic flap, reflected. (D) Extent of bone removed in subtemporal craniectomy. Source: Adapted from Ref. 61.

Radiation therapy is also considered for recurrences not amenable to repeat surgery, and after surgery when subtotal resection is necessary to avoid risk to vital structures. It is important to note that radiation therapy usually provides relatively rapid relief of symptoms, but complete involution of the tumor can take years and recurrence is more likely when residual tumor persists more than 2 years (68). Thus, meticulous follow-up with serial imaging is particularly important when patients are treated with this modality. The largest series of JNA patients treated with radiation therapy (55 patients treated with 30–35 Gy) described an 80% control rate (54). More recent series have described similar control rates, including one of 22 patients treated with

definitive radiotherapy (30–36 Gy, conventional fractionation) from 1975 to 2003 which found a 90% control rate at 10 years, with all tumors ultimately controlled by salvage surgery or re-irradiation (69). All patients tolerated therapy without interruptions in treatment; however, there was a significant rate of late complications (32%), including six patients with cataracts, two with in-field basal cell carcinomas, and two with transient central nervous system syndrome. Another series that reported 27 patients treated with definitive radiation therapy for advanced JNA (30–55 Gy) reported a recurrence rate of 15%, with all recurrences occurring within the first 2 to 5 years (70). An additional 15% of patients had severe complications including temporal lobe radionecrosis,


cataracts, panhypopituitarism, and growth retardation. None of these series reported unfavorable alterations of facial growth, although that is a theoretical concern of radiation in this population. Due to the development of skull base surgery techniques, which allow the reliable and safe resection of advanced JNA, and the risk of significant delayed complications of radiotherapy, we consider surgery frontline therapy for the majority of patients with JNA and reserve radiation therapy for recurrent and residual disease and in combination with therapy to address infiltration of tumor into areas where resection is not desirable. While most published studies are of JNA, treated with conventional radiotherapy, small case series treated with conformal or intensity-modulated radiotherapy have shown similar control rates and an initially favorable complication rate (71,72). However, both patient numbers and duration of follow-up are still too limited to draw conclusions about these relatively new techniques. Several one- and two-patient series of patients with recurrent or residual disease after surgery treated with single-dose stereotactic radiotherapy (“gamma knife” or “cyber knife”) described 100% control (defined as failure of disease to progress) at 2 to 3 years with no complications, but the small number of patients and limited follow-up preclude drawing conclusions from these interesting pilot studies (73,74). For other vascular skull base tumors (HPC, hemangioma, and vascular malformations), radiotherapy has an even less-defined role. Surgery is the usual treatment of these predominantly benign lesions.

Chemotherapy and other Therapies Chemotherapy (with doxorubicin and dacarbazine or Adriamycin and dacarbazine) has been described in one 5-patient series (75), and a single patient who was part of a larger series of chemotherapy for pediatric tumors (76), as a potential treatment for patients with advanced, unresectable disease due to significant intracranial involvement, extensive blood supply from intracranial vessels, or recurrence after adequate surgical treatment. However, it is infrequently used. The presence of hormone receptors on JNA and the obvious hormonal component of its regulation have led to substantial interest in hormonal treatment either alone or to facilitate surgical resection. Estrogen has been shown to decrease the size and vascularity of JNA, but it is impossible to predict which tumors will respond, and the feminizing effects of estrogen are highly undesirable in this patient population. The androgen receptor blocker flutamide has been examined in two small series of five (77)and seven (78) patients; in the first series four patients had an average 44% reduction of tumor volume, and in the second, no reduction in tumor volume or intraoperative blood loss was observed. Currently, the use of chemotherapy and hormonal therapies in the treatment of JNA should be restricted to investigational clinical trials. These therapies have no role in the treatment of other benign vascular skull base lesions.

OUTCOME AND PROGNOSIS The overall prognosis for limited JNA is excellent, with control rates for lesions without intracranial involvement approaching 90 to 100% after surgery alone. Recurrence rates for stage III and IV disease are considerably worse, and historically have ranged from 30% to almost 60% (see discussion of recurrence in section on complications, above) (36,79,80). More recent series suggest that recurrence is lower with mod-

491

ern advanced craniofacial techniques. A report describing a single-center’s experience treating 17 patients with Fisch stage III and IV JNA with preoperative embolization followed by an infratemporal fossa approach, and comparing results in this group to 16 additional patients previously reported by the same institution, found a recurrence rate of 6% (1 patient) after a median 28-month follow-up, despite the fact that 8 patients had had previous operations (64). Complications included wound infection (6%), facial palsy involving the frontal branch (25%, with 4/5 recovering by 6 months), and permanent conductive hearing loss in all patients treated with a type C approach; one patient had partial vision loss after embolization. There was no incidence of death, stroke, CSF leak, meningitis, or injury to the brain parenchyma. The senior author’s experience using preoperative embolization followed by a craniofacial approach to treat five patients with advanced (Radkowski stage IIIB) JNA found one recurrence (20%) after 28 to 63 months of follow-up (61). Complications were minimal, including nasal crusting and discharge, sinusitis, temporary serous otitis media (60%), and facial anesthesia in the distribution of V2 and V3. Less than 50% of recurrences of JNA occur within 1 year of surgery (20), and recurrences can present at any time before young adulthood. Thus, cautious posttreatment follow-up, including serial endoscopic and CT and/or MRI imaging, is necessary to detect and treat recurrence before it results in complications. We recommend physical examination and imaging every 3 months during the first year, every 6 months during the second year following surgery, and yearly examinations for the next 3 years. Recurrence is extremely unlikely after 5 years of uneventful follow-up. The prognosis of other vascular skull base lesions is hard to estimate due to their rarity. For HPC, the local control recurrence rate has been estimated to be 40%, with metastases developing in an additional 15% of patients; 10-year survival is estimated to be 70% for HPC at all sites (81). As is the case with JNA, sinonasal HPC demands close endoscopic and radiographic follow-up after treatment. Clinicians should be mindful of the potential for metastases to lung, liver, and bone even in disease with apparently benign behavior, and perform symptom-directed evaluation of these areas as necessary.

REFERENCES 1. Batsakis JG. Tumors of the Head and Neck: Clinical and Pathological Correlations, 2nd ed. Baltimore, MD: Williams and Wilkins, 1979. 2. Bremer JW, et al. Angiofibroma: Treatment trends in 150 patients during 40 years. Laryngoscope. 1986;96(12):1321–1329. 3. Gatumbi I, Linsell CA. Nasopharyngeal angiofibromas in kenya. Br J Cancer. 1964;18:69–73. 4. Backwinkel KD, Diddams JA. Hemangiopericytoma. Report of a case and comprehensive review of the literature. Cancer. 1970;25:896–901. 5. Walike JW, Bailey BJ. Head and neck hemangiopericytoma. Arch Otolaryngol. 1971;93:345–353. 6. Mulliken JB, Glowacki J. Hemangiomas and vascular malformations in infants and children: A classification based on endothelial characteristics. Plast Reconstr Surg. 1982;69(3):412–422. 7. Jacobsson M, et al. Involution of juvenile nasopharyngeal angiofibroma with intracranial extension. A case report with computed tomographic assessment. Arch Otolaryngol Head Neck Surg. 1989;115(2):238–239. 8. Stansbie JM, Phelps PD. Involution of residual juvenile nasopharyngeal angiofibroma (a case report). J Laryngol Otol. 1986;100(5):599–603.

492

Sikora and Weber

9. Abraham SC, et al. Frequent beta-catenin mutations in juvenile nasopharyngeal angiofibromas. Am J Pathol. 2001;158(3):1073– 1078. 10. Valanzano R, et al. Genetic evidence that juvenile nasopharyngeal angiofibroma is an integral FAP tumour. Gut. 2005;54(7):1046–1047. 11. Guertl B, et al. Nasopharyngeal angiofibroma: An APC-geneassociated tumor? Hum Pathol. 2000;31(11):1411–1413. 12. Ferouz AS, Mohr RM, Paul P. Juvenile nasopharyngeal angiofibroma and familial adenomatous polyposis: An association? Otolaryngol Head Neck Surg. 1995;113(4):435–439. 13. Giardiello FM, et al. Nasopharyngeal angiofibroma in patients with familial adenomatous polyposis. Gastroenterology. 1993;105(5):1550–1552. 14. Farag MM, et al. Hormonal receptors in juvenile nasopharyngeal angiofibroma. Laryngoscope. 1987;97(2):208–211. 15. Hwang HC, et al. Expression of androgen receptors in nasopharyngeal angiofibroma: An immunohistochemical study of 24 cases. Mod Pathol. 1998;11(11):1122–1126. 16. Brentani MM, et al. Multiple steroid receptors in nasopharyngeal angiofibromas. Laryngoscope. 1989;99(4):398–401. 17. Gatalica Z. Immunohistochemical analysis of steroid hormone receptors in nasopharyngeal angiofibromas. Cancer Lett. 1998;127(1–2):89–93. 18. Neel HB III, et al. Juvenile angiofibroma. Review of 120 cases. Am J Surg. 1973;126(4):547–556. 19. Harrison DF. The natural history, pathogenesis, and treatment of juvenile angiofibroma. Personal experience with 44 patients. Arch Otolaryngol Head Neck Surg. 1987;113(9):936–942. 20. Lloyd G, et al. Juvenile angiofibroma: The lessons of 20 years of modern imaging. J Laryngol Otol. 1999;113(2):127–134. 21. Wilms G, et al. Pre-operative embolization of juvenile nasopharyngeal angiofibromas. J Belge Radiol. 1989;72(6):465–470. 22. Sternberg SS. Pathology of juvenile nasopharyngeal angiofibroma; a lesion of adolescent males. Cancer. 1954;7(1):15–28. 23. Svoboda DJ, Kirchner F. Ultrastructure of nasopharyngeal angiofibromas. Cancer. 1966;19(12):1949–1962. 24. Hasegawa T, et al. Solitary fibrous tumor of the soft tissue. An immunohistochemical and ultrastructural study. Am J Clin Pathol. 1996;106(3):325–331. 25. Alobid I, et al. Solitary fibrous tumour of the nasal cavity and paranasal sinuses. Acta Otolaryngol. 2003;123(1):71–74. 26. Thompson LD, Miettinen M, Wenig BM. Sinonasal-type hemangiopericytoma: A clinicopathologic and immunophenotypic analysis of 104 cases showing perivascular myoid differentiation. Am J Surg Pathol. 2003;27(6):737–749. 27. Stout AP, Murry MR. Hemangiopericytoma: A vascular tumor featuring Zimmermann’s pericytes. Ann Surg. 1942;116:26–33. 28. Gengler C, Guillou L. Solitary fibrous tumour and haemangiopericytoma: Evolution of a concept. Histopathology. 2006;48(1):63–74. 29. Compagno J. Hemangiopericytoma-like tumors of the nasal cavity: A comparison with hemangiopericytoma of soft tissues. Laryngoscope. 1978;88(3):460–469. 30. Carew JF, Singh B, Kraus DH. Hemangiopericytoma of the head and neck. Laryngoscope. 1999;109(9):1409–1411. 31. Batsakis JG, Rice DH. The pathology of head and neck tumors: Vasoformative tumors, part 9B. Head Neck Surg. 1981;3(4):326– 339. 32. Chang J, et al. Proliferative hemangiomas: Analysis of cytokine gene expression and angiogenesis. Plast Reconstr Surg. 1999;103(1):1–9; discussion 10. 33. Elluru RG, Azizkhan RG. Cervicofacial vascular anomalies. II. Vascular malformations. Semin Pediatr Surg. 2006;15(2):133– 139. 34. Maher ER, Kaelin WG Jr. von Hippel-Lindau disease. Medicine (Baltimore). 1997;76(6):381–391. 35. Werner JA, et al. Current concepts in the classification, diagnosis and treatment of hemangiomas and vascular malformations of the head and neck. Eur Arch Otorhinolaryngol. 2001;258(3):141– 149. 36. Radkowski D, et al. Angiofibroma. Changes in staging and treatment. Arch Otolaryngol Head Neck Surg. 1996;122(2):122–129.

37. Sessions RB, et al. Radiographic staging of juvenile angiofibroma. Head Neck Surg. 1981;3(4):279–283. 38. Yeh S, Foroozan R. Orbital apex syndrome. Curr Opin Ophthalmol. 2004;15(6):490–498. 39. Schick B, Kahle G. Radiological findings in angiofibroma. Acta Radiol. 2000;41(6):585–593. 40. Chong VF, Fan YF. Radiology of the nasopharynx: Pictorial essay. Australas Radiol. 2000;44(1):5–13. 41. Vilanova JC, Barcelo J, Villalon M. MR and MR angiography characterization of soft tissue vascular malformations. Curr Probl Diagn Radiol. 2004;33(4):161–170. 42. Li JR, et al. Evaluation of the effectiveness of preoperative embolization in surgery for nasopharyngeal angiofibroma. Eur Arch Otorhinolaryngol. 1998;255(8):430–432. 43. Siniluoto TM, et al. Value of pre-operative embolization in surgery for nasopharyngeal angiofibroma. J Laryngol Otol. 1993;107(6):514–521. 44. Moulin G, et al. Juvenile nasopharyngeal angiofibroma: Comparison of blood loss during removal in embolized group versus nonembolized group. Cardiovasc Intervent Radiol. 1995;18(3):158–161. 45. George B, et al. Intratumoral embolization of intracranial and extracranial tumors: Technical note. Neurosurgery. 1994;35(4):771– 773; discussion 773–774. 46. Liang Y, et al. Direct intratumoral embolization of hypervascular tumors of the head and neck. Chin Med J (Engl). 2003;116(4):616–619. 47. da Costa DM, et al. Surgical experience with juvenile nasopharyngeal angiofibroma. Ann Otolaryngol Chir Cervicofac. 1992;109(5):231–234. 48. Duvall AJ III, Moreano AE. Juvenile nasopharyngeal angiofibroma: Diagnosis and treatment. Otolaryngol Head Neck Surg. 1987;97(6):534–540. 49. Jackson IT, et al. Hemangiomas, vascular malformations, and lymphovenous malformations: Classification and methods of treatment. Plast Reconstr Surg. 1993;91(7):1216–1230. 50. Hovius SE, et al. The diagnostic value of magnetic resonance imaging in combination with angiography in patients with vascular malformations: A prospective study. Ann Plast Surg. 1996;37(3):278–285. 51. Scholtz AW, et al. Juvenile nasopharyngeal angiofibroma: management and therapy. Laryngoscope. 2001;111(4 Pt 1):681–687. 52. Burkey B, Koopmann CF, Brunberg J. The use of biopsy in the evaluation of pediatric nasopharyngeal masses. Int J Pediatr Otorhinolaryngol. 1990;20(2):169–179. 53. Cummings BJ. Relative risk factors in the treatment of juvenile nasopharyngeal angiofibroma. Head Neck Surg. 1980;3(1):21– 26. 54. Cummings BJ, et al. Primary radiation therapy for juvenile nasopharyngeal angiofibroma. Laryngoscope. 1984;94(12 Pt 1):1599–1605. 55. Denys D, et al. The effects of radiation on craniofacial skeletal growth: A quantitative study. Int J Pediatr Otorhinolaryngol. 1998;45(1):7–13. 56. Roger G, et al. Exclusively endoscopic removal of juvenile nasopharyngeal angiofibroma: Trends and limits. Arch Otolaryngol Head Neck Surg. 2002;128(8):928–935. 57. Hofmann T, et al. Endoscopic resection of juvenile angiofibromas–long term results. Rhinology. 2005;43(4):282– 289. 58. Pryor SG, Moore EJ, Kasperbauer JL. Endoscopic versus traditional approaches for excision of juvenile nasopharyngeal angiofibroma. Laryngoscope. 2005;115(7):1201–1207. 59. Onerci TM, Yucel OT, Ogretmenoglu O. Endoscopic surgery in treatment of juvenile nasopharyngeal angiofibroma. Int J Pediatr Otorhinolaryngol. 2003;67(11):1219–1225. 60. Sciarretta V, et al. Endoscopic sinus surgery for the treatment of vascular tumors. Am J Rhinol. 2006;20(4):426–431. 61. Bales C, et al. Craniofacial resection of advanced juvenile nasopharyngeal angiofibroma. Arch Otolaryngol Head Neck Surg. 2002;128(9):1071–1078. 62. Powell DM, et al. Maxillary removal and reinsertion in pediatric patients. Arch Otolaryngol Head Neck Surg. 2002;128(1):29–34.

Chapter 34: Angiofibromas and Vascular Tumors of the Skull Base 63. Lang DA, et al. Craniofacial access in children. Acta Neurochir (Wien). 1998;140(1):33–40. 64. Zhang M, et al. Update on the infratemporal fossa approaches to nasopharyngeal angiofibroma. Laryngoscope. 1998;108(11 Pt 1):1717–1723. 65. Fisch U, Fagan P, Valavanis A. The infratemporal fossa approach for the lateral skull base. Otolaryngol Clin North Am. 1984;17(3):513–552. 66. Fisch U, Pillsbury HC. Infratemporal fossa approach to lesions in the temporal bone and base of the skull. Arch Otolaryngol. 1979;105(2):99–107. 67. Tyagi I, Syal R, Goyal A. Staging and surgical approaches in large juvenile angiofibroma–study of 95 cases. Int J Pediatr Otorhinolaryngol. 2006;70(9):1619–1627. 68. Reddy KA, et al. Long-term results of radiation therapy for juvenile nasopharyngeal angiofibroma. Am J Otolaryngol. 2001;22(3):172–175. 69. McAfee WJ, et al. Definitive radiotherapy for juvenile nasopharyngeal angiofibroma. Am J Clin Oncol. 2006;29(2):168–170. 70. Lee JT, et al. The role of radiation in the treatment of advanced juvenile angiofibroma. Laryngoscope. 2002;112(7 Pt 1):1213–1220. 71. Kuppersmith RB, et al. The use of intensity modulated radiotherapy for the treatment of extensive and recurrent juvenile angiofibroma. Int J Pediatr Otorhinolaryngol. 2000;52(3):261– 268. 72. Beriwal S, Eidelman A, Micaily B. Three-dimensional conformal radiotherapy for treatment of extensive juvenile angiofibroma: report on two cases. ORL J Otorhinolaryngol Relat Spec. 2003;65(4):238–241.

493

73. Park CK, et al. Recurrent juvenile nasopharyngeal angiofibroma treated with gamma knife surgery. J Korean Med Sci. 2006;21(4):773–777. 74. Dare AO, et al. Resection followed by radiosurgery for advanced juvenile nasopharyngeal angiofibroma: Report of two cases. Neurosurgery. 2003;52(5):1207–1211; discussion 1211. 75. Goepfert H, Cangir A, Lee YY. Chemotherapy for aggressive juvenile nasopharyngeal angiofibroma. Arch Otolaryngol. 1985;111(5):285–289. 76. Cangir A, et al. Combination chemotherapy with adramycin (NSC-123127) and dimethyl triazeno imidazole carboxamide (DTIC) (NSC-45388) in children with metastatic sold tumors. Med Pediatr Oncol. 1976;2(2):183–190. 77. Gates GA, et al. Flutamide-induced regression of angiofibroma. Laryngoscope. 1992;102(6):641–644. 78. Labra A, et al. Flutamide as a preoperative treatment in juvenile angiofibroma (JA) with intracranial invasion: Report of 7 cases. Otolaryngol Head Neck Surg. 2004;130(4):466–469. 79. Gullane PJ, et al. Juvenile angiofibroma: A review of the literature and a case series report. Laryngoscope. 1992;102(8):928– 933. 80. Howard DJ, Lloyd G, Lund V. Recurrence and its avoidance in juvenile angiofibroma. Laryngoscope. 2001;111(9):1509–1511. 81. Hasson O, Kirsch G, Lustmann J. Hemangiopericytoma of the tongue in an 11-year-old girl: Case report and literature review. Pediatr Dent. 1994;16(1):49–52. 82. Fisch U. The infratemporal fossa approach for nasopharyngeal tumors. Laryngoscope. 1983;93(1):36–44.

35 Chordoma and Chondrosarcoma of the Skull Base Gordon T. Sakamoto and Griffith R. Harsh

chondrocranium (14). The petroclival synchondrosis is the most common site (15). Chondrosarcomas are also extradural tumors. Most grow slowly, destroy bone, and extend into surrounding soft tissue. Distant metastases occur in 7% to 12% of patients (14,16). Most chondrosarcomas are low-grade malignancies (15) and thus have a better prognosis than chordomas (17,18).

INTRODUCTION Chordomas and chondrosarcomas of the skull base can present formidable challenges to effective treatment. These challenges include the relative inaccessibility of the skull base, tumor involvement of critical neural and vascular structures, a tendency to recur locally, and possible metastatic dissemination. Despite such challenges, the combination of aggressive skull base surgical resection and high-dose radiation can cure almost all low-grade chondrosarcomas and provide meaningful intervals of disease control of both chordomas and high-grade chondrosarcomas.

INCIDENCE AND EPIDEMIOLOGY Chordomas

Chordomas were first recognized at autopsy by Lushka in 1856 (1) and Virchow in 1857 (2). Believing that these tumors were cartilaginous in origin, Virchow named them “ecchondrosis physaliphora” (2). In 1858, Muller proposed that these tumors were related to the notochord (3). In 1864, Klebs described the first symptomatic case (4). In 1894, Ribbert coined the term “chordoma” after he found these tumors in the nucleus pulposus and correctly surmised their notochordal origin (5). Chordomas are rare primary malignant tumors of bone that arise from notochordal remnants (6). They are locally aggressive. Although they can occur anywhere along the axial skeleton, chordomas most commonly occur at either end, the sacrococcyx (50%) or the clivus (35%) (6,7). Most grow slowly, expanding and destroying bone. Although chordomas usually arise outside the dura, they may infiltrate and penetrate the dura to spread intracranially or intraspinally. Dural invasion usually occurs late in the course of aggressive tumors. Chordomas can also extend intradurally through surgical durotomies. There are rare reports of primary intradural intracranial chordomas (8,9). Metastases, which become clinically evident in 10% to 20% of patients, usually occur late in the course of the disease (10). At autopsy, metastases can be found in up to 40% of patients (10). Patients usually die from the consequences of locoregional disease, rather than from the metastases.

Chordomas are the most common extradural clival tumors. The overall incidence of chordomas is less than 0.1 per 100,000 persons per year (19), and they account for about 0.15% of all intracranial tumors (20). Skull base chordomas have an equal gender distribution (21–23). Although chordomas can occur at any age, they are rare in patients younger than 30 years old (24). The peak incidence is in the fourth or fifth decade of life (25). The median age at diagnosis in a large series was 46 years (22). There is no known association between the development of chordomas and potential risk factors such as radiation or other environmental carcinogens. Chordomas occur in isolation and are not part of any known systemic syndrome. Although chordoma occurrence in one family has been linked to chromosome 7q33 (26), no gene mutation specific to chordoma has been identified. Intracranial chordomas most commonly arise from the midline caudal third of the clivus, below the spheno-occipital synchondrosis (22,27). Chordomas may extend in all directions from their notochordal origin (27). Chordomas from the rostral notochord often extend into the dorsum sellae and present as sellar, suprasellar, or cavernous sinus tumors, which compress the pituitary gland, optic nerves and chiasm, and the midbrain (28). Chordomas extending ventrally through the clivus can present as nasopharyngeal masses causing nasal obstruction or dysphasia (29). Chordomas extending from the dorsal clivus can compress the pons and medulla; dorsolaterally extending tumors can involve the spheno-occipital or petrosal temporal bone.

Chondrosarcomas

Chondrosarcomas

Chondrosarcoma was first recognized as a distinct pathologic entity in 1939 by the American College of Surgeons (11). Previously, these tumors were classified as osteosarcomas. Chondrosarcomas are rare cartilaginous tumors of different grades of malignancy. Their origin is controversial; possibilities include embryonal cartilaginous rests, mesenchymal pluripotent cells, and metaplasia of fibroblasts (12,13). Approximately 50% of chondrosarcomas occur at skull base synchondroses, sites of fusion of separate cartilages forming the

Chondrosarcomas are rare. They account for 0.02% of all intracranial neoplasms (7,14,30). Chondrosarcomas are slightly more common in men than in women (16). Although skull base chondrosarcomas can occur in any age group, there is a peak incidence in the second and third decades (14,31). The mean age at diagnosis is 40.7 years (13). Although chondrosarcomas are usually isolated tumors, they may occur as part of a systemic syndrome, such as Paget disease, Ollier disease, and Maffucci syndrome (15,32). Ollier disease involves

Chordomas

495

496

Sakamoto and Harsh

multiple enchondral bone cysts and Maffucci syndrome involves multiple enchondromas and cutaneous and visceral hemangiomas. Ploidy ranges from hyperhaploidy to pentaploidy (33). Although loss (and gain) of genetic material from many different chromosomes occurs, cytogenetic studies have identified no single characteristic aberration (33). Isolated chondrosarcomas are usually paramedian. However, when part of Ollier or Maffucci syndrome, chondrosarcomas may be midline (34). Embryologically, the skull base forms from a cartilaginous matrix (35). During ossification, some cartilage may fail to form bone and remain as a rest. These cartilaginous rests may transform into chondrosarcomas. Although chondrosarcomas can arise from cartilage forming the anterior, middle, or posterior fossa, most skull base chondrosarcomas arise near the clivus. Sixty-six percent arise from the petro-occipital junction, 28% from the clivus and 6% from the sphenoethmoid complex (15).

PATHOLOGY Chordomas Chordomas are lobulated, grayish tan to bluish white tumors. They are usually well delineated from surrounding soft tissues. They range in consistency from firm to gelatinous. There may be foci of calcification or hemorrhage. The tumor’s size varies greatly. In bone, the tumor infiltrates the marrow space and expands the cortex to form a well-demarcated mass. The tumor may also penetrate the cortex and grow into neighboring soft tissue. Chordoma are composed of lobules and nests of large epithelial-appearing cells separated by fibrous bands. The neoplastic cells are arranged in sheets or cords or float individually in the myxoid stroma. The nuclei are of moderate size and show mild to moderate atypia. They have abundant pink cytoplasm. Variable numbers of cells have clear vacuoles, which impart a “bubbly” appearance to the cytoplasm (Fig. 1). Mitoses are limited and foci of necrosis are common. The neoplastic cells contain periodic acid-Schiff diastase-sensitive glycogen. Immunohistochemically, chordomas express S-100 and epithelial markers such as cytokeratin and epithelial membrane antigen (6).

Figure 1 H & E stained chordoma at 200×. Physaliferous cells with multiple clear cytoplasmic vacuoles are arranged in chords.

Chondrosarcomas Macroscopically, chondrosarcomas consist of gray-to-tan white nodules. The tumor consistency ranges from firm and gritty to mucinous. Additionally, there may be yellow-white chalky areas of calcification. They infiltrate the normal marrow and encase cancellous bone. They may transgress the cortex and form a soft tissue mass. Microscopically, four primary types of chondrosarcoma have been described: conventional, clear cell, dedifferentiated, and mesenchymal (15,36). Almost all skull base chondrosarcomas are of the conventional type. The dedifferentiated and mesenchymal variants are more aggressive tumors (25) and rarely affect the skull base. Conventional chondrosarcoma is composed of hyaline, myxoid, or a combination of hyaline and myxoid cartilage. Mixed hyaline and myxoid chondrosarcomas contain variable amounts of both matrices. Hyaline chondrosarcomas (Fig. 2) are characterized by hypercellular hyaline cartilage. The neoplastic chondrocytes lie in clear lacunae within the hyaline matrix. The chondrocytes vary in size and shape. The chondrocyte nuclei have fine chromatin and small nucleoli and vary in size and shape from small and round to medium size and ovoid. The cytoplasm may be clear or eosinophilic. The cytoplasm may also have a bubbly appearance which mimics that of the physaliphorous cells of a chordoma (15). Mitotic activity is usually very low and foci of necrosis may be present. Myxoid chondrosarcomas have neoplastic cells that appear to float in a mucinous matrix. The tumor cells may be bipolar or stellate. The cells are typically arranged in a honeycomb network of interconnecting strands and cords of cells. Mitoses are rare. Immunohistochemically, conventional chondrosarcomas express vimentin and S-100, as do chordomas (15). However, chondrosarcomas typically do not express epithelial markers such as keratin and epithelial membrane antigen (15). This is useful in distinguishing low-grade chondrosarcomas from chordomas.

Grading Grading is important in chondrosarcomas, as it has been shown to predict prognosis (37). Conventional

Figure 2 H & E stained hyaline chondrosarcoma at 100×. There are slightly atypical cells within the lacunar spaces of the hyaline matrix.

Chapter 35: Chordoma and Chondrosarcoma of the Skull Base

chondrosarcomas are graded according to the degree of cellularity, atypia, and mitotic activity (38). Most grading systems employ a three- or four-tiered system. Most chondrosarcomas are well to moderately differentiated. In a series of 200 skull base chondrosarcomas, 50.5% were grade I, 28.5% were a mixture of grade I and grade II, and 21% were pure grade II tumors. There were no grade III tumors (15). Grade I chondrosarcomas are very similar to enchondromas. However, the cellularity is higher, and there is mild cellular pleomorphism. The nuclei are small but often have an open chromatin pattern and small nucleoli. Mitoses are very rare. Grade I chondrosarcomas usually do not metastasize. Grade II chondrosarcomas have higher cellularity than do grade I tumors. They also have moderate cellular pleomorphism, larger nuclei, and more nuclear atypia. Mitoses are rare. Unlike grade I tumors, about 10% to 15% of grade II chondrosarcomas metastasize. Grade III chondrosarcomas have high cellularity, marked cellular pleomorphism, a high nuclear to cytoplasm ratio, and frequent mitoses. Grade III tumors have significant metastatic potential.

CLINICAL PRESENTATION The presentation of chordomas is variable and can be influenced by the tumor’s location and the patient’s age. Symptoms can manifest by compression or invasion of neural tissue. Clival tumors commonly present with pain and cranial neuropathies (6,13,39). The headaches are usually occipital and are worsened by changes in neck position. The cranial neuropathies vary with tumor location: upper clival lesions may affect cranial nerves II–VI, and lower clival lesions may affect cranial nerves VI–XII. Myelopathy may be caused by tumors compressing the brain stem or upper cervical spinal cord. Diplopia and headache are the most common symptoms of skull base chordomas (6,39). In one series, more than 90% of 155 skull base chordomas presented with diplopia (6), in most cases secondary to a sixth nerve palsy. Fifty percent of patients had headache. Fifty percent had one or more palsies of cranial nerves VII through XII. Twenty-five percent had loss of visual acuity and diplopia, and 25% had nasal involvement (6). Additionally, pituitary dysfunction, seizures, and fifth nerve palsy occur (6,13,39). Diplopia (64%), tongue weakness (60%), and headache (45%) were the most common signs and symptoms in one series of pediatric skull base chordomas (40). In children 5 years or younger, long tract signs (88%), lower cranial nerve deficits (62%), and signs of increased intracranial pressure (50%) were the most frequent presentations (40). The presentation of chondrosarcomas is very similar to that of chordomas. Patients commonly present with cranial nerve palsies, headaches, and gait disturbance (15). In a mixed series of skull base chondrosarcomas and chordomas, diplopia (60%), headache (60%), dysphagia (40%), and facial numbness (33%) were the most common symptoms (13). Tumor compression of the brain stem and fourth ventricle can cause lower cranial neuropathy (31), gait ataxia, and increased intracranial pressure (14).

DIFFERENTIAL DIAGNOSIS Most tumors of the skull base arise from its bone or surrounding soft tissue (41,42). Skull base tumors other than chordoma or chondrosarcoma include chondroma, craniopharyngioma, eosinophilic granuloma, fibrous dysplasia, giant cell

497

tumor, lymphoma, meningioma, metastases, osteoblastoma, pituitary adenoma, and plasmacytoma/multiple myeloma. Chondromas are benign tumors composed of mature hyaline cartilage. They grow and usually become symptomatic during adolescence. Plasmacytomas and multiple myeloma are malignant plasma cell tumors that grow within bone. Pituitary adenomas arise within the sella, which is usually expanded. Lymphomas are more likely to involve adjacent soft tissue. In fibrous dysplasia, normal bone matrix is replaced with abnormal calcified tissue containing collagen and fibroblasts (43). Fibrous dysplasia normally presents in late childhood or adolescence. The monostotic form is more common; the polyostotic form is associated with Albright syndrome. Meningiomas in this region can arise from the dura of the clivus (0.6–0.8%) (44), sella, cavernous sinus, petrous apex, or the foramen magnum (2–3%) (42,44). They occur more frequently in women. Clival meningiomas may present with cranial nerve palsies or myelopathy. Foramen magnum meningiomas may also cause local pain. Eosinophilic granulomas usually occur during childhood and present as an enlarging tender mass that appears lytic without a rim of sclerosis on skull X-rays or computed tomography (CT) scan (45). Eosinophilic granulomas rarely affect the skull base, but those that do can cause otorrhea and cranial nerve palsies (45). Osteomas and osteoblastomas are blastic lesions that rarely involve the skull base. Osteoid osteomas are smaller than osteoblastomas and present with pain that is relieved by aspirin. Osteoblastomas present with pain that is often nocturnal. Metastases to the bone, particularly those from breast, lung, and prostate cancer, are quite common.

IMAGING Imaging is critical to the diagnosis and monitoring of chordomas and chondrosarcomas. Chordomas of the skull base are usually found in the midline arising from the clivus (Fig. 3). On noncontrast CT, they are usually isodense or slightly hypodense to the brain. They expand and erode bone (46). Chordomas show moderate-to-marked contrast enhancement on CT. On magnetic resonance imaging (MRI), chordomas are isointense to brain on T1-weighted images, hyperintense on T2 images, and enhance with gadolinium. Blood products from tumoral hemorrhage may change the tumor’s signal characteristics. Chondrosarcomas of the skull base are usually paramedian (Fig. 4) at the petro-occipital junction (66%), the clivus (28%), and the sphenoethmoid complex (6%) (15). Contrast CT shows calcified tumor destroying bone (47). Solid portions of the tumor enhance with contrast. On MRI, chondrosarcomas are hypointense to brain on T1-weighted images and hyperintense on T2-weighted, proton-density, and FLAIR images (48). Chondrosarcomas show heterogeneous contrast enhancement, which may give them a salt and pepper appearance.

TREATMENT Options for the management of skull base chordomas and chondrosarcomas include clinical and radiological observation, biopsy followed by observation, biopsy followed by radiation, surgical removal, and surgery followed by radiation. In addition to these options, chemotherapy has also been used. Although these tumors have characteristic appearances on CT and MRI, the differential diagnosis (Table 1)

498

Sakamoto and Harsh Table 1 Differential Diagnosis Chondroma Craniopharyngioma Eosinophilic granuloma Fibrous dysplasia Lymphoma Meningioma Metastasis Nasopharyngeal carcinoma Neurofibroma Pituitary adenoma Plasmacytoma/Multiple myeloma

(A)

(B)

(C)

(D)

logic function. The goals of surgical resection may vary: en bloc excision of tumor, piecemeal gross total resection of the tumor, or a radical subtotal removal to decompress critical neurovascular structures or to improve tumor geometry for postoperative radiotherapy. The surgical approach should be tailored to the goal chosen for each individual patient. Choice of surgical approach should also consider tumor size, site of origin, direction of expansion, relationships with cranial nerves and arteries, extent of tumor invasion, the patient’s preoperative health, the surgeon’s familiarity with the approach, and prior treatments. Many of surgical approaches to the central skull base have been described. Schematically, they can be broken down into three general approaches: anterior, anterolateral, and

Figure 3 Chordoma. (A) Contrast-enhanced CT shows a midclival mass destroying the clivus and extending posteriorly to compress the basilar artery. (B) T1-weighted post gadolinium axial MR image shows this chordoma to be moderately enhancing. (C) T2-weighted axial MR image shows the chordoma to be hyperintense to brain and hypointense to CSF. (D) T1-weighted coronal image shows destruction of the clivus by a midline tumor.

of these skull base tumors is usually broad enough to warrant tissue examination. Since these lesions are predominantly extradural, standard stereotactic biopsy techniques may not be useful and biopsy is more likely to be performed through the nose, mouth, or mastoid.

(A)

(B)

(C)

(D)

Surgery When the diagnosis is firmly established as chordoma or chondrosarcoma, surgery will usually play a prominent role in treatment. Surgery can provide a definitive pathologic diagnosis, improve neurologic function by decompressing critical structures, lengthen time to recurrence and patient survival, and optimize spatial relationships and tumor geometry for postoperative radiotherapy. Surgical resection is usually indicated in skull base chordomas and chondrosarcomas. Some low-grade chondrosarcomas may be monitored if they are small and asymptomatic, but most chondrosarcomas and all chordomas warrant treatment. The multidisciplinary treatment team of neurosurgeon, otolaryngologist, and radiation oncologist should develop a comprehensive treatment plan which maximizes tumor control and patient survival while limiting the risk of iatrogenic complications. Numerous reports suggest the value to both tumor control and patient survival of extensive resection and high-dose radiotherapy for both chordomas and chondrosarcomas (13,17,18,22,49). Surgery is also often indicated for the restoration or preservation of neuro-

Figure 4 Chondrosarcoma. (A) On a T1-weighted axial MR image, this chondrosarcoma is isointense to brain. (B) Postgadolinium MR imaging shows the heterogeneous “salt and pepper” enhancement of the tumor. (C) A T2weighted axial MR image shows the tumor to be hyperintense to brain. (D) A contrast-enhanced CT demonstrates calcification and enhancement of the tumor.

Chapter 35: Chordoma and Chondrosarcoma of the Skull Base Table 2 Operative Approaches to Chordomas and Chondrosarcomas Anterior approaches Extended subfrontal (modified transbasal) Maxillotomy and extended maxillotomy Transethmoidal Transsphenoidal and extended transsphenoidal Transoral Anterolateral approaches Frontotemporal Lateral Approaches Frontotemporal transsylvian Subtemporal anterior transpetrosal Subtemporal infratemporal preauricular

499

defect should be obliterated by an inlay graft whose edges are placed deep to the margins of the dura. Repaired dura should then be covered by an onlay graft of fascia or dural allograft held snugly by a firm strut (thin bone, polyethylene glycol plate, or titanium wire mesh) wedged beneath the deep margins of the bone opening. This can then be covered by tissue adhesive, fat, and vascularized soft tissue such as nasal mucosa, pharyngeal musculature, pericranium, or temporalis fascia or muscle. With large, insecurely repaired openings, drainage of CSF through a ventriculostomy or lumbar drain is indicated. At the craniocervical junction, the combination of tumor erosion and surgical removal of involved bone may cause instability, which warrants fusion. Instrumentation is then often needed because most patients will receive radiotherapy, which will retard bone fusion.

Posterolateral approaches Extreme lateral Presigmoid petrosal Retrolabyrinthine or partial translabyrinthine

Radiation

posterolateral Table 2. A combination of approaches may be required to obtain satisfactory tumor exposure, especially for larger tumors. Microscopic and endoscopic visualization may both be helpful (50). Since most chordomas and chondrosarcomas are predominantly extradural, an extradural approach is usually preferred (51). This allows direct access to affected bone. Often involved bone is removed in the approach to infiltrated dura. Examples include a transclival approach to a chordoma compressing the pons and a transpetrosal approach to a petroclival chondrosarcoma. Chondrosarcomas tend to be more discrete and thus more completely resectable than chordomas. Some chondrosarcomas, however, may be so heavily calcified and incorporated into the skull base that they can be removed only by fragmenting them into dense parcels or drilling. In manipulating these calcified fragments, it is essential to first identify and then dissect tumor from nearby cranial nerves and critical vessels. Incorporation of fine cranial nerves or vessels within a calcified mass may preclude safe tumor removal and warrant subtotal resection. This portion of the tumor is often relatively indolent and long-term tumor control rates with adjuvant radiotherapy are quite high. Chordomas, grossly, often have two intermixed components: a soft, gelatinous portion within expanded bone or dura and a more sinewy infiltration and expansion of dura or extracranial soft tissue. The gelatinous part can be easily removed by suction or gentle dissection and curettage; often thickened arachnoid protects cranial nerves, brain stem, and vessels. The exception occurs in reoperations in which this protection has been violated by prior dissection. Then, tumor may surround cranial nerves and extend between brain stem arteries and the pia. Inappropriately aggressive resection then risks cranial neuropathies and brain stem stroke from injury to perforating arteries. The sinewy portion requires more sharp dissection. The plane between tumor and surrounding soft tissue is often obscure. Where possible, thickened, potentially infiltrated dura and extradural soft tissue should be excised. After dural and intradural tumor has been removed, any remaining tumor-infiltrated bone at the margins of resection should be aggressively drilled away. The result is usually a large fistula through dura and bone, which must be repaired to prevent leakage of CSF. Dural openings should be sutured closed, primarily if possible, or by using autologous fascia or dural allograft if needed. If a graft cannot be sewn in, the

Few chondrosarcomas and almost no chordomas will be removed with such confidence of microscopically complete resection that adjuvant therapy can be eschewed. Incomplete resection is usually followed by regrowth of tumor (6,13,23,52,53). High-dose radiation can delay or prevent this recurrence. Efficacy is highly dependent on dose (55–70 Gy) (39,54,55). Doses of 45 to 60 Gy are associated with poor progression-free and overall survival rates. Recurrence rates from 50% to 100% have been reported in chordomas previously treated with conventional irradiation (52,53). The optimal dose for cranial chordomas is unknown. The Proton Radiation Oncology Group is studying doses of 75.6 and 82.9 CGE (56). The Proton Radiation Oncology Group also found excellent results in patients with skull base chondrosarcomas treated with 69.9 CGE (56). After mixed photon–proton beam irradiation of chondrosarcomas at 66 to 83 CGE, the 5- and 10year local recurrence-free survival rates were 97% and 92%, respectively (57). For chordomas, the 5- and 10-year local recurrence-free survival rates were 64% and 42%, respectively (57). The goal is to deliver these doses in a highly conformal way which spares adjacent radiosensitive structures from exposure beyond their tolerance. For example, exclusion of the optic nerve/chiasm at 55 to 60 CGE (58) may protect vision, and exclusion of the pituitary at 50 CGE may prevent endocrinopathies (59). Currently popular techniques include proton beam, intensity modulation radiotherapy, and stereotactic radiosurgery. Proton beam irradiation achieves high conformality by virtue of the Bragg peak effect of energy deposition by protons traversing tissue. The energy deposition profile of protons through tissue has a low entrance dose, a peak whose depth can be modulated, and no exit dose (60). This permits planning of treatments with high dose to tumor and rapid falloff in the surround. This high conformality is important for irregularly shaped tumors such as chordomas and chondrosarcomas at the skull base. In one series, 90% of the patients underwent pure proton beam treatment of chordomas with 65 to 79 CGE; they demonstrated a local control rate of 59% and an overall survival rate of 79% at 5 years (61). For chondrosarcomas, the 5-year local control rate and overall survival rate at 5 years are 75% and 100%, respectively (61). Particles other than protons, such as helium, neon, carbon, and neutrons, have also been used to treat chordomas and chondrosarcomas (62–66). Intensity modulation radiotherapy achieves the same goal by modulating the dose intensity pattern of the radiation to match the tumor’s shape. Treatment is carefully planned

500

Sakamoto and Harsh

using 3-D CT images and computerized dose calculations. Combinations of several intensity-modulated fields from different beam directions produce a custom tailored radiation dose that maximizes tumor dose while minimizing exposure of adjacent normal tissues. Stereotactic radiosurgery targets high doses of X-rays from a linear accelerator (LINAC or Cyberknife) or gamma rays from multiple cobalt sources (Gamma Knife) to a tumor. Numerous low-intensity beams from multiple directions intersect within the tumor volume so as to deliver high dose to the tumor while sparing surrounding tissue. Radiosurgery, traditionally, uses one to five treatment sessions. The precision possible permits treatment of tumor very close (within 2–3 mm) to critical structures (67). In that the risk of injury rises with targeted volume, radiosurgery is usually limited to tumor volumes less than 10 mL. Radiosurgery may be, particularly, useful to treat small unresectable residua of chondrosarcomas or recurrences of either chondrosarcomas or chordomas. All three of these modalities share the risk of neurologic deficit from injury to adjacent structures, such as delayed visual deterioration (2% in one proton beam series) (61), hypopituitarism (as high as 72% at 5 years and 84% at 10 years in one proton beam series and as low as 7% in another) (59,61), and hearing loss (10% in one proton beam series) (61). Use of lower doses may prevent these deficits. Use of radiation sensitizers such as razoxane may also improve the risk benefit calculus (68). Interstitial brachytherapy employs a different strategy. It involves permanent placement of radioactive seeds such as 125 I within a tumor or along a tumor resection margin. Gold foil implants are used to shield vital structures such as the brain stem from radiation. The half-life of 125 I is 60.2 days and almost all of the dose is delivered within four halflives. The dose to the margin is usually at least 50 Gy. The continuous low dose of radiation theoretically targets more cells as they move from radiation resistant to more radiation sensitive phases of the cell cycle. Additionally, the low dose should allow normal tissue to repair the damage caused by the sublethal doses of radiation.

Chemotherapy As therapies improve the local control of chordomas and chondrosarcomas, effective treatment of metastatic disease becomes increasingly important. Approximately 30% to 40% of patients with chordomas develop metastases (69,70), most commonly to the lungs, liver, and bone (71). Chemotherapy is an option for surgically inaccessible, previously irradiated recurrent local or metastatic tumor. Historically, it has had poor efficacy (44). Newer agents with tumor-specific rationales, such as imatinib mesylate, an inhibitor of tyrosine kinases activated by growth factors expressed in chordomas, may be an improvement (72).

OUTCOME AND PROGNOSIS Chordomas and chondrosarcomas, despite being lumped together historically, have distinct natural histories. The natural history of skull base chordomas is dismal. If not treated, the average affected patient will live approximately 18 months (6,73). Even with treatment, the 5-year survival rate ranges from 51% to 79% (13,18,22,61) and the 10-year survival rate ranges from 35% to 69% (18,22) (Fig. 5). Treatment usually involves attempted radical resection and radiotherapy. The importance of the extent of the initial resection is empha-

Figure 5 Local recurrence-free survival for skull base chordomas and chondrosarcomas after surgical resection (if necessary) and mixed proton– photon radiotherapy. Source: Reprinted with permission from Harsh G, ed., Chordomas and Chondrosarcomas of the Skull Base.

sized in one study which had a 5-year survival rate of 100% in patients with a radical or total excision without immediate postoperative radiotherapy (18). The overall operative mortality rate is approximately 1.9% to 5% (13,18,74). Subsequent operations carry greater risk to neurologic function and survival. A recent study demonstrated a 0% operative mortality rate for the first operation and a 7.1% mortality rate for the second (18). Surgeryrelated complications include cranial nerve palsies, stroke causing brain stem injury, CSF leakage, and meningitis. Permanent neurologic deficits were observed in 28.6% of patients in one series (74). Postoperative radiotherapy is almost always indicated. In one series, patients treated with surgery alone had a mean survival of 18 months versus 63 months for those who received postoperative radiotherapy (6). In patients with subtotal resections, postoperative conventional radiotherapy confers a 5-year survival rate of 65% (18). Patients who received postoperative proton beam therapy had an actuarial 5-year survival rate of 79% (61). Recurrences were felt to be due to limitations on delivered dose by critical structures. A larger series demonstrated 5-year and 10year local recurrence-free rates of 64% and 42%, respectively (57), as well as a complication rate of 8% (57). Although the series is small and the follow-up is limited, carbon ion therapy provided a 4-year survival rate of 86% (65). Initial results for patients who underwent subtotal resection followed by stereotactic radiosurgery appear promising: a 97 to 100% 2-year survival rate (75,76) and a 5-year survival rate of 82% (76). These series were small and the follow-up is limited. A more recent study shows actuarial tumor control rates at 2 years and 5 years of 93% and 52%, respectively (77). Recurrences are associated with a worse prognosis. The actuarial 3-year and 5-year survival rates for a chordoma, locally recurrent after surgery and radiotherapy, are 44% and 5%, respectively (78). Other factors associated with poor prognosis are tumor volume (>70cm3 ) (21,61) and older age of the patient (22). In one series, younger patients had 5-year and 10year survival rates of 75% and 63%, respectively, compared to 30% and 11%, respectively, in older patients (18).

Chapter 35: Chordoma and Chondrosarcoma of the Skull Base

Various immunohistochemical stains have been used as a prognostic indicator. A high MIB labeling index and overexpression of the tumor suppressor, p53, carry poor prognoses (79,80). Conventional chondrosarcomas have a better prognosis than do chordomas. Five-year survival rates depend on the histological grade of the tumor. Grade I, II, and III lesions had, respectively, 90%, 81%, and 43% 5-year survival rates in one series (37). Overall, the 5-year survival rate is 90% to 99% and the 10-year survival rate is 71% to 99% (13,15,81,82). Most surgical series advocate radical resection when feasible. Postoperative proton beam radiotherapy and radiosurgery are the most commonly used adjunctive treatments for small residual or recurrent tumors. Radical excision of low-grade chondrosarcomas may obviate the need for immediate postoperative irradiation. Among patients whose tumor was completely removed but who were not given postoperative radiotherapy, 78.3% experienced 5 years of recurrence-free survival (17,81). Morbidity and mortality rates for surgical resection of chondrosarcomas are similar to those for chordomas. Postoperative radiotherapy is indicated for most subtotally resected low-grade chondrosarcomas and all highergrade chondrosarcomas. It is highly effective. In a large series using proton beam radiotherapy, the 5-year and 10-year local recurrence-free survival rates were 97% and 92%, respectively, for chondrosarcomas (57). A smaller, more recent series demonstrated a 5-year survival rate of 100% (61). Radiosurgery is an attractive alternative for smaller residual tumors (82). In one series, postoperative fractionated stereotactic radiotherapy provided a 100% 5-year recurrence-free survival rate (76).

SUMMARY Chordomas and chondrosarcomas of the skull base are rare and challenging tumors. The relative inaccessibility of the skull base, the tumors’ proximity to critical neurovascular structures, their relative radioresistance to standard doses, their lack of chemosensitivity, and their tendency to recur locally require thoughtful choice and meticulous administration of therapy. Despite these challenges, the combination of surgical resection and high-dose radiation can cure almost all low-grade chondrosarcomas and provide meaningful intervals of control of both chordomas and high-grade chondrosarcomas.

REFERENCES 1. Luschka H. Die Altersveranderungen der Zwischenwirbelknorpel. Virchows Arch Path Anat. 1856;9:312–327. 2. Virchow R. Untersuchungen uber die entwickelung des schadelgrundes im gesunden und krankhaften zustande, und uber den einfluss derselben auf schadelform, gesichtsbildung und gehirnbau. G. Zeimer, 1857: Berlin. 3. Muller H. Ueber das Vorkommen von Resten der Chorda dorsalis bei Mecschen nach der Geburt und uber ihr Verhaltnis zu den Gallertgeschwulsten am Clivus. Ztschr Rat Med. 1858;2:202–229. 4. Klebs E. Ein Fall von Ecchondrosis spheno-occipitalise amylacea. Virchows Arch Path Anat. 1864;31:396–399. 5. Ribbert H. Ueber die Ecchondrosis physlifora sphenooccipitalis. Centralbl Allg Path Path Anat. 1894;5:457–461. 6. Heffelfinger MJ, et al. Chordomas and cartilaginous tumors at the skull base. Cancer. 1973;32(2):410–420.

501

7. Borba L, Al-Mefty O. Skull-base chordomas. Contemp Neurosurg. 1998;20:1–6. 8. Steenberghs J, et al. Intradural chordoma without bone involvement. Case report and review of the literature. J Neurosurg. 2002;97(1 Suppl):94–97. 9. Nishigaya K, et al. Intradural retroclival chordoma without bone involvement: No tumor regrowth 5 years after operation. Case report. J Neurosurg. 1998;88(4):764–768. 10. Laws E, Thapar K. Parasellar lesions other than pituitary adenomas. In: Powell M, Lightman SL, eds. Management of Pituitary Tumors: A Handbook. New York: Churchill-Livignstone, 1996:175–222. 11. Ewing J. A review of the classification of bone tumors. Surg Gynecol Obstet. 1939;68:971–976. 12. Neff B, et al. Chondrosarcoma of the skull base. Laryngoscope. 2002;112(1):134–139. 13. Gay E, et al. Chordomas and chondrosarcomas of the cranial base: Results and follow-up of 60 patients. Neurosurgery. 1995;36(5):887–896; discussion 896–897. 14. Hassounah M, et al. Primary cranial and intracranial chondrosarcoma. A survey. Acta Neurochir (Wien). 1985;78(3– 4):123–132. 15. Rosenberg AE, et al. Chondrosarcoma of the base of the skull: A clinicopathologic study of 200 cases with emphasis on its distinction from chordoma. Am J Surg Pathol. 1999;23(11):1370– 1378. 16. Koch BB, et al. National cancer database report on chondrosarcoma of the head and neck. Head Neck. 2000;22(4):408–425. 17. Crockard HA, et al. A multidisciplinary team approach to skull base chondrosarcomas. J Neurosurg. 2001;95(2):184–189. 18. Crockard HA, et al. A multidisciplinary team approach to skull base chordomas. J Neurosurg. 2001;95(2):175–183. 19. McMaster ML, et al. Chordoma: Incidence and survival patterns in the United States, 1973–1995. Cancer Causes Control. 2001;12(1):1–11. 20. Berkmen YM, Blatt ES. Cranial and intracranial cartilaginous tumours. Clin Radiol. 1968;19(3):327–333. 21. O’Connell JX, et al. Base of skull chordoma. A correlative study of histologic and clinical features of 62 cases. Cancer. 1994;74(8):2261–2267. 22. Forsyth PA, et al. Intracranial chordomas: A clinicopathological and prognostic study of 51 cases. J Neurosurg. 1993;78(5):741– 747. 23. Watkins L, et al. Skull base chordomas: A review of 38 patients, 1958–1988. Br J Neurosurg. 1993;7(3):241–248. 24. Wold LE, Laws ER Jr. Cranial chordomas in children and young adults. J Neurosurg. 1983;59(6):1043–1047. 25. Unni K. Dahlin’s Bone Tumors: General Aspects and Data on 11,087 Cases, 5th ed. Philadelphia, PA: Lippincott-Raven, 1996:291–303. 26. Yang XR, et al. Corroboration of a familial chordoma locus on chromosome 7q and evidence of genetic heterogeneity using single nucleotide polymorphisms (SNPs). Int J Cancer. 2005;116(3):487–491. 27. Laws E Jr. Cranial Chordomas. In: Wilkins R, Rengachary S, eds. Neurosurgery. New York: McGraw-Hill, 1985:927–929. 28. Thodou E, et al. Intrasellar chordomas mimicking pituitary adenoma. J Neurosurg. 2000;92(6):976–982. 29. Campbell WM, et al. Nasal and paranasal presentations of chordomas. Laryngoscope. 1980;90(4):612–618. 30. Cianfriglia F, Pompili A, Occhipinti E. Intracranial malignant cartilaginous tumours. Report of two cases and review of literature. Acta Neurochir (Wien). 1978;45(1–2):163–175. 31. Oguro K, et al. Chondrosarcoma of the posterior fossa—case report. Neurol Med Chir (Tokyo). 1989;29(11):1030–1038. 32. Korten AG, et al. Intracranial chondrosarcoma: Review of the literature and report of 15 cases. J Neurol Neurosurg Psychiatry. 1998;65(1):88–92. 33. Mandahl N, et al. Cytogenetic aberrations and their prognostic impact in chondrosarcoma. Genes Chromosomes Cancer. 2002;33(2):188–200. 34. Tachibana E, et al. Surgical treatment of a massive chondrosarcoma in the skull base associated with Maffucci’s syndrome:

502

35. 36. 37.

38. 39.

40.

41. 42. 43. 44. 45.

46. 47.

48. 49. 50.

51.

52. 53.

54. 55.

56.

57. 58. 59.

Sakamoto and Harsh A case report. Surg Neurol. 2000;54(2):165–169; discussion 169– 170. Ariel IM, Verdu C. Chordoma: An analysis of twenty cases treated over a twenty-year span. J Surg Oncol. 1975;7(1):27–44. Richardson MS. Pathology of skull base tumors. Otolaryngol Clin North Am. 2001;34(6):1025–1042, vii. Evans HL, Ayala AG, Romsdahl MM. Prognostic factors in chondrosarcoma of bone: A clinicopathologic analysis with emphasis on histologic grading. Cancer. 1977;40(2):818–831. Finn DG, Goepfert H, Batsakis JG. Chondrosarcoma of the head and neck. Laryngoscope. 1984;94(12 Pt 1):1539–1544. Raffel C, et al. Cranial chordomas: Clinical presentation and results of operative and radiation therapy in twenty-six patients. Neurosurgery. 1985;17(5):703–710. Matsumoto J, Towbin RB, Ball WS Jr. Cranial chordomas in infancy and childhood. A report of two cases and review of the literature. Pediatr Radiol. 1989;20(1–2):28–32. Menezes AH, Traynelis VC, Gantz BJ. Surgical approaches to the craniovertebral junction. Clin Neurosurg. 1994;41:187–203. Meyer FB, Ebersold MJ, Reese DF. Benign tumors of the foramen magnum. J Neurosurg. 1984;61(1):136–142. Levy ML, Chen TC, Weiss MH. Monostotic fibrous dysplasia of the clivus. Case report. J Neurosurg. 1991;75(5):800–803. Castellano F, Ruggiero G. Meningiomas of the posterior fossa. Acta Radiol Suppl. 1953;104:1–177. Brisman JL, et al. Eosinophilic granuloma of the clivus: Case report, follow-up of two previously reported cases, and review of the literature on cranial base eosinophilic granuloma. Neurosurgery. 1997;41(1):273–278; discussion 278–279. Meyer JE, Oot RF, Lindfors KK. CT appearance of clival chordomas. J Comput Assist Tomogr. 1986;10(1):34–38. Grossman RI, Davis KR. Cranial computed tomographic appearance of chondrosarcoma of the base of the skull. Radiology. 1981;141(2):403–408. Brown E, Hug EB, Weber AL. Chondrosarcoma of the skull base. Neuroimaging Clin N Am. 1994;4(3):529–541. Al-Mefty O, Borba LA. Skull base chordomas: A management challenge. J Neurosurg. 1997;86(2):182–189. Frank G, et al. The endoscopic transnasal transsphenoidal approach for the treatment of cranial base chordomas and chondrosarcomas. Neurosurgery. 2006;59(1 Suppl 1):ONS50–7; discussion ONS50–7. Sen CN, Sekhar LN. An extreme lateral approach to intradural lesions of the cervical spine and foramen magnum. Neurosurgery. 1990;27(2):197–204. Catton C, et al. Chordoma: Long-term follow-up after radical photon irradiation. Radiother Oncol. 1996;41(1):67–72. Cummings BJ, Hodson DI, Bush RS. Chordoma: The results of megavoltage radiation therapy. Int J Radiat Oncol Biol Phys. 1983;9(5):633–642. Higinbotham NL, et al. Chordoma. Thirty-five-year study at Memorial Hospital. Cancer. 1967;20(11):1841–1850. Pearlman AW, Friedman M. Radical radiation therapy of chordoma. Am J Roentgenol Radium Ther Nucl Med. 1970;108(2):332–341. Liebsch NJ, Munzenrider JE. Proton Radiotherapy for Cranial Base Chordomas. In: Harsh G, ed. Chordomas and Chondrosarcomas of the Skull Base and Spine. New York: Thieme, 2003:307– 314. Munzenrider JE, Liebsch NJ. Proton therapy for tumors of the skull base. Strahlenther Onkol. 1999;175(Suppl 2):57–63. Habrand IL, et al. Neurovisual outcome following proton radiation therapy. Int J Radiat Oncol Biol Phys. 1989;16(6):1601–1606. Pai HH, et al. Hypothalamic/pituitary function following highdose conformal radiotherapy to the base of skull: Demonstration of a dose-effect relationship using dose-volume histogram analysis. Int J Radiat Oncol Biol Phys. 2001;49(4):1079–1092.

60. Hug EB. Review of skull base chordomas: Prognostic factors and long-term results of proton-beam radiotherapy. Neurosurg Focus. 2001;10(3):E11. 61. Hug EB, et al. Proton radiation therapy for chordomas and chondrosarcomas of the skull base. J Neurosurg. 1999;91(3):432– 439. 62. Berson AM, et al. Charged particle irradiation of chordoma and chondrosarcoma of the base of skull and cervical spine: The Lawrence Berkeley Laboratory experience. Int J Radiat Oncol Biol Phys. 1988;15(3):559–565. 63. Castro JR, et al. Experience in charged particle irradiation of tumors of the skull base: 1977–1992. Int J Radiat Oncol Biol Phys. 1994;29(4):647–655. 64. Saunders WM, et al. Precision, high dose radiotherapy. II. Helium ion treatment of tumors adjacent to critical central nervous system structures. Int J Radiat Oncol Biol Phys. 1985;11(7):1339– 1347. 65. Schulz-Ertner D, et al. Carbon ion radiation therapy for chordomas and low grade chondrosarcomas—current status of the clinical trials at GSI. Radiother Oncol. 2004;73(Suppl 2): S53–S56. 66. Schulz-Ertner D, et al. Results of carbon ion radiotherapy in 152 patients. Int J Radiat Oncol Biol Phys. 2004;58(2):631–640. 67. Muthukumar N, et al. Stereotactic radiosurgery for chordoma and chondrosarcoma: Further experiences. Int J Radiat Oncol Biol Phys. 1998;41(2):387–392. 68. Rhomberg W, et al. Combined radiotherapy and razoxane in the treatment of chondrosarcomas and chordomas. Anticancer Res. 2006;26(3B):2407–2411. 69. Chambers PW, Schwinn CP. Chordoma. A clinicopathologic study of metastasis. Am J Clin Pathol. 1979;72(5):765–776. 70. Sundaresan N, et al. Spinal chordomas. J Neurosurg. 1979;50(3):312–319. 71. O’Neill P, et al. Fifty years of experience with chordomas in southeast Scotland. Neurosurgery. 1985;16(2):166–170. 72. Casali PG, et al. Imatinib mesylate in chordoma. Cancer. 2004;101(9):2086–2097. 73. Eriksson B, Gunterberg B, Kindblom LG. Chordoma. A clinicopathologic and prognostic study of a Swedish national series. Acta Orthop Scand. 1981;52(1):49–58. 74. Colli BO, Al-Mefty O. Chordomas of the skull base: Follow-up review and prognostic factors. Neurosurg Focus. 2001;10(3):E1. 75. Miller RC, et al. The role of stereotactic radiosurgery in the treatment of malignant skull base tumors. Int J Radiat Oncol Biol Phys. 1997;39(5):977–981. 76. Debus J, et al. Stereotactic fractionated radiotherapy for chordomas and chondrosarcomas of the skull base. Int J Radiat Oncol Biol Phys. 2000;47(3):591–596. 77. Krishnan S, et al. Radiosurgery for cranial base chordomas and chondrosarcomas. Neurosurgery. 2005;56(4):777–784; discussion 777–784. 78. Fagundes MA, et al. Radiation therapy for chordomas of the base of skull and cervical spine: Patterns of failure and outcome after relapse. Int J Radiat Oncol Biol Phys. 1995;33(3):579– 584. 79. Naka T, et al. Alterations of G1-S checkpoint in chordoma: The prognostic impact of p53 overexpression. Cancer. 2005;104(6):1255–1263. 80. Saad AG, Collins MH. Prognostic value of MIB-1, E-cadherin, and CD44 in pediatric chordomas. Pediatr Dev Pathol. 2005;8(3):362–368. 81. Tzortzidis F, et al. Patient outcome at long-term follow-up after aggressive microsurgical resection of cranial base chondrosarcomas. Neurosurgery. 2006;58(6):1090–1098; discussion 1090– 1098. 82. Wanebo JE, et al. Management of cranial base chondrosarcomas. Neurosurgery. 2006;58(2):249–255; discussion 249–255.

36 Meningioma Ashwin Viswanathan and Franco DeMonte

INCIDENCE/EPIDEMIOLOGY Incidence

the most common radiation-induced neoplasm (14). Studies of immigrants to Israel who had received radiotherapy for tinea capitis between 1948 and 1960 clearly demonstrate therapeutic radiation as an etiological factor (15). Patients who had received radiation showed a relative risk of 9.5 for the development of meningioma, oftentimes after a latency period of 20 to 40 years. Similarly, studies of atomic bomb survivors in Hiroshima and Nagasaki have shown a relative risk of 6.48 when compared with non-exposed populations (16). Both increasing time from exposure and increasing dose of radiation are associated with a higher incidence of meningiomas (17). As compared with non-radiation–induced meningiomas, radiation-induced meningiomas tend to possess atypical histological features, have a more aggressive clinical course, occur at multiple locations, and have different cytogenetic characteristics (14). Cellular phone usage and the concomitant exposure to low-dose radiation exposure has recently generated attention as a possible etiological factor in the formation of brain tumors. However, to date no conclusive evidence has shown a link between cell phone usage and the development of meningiomas (18,19).

Meningiomas account for 32% of all primary brain tumors and represent the second most common central nervous system (CNS) neoplasm in adults after gliomas (1). The incidence of meningioma is 4.7 cases per 100,000 person-years, and women outnumber men by 2.1:1. The median age at diagnosis is 63 and the incidence of meningioma increases with increasing age. The incidence of diagnostically and nondiagnostically confirmed meningioma has increased between 1985 and 1999, with the incidence of nondiagnostically confirmed meningioma increasing by 4.1% per year. This trend likely reflects the increased use of MRI in the diagnosis of incidental meningioma (2). Data from the population-based Rotterdam study found incidental meningiomas in 18 out of 2000 (0.9%) MRI scans performed. These incidental meningiomas ranged from 5 mm to 60 mm in diameter and the prevalence was 1.1% in women and 0.7% in men (3). Intracranial meningiomas outnumber spinal meningiomas by approximately 10:1 (4). Pediatric meningiomas are rare and comprise less than 2% of all meningiomas and less than 5% of all pediatric brain tumors (5,6). The most common locations for intracranial meningiomas are parasagittal, sphenoid ridge, and convexity. Forty percent of all meningiomas arise from the base of the anterior, middle, or posterior fossa and are the most common skull base tumors. Sphenoid wing meningiomas make up about half of these; tuberculum sella and olfactory groove tumors the other half. Ectopic meningiomas have been described in the orbit, paranasal sinuses, skin, subcutaneous tissues, lung, mediastinum, and adrenal glands. Table 1 details the common sites for meningiomas and their incidence.

Infection The role of infectious agents in the development of brain tumors has also been investigated. The number of siblings in a family has been proposed as an indicator for exposure to infectious agents (more siblings relates to a greater risk or exposure to infectious agents). For patients less than 15 years of age at the time of meningioma diagnosis, a rate ratio of 3.71 was found when compared with patients with no siblings (20). Viruses, and in particular Simian virus 40 (SV40), have been studied as an etiological agent as well. SV40, a polyomavirus, is capable of transforming cells into those with a neoplastic phenotype (21). The oncogenic and transforming properties of SV40 are related to expression of large tumor antigen (Tag), which is postulated to have a role in inactivating the tumor suppressor functions of p53, pRb, p107, and others (22). Of 10 human meningioma samples analyzed, SV40 Tag was identified in 7 samples, the Tag-p53 complex was identified in 3, and the Tag-pRb complex in 2 (23). Rollison et al., however, were not able to identify SV40 in any of the 15 human meningioma tumor samples they analyzed (24). Further work is necessary to delineate and characterize a potential relationship.

Etiology Trauma Several case-control studies have shown an elevated odds ratio for the development of meningioma in patients with a history of head trauma, though no clear association has been found to date (7–10). A recent population-based case-control study of 200 patients with meningioma found an elevated odds ratio of 2.62 for the development of meningioma given a history of head trauma (11). In patients with a history of head trauma occurring 10 to 19 years prior to the date of diagnosis, an odds ratio of 4.33 was found. Yet other studies have found no such elevated risk (12,13). Recall bias has been suggested as a confounding factor limiting the effectiveness of case-control studies. Further epidemiological studies are necessary to validate and delineate a relationship.

Genetics NF-2 Mutations The majority of meningiomas are sporadic tumors in patients with no history of brain tumors. However, familial meningiomas have been identified in a number of conditions including neurofibromatosis type 2 (NF2), NF1, Gorlin/nevoid basal cell carcinoma syndrome, Rubinstein–Taybi

Radiation Exposure to ionizing radiation is a known etiological factor in the development of meningiomas, and meningiomas are 503

504

Viswanathan and DeMonte Table 1

Location and Incidence of Intracranial Meningioma

Site Parasagittal/Falcine Convexity Sphenoid ridge Suprasellar (tuberculum) Posterior fossa Olfactory groove Middle fossa/Meckel cave Tentorial Peritorcular Lateral ventricle Foramen magnum Orbit/optic nerve sheath

Incidence (%) 25 19 17 9 8 8 4 3 3 1–2 1–2 1–2

Source: Based on data from Refs. 55, 91–94.

syndrome, Li–Fraumeni syndrome, von Hippel–Lindau syndrome, Werner syndrome, Gardner syndrome, and the melanoma/astrocytoma-brain tumor syndrome (25). Meningiomas were one of the first solid tumors to be associated with a characteristic cytogenetic change—loss of heterozygosity (LOH) of chromosome 22. Forty to seventy percent of meningiomas exhibit loss of heterozygosity (LOH) for markers from the chromosomal region 22q12.2 that encompasses the NF2 gene. Mutations specifically in the NF2 gene have been reported in 30% to 60% of sporadic meningiomas. The frequency of NF2 mutations is similar among WHO grades I, II, and III meningiomas (26). This finding suggests that NF2 gene inactivation may be an important initiation step in the formation of meningiomas, but may not play a role in tumor progression (25). Inactivation of both NF2 gene alleles is necessary for tumor formation. The NF2 protein merlin belongs to the protein 4.1 families that link membrane proteins to the cytoskeleton. Hannson et al. utilized microarray-based comparative genomic hybridization to study 126 sporadic meningioma specimens (27). They found the incidence of biallelic NF2 inactivation to be 52% in fibroblastic variants as compared with 18% in meningothelial histologies. This finding suggests that NF2 inactivation may not be a critical step in the formation of meningothelial meningiomas (27). In addition to NF2 mutations, other protein 4.1 family members are also downregulated in meningiomas. The EPB41L3 or DAL1 gene product protein 4.1B belongs to the same superfamily as merlin, and is thought to act as a tumor suppressor. Loss of protein 4.1B expression is common in meningiomas of all grades, though one group found the loss predominantly in meningiomas of higher grade (28). However, no propensity to develop tumors has been observed in transgenic mice lacking EPB41L3/DAL1 (29).

Other Genetic Abnormalities Deletion of 1p is the second most common genetic change seen in meningiomas (27), and has been associated with meningioma progression and recurrence. Co-deletion of 1p and 22q is seen in 67% of meningiomas. Transition to atypical meningioma has been associated with losses on chromosomes 1p, 6q, 10, 14q, and 18q and gains on chromosomes 1q, 9q, 12q, 15q, 17q, and 20q (30). Grade III or anaplastic tumors are associated with gains on 17q23 and losses on 9p. They may also demonstrate more frequent losses on 6q, 10, and 14q as compared with atypical tumors. Loss of chromosome 14 represents the third most common genetic abnormality after mutations in chromosomes 22 and 1. Deletions on chromosome 14 have been associated

with an increased risk of relapse and poorer prognosis (31). More recent DNA microarray assays have shown losses on chromosomes 10 and 14 in high-grade meningiomas with increased expression of several genes related to IGF (IGF2, IGFBP3, AKT3) or wingless (WNT, CTNNB1, CDK5R1, ENC1, CCND1) pathways (32). Other studies have demonstrated amplification of MSH2 (2p22.3-p22.1) in 16 of 31 meningiomas (51.6%), deletion of GSCL (22q11.21) in 41.9%, amplification of INS (12ptel) and TCL1A (14q32.1), and deletions of HIRA (22q11.21) and IGH (14qtel) (33). Recently, Pecina-Slaus et al. showed loss of heterozygosity of the Adenomatous polyposis coli (APC) gene in 47% of 32 specimens (34). APC acts as a tumor suppressor gene and has previously been recognized as a colon-specific tumor suppressor.

Tumor Biology Growth Factors Meningiomas have been found to express a number of growth factors and their receptors including epidermal growth factor receptor (EGFR), basic fibroblast growth factor (BFGF) receptor, platelet-derived growth factor (PDGF) receptor, and vascular endothelial growth factor-A (VEGF-A). This finding has led to the possibility that autocrine growth factor secretion and autocrine loops may play a role in the growth of meningiomas (25). Recent investigation by Smith et al. demonstrated PDGFR-β expression in all of the 84 meningioma samples studied (35). In addition, expression of BFGFR was found in 89% of benign meningiomas, while EGFR immunoreactivity was detected in 47% of benign meningiomas. EGFR immunoreactivity was found as a strong predictor of prolonged survival in patients with atypical meningioma (35). VEGF-A, which is also known as vascular permeability factor, is considered to be a key factor in angiogenesis and edema formation for meningiomas. Several studies have demonstrated VEGF-A levels in meningiomas to be associated with the extent of peritumoral edema (36,37) and some smaller studies have postulated that VEGF-A mRNA expression may correlate with meningioma vascularity (37,38). Other studies however have found no association between VEGF-A protein levels and microvessel density (39).

Receptors The increased incidence of meningioma in women, along with early studies reporting accelerated tumor growth during pregnancy (40) and associations with breast cancer (41), has led to investigation of the role of sex steroids in the pathogenesis of meningioma. Most studies have found meningiomas to lack estrogen receptors (ERs), though some have found low concentrations of ERs in 5% to 33% of meningiomas (42,43). Progesterone receptors are much more common, being expressed in about two-thirds of meningiomas. As seen in breast cancer, expression of the progesterone receptor tends to decrease during malignant progression (25,44). Lusis et al. utilized high-throughput tissue microarray immunohistochemistry (TMA-IHC) to study 41 meningioma samples of all histological grades. They found PR reactivity in all benign specimens, as compared with 67% of atypical meningiomas, and 56% of anaplastic meningiomas (44). Pravdenkova et al. found the expression of the PR alone indicates a more favorable clinical and biological outcome (45).

PATHOLOGY Meningiomas arise from the arachnoidal (meningothelial) cells, and most are well-demarcated tumors with a broad

Chapter 36: Meningioma

505

Table 2 World Health Organization 2007 Classification of Meningiomas Meningiomas with low risk of recurrence and aggressive growth Meningothelial meningioma WHO grade I Fibrous (fibroblastic) meningioma WHO grade I Transitional (mixed) meningioma WHO grade I Psammomatous meningioma WHO grade I Angiomatous meningioma WHO grade I Microcystic meningioma WHO grade I Secretory meningioma WHO grade I Lymphoplasmacyte-rich meningioma WHO grade I Metaplastic meningioma WHO grade I Meningiomas with greater likelihood of recurrence and/or aggressive behavior Chordoid meningioma WHO grade II Clear cell meningioma (intracranial) WHO grade II Atypical meningioma WHO grade II Papillary meningioma WHO grade III Rhabdoid meningioma WHO grade III Anaplastic (malignant) meningioma WHO grade III Meningiomas of any subtype or grade with high proliferation index and/or brain invasion

dural attachment. The World Health Organization (WHO) 2007 classification divides meningiomas into three grades: benign (WHO grade I), atypical (WHO grade II), and anaplastic or malignant (WHO grade III) (46). Approximately 80% of meningiomas are WHO grade I and possess a low risk of recurrence or aggressive growth. Grade II meningiomas account for 15% to 20% of all meningiomas, while grade III tumors comprise only 1% to 3% meningiomas (31). Grade II and grade III meningiomas are characterized by a greater likelihood of recurrence or of aggressive growth. Table 2 groups the subtypes of meningiomas based on their likelihood of recurrence. The WHO 2007 classification is based upon the number of mitotic figures (MFs) per 10 high-powered fields (HPF) in the area of highest mitotic activity. Meningiomas of any grade may exhibit invasion of brain parenchyma, which is characterized by finger-like projections of tumor cells without an intervening layer of leptomeninges. The presence of brain invasion does confer a greater risk for recurrence similar to that seen for atypical meningiomas. Proliferative indices are not currently included in the grading criteria for meningiomas due to significant difference in technique and interpretation between laboratories (46).

WHO Grade I Meningothelial, fibrous, and transitional meningiomas are the three most common histological variants of grade I meningiomas. Less common subtypes include psammomatous, angiomatous, microcystic, and secretory meningiomas. Chordoid, clear-cell, papillary, and rhabdoid variants of meningiomas are often associated with more aggressive tumor behavior and are consequently not classified as grade I tumors. Benign meningiomas may also invade surrounding structures including the skull, dural sinuses, orbit, and soft tissues. Meningothelial meningioma is a common variant in which tumor cells form lobules surrounded by thin collagenous septae (Fig. 1). Tumor cells are uniform, resembling normal arachnoid cap cells, and may show central clearing. Rounded eosinophilic cytoplasmic protrusions termed pseudoinclusions are also seen. Whorls and psammoma bodies, when present, are less well defined than those seen in fibrous, psammomatous, or transitional meningiomas. Fibrous meningiomas are formed by spindle-shaped cells that resem-

Figure 1 WHO grade I meningioma with multiple “whorls” (arrows, H&E x200). Source: Property of the Department of Neurosurgery, The University of Texas M.D. Anderson Cancer Center—Used with permission.

ble fibroblasts. These spindle-shaped cells form intersecting fascicles and are embedded in a collagen-rich and reticulinrich matrix. Transitional meningioma (mixed meningioma) is a common histological subtype which contains features of both meningothelial and fibrous meningiomas. Both the lobular arrangement of meningothelial meningiomas and the fascicular pattern of fibrous variants may be seen next to one another. They usually demonstrate extensive whorl formation in which the tumor cells form concentric cell layers by wrapping around each other. When the whorl formations hyalinize and calcify, they form structures known as psammoma bodies.

WHO Grade II By the WHO 2007 criteria, a meningioma is classified as atypical if it has ≥4 mitoses per 10 high-power fields, or meets three of the five following criteria: increased cellularity, high nuclear to cytoplasmic ratio (small cells), prominent nucleoli, uninterrupted patternless or sheet-like growth, or foci of spontaneous (not induced by embolism) necrosis (46) (Fig. 2). Two meningioma variants, clear-cell and chordoid, have been found to have higher recurrence rates even in the absence of the above criteria, leading to their classification as atypical meningiomas. Clear cell meningioma is a rare meningioma variant which is composed of sheets of polygonal cells with clear, glycogen-rich cytoplasm that is positive for periodic acid Schiff. They also demonstrate dense perivascular and interstitial collagen deposition. Clear cell tumors are more commonly found in the cauda equina or the cerebellopointine angle, and have a tendency to affect younger patients. Chordoid meningiomas are typically supratentorial tumors and have regions that are histologically similar to chordoma. They may have cords of small epithelioid tumor cells that contain eosinophilic cytoplasm residing in a basophilic, mucinrich matrix (30).

WHO Grade III Malignant meningiomas are characterized by a highly elevated mitotic index of >20 mitoses per 10 high-power

506

Viswanathan and DeMonte

Immunohistochemistry

Figure 2 Atypical meningioma, WHO grade II. This photomicrograph illustrates cellular sheeting, a feature indicative of increased tumor aggressiveness (H&E x200). Source: Property of the Department of Neurosurgery, The University of Texas M.D. Anderson Cancer Center—Used with permission.

fields and can pathologically resemble sarcoma, carcinoma, or melanoma (Fig. 3). In addition, large areas of necrosis can usually be seen. Papillary and rhabdoid meningiomas are consistently associated with malignant behavior and are therefore classified as WHO grade III tumors (47). Papillary meningioma has a propensity to affect younger patients, commonly exhibits brain invasion, and may show distant metastases. The rhabdoid variant is a rare tumor which contains sheets of large rhabdoid cells with eosinophilic cytoplasm containing whorled intermediate filaments and eccentric nuclei. The median survival is less than 2 years and the recurrence rate is 50% to 80% after surgical resection (48).

Figure 3 Anaplastic meningioma, WHO grade III. Mitoses are easily identified in this anaplastic meningioma (arrows) as is the marked nuclear pleomorphism (H&E x200). Source: Property of the Department of Neurosurgery, The University of Texas M.D. Anderson Cancer Center—Used with permission.

Epithelial membrane antigen (EMA) is the most commonly used marker in pathological analysis of meningiomas and most tumors show at least scattered positivity for this antigen (49). All meningiomas also strongly express vimentin (50). Tissue microarray immunohistochemistry (TMA-IHC) has proven to be an efficient and reliable method for analyzing biomarkers in meningioma. Using TMA-IHC, Lusis et al. found EMA reactivity in 100% of meningiomas regardless of grade and E-cadherin immunoreactivity in 91% of all meningiomas and 90% of anaplastic meningiomas (44). However, these markers are not specific for meningiomas. Immunohistochemistry has also proved useful in quantifying the proliferative index for meningiomas with the antibody MIB-1, which targets the proliferation marker Ki-67. Elevated proliferative indices as measured by MIB-1 labeling have been associated with an increased risk of recurrence (51). Mean MIB-1 labeling indices are 0.7% to 2.2% for grade I meningiomas, 2.1% to 9.3% for grade II meningiomas, and 11% to 16.3% for grade III meningiomas (52). MIB-1 labeling indices of greater than 5% suggest a greater likelihood of recurrence. Similarly, an antibody that specifically recognizes the phosphorylated histone H3 (PHH3) has been effectively used as an aid in grading meningiomas (53). During mitosis, phosphorylation of the Ser-10 residue of histone H3 reaches a maximum. Consequently, immunostaining with anti-PHH3 allows the observer to rapidly focus on the most mitotically active areas of the tumor (Fig. 4).

TREATMENT Surgery Complete surgical excision is the treatment of choice for meningiomas. The ability to achieve a safe, complete resection is influenced by tumor involvement of major dural sinuses, association with eloquent neurovascular structures, tumor location and size, and previous surgery or radiation (54). The vast majority of meningiomas are surrounded by a layer of arachnoid which separates the tumor from the brain, cranial nerves, and blood vessels. Meningiomas may attach to or

Figure 4 pHH3 immunostaining readily identifies the presence of mitoses (x200). Source: Property of the Department of Neurosurgery, The University of Texas M.D. Anderson Cancer Center—Used with permission.


surround cerebral arteries, but only very rarely do they invade the arterial walls. By accessing this arachnoidal plane, the surgeon is able to minimize the chance of injury to neurovascular structures. Internal debulking of the tumor facilitates the delineation of the arachnoid plane by allowing the edge of the tumor to collapse inward.

Convexity Meningiomas Convexity meningiomas comprise approximately 15% of all meningiomas and they possess the greatest potential for cure. By definition, convexity meningiomas do not arise from the skull base and do not involve the dural sinuses, and hence they allow for excision of a wide dural margin. Recurrent and/or aggressive meningiomas may not have the normal arachnoidal layer separating them from the cerebrum and require careful sharp dissection under the operating microscope to minimize cortical injury. Once the tumor and a wide dural margin have been resected, the dural defect may be repaired with pericranium, fascia lata, temporalis fascia, cadaveric dura, synthetic collagen matrix, or bovine pericardium.

Parasagittal Meningiomas Cushing and Eisenhardt defined the parasagittal meningioma as one that fills the parasagittal angle, with no brain tissue between the tumor and the superior sagittal sinus (55). Parasagittal meningiomas account for 17% to 32% of meningiomas and the primary consideration in their removal is management of the superior sagittal sinus and the cerebral veins that drain into it. Surgical approaches and management include simple dissection of the meningiomas off the lateral wall of the sinus, sagittal sinus reconstruction, and excision of the sinus in the case of a totally occluded sinus.

Olfactory Groove and Tuberculum Sellae Meningiomas Olfactory groove and tuberculum sellae tumors each comprise approximately 10% of meningiomas. Both tumors are midline lesions that derive their blood supply from the ethmoidal branches of the ophthalmic arteries, the anterior branch of the middle meningeal artery, and the meningeal branches of the ICA. Early division of this vascular supply is the first step in tumor removal. A low basal approach is preferred with a unilateral supraorbital craniotomy usually being sufficient. The optic nerves and chiasm are typically displaced and fine microdissection is required to free these structures. Inspection of both optic canals is necessary to detect tumor extension into this area—a common occurrence that can easily be missed.

Sphenoid Wing Meningiomas Sphenoid wing meningiomas are the second most common type of meningiomas after the parasagittal type. These meningiomas are classified according to their point of origin along the sphenoid ridge and include the meningiomas en plaque that are characterized by hyperostosis of the sphenoid bone that causes progressive painless proptosis and occasionally cranial neuropathies secondary to foraminal encroachment. The internal carotid, the middle and anterior cerebral arteries and their branches, as well as the optic, oculomotor, and olfactory nerves are the neurovascular structures at greatest risk during the surgical removal of sphenoid wing meningiomas. The presence of the arachnoidal layer allows for meningiomas to be microsurgically separated from these structures even though there may be a marked distortion of the normal anatomy of the region.

507

Posterior Fossa Meningiomas Posterior fossa meningiomas account for 10% of all intracranial meningiomas. Almost half of these meningiomas are located in the cerebellopontine angle, 40% are tentorial or cerebellar convexity tumors, and 9% and 6% are petroclival or at the foramen magnum. A standard retrosigmoid craniotomy allows sufficient exposure for the removal of most meningiomas of the cerebellopontine angle while supra- and infratentorial and presigmoid approaches may be necessary for petroclival meningiomas. Foramen magnum meningiomas may require a far lateral or trans-occipital condylar approach for optimal access. Following tumor debulking and devascularization the tumor is dissected from the brain stem the basilar, vertebral and cerebellar arteries and the trochlear, trigeminal, abducens, facial, vestibulocochlear, and lower cranial nerves. Meningiomas involving the tentorium and cerebellar convexity have the transverse sinus as their main area of concern. Management of the transverse sinus is similar to that of the sigmoid sinus in parasagittal meningiomas described above.

Radiation External Beam Radiation Therapy Though complete surgical resection is the ultimate goal in treating patients with meningioma, this is not always possible with an acceptable level of morbidity. Specifically, meningiomas of the sphenoid wing, cavernous sinus, clivus, cerebellopontine angle, and sellar regions are more likely to be subtotally excised (56). In particular, external beam radiation therapy (EBRT) has become an integral part of the management of optic nerve sheath meningiomas. In their evaluation of 64 patients with long-term follow-up, Turbin et al. concluded that EBRT led to more favorable outcomes as compared with surgical resection, observation, or surgical resection plus EBRT (57). Several other studies have shown similar results (58). Radiation therapy has evolved as an additional means for controlling meningioma in patients who have either undergone a subtotal resection or in patients with atypical or anaplastic histologies (59). Recommended doses generally range from 50 to 55 Gy in fractions of 1.8 to 2.0 Gy. The planning target volume can include only the gross tumor volume or the gross tumor volume plus a margin depending on the grade of the meningioma (2 cm margin recommended for anaplastic meningioma). Targeting the dural tail of meningiomas remains a subject of controversy (58).

Stereotactic Radiosurgery and Radiotherapy Stereotactic radiation techniques have emerged over the past 20 years as an important alternative to conventional external beam radiation therapy. Three modalities exist for stereotactic radiosurgery (SRS): LINAC, Gamma Knife, and protons. Tumors most appropriate for SRS are those that are smaller than 3.5 cm with little surrounding edema, and in locations where dose constraints for critical structures including the optic apparatus and the brainstem can be respected (60). Khoo et al. compared the clinical target volumes using CT and MRI for patients with skull base meningiomas undergoing radiation therapy. They found the MR and CT based target volumes provided complementary data regarding tumor involvement in soft tissue and bony regions respectively (61). Consequently, MR and CT fusion images are optimal for treatment planning of smaller meningiomas. For larger meningiomas, CT-based planning is usually adequate. Fractionated stereotactic radiotherapy (SRT) allows for precise stereotactic targeting, steep dose gradients, and the

508


benefit of allowing normal tissues to heal between radiation fractions. LINAC is the primary modality used for SRT, and is used with a relocatable frame. Immobilization systems used with SRT include bite block devices such as the Gill–Thomas– Cosman frame for adults and the Tarbell–Loeffler–Cosman frame for children. After irradiation of benign meningiomas, surveillance MR images should be obtained at 6 and 12 months, and then annually thereafter.

Table 3 Simpson’s Classification for the Extent of Resection of Intracranial Meningioma

Chemotherapy

Grade IV Grade V

Chemotherapy has a very limited role in the treatment of meningiomas. No clear evidence of efficacy has been shown for chemotherapeutic agents, though a number of different agents have been investigated. Hydroxyurea is an oral ribonucleotide reductase inhibitor which arrests meningioma cell division in the S-phase of the cell cycle and induces apoptosis (62). Though this agent has been effective in in vitro and in vivo studies, treatment of patients with recurrent or unresectable meningiomas has shown little benefit (63–65). Combination of hydroxyurea with calcium channel blockers has also been investigated in vitro with positive results (65). In vitro studies of the combination of interferon-alpha and 5fluorouracil have been promising (66), and interferon-alpha has been shown effective in prolonging the time to recurrence in a small group of patients with aggressive meningioma (67). Temozolamide, an alkylating agent which has been used in malignant gliomas, showed no benefit in treating refractory meningiomas in a phase II trial (68). Mifepristone (RU486) is a progesterone blocker that has been shown to inhibit the growth of cultured human meningioma tissue and in animal models (69). Long-term administration of mifepristone showed only mild clinical benefit in 8 of 28 patients. Endometrial hyperplasia was noted in several patients after long-term administration (70). Irinotecan (CPT11), a topoisomerase-1 inhibitor, has been shown to inhibit in-vitro cultures of human meningioma cell lines and invivo studies using a subcutaneous tumor model (71). However, a phase II study evaluating CPT-11 in patients with recurrent meningioma was stopped prematurely as all patients demonstrated tumor progression within 6 months (72).

Grade I Grade II Grade III

Gross total resection of tumor, dural attachments, and abnormal bone Gross total resection of tumor, coagulation of dural attachments Gross total resection of tumor, without resection or coagulation of dural attachments, or alternatively of its extradural extensions (e.g., invaded sinus or hyperostotic bone) Partial resection of tumor Simple decompression (biopsy)

ing (73). Careful observation with an initial imaging study 3 months following the first is recommended to identify atypical or anaplastic growth patterns; a second scan 6 months later to detect any growth; and then yearly thereafter is a reasonable method of managing patients with asymptomatic tumors.

Surgery In his landmark 1957 paper, Simpson retrospectively graded the extent of meningioma resection in an attempt to find a correlation with recurrence (75). Table 3 details Simpson’s methodology in which surgical resection is graded from 1 (complete resection) to 5 (decompression only). Kinjo et al. further suggested a grade zero resection in which an additional 2 cm margin of dura is excised as a means for further reducing the rate of recurrence (76). From Simpson’s original work, the recurrence rate after surgical resection ranged from 9% for a grade I resection, 19% for grade II, 29% for grade III, to 44% for a grade IV resection (75). The extent of resection has since been strongly confirmed as the primary factor influencing meningioma recurrence rate. More recent studies have shown the 5-year progression-free survival to be between 77% and 93% after complete resection, and between 52% and 63% in patients with a subtotal resection. The 10year progression-free survival data ranges from 61% to 80% for a gross total resection and from 37% to 45% for a subtotal resection (56,77,78).

Adjuvant Therapy OUTCOME AND PROGNOSIS Asymptomatic Meningioma As the discovery of incidental meningiomas grows, understanding the natural history is becoming increasingly more important. One-third to two-fifths of meningiomas are asymptomatic (54,73), and several studies have assessed the growth rate of incidental meningiomas. Olivero et al. found that 10 of 45 patients with asymptomatic meningiomas exhibited tumor growth. Over an average imaging follow-up of 47 months, the average tumor growth in these 10 patients was 2.4 mm/yr (74). Yano et al. found that only 37% of asymptomatic meningiomas showed tumor growth and only 6% of patients became symptomatic over a mean follow-up of 3.9 years. Patients with tumors larger than 3 cm at diagnosis or T2-hyperintense tumors were more likely to become symptomatic over time while patients with calcified tumors were less likely to (73). In the subgroup of patients greater than 70 years old, the surgical morbidity associated with asymptomatic tumors was 9.4% as compared with 4.4% in patients less than 70 years. Further the surgical morbidity in this group exceeded the morbidity in the observation alone cohort (6%). Hence, for asymptomatic meningiomas, Yano et al. recommend serial neuroimaging and close clinical monitor-

Radiation therapy is an integral part of the treatment of meningiomas. In patients who have undergone a subtotal resection followed by adjuvant external beam radiation therapy, 5-year progression-free survival has been shown to be between 77% and 91% (56,59,79–81). In a retrospective analysis of 140 patients treated with subtotal resection followed by EBRT, Goldsmith et al. reported a 5-year progression-free survival rate of 85% for benign meningiomas and 58% for malignant tumors (59). Soyeur et al. more recently compared gross total resection, subtotal resection plus adjuvant EBRT, and subtotal resection followed by radiotherapy at tumor progression (56). Over a mean follow-up of 7.7 years, the 5year progression-free survival for gross total resection was 77%, while that for subtotal resection alone was 38%. Those patients who underwent subtotal resection and adjuvant radiotherapy had a 5-year progression-free survival rate of 91%. The overall survival for the three groups was not statistically different and was no different than the age-match general population. Current methods of treatment planning and delivery have led to decreased toxicity associated with EBRT than the 38% rate reported in earlier literature (82). The complication rate for radiation therapy is between 2.2% and 3.6% and includes cognitive decline, pituitary insufficiency, and radiation-induced neoplasms (59,82,83).


Stereotactic radiosurgery (SRS) and stereotactic radiotherapy have shown promising results as both primary and adjuvant therapies for meningioma. Numerous retrospective studies since the 1990s have demonstrated 5-year local control rates with SRS of between 86% and 99%, tumor regression rates of 28% to 70%, and symptom improvement in 8% to 65% of patients (60). In their experience in treating patients with benign meningiomas less than 3.5 cm in average diameter, Pollock et al. showed radiosurgery to yield comparable results to those seen with a Simpson’s grade I surgical resection (84). However, compared with the population who underwent a Simpson’s grade II, III, or IV resection, stereotactic radiosurgery yielded a higher rate of progression-free survival (84). The 3 and 7 year rate of progression-free survival for stereotactic radiosurgery was 100% and 95% respectively, while that seen for Simpson’s grade I was 100% and 96%, Simpson’s grade II was 91% and 82%, and Simpson’s grade III and IV was 68% and 34% respectively. More recently, Kollova et al. reported their experience in treating 325 benign meningiomas with either primary or adjuvant SRS (85). Patients had a mean tumor volume of 4.4 cm3 and the authors achieved a tumor control rate of 97.9% at 5 years. Improvement in neurological symptoms such as imbalance, oculomotor palsy, trigeminal symptoms, hemiparesis, and vertigo occurred in 61.9% of patients. The permanent toxicity rate was 5.7%, which included seizures, trigeminal symptoms, hemiparesis, and others. Toxicity after radiosurgery is usually due to either symptomatic edema or cranial neuropathies. In particular, the special sensory nerves (optic and vestibulocochlear) appear the most sensitive (86). Vascular occlusion after stereotactic radiosurgery is a rare complication, but is estimated to occur in 1% to 2% of cases (87). The pathogenesis is thought to involve luminal narrowing after radiation-induced endothelial damage. As with stereotactic radiosurgery, stereotactic radiotherapy has shown high rates of progression-free survival of between 98% and 100% over a mean follow-up of 21 to 68 months (88,89). Studies have also shown an average reduction in tumor volume of 33% at 24 months and 36% at 36 months with stereotactic radiotherapy (90). Acute toxicities of stereotactic radiotherapy are generally mild and can include alopecia, skin erythema, and fatigue. The rate of late toxicity ranges between 2% and 13%. Late complications include hypopituitarism, visual deterioration, cognitive impairment, and tinnitus (60). Timing of adjuvant therapy for patients with benign meningioma who have undergone subtotal resection is still a matter of controversy. There is no randomized data to date to support postoperative radiation therapy versus radiation therapy at the time of tumor progression. The current Phase III prospective randomized control trial by the European Organisation of Research and Treatment for Cancer should aid in answering this question. REFERENCES 1. CBTRUS (2008). Statistical report: Primary brain tumors in the United States, 2000–2004. Central Brain Tumor Registry of the United States. Hinsdale, IL: 2008. 2. Hoffman S, Propp JM, McCarthy BJ. Temporal trends in incidence of primary brain tumors in the United States, 1985–1999. Neuro Oncol. 2006;8(1):27–37. 3. Vernooij MW, Ikram A, Tanghe HL. Incidental findings on brain MRI in the general population. N Eng J Med. 2007;357:1821– 1828. 4. Sloof J, Kernohan J, MacCarthy C. Primary Intramedullary Tumors of the Spinal Cord and Filum Terminale. Philadelphia, PA: WB Saunders, 1964.

509

5. Baumgartner JE, Sorenson JM. Meningioma in the pediatric population. J Neurooncol. 1996;29:223–228. 6. Rochat P, Johannesen HH, Gjerris F. Long-term follow-up of children with meningiomas in Denmark: 1935 to 1984. J Neurosurg. 2004;100:179–182. 7. Monteiro GT, Pereira RA, Kiofman RJ, et al. Head injury and brain tumors in adults: A case-control study in Rio de Janeiro, Brazil. Eur J Cancer. 2006;42(7):917–921. 8. Preston-Martin S, Paganini-Hill A, Henderson BE, et al. Casecontrol study of intracranial meningiomas in women in Los Angeles County, California. J Natl Cancer Inst. 1980;65(1):67– 73. 9. Preston-Martin S, Pogoda JM, Schlehofer B, et al. An international case-control study of adult glioma and meningioma: The role of head trauma. Int J Epidemiol. 1998;27:579–586. 10. Preston-Martin S, Yu MC, Henderson BE, et al. Risk factors for meningiomas in men in Los Angeles County. J Natl Cancer Inst. 1983;70(5):863–866. 11. Phillips LE, Koepsell TD, van Belle G, et al. History of head trauma and risk of intracranial meningioma: Population-based case-control study. Neurology. 2002;58:1849–1852. 12. Annegers JF, Laws ER Jr, Kurland LT, et al. Head trauma and subsequent brain tumors. Neurosurgery. 1979;4:203–206. 13. Inskip PD, Mellemkjaer L, Gridley G, et al. Incidence of intracranial tumors following hospitalization for head injuries (Denmark). Cancer Causes Control. 1998;9:109–116. 14. Al-Mefty O, Topsakal C, Pravdenkova S, et al. Radiationinduced meningiomas: Clinical, pathological, cytokinetic, and cytogenic characteristics. J Neurosurg. 2004;100:1002–1013. 15. Ron E, Modan B, Boice JD, et al. Tumors of the brain and nervous system after radiotherapy in childhood. N Engl J Med. 1988;319(16):1033–1039. 16. Shintani T, Hayakawa N, Hoshi M, et al. High incidence of meningioma among Hiroshima atomic bomb survivors. J Radiat Res. 1999;40:40–57. 17. Sadamori N, Shibata S, Mine M, et al. Incidence of intracranial meningiomas in Nagasaki atomic-bomb survivors. Int J Cancer. 1996;67(3):318–322. ¨ J, Kosteljanetz M, et al. Cellular tele18. Christensen HC, Schuz phones and risk for brain tumors: A population-based, incident case-control study. Neurology. 2005;64(7):1189–1195. ¨ J, Bohler ¨ 19. Schuz E, Berg G, et al. Cellular phones, cordless phones, and the risks of glioma and meningioma (Interphone Study Group, Germany). Am J Epidemiol. 2006;163(6):512–520. 20. Altieri A, Castro F, Bermejo JL, et al. Association between number of siblings and nervous system tumors suggests an infectious etiology. Neurology. 2006;67:1979–1983. 21. Topp WC, Lane D, Pollack R. Transformation by SV40 and polyomavirus. In: Tooze J, ed. DNA Tumor Viruses. New York, NY: Cold Spring Harbor Laboratory Press, Inc., 1981: 200–301. 22. Fanning E. Simian virus 40 large T antigen: The puzzle, the pieces, and the emerging picture. J Virol. 1992;66:1289–1293. 23. Zhen H, Zhang X, Bu X, et al. Expression of the simian virus 40 large tumor antigen (Tag) and formation of Tag-p53 and TagpRb complexes in human brain tumors. Cancer. 1999;86:2124– 2132. 24. Rollison DE, Utaipat U, Ryschkewitsch C, et al. Investigation of human brain tumors for the presence of polyomavirus genome sequences by two independent laboratories. Int J Cancer. 2005;113(5):769–774. 25. Simon M, Bostrom JP, Hartmann C. Molecular genetics of meningiomas: From basic research to potential clinical applications. Neurosurgery. 2007;60:787–798. 26. Wellenreuther R, Kraus JA, Lenartz D, et al. Analysis of the neurofibromatosis 2 gene reveals molecular variants of meningioma. Am J Pathol. 1995;146:827–832. 27. Hansson CM, Buckley PG, Grigelioniene G, et al. Comprehensive genetic and epigenetic analysis of sporadic meningioma for macro-mutations on 22q and micro-mutations within the NF2 locus. BMC Genomics. 2007;8:16. 28. Gutmann DH, Donohoe J, Perry A, et al. Loss of DAL-1 a protein 4.1-related tumor suppressor is an important early event in the pathogenesis of meningiomas. Hum Mol Genet. 2000;9:1495– 1500.

510


29. Yi C, McCarty JH, Troutman SA, et al. Loss of the putative tumor suppressor band 4.1B/Dal1 gene is dispensible for normal development and does not predispose to cancer. Mol Cell Biol. 2005;25:10052–10059. 30. Riemenschneider M, Perry A, Reifenberger G. Histological classification and molecular genetics of meningiomas. Lacet Neurol. 2006;5:1045–1054. 31. Lamszus K. Meningioma pathology, genetics, and biology. J Neuropathol Exp Neurol. 2004;63(4):275–286. 32. Wrobel G, Roerig P, Kokocinski F, et al. Microarray-based gene expression profiling of benign, atypical and anaplastic meningiomas identifies novel genes associated with meningioma progression. Int J Cancer. 2005;114:249–256. 33. Wada K, Maruno M, Suzuki T, et al. Chromosomal and genetic abnormalities in benign and malignant meningiomas using DNA microarray. Neurol Reas. 2005;27:747–754. 34. Pecina-Slaus N, Martic TN, Tomas D. Meningiomas exhibit loss of heterozygosity of the APC gene. J Neuroonc. 2007, epub. 35. Smith JS, Lal A, Harman-Smith M, et al. Association between absence of epidermal growth factor receptor immunoreactivity and poor prognosis in patients with atypical meningioma. J Neurosurg. 2007;106;1034–1040. 36. Goldman CK, Bharara S, Palmer CA, et al. Brain edema in meningiomas is associated with increased vascular endothelial growth factor expression. Neurosurgery. 1997;40:1269–1277. 37. Provias J, Claffey K, delAguila L, et al. Role of vascular endothelial growth factor/vascular permeability factor in angiogenesis and peritumoral edema. Neurousurgery. 1997;40:1016– 1026. 38. Samoto K, Ikezaki K, Ono M, et al. Expression of vascular endothelial growth factor and its possible relation with neovascularization in human brain tumors. Cancer Res. 1995;55:1189– 1193. 39. Lamszus K, Lengler U, Schmidt NO, et al. Vascular endothelial growth factor, hepatocyte growth factor/scatter factor, basic fibroblast growth factor, and placenta growth factor in human meningiomas and their relation to angiogenesis and malignancy. Neurosurgery. 2000;46:938–947. 40. Bickerstaff ER, Small JM, Guest IA. The relapsing course of certain meningiomas in relation to pregnancy and menstruation. J Neurol Neurosurg Psychiatry. 1958;21:89–91. 41. Mehta D, Khatib R, Patel S. Carcinoma of the breast and meningioma. Association and management. Cancer. 1983;51:1937– 1940. 42. Hsu DW, Efird JT, Hedley-White ET. Progesterone and estrogen receptors in meningiomas: Prognostic considerations. J Neurosurg. 1994;86:113–120. 43. Rubinstein AB, Loven D, Geier A, et al. Hormone receptors in initially excised versus recurrent intracranial meningiomas. J Neurosurg. 1994;81:184–187. 44. Lusis EA, Chicoine MR, Perry A. High throughput screening of meningioma biomarkers using a tissue microarray. J Neurooncol. 2005;73:219–223. 45. Pravdenkova S, Al-Mefty O, Sawyer J, et al. Progesterone and estrogen receptors: Opposing prognostic indicators in meningiomas. J Neurosurg. 2006;105:163–173. 46. Perry A, Louis DN, Scheithauer BW, et al. Meningiomas. In: Kleihues P, Cavanee WK, eds. Pathology and Genetics of Tumors of the Nervous System: World Health Organization Classification of Tumors Lyon, France: IARC Press, 2007. 47. Perry A, Scheithauer BW, Stafford SI, et al. “Rhabdoid” meningioma: An aggressive variant. Am J Surg Pathol. 1998;22;1482– 1490. 48. Perry A, Scheithauer BW, Stafford SI, et al. “Malignancy” in meningiomas: A clinicopathological study of 116 patients, with grading implications. Cancer. 1999;85:2046–2056. 49. Schnitt SJ, Vogel H. Meningiomas. Diagnostic value of immunoperoxidase staining for epithelial membrane antigen. Am J Surg Pathol. 1986;10:640–649. 50. Artlich A, Schmidt D. Immunohistochemical profile of meningiomas and their histological subtypes. Hum Pathol. 1990;21:843–849.

51. Bruna J, Brell M, Ferrer I, et al. Ki-67 proliferative index predicts clinical outcome in patients with atypical or anaplastic meningioma. Neuropathology. 2007;27:114–120. 52. Matsuno A, Hiroshi N, Nagashima T. Histopathological analysis and proliferative potentials of intracranial meningiomas using bromodeoxyuridine and MIB-1 immunohistochemistry. Acta Histochem Cytochem. 2005;38:9–15. 53. Ribalta T, McCutcheon I, Aldape K, et al. The mitosis-specific antibody anti-phosphohistone-H3 (PHH3) facilitates rapid reliable grading of meningiomas according to WHO 2000 criteria. Am J Surg Pathol. 2004;28:1532–1536. 54. Drummond KJ, Zhu J, Black PM. Meningiomas: Updating basic science, management and outcome. Neurol. 2004;10(3):113– 130. 55. Cushing H, Eisenhardt L. Meningiomas: Their Classification, Regional Behaviour, Life History, and Surgical End Results. Springfield, IL: Charles C Thomas, 1938. 56. Soyuer S, Chang EL, Selek U, et al. Radiotherapy after surgery for benign cerebral meningioma. Radiother Oncol. 2004;71:85– 90. 57. Turbin RE, Thompson CR, Kennerdell JS. A long-term visual outcome comparison in patients with optic nerve sheath meningioma managed with observation, surgery, radiotherapy, or surgery and radiotherapy. Ophthalmology. 2002;109:890–900. 58. Rogers L, Mehta M. Role of radiation therapy in treating intracranial meningiomas. Neurosurg Focus. 2007;23(4):E4. 59. Goldsmith BJ, Wara WM, Wilson CB. Postoperative irradiation for subtotally resected meningiomas. A retrospective analysis of 140 patients treated from 1967 to 1990. J Neurosurg. 1994;80:195– 201. 60. Elia AE, Shih HA, Loeffler JS. Stereotactic radiation treatment for benign meningiomas. Neurosurg Focus. 2007;23(4):E5. 61. Khoo VS, Adams EJ, Saran F, et al. A comparison of clinical target volumes determined by CT and MRI for the radiotherapy planning of base of skull meningiomas. Int J Radiat Oncol Biol Phys. 2000; 46(5):1309–1317. 62. Schrell UM, Rittig MG, Anders M, et al. Hydroxyurea for treatment of unresectable and recurrent meningiomas. I. Inhibition of primary human meningioma cells in culture and in meningioma transplants by induction of the apoptotic pathway. J Neurosurg. 1997;86:845–852. 63. Loven D, Hardoff R, Sever ZB, et al. Non-resectable slowgrowing meningiomas treated by hydroxyurea. J Neurooncol. 2004;67:221–226. 64. Newton H, Slivka M, Stevens C. Hydroxyurea chemotherapy for unresectable or residual meningioma. J Neurooncol. 2000;49:165–170. 65. Ragel BT, Gillespie DL, Kushnir V, et al. Calcium channel antagonists augment hydroxyurea- and RU486-induced inhibition of meningioma growth in vivo and in vitro. Neurosurgery. 2006;59:1109–1121. 66. Zhang Z, Wang J, Muhr C, et al. Synergistic inhibitory effects of interferon-alpha and 5-fluorouracil in meningioma cells in vitro. Cancer Lett. 1996;100:99–105. 67. Kaba SE, DeMonte F, Bruner JM, et al. The treatment of recurrent unresectable and malignant meningiomas with interferon alpha-2B. Neurosurgery. 1997;40:271–275. 68. Chamberlain MC, Tsao-Wei DD, Groshen S. Temozolomide for treatment-resistant recurrent meningioma. Neurology. 2004;62:1210–1212. 69. Matsuda Y, Kawamoto K, Kiya K, et al. Antitumor effects of antiprogesterones on human meningioma cells in vitro and in vivo. J Neurosurg. 1994;80:527–534. 70. Grunberg S, Weiss M, Spitz I, et al. Treatment of unresectable meningiomas with the antiprogesterone agent mifepristone. J Neurosurg. 1991;74:861–866. 71. Gupta V, Su YS, Samuelson CG, et al. Irenotecan: A potential new chemotherapeutic agent for atypical or malignant meningiomas. J Neurosurg. 2007;106:455–462. 72. Chamberlain MC, Tsao-Wei DD, Groshen S. Salvage chemotherapy with CPT-11 for recurrent meningioma. J Neurooncol. 2006;78:271–276.

Chapter 36: Meningioma 73. Yano S, Kuratsu J. Kumamoto Brain Tumor Research Group. Indications for surgery in patients with asymptomatic meningiomas based on an extensive experience. J Neurosurg. 2006;105:538–543. 74. Olivero WC, Lister JR, Elwood PW. The natural history and growth rate of asymptomatic meningiomas: A review of 60 patients. J Neurosurg. 1995;83(2):222–224. 75. Simpson D. The recurrence of intracranial meningiomas after surgical treatment. J Neurol Neurosurg Psychiatry. 1957;20:22– 39. 76. Kinjo T, Al-Mefty O, Kanaan I. Grade zero removal of supratentorial convexity meningiomas. Neurosurgery. 1993;33(3):394– 399. 77. Condra KS, Buatti JM, Mendenhall WM, et al. Benign meningiomas: Primary treatment selection affects survival. Int J Radiat Oncol Biol Phys. 1997;39:427–436. 78. Mirimanoff RO, Dosoretz DE, Linggood RM, et al. Meningioma: Analysis of recurrence and progression following neurosurgical resection. J Neurosurg. 1985;80:191–194. 79. Barbaro NM, Gutin PH, Wilson CB, et al. Radiation therapy in the treatment of partially resected meningiomas. Neurosurgery. 1987;20:525–528. 80. Miralbell R, Linggood RM, De la Monte S, et al. The role of radiotherapy in the treatment of subtotally resected benign meningiomas. J Neurooncol. 1992;13:157–164. 81. Taylor BW Jr, Marcus RB Jr, Friedman WA, et al. The meningioma controversy: Postoperative radiation therapy. Int J Radiat Oncol Biol Phys. 1988;15:299–304. 82. Al-Mefty O, Kersh JE, Routh A, et al. The long-term side effects of radiation therapy for benign brain tumors in adults. J Neurosurg. 1990;73:502–512. 83. Glaholm J, Bloom HJG, Crow JH. The role of radiotherapy in the management of intracranial meningiomas: The Royal Marsden Hospital experience with 186 patients. Int J Radiat Oncol Biol Phys. 1990;18:755–761.

511

84. Pollock BE, Stafford SL, Utter A, et al. Stereotactic radiosurgery provides equivalent tumor control to Simpson Grade 1 resection for patients with small- to medium-size meningiomas. Int J Radiat Oncol Biol Phys. 2003;55:1000–1005. 85. Kollova A, Liscak R, Novotny J Jr, et al. Gamma knife surgery for benign meningioma. J Neurosurg. 2007;107(2):325– 336. 86. Tishler RB, Loeffler JS, Lunsford LD, et al. Tolerance of cranial nerves of the cavernous sinus to radiosurgery. Int J Radiat Oncol Biol Phys. 1993;27:215–221. 87. Barami K, Grow A, Brem S, et al. Vascular complications after radiosurgery for meningiomas. Neurosurg Focus. 2007;22(3):E9. 88. Debus J, Wuendrich M, Pirzkall A, et al. High efficacy of fractionated stereotactic radiotherapy of large base-of-skull meningiomas: Long-term results. J Clin Oncol. 2001;19:3547– 3557. 89. Jalali R, Loughrey C, Baumert B, et al. High precision focused irradiation in the form of fractionated stereotactic radiotherapy (SCRT) for benign meningiomas predominantly in the skull base location. Clin Oncol (R Coll Radiol). 2002;14:103– 109. 90. Henzel M, Gross M, Hamm K, et al. Significant tumor volume reduction of meningiomas after stereotactic radiotherapy: Results of a prospective multicenter study. Neurosurgery. 2006;59:1188– 1194. 91. Gautier-Smith C, Lamouche M. Parasagittal and Falx Meningiomas. London: Butterworths, 1987. 92. Quest D. Meningiomas: An update. Neurosurgery. 1978;3:219– 225. 93. MacCarty C, Taylor W. Intracranial meningiomas: Experiences at the Mayo Clinic. Neurologica Medico-Chir (Tokyo). 1979;19:569–574. 94. Rohringer M, Sutherland GR, Louw DF, et al. Incidence and clinicopathological features of meningioma. J Neurosurg. 1989;71:665–672.

37 Schwannomas of the Skull Base Daniel W. Nuss and Emily Lifsey Burke

considered in aggregate). Rarely, schwannomas involve the hypoglossal nerve, and extraocular motor nerves. Other sites have been reported as well; these are discussed in more depth later in the chapter. Both the genders are equally affected across the gamut of most anatomic sites, but intracranial schwannomas affect women more often than men, in a ratio estimated at 3:2 to 2:1. Schwannomas can occur at any age, but 75% occur in the third to fourth decades of life. There is no racial predisposition for schwannomas (3,5). Epidemiologically, it is essential to recognize that schwannomas occur across a spectrum of clinical scenarios. Certainly, schwannomas occur sporadically as isolated tumors in individuals with no history of familial disease, but they are more frequent in the setting of the neurofibromatosis syndromes. Therefore, a patient with a newly diagnosed or suspected schwannoma must be carefully evaluated for the possible presence of one of the genetic neurofibromatosis syndromes. This is true not only because early diagnosis is desirable, but because prognosis depends on accurate assessment, and the treatment options and decision-making are more complicated when multiple tumors are present, or expected to develop over the patient’s lifetime.

INTRODUCTION The schwannoma, in the simplest terms, is a tumor of nervesheath cells. While schwannomas of the skull base are uncommon, they are not rare. Intuitively this makes sense, given the rich network of nerves in the head and neck region. In fact, about 25% to 45% of all reported schwannomas are found in the head and neck (1,2), and since most of these originate from cranial nerves, a significant proportion will involve the skull base. Therefore, most practitioners in otolaryngology, neurosurgery, and related specialties who care for patients with skull base problems will encounter schwannomas in clinical practice. Despite their generally benign nature, schwannomas at or near the skull base can be responsible for life-changing morbidity due to associated neurologic deficits, and also due to frequent association with the numerous manifestations of the neurofibromatosis syndromes. They can be lethal as well, because of proximity to, or encroachment upon, neurovascular structures and the airway. Successful management of these benign neoplasms demands a sophisticated understanding of their origin, behavior, clinical features, and treatment options, along with clinical judgment that takes into consideration the natural history of the lesion as well as the likely morbidity of therapy. The purpose of this chapter is to provide the reader with a practical overview of the incidence, pathology, clinical features, treatment options, and outcomes for management of skull base schwannomas. Vestibular (eighth cranial nerve) schwannomas, discussed elsewhere in this text, will be excluded from this discussion. Specific surgical techniques, also not the focus of this chapter, will be addressed elsewhere.

The Neurofibromatoses There are now at least three major types (and eight or more subtypes) of neurofibromatosis syndromes: type 1 (NF1), type 2 (NF2), and the recently described entity referred to as schwannomatosis (6–9). As a matter of perspective, the NF1 syndrome is much more common than NF2. Worldwide, the incidence of NF1 is approximately 1 in 4000, irrespective of race, gender, or ethnic background. The incidence of NF2 has recently been estimated at around 1 in 25,000 (10); this recent estimate is higher than historical reports, likely because of increasing awareness of diagnostic criteria and perhaps in part because of improved diagnostic imaging. Because schwannomatosis has only recently been recognized as a separate entity, its real incidence is not yet well understood. A Finnish population-based review in 2000 (11) estimated the annual incidence at 1 in 1,700,000, but MacCollin et al. (12) more recently postulated that it may be just as common as NF2. All of these syndromes reflect an underlying genetic abnormality, discussed in detail below, which predisposes patients to develop tumors of the nervous system (see “Genetics and Molecular Features”). Specific criteria for diagnosis of NF1 and NF2, promulgated by The National Institutes of Health (13), are as follows:

INCIDENCE AND EPIDEMIOLOGY While the exact incidence and prevalence of benign schwannomas in all sites are not known, it is clear that nerve sheath tumors make up 8% to 10% of all intracranial, extra-axial tumors (3). Of those, schwannomas are the majority (5). As noted above, 25% to 45% of all schwannomas are found in the head and neck region. Although this discussion will focus on schwannomas affecting the skull base, it should be noted that schwannomas can occur in many other sites in the head and neck as well (face, skin, orbit, lips, maxilla, mandible, oral cavity, sinuses, parotid, nasopharynx, and larynx) (5). Schwannomas in general have a predilection for sensory nerves, although they do also occur along motor nerves. They arise most commonly (in descending order) from the vestibular component of the eighth nerve (>90%), sensory divisions of the trigeminal nerve, the facial nerve, and the nerves of the jugular foramen (cranial nerves IX, X, and XI

The criteria for the diagnosis of NF1 are met in an individual if two or more of the following signs are found: 513

514

Nuss and Burke

r Six or more café au lait macules larger than 5 mm in the greatest diameter in prepubertal children and larger than 1.5 cm in postpubertal individuals r Two or more neurofibromas of any type, or 1 plexiform neurofibroma r Multiple freckles (Crowe sign) in the axillary or inguinal region r A distinctive osseous lesion, such as sphenoid dysplasia or thinning of long bone cortex, with or without pseudoarthrosis r Optic glioma r Two or more iris hamartomas (Lisch nodules) seen on slit-lamp or biomicroscopy examination r A first-degree relative (parent, sibling, offspring) with NF1, as diagnosed by using the above criteria. To diagnose NF2, the following criteria must be met: r Bilateral vestibular schwannomas, or r A family history of NF2 (parent, sibling, or child) plus a unilateral vestibular schwannoma before age 30 r or any two of the following: ◦ glioma ◦ meningioma ◦ schwannoma ◦ juvenile posterior subcapsular lenticular opacity (juvenile cortical cataract).

Schwannomatosis In recent years, the separate clinical entity known as schwannomatosis has been defined (6–8,12,14–16). According to MacCollin and colleagues, “schwannomatosis is a recently recognized third major form of neurofibromatosis that causes multiple schwannomas without vestibular tumors.” Diagnosis is based on the presence of multiple schwannomas without the stigmata of neurofibromatosis NF1 or NF2. It has been estimated that patients with schwannomatosis may represent up to 5% of all patients requiring schwannoma resection (12). Epidemiologically the condition may be just as common as NF2, but for unclear reasons, it is not apparently familial. Schwannomatosis patients have been noted to have tumors of numerous sites, including intraspinal, paraspinal, brachial plexus, femoral nerve, sciatic nerve, calf, forearm, retroperitoneum, and middle cranial/infratemporal fossa region. The common presenting symptoms included paresthesias, palpable mass, pain, or weakness, with pain being the dominant clinical problem and indication for surgery. Huang et al., in reporting a series of six schwannomatosis patients, concluded that surgery is indicated for symptomatic lesions, while asymptomatic tumors should be followed conservatively (17). Because of the increased risk for developing multiple schwannomas, they recommend regular surveillance and consideration of genetic counseling, even though the exact mechanism of genetic alteration is unknown. The relative paucity of information on this condition makes it especially important that such patients be studied prospectively.

Associated Conditions and Considerations In discussing schwannomas, neurofibromas, and their associated NF syndromes, it should be emphasized that persons who have any of these clinical syndromes are members of populations who are at risk for other very serious problems, including tumors and nontumor conditions as well. Other tumors that occur with more frequency in these syndromes include gliomas of the optic nerve, astrocytoma, meningioma, intramedullary glioma, ependymoma, soft tissue sarcomas,

and juvenile myeloid leukemia. Also, some individuals with neurofibromatosis have below average intelligence; 25% to 40% have learning disabilities (e.g., attention-deficit hyperactivity disorder/ADHD, neuromotor dysfunction, visualspatial processing disorders); and 5% to 10% have mental retardation. Certain endocrine problems can also be associated with NF. Short stature, growth hormone deficiency, sexual precocity, and pheochromocytomas all can occur more commonly in NF patients than in the general population (1,18). Care of patients with any of the neurofibromatoses, for all the above reasons, crosses multiple disciplines and requires a high level of awareness among all involved clinicians.

PATHOLOGY Definition and Nomenclature Historically, many different names have been used to describe what is now classified as schwannoma, including neuroma, neurinoma, neurilemmoma, neurolemmoma, perineural fibroblastoma, and others. These older terms are now considered inaccurate and their use is discouraged, but they are acknowledged here since they do appear in older but stillimportant reference texts and clinical literature.

Distinguishing Between Schwannomas and Neurofibromas In any discussion of schwannomas, it is essential to distinguish the schwannoma from the one other common nerveassociated tumor that affects the skull base, namely the neurofibroma. They are both encountered with some regularity in clinical practice, yet the prognostic implications as well as the therapeutic decision-making are sometimes significantly different. Although schwannomas and neurofibromas do share certain clinical characteristics, as well as a common cell-type in the progenitor Schwann cell, the two entities are really quite distinct. Clinicians and pathologists alike must be mindful of the features that set them apart, for those features straddle both clinical and pathological lines. In actual practice, when all of the clinical and pathological features are taken into account, precise diagnosis is usually straightforward. Clinically, schwannomas are usually slow-growing masses that may be sensitive or painful, especially when they are in locations subject to pressure or manipulation (e.g., the upper neck or parapharyngeal space), and they are often tender to palpation. Neurofibromas are usually asymptomatic. The Schwann cell is the parent cell of both schwannomas and neurofibromas. The Schwann cell is a sheath cell, not a nerve cell. Schwann cells, which can be regarded as the nervous system’s cellular maintenance team, are important for nerve sheath integrity and myelin production. According to Batsakis (1), the Schwann cell is regarded by many neuropathologists as homologous with the oligodendroglia of the central nervous system, and it is believed to be derived from the neural crest, and therefore considered as neuroectodermal in origin. This is supported by immunohistochemical findings (see below), with schwannoma strongly positive for S-100 protein, which is helpful in distinguishing schwannoma from other soft-tissue tumors. Pathologically, the typical schwannoma is an encapsulated, solitary, and expansile tumor that is often fusiform and attached to a recognizable nerve (Fig. 1). On cut section, they are usually mostly white or yellowish-white and firm. The tumor appears to push axons aside rather than enveloping them (“centrifugal” growth), at least early on in its growth. Initially fusiform, these tumors at the skull base tend to assume the shape of the confining space in the area of origin,

Chapter 37: Schwannomas of the Skull Base

Figure 1 Gross specimen of resected schwannoma from parapharyngeal space at base of skull. Note fusiform shape tapering to nerve trunk on each end.

and then progressively erode adjacent bone. When in proximity to a foramen or fissure of the skull, they assume a bi-lobed or “dumbbell” shape with expansion on either side of the bony threshold [Fig. 2(A)and 2(B)]. Usually there is a distinctive biphasic histological pattern of well-developed cylindrical structures (known as Antoni type A tissue), with palisading nuclei around a central mass of cytoplasm, the so-called Verocay body (1,5,19). These well-recognized features often are embedded within a more loosely textured, less well-developed, hypocellular stroma in which fibers and cells form no distinctive pattern (Antoni B tissue) (1). Another very common feature of schwannomas is their rich vascularity. As discussed below, this gives rise to certain distinctive histological and radiographic characteristics, as well as the tendency for schwannomas to bleed. Fine pericellular reticulin can be seen with special stains. Histopathological findings of schwannoma are depicted in Figure 3. “Retrogressive features” (1) are very common in schwannomas, including necrosis, areas of cystic degeneration, and vascular changes. Owing to their rich supply of blood vessels that have a tendency to be abnormal, schwannomas usually feature angiomatous clusters of blood vessels that are prone to focal thrombosis, and areas of recent and old hemorrhage. The term “cystic schwannoma” is applied to tumors that, because of the retrogressive/degenerative

515

phenomena, are filled with serous fluid. The related term, “ancient schwannoma” describes tumors in which there is extensive hyalinization, again, believed to be related to degenerative phenomena. These kinds of features can often be appreciated even on examination of the bisected gross tumor, but are more apparent at light microscopy. Sometimes these retrogressive features, because of the resulting inhomogeneity within a given tumor, can be misinterpreted to represent malignancy, which is extremely rare in true schwannomas (see discussion on malignancy below), and must be evaluated with caution. Neurofibromas do not share most of the above features. Neurofibromas do not have the collagenous capsule that schwannomas have; they are instead surrounded by thickened perineurium and epineurium. They incorporate axons into the tumor and envelop them (“centripetal” growth). Typically they are more cellular, manifesting a spindle cell pattern of growth. They do not often exhibit retrogressive features. Neurofibromas do not generally display the Antoni type A and B patterns, nor do they have Verocay bodies as schwannomas do. Instead, the matrix of the tumor is made up of type IV collagen fibers and myxoid material. The interspersed cell populations (Schwann cells, fibroblasts, and perineural cells) are scattered within this matrix. Many are seen in association with the overall clinical picture of generalized neurofibromatosis. And as discussed below, malignant degeneration is well established with neurofibromas.

Immunohistochemistry and Special Stains As derivatives of embryonic neural crest cells, Schwann cells are of neuroectodermal origin. Consequently, tumors derived from these cells will stain strongly positive for S-100 protein on immunohistochemistry. While S-100 protein is not specific for schwannoma, it is helpful in diagnosis, since schwannomas will stain much more heavily for S-100 than will neurofibromas. (Other cell types that are S-100 positive include glial cells, melanocytes, chondrocytes, adipocytes, myoepithelial cells, macrophages, Langerhans cells, dendritic cells, and keratinocytes, all of which are of neuroectodermal origin.) The immunoreactivity for S-100 protein is seen in fewer cells making up the neurofibroma, as opposed to more blanketed reactivity of the schwannoma. Trichrome and Alcian blue stains will highlight the collagen and mucinous matrix of neurofibromas (3,5,18,19). Schwannomas may also variably stain positive for GFAP (glial fibrillary acid protein), keratin, and EMA (epithelial membrane antigen), but because of this variability, they are rarely helpful in diagnosis.

Genetics and Molecular Features

Figure 2 (A) Parasagittal T1-weighted MR image of trigeminal schwannoma demonstrating bilobed or “dumbbell ” shape typical of schwannoma that develops on both sides of a foramen or fissure. Note that the tumor is heterogeneous in its signal characteristics, another common feature of schwannomas (see text). (B) Axial T2-weighted MR image of same patient, showing dumbbell shape with smaller portion of tumor in posterior fossa (slightly compressing brainstem), and larger portion in middle fossa. Tumor demonstrates higher signal intensity on T2 series than on T1 (see text).

Much has been learned in recent years about the genetics of neurofibromatosis, and in aggregate, discoveries concerning the neurofibromatosis syndromes have provided unique insight into the genetic basis of tumor formation in general. Neurofibromatosis type 1 (NF1) has been traced to a defect in the neurofibromin gene and has an autosomal dominant inheritance pattern. This is a large gene located on chromosome 17. The fact that the gene is of large size may be the cause of the high number (40%) of sporadic cases of NF1; defects in the germ cell line cause the sporadic cases of disease. The neurofibromin protein normally functions to inhibit Ras, which is commonly activated in human cancers. The loss of function of neurofibromin causes uninhibited activity of Ras and thus unregulated cell growth. NF1, also known as von Recklinghausen disease, results in neurofibromas of cranial and (more commonly) peripheral nerves, frequently arising

516

Nuss and Burke

Figure 3 Histopathological sections demonstrating typical features of schwannoma. Biphasic pattern (A) with Verocay bodies often seen in the compact Antoni A type tissue, along with pallisading nuclei (B). (C) Nuclear pleomorphism, which can be seen as part of so-called ancient change, sometimes mistaken as features of malignancy. (D) Microcystic changes, inflammation, and hemosiderin characteristic of schwannoma. (E) Strong, diffusely positive stain for S-100 protein. (F) Fine pericellular reticulin. Source: From Ref. 19.

in the skin. Overall, the penetrance of NF1 is variable, from very mild to severe. Neurofibromatosis type 2 is caused by a defect in the merlin or schwannomin gene, located on chromosome 22. The loss of chromosome 22q is seen in sporadic cases of schwannomas, but the NF2 defect usually causes premature truncation or complete deletion of the entire gene. Schwannomin acts differently than neurofibromin, in that it is a membrane/cytoskeleton protein involved in cell motility and proliferation, but it also affects Ras/Rac activity. The hallmark of this disorder is bilateral vestibular schwannomas, and affected persons may also have peripheral schwannomas and meningiomas. There is a spectrum of disease, but in general, NF2 is considered a more severe and morbid disease than NF1. The newer entity of schwannomatosis has also been traced to a defect in the schwannomin gene on chromosome

22 as its cause. What distinguishes this from NF2 has not yet been determined, but the two seem to have significant molecular overlap. In all of the neurofibromatosis syndromes, the affected genes basically fail to perform in their protective role as tumor suppressors. NF-related tumorigenesis takes place in what has become known as the “two-hit” model of genetic disease. As such, one copy of the gene, inherited from a parent, carries the mutation, but when another “hit,” or spontaneous mutation on the normal (second) copy occurs, the end result is a failure of tumor suppression. Eventually tumors become clinically manifest. While a more detailed discussion of the molecular genetics of neurofibromatosis is beyond the scope of this chapter, the molecular basis of these disease processes is fascinating and has provided invaluable information on tumor biology (14,18,19).


517

Figure 4 Malignant peripheral nerve sheath tumor. (A) Clinical photograph of patient with NF type I, who experienced sudden and dramatic growth of longstanding facial tumor, ultimately diagnosed as malignant peripheral nerve sheath tumor. (B) CT of the same patient showing deeply infiltrating nature of tumor with invasion of temporal fossa, orbit, skull, temporomandibular joint and facial soft tissues.

Malignancy and Schwannomas Although all neurofibromatosis patients will develop benign tumors, approximately 5% to 10% will also develop malignant nerve-sheath tumors. A smaller number will develop other malignancies including leukemia and soft tissue sarcomas. Therefore, appropriate suspicion and surveillance for malignancy must be part of their regular follow-up. The nomenclature of nerve-sheath malignancies is sometimes a source of confusion and even among experts is not without some controversy. Pathologists and surgeons have historically applied the terms differently, leading to many clinical descriptions of so-called “malignant schwannoma” in cases for which pathologists would prefer to use other terms. The explanation for this relates to the process by which these malignant tumors develop. Undoubtedly the majority of malignant neoplasms of the peripheral nerves are of Schwann cell origin. However, the majority of the pathological literature on the subject dismisses any relationship between the benign schwannoma and any malignant degeneration. In other words, benign schwannomas do not degenerate into malignant tumors, or at least if they do it happens exceedingly rarely (1,5,20), and it is nearly always in patients who have neurofibromatosis (21). Therefore, the term “malignant schwannoma,” technically speaking, should rarely be used, even though it often appears in clinical case reports. “Neurogenous sarcoma” and more recently “malignant peripheral nerve sheath tumor” (MPNST) have been proposed as more accurate terms for malignant tumors of peripheral nerves (1,5). While schwannomas are rarely if ever associated with malignant degeneration, neurofibromas frequently are. Malignant nerve-sheath tumors have clearly been linked to neurofibromatosis, and have consistently been observed evolving from established neurofibromas, especially in the clinical setting of generalized neurofibromatosis (as opposed to solitary neurofibroma cases). The incidence of malignancy in association with neurofibromas is generally accepted to be around 8%, with estimates ranging from 5% to 16.4% (22,23). Malignant peripheral nerve-sheath tumors (MPNSTs) are most appropriately classified as highly aggressive softtissue sarcomas, and are often lethal. Their clinical course is different from most schwannomas and neurofibromas. Unlike the benign nerve-sheath lesions, malignant ones often

present with rapidly growing, painful tumors. They may reach extreme proportions in short periods of accelerated growth [Fig. 4(A)and 4(B)]. Because this chapter is devoted to benign skull base schwannomas, a more detailed discussion of malignant peripheral nerve tumors will not be included here. The reader is referred to several excellent references for more information on that subject (24–26).

RADIOGRAPHIC FEATURES OF SKULL BASE SCHWANNOMAS As with all skull base lesions, radiographic imaging of schwannomas must be thorough in order to give the best chance of accurate diagnosis. Both CT and MRI should be obtained, as their findings are complimentary; images should be obtained in three planes.

Features Common to Both CT and MR On both CT and MRI, schwannomas may be heterogeneous because of their tendency for cystic degeneration and intratumoral hemorrhage, as detailed above in the section on pathology. Smaller tumors (less than 1.5 cm) tend to be mostly solid, but larger ones will have both cystic and solid areas (27). The solid component of a schwannoma will enhance brightly with contrast due to inherent vascularity, but the cystic components may not. Most schwannomas will be well-defined lesions with smooth borders. The dumbbell shape, along with smooth expansion of bony foramina, is highly suggestive of, if not pathognomonic for, schwannoma [Figs. 2(A), 2(B), 5, and 6].

CT Characteristics CT images typically demonstrate lesions that are isodense to hyperdense compared to muscle, although cystic components will be hypodense. Thin sections are essential to accurately assess bone erosion and related bony anatomy of the adjacent skull. Most schwannomas will erode bone gradually and smoothly, enlarging foramina of affected nerves, and causing remodeling of adjacent bone as tumor expands. Often there will be a sclerotic rim of bone at the margins. In many cases, the affected foramen will be greatly expanded

518

Nuss and Burke

MR Characteristics On MRI, most schwannomas will appear hypo- to isointense relative to brain (white matter) on T1 sequences, and iso- to hyperintense on T2 sequences (27,28) [Figs. 2(A), 2(B), and 5(C)].

CLINICAL FEATURES OF SKULL BASE SCHWANNOMAS General Considerations Regarding Nonvestibular Schwannomas Anatomic Distribution Nonvestibular schwannomas (NVSs) (i.e., tumors of nerves other than cranial nerve VIII) have been reported to involve virtually every nerve associated with the skull base, including all of the cranial nerves as well as the sympathetic, parasympathetic, and upper cervical nerves. They may develop as purely intracranial, purely extracranial, or transcranial entities. The clinical symptoms and differential diagnosis, discussed in the sections that follow, vary with the specific site of origin and direction of growth.

Natural History

Figure 5 (A) Axial CT and (B) axial T1-weighted MR of patient with V2 schwannoma. In this case, the tumor has widened the pterygomaxillary fissure, with bilobed growth into infratemporal fossa and nasal cavity. (C) Coronal T2-weighted MR demonstrating dumbbell tumor of V2 (maxillary division of trigeminal) straddling pterygomaxillary fissure on patient’s right side. Arrow points to the normal V2 on the patient’s left side. V2 is not visible on right side because it is the nerve of origin of tumor.

such that it may be difficult to recognize, or may appear to be absent (Figs. 5 and 6). Ancient schwannomas may show irregular calcification.

Patients with neurofibromatosis type 2 (NF2) are at particularly high risk of developing schwannomas, and much can be learned about schwannomas by examining this population. As part of the NF2 Natural History Consortium project, Fischer et al. (29) studied the prevalence and location of nonvestibular cranial nerve schwannomas among NF2 patients. Magnetic resonance imaging findings were prospectively gathered for 83 patients, over three consecutive annual evaluations. More than half (51%) were found to have nonvestibular schwannomas (NVSs) of the cranial nerves. Of these, 25 (60%) also had cranial meningiomas, and 21 of those without NVS (25% of 83) had at least 1 meningioma. Several other interesting observations came from Fischer’s review. The average size of the incidental NVS was small, at 0.4 cm3 , and overall, there was no significant change in NVS during the roughly 3 year interval of the study. The most common locations of the NVSs in this NF2 population were oculomotor and trigeminal. Neuropathies associated with tumors of the upper cranial nerves were few in this series. In contrast, lower cranial nerve schwannomas were

Figure 6 CT-derived images of a 4-year-old patient with V3 tumor, showing extensive smooth expansion of foramen ovale, typical for schwannoma. (A) and (B), Three-dimensional CT reconstructions showing intracranial and extracranial views, respectively. Arrows point to normal foramen ovale on unaffected side. (C) Coronal CT scan showing extensive expansion of foramen ovale and also considerable remodeling of adjacent mandible due to tumor in infratemporal fossa.


often symptomatic, associated with swallowing difficulty, aspiration, and other sequelae. Mautner, Samii, and colleagues studied the clinical spectrum of disease, and radiographic findings, in a population of 48 NF2 patients, using a protocol that included gadolinium-enhanced MRI of the brain and spine as well as neurological, dermatological, and ocular examinations (30). Their study revealed that 96% had vestibular schwannomas, 90% had spinal tumors, 63% had subcapsular cataracts, 58% had meningiomas, and 29% had trigeminal schwannomas. They emphasized that the incidence of spinal tumors, a major source of morbidity and mortality, was higher in this report than in previous studies. These reviews illustrate that NF2 patients are highly likely to develop nonvestibular schwannomas, and that these patients are also very likely to develop multiple nonschwannoma tumors (especially spinal tumors, neurofibromas, and meningiomas) over time. This latter observation underscores the importance of long-term clinical follow-up with serial imaging, and it also emphasizes the need for careful consideration of all treatment options, including observation, for newly diagnosed schwannomas in NF2 patients.

Schwannomas in Children For pediatric patients with schwannomas in the setting of NF2, the multiplicity and variety of tumors—and the resulting morbidity—appears to be even worse than for adults. Nunes and MacCollin reported their findings from a series of 12 patients with neurofibromatosis 2 (31). One-third of the patients presented with hearing loss and another third presented with other cranial neuropathies. Tumor-related disability in many patients was high, with documented cranial meningiomas in 75%, cranial schwannomas other than vestibular schwannomas in 83%, and spinal cord tumors in 75%. Functionally, 75% of children had hearing loss, 83% had visual impairment, 25% had abnormal ambulation, and 25% were performing below their academic grade level. The authors concluded that our ability to diagnose pediatric patients with NF2 is improving, but outcomes appear to be significantly worse than in adult patients, and that work is needed to determine optimal management of pediatric neurofibromatosis 2.

Symptoms Vary with Location Schwannomas of the skull base exhibit a wide variation in the degree to which they produce signs and symptoms. While some tumors are essentially asymptomatic, others produce local symptoms that are directly attributable to the tumor’s nerve of origin. Still others become symptomatic by virtue of the pressure they place on surrounding structures, while the nerve of origin may remain free of any demonstrable signs of deficit until later in the course of progression. One of the chief determinants of a tumor’s clinical impact is whether the tumor volume is mostly intracranial, mostly extracranial, or a combination of both. In fact, symptoms may be more related to the tumor’s pattern of growth than to its nerve of origin. As a general rule, when the tumor’s volume is mostly intracranial, the patient is more likely to have symptoms that are global or due to central nervous system effects; when most of the tumor’s volume is extracranial, the likelihood of specific focal cranial nerve deficits increases. Many schwannomas exhibit combinations of intra- and extracranial effects. Because of these observations, it is useful to consider schwannomas not only according to the nerve of origin, but

519

also according to the anatomic compartments in which they occur and subsequently grow.

Specific Anatomic Sites By far the most common cranial nerves of origin for nonvestibular skull base schwannomas are the trigeminal (CN V), the facial (CN VII), and the jugular foramen group (CN IX, X, and XI). Therefore, these sites will be discussed in detail before addressing the other cranial nerve sites and noncranial nerve sites of origin, which are rare.

Trigeminal Nerve Schwannomas The trigeminal nerve is the second most common site for skull base schwannomas, after the vestibular nerve. The precise incidence is not known, but Kouyialis et al. in 2007 stated that a literature review dating back to 1935 turned up at least 580 published cases of trigeminal nerve schwannomas (TNSs) that had been surgically treated; undoubtedly many more were unreported or not treated surgically (32). Symptoms vary with exact site of origin, and manifestations are many, reflecting the long course and rich distribution of the trigeminal nerve. Patients with TNS present most often with trigeminal nerve-related dysfunction, including facial pain, headache, and numbness of the affected nerve segment(s). The character of symptoms varies considerably. In some patients pain is predominant and may be sharp or dull, intermittent or constant. In others, sensory disturbance is the chief concern. Patients often describe a sensation of numbness, burning, creeping, pins-and-needles, or other vague dysesthesias. Initially symptoms may be focal, but as tumor grows, deficits may become manifest in all three divisions. TNSs may develop essentially anywhere along the course of the trigeminal nerves, including the root, the ganglion, and the peripheral branches. According to Eisenman, the majority of trigeminal schwannomas develop in the gasserian ganglion (33), where they expand gradually from Meckel’s cave and initially involve only the middle fossa. With further enlargement, they may extend into the posterior fossa secondarily, with one component or the other being larger, creating a dumbbell shape [Figs. 7 and 2(B)]. Petrous apex erosion is common with such tumors, usually preserving the internal auditory canal, a finding which helps distinguish them from vestibular schwannomas (34).

Figure 7 Axial MR of patient with large trigeminal schwannoma involving posterior fossa (with brainstem compression) and middle fossa (at region of gasserian ganglion).

520

Nuss and Burke

and width of petrous erosion, classifying tumors as noted below (37): Type M, tumor confined to middle fossa Type Mp, tumor mainly in middle fossa Type M = P, tumor equally distributed into both middle and posterior fossae Type Pm, tumor mainly in cisternal space of posterior fossa

Management of Trigeminal Nerve Schwannomas

Figure 8 Coronal CT of giant V3 schwannoma extensively involving middle cranial fossa and infratemporal fossa. The patient reported only minimal numbness and no other symptoms.

TNSs may also arise directly from the trigeminal root near the brainstem in the posterior fossa. When tumor is confined to the posterior fossa, symptoms may mimic cerebellopontine angle tumors, causing hearing loss, vertigo, tinnitus, and facial weakness, in addition to trigeminal dysfunction, but in such cases the trigeminal dysfunction usually overshadows other deficits. Still other TNSs will arise more distally, and may even be entirely extracranial, in which case symptoms are very site-specific. For example, V1 tumors may arise either in the cavernous sinus or in the orbit, where they cause diplopia and proptosis due to pressure; the corneal reflex may be absent. They may also arise in V2, causing midfacial or palatal numbness, pain, and dysesthesias. Some patients may have xerophthalmia due to reduced lacrimation. V3 tumors result in numbness, pain, and dysesthesias of the lower face and jaw as well as chewing problems, malocclusion, and atrophy. Such tumors may become quite large as they extend into the infratemporal fossa, where there is ample room for slow growth before symptoms appear. Typically slow-growing, they may reach impressive proportions before some patients perceive symptoms (Fig. 8). Distal trigeminal schwannomas may also present in the frontal, ethmoid, sphenoid or maxillary sinuses, pterygopalatine fossa, or even in subcutaneous or submucosal tissues. Rarely, TNSs may involve more than two compartments, typically middle fossa, posterior fossa, and infratemporal fossa simultaneously (32).

Radiographic Appearance of Trigeminal Schwannomas A variety of radiographic appearances are possible for TNS, depending on the site of origin and direction of growth (Figs. 2, 5–8).

Staging of Trigeminal Schwannomas The generally accepted staging system used for TNS is that adopted by Samii et al. (35,36), summarized here: Type A: tumor predominantly middle fossa Type B: tumor predominantly posterior fossa Type C: tumor in both middle and posterior fossae Type D: tumor predominantly extracranial Gwak et al. proposed a different staging system, focusing on degree of petrous erosion, and felt that their system helped predict which surgical approach would be most successful. They used CT and MR to determine tumor diameter

Historically, TNS management has mostly been surgical, which is understandable since many patients initially present with symptoms caused by tumor compressing vital structures. Reports of successful radiation treatment are increasing in frequency, and will be discussed in the section on radiation. Surgical approaches are determined on the basis of tumor location and extent. A myriad of techniques and approaches have been advocated in the literature. The generally accepted approaches are as follows: Type A tumors are approached via temporal craniotomy, often with zygomatic or orbitozygomatic osteotomy for more basal access (Fig. 9). Type B tumors are often approached via suboccipital craniotomy or variations thereof. Type C tumors are best treated using combined middle-fossa/posterior fossa approaches such as combined subtemporal and presigmoid craniotomy. Successful one-stage resection can be accomplished (38–40), and appears to be becoming the standard of care, but some of these larger tumors may be best managed with two-stage operations, and treatment must be individualized. Type D tumors that are infratemporal are usually removed via infratemporal fossa approaches (33). Those Type D tumors that present in other areas, such as the pterygopalatine fossa, orbit, or sinuses, can be approached using transfacial, degloving, endoscopic transnasal, orbitotomy approaches, or combinations thereof (41–46). Details of surgical technique are reviewed elsewhere in this text.

Outcomes for Surgical Treatment of Trigeminal Nerve Schwannomas As in other areas of skull base surgery, major advances have been realized in the past two decades for TNS. Zhou et al. published their single-institution series of 57 cases of dumbbellshaped schwannomas in 2007 (39). Their analysis was divided into two groups, one treated pre-1984 using traditional (“non-skull base”) approaches, and the other after 1984, using modern microsurgical combined approaches. The two groups differed substantially in outcomes, with the pre-1984 cohort faring worse in all respects, most notably likelihood of total tumor resection (42% vs. 87%); long-term cranial nerve morbidity (55% vs. 18.6%); recurrences (3 of 12 vs. 1 of 45); and overall performance status. The authors concluded that the best treatment for large or giant TNS is microsurgery, via single-stage skull-base craniotomy (using a technique which they called an extraduro-transduro-transtrigeminal pore approach). The authors felt it was not necessary to resect the petrous apex for removal of the tumor in the posterior fossa. They recommended that radiosurgery be reserved for residual or recurrent tumors. Al-Mefty et al. described experience with 25 large dumbbell TNS in 2002, emphasizing a philosophy of singlestage, total tumor resection via an extradural zygomatic middle fossa approach through the expanded Meckel cave (40). All tumors involved the cavernous sinus, and most had more than one preoperative cranial nerve deficit, including abducens paralysis in 40%. In patients who had not undergone previous surgery, the preoperative trigeminal sensory


521

Figure 9 Overview of orbitozygomatic-temporal approach as would be used for Type A trigeminal schwannoma. (A) Exposure for temporal craniotomy. (B) Orbitozygomatic osteotomies to facilitate infratemporal exposure. (C) Tumor (arrows) in infratemporal fossa seen protruding from foramen ovale before craniotomy [same orientation as in 9(A)]. (D) Preoperative (left) and postoperative MR images, demonstrating complete resection. (E) Appearance of patient approximately six months after surgery. Temporal fossa depression resulted from the use of the temporalis muscle as a reconstructive flap, and was not due to atrophy; patient declined temporal fossa reconstruction and remains tumor-free more than five years later.

522

Nuss and Burke

deficit improved in 44%, facial pain decreased in 73%, and trigeminal motor deficit improved in 80%. Among patients with preoperative abducent nerve paresis, recovery was attained in 63%. Only three patients (12%) experienced a persistent new or worse cranial nerve deficit, all confined to the trigeminal, postoperatively. Three patients experienced recurrences. The authors noted that, as with patients who have vestibular schwannomas, advances in surgical procedures have markedly improved outcomes in patients with trigeminal schwannomas, and that “preservation or improvement of cranial nerve function can be achieved through total removal of trigeminal schwannoma” (40). In a similar study that included not only 13 cases of TNS but also 27 other cases of benign nonmeningeal cavernous sinus tumors (47), improved cranial nerve outcomes were also noted. Postoperatively, 89.7% of the patients had either stable or improved extraocular muscle function compared with their preoperative statuses. Forty percent of the patients experienced improvement of their preoperative extraocular muscle deficits. From this experience, it was felt that with microsurgical skull base techniques, benign nonmeningeal tumors of the cavernous sinus, including schwannomas, can be safely and radically removed, and with good long-term neuro-ophthalmological function and low morbidity. In 2007, Pamir et al. reviewed a similar experience with 18 cases of TNS treated since 1992. Total excision was achieved in 17 of 18 cases, with minimal morbidity (36). Schwannomas affecting the orbit present special problems. TNS involving V1 and sometimes V2 can directly involve the orbit, with ocular complaints as the presenting symptoms. A wide spectrum of clinical findings can be observed (48), but usually progressive proptosis is the earliest sign (49). Vision is not usually impacted early on, but direct optic nerve compression, globe indentation with induced hyperopia, or increased intracranial pressure with optic nerve compromise may be responsible for visual decline as tumors progress. Careful documentation of neuro-ophthalmological findings is essential in all patients with trigeminal schwannomas. The ophthalmologist must play an active role in the multidisciplinary team to ensure optimal assessment, treatment planning, and visual outcomes, including rehabilitation of gaze deficits. In surgical management of TNS, the philosophy of total tumor resection appears to be extremely important. Multiple reviews have noted that when gross total tumor removal has not been accomplished, TNS tumors are highly likely to recur. In Gwak’s series of 29 consecutive TNSs, recurrence was noted in 8 of 10 patients in whom total resection was not achieved (37). Moffat et al. reported a series of eight TNS patients, among whom they elected to do subtotal resection in five (50). In those cases, tumor was left behind with the goal of minimizing postoperative deficit and improving quality of life. All five recurred and required revision surgery. Surgical resection remains the mainstay of treatment for trigeminal schwannomas. With advances in microsurgical techniques, creative skull base approaches, endoscopy and related technological improvements, morbidity of resection is decreasing, and the likelihood of improvement of preoperative neurological deficits is increasing. The role of radiation will be discussed later in this chapter.

Facial Nerve Schwannomas Facial nerve schwannomas (FNS) are the next most frequent after trigeminal nerve schwannomas, and are also uncommon. A literature review in 1996 revealed just over 300 published cases (51), but since that time over a hundred new cases have been added, probably reflecting increasing diagnostic

accuracy and awareness. FNS may occur anywhere along the course of the facial nerve, and like schwannomas elsewhere, symptoms are dependent upon which area is involved and to what extent the tumor has grown. Facial paralysis is the eventual hallmark of these tumors and a major source of morbidity among affected patients. Like the trigeminal nerve, the facial travels a long course and is anatomically complex in its distribution. Dort and Fisch (52) classified FNS tumors as intracranial, intratemporal, and extratemporal, with symptoms being different for each. Intracranial lesions (i.e., at the cerebellopontine angle) and those in the internal auditory canal frequently result in sensorineural hearing loss, tinnitus, and vestibular symptoms because of compressive effects on the eighth nerve. Intratemporal tumors often present with facial palsy and with conductive hearing loss, which occurs as tumor expands into the middle ear space. Extracranial tumors basically present as parotid or retromandibular masses at the skull base. Schaitkin authored a scholarly review of FNS in 2000 (53), and the reader is referred to that source for a detailed historical summary of reported cases of FNS. In his review, and those of O’Donaghue and Wiggins (54,55), the frequency of site of origin was examined. The majority of FNS tumors originate in the region of the geniculate ganglion (up to 83%), followed by the labyrinthine and tympanic segments of the facial canal (54% for both). Nearly 30% of these tumors will cause erosion of the otic capsule, visible on CT. Very few FNSs appear to originate in the CP angle, unlike vestibular schwannomas, but when they do they may be indistinguishable from VS both clinically and radiographically. Schaitkin observed that multiple segments of the facial nerve are commonly involved at diagnosis, most likely because FNS tumors are so indolent that they become extensive before they cause symptoms. In fact, multiple authors have noted that FNSs are extremely slow-growing tumors (56,57), in some cases with many years transpiring before definitive diagnosis is made. A significant number of FNSs have been noted at autopsy without ever having caused clinical symptoms (53,58). Lipkin reviewed 238 previously reported cases of FNS, and many of the findings were similar to those of schwannomas at other sites (59). The mean age at diagnosis was 39 years; there was no gender predilection; there was no distinct laterality of incidence; and the exact location of the tumor was the best predictor of symptomatology.

Clinical Features of Facial Nerve Schwannomas The most common clinical presentation of FNS is that of slowly progressive facial paralysis, which may sometimes be preceded by facial twitching, spasm, tic, or pain. This evolution is in contradistinction to the classic Bell’s palsy, which presents as facial palsy of sudden onset and rapid progress. That said, however, in several of the FNS series, a minority of patients did not have slowly progressive paralysis but instead had either sudden complete paralysis (14–21%) or recurring ipsilateral paralysis (up to 10%) (53,59,60). Thus, facial nerve tumors may mimic Bell’s palsy in a minority of cases, and clinicians must be cognizant and thorough in evaluation and follow-up of such patients. Otologic symptoms including hearing loss, tinnitus, and disequilibrium may be prominent in some patients with FNS. Although facial nerve symptoms usually occur first, Lipkin found that up to 13% had tinnitus initially. McMenomy reported on 12 FNS patients in whom 100% had hearing loss, 50% had tinnitus, and none had facial nerve symptoms, essentially presenting with symptoms indistinguishable from


523

praxia only, but paralysis has been of a slowly progressive nature.

Radiologic Characteristics of FNS

Figure 10 Imaging of facial schwannoma involving multiple segments of nerve. (A) Axial T1-weighted, contrast-enhanced MRI at the level of the leftside CPA. Left-side facial schwannoma is seen as an enhancing tumor mass involving the CPA, IAC, labyrinthine segment (arrowhead), and geniculate fossa (arrow). The overall shape of the tumor is that of a dumbbell. (B) Axial computed tomography scan filmed in bone window at the same level as in A. The widened labyrinthine segment of the facial nerve (arrowhead) and expanded geniculate fossa (arrow) are evident. Source: From Ref. 66.

vestibular schwannomas (61). (These differences in reported incidence may reflect practice/referral patterns more than actual incidence, but they are noteworthy nonetheless. The point is that FNS can mimic VS, Bell palsy, and other entities.) Park et al. (62) emphasized the diagnostic dilemma of preoperatively distinguishing FNS from VS, and analyzed multiple parameters including preoperative symptoms, pure-tone audiometry, auditory brainstem response, caloric test, electroneuronography, and magnetic resonance imaging; ultimately none could reliably predict FNS. Kubota et al. reported two cases of facial nerve schwannoma in which there was no paralysis despite massive tumor involving both middle and posterior fossa, with the only symptoms being otologic (63). The diagnostic dilemma applies as well to intraparotid FNS, where definitive preoperative diagnosis is rare. Fine needle biopsy of FNS is often inconclusive because these tumors are hypocellular or heterogeneous. There have traditionally been no universally accepted imaging criteria that favor FNS over other more common tumors of the gland. Recently, however, Shimizu et al. described the so-called “target sign” in which T2-weighted MRI images reveal higher signal intensity around the tumor periphery, which they believe is suggestive of schwannoma (64). Whether this will be of benefit in the prospective recognition of FNS is not yet clear. At present, most intraparotid schwannomas are not diagnosed definitively before surgery. Caughey suggested that schwannoma should be suspected if the facial nerve cannot be found intraoperatively or if the tumor is intimately associated with the facial nerve (65). Other regional signs and symptoms of FNS include 11% with pain in or around the ear, 13% with visible mass in the auditory canal (caused by tumor in vertical segment), and 6% with otorrhea. Occasional patients have reported disturbed taste and salivation when there is involvement of the chorda tympani. With schwannoma, as with other causes of facial weakness, electrophysiological testing may provide helpful clues. Schaitkin has recommended a protocol for facial paralysis evaluation that includes facial EMG and evoked EMG (EEMG) (53). He proposed that tumor should be suspected if (i) prolonged conduction latency is present on EEMG, even with normal amplitude; (ii) a patient has an incomplete facial lesion with normal EMG but EEMG amplitude less than 10%, and (iii) there is EEMG and EMG evidence of neuro-

Schwannomas of the facial nerve share the same general characteristics as schwannomas at other sites. As with trigeminal and jugular fossa schwannomas, it is essential to obtain highly detailed, thin-section images in three planes, including CT and MRI, with contrast. Typical findings include increased diameter of any segment of the facial nerve canal in a fusiform shape; sharply defined bone erosion around the geniculate ganglion, otic capsule, or other nerve canal segment; and middle ear mass, cerebellopontine angle mass, or parotid mass (55,65,66). Various imaging appearances are shown in Figures 10 to 12. Often, intraparotid FNS will be contiguous with tumor involving the vertical and sometimes horizontal segments of the facial nerve in the temporal bone, and the facial canal in the mastoid will be widened (Fig. 13).

Figure 11 Imaging of facial nerve schwannoma continued. (A) Axial CT scan at the level of the epitympanum. Left-side facial schwannoma involving the geniculate fossa (closed arrow) and tympanic segment of the facial nerve (arrowheads) can be seen pedunculating into the middle ear. The lateral displacement of ossicles by tumor (open arrow) is evident. (B) Axial T1weighted, contrast-enhanced MRI at the same level as in Fig. 4(A) shows the schwannoma as an avidly enhanced oval mass (arrow). Source: From Ref. 66.

524

Nuss and Burke

Figure 12 Imaging of facial nerve schwannoma continued. (A) Axial CT scan filmed in bone window at the level of the mid-mastoid segment of the facial nerve canal. The left-side mastoid segment facial schwannoma is seen as an oval mass breaking into surrounding mastoid air cells (arrow). The tumor also dehisces the posterior wall of the external auditory canal (arrowhead). When the schwannoma breaks into surrounding mastoid air cells, it gives the impression of irregular “invasive” tumor margins. (B) Coronal CT scan filmed in bone window shows the same tumor in its craniocaudal extent. The enlarged stylomastoid foramen (arrow) is easily identified from this vantage point. (C) T1-weighted, contrast-enhanced coronal MRI of the same tumor. The inferior limit of tumor enhancement is the stylomastoid foramen (arrowhead), differentiating the schwannoma from a primary parotid neoplasm with perineural spread. Source: From Ref. 66.

(A)

(B)

Figure 13 Imaging of facial nerve schwannoma continued. (A) Axial CT showing facial nerve schwannoma (arrows) presenting as parotid mass. (B) Coronal MR of the same patient showing tumor (fusiform area of high signal intensity) involving vertical segment of facial nerve in mastoid.


525

Figure 14 Topographic classification of facial nerve schwannomas as proposed by Litre et al. Images are obtained from a fused bone window computed tomographic scan with magnetic resonance imaging sequences. The tumor margin is delimited by the outline. Source: From Ref. 67.

Litre et al. described a classification for facial nerve schwannoma, based on a fused CT-MR algorithm (67) (Fig. 14) designed to aid in treatment planning. Classification of facial nerve schwannomas (67) Type I—tumor localized on geniculate ganglion. Type II—tumor is dumbbell-shaped on the geniculate ganglion, labyrinthine segment, internal auditory canal, and cerebellopontine angle cistern. Type III—tumor develops in tympanic and/or vertical segments of the facial nerve. Type IV—tumor develops in internal auditory canal or cerebellopontine angle without invasion of fallopian canal or geniculate ganglion. This category is difficult to distinguish from the vestibular schwannoma using radiological criteria. In this group, diagnosis was usually based on a previous microsurgical attempt.

regions. Intracranially, vestibular schwannoma may be indistinguishable; meningioma usually will show a “dural tail” sign, although that sign has been shown to be occasionally unreliable (68,69). In the temporal bone, cholesteatoma can cause similar bone erosion and facial paralysis. Traumatic neuroma, related to prior chronic otitis or temporal bone trauma, looks very similar to FNS. Granular cell tumor and osseous hemangioma are two entities that occur in the geniculate region. The former usually destroys bone irregularly, as opposed to the FNS. The latter is distinctive on CT, with “salt and pepper” density and a margin of new bone formation. Metastatic lesions to the temporal bone may also be considered, but the bone erosion is typically less well circumscribed. Extracranially, the chief differential diagnosis is that of parotid neoplasia.

Differential Diagnosis of Facial Nerve Schwannomas

Management and Outcomes for Facial Nerve Schwannomas

The differential diagnosis of FNS includes a variety of other common and uncommon pathologies of the same anatomic

As with trigeminal schwannomas, FNS has traditionally been managed surgically. Usually, achievement of gross total

526

Nuss and Burke

tumor resection has required resection of the nerve of origin. The rationale for aggressive total tumor removal, and nerve resection if necessary, is that as the tumor grows, there is progressive degeneration of axons and collagen deposition, which ultimately decrease the likelihood of good functional outcome (70). Therefore, better outcomes could theoretically be expected with earlier intervention. With FNS, the obvious concern is the facial paralysis that is inevitable after the facial nerve has been resected and reconstructed. Even in expert hands, nerve resection and grafting yields imperfect facial animation, and in most series the best outcome can be expected to result in function at House-Brackmann grade 3 or 4; many fare worse (57,71–73). While resection and nerve reconstruction are clearly necessary in patients who initially present with moderate or severe paralysis, management of patients with little or no paralysis is more challenging. In peripheral nerve surgery, it is often said that schwannomas can be separated from their nerve of origin and functional axons can be preserved intact. With schwannomas of the cranial nerves, this has not often proved to be possible in the senior author’s experience (DN), and in the reported experience of others (53,56,74). However, of all locations in which one would want to remove tumor without disrupting the affected nerve, the facial nerve would be a high priority for preservation. Conley and Janecka recommended that nerve preservation surgery for facial schwannoma “should be attempted every time,” but noted that in the majority it was not possible to preserve intact nerve. Of nine intratemporal FNS in their series, only one allowed preservation of the nerve (74). However, a number of subsequent authors have published increasing experience with nerve-preserving techniques, which can yield excellent functional outcomes in selected patients (57,70,75). In 2007 Lee et al. reviewed their experience with nervepreserving approaches for FNS (75). Using microsurgical technique and EMG monitoring, they employed what they termed “stripping surgery,” separating tumor from intact axons. The technique was used in six patients who had good preoperative facial function (HB Grade 1 or 2); it was possible to completely remove tumor in four. In the remaining two they performed decompressive surgery only (see below). Tumors were located in the geniculate ganglion (1 case), mastoid (2 cases), and IAC (3 cases). In all cases facial nerve integrity was preserved. Good facial function was preserved in all; two patients achieved HB Grade 1, four achieved Grade 2. There were no recurrences in the follow-up period, which ranged from 6 to 128 months (mean, 53 months). Size and location of tumor did not affect outcome. Their recommendation was that nerve-sparing approaches should be attempted in all patients with good preoperative facial function. In cases where tumor cannot be separated from intact nerve fibers, decompression may be a good alternative (75,76). Decompression applies to tumors that are expanding within the bony confines of the temporal bone. For patients with intratemporal FNS who have little or no preoperative paralysis, this technique simply removes bone from the areas around the tumor, with the goal of delaying compressive symptoms. Angeli and Brackmann reported excellent facial function in 4 of 4 patients in whom this technique was used and negligible or no tumor progression was noted during postoperative surveillance averaging 45 months. Interestingly, one of their patients presented with HB Grade 5 facial weakness, and improved to Grade 2 (75,76).

A wide variety of surgical approaches have been used to remove or decompress FNS, depending on tumor site and extent. Temporal bone approaches, middle fossa and posterior fossa approaches, and parotidectomy approaches have been used in various combinations; these are presented in detail elsewhere in this book. The complex subject of facial nerve reconstruction is relevant to this discussion but beyond the scope of this chapter.

Observation for Facial Nerve Schwannomas A cogent argument can be made for observation alone in selected FNS patients. Again, it is generally agreed that patients presenting with moderate or severe facial paralysis should undergo nerve resection and reconstruction, but for those with little or no paralysis, observation is an option. Proponents for observation (also referred to as conservative or expectant management) correctly point out that FNS is an extremely indolent tumor in most cases, perhaps more so than schwannomas at other sites, and that even when no therapy is given, patients may not experience progression of symptoms for years. Liu and Fagan (71) reported a series of 22 FNS cases, in which 12 were excised and 10 were observed. The best postoperative facial function in the tumor removal group was HB Grade 3, while 8 of the 10 conservatively treated patients had normal facial function up to 10 years later. No significant tumor growth was noted in any of the observed patients. The authors concluded that, for patients with little or no facial nerve symptoms, “delaying surgical resection of facial nerve schwannomas may allow patients to retain normal facial function indefinitely.” Not all observed patients will fare well, however, Perez et al. reviewed 24 FNS patients, of whom 11 underwent surgery and 13 were observed. Of the 11 in the operated group, 6 had unchanged postoperative facial function, 4 were improved, and one was worse; there were no postsurgical recurrences (57). Of the 13 in the observed group, facial function remained unchanged in 8 but was worse in 5 patients. In 4 patients, tumor progression was noted, and 3 underwent subsequent surgery. The authors concluded that “the decision on how to treat these patients should be individualized and based on initial facial function, growth rate, surgical experience, and informed patient consent.” A significant issue in considering the observation alternative is the establishment of a definitive diagnosis. Certainly, patients who present with classic findings of FNS may be candidates for observation without invasive diagnosis, but it must be acknowledged that the diagnosis can be elusive, and a close follow-up protocol must be strictly adhered to. As noted above, many FNS tumors are clinically and radiographically similar to other pathologic entities for which observation alone would be a poor choice. Therefore, a prudent approach would be to obtain a tissue diagnosis in cases where there is doubt. In all cases of FNS, the choice of “observation” should be viewed as a firm commitment to regular and detailed clinical and radiographic follow-up for the rest of the patient’s life. In addition to surgery and observation, stereotactic radiosurgery has received increasing attention in recent years. This will be reviewed in the section on radiation.

Decision-Making in Management of Facial Nerve Schwannomas Patients with small, minimally symptomatic FNS in whom the diagnosis is confirmed or considered highly likely may be offered several good alternatives, including nerve-preserving


surgery, decompression if tumor is not separable, observation, or stereotactic radiation. Significant controversy remains as to which of those options is best. Patients with larger tumors and/or moderate to severe symptoms should undergo surgery to remove the tumor, and in most cases to resect and reconstruct the facial nerve. Radiosurgery may be considered as an alternative in selected cases, or as an adjunct for residual or recurrent tumor.

Jugular Foramen Schwannomas It is important to realize that while jugular foramen schwannomas are among the more commonly encountered nonvestibular schwannomas, they are still rare. The world’s literature on these tumors recorded only about 200 cases as of the year 2000, and most of what is written has come from single case reports and small series. Of the largest reported series of jugular foramen schwannomas, most describe experience with fewer than 20 patients. Therefore, what is known of these tumors is based on relatively limited information. An elegant and thorough review of this subject was presented by Von Doersten, who did a meta-analysis of all previously published cases in 2000 (77). He estimated that the incidence of jugular foramen schwannomas was likely in the range of 3to 5 tumors per 10 million people per year. Jugular foramen schwannomas (JFSs) originate from the nerves of the jugular foramen, namely cranial nerves IX, X, and XI. These tumors are generally grouped together for discussion purposes because it is sometimes difficult to precisely determine the exact nerve of origin, given the tight proximity of the three nerves in and around the jugular fossa, and also because of the tendency for tumors to involve more than one nerve simultaneously. The jugular foramen is a region of densely compacted, intricate anatomy. Basically, it is an intracranial/extracranial conduit from the posterior fossa to the parapharyngeal space and upper neck, delineated by a smooth opening between the occipital bone and the petrous portion of the temporal bone. Traditional surgical anatomy describes two parts, the pars nervosum anteromedially, and the pars venosum posterolaterally. In the pars nervosum, the nerves are situated such that CN IX is most anterolateral, and CN XI is most posteromedial, with the vagus in between. Surgical approaches to this compartment are challenging because the facial nerve, internal carotid artery, mastoid, middle ear, and inner ear structures are all within a few millimeters of the jugular foramen, and therefore are vulnerable to injury. For this very same reason, untreated tumors in this region can be treacherous and even lethal. A critical issue in management of these tumors is the status of the jugular vein. In the majority of persons the right jugular vein is larger, reflecting the dominant venous outflow from the posterior fossa; this is considered a normal finding. In some individuals the left vein will be larger. Generally if the vein is known to be completely occluded preoperatively, it is assumed that sacrifice will not be likely to create new problems postoperatively. However, it is interesting that these tumors, despite their expansion in a tight space, do not always lead to complete obstruction of the venous circulation. Injury to, or sacrifice of, a functioning jugular vein can lead to disastrous cerebral venous infarction. Symptoms and signs of JFSs are quite variable. When tumors are small they may be asymptomatic. As mentioned above, the pattern of growth determines what symptoms will evolve. Early extracranial growth results in typical deficits of CN IX, X, and XI, with dysphagia, hoarseness, and shoulder weakness, respectively. When all three nerves are affected, the

527

patient is said to have Vernet syndrome. When extracranial expansion exerts pressure on CN XII, the patient also develops tongue weakness and is said to have the Collet–Sicard syndrome. A variety of other eponyms have been applied to related symptom combinations of this region (77). When the tumor grows intracranially, the symptoms relate to expanding tumor causing pressure on nearby structures of the posterior fossa and cerebellopontine angle. In some of these tumors, the intracranial symptoms will predominate. Thus, hearing loss, vertigo, tinnitus, facial weakness, tongue weakness, and signs of elevated intracranial pressure (headache, hydrocephalus) develop. With progression, patients develop gait disturbance, ataxia, long tract signs including motor weakness in the extremities, and potentially may die as tumor compresses the brainstem. In some patients, these intracranial manifestations may be present without neuropathies of cranial nerves IX, X, or XI. Schwannoma cases have been reported in which the presenting sign was foramen magnum syndrome (78) (defined by unilateral arm sensory and later motor deficits, progressing to ipsilataeral leg, then contralateral leg, and finally contralateral upper extremity deficits) (79); schwannoma has also been reported to mimic tumor of the fourth ventricle (80). Typically, these tumors will grow in both directions to some extent. The path of least resistance allows bulbous expansion above and below the foramen, with bony erosion about the foramen, giving the classic dumbbell shape (77) (Fig. 15). Von Doersten compiled a database of 164 analyzable JFS cases from the existing literature (i.e. those reports that contained sufficient detail for meaningful review) (77). Of those cases, the mean age at diagnosis was 41.7 years, with equal gender distribution. The nerve of origin of the JFS was determined to be CN X in 43.5%, CN IX in 29%, and CN XI in 14.1%, with the remainder involving two or more of the nerves. Thus, in descending order of frequency, tumors appear to arise from the vagus, the glossopharyngeal, and the spinal accessory nerves, followed by combinations thereof. In Van Doersten’s review, the most common symptoms were hearing loss (35%), hoarseness (26%), dysphagia (12%), and vertigo (10%), with other less common complaints occurring in many combinations of both intracranial and extracranial symptoms. Depending on size and extent of tumor, cranial nerve deficits from V to XII were all reported, individually (except that there were no isolated presentations of sixth nerve palsy) and in various combinations. Clinical evaluation of patients with suspected lesions of the jugular foramen includes careful history and physical examination, with detailed neurological focus on cranial nerves, cerebellum, and long tract signs. Audiologic and vestibular studies quantify sensory deficits of the ear. Flexible endoscopy of the upper airway documents details of lower cranial nerve function. Inquiry must be made as to any history of chronic lung disease and gastroesophageal reflux disease, since these disorders may place the patient at higher risk of poor outcome from aspiration, which commonly occurs with JFSs. All of these details influence preoperative planning and patient counseling regarding the need for feeding tubes, laryngoplasty, facial reanimation procedures, and hearing and balance rehabilitation (77). Radiologic investigation of suspected jugular fossa tumors must be thorough. Thin-section CT and MR in all three planes are necessary to optimally evaluate these lesions; often MR arteriography and MR venography are necessary to adequately image the regional bloodflow. In addition to the

528

Nuss and Burke

Figure 15 Comparison of early and later stages in development of jugular foramen schwannomas in three planes (from top to bottom, parasagittal, axial, and coronal planes). Images on left side of page show small tumor within jugular foramen. Note that with small tumor, venous system remains patent. Images on right side show larger schwannoma, now bilobed and involving both intra- and extracranial compartments, with compromise of venous system. The degree of intra- versus extracranial growth is highly variable. Source: from Ref. 77.

general radiographic features described for schwannomas, JFSs will exhibit specific findings that relate to the unique regional anatomy. On CT, the jugular fossa will be smoothly enlarged, and tumor will be isodense with muscle. On MR, T1 images will typically show low signal intensity; T2 will show higher signal. MR can give much information about flow in the jugular vein, which gives a dark signal void in high-flow states and a bright or mixed signal in low-flow states, but MR venography and arteriography can more accurately define regional circulation. On both CT and MR, dumbbell shape of the lesion is considered to be virtually pathognomonic for schwannoma, and the tumor will enhance brightly with contrast due to vascularity. Figures 16 to 18 show common radiographic appearances of JFS.

Differential Diagnosis of Jugular Foramen Schwannomas The differential diagnosis of lesions at the jugular foramen is shown in Table 1 (77). Paraganglioma is differentiated from JFS by a more irregular pattern of bone destruction on CT, and large flow-voids within the tumor, along with the classic “salt and pepper” appearance on MR. Meningioma will often be associated with the “dural tail” sign. Appropriate imaging will sometimes include traditional angiography to better define vascularity and regional bloodflow, and to help differentiate the likely tumor type. JFSs do not usually show an angiographic “blush,” typical of the much more vascular paragangliomas. Embolization can be helpful to preoperatively occlude the inferior petrosal sinus, which can be a troublesome source of bleeding


529

Figure 16 Radiographic appearance of jugular foramen schwannomas. (A) Axial CT scan with bone algorithm shows an enlarged right jugular foramen (arrows). Note the sharp, rounded bone borders of the intraosseous extension, including a thin sclerotic rim and slightly bulging and eroded cortex (arrowheads). (B) Coronal contrast-enhanced T1-weighted MR image shows dumbbell-shaped tumor extending both into the posterior cranial fossa (arrows) and below the skull base (star). (C) Axial CT scan with bone algorithm shows a flared IAC meatus (arrow), which may be normal, although it is probably due in part to tumor eroding its posterior margin (compare with D). (D) Axial T2-weighted MR image shows a large tumor with high, slightly inhomogeneous signal in the posterior cranial fossa abutting but not extending into the right IAC (arrow). Note deformity of the brain stem, fourth ventricle, and cerebellum by the tumor. Source: From Ref. 28.

when the tumor is resected from the jugular bulb. Minimizing this bleeding may help limit injury to uninvolved cranial nerves (77). Despite the rarity of JFS several staging systems have been described. Kaye advocated a simple system in which stage A indicates tumor that is mostly intracranial; stage B mainly involves the bone surrounding the foramen; stage C is mostly extracranial (81) Pellet (82) added a stage D, defined as dumbbell tumor with both intra- and extracranial extent. Franklin (83) adapted the paraganglioma staging system of Fisch, describing four stages (and multiple substages of each), based on extension from the neck to the petrous carotid to the intracranial space, as outlined below: A: B: C: D:

tumor only in the neck tumor extends to jugular fossa but is primarily extracranial tumor progresses along petrous carotid tumor extends intracranially

Treatment Options for Jugular Foramen Schwannomas As with schwannomas elsewhere, the treatment options for JFS include surgery, radiation, and observation. Decision

making is complex for several reasons. First, the natural history for JFS, in contrast to the much more common vestibular schwannoma (VS), is not well known. With increasing access to imaging, these tumors are sometimes diagnosed in minimally symptomatic persons, or asymptomatic persons undergoing imaging studies for unrelated causes. There are insufficient studies of JFS upon which to advise such patients. Von Doersten postulated that if one were to extrapolate what is known from the abundant natural history studies of untreated VS, it would be reasonable to assume that JFSs can be expected to grow 0.1 to 0.2 cm per year. If that assumption is true, then very small, manageable tumors identified in young healthy persons would likely progress by 1to 2 cm per decade, eventually becoming formidable lesions. Theoretically, such tumors could be removed with less morbidity while small. However, observation might also be a valid option for patients with small tumors and minimal symptoms. Observation, or so-called “expectant management,” will be addressed in more detail later in this chapter.

530

Nuss and Burke

Figure 17 Radiographic appearance of jugular foramen schwannomas, continued. 21-year-old woman with otalgia and sudden seventh cranial nerve paralysis caused by jugular foramen schwannoma. (A) Axial CT scan with bone algorithm shows marked tumor growth eroding the bone (arrows). Note rounded, sharply demarcated margins as well as erosion and bulging of cortex along the medial border (arrowheads). (B) Coronal CT scan with bone algorithm shows marked intraosseous tumor growth, rounded and sharp bone borders, and erosion and slight bulging of cortex (arrows). (C) Sagittal noncontrast T1-weighted MR image shows the patient’s normal right side (arrow points to jugular foramen). (D) Sagittal noncontrast T1-weighted MR image shows a tumor with low signal intensity in the left jugular foramen, extending below the skull base (arrows). (E) Axial contrast-enhanced T1-weighted MR image shows tumor growth into the IAC (arrow), which was confirmed on coronal MR images (not shown). (F) Axial contrast-enhanced T1-weighted MR image shows that the main portion of the tumor is located within the jugular foramen and below the skull base (arrows) (compare with D). Source: From Ref. 28.

With larger tumors that are causing deficits, the decision to treat is not as difficult. Hence, the decision must focus on surgery versus radiation.

Surgical Treatment of Jugular Foramen Schwannomas In deciding how best to manage JFSs, outcomes of reported series managed surgically must be thoughtfully considered. In Von Doersten’s review, in which all cases were treated surgically with curative intent, it was noted that many papers did not specifically record preoperative cranial nerve status; some did not record postoperative status either. He stated that “it was very difficult to interpret whether any preoperative cranial nerves actually improved after resection of the tumor, or whether the status simply was not recorded.” (Intuitively, it seems that most surgeons would make note of clinical improvement in such reports if improvement did in fact occur.) Von Doersten also noted that in cases where CN IX, X, and XI deficits were noted preoperatively, there was no recovery of any cranial nerve function postoperatively. In

his review, cranial nerve function was possible but not likely when multiple deficits were present preoperatively. Moreover, the likelihood that additional (i.e., new) cranial nerve deficits would be caused by surgery was significant. Again, the inhomogeneity of the reports made interpretation difficult, but a few observations were insightful. Of the 31 cases in which little or no facial nerve deficit was present postoperatively (House-Brackmann score of 1 or 2) (84), none had any preoperative deficit. There were 10 patients with HB scores of 3 or 4, and only five of those had any preoperative deficit. Thus, all of the patients with the best facial nerve outcomes were those who had no preoperative deficit, and among patients with worse outcomes there were some who had normal facial function preoperative. Surgical approaches to jugular foramen tumors are numerous and addressed elsewhere in this text. The most frequently employed methods have included modifications of the Fisch infratemporal approach, suboccipital approach, petro-occipital trans-sigmoid (POTS) approach, retrosigmoid


Figure 18 Radiographic appearance of jugular foramen schwannomas, continued. 63-year-old woman with jugular foramen schwannoma causing hearing loss, reduced gag reflex, and nystagmus. (A) Axial T2-weighted MR image shows tumor with high signal intensity within the enlarged left jugular foramen. The tumor is well demarcated with smooth borders. There is only a small tumor bulge into the posterior cranial fossa (arrow). (B) Coronal contrast-enhanced T1-weighted MR image with fat suppression shows strong and homogeneous contrast enhancement in schwannoma located below the skull base (arrow indicates level of the jugular foramen). Source: From Ref. 28.

approach, and combinations thereof. Unfortunately, nomenclature in existing reports is not uniform, making comparisons difficult. Tumor outcomes, reported for 164 cases, were such that only seven recurrences were documented. Of these, five had been operated via a suboccipital approach, one via a cervical approach, and one by an approach not documented. No recurrences were reported after infratemporal fossa resection. The conclusion was that combined procedures that afford access to both intracranial and extracranial aspects of the jugular fossa, above and below the tumor, are more likely to be curative. In a more recent report, Wilson et al. (85) reviewed their experience with seven JFS patients, of whom six were managed surgically. Patients ranged from ages 24 to 69 years. Presenting symptoms included dizziness, hearing loss, dysphagia, diplopia, tongue paresis, and hoarseness. All operated patients had complete tumor excision. Lower cranial nerve Table 1 Differential Diagnosis of Tumors of the Jugular Foramen Primary tumors r Paraganglioma r Meningioma r Schwannoma, neurofibroma r Hemangiopericytoma r Chondrosarcoma r Plasmacytoma Secondary tumors r Endolymphatic sac tumor r Nasopharyngeal carcinoma r Malignant tumors of the temporal bone (SCCA) r Parotid neoplasms r Langerhans cell histiocytosis Metastatic disease r Squamous cell carcinoma r Breast cancer r Prostate cancer Source: Modified from Ref. 77.

531

dysfunction that was present preoperatively did not resolve, whereas preoperative deficits of CN V and VI did resolve. The incidence of new lower cranial nerve deficits was 15%, and these were only in the nerves that were determined to be the nerve of origin. No recurrences were seen. In two cases, temporary feeding tubes were needed. The authors concluded that JFS tumors can be successfully managed with surgery, and with low morbidity. In 2006, Sanna et al. reported one of the world’s largest series of JFS patients, 23 cases treated over 18 years (86), in which 22 patients were treated surgically. A wide variety of approaches were used in various combinations, most often the petro-occipital trans-sigmoid approach (POTS), plus subtotal petrosectomy, translabyrinthine, transotic, and transcervical approaches, and infratemporal fossa type A approach, among others. Some required two-staged surgeries to address intradural and extradural tumor separately. Complete resection was accomplished in 21 patients, and there were no recurrences. There were no deaths, and only one CSF leak. Facial nerve function was preserved in all patients operated by the POTS approach, but a few patients operated by other approaches experienced HB scores of 3 and 4. Of patients who were operated with the intent of hearing preservation, good hearing was preserved in 83.3%. No patients recovered the function of the preoperatively paralyzed lower cranial nerves. A new deficit of one or more of the lower cranial nerves was recorded in 50% of cases. There was a single CSF leak and no perioperative mortality. The authors concluded that “surgical resection is the treatment of choice for JFS,” and that “the POTS approach allowed single-stage, total tumor removal with preservation of the facial nerve and of the middle and inner ear functions in the majority of cases. Despite the advances in skull base surgery, new postoperative lower cranial nerve deficits still represent a challenge.” A 2004 report from Kadri and Al-Mefty reviewed the experiences of six patients with JFS, who all presented with intra- and extracranial extensions through an enlarged jugular foramen (classic dumbbell tumors). Symptoms included deficits of CN IX, X, XII, and XI, in that order (87). All patients had two or more deficits upon presentation. A transcondylar suprajugular approach was used without sacrificing the labyrinth or the integrity of the jugular bulb. Complete resection was accomplished in all. There were no deaths, no new cranial nerve deficits, and no recurrences. There was one case of aspiration pneumonia. Two patients with preoperative deficits of CN IX and X improved and three patients recovered tongue mobility. Their conclusion from this series was that “with careful, extensive preoperative evaluation and appropriate planning of the surgical approach, dumbbell-shaped jugular foramen schwannomas can be radically and safely resected without creating additional neurological deficits. Furthermore, recovery of function in the affected cranial nerves can be expected.” To put these differing outcomes into perspective, and to make fair comparisons, a few observations are relevant. The “meta-series” reviewed by Von Doersten represents a great number of cases scattered across numerous centers worldwide with no uniformity of surgical approach, treatment philosophy, operating surgeons, or reporting strategies. It also spans many decades, across which skull base techniques, imaging studies, and instrumentation have changed dramatically. The series of Sanna et al. (among the most respected skull base surgeons of the world) is one of the largest series in the literature, but it shares some of the same characteristics,

532

Nuss and Burke

namely a diverse group of tumors, treated over a long timeframe spanning (and contributing to) the evolution of many techniques. The series of Wilson et al., and also the series of Kadri and Al-Mefty, are both smaller series but may actually be more relevant in some ways, because they represent more closely the current state of the art in skull base surgery, having taken place entirely in the “modern era of skull base surgery.” In Wilson’s series, the incidence of new cranial nerve deficit was just 15%, and all of those were confined to nerves of tumor origin. Essentially there were no unexpected or avoidable new palsies. Furthermore, associated upper cranial nerve deficits resolved. In Kadri and Al-Mefty’s series, unlike the others, improvement of lower cranial nerve deficits was observed postoperatively. The recent trends toward improvement in functional outcomes are encouraging. It should be emphasized, however, that these outcomes were reported by specialized skull base teams whose senior surgeons are vastly experienced. The rarity of JFS tumors mandates that if surgical treatment is to be recommended, it must be rendered by such teams.

Decision-Making in Management of Jugular Foramen Schwannomas As a general rule, surgical treatment for JFS is arguably the optimal treatment in younger patients or those who are symptomatic with progressive deficits, and who are reasonably good surgical candidates. With symptomatic patients who are elderly, or those who are medically infirm or otherwise poor surgical candidates, the best treatment option is probably radiation, which will be discussed in more detail later in the chapter. JFS patients, who are minimally symptomatic with small tumors, or asymptomatic, can be managed expectantly (observed) with serial imaging and clinical examinations.

Schwannomas of Other Skull Base Sites Olfactory and Optic Nerve Schwannomas: Controversy The pathological literature and the surgical literature are at odds as to whether olfactory and optic nerve schwannomas even exist. In traditional anatomical and pathological terms (1,5), these nerves are devoid of Schwann cell sheaths; therefore, theoretically they should never give rise to schwannomas. Despite this scientific assertion, rare cases of schwannoma involving the olfactory (88,89) and optic (90,91) nerves have been apparently well documented. This discrepancy might be explained by the recent characterization of specialized glial cells known as “olfactory ensheathing cells” (and presumably optic ensheathing cells exist as well), which share many features in common with Schwann cells, including neural crest origin as well as a number of molecular markers (92,93). When such cells become neoplastic, they may well be indistinguishable from Schwann cells. These are of course extremely rare tumors (94,95) Presenting signs include hyposmia and visual blurring.

Schwannomas of the Nerves of Ocular Motility Orbital schwannomas most often arise from the trigeminal nerve (49), but rare cases of oculomotor (96–98) trochlear (99–100), and abducens origin (101–103) have been welldocumented. These lesions may arise anywhere along the course of these nerves and may therefore be intracisternal (where they cause brainstem effects), intracavernous (causing diplopia), intraorbital (causing diplopia and proptosis) (46), or may occupy more than one compartment.

Schwannomas of the Hypoglossal Nerve The hypoglossal nerve is another rare site for schwannoma, but one in which substantial morbidity ensues (104–106). Only 26 cases of transdural (dumbbell) hypoglossal schwannomas are reported, and their management is challenging for many of the same reasons that jugular foramen lesions are challenging. Typical symptoms are dysarthria, dysphagia, and symptoms of brainstem compression. Treatment may require combined microsurgical approaches to decompress the brainstem, which is frequently affected, and adjuvant stereotactic radiosurgery.

Unusual and Curious Manifestations of Schwannomas Schwannomas, as we have noted, may occur in virtually any nerve of the head and neck. The most common sites and presentations have been reviewed above. To complete the picture, we consider here some examples of schwannomas that have been reported in the most unusual locations, or those causing unusual symptoms and manifestations. Halefoglu et al. reported a severe case of NF2 in which the unfortunate patient had synchronous schwannomas involving both hypoglossal nerves, both vestibular nerves, the right trigeminal, the left oculomotor, and the right abducens (107). Cheong et al. observed bilateral vidian nerve schwannomas, associated with facial palsy, which resolved after transnasal resection of the tumors (108). Schwannomas have also been reported originating from the greater petrosal nerve (109), Jacobson nerve (110) and the sympathetic plexus of nerves about the cavernous segment of the internal carotid artery (111). In the latter case, microsurgical resection resulted in excellent outcome except for Horner syndrome. Kamel at al. reported a vagal schwannoma of the cerebellomedullary cistern causing severe refractory neurogenic hypertension (112), believed to be secondary to compression of medullary centers that coordinate sympathetic control of blood pressure. Jagetia et al. described a case of a dumbbellshaped trigeminal schwannoma in which the patient’s predominant symptom was pathological laughter, which resolved immediately after surgical resection (113). The authors opined that this outcome supports the theory that the brainstem and perhaps the medial temporal lobe play some role in the control of spontaneous laughter. Hsu et al. studied eight cases of pathology-proven orbital schwannomas, and four were found to have associated pneumosinus dilatans affecting paranasal sinuses adjacent to the orbit (114).

RADIATION IN THE TREATMENT OF SKULL BASE SCHWANNOMAS Historically, surgery has been the treatment of choice for most patients with skull base schwannomas, and with good reason. For most of the 20th century, the only nonsurgical option was external beam radiation, with which it was impossible to discretely irradiate skull base tumors without delivering unwanted high doses to nearby uninvolved structures, which would result in unacceptably high incidence of neurologic injury. With increasing availability of stereotactic and imageguided radiation delivery systems in recent years, however, and the consequent ability to precisely deliver high doses of radiation to well-defined target areas, growing attention has been paid to radiation as both primary and adjuvant treatment for schwannomas. The vast majority of this experience has been with vestibular schwannomas, and incontrovertibly radiation is


now a well-established alternative for treatment of VS. Although stereotactic radiation systems have been available for decades, clinical experience with the very uncommon nonvestibular schwannomas has until recently been more limited, and clinical reports with adequate long-term follow-up were insufficient. A significant and growing body of new literature has now begun to improve our understanding of how radiation fits into the scheme of management of these benign tumors. The fundamentals of physics, radiobiology, and details of radiation delivery systems are available elsewhere and will not be addressed here. One important technical point, however, is that the vast majority of data regarding radiation of skull base schwannomas comes from experience with the Leksell Gamma Knife system and variations thereof, including linear accelerator-based units. The term stereotactic radiosurgery refers to these kinds of extremely precise delivery systems. While other radiation delivery systems (e.g., fractionated radiation therapy and proton beam radiation) have been used to treat schwannomas, outcomes data for those are more limited. Therefore, this section will review reported clinical outcomes regarding the effectiveness, limitations, and risks of stereotactic radiosurgery in the treatment of skull base schwannomas. In evaluating the role of any treatment, the critical issues are two: efficacy and safety. Efficacy is best measured by the degree of symptom relief and tumor control. Safety is assessed by noting whether existing deficits or symptoms worsen; whether any new deficits are incurred as a result of treatment; by the appearance of early or late complications; and by the implications for any needed future treatment. The following sections will attempt to examine these issues.

Radiation for Nonvestibular Schwannoma in General In 2006 Flickinger and Barker reviewed a broad experience with radiosurgery for cranial nerve schwannomas (115). They analyzed outcomes from published series involving thousands of patients worldwide, most of whom were treated for vestibular schwannoma, but focusing on the issues of tumor control rates, dosing, and morbidity of adjacent cranial nerves. Although this study involved predominantly vestibular schwannoma patients, the data are relevant here with respect to functional outcomes for cranial nerves in proximity to tumor. The authors concluded that “the low morbidity and high long-term tumor control rates with radiation treatment have made it the choice of many patients who opt for active initial management for small- or medium-sized cranial nerve schwannomas.” Pollock et al. reviewed the Mayo Clinic experience with 23 patients treated with radiosurgery between 1992 and 2000 for a variety of nonvestibular skull base schwannomas of various cranial nerves, including the trigeminal (n = 10), jugular foramen group (n = 10), hypoglossal (n = 2), and trochlear (n = 1) (116). Nine of these had undergone prior surgery. With a median follow-up of 43 months, 22 of 23 tumors were smaller (n = 12) or unchanged in size (n = 10). The only tumor that failed to respond was the one “malignant schwannoma” included in the series. Four patients (17%) suffered radiation-related morbidity, including three with trigeminal tumors who suffered new or worsened trigeminal dysfunction. One patient developed Eustachian tube dysfunction after treatment for hypoglossal schwannoma. Of note, however, no patient with lower cranial nerve schwannoma developed any treatment-related hearing loss, facial palsy, or dysphagia after therapy. The authors pointed out that the high tumor control rates reported for vestibular schwanno-

533

mas could be expected to apply to the nonvestibular schwannoma (NVS) population, and asserted that compared to historical controls treated surgically, radiosurgery appears to result in less treatment-related morbidity, especially for tumors of the lower cranial nerves. Symptom relief in this series was not specifically detailed. The authors made comparisons to several reported surgical series that summarized results after microsurgical resection of NVS, and concluded that the radiosurgery treatment-related morbidity and tumor control rates compared favorably. Further, they asserted that patients with jugular fossa schwannomas did better with radiosurgery than with surgery, in terms of reduced cranial nerve morbidity. They did stipulate, however, that for patients with significant mass effect, or those with primarily cystic tumors, primary surgical resection should be the first choice.

Radiation for Trigeminal Nerve Schwannomas Sheehan et al. published their experience treating trigeminal nerve schwannomas (TNS) using radiosurgery in 26 patients, from 1989 to 2005 (117). The median follow-up was 48.5 months. Clinically, 18 patients improved (72%), four were stable (16%), and three were worse (12%). Imaging studies revealed tumor shrinkage in 12 patients (48%), no change in 10 (40%), and tumor growth in three (12%). They concluded that the risk/benefit ratio with radiosurgery was favorable for TNS patients, but that larger studies are needed to better evaluate long-term outcomes. Hasegawa et al. also examined outcomes after treatment of TNS with radiosurgery, in 37 cases, with a mean follow-up of 54 months (118). Clinically, 40% of patients had improvement in symptoms, but one patient worsened despite good radiographic tumor control. Radiographically, 20 patients (54%) showed tumor regression and 8 (22%) showed stable findings. In 5 patients (14%), however, tumor enlarged or uncontrollable facial pain developed with radiationinduced edema requiring surgical resection. As in Pollock’s series, the authors concluded that gamma knife was safe and effective for select patients, but that large tumors or those that are cystic or compressing brainstem or 4th ventricle should be treated with surgery as the first choice. Huang et al. reviewed outcomes of 16 TNS patients treated with gamma knife radiosurgery; six had prior surgery and 10 were treated primarily; mean follow-up was 44 months (119). Clinically, five patients improved and 11 were stable. Radiographically, the tumor control rate was 100%, with regression in nine and stability in seven. Significantly, there were no new cranial nerve deficits of any kind after treatment. The conclusion of the authors was that radiotherapy offered a reasonable alternative to microsurgery, either as primary or adjuvant treatment, which “controlled tumor growth, did not cause new deficits, and often improved presenting symptoms.” Pan et al. examined long-term results of radiosurgery for TNS in 56 patients, in one of the largest series to date (120). Fourteen had undergone prior surgery, 42 were treated primarily. Clinically, 14 patients had complete relief of symptoms (numbness or diplopia), and improvement of other deficits was seen in 25 patients. In 13 patients, trigeminal dysfunction either did not change or got slightly worse, and in 4 patients worsening symptoms were related to tumor progression. Radiographically, tumors disappeared in 7 patients, regressed in 41 patients, and were unchanged in 4 patients. Four patients experienced tumor progression, and one of those died 36 months following treatment. The overall tumor growth control rate was 93% (52 of 56 cases). The authors concluded that radiosurgery is effective for small and

534

Nuss and Burke

medium-sized TNS tumors, but larger tumors should be surgically removed or decompressed, particularly when in close proximity to the brainstem. Also on the subject of TNS, Peker et al. studied gamma knife treatment of 15 patients followed for a mean of 61 months post-treatment (121). As in other reports, the cohort included patients treated primarily as well as adjuvantly. As in Huang’s series, 100% radiographic control rate was noted, with regression in 13 and no change in 2. One patient developed transient facial numbness and diplopia. Again, the conclusion was that radiosurgery is associated with good tumor control and minimal risk of adverse effects. As with all forms of radiation, stereotactic radiosurgery has been associated with complications, duly noted in several of the above series. Akiyama et al. reported an unusual complication 15 months after gamma knife treatment, involving rapid tumor regrowth with extensive cyst formation and severe brainstem compression requiring urgent surgery (122). At operation, the trigeminal nerve had to be sacrificed because of dense pseudocapsule formation that was attributed to radiosurgery. The patient suffered chronic facial pain afterwards. The authors admonished that radiosurgery can induce fibrosis or degenerative change that complicates subsequent surgery.

Radiation for Facial Schwannomas Kida et al. reviewed the stereotactic radiosurgical treatment of 14 patients with facial nerve schwannoma (FNS) (123). Eleven of 14 presented with facial palsy; nine presented with hearing loss. After a mean follow-up of 31.4 months, facial nerve function was improved in five, stable in eight, and worse in one patient. None suffered any new hearing loss. One patient developed facial palsy immediately post-treatment that recovered to House-Brackmann Grade 3. Radiographically, 10 tumors regressed and 4 were stable (100% control rate). No tumors progressed. The authors concluded that radiosurgery should be the “treatment of first choice for facial schwannomas.” Litre et al. presented their experience in irradiating FNS in 11 patients (67). These were identified from a large population of 1783 patients with CP angle tumors treated in the same institution. Favorable previous experience with facial nerve outcomes among VS patients provided rationale for using radiosurgery to treat FNS. Mean follow-up was 39 months. Clinically, three patients with facial weakness improved, and none developed any new palsy or worsening of previous palsy. Radiographically, 10 were stable or regressed, but 1 required microsurgery due to cyst development. The authors proposed a classification of FNS into four anatomic subtypes stratified according to different clinical and surgical difficulties (Fig. 14). They felt that gamma knife radiosurgery could become “a first treatment option” for small or medium-sized FNS.

Radiation for Jugular Foramen Schwannomas Martin et al. examined outcomes for 34 patients with 35 jugular foramen schwannomas (one patient had bilateral tumors) treated with radiosurgery (124). Noting that JFS often presents with multiple lower cranial nerve deficits, and that surgical resection may be associated with significant morbidity, their hypothesis was that radiosurgery might reduce cranial nerve morbidity or at least prevent additional deficits compared with surgical resection. Twenty-two patients had previously undergone surgery and all had pretreatment cranial neuropathies. Median follow-up was 83 months, one of the longest in the radiosurgery literature for schwannoma.

Clinically, cranial nerve deficits improved in 20% and were stable in 77% after radiosurgery, but worsened in one patient. All nerves that were functioning pretreatment were intact after treatment (i.e., no new cranial nerve deficits). Radiographically, tumors regressed in 17 patients, were stable in 16, and progressed in 2. The authors submitted that their expectation, in terms of long-term tumor control and preservation or improvement of cranial nerve function, was confirmed.

The Current Role of Stereotactic Radiosurgery for Schwannomas Stereotactic radiosurgery clearly has a role in the treatment of nonvestibular skull base schwannoma, in both primary and adjuvant settings. The specifics of this role, however, remain controversial. It is possible to achieve high tumor control rates and significant symptom relief in many patients. The challenge is to identify who will benefit, who will not, and perhaps most importantly, who will be likely to be harmed in the process. With stereotactic radiosurgery, as with any modality of treatment for these challenging problems, complications can be devastating. In contrast with surgery, for irradiated patients the risk of serious complications extends well into the future. Late-onset brainstem edema, facial palsy and other cranial nerve deficits, cystic degeneration, and progressive fibrosis have all been reported. The risk of radiation-induced malignancy should be low with radiosurgery, but will not be accurately known for years to come. The incidence of radionecrosis affecting the brain and skull would also be expected to be low, but late-effects outcomes will take years if not decades to define. Currently, we have little means of identifying which patients will develop these long-term problems, but the concern is real, especially when radiation is considered for younger patients. With regard to radiation-related cranial nerve palsies, it is known that these phenomena are dose-dependent. For effective tumor control, treatment protocols most commonly employ a prescription dose, or so-called marginal tumor dose, in the range of 12 to 13 Gy (115) but significantly higher doses have been used in the past and continue to be used in some centers. Miller et al. (125) reported the Mayo Clinic experience with a reduced-dose protocol in treatment of vestibular schwannomas, in a study designed to lower the incidence of radiation-related cranial neuropathies. Comparing cohorts treated with standard doses vs. reduced doses, they found that the incidence of facial nerve morbidity and trigeminal nerve morbidity could be substantially reduced when the tumor-margin dose was lowered. This was statistically quite significant for facial palsy in particular, in which facial neuropathy went from 38% to 8%. During the median follow-up of 2.3 years, there were no cases of tumor progression in the reduced-dose group, but the authors noted that longer follow-up is needed to determine whether reduction of therapeutic dose will reduce ultimate tumor control rates. Also, as noted before, much is known about radiosurgery for VS, but much remains to be learned, and it is unclear to what degree VS data apply to NVS patients. A multitude of detailed issues beyond the scope of this discussion will continue to deserve scrutiny. As in Miller’s study, precise dose comparisons are needed to determine optimal “compromise dose” strategies that give good tumor control with minimal risk of complications. Comparison of other parameters, including tumor volumes, geometric tumor variations, and specific risks that vary by anatomic site must all be carefully considered.


DECISION-MAKING IN THE MANAGEMENT OF SKULL BASE SCHWANNOMAS The rarity of nonvestibular skull base schwannomas does not diminish the grave implications of their diagnosis. These are devastating clinical problems. Perhaps it is this reality that has led to the vast number of publications on the subject, the number of which almost certainly exceeds the number of patients with the disease. Clearly, there is no perfect treatment that can be universally applied to all schwannoma patients. As outlined in this chapter, there are numerous factors that must be taken into account in order to determine the best course of action for a particular patient. The fact that a patient has schwannoma does not necessarily mandate treatment. Observation, or expectant management, may be the best option for select patients with skull base schwannoma. If observation is to be a serious consideration, it is essential that the diagnosis be established with as much certainty as the situation permits, using clinical, radiographic and (when appropriate) invasive means. Observation may be the best option for patients who are asymptomatic or minimally symptomatic; who are advanced in age or have medical contraindications to, or are unwilling to accept, active intervention; who would be likely to experience treatment-related deficits that are worse than their existing symptoms; or who have one of the neurofibromatosis syndromes in which multiple tumors exist and a new treatment-related deficit would be debilitating. Observation is not a passive choice. The decision to follow this course requires a firm commitment to detailed clinical and radiographic follow-up, at regular intervals and over the long term, so that tumor progression can be detected early enough to intervene with the least possible morbidity. As we have noted, there will be some patients who may never need intervention, but clearly many will. Intervention, be it with surgery or radiation, should be considered when the patient under observation becomes increasingly symptomatic, or when significant tumor progression is noted radiographically. One of the challenges lies in deciding what amount of growth is significant. Observation is not a risk-free choice. One potential problem with observation is the fact that some tumors will not follow the steady course of slow progression, and new deficits may come on quite suddenly. This may rarely be from a sudden tumor growth, or it may be due to intratumoral hemorrhage or rapid cyst formation. Deficits brought on under these circumstances may not be reversible. Surgical treatment of NVS historically carried with it a high incidence of morbidity, especially with respect to cranial nerve deficits. Substantial progress in the areas of microsurgical technique and cranial nerve rehabilitation has drastically reduced the morbidity of surgery when performed by expert skull base surgery teams. Many patients with complex tumor problems can be offered the possibility of complete tumor removal and improvement in cranial nerve function. Unquestionably, surgery is the first treatment of choice for very large or predominantly cystic schwannomas, or for those that compromise the brainstem or 4th ventricle. Even in cases where total resection is inadvisable or impossible, decompression or subtotal resection can greatly improve the patient’s outcome. A growing body of evidence suggests that stereotactic radiosurgery is a very reasonable and effective option for patients who have small NVS tumors that are noncystic and not in close proximity to the brainstem or 4th ventricle. It is both logical and appealing to assume that the generally high success rates that have been reported for vestibular schwannoma

535

will be applicable to the NVS population. The recent literature is encouraging in that regard, but experience is still limited. Given the rarity of NVS, multi-institutional prospective trials, or at least collaborative database sharing, are needed in order to further define the optimal use of this modality specifically for NVS. Currently, as a practical matter, treatment decisions should best be made in the context of a multidisciplinary discussion in which patients have the benefit of expert consultation from all relevant specialists. As Wackym stated, “An informed decision to pursue observation, microsurgery, stereotactic radiosurgery, or a combination of these methods must be made, and it remains the responsibility of the surgeon to provide a balanced view of the relative advantages and disadvantages of each method.” (126)

ACKNOWLEDGEMENT The senior author (DWN) gratefully acknowledges the following individuals for their roles in the genesis of this chapter: Wilma Rowley Nuss, Michael Artigues, Danny Douglass, Eric Dusang, Stacey Liljeberg, and Les Nettleton. REFERENCES 1. Batsakis JG. Tumors of the peripheral nervous system. In: Batsakis JG, ed. Tumors of the Head and Neck: Clinical and Pathological Considerations. Baltimore: Williams & Wilkins, Baltimore, 1982:313–333. 2. Katz AD, Passy V, Kaplan L. Neurogenous neoplasms of major nerves of face and neck. Arch Surg. 1971;103:51. 3. Kao G. Neurolemmoma. Nov. 27 , 2007. 4. Asthagiri AR, Helm GA, Sheehan JP. Current concepts in management of meningiomas and schwannomas. Neurol Clin North Am. 2007;25(4):1209–1230. 5. El-Mofty SK, Kyriakos M. Soft tissue and bone lesions. In: Gnepp DR, ed. Diagnostic Surgical Pathology of the Head and Neck. Philadelphia: W.B. Saunders, 2001:520–525. 6. MacCollin M, Woodfin W, Kronn D, et al. Schwannomatosis: A clinical and pathologic study. Neurology. 1997;48(3):787–788. 7. Mrugala MM, Batchelor TT, Plotkin SR. Peripheral and cranial nerve sheath tumors. Curr Opin Neurol. 2005;18(5):604–610. 8. Hanemann CO, Evans DG. News on the genetics, epidemiology, medical care and translational research of Schwannomas. J Neurol. 2006;253(12):1533–1541. 9. Fraenzer JT, Pan H, Minimo L Jr, et al. Overexpression of the NF2 gene inhibits schwannoma cell proliferation through promoting PDGFR degradation. Int J Oncol. 2003;23(6):1493–1500. 10. Evans DG, Moran A, King A, et al. Incidence of vestibular schwannoma and neurofibromatosis 2 in the North West of England over a 10-year period: Higher incidence than previously thought. Otol Neurotol. 2005;26:93–97. 11. Antinheimo J, Sankila R, Carpen O, et al. Population-based analysis of sporadic and Type 2 neurofibromatosis-associated meningiomas and schwannomas. Neurology. 2000;54:71–76. 12. MacCollin M, Chiocca EA, Evans DG, et al. Diagnostic criteria for schwannomatosis. Neurology. 2005;64(11):1838–1845. 13. Kahn AN. Neurofibromatosis: Conference statement. National Institutes of Health Consensus Development Conference. Arch Neurol. 1988;45(5):575–578. 14. Brooks DG. The neurofibromatoses: Hereditary predisposition to multiple peripheral nerve tumors. Neurosurg Clin N Am. 2004;15(2):145–155. 15. Ahn JY, Kwon SO, Shin MS, et al. A case of multiple schwannomas of the trigeminal nerves, acoustic nerves, lower cranial nerves, brachial plexuses and spinal canal: Schwannomatosis or neurofibromatosis? Yonsei Med J. 2002;43(1):109–113. 16. Purcell SM, Dixon SL. Schwannomatosis. An unusual variant of neurofibromatosis or a distinct clinical entity? Arch Dermatol. 1989;125(3):390–393.

536

Nuss and Burke

17. Huang JH, Simon SL, Nagpal S, et al. Management of patients with schwannomatosis: Report of six cases and review of the literature. Surg Neurol. 2004;62(4):353–361; discussion 361. 18. Kam J, Helm T. Neurofibromatosis. www.emedicine.com, January 2, 2007. 19. Skovronsky DM, Oberholtzer JC. Pathologic classification of peripheral nerve sheath tumors. Neurosurg Clin North Am. 2004;15:157–166. 20. Dodd LG, Marom EM, Dash RC, et al. Fine-needle aspiration cytology of "ancient" schwannoma. Diagn Cytopathol. 1999;20(5):307–311. 21. Maertens O, Brems H, Vandesompele J, et al. Comprehensive NF1 screening on cultured Schwann cells from neurofibromas. Hum Mutat. 2006;27(10):1030–1040. 22. Stout AP. Tumors of the peripheral nervous system. In: Atlas of Tumor Pathology, Section 2, Fascicle 6. Armed Forces Institute of Pathology, Washington, D.C., 1946. 23. Preston FW, Walsh WS, Clarke TH. Cutaneous neurofibromatosis: Clinical manifestations and incidence of sarcoma in 61 male patients. Arch Surg. 1952;64:813. 24. Gupta G, Maniker A. Malignant peripheral nerve sheath tumors. Neurosurg Focus. 200715;22(6):E12. 25. Ferrari A, Bisogno G, Carli M. Management of childhood malignant peripheral nerve sheath tumor. Paediatr Drugs. 2007;9(4):239–248. 26. Minovi A, Basten O, Hunter B, et al. Malignant peripheral nerve sheath tumors of the head and neck: Management of 10 cases and literature review. Head Neck. 2007;29(5):439–445. 27. Huang SQ, Liang BL, Yuan JP, et al. MRI performance and diagnosis of schwannomas from cranial nerves in posterior cranial fossae. Ai Zheng. 2006;25(9):1178–1182. 28. Eldevik OP, Gabrielsen TO, Jacobsen EA. Imaging findings in schwannomas of the jugular foramen. Am J Neuroradiol. 2000;21(6):1139–1144. 29. Fisher LM, Doherty JK, Lev MH, et al. Distribution of nonvestibular cranial nerve schwannomas in neurofibromatosis 2. Otol Neurotol. 2007;28(8):1083–1090. 30. Mautner VF, Lindenau M, Baser ME, et al. The neuroimaging and clinical spectrum of neurofibromatosis 2. Neurosurgery. 1996;38(5):880–885; discussion 885–886. 31. Nunes F, MacCollin M. Neurofibromatosis 2 in the pediatric population. J Child Neurol. 2003;18(10):718–724. 32. Kouyialis AT, Stranjalis G, Papadogiorgakis N, et al. Giant dumbbell-shaped middle cranial fossa trigeminal schwannoma with extension to the infratemporal and posterior fossae. Acta Neurochir. 2007;149(9):959–964. 33. Eisenman D, Voigt E, Selesnick S. Unusual tumors of the internal auditory canal and cerebellopontine angle. In: Jacker R, Driscoll CLW, eds. Tumors of the Ear and Temporal Bone. Philadelphia: Lippincott Williams & Wilkins, 2000:259. 34. Nager GT. Neurinomas of the trigeminal nerve. Am J Otolaryngol. 1984;5:301–333. 35. Samii M, Migliori MM, Tatagiba M, et al. Surgical treatment of trigeminal schwannomas. J Neurosurg. 1995;82(5):711–718. 36. Pamir MN, Peker S, Bayrakli F, et al. Surgical treatment of trigeminal schwannomas. Neurosurg Rev. 2007;30(4):329–337; discussion 337. epub 2007, Aug 4. 37. Gwak HS, Hwang SK, Paek SH, et al. Long-term outcome of trigeminal neurinomas with modified classification focusing on petrous erosion. Surg Neurol. 2003;60(1):39–48; discussion 48. 38. Verstappen CC, Beems T, Erasmus CE, et al. Dumbbell trigeminal schwannoma in a child: Complete removal by a one-stage pterional surgical approach. Childs Nerv Syst. 2005;21(11):1008– 1011. Epub 2005 Mar 16. 39. Zhou LF, Mao Y, Zhang R. Surgical treatment of dumbbellshaped neurinomas: Report of an experience with 57 cases in a single hospital. Surg Neurol. 2007;68(6):594–602. 40. Al-Mefty O, Ayoubi S, Gaber E. Trigeminal schwannomas: Removal of dumbbell-shaped tumors through the expanded Meckel cave and outcomes of cranial nerve function. J Neurosurg. 2002;96(3):453–463. 41. Boustred AM. Endoscopic excision of a supraorbital schwannoma. Ophthal Plast Reconstr Surg. 2007;23(1):66–68.

42. Roh JL. Removal of infratemporal fossa schwannoma via a transmandibular transpterygoid approach. Eur Arch Otorhinolaryngol. 2005;262(5):428–431. Epub 2004 Sep 18. 43. Park EH, Lee SS, Byun SW. A schwannoma in the nasal septum. Eur Arch Otorhinolaryngol. 2007 [Epub ahead of print]. 44. Sciarretta V, Pasquini E, Frank G, et al. Endoscopic treatment of benign tumors of the nose and paranasal sinuses: A report of 33 cases. Am J Rhinol. 2006;20(1):64–71. 45. Har-El G. Combined endoscopic transmaxillary-transnasal approach to the pterygoid region, lateral sphenoid sinus, and retrobulbar orbit. Ann Otol Rhinol Laryngol. 2005;114(6):439– 442. 46. Kennerdell JS, Maroon JC, Celin SE. The posterior inferior orbitotomy. Ophthalmic Plastic and Reconstr Surg. 1998;14(4):277–280. 47. Eisenberg MB, Al-Mefty O, DeMonte F, et al. Benign nonmeningeal tumors of the cavernous sinus. Neurosurgery. 1999;44(5):949–954; discussion 954–955. 48. Cockerham KP, Cockerham GC, Stutzman R, et al. The clinical spectrum of schwannomas presenting with visual dysfunction: A clinicopathologic study of three cases. Surv Ophthalmol. 1999;44(3):226–234. 49. Cantore WA. Neural orbital tumors. Curr Opin Ophthalmol. 2000;11(5):367–371. 50. Moffat D, De R, Hardy D, et al. Surgical management of trigeminal neuromas: A report of eight cases. J Laryngol Otol. 2006;120(8):631–637. Epub 2006 May 17. 51. Hajjaj M, Linthicum FH Jr. Facial nerve schwannoma: Nerve fibre dissemination. J Laryngol Otol. 1996;110(7):632–633. 52. Dort JC, Fisch U. Facial nerve schwannomas skull base surg. 1991;51–56. 53. Schaitkin BM. Facial nerve schwannoma. In: Jackler R, Driscoll CLW, eds. Tumors of the Ear and Temporal Bone. Philadelphia: Lippincott Williams & Wilkins, 2000: 276–289. 54. O’Donaghue GM, Brackmann DE, House JW, et al. Neuromas of the facial nerve. Am J Otol. 1989;10:49–54. 55. Wiggins RH III, Harnsberger HR, Salzman KL, et al. The many faces of facial nerve schwannoma. AJNR Am J Neuroradiol 2006;27:694–699. 56. Kreeft A, Schellekens PP, Leverstein H. Intraparotid facial nerve schwannoma. What to do? Clin Otolaryngol. 2007;32(2):125– 129. 57. Perez R, Chen JM, Nedzelski JM. Intratemporal facial nerve schwannoma: A management dilemma. Otol Neurotol. 2005;26(1):121–126. (Toronto) 58. Elahi M, Audet N, Rochon L. Intraparotid facial nerve schwannoma. J Otolaryngol. 1995;24:364–367. 59. Lipkin AF, Coker NJ, Jenkins HA, et al. Migraine and sudden sensorineural hearing loss. Arch Otolaryngol Head Neck Surg. 1987;113:325–326. 60. Fisch V, Ruttner J. In: Fisch U, ed. Pathology of Intratemporal Tumors Involving the Facial Nerve. The Netherlands: Kugler Medical Publishing, 1997. 61. McMenomey SO, Glasscock ME 3rd, Minor LB, et al. Facial nerve neuromas presenting as acoustic tumors. Am J Otol. 1994;15(3):307–312. 62. Park HY, Kim SH, Son EJ, et al. Intracanalicular facial nerve schwannoma. Otol Neurotol. 2007;28(3):376–380. 63. Kubota Y, Kawamata T, Kubo O, et al. Large facial nerve schwannomas without facial palsy: Case reports and review of the literature. Neurosurg Rev. 2005;28(3):234–238. Epub 2005 Mar. 64. Shimizu K, Iwai H, Ikeda K, et al. Intraparotid facial nerve schwannoma: A report of five cases and an analysis of MR imaging results. AJNR Am J Neuroradiol. 2005;26(6):1328–1330. 65. Caughey RJ, May M, Schaitkin BM. Intraparotid facial nerve schwannoma: Diagnosis and management. Otolaryngol Head Neck Surg. 2004;130(5):586–592. 66. Kertesz TR, Shelton C, Wiggins R, et al. Intratemporal facial nerve neuroma: Anatomical location and radiological features laryngoscope. 2001;111(7):1250–1256. 67. Litre CF, Gourg GP, Tamura M, et al. Gamma knife surgery for facial nerve schwannomas. Neurosurgery. 2007;60(5):853–859; discussion 853–859.

Chapter 37: Schwannomas of the Skull Base 68. Guermazi A, Lafitte F, Miaux Y, et al. The dural tail sign: Beyond meningioma. Clin Radiol. 2005; 60:171–188. 69. Stasolla A, Bellussi A, Bibbolino C. Dural tail: Another face of facial nerve schwannoma? AJNR Am J Neuroradiol. 2006;27(3):694–699. 70. Sataloff RT, Frattali MA, Myers DL. Intracranial facial neuromas: Total tumor removal with facial nerve preservation: A new surgical technique. Ear Nose Throat J. 1995;74:244–246, 248–256. 71. Liu R, Fagan P. Facial nerve schwannoma: Surgical excision versus conservative management. Ann Otol Rhinol Laryngol. 2001;110(11):1025–1029. 72. Valentino J, Boggess MA, Ellis JL, et al. Expected neurologic outcomes for surgical treatment of cervical neurilemomas. Laryngoscope. 1998;108:1009–1013. 73. Kim CS, Chang SO, Oh SH, et al. Management of intratemporal facial nerve schwannoma. Otol Neurotol. 2003;24(2):312–316. 74. Conley J, Janecka I. Schwann cell tumors of the facial nerve. Laryngoscope. 1974;84(6):958–962. 75. Lee JD, Kim SH, Song MH, et al. Management of facial nerve schwannoma in patients with favorable facial function. Laryngoscope. 2007;117(6):1063–1068. 76. Angeli SI, Brackmann DE. Is surgical excision of facial nerve schwannomas always indicated? Otolaryngol Head Neck Surg 1997;117:S144–S147. 77. Von Doersten PG. Jugular foramen schwannoma. In: Jacker R, Driscoll CLW, eds. Tumors of the Ear and Temporal Bone. Philadelphia: Lippincott Williams & Wilkins, 2000:374–387. 78. Tatebayashi K, Tanaka Y, Numata H, et al. Schwannoma of the spinal accessory nerve in the cisterna magna. Surg Neurol. 2003;59(3):217–222; discussion 222. 79. Boulton MR, Cusimano MD. Foramen magnum meningiomas: Concepts, classifications, and nuances. Neurosurg Focus. 2003;14(6): Article 10. 80. Kurokawa R, Tabuse M, Yoshida K, et al. Spinal accessory schwannoma mimicking a tumor of the fourth ventricle: Case report. Neurosurgery. 2004;54(2):510–514; discussion 514. 81. Kaye AH, Hahn JF, Kinney SE, et al. Jugular foramen schwannomas. J Neurosurg. 1984;60:1045–1053. 82. Pellet W, Cannoni M, Perch A. The widened transcochlear approachto jugular foramen tumors. J Neurosurg. 1988;69:887– 894. 83. Franklin DJ, Moore GF, Fisch U. Jugular foramen peripheral nerve sheath tumors. Laryngoscope. 1989;99:1081–1087. 84. House JW, Brackmann DE. Facial nerve grading system. Otolaryngol Head Neck Surg. 1985;93:146–147. 85. Wilson MA, Hillman TA, Wiggins RH, et al. Jugular foramen schwannomas: Diagnosis, management, and outcomes. Laryngoscope. 2005;115(8):1486–1492. 86. Sanna M, Bacciu A, Falcioni M, et al. Surgical management of jugular foramen schwannomas with hearing and facial nerve function preservation: A series of 23 cases and review of the literature. Laryngoscope. 2006;116(12):2191–2204. 87. Kadri PAS, Al-Mefty O. Surgical treatment of dumbbellshaped jugular foramen schwannomas. Neurosurgical Focus. 2004;17(2):56–62. 88. Ahmad FU, Gupta A, Sharma MC, et al. The enigmatic origin of subfrontal schwannomas: Report of a case without hyposmia. Acta Neurochir (Wien). 2006;148(6):671–672; discussion 672. Epub 2006 Feb 9. 89. Adachi K, Yoshida K, Miwa T, et al. Olfactory schwannoma. Acta Neurochir. 2007;149(6):605–611. 90. Kinoshita Y, Kurosaki M, Kamitani H, et al. Schwannoma originating in the optic canal. Acta Neurochir (Wien). 2008;150(1):89– 90; discussion 90. Epub 2008 Jan 10. 91. Kim DS, Choi JU, Yang KH, et al. Optic sheath schwannomas: Report of two cases. Childs Nerv Syst. 2002;18(12):684–689. Epub 2002 Oct 16. 92. Wewetzer K, Verdu´ E, Angelov DN, et al. Olfactory ensheathing glia and Schwann cells: Two of a kind? Cell Tissue Res. 2002;309(3):337–345. Epub 2002 Jul 17. 93. Thompson RJ, Roberts B, Alexander CL, et al. Comparison of neuregulin-1 expression in olfactory ensheathing cells, Schwann cells and astrocytes. J Neurosci Res. 2000;61(2):172–185.

537

94. Yuen A, Trost N, McKelvie P, et al. Subfrontal schwannoma: A case report and literature review. J Clin Neurosci. 2004;11(6):663–665. 95. Miller NR. Primary tumours of the optic nerve and its sheath. Eye. 2004;18(11):1026–1037. 96. Tanriover N, Kemerdere R, Kafadar AM, et al. Oculomotor nerve schwannoma located in the oculomotor cistern. Surg Neurol. 2007;67(1):83–88; discussion 88. Epub 2006 Nov 3. 97. Ohata K, Takami T, Goto T, et al. Schwannoma of the oculomotor nerve. Neurol India. 2006;54(4):437–439. 98. Mariniello G, Horvat A, Dolenc VV. En bloc resection of an intracavernous oculomotor nerve schwannoma and grafting of the oculomotor nerve with sural nerve. Case report and review of the literature. J Neurosurg. 1999;91(6):1045–1049. Review. 99. Gerganov V, Amir S, Koerbel A, et al. Cystic trochlear nerve schwannoma. Case report. Surg Neurol. 2007;68(2):221–225. Epub 2007 Jun 21. 100. Ohba S, Miwa T, Kawase T. Trochlear nerve schwannoma with intratumoral hemorrhage: Case report. Neurosurgery. 2006;58(4):E791; discussion E791. 101. Nakamura M, Carvalho GA, Samii M. Abducens nerve schwannoma: A case report and review of the literature. Surg Neurol. 2002;57(3):183–188; discussion 188–189. 102. Acharya R, Husain S, Chhabra SS, et al. Sixth nerve schwannoma: A case report with literature review. Neurol Sci. 2003;24(2):74–79. 103. Nakagawa T, Uchida K, Ozveren MF, et al. Abducens schwannoma inside the cavernous sinus proper: Case report. Surg Neurol. 2004;61(6):559–563; discussion 563. 104. Ho CL, Deruytter MJ. Navigated dorsolateral suboccipital transcondylar (NADOSTA) approach for treatment of hypoglossal schwannoma. Case report and review of the literature. Clin Neurol Neurosurg. 2005;107(3):236–242. 105. Sato K, Shimizu S, Oka H, et al. Usefulness of transcervical approach for surgical treatment of hypoglossal schwannoma with paraspinal extension: Case report. Surg Neurol. 2006;65(4):397– 401, discussion 401. 106. Tucker A, Miyake H, Tsuji M, et al. Intradural microsurgery and extradural gamma knife surgery for hypoglossal schwannoma: Case report and review of the literature. Minim Invasive Neurosurg. 2007;50(6):374–378. 107. Halefoglu ˘ AM. Neurofibromatosis type 2 associated with multiple cranial nerve schwannomas: A case report. Kulak Burun Bogaz Ihtis Derg. 2007;17(3):171–175. 108. Cheong JH, Kim JM, Bak KH, et al. Bilateral vidian nerve schwannomas associated with facial palsy. Case report and review of the literature. J Neurosurg. 2006;104(5):835–839. 109. Wittekindt C, Liu WC, Hampl JA, et al. Radiology quiz case 1: Intracranial schwannoma originating from the greater petrosal nerve. Arch Otolaryngol Head Neck Surg. 2006;132(1):102,104. 110. Vasama JP, Rahnasto J, Dastidar P. Jacobson’s nerve schwannoma. Otol Neurotol. 2007;28(1):141. ¨ U, Seker A, Kurtkaya O, et al. Internal carotid plexus 111. Ture schwannoma of the cavernous sinus: Case report. Neurosurgery. 2003;52(2):435–438; discussion 438–439. 112. Kamel MH, Mansour NH, Mascott C, et al. Compression of the rostral ventrolateral medulla by a vagal schwannoma of the cerebellomedullary cistern presenting with refractory neurogenic hypertension: Case report. Neurosurgery. 2006;58(6):E1212; discussion E1212. 113. Jagetia A, Pratap Kumar N, Singh D, et al. Pathological laughter in trigeminal schwannoma: Case report and review of the literature. Neurosurg Rev. 2006;29(4):348–352; discussion 352. Epub 2006 Sep 5. 114. Hsu WM, Wang AG. Pneumosinus dilatans associated with orbital neurilemmoma. Zhonghua Yi Xue Za Zhi (Taipei). 2002;65(5):218–224. 115. Flickinger JC, Barker FG 2nd. Clinical results: Radiosurgery and radiotherapy of cranial nerve schwannomas. Neurosurg Clin N Am. 2006;17(2):121–128, vi. 116. Pollock BE, Foote RL, Stafford SL. Stereotactic radiosurgery: The preferred management for patients with nonvestibular schwannomas? Int J Radiat Oncol Biol Phys. 2002;52(4):1002–1007.

538

Nuss and Burke

117. Sheehan J, Yen CP, Arkha Y, et al. Gamma knife surgery for trigeminal schwannoma. J Neurosurg. 2007;106 (5):839– 845. 118. Hasegawa T, Kida Y, Yoshimoto M, et al. Trigeminal schwannomas: Results of gamma knife surgery in 37 cases. J Neurosurg. 2007;106(1):18–23. 119. Huang CF, Kondziolka D, Flickinger JC, et al. Stereotactic radiosurgery for trigeminal schwannomas. Neurosurgery. 1999;45(1):11–16; discussion 16. 120. Pan L, Wang EM, Zhang N, et al. Long-term results of Leksell gamma knife surgery for trigeminal schwannomas. J Neurosurg. 2005;102 (Suppl):220–224. 121. Peker S, Bayrakli F, Kiliç T, et al. Gamma-knife radiosurgery in the treatment of trigeminal schwannomas. 2007;149(11):1133– 1137; discussion 1137. Epub 2007, Aug 31.

122. Akiyama T, Ikeda E, Kawase T, et al. Pseudocapsule formation after gamma knife radiosurgery for trigeminal neurinoma— case report. Neurol Med Chir (Tokyo). 2005;45(10):526–529. 123. Kida Y, Yoshimoto M, Hasegawa T. Radiosurgery for facial schwannoma. J Neurosurg. 2007;106(1):24–29. 124. Martin JJ, Kondziolka D, Flickinger JC, et al. Cranial nerve preservation and outcomes after stereotactic radiosurgery for jugular foramen schwannomas. Neurosurgery. 2007; 61(1): 76– 81. 125. Miller RC, Foote RL, Coffey RJ, et al. Decrease in cranial nerve complications after radiosurgery for acoustic neuromas: A prospective study of dose and volume. Int J Radiat Oncol Biol Phys. 1999;43(2):305–311. 126. Wackym PA. Stereotactic radiosurgery, microsurgery, and expectant management of acoustic neuroma: Basis for informed consent. Otolaryngol Clin North Am. 2005;38(4):653–670.

38 Paragangliomas of the Head and Neck David P. Goldstein, Mark G. Shrime, Bernard Cummings, and Patrick J. Gullane

(nodose) ganglion (5). They may be found within or below the perineurium of the vagus nerve, or dispersed among the nerve fibers (4,5). Smaller collections of paraganglia are located within the larynx, in the supraglottis (superior paraganglia), and in the subglottis, in close proximity to the cricoid (inferior paraganglia) (16).

INTRODUCTION Paragangliomas of the head and neck represent a rare group of neoplasms, arising from paraganglionic tissue located throughout the head and neck. They comprise 0.6% of all head and neck tumors, and only 0.01% to 0.03% of all tumors diagnosed in humans (1,2). Incidence in the general population ranges from 1 in 30,000 to 1 in 100,000 (1,3). Only 3% of all paragangliomas arise in the head and neck (4,5). Since the early 20th century there have been numerous reports of paragangliomas arising at various locations throughout the head and neck. Von Haller’s description of the carotid body in 1743 marks the first recorded mention of paraganglionic tissue in the literature (6). In 1862, von Luschka described tumors arising from the carotid body; surgical excision of which was first reported by Scudder in 1903 (6,7). In 1946, Rossenwasser described removal of a “carotid body tumor” of the middle ear and proposed that the tumor may have arisen from recently described glomus bodies in the temporal bone (8,9). The first description of a vagal paraganglioma was by Stout in 1935 (10).

Physiology and Function of Paraganglia Paraganglia contain two cell types: chief cells and sustentacular cells, which organize in clusters called zellballen (17). Chief cells are filled with catecholamines and tryptophanrich proteins (17). Within the chief cells of the extra-adrenal paraganglia, only norepinephrine is present, because these cells lack methyltransferase, the enzyme necessary for the conversion of norepinephrine to epinephrine (18). Sustentacular cells function as support cells. The rich microvasculature within paraganglia facilitates secretion of the granular products into the bloodstream (17). During embryogenesis, paraganglia serve as the major source of catecholamines (19). In adulthood, their role changes, and they become chemoreceptors, responding to alterations in homeostasis (20,21). The carotid body is involved in the reflex regulation of arterial pH, pO2 and pCO2 . The role of the other paraganglia is not well understood, but it may be a similar role to that of the carotid body (11).

THE PARAGANGLION SYSTEM Paraganglia, aggregates of cells located within neuronal and vascular adventitia throughout the body (11), are part of the diffuse neuroendocrine system, or amine precursor decarboxylate system (12). These cells arise from the neural crest and migrate during embryogenesis to concentrate around autonomic ganglia (3). The largest collection remains in the adrenal medulla (3), which secretes catecholamines in association with the sympathetic nervous system (11). Paraganglia located within the head and neck are associated with the parasympathetic nervous system (11). The carotid bodies constitutes the largest collection of paraganglia in the head and neck and are located in the adventia or periadvential tissue of the posteromedial wall of the carotid bifurcation (6). In the vast majority of patients, a single carotid body is found at each carotid bifurcation (13,14). In the temporal bone paraganglia are found arising along Arnold nerve (the auricular branch of the vagus nerve), and Jacobson nerve (the tympanic branch of the glossopharyngeal nerve), and juxtaposed to the jugular bulb (15). On average, there are three such glomus bodies in each ear, although the number decreases after the age of 60 (15). Slightly more than 50% of temporal bone paraganglia are located in the region of the jugular fossa, approximately 30% are within the mucosa of the cochlear promontory, and 10% lie within the inferior tympanic canaliculus (15). Vagal paraganglia may occur anywhere along the vagus nerve but are most commonly located around the inferior

Paraganglioma Nomenclature Paragangliomas are benign neuroendocrine tumors arising from the paraganglionic system (22). These tumors have, at various times, been called nonchromaffin tumors, chemodectomas, glomus tumors, and carotid body tumors. They have been described by their morphology (glomus refers to the tuft-like appearance of the tumor vasculature), by the physiologic function of the organ from which they derive (chemodectoma is derived from the Greek ch¯emeia- “chemical” and dektos- “to receive”), or by their histological properties (e.g., chromaffin vs. nonchromaffin) (11). Glenner and Grimley developed a classification system based on embryology, anatomic location, and histology which is presented in Table 1 (4,18). Under the World Health Organization Classification of Tumors, paragangliomas are classified by their anatomic location (Table 2) (11,23). Currently, the latter is the preferred classification system.

Location and Routes of Spread Paragangliomas have been described in nearly 20 distinct locations in the head and neck, with the carotid body being the most common location (12). Carotid body and jugulotympanic paragangliomas account for 80% of all head and neck paragangliomas (24) and vagal paragangliomas account for another 5% (1,5,25). Less common head and neck sites include 539

540

Goldstein et al. Table 1 Glenner and Grimley Classification Adrenal

Extra-adrenal

Pheochromocytoma

Branchiomeric Aorticopulmonary Coronary Intercarotid Jugulotympanic Laryngeal Nasal Orbital Pulmonary Subclavian Intravagal Aorticosympathetic Visceroautonomic

the trachea (26), larynx (27–29), paranasal sinuses (30–32), orbit (33), sympathetic trunk (34), and thyroid (35). Carotid body tumors (CBTs), as their nomenclature implies, arise from the paraganglia forming the carotid body (6,36). Their blood supply is derived principally from branches of the external carotid artery (ECA); they can, however, receive blood supply from branches of other major vessels, including the internal carotid (ICA) and the vertebral arteries (37). These tumors typically cause splaying of the carotid bifurcation, displacing the ICA posterolaterally and the ECA anterolaterally or anteromedially. As they enlarge, they tend to encase the ICA and ECA, without narrowing their caliber (5). Based on their location, growth may involve surrounding nerves such as the hypoglossal and vagus nerves or the sympathetic chain, or may impinge on the skull base itself. Medial extension into the parapharyngeal space occurs in up to 20% of cases (38,39). Tympanic paragangliomas develop along the tympanic canaliculus, in association with Jacobson nerve, and over the cochlear promontory within the tympanic cavity of the middle ear, in association with Arnold nerve. As these tumors enlarge, they surround the ossicles, fill the tympanic cavity, occlude the eustachian tube, and expand through the aditus ad antrum into the mastoid cavity. They may also protrude through the tympanic membrane into the external auditory canal (15). Large tumors can extend inferiorly to involve the jugular foramen, making them difficult to differentiate from jugular paragangliomas. Although bone destruction is unusual (40–42), these tumors can involve the cochlea, facial nerve, jugular bulb, sigmoid sinus, and petrous carotid (43). Tympanic paragangliomas typically derive their vascular supply from the inferior tympanic artery. Arterial supply may also come from other branches of the ICA and ECA supplying the middle ear and temporal bone (15). Jugular paragangliomas originate within the adventitia of the jugular bulb, in the lateral portion of the jugular foramen. As they grow, they may extend into the medial portion of the jugular canal. In the early stage of tumor growth, Table 2 WHO Classification of Extra-adrenal Paraganglioma of the Head and Neck Carotid body Jugulotympanic Vagal Laryngeal Aortico-pulmonary Orbital nasopharyngeal

there may only be slight erosion of the bony cortex, usually along the lateral border of the jugular fossa (15). Continued growth leads to further destruction and irregular enlargement of fossa. Tumors may grow to involve the hypoglossal canal and nerve or the intrapetrous segment of the ICA (37). Intracranial extension (ICE) is postulated to occur via spread through the jugular or hypoglossal canals into the posterior fossa or through the carotid canal into the middle cranial fossa (44). The presence of ICE is reported to be between 14% and 72% (45,46). These tumors receive their blood supply primarily from the ascending pharyngeal artery but may also receive contributions from the occipital and post-auricular arteries, and, less frequently, branches of the vertebral and posterior inferior cerebellar (PICA) arteries (37). Vagal paragangliomas (VP) can arise at any point along the course of the vagus nerve from either the superior (jugular) ganglion within the jugular fossa, or the inferior (nodose) ganglion (15). Most commonly, they originate from the latter, approximately 2 cm below the jugular foramen (47). In contrast with CBTs, these tumors typically cause displacement of the ECA and ICA anteromedially, without splaying at the carotid bifurcation. Tumors of the inferior ganglion tend to grow to involve the post-styloid parapharyngeal space (12); superior growth may lead to involvement of the skull base and jugular foramen. Tumors arising in the middle or superior vagal ganglia, in contrast, are associated with early skull base involvement and intracranial extension (12). Further extension through the jugular foramen allows posterior fossa involvement (37). The vascular supply of vagal paragangliomas is derived from the occipital and ascending pharyngeal arteries. Laryngeal paragangliomas are very rare and most commonly arise from the superior paraganglia, located within the supraglottis (16,27). Paragangliomas of the inferior laryngeal paraganglion, depending on their anatomic location, may give rise to one of two different clinical entities, namely the so-called thyroid paragangliomas, which arise within the thyroid parenchyma itself (35), and subglottic paragangliomas which are in intimate association with the cricoid cartilage (48). Paragangliomas of the sinonasal cavity are also extremely rare (16). They have been reported in all sites within the nose and paranasal sinuses, including the frontal and sphenoid sinuses, and at the terminal portion of the pterygopalatine fossa, in close association with the ptyergoid ganglion (49).

Epidemiology and Etiology Head and neck paragangliomas tend to occur in women three to four times more frequently than in men (15,18,50,51), which is in contrast to paragangliomas of other extra-adrenal sites, in which men are more commonly affected (50). Two-thirds of patients are in the fourth and fifth decades of life when diagnosed, although age at diagnosis ranges from 6 months to 88 years (15,22). An increased incidence of CBTs have been noted in patients living with chronic sustained hypoxemia (52–56), most commonly in those living at high altitudes (22,54–56). Paragangliomas within this specific group of patients have an evident female predominance (8.3:1), low rate of bilaterality (5%), and a family history of 1% (55). Emerging evidence from genetic studies implicates heredity in 35% to 55% of individuals presenting with a head and neck paraganglioma (23,57). Prior estimates of 10% are likely miscalculations, related to the mode of genetic transmission. Paragangliomas are now recognized to be transmitted in a manner influenced by maternal imprinting, which

Chapter 38: Paragangliomas of the Head and Neck

can cause the phenotypic expression of germline mutations to skip multiple generations (57). Patients with familial paragangliomas have a significantly lower age of onset (25,57–62), a higher rate of multicentricity (57,58), and are more likely to have a CBT (59), or VP (63), than paragangliomas at other sites. Similarly, patients with CBTs are 5.8 times more likely to have familial tumors than those diagnosed with paragangliomas at other sites (59). Multicentricity occurs in 10% to 15% of nonfamilial cases (3,64); in familial cases, the incidence of multicentricity ranges from 25% to 87% of patients (12,25,62,65–68).

Genetics of Paragangliomas Familial paragangliomas may occur as part of a familial tumor syndrome or as isolated hereditary tumors (57). Familial syndromes that are known to be associated with the development of paragangliomas include von-Hippel-Lindau, Multiple Endocrine Neoplasia IIA and IIB , and the Carney triad (paragangliomas, pulmonary chondromas, and gastrointestinal stroma tumors) (19,23,25,69–75). Nonsyndromic familial paragangliomas are inherited in an autosomal dominant fashion, with genomic imprinting of the maternal allele (5,19,51,62,76). Genomic imprinting implies that affected men, who pass on the unimprinted genes, have a 50% chance of having an affected child, whereas affected women will not have affected children but can pass an inactivated gene to the next generation. A male child who inherits that gene will then produce children with a 50% chance of developing the tumor (51). Inheritance is currently believed to be related to mutations in the succinate dehydrogenase (SDH) gene (77). SDH is a mitochondrial enzyme complex, playing a role in key functions of the Krebs cycle, oxidative phosphorylation, and intracellular oxygen sensing and signaling (3,77– 79). Three of the four genes encoding subunits of the mitochondrial II complex—SDHB (pgl4 on 1p35-36), SDHC (pgl3 on 1q21), and SDHD (pgl1 on 11q23)—have been implicated in the pathogenesis of hereditary head and neck paragangliomas (63,77,80–82). It is thought that these mutations lead to a chronic hypoxic signal within the cell, causing cellular proliferation and tumor formation, a theory supported by the higher incidence of paragangliomas in patients living with chronic hypoxia (57,78,83). Schiavi et al. reviewed the prevalence of different mutations in 121 symptomatic, unrelated cases, in the International Head and Neck Paraganglioma Registry (84). The prevalence of SDHC mutation was 4%, SDHB was 7%, and SDHD was 17%. In three other non-population–based studies overall mutation frequencies ranged from 12% to 41% (63,81,82). In familial cases SDHD mutations account for 50% of cases, and SDHB mutations account for 20% (57,82). SDHD mutations have also been shown to predispose to the development of multifocal paragangliomas (85), while patients with mutations in SDHB are at increased risk for malignant paragangliomas (43,85–87) and may also be at increased risk for renal cell carcinoma and papillary thyroid carcinoma (85). Familial paragangliomas constitute approximately 20% of lesions for which genetic defects are known (3). Sporadic mutations in SDHB and SDHD each occur in less than 10% of all nonfamilial cases of paragangliomas (3,57,82,88), while the pathogenesis of the remaining 80% remains unknown (3).

Genetic Counseling Genetic counseling and/or radiologic screening should be considered in family members of patients with familial paragangliomas (57,58,72,89,90). It can help identify those at risk

541

of disease, allowing for early identification and treatment (57). Carriers could also be identified, allowing their offspring to benefit from genetic counseling (57). Dundee et al. recommend that genetic counseling be offered to all patients from the age of 5 who have a family history of paraganglioma (57). Those positive for the paternal PGL gene should then undergo radiological screening every 3 years. Family members of those with the carrier state should also be offered genetic counseling from 5 years of age. In those who choose not to undergo genetic counseling, radiologic screening should be offered from the age of 10 years. Patients presenting with sporadic disease should be offered genetic counseling, particularly those presenting at a young age or with multiple tumors. Genetic screening should then be extended to family members at risk of transmission.

Pathology Paragangliomas are solid neoplasms with a homogenous tan to red-brown appearance and may be partially or completely encapsulated (22,23). On microscopic examination, paragangliomas are composed predominantly of chief cells and sustentacular cells. Chief cells are arranged in distinctive clusters of cells, referred to as zellballen, and surrounded by extensive vascular sinusoids, sustentacular cells, and a stroma composed of a prominent fibrovascular tissue (Fig. 1) (2). Chief cells are round or oval cells with uniform nuclei, dispersed chromatin, and abundant, eosinophilic, granular or vacuolated cytoplasm. Sustentacular cells are spindle-shaped, basophilic cells (modified Schwann cells). Head and neck paragangliomas tend to have the same histologic appearance irrespective of their location. In most instances, tumor cells are relatively homogenous in their appearance. Infrequently, nuclear pleomorphism, necrosis, and increased mitotic activity may be found. These cellular features do not imply malignancy but may indicate a more aggressive neoplasm (51). On immunohistochemistry tumor cells are agyrophilic and cell nests may be delineated by reticular staining (51). The chief cells stain for chromogranin, synaptophysin, and neuron-specific enolase, while sustentacular cells stain positive for S-100 (22). Argentaffin, mucin, and periodic acid Schiff stains are negative (22). On electron microscopic examination, the hallmark of these tumors is presence of neurosecretory granules within the chief cells. The differential diagnosis of these tumors on light microscopy includes carcinoid tumors, neuroendocrine carcinomas, medullary thyroid carcinomas, middle ear adenomas, meningiomas, hemangiopericytomas, alveolar soft part sarcomas, and metastatic renal cell carcinomas (22).

Malignant Paragangliomas The diagnosis of malignant paragangliomas rests on clinical behavior rather than histologic appearance. Prominent sustentacular cells, necrosis, mitotic activity, nuclear pleomorphism, and perineural, bony, and vascular invasion may be seen in both benign and malignant paragangliomas (2,25,91,92). A paraganglioma is determined to be malignant only if metastases to non-neuroendrocine tissue can be demonstrated (Fig. 2) (2,25,91,92). Nodal metastases occur more commonly than distant metastases to the lungs, liver, bone, and skin (2,23,93,94). Examples of metastasizing paragangliomas have been reported in all head and neck locations from which paragangliomas arise (91). In general, less than 5% of all paragangliomas are malignant (2,91); however, prevalence depends on the site of the primary (12). Vagal paragangliomas appear to be associated with the highest rates

542

Goldstein et al.

Figure 1 Carotid body paraganglioma showing nests of clear cells surrounded by thin capillaries. This appearance represents the classical “Zellballen” pattern (A). The tumor cells are diffusely positive for the neuroendocrine marker synaptophysin (B). S-100 immunohistochemical stain highlighting peripheral sustentacular cells (C).

(10–19%) of malignancy among the more common head and neck sites (12,18,91,92), while orbital and laryngeal paragangliomas have slightly higher reported rates (20–25%) (12,95). Malignant CBTs and jugulotympanic paragangliomas have been reported to occur in about 3% to 6% of cases (7,91,96). In a National Cancer Database review of 59 cases of malignant paragangliomas, 68.6% were confined to regional nodes, and 31.4% of patients had distant disease (2). The overall 5-year relative survival rate was 59.5%. The 5-year survival rate for patients with metastases limited to lymph nodes was 76.8%, significantly higher than that for patients with distant metastases (11.8%) (2)

Secreting Paragangliomas Fewer than 5% of head and neck paragangliomas secrete catecholamines in quantities sufficient to produce symptoms (5,12,15,63,66,97,98). The vast majority of secreting tumors produce norepinephrine. Tumors producing serotonin, kallikrein, and histamine precursors have also been described, often causing carcinoid-like syndromes (15,99). Symptoms associated with secreting paragangliomas are the same as those seen with pheochromocytomas, namely excessive sweating, hypertension, tachycardia, nervousness, and weight loss (12,97). Breakdown products of norepinephrine, including metanephrine (normal: 5 cm) compressive tumors (51,110). Table 5 Antonio De La Cruz Classification for Jugulotympanic Paragangliomas Anatomic classification

Surgical approach

Tympanic Tympanomastoid Jugular bulb

Transcanal Mastoid-extended facial recess Mastoid-neck (possible limited facial nerve rerouting) Infratemporal fossa Infratemporal fossa/intracranial

Carotid artery Transdural

544

Goldstein et al.

Table 6 McCabe/Fletcher Classification of Temporal Bone Paragangliomas Group 1 1. Absence of bone destruction 2. Intact eight nerves 3. Intact jugular foramen nerves 4. Absence of facial weakness Group 2 1. Bone destruction confined to mastoid 2. Facial nerve normal or paretic 3. Jugular foramen nerves intact 4. Superior bulb of jugular vein uninvolved ny retrograde jugulography Group 3 1. Destruction on roentenogram involving petrous bone, jugular fossa, and occipital bone 2. Positive retrograde jugulography 3. Jugular foramen syndrome 4. Presence of metastases 5. Carotid arteriogram evidence of destruction of pertrous/ occipital bone

Jugulotympanic Paragangliomas Hearing loss and pulsatile tinnitus are the most common presenting symptoms, manifesting in 55% to 58% and 56% to 82% of patients respectively (15,111–115). Hearing loss is typically conductive, but, with tumor invasion into the labyrinth, patients may develop a sensorineural hearing loss and vertigo (15). On examination a reddish-blue, pulsating mass may be seen behind the inferior portion of an intact tympanic membrane (Fig. 6); pneumatic otoscopy causes the mass to blanch (Brown sign). Tympanic paragangliomas tend to remain well-defined, intra-tympanic soft-tissue masses (Fig. 7). In contrast the presence of intracranial extension in jugular paragangliomas at presentation has been reported to be as high as 72% (45,46,116). Whereas tympanic paragangliomas commonly manifest themselves early with pulsatile tinnitus and conductive hearing loss, jugular paragangliomas typically present late, frequently with at least one CN neuropathy (111). Palsy of the lower CNs usually precludes the diagnosis of a tympanic paraganglioma (117); neuropathies in tympanic para-

Figure 4 Clinical appearance of a carotid body tumor presenting as a lateral neck mass anterior to the sternomastoid muscle at the level of the carotid bifurcation.

Figure 5 CT scan of a large left carotid body tumor extending into the parapharyngeal space with bulging of the oropharyngeal wall.

gangliomas occur uncommonly and are usually limited to the seventh and eighth CN (40). The reported rates of CN neuropathies in jugular paragangliomas ranges from 8% to 60% (38,45,112,115,118–121). Palsies of the ninth, tenth, and eleventh CNs, producing jugular foramen (Vernet’s) syndrome may be occasionally encountered (38). Further skullbase extension can result in hypoglossal nerve involvement

Figure 6 Otoscopic examination of a patient with a jugular paraganglioma demonstrating a reddish-blue mass behind the inferior portion of an intact tympanic membrane.


545

presentation, with the vagus nerve most commonly affected (47,66,130,131). Miller et al. reported that at presentation at least 50% of patients will have 1 or more CN palsies (47). Netterville et al. noted that 36% of patients presented with CN deficits (66). In 28% of patients the vagus nerve was totally or partially paralyzed. Other CN deficits upon presentation included CN XII (17%), CN XI (11%), CN IX (11%), CN VII (6%) and sympathetic chain (4%) (66).

Laryngeal and Sinonasal Paragangliomas Patients with laryngeal paragangliomas commonly present with varying degrees of dysphonia, stridor, dysphagia, and dyspnea (16,51). Supraglottic paragangliomas cause hoarseness and impair deglutition, whereas subglottic tumors produce airway obstruction (16,132). Fiberoptic examination classically reveals a submucosal mass with normal vocal cord function (16). Sinonasal paragangliomas present in a manner similar to that of any expansile mass lesion of the nasal cavity with nasal obstruction and epistaxis (16,51). These tumors are usually painless and may be misdiagnosed as nasal polyps (51). With advanced disease local symptoms will result from involvement of the facial skin and/or orbit (16,51).

Diagnostic Imaging Figure 7 Otoscopic examination of a patient with a tympanic paraganglioma immediately inferior to the umbo of the malleus. Unlike a jugular paraganglioma the mass is situated over the promontory and well-circumscribed with all margins visible.

(Collet–Sicard syndrome). Facial nerve paralysis is associated with advanced disease (114) causing medial and posterior extension of the tumor through the facial recess and retrofacial air cell tract to involve the facial nerve in its horizontal and vertical segment (122). Horner syndrome may occur with involvement of the sympathetic plexus, particularly when the carotid artery is involved in its petrous or cervical portion (44). The range in percentages of CN deficits at presentation is presented in Table 7 (114,116,118,122–128). In a literature review of 509 cases of jugular paragangliomas, Lustig et al. reported the rate of preoperative paralysis for the following CNs to be: 19% for CN IX, 24% for CN X, 16% for CN XI, and 21% for CN XII (129).

Vagal Paragangliomas VPs most commonly present as a painless upper neck mass, which is typically located more cephalad than CBTs (130). Tumors originating from the nodose ganglion manifest clinically as a mass behind the angle of the mandible, with associated hoarseness and vocal cord paralysis. Those that arise more superiorly impinge on the jugular foramen, producing Vernet or Collet–Siccard syndrome (15). With temporal bone involvement, hearing loss and pulsatile tinnitus may be present (47,66). Overall, CN paralysis is common at Table 7 Rate of Cranial Nerve Deficits at Presentation of Jugulotympanic Paragangliomas Cranial nerve IX X XI XII

Preoperative paralysis rate 4–43% 2–57.1% 4–43% 7–57%

Source: From Refs. 114,116,118,122–128.

Diagnostic imaging is an integral aspect of the work-up of patients presenting with symptoms and signs suggestive of a paraganglioma. Imaging aids diagnosis and treatment by delineating tumor location, displacement of major vessels, involvement or invasion of surrounding structures, and detection of multicentric tumors and metastases.

Ultrasound Ultrasound (US) can be used in the initial assessment of a neck mass suspected to be a paraganglioma, and is often sufficient to make the preliminary diagnosis of a CBT. In B-mode US paragangliomas present as a solid, well-defined, hypoechoic, heterogeneous mass (133). Color Doppler US can define the vascularity of the mass and its relationship to the ICA and ECA133. Because most VPs extend above the hyoid bone, the value of US for these lesions becomes limited (134). The direction of blood flow can, however, help differentiate CBTs from VPs; CBTs exhibit upward intratumoral blood flow (133), while downward intratumoral flow of a similarly located neck mass suggests a VP (133).

Computed Tomography High-resolution CT scanning is used to evaluate most patients with suspected head and neck paragangliomas. CT shows the anatomic relationship of the tumors with surrounding structures, and their relationship with the major vessels of the neck. Accuracy diminishes with lesions less than 8 mm in size (12,76). In general, paragangliomas appear as homogenous masses with intense enhancement following contrast administration. CT can differentiate between CBTs and VPs based on the relationship to the ICA and ECA (Figs. 8 and 9) (15,134). CT is particularly valuable in assessing bone invasion at the skull base (Fig. 2), which aids in differentiating jugular and tympanic paragangliomas, as well as in staging these tumors. If CT demonstrates an intact bone plate separating a middle ear mass from the dome of the jugular bulb a tympanic paraganglioma is the most likely diagnosis (15). If the caroticojugular spine, which separates the petrous carotid artery from the jugular bulb, is destroyed, a jugular paraganglioma extending into middle ear is more likely (15). However, the corollary is not always true; its preservation

546

Goldstein et al.

Magnetic Resonance Imaging

Figure 8 CT demonstrating the typical relationship of a left carotid body tumor to the carotid artery with splaying of the bifurcation, and displacement of the ICA posterolaterally and the ECA anterolaterally or anteromedially.

does not necessarily preclude the existence of a jugular paraganglioma (43,135). Finally, CT may also help rule out other causes of pulsatile tinnitus, such as an aberrant carotid artery, a high jugular bulb, or a persistent stapedial artery.

Figure 9 CT of a patient with bilateral vagal paragangliomas demonstrating displacement of the ECA and ICA anteromedially (circled), without splaying at the carotid bifurcation.

MRI is complementary to CT in patients being investigated for paragangliomas. MRI offers several benefits including (a) detection of smaller paragangliomas (12,37,136–138) (b) increased sensitivity for invasion or erosion of adjacent vessels, dura, or brain (12), and (c) the ability of magnetic resonance angiography (MRA) and venography (MRV) to provide noninvasive delineation of adjacent arterial and venous structures. On T1- and proton-weighted images paragangliomas display a low-to-intermediate signal; they are relatively hyperintense on T2-weighted images (12,15). On postcontrast T1-weighted spin-echo images, the tumors enhance strongly and homogenously (15,134). Paragangliomas exhibit a characteristic “salt and pepper” pattern of hyper- and hypointensity on T1- and T2-weighted images, most often evident in tumors greater than 1.5 cm in diameter (15,136,138). This appearance, which represents the existence of sinusoidal flow-voids within the tumor, is accentuated with gadolinium (Fig. 10). While characteristic, these findings are not specific to paragangliomas and may be seen with other hypervascular lesions, such as metastatic renal cell carcinoma (134,139). Short T1-inversion recovery (STIR) sequences are now also commonly used to reduce signal from fat, aiding in lesion detection (15). MRA has the ability to facilitate detection of multicentric paragangliomas and to determine feeding vessels preoperatively (Fig. 11) (138,140). Three-dimensional time of flight (TOF) MRA allows detection of vessel displacement, gross tumor involvement, and compromised blood flow (12). The combined use of spin-echo images and 3D TOF MRA has a higher sensitivity and similar specificity (90%/92%) to conventional 3D phase contrast angiography (72%/97%) (37,124,138,141,142). MRV can also demonstrate compression or invasion of the jugular bulb (15). Even with the techniques

Figure 10 MRI of a left jugular paraganglioma, axial T1-weighted study with gadolinium demonstrating the classic “salt and pepper” pattern of hyperand hypointensity.


547

Figure 12 Conventional angiography of a jugular paraganglioma demonstrating an intense tumor blush confirming the tumor’s vascularity. Feeding vessels can also be assessed. Figure 11 MRA of a carotid body tumor demonstrating the “lyre sign,” which results from splaying of the internal and external carotid arteries.

discussed, MRA is not sensitive enough to demonstrate the detailed vascular supply to these tumors, which are best evaluated with digital subtraction superselective angiography when required (140).

Diagnostic Angiography Digital subtraction angiography is able to delineate, in exquisite detail, the vascular anatomy of these tumors, their feeding blood supply, and any potential vascular invasion or compromise. It is a sensitive test for small and multicentric paragangliomas (19,74,118,143). The hallmark appearance of a carotid body tumor on angiography is the “lyre sign,” which results from splaying of the carotids, and an intense tumor blush confirming the tumor’s vascularity. Temporal bone paragangliomas also reveal a characteristic angiographic appearance, consisting of enlarged feeding arteries, an early, intense, and slightly inhomogeneous tumor blush, and early-appearing draining veins (Fig. 12) (15). Despite its benefits, angiography is an invasive procedure with potential risks and is rarely required for diagnosis, assessment, and treatment planning. Its use is therefore currently limited to cases in which the diagnosis remains in question after CT and MRI (15,144), cases in which preoperative embolization is desired, or for preoperative planning in cases when there is suspicion of carotid artery involvement. In the latter circumstance angiography allows for assessment of the adequacy of the patency of the intracranial circulation (134), and, when used in conjunction with balloon occlusion to assess the adequacy of contralateral cerebral blood flow, can guide decisions regarding carotid artery sacrifice or bypass.

Octreotide Scintigraphy Paragangliomas, as with other neuroendocrine tumors, express somatostatinergic receptors and are well suited for imaging with octreotide scintigraphy (5,145). Octreotide is a somatostatin analog that, when coupled to a radioisotope

(111 indium-labeled-DPTA), produces a scintigraphic image (134,146). Areas of increased uptake can then be examined further with MRI. Octreotide scintigraphy appears to be a safe and relatively noninvasive method for early diagnosis of familial paragangliomas and is useful for the detection of multicentric tumors (134,147). The reported sensitivity and specificity in diagnosing paragangliomas (usually >1 cm) ranges from 94% to 97% and 75% to 82%, respectively (146,148,149). It has also been shown to have a role in detecting recurrent paragangliomas since octreotide binding is not affected by postsurgical or radiation changes (145,149).

Management Treatment options in the management of head and neck paragangliomas include observation, surgery, radiation, singly or in combination. Traditionally, surgery has been the preferred modality of treatment (12), while radiation was reserved for unresectable tumors or those occurring in elderly or debilitated patients. Advances in radiation therapy have led to improved long-term response with acceptable complications (12), allowing it to be considered a primary treatment option (112). Decisions regarding treatment modality must take into consideration the biologic behavior, size, and site of the tumor, the age and general medical status of the patient, and potential treatment-related morbidity (51). An evidence-based approach to optimal treatment for patients with paragangliomas is limited by the rarity of the tumor and the nature of available studies. The vast majority of studies are retrospective with small subject numbers. They often include a heterogeneous population of tumors arising at different sites and treated in nonuniform ways over many decades, during which time there have been significant advances in interventional radiology, surgery, and radiotherapy. Direct comparison between surgery and radiation is made difficult by the fact that the two treatment modalities measure success by different standards: in general, surgical success is measured by total tumor resection without recurrence, whereas radiation therapy judges treatment response by the

548

Goldstein et al.

absence of radiologic evidence of tumor growth. In addition, there is often a selection bias in these studies, with more advanced tumors referred for radiation and smaller tumors approached surgically. Based on current literature control rates for surgery and radiation are similar. Treatment modality decisions are therefore often based primarily on the risk of treatment-related complications.

Observation Based on the natural history of these tumors observation may be sufficient in selected cases, such as asymptomatic elderly patients with jugular paragangliomas, and patients with multiple, small, synchronous, and asymptomatic tumors. Serial imaging should be performed on a yearly basis; if significant tumor growth or clinical symptoms are noted, consideration can then be given to other interventions.

Surgery Surgery should be considered when tumors can be completely excised without significant morbidity. This group is usually limited to patients with small to moderate size CBT and tympanic paragangliomas who present without cranial nerve dysfunction. Surgery is also indicated for patients with jugular or VPs who present with lower CN dysfunction, in whom tumor resection will not result in additional significant functional deficits. Relative contraindications to surgery include extensive skull base or intracranial involvement, advanced patient age, medical comorbidities, and bilateral or multiple paragangliomas, the resection of which may result in unacceptable postoperative morbidity and bilateral lower CN palsies (12). If surgery results in significant disability requiring long-term rehabilitation, radiation should be considered (112). Since paragangliomas are often situated near or involve the skull base and carotid artery, they are best managed by a team of head and neck surgeons, neurosurgeons, neuro-otologists, and vascular surgeons. Speech therapists and physiotherapists are also integral members of the multidisciplinary team, addressing the potential postoperative speech, swallowing, and shoulder difficulties that may arise. Appropriate preoperative management of secreting tumors is essential. In cases in which tumors being considered for surgery involve the carotid artery, consideration should also be given to assessment of cerebral blood flow and possible balloon occlusion.

Superselective Angiography and Preoperative Embolization There is considerable debate in the literature about the usefulness of preoperative embolization (39,67,106,150). Advantages include decreased blood loss, a lower transfusion requirement (38,118), and, consequently, a reduction in operative time (15). However, embolization of paragangliomas is associated with the risk of stroke or blindness due to potential anastomosis between the ICA and ECA. The decision to use preoperative embolization depends on tumor location and size. Embolization has a well-recognized role in the surgical management of jugular paragangliomas (118), whereas tympanic paragangliomas usually do not require it. The use of preoperative embolization in the treatment of CBTs and VPs is a matter for debate. When indicated, surgery should be performed within two days of embolization in order to avoid recruitment of collateral tumor blood supply in the postinflammatory phase (118). Intraoperative ligation

of the ECA must also be avoided, as doing so prevents future embolization for recurrences.

Surgery for Carotid Body Paragangliomas Surgery is the preferred modality for treatment of CBTs (6,7,21,108,110,124,151). Relative contraindications to surgery particular to CBTs include Shamblin type III tumors and patients with bilateral tumors who have sustained a cranial nerve or sympathetic trunk injury on one side (6). Dissection of the tumor off of the carotid artery can be performed in either a subadventitial or peri-adventitial plane (152). The former carries a risk of arterial wall injury, whereas dissection in the capsular or peri-adventitial plane may carry a lower risk of vessel injury (153). Complete resection can usually be accomplished in most tumors and nerves adherent to the tumor can often be mobilized without injury (21). The extent of surgery and the risk of complications, such as stroke, carotid artery, and CN injury, depend on the Shamblin classification (7). Type I tumors can often be removed with few complications and without the need for shunting or carotid reconstruction (6). For complete removal of type II tumors, carotid shunting may be required, and type III tumors may require carotid reconstruction (6). When indicated, vascular bypass and reconstruction also allows resection of tumors with a low incidence of stroke (51,154). Intraluminal shunting should be employed in circumstances in which the common or ICA must be sacrificed or repaired, in order to reduce the risk of stroke (153,155,156). Ligation of the carotid artery without shunting must be avoided as it carries up to a 66% chance of stroke (21). Even patients who pass balloon occlusion testing still run a 25% chance of delayed stroke after carotid artery ligation without shunting (21,157). In over 75% of cases CBTs can be resected without the need for arterial reconstruction (156) and is more common in class II and III tumors, malignant lesions, and tumors over 6 cm in size (38,108,154,158). The most common complications are stroke and CN injury (110). The incidence of stroke is less than 2% in the current literature (12,108,155,156). The reported incidence of carotid artery injury is between 2% to 12.5% (39,107,159–161). Rates of postoperative CN injury range between 0% to 71% (162), with the more recent studies reporting rates between 13% and 56% and the majority occurring with type II or III tumors (106–108,110,160). Nerves at risk of injury include the vagus nerve, the superior laryngeal nerve, the glossopharyngeal nerve, the hypoglossal nerve, and the marginal mandibular branch of the facial nerve, with the vagus nerve being the most commonly injured (110).

Management of Bilateral Carotid Body Tumors Patients with bilateral CBTs present a management dilemma due to the potential for bilateral vagus nerve injury and baroreflex failure syndrome. Denervation of both carotid baroreceptors results in tonic inhibition of the parasympathetic input to arterial blood pressure and unopposed sympathetic activity (163,164). Patients experience sudden hypertension associated with episodes of flushing, headache, diaphoresis, and emotional lability, followed by sharp fluctuations in arterial pressure and heart rate typically within 24 to 72 hours after surgery (165). Marked hypotension and bradycardia may also occur when the patients are drowsy or sedated (165,166). There appears to be a wide spectrum of symptom severity, and timing of onset and resolution between patients (163,167). While there is an apparent heterogeneous response to arterial baroreflex dysfunction, the normocapnic hypoxic drive is invariably lost (164). The


long-term clinical course is also variable: some patients experience a lifetime of hypertension while others improve without medical control (167). Clonidine is an excellent agent since it acts as a central alpha-2 agonist, thereby decreasing circulating norepinephrine levels (163,165). Episodes of hypotension have been shown to respond to the administration of low dose corticosteroids (68,165,168). There is no standard management protocol for bilateral carotid tumors. The decision of which modality to use on each CBT depends on the tumor stage, morbidity associated with its management and the preoperative status of the function of the vagus nerve. Whichever treatment modality is chosen every effort must be made to preserve the function of at least one vagal nerve and its laryngeal branches. One tumor can be resected with the contralateral CBT managed with observation, surgery, or radiation. If surgery is considered for both sides, the smaller tumor should be removed first with assessment of the function of the vagal nerve postoperatively before undertaking resection of the contralateral tumor. In cases where there is either pre- or postoperative vagal nerve paralysis, the contralateral tumor should be managed conservatively with either observation or radiation. Another treatment approach would be radiation to the tumor on both sides. However, baroreflex failure can also be a long-term sequelae of radiation therapy (169,170).

Surgery for Vagal Paragangliomas VPs confined to the neck have been frequently managed with surgery (51). Relative contraindications include elderly patients and bilateral tumors, since surgery inevitably results in sacrifice of the vagus nerve (66,160,171). The majority of tumors can be removed with a lateral cervical approach, even in tumors greater than 5 cm (47). With large tumors extending up to the skullbase, a mandibulotomy may be required for either adequate exposure or vascular control at the skull base (51). When surgery is performed for tumors that traverse the skull base, a combined cervical and mastoid approach may be indicated in order to achieve safe and wide exposure. The major morbidity following surgery is related to postoperative CN dysfunction, which is significantly more common with VPs than for CBTs (51). While an attempt can be made to preserve the vagus nerve, excision of these tumors almost always requires its sacrifice (47,66,102). Nerves which are spared frequently do not recover function (47). As a result, postsurgical CN deficits should be the main consideration when formulating a treatment plan. Vagal nerve paralysis is frequently complicated by an associated hypoglossal and glossopharyngeal nerve paralysis adding additional morbidity (47,66,102). Surgical resection may be also associated with vascular injury, especially in larger tumors and those with skull base involvement (51), although less frequently than CBTs (12). Biller et al. noted that the risk of carotid injury increases with increasing tumor size, as tumors larger than 5 cm tend to cause significant displacement of the carotid artery at the skull base (172). As a result of this high potential for morbidity from CN neuropathies and vascular injury, patients should be selected carefully for surgical management of VPs. Other management options such as radiation or observation should be considered, particularly in asymptomatic patients.

Surgery for Glomus Tympanicum Because the rate of intra- and postoperative morbidity is low in these patients, surgical resection is the treatment of choice for tympanic paragangliomas confined to the middle ear or mastoid (Fisch Class A and B) (45,67). Tumors confined to the

549

middle ear, in which the entire margin is visible (Glasscock– Jackson type 1, Fisch Class A), can be removed through a transcanal approach. Laser-assisted excision with either a KTP or an Nd:Yag laser has also been described for these tumors (173–175). Large tumors or those whose margins cannot be visually confirmed usually require an extended facial recess approach (117). This approach allows assessment of the relationship of the tumor to the ossicular chain, tympanic membrane, labyrinth, facial nerve, and jugular bulb (117). Surgical management results in total tumor removal and long-term tumor control in the majority of patients (117). Recurrence rates after surgical resection ranges from 2.5% to 5% (117,176); postoperative residual tumor is the main risk factor for recurrent disease. Complications are uncommon with Fisch Class A and B tumors (176,177). With more extensive tumors, complications include CSF leak, stroke, bleeding, hearing loss (conductive and/or sensorineural), and facial paralysis.

Surgery for Jugular Paragangliomas Management of jugular paragangliomas remains controversial. Surgery is curative, offering the possibility of immediate and complete tumor elimination; there is, however, considerable risk of morbidity. Advances in skull-base and microsurgical techniques, neuromonitoring, embolization, and preoperative evaluation of cerebral blood flow has given surgeons the ability to resect the majority of tumors with fewer complications (178). It is, therefore, the extent of morbidity, mainly related to CN dysfunction, that determines the role of surgery in the management of tumors in this location. The surgical approach to jugular paragangliomas must be tailored based on the routes of extension of the tumor, which are highly variable. The approach must take into account the tumor size, extent of distal ICA control necessary, and must provide access to all tumor margins. A combined transmastoid and transcervical approach can be used for small jugular paragangliomas that do not involve the carotid artery or posterior cranial fossa (44). Cranial nerves can be preserved if they can be adequately separated from the tumor (44). For larger tumors impinging on the facial nerve but still not involving the carotid artery or posterior cranial fossa, a similar approach can be used with limited facial nerve rerouting (44). More extensive tumors will require additional exposure with an infratemporal approach, affording access to the posterior and middle cranial fossa and petrous carotid artery. In these cases, complete exposure of the vertical portion of petrous carotid requires rerouting of the facial nerve from the geniculate ganglion laterally. Surgical exposure of the ICA must guarantee its proximal and distal control as well as visualization of its entire circumference to enable its mobilization. For tumors abutting or involving the ICA, accurate preoperative assessment of the extent of involvement and of cross-perfusion from contralateral vessels is imperative for safe resection of this tumor (116). When there is potential for carotid sacrifice, consideration should be given to balloon occlusion or revascularization (179). Involvement of the brain or cavernous sinus often makes complete resection impossible. Embolization continues to have a role in the surgical management of large tumors involving the jugular foramen and skull base (180). Gross tumor removal is possible in 70% to 96% of cases (45,116,123,178,181). The extent of resectability depends upon the stage and complexity of the tumor. Complex tumors (i.e., giant tumors, multiple, malignant or secreting tumors, previous treatments) have a significantly lower rate of complete resection than their noncomplex counterparts (178). Long-term

550

Goldstein et al. Table 8 Rate of Postoperative New Cranial Nerve Deficits Following Resection of Jugulotympanic Paragangliomas Cranial nerve VII IX X XI XII

Range 4.4–11% 26–42% 13–28% 25–26% 12.5–21%

Source: From Ref. 178.

local control is achieved in 80% to 95% of cases in most surgical series, with surgical control in noncomplex tumors approaching 95% (45,67,116,123,126,178,179,182–184). The overall recurrence rate is 2% to 3% when total tumor removal is possible, with a mean time to recurrence of 82.8 months (45). Because recurrences after surgery have been documented up to 23 years later, lifelong follow-up is required (179). Bleeding, stroke, CSF leak, and CN dysfunction are the most significant complications. Total resection without CN injury is only achievable in 31% of patients (51). Table 8 shows the rate of development of new lower CN injuries in more recent surgical series of jugular paragangliomas resections as reported by Gottrified et al. (178). The lack of preoperative CN dysfunction does not correlate well with intraoperative neural involvement and therefore does not predict postoperative deficits (116). Serviceable hearing is salvageable in over 90% of patients (126,178). Facial nerve injury most commonly results from manipulation of the nerve for access. Rates of recovery to House-Brackmann grade 1 or 2 levels of function vary from 60% to 87% depending on extent of rerouting (123,183–186). In most series of noncomplex tumors, the rate of CSF leak ranges from 3% to 11%, occurring most frequently in Fisch class D tumors (45,116,123,178,179,183). The majority resolve without surgical intervention (123,126,178). On a review of 384 patients from seven surgical series on jugular paragangliomas, Gottfried et al. noted the following complication rates: CSF leak occurred in 8.3% of patients, aspiration in 5.5%, and stroke in 1.6% (178). The overall mortality rate was 1.3%. Death usually resulted from injury to the ICA, intracerebral hemorrhage, or pulmonary embolism. In an attempt to reduce complications associated with intradural (Fisch class D2) tumors, some authors recommend staging the procedure, removing the extradural tumor first, followed by removal of the intradural at a later date (116,123). Rehabilitation is imperative in patients developing CN dysfunction following surgical resection of jugular paragangliomas. Full rehabilitation following surgical resection of class C and D tumors may take up to 1 to 2 years (181). With time and aggressive postoperative speech and swallowing therapy, patients may be able to compensate well (126,128,181). This, however, depends upon age, the CNs involved, and the preoperative status of the CNs. Younger patients and those with preoperative dysfunction have better outcomes with rehabilitation (126,187). There is wide variation in the role surgery plays in the management of jugular paragangliomas. Some authors suggest that gross total tumor removal should be performed in all young patients since they have a greater capacity to compensate for loss of CN function. In patients in whom tumor cannot be resected completely, subtotal resection followed by observation with serial imaging has been suggested as an option, since growth after surgical devascularization is rare; others follow with stereotactic radiation (116,185). Overall, surgery

can be considered in young patients with pre-existing CN dysfunction, in whom surgery is unlikely to cause additional morbidity, and it should generally not be offered in patients older than 60 years when preoperative CN deficits are not present. For all other tumors considerable debate still exists.

Radiation Advances in radiation therapy have led to improved longterm response with minimal morbidity (12). Currently, radiation therapy may be considered a reasonable primary treatment option for paragangliomas, particularly those in elderly patients, patients with multiple or severe medical conditions, or patients with extensive skull base or intracranial involvement (112). Radiotherapy is also indicated in patients with jugular or vagal paragangliomas without evidence of lower CN dysfunction or patients with multiple or bilateral tumors, which if surgically resected would result in significant morbidity and disability (12,68,120,171). Both conventional fractionated radiotherapy and stereotactic radiosurgery have been used to treat paragangliomas. Unlike with surgery, successful treatment with radiotherapy is defined as stability or regression of tumor size and neurological symptoms, not gross tumor removal (112). The exact mechanism by which radiation prevents tumor growth is not completely understood, although it is theorized that radiation induces an obliterative endarteritis with resultant fibrosis (112,178,188). It is difficult, however, to determine if radiation has a primary effect on the chief cells themselves: because morphologically intact but persistent cells may lose their ability to reproduce (189).

Conventional Fractionated Radiotherapy The doses of conventional radiotherapy commonly used in the literature ranges between 35 to 60 Gy. With these doses the majority of lesions remain stable in size or show modest regression on radiographic evaluation after completion of radiation (115,190,191). In general a total dose of 45 Gy in 25 fractions, using once-daily fractions, in continuous course delivered over 5 weeks is recommended to control benign paragangliomas (112,192,193). A schedule of 35 Gy delivered over 3 weeks has also been shown to be effective (113). Patients with unilateral disease can almost always be treated with an ipsilateral field arrangement, allowing for the minimization of long-term xerostomia. Patients with bilateral tumors are usually treated with more complex techniques. For malignant paragangliomas treated with surgery, postoperative radiation is often indicated at a more intensive dose fractionation schedule, typically 60 to 66 Gy in 6 to 6.5 weeks. There is an extensive body of literature with over 30 years of follow-up assessing the effectiveness of radiation therapy in controlling head and neck paragangliomas (112). Similar to the surgical literature, most of the radiation studies are single-institution, retrospective case series. There is significant heterogeneity both within and between studies in terms of years of analysis, dose levels, radiation source (electrons vs. photons), mode of delivery (external beam, stereotactic radiosurgery), and beam energies (superficial, cobalt, and 4–6 MV) (112). Despite these limitations, long-term control rates have been reported with radiation therapy (19). A review, by Hu et al., of 1000 tumors from 35 studies noted a 10-year local control rate for radiation therapy ranging from 65% to 100% with a mean of 90% (112). Schild et al. reported local control of 91% in a review of series published between 1965 to 1992 (194). Springate et al. also reviewed the literature on the management of head and neck paragangliomas published from 1965 to 1988; they found a control rate of 93%


in patients treated with radiotherapy alone (195). It is worth noting, however, that studies reviewed often include patients managed with combined modality therapy. Radiation therapy has also been shown to result in resolution of symptoms and CN neuropathies in some patients. Tinnitus may improve or resolve in up to 88% of patients (112), and rates of improvement of CN neuropathies vary from 0% to 83%, with an average of 35% (112). Complete resolution of CN dysfunction occurs in about 10% of cases (range 8–20%). Notably about 1% of CN neuropathies worsen after radiotherapy (112,196,197). There is a growing body of literature to suggest that CBTs and VPs can be treated with radiation therapy when surgery will be associated with significant morbidity (6,112,162). Based on observational data, with limited numbers of patients, the overall likelihood of local control after radiation for cervical paragangliomas is essentially the same as surgery and complications are infrequent (104,192,198). While local control in a number of small series ranges from 96% to 100% and complications are rare (104,196,198–200), surgery is still considered the primary treatment for CBTs since the majority can be resected without significant morbidity. The most significant role for radiation is in the management of temporal bone paragangliomas, or VPs in which surgical removal can result in injury to the lower CNs and attendant significant morbidity. Hinerman et al. treated 55 temporal bone paragangliomas (46 JP, 9 GT) with radiation alone (104). Local control was obtained in 40 of 43 (93%) of previously untreated lesions. All Macabe-Fletcher Group I and II patients were controlled as opposed to a control rate of 87% for Group II tumors. The overall cause-specific survival was 98%, and no patients developed regional or distant metastasis. Two patients died, one of whom was treated with radiation for recurrence after surgery, and one of whom discontinued treatment. In the absence of tumor progression, no patients developed CN deficits. Local control rates with radiation alone for temporal bone paragangliomas range from 90% to 100% (43,104,113,195,201–204). Although some authors voice concern that tumor growth may resume 10 to 15 years later (43), literature supporting this notion is limited. Pemberton et al. assessed the management of glomus jugulare and glomus tympanicum tumors with external beam radiation therapy in 49 patients given a median dose of 45 Gy (201). At 5 and 10 years, 92% of patients were recurrence-free, and diseasespecific survival was 96%.

Stereotactic Radiosurgery First described in the mid-1990s (205,206), stereotactic radiosurgery or hyperfractionated stereotactic radiation has been used with increasing frequency in the treatment of paragangliomas of the head and neck. The initial small studies focused on jugular and tympanic tumors, and had limited follow-up times, but their results were promising enough to engender interest in this treatment modality. The earliest multicenter trial in jugular paragangliomas demonstrated that, in the majority of cases, stereotactic radiosurgery arrested tumor growth in 60% (28 of 47 patients) of patients and provided actual tumor regression in 40% (19/47). Symptoms improved in approximately 30% of patients, remained stable in 65% of patients, and deteriorated in approximately 5% of patients (207). Unfortunately, this trial was also limited in follow-up and utilized stereotactic radiosurgery both as a primary and secondary treatment modality, a difficulty that has plagued studies.

551

Further trials in jugular paragangliomas, with followup ranging from 2 to 7.2 years, have corroborated these findings. On average, tumors regress in 10% to 67% of patients and increase in size in only 5%. Subjective symptoms improve in up to 67% of patients, remain stable in approximately 30% to 40%, and increase in fewer than 10% (208–211). In single fraction schedules, dosages range from 10 to 30 Gy to the tumor margin, with a mean dose in most studies at approximately 15 Gy (207,209,212). As with the initial trials, however, most of these studies do not differentiate between radiosurgery as primary or secondary treatment. A review of published data comparing surgery with stereotactic radiosurgery as the primary treatment modality has found a similar recurrence rate (2.1% vs. 3.1%) between the two modalities, a similar rate of morbidity (8.3% vs. 8.5%), but an increased mortality rate (0% vs. 1.3%) with patients undergoing primary surgical treatment (178). However, in 92% of surgical cases, complete tumor removal was possible, whereas 100% of patients undergoing radiotherapy had residual tumor. Mean follow-up was approximately 4 years in this review, leading the authors to conclude that long-term control rates from stereotactic radiosurgery are still in question (178). In the single retrospective study with adequate followup (mean 13 years) (213), however, the 10-year tumor control rate has been found to be as high as 92%. This study followed 33 patients treated with either conventional (76%) or stereotactic (24%) radiotherapy. Nineteen patients (56%) had jugular paragangliomas, eight (24%) had tympanic tumors, and the remainder had carotid body or retroperitoneal tumors. In half the patients, radiotherapy was the primary modality. In one patient with a jugular paraganglioma, the tumor began to re-grow after 8.5 years; this patient was successfully treated with re-irradiation.

Complications Severe complications—defined as those requiring surgical intervention, hospitalization, hyperbaric oxygen, or death— are uncommon, and the majority of side effects are acute and mild, including mucositis, acute dermatitis, alopecia, otitis externa, serous otitis media, dysgeusia, and xerostomia (209,214). Severe complications reported include osteoradionecrosis of the temporal bone (1.7%), brain necrosis (0.84%), and radiation-induced malignant disease (0.28%) (19). Cranial neuropathies are also a potential risk of radiation, with up to 4% of patients undergoing radiotherapy for paragangliomas of the temporal bone developing facial nerve palsy (197). However, both osteoradionecrosis and cranial neuropathies are associated with doses of above 50 Gy (215). Cummings reported two major complications in 45 patients (4%): temporal bone necrosis after 70 Gy using electron beam and fatal brain necrosis after delivery of 70 Gy in 3 weeks in error (113). It should be noted that a number of authors have reported no complications in the management of CBT and VP with radiation (192,198,214,216). The risk of malignant transformation of the radiated paraganglioma, the development of radiation-induced malignancies, and difficulty with surgical salvage are all possible disadvantages of radiation therapy. The former two concerns, however, are not necessarily supported by evidence. Krych et al. reported on long-term outcomes following radiation for 33 paragangliomas treated over 28 years (25 with conventional radiotherapy and 8 with stereotactic radiosurgery) with a median follow-up of 161 months (range 4–429 mo) (213). No patient in their series developed a radiation-induced malignancy. The authors performed a MedLine search for second primary malignancies following radiation for

552

Goldstein et al.

paragangliomas and found only three reported cases that were possibly related to radiation (213). Lalwani et al. estimated the risk of radiation-induced malignancy to be 0.5–1 in 1000 (217), while others have placed the risk as high as 2 to 3 per 1000 (19,124,195,213). Complications from stereotactic radiosurgery so far have been rare. Isolated reports do exist of intractable vertigo (190,209) and new transient cranial nerve palsies (210), but no reports of permanent cranial nerve palsies or radiationinduced malignancies have been published, although current follow-up is still limited.

Comparison Between Radiation and Surgery Studies comparing the results of surgery and with those of radiation in the management of head and neck paragangliomas typically compare historical results published on each modality from retrospective case series. Selection bias also limits the interpretation of these results. Since we do not have randomized trials or well-designed observational studies directly comparing the two modalities in homogenous populations, one cannot determine if one treatment is superior to others. From the available literature, however, it appears that the rates of control after radiation or surgery are both around 90% (214). Therefore, treatment decisions must be based on the ability to control the tumor with minimal short- and longterm morbidity. Patient and physician preference both affect the choice of treatment. In most cases, the likelihood of cure after subtotal resection and radiation is similar to that after radiation alone; there therefore does not appear to be an advantage of planned combined therapy (193,197,214). Hu et al. performed a combined analysis of five studies which reported on results of surgery or radiation where tumors were staged using the McCabe/Fletcher system. Ninety-four patients were treated with radiation alone, 45 with surgery alone, and 39 with surgery and radiation. The average local control rates were 93%, 78%, and 85% for the three groups respectively. Sixty-five percent of tumors managed with radiation were stage III compared with 13% of tumors managed with surgery alone, and 51% of those treated with a combined approach (112). For recurrent chemodectomas, radiotherapy may be a better treatment modality than surgery. In a review of 29 patients with recurrent jugular and tympanic tumors (218), patients treated with stereotactic, conventional, or intensitymodulated radiotherapy fared better in terms of long-term disease-free survival (100% vs. 62% at 5 years) and posttreatment side effects (0% vs. 47%) than patients who were treated with a re-operation. All recurrent carotid body tumors, on the other hand, were successfully treated surgically.

CONCLUSION In general, carotid body tumors and low vagal paragangliomas can be treated surgically. Management of jugulotympanic tumor is, however, more controversial. Small tympanic tumors can be resected easily and with minimal morbidity, while large tumors require lateral skull base or infratemporal fossa approaches that are associated with morbidity; radiation therapy is often therefore considered. Radiation therapy is also advocated for patients with multiple paragangliomas, in patients with poor performance status or in elderly patients in whom disability from cranial nerve injury would be poorly tolerated. Observation may also be an option in the latter two groups of patients.

ACKNOWLEDGMENT We thank Cheryl Volling for drawing Shamblin classification.

REFERENCES 1. Lack EE, Cubilla AL, Woodruff JM, Farr HW. Paragangliomas of the head and neck region: A clinical study of 69 patients. Cancer. 1977;39:397–409. 2. Lee JH, Barich F, Karnell LH, et al. National Cancer Data Base report on malignant paragangliomas of the head and neck. Cancer. 2002;94:730–737. 3. Heth J. The basic science of glomus jugulare tumors. Neurosurg Focus 2004;17:E2. 4. Glenner GG, Grimley PM. Tumors of the extra-adrenal paraganglion system (including chemoreceptors). In: Atlas of Tumor Pathology. Washington, DC: Armed Forces Institute of Pathology, 1974:1–90. 5. Rao AB, Koeller KK, Adair CF. From the archives of the AFIP. Paragangliomas of the head and neck: radiologic-pathologic correlation. Armed Forces Institute of Pathology. Radiographics. 1999;19:1605–1632. 6. Knight TT, Jr., Gonzalez JA, Rary JM, et al. Current concepts for the surgical management of carotid body tumor. Am J Surg. 2006;191:104–110. 7. Shamblin WR, ReMine WH, Sheps SG, et al. Carotid body tumor (chemodectoma). Clinicopathologic analysis of ninety cases. Am J Surg. 1971;122:732–739. 8. Rosenwasser H. Carotid body tumor of the middle ear and mastoid. Arch Otolaryngol. 1945;41:64–67. 9. Guild SR. A hitherto unrecognized structure, the glomus jugularies in man. Anat Rec. 1941;79:28. 10. Gulya AJ. The glomus tumor and its biology. Laryngoscope. 1993;103:7–15. 11. Martin TP. What we call them: the nomenclature of head and neck paragangliomas. Clin Otolaryngol. 2006;31:185– 186. 12. Hu K, Persky MS. Multidisciplinary management of paragangliomas of the head and neck, Part 1. Oncology. (Williston Park) 2003;17:983–993. 13. Khan Q, Heath D, Smith P. Anatomical variations in human carotid bodies. J Clin Pathol. 1988;41:1196–1199. 14. Smith P, Jago R, Heath D. Anatomical variation and quantitative histology of the normal and enlarged carotid body. J Pathol. 1982;137:287–304. 15. Noujaim SE, Pattekar MA, Cacciarelli A, et al. Paraganglioma of the temporal bone: role of magnetic resonance imaging versus computed tomography. Top Magn Reson Imaging. 2000;11:108– 122. 16. Myssiorek D, Halaas Y, Silver C. Laryngeal and sinonasal paragangliomas. Otolaryngol Clin North Am. 2001;34:971–982, vii. 17. Wasserman PG, Savargaonkar P. Paragangliomas: Classification, pathology, and differential diagnosis. Otolaryngol Clin North Am. 2001;34:845–862, v–vi. 18. Myssiorek D. Head and neck paragangliomas: An overview. Otolaryngol Clin North Am. 2001;34:829–836, v. 19. Gujrathi CS, Donald PJ. Current trends in the diagnosis and management of head and neck paragangliomas. Curr Opin Otolaryngol Head Neck Surg. 2005;13:339–342. 20. McCaffrey TV, Myssiorek D, Marrinan M. Head and neck paragangliomas: physiology and biochemistry. Otolaryngol Clin North Am. 2001;34:837–844, v. 21. van der Mey AG, Jansen JC, van Baalen JM. Management of carotid body tumors. Otolaryngol Clin North Am. 2001;34:907– 924, vi. 22. Wenig B. Atlas of Head and Neck Pathology. Philadelphia: W.B. Saunders Company, 1993. 23. Kimura N, Chetty R, Capella C, et al. Extra-adrenal paraganglioma: Carotid body, jugulotympanic, vagal, laryngeal, aorticopulmonary. In: DeLellis RA LR, Heitz PU, Eng C, eds. World Health Organiztion: Pathology and Genetics: Tumors of Endocrine Organs. Lyon: IARC Press, 2004.

Chapter 38: Paragangliomas of the Head and Neck 24. Kliewer KE, Wen DR, Cancilla PA, et al. Paragangliomas: assessment of prognosis by histologic, immunohistochemical, and ultrastructural techniques. Hum Pathol. 1989;20:29–39. 25. Arts HA, Fagan PA. Vagal body tumors. Otolaryngol Head Neck Surg. 1991;105:78–85. 26. Michaelson PG, Fowler CB, Brennan J. Tracheal paraganglioma presenting with acute airway obstruction. Otolaryngol Head Neck Surg. 2005;132:661–662. 27. Del Gaudio JM, Muller S. Diagnosis and treatment of supraglottic laryngeal paraganglioma: report of a case. Head Neck. 2004;26:94–98. 28. Peterson KL, Fu YS, Calcaterra T. Subglottic paraganglioma. Head Neck. 1997;19:54–56. 29. Sanders KW, Abreo F, Rivera E, et al. A diagnostic and therapeutic approach to paragangliomas of the larynx. Arch Otolaryngol Head Neck Surg. 2001;127:565–569. 30. Ketabchi S, Massi D, Santoro R, et al. Paraganglioma of the nasal cavity: A case report. Eur Arch Otorhinolaryngol. 2003;260:336– 340. 31. Lecanu JB, Arkwright S, Halimi PH, et al. Multifocal malignant paraganglioma of the paranasal sinuses: A case report. Otolaryngol Head Neck Surg. 2002;126:445–447. 32. Mouadeb DA, Chandra RK, Kennedy DW, et al. Sinonasal paraganglioma: Endoscopic resection with a 4-year follow-up. Head Neck. 2003;25:1077–1081. 33. Sharma MCMC, Epari SS, Gaikwad SS, et al. Orbital paraganglioma: Report of a rare case. Can J Ophthalmol. 2005;40:640– 644. 34. Moyer JS, Bradford CR. Sympathetic paraganglioma as an unusual cause of Horner’s syndrome. Head Neck. 2001;23:338–342. 35. Kronz JD, Argani P, Udelsman R, et al. Paraganglioma of the thyroid: Two cases that clarify and expand the clinical spectrum. Head Neck. 2000;22:621–625. 36. Maxwell JG, Jones SW, Wilson E, et al. Carotid body tumor excisions: adverse outcomes of adding carotid endarterectomy. J Am Coll Surg. 2004;198:36–41. 37. van den Berg R, van Gils AP, Wasser MN. Imaging of head and neck paragangliomas with three-dimensional time-of-flight MR angiography. Am J Roentgenol. 1999;172:1667–1673. 38. Persky MS, Hu KS, Berenstein A. Paragangliomas of the head and neck In: Harrison LB, Sessions RB, Hong WK, eds. Head and Neck Cancer: A Multidisciplinary Approach. Philadelphia: Lippincott Williams & Wilkins, 2004:678–713. 39. Leonetti JP, Donzelli JJ, Littooy FN, et al. Perioperative strategies in the management of carotid body tumors. Otolaryngol Head Neck Surg. 1997;117:111–115. 40. Larson TC 3rd, Reese DF, Baker HL Jr., et al. Glomus tympanicum chemodectomas: Radiographic and clinical characteristics. Radiology. 1987;163:801–806. 41. Hesselink JR, Davis KR, Taveras JM. Selective arteriography of glomus tympanicum and jugulare tumors: Techniques, normal and pathologic arterial anatomy. Am J Neuroradiol. 1981;2:289– 297. 42. Mafee MF, Kumar A, Yannias DA, et al. Computed tomography of the middle ear in the evaluation of cholesteatomas and other soft-tissue masses: Comparison with pluridirectional tomography. Radiology. 1983;148:465–472. 43. Arriaga M. Paraganglioma (Glomus tympanicum). In: Jackler RK, Driscoll CLW, eds. Tumors of the Ear and Temporal Bone. Philadelphia: Lippincott Williams & Wilkins, 2000:112–127. 44. Friedman R, Brackmann DE. Jugulotympanic paragangliomas. In: Jackler R, Driscoll CLW, eds. Tumors of the Ear and Temporal Bone. Philadelphia: Lippincott Williams & Wilkins, 2000:343– 360. 45. Moe KS, Li D, Linder TE, et al. An update on the surgical treatment of temporal bone paraganglioma. Skull Base Surg. 1999;9:185–194. 46. Rigby P, Jackler RK. Clinicopathologic presentation and diagnostic imaging of jugular foramen tumors. Operat Tech Otolaryngol Head Neck Surg. 1996;7:99–105. 47. Miller RB, Boon MS, Atkins JP, et al. Vagal paraganglioma: The Jefferson experience. Otolaryngol Head Neck Surg. 2000;122:482–487.

553

48. Hinojar AG, Prieto JR, Munoz E, et al. Relapsing paraganglioma of the inferior laryngeal paraganglion: Case report and review of the literature. Head Neck. 2002;24:95–102. 49. Kuhn JA, Aronoff BL. Nasal and nasopharyngeal paraganglioma. J Surg Oncol 1989;40:38–45. 50. Lack E. Tumors of the adrenal gland and extra-adrenal paraganglioma. Atlas of Tumor Pathology, series 3, fasc 19. Washington, DC: Armed Forces Institute of Pathology, 1997. 51. Pellitteri PK, Rinaldo A, Myssiorek D, et al. Paragangliomas of the head and neck. Oral Oncol. 2004;40:563–575. 52. Astrom KK, Cohen JEJE, Willett-Brozick JEJE, et al. Altitude is a phenotypic modifier in hereditary paraganglioma type 1: Evidence for an oxygen-sensing defect. Human Genet. 2003;113:228–237. 53. Nissenblatt MJ. Cyanotic heart disease: “low altitude” risk for carotid body tumor? Johns Hopkins Med J. 1978;142:18–22. 54. Rodriguez-Cuevas HH, Lau II, Rodriguez HHP. High-altitude paragangliomas diagnostic and therapeutic considerations. Cancer. 1986;57:672–676. ´ 55. Rodr´ıguez-Cuevas SS, Lopez-Garza JJ, Labastida-Almendaro SS. Carotid body tumors in inhabitants of altitudes higher than 2000 meters above sea level. Head Neck. 1998;20:374–378. 56. Saldana MMJ, Salem LLE, Travezan RR. High altitude hypoxia and chemodectomas. Human Pathol. 1973;4:251–263. 57. Dundee P, Clancy B, Wagstaff S, et al. Paraganglioma: The role of genetic counselling and radiological screening. J Clin Neurosci. 2005;12:464–466. 58. Petropoulos AE, Luetje CM, Camarata PJ, et al. Genetic analysis in the diagnosis of familial paragangliomas. Laryngoscope 2000;110:1225–1229. 59. Drovdlic CM, Myers EN, Peters JA, et al. Proportion of heritable paraganglioma cases and associated clinical characteristics. Laryngoscope. 2001;111:1822–1827. 60. Grufferman S, Gillman MW, Pasternak LR, et al. Familial carotid body tumors: Case report and epidemiologic review. Cancer. 1980;46:2116–2122. 61. Parry DM, Li FP, Strong LC, et al. Carotid body tumors in humans: Genetics and epidemiology. J Natl Cancer Inst. 1982;68:573–578. 62. van der Mey AG, Maaswinkel-Mooy PD, Cornelisse CJ, et al. Genomic imprinting in hereditary glomus tumours: Evidence for new genetic theory. Lancet. 1989;2:1291–1294. 63. Badenhop RF, Jansen JC, Fagan PA, et al. The prevalence of SDHB, SDHC, and SDHD mutations in patients with head and neck paraganglioma and association of mutations with clinical features. J Med Genet. 2004;41:e99. 64. Spector GJ, Ciralsky R, Maisel RH, et al. Multiple glomus tumors in the head and neck. Laryngoscope 1975;85:1066–1075. 65. Hodge KM, Byers RM, Peters LJ. Paragangliomas of the head and neck. Arch Otolaryngol Head Neck Surg. 1988;114:872–877. 66. Netterville JL, Jackson CG, Miller FR, et al. Vagal paraganglioma: A review of 46 patients treated during a 20-year period. Arch Otolaryngol Head Neck Surg. 1998;124:1133–1140. 67. Boedeker CC, Ridder GJ, Schipper J. Paragangliomas of the head and neck: Diagnosis and treatment. Fam Cancer. 2005;4:55– 59. 68. Netterville JL, Reilly KM, Robertson D, et al. Carotid body tumors: A review of 30 patients with 46 tumors. Laryngoscope. 1995;105:115–126. 69. Carney JA, Sheps SG, Go VL, et al. The triad of gastric leiomyosarcoma, functioning extra-adrenal paraganglioma and pulmonary chondroma. N Engl J Med. 1977;296:1517–1518. 70. Hull MT, Roth LM, Glover JL, et al. Metastatic carotid body paraganglioma in von Hippel-Lindau disease. An electron microscopic study. Arch Pathol Lab Med. 1982;106:235–239. 71. Benn DE, Gimenez-Roqueplo AP, Reilly JR, et al. Clinical presentation and penetrance of pheochromocytoma/paraganglioma syndromes. J Clin Endocrinol Metab. 2006;91:827–836. 72. Neumann HP, Bausch B, McWhinney SR, et al. Germ-line mutations in nonsyndromic pheochromocytoma. N Engl J Med. 2002;346:1459–1466. 73. Baysal BE. Genetics of familial paragangliomas: Past, present, and future. Otolaryngol Clin North Am. 2001;34:863–879, vi.

554

Goldstein et al.

74. Maier W, Marangos N, Laszig R. Paraganglioma as a systemic syndrome: Pitfalls and strategies. J Laryngol Otol. 1999;113:978– 982. 75. Schimke RN, Collins DL, Rothberg PG. Functioning carotid paraganglioma in the von Hippel-Lindau syndrome. Am J Med Genet. 1998;80:533–534. 76. van Gils AP, van der Mey AG, Hoogma RP, et al. MRI screening of kindred at risk of developing paragangliomas: Support for genomic imprinting in hereditary glomus tumours. Br J Cancer. 1992;65:903–907. 77. Pawlu C, Bausch B, Neumann HP. Mutations of the SDHB and SDHD genes. Fam Cancer. 2005;4:49–54. 78. Baysal BE. Hereditary paraganglioma targets diverse paraganglia. J Med Genet. 2002;39:617–622. 79. Baysal BE, Ferrell RE, Willett-Brozick JE, et al. Mutations in SDHD, a mitochondrial complex II gene, in hereditary paraganglioma. Science. 2000;287:848–851. 80. Mariman EC, van Beersum SE, Cremers CW, et al. Fine mapping of a putatively imprinted gene for familial non-chromaffin paragangliomas to chromosome 11q13.1: Evidence for genetic heterogeneity. Hum Genet. 1995;95:56–62. 81. Mhatre AN, Li Y, Feng L, et al. SDHB, SDHC, and SDHD mutation screen in sporadic and familial head and neck paragangliomas. Clin Genet. 2004;66:461–466. 82. Baysal BE, Willett-Brozick JE, Lawrence EC, et al. Prevalence of SDHB, SDHC, and SDHD germline mutations in clinic patients with head and neck paragangliomas. J Med Genet. 2002;39:178– 183. 83. Dluhy RG. Pheochromocytoma—death of an axiom. N Engl J Med. 2002;346:1486–1488. 84. Schiavi F, Boedeker CC, Bausch B, et al. Predictors and prevalence of paraganglioma syndrome associated with mutations of the SDHC gene. JAMA. 2005;294:2057–2063. 85. Neumann HP, Pawlu C, Peczkowska M, et al. Distinct clinical features of paraganglioma syndromes associated with SDHB and SDHD gene mutations. JAMA. 2004;292:943–951. 86. Gimenez-Roqueplo AP, Favier J, Rustin P, et al. Mutations in the SDHB gene are associated with extra-adrenal and/or malignant phaeochromocytomas. Cancer Res. 2003;63:5615– 5621. 87. Young AL, Baysal BE, Deb A, et al. Familial malignant catecholamine-secreting paraganglioma with prolonged survival associated with mutation in the succinate dehydrogenase B gene. J Clin Endocrinol Metab. 2002;87:4101–4105. 88. Gimenez-Roqueplo AP, Favier J, Rustin P, et al. Functional consequences of a SDHB gene mutation in an apparently sporadic pheochromocytoma. J Clin Endocrinol Metab. 2002;87:4771– 4774. 89. Bikhazi PH, Roeder E, Attaie A, et al. Familial paragangliomas: The emerging impact of molecular genetics on evaluation and management. Am J Otol. 1999;20:639–643. 90. Renard L, Godfraind C, Boon LM, et al. A novel mutation in the SDHD gene in a family with inherited paragangliomas— implications of genetic diagnosis for follow up and treatment. Head Neck. 2003;25:146–151. 91. Rinaldo A, Myssiorek D, Devaney KO, et al. Which paragangliomas of the head and neck have a higher rate of malignancy? Oral Oncol. 2004;40:458–460. 92. Walsh RM, Leen EJ, Gleeson MJ, et al. Malignant vagal paraganglioma. J Laryngol Otol. 1997;111:83–88. 93. Martin CE, Rosenfeld L, McSwain B. Carotid body tumors: A 16-year follow-up of seven malignant cases. South Med J. 1973;66:1236–1243. 94. Zbaren P, Lehmann W. Carotid body paraganglioma with metastases. Laryngoscope. 1985;95:450–454. 95. Sykes JM, Ossoff RH. Paragangliomas of the head and neck. Otolaryngol Clin North Am. 1986;19:755–767. 96. Manolidis S, Shohet JA, Jackson CG, et al. Malignant glomus tumors. Laryngoscope. 1999;109:30–34. 97. Schwaber MK, Glasscock ME, Nissen AJ, et al. Diagnosis and management of catecholamine secreting glomus tumors. Laryngoscope 1984;94:1008–1015.

98. Erickson D, Kudva YC, Ebersold MJ, et al. Benign paragangliomas: Clinical presentation and treatment outcomes in 236 patients. J Clin Endocrinol Metab. 2001;86:5210–5216. 99. Jensen NF. Glomus tumors of the head and neck: Anesthetic considerations. Anesth Analg. 1994;78:112–119. 100. van der Mey AG, Frijns JH, Cornelisse CJ, et al. Does intervention improve the natural course of glomus tumors? A series of 108 patients seen in a 32-year period. Ann Otol Rhinol Laryngol. 1992;101:635–642. 101. Jansen JC, van den Berg R, Kuiper A, et al. Estimation of growth rate in patients with head and neck paragangliomas influences the treatment proposal. Cancer. 2000;88:2811–2816. 102. Bradshaw JW, Jansen JC. Management of vagal paraganglioma: Is operative resection really the best option? Surgery. 2005;137:225–228. 103. Jackson CG, Glasscock ME, 3rd, Harris PF. Glomus tumors. Diagnosis, classification, and management of large lesions. Arch Otolaryngol. 1982;108:401–410. 104. Hinerman RW, Mendenhall WM, Amdur RJ, et al. Definitive radiotherapy in the management of chemodectomas arising in the temporal bone, carotid body, and glomus vagale. Head Neck. 2001;23:363–371. 105. Monabati A, Hodjati H, Kumar PV. Cytologic findings in carotid body tumors. Acta Cytol. 2002;46:1101–1104. 106. Patetsios P, Gable DR, Garrett WV, et al. Management of carotid body paragangliomas and review of a 30-year experience. Ann Vasc Surg. 2002;16:331–338. 107. Luna-Ortiz K, Rascon-Ortiz M, Villavicencio-Valencia V, et al. Carotid body tumors: Review of a 20-year experience. Oral Oncol. 2005;41:56–61. 108. Plukker JT, Brongers EP, Vermey A, et al. Outcome of surgical treatment for carotid body paraganglioma. Br J Surg. 2001;88:1382–1386. 109. Nora JD, Hallett JW Jr., O’Brien PC, et al. Surgical resection of carotid body tumors: Long-term survival, recurrence, and metastasis. Mayo Clin Proc. 1988;63:348–352. 110. Najibi S, Terramani TT, Brinkman W, et al. Carotid body tumors. J Am Coll Surg. 2002;194:538–539. 111. Holden PK, Linthicum FH, Jr. Glomus jugulare tumor. Otol Neurotol. 2005;26:312–313. 112. Hu K, Persky MS. The multidisciplinary management of paragangliomas of the head and neck, Part 2. Oncology (Williston Park). 2003;17:1143–1153; discussion 1154, 1158, 1161. 113. Cummings BJ, Beale FA, Garrett PG, et al. The treatment of glomus tumors in the temporal bone by megavoltage radiation. Cancer. 1984;53:2635–2640. 114. Jackson CG. Glomus tympanicum and glomus jugulare tumors. Otolaryngol Clin North Am. 2001;34:941–970, vii. 115. Pryzant RM, Chou JL, Easley JD. Twenty year experience with radiation therapy for temporal bone chemodectomas. Int J Radiat Oncol Biol Phys. 1989;17:1303–1307. 116. Sanna MM, Jain YY, De Donato GG, et al. Management of jugular paragangliomas: The Gruppo Otologico experience. Otology Neurotol. 2004;25:797–804. 117. Forest JA, 3rd, Jackson CG, McGrew BM. Long-term control of surgically treated glomus tympanicum tumors. Otol Neurotol. 2001;22:232–236. 118. Persky MS, Setton A, Niimi Y, et al. Combined endovascular and surgical treatment of head and neck paragangliomas—a team approach. Head Neck. 2002;24:423–431. 119. Jackson CG, Harris PF, Glasscock ME, 3rd, et al. Diagnosis and management of paragangliomas of the skull base. Am J Surg. 1990;159:389–393. 120. Gardner G, Cocke EW, Jr., Robertson JH, et al. Skull base surgery for glomus jugulare tumors. Am J Otol. 1985;Suppl:126–134. 121. Ogura JH, Spector GJ, Gado M. Glomus jugulare and vagale. Ann Otol Rhinol Laryngol. 1978;87:622–629. 122. Spector GJ, Gado M, Ciralsky R, et al. Neurologic implications of glomus tumors in the head and neck. Laryngoscope. 1975;85:1387–1395. 123. Pareschi R, Righini S, Destito D, et al. Surgery of glomus jugulare tumors. Skull Base. 2003;13:149–157.

Chapter 38: Paragangliomas of the Head and Neck 124. Alford BR, Guilford FR. A comprehensive study of tumors of the glomus jugulare. Laryngoscope. 1962;72:765–805. 125. Makek M, Franklin DJ, Zhao JC, et al. Neural infiltration of glomus temporale tumors. Am J Otol. 1990;11:1–5. 126. Patel SJ, Sekhar LN, Cass SP, et al. Combined approaches for resection of extensive glomus jugulare tumors. A review of 12 cases. J Neurosurg. 1994;80:1026–1038. 127. Pensak ML, Jackler RK. Removal of jugular foramen tumors: The fallopian bridge technique. Otolaryngol Head Neck Surg. 1997;117:586–591. 128. Woods CI, Strasnick B, Jackson CG. Surgery for glomus tumors: The Otology Group experience. Laryngoscope. 1993;103:65– 70. 129. Lustig LLR, Jackler RRK. The variable relationship between the lower cranial nerves and jugular foramen tumors: Implications for neural preservation. Am J Otol. 1996;17:658–668. 130. Sniezek JC, Netterville JL, Sabri AN. Vagal paragangliomas. Otolaryngol Clin North Am. 2001;34:925–939, vi. 131. Lawson W. Glomus bodies and tumors. N Y State J Med. 1980;80:1567–1575. 132. Myssiorek D, Rinaldo A, Barnes L, et al. Laryngeal paraganglioma: An updated critical review. Acta Otolaryngol. 2004;124:995–999. 133. Stoeckli SJ, Schuknecht B, Alkadhi H, et al. Evaluation of paragangliomas presenting as a cervical mass on color-coded Doppler sonography. Laryngoscope. 2002;112:143–146. 134. Lustrin ES, Palestro C, Vaheesan K. Radiographic evaluation and assessment of paragangliomas. Otolaryngol Clin North Am. 2001;34:881–906, vi. 135. Phelps PD, Stansbie JM. Glomus jugulare or tympanicum? The role of CT and MR imaging with gadolinium DTPA. J Laryngol Otol. 1988;102:766–776. 136. Olsen WL, Dillon WP, Kelly WM, et al. MR imaging of paragangliomas. Am J Roentgenol. 1987;148:201–204. 137. Phelps PD, Cheesman AD. Imaging jugulotympanic glomus tumors. Arch Otolaryngol Head Neck Surg. 1990;116:940–945. 138. Vogl TJ, Juergens M, Balzer JO, et al. Glomus tumors of the skull base: Combined use of MR angiography and spin-echo imaging. Radiology. 1994;192:103–110. 139. Som PM, Braun IF, Shapiro MD, et al. Tumors of the parapharyngeal space and upper neck: MR imaging characteristics. Radiology. 1987;164:823–829. 140. van den Berg R, Wasser MN, van Gils AP, et al. Vascularization of head and neck paragangliomas: Comparison of three MR angiographic techniques with digital subtraction angiography. Am J Neuroradiol. 2000;21:162–170. 141. van den Berg R, Verbist BM, Mertens BJ, et al. Head and neck paragangliomas: Improved tumor detection using contrastenhanced 3D time-of-flight MR angiography as compared with fat-suppressed MR imaging techniques. Am J Neuroradiol. 2004;25:863–870. 142. van den Berg R, Schepers A, de Bruine FT, et al. The value of MR angiography techniques in the detection of head and neck paragangliomas. Eur J Radiol. 2004;52:240–245. 143. Wax MK, Briant TD. Carotid body tumors: A review. J Otolaryngol. 1992;21:277–285. 144. van den Berg R. Imaging and management of head and neck paragangliomas. Eur Radiol. 2005;15:1310–1318. 145. Duet M, Sauvaget E, Petelle B, et al. Clinical impact of somatostatin receptor scintigraphy in the management of paragangliomas of the head and neck. J Nucl Med. 2003;44:1767–1774. 146. Bustillo A, Telischi F, Weed D, et al. Octreotide scintigraphy in the head and neck. Laryngoscope. 2004;114:434–440. 147. Myssiorek D, Palestro CJ. 111 Indium pentetreotide scan detection of familial paragangliomas. Laryngoscope. 1998;108:228– 231. 148. Telischi FF, Bustillo A, Whiteman ML, et al. Octreotide scintigraphy for the detection of paragangliomas. Otolaryngol Head Neck Surg. 2000;122:358–362. 149. Bustillo A, Telischi FF. Octreotide scintigraphy in the detection of recurrent paragangliomas. Otolaryngol Head Neck Surg. 2004;130:479–482.

555

150. Litle VR, Reilly LM, Ramos TK. Preoperative embolization of carotid body tumors: When is it appropriate? Ann Vasc Surg. 1996;10:464–468. 151. Mitchell RO, Richardson JD, Lambert GE. Characteristics, surgical management, and outcome in 17 carotid body tumors. Am Surg. 1996;62:1034–1037. 152. Witterick IJ, Gullane PJ, Keller MA. Postoperative carotid body tumor evaluation: Analysis using MR angiography. Laryngoscope. 1995;105:764–767. 153. Shibuya Y, Umeda M, Yoshikawa T, et al. Carotid body tumour. Oral Oncol. 2002;38:313–317. 154. Smith JJ, Passman MA, Dattilo JB, et al. Carotid body tumor resection: Does the need for vascular reconstruction worsen outcome? Ann Vasc Surg. 2006;20:435–439. 155. Dickinson PH, Griffin SM, Guy AJ, et al. Carotid body tumour: 30 years experience. Br J Surg. 1986;73:14–16. 156. Hallett JW, Jr., Nora JD, Hollier LH, et al. Trends in neurovascular complications of surgical management for carotid body and cervical paragangliomas: A fifty-year experience with 153 tumors. J Vasc Surg. 1988;7:284–291. 157. de Vries EJ, Sekhar LN, Horton JA, et al. A new method to predict safe resection of the internal carotid artery. Laryngoscope. 1990;100:85–88. 158. Pantanowitz D, Davidge-Pitts K, Demetriades D. The significance of the carotid bifurcation angle in carotid body tumours. S Afr Med J. 1991;80:318–321. 159. Anand VK, Alemar GO, Sanders TS. Management of the internal carotid artery during carotid body tumor surgery. Laryngoscope. 1995;105:231–235. 160. Thabet MH, Kotob H. Cervical paragangliomas: Diagnosis, management and complications. J Laryngol Otol. 2001;115:467– 474. 161. Rosen IB, Palmer JA, Goldberg M, et al. Vascular problems associated with carotid body tumors. Am J Surg. 1981;142:459– 463. 162. Williams MD, Phillips MJ, Nelson WR, et al. Carotid body tumor. Arch Surg. 1992;127:963–967; discussion 967–968. 163. Maturo S, Brennan J. Baroreflex failure: A rare complication of carotid paraganglioma surgery. Laryngoscope. 2006;116:829– 830. 164. Timmers HJ, Karemaker JM, Wieling W, et al. Baroreflex control of muscle sympathetic nerve activity after carotid body tumor resection. Hypertension. 2003;42:143–149. 165. De Toma G, Nicolanti V, Plocco M, et al. Baroreflex failure syndrome after bilateral excision of carotid body tumors: An underestimated problem. J Vasc Surg. 2000;31:806–810. 166. Robertson D, Beck C, Gary T, et al. Classification of autonomic disorders. Int Angiol. 1993;12:93–102. 167. Smit AA, Timmers HJ, Wieling W, et al. Long-term effects of carotid sinus denervation on arterial blood pressure in humans. Circulation. 2002;105:1329–1335. 168. Robertson D, Hollister AS, Biaggioni I, et al. The diagnosis and treatment of baroreflex failure. N Engl J Med. 1993;329:1449– 1455. 169. Sharabi YY, Dendi RR, Holmes CC, et al. Baroreflex failure as a late sequela of neck irradiation. Hypertension. 2003;42:110– 116. 170. Timmers HHJ, Karemaker JJM, Lenders JJW, et al. Baroreflex failure following radiation therapy for nasopharyngeal carcinoma. Clin Auton Res. 1999;9:317–324. 171. Urquhart AC, Johnson JT, Myers EN, et al. Glomus vagale: Paraganglioma of the vagus nerve. Laryngoscope. 1994;104:440–445. 172. Biller HF, Lawson W, Som P, et al. Glomus vagale tumors. Ann Otol Rhinol Laryngol. 1989;98:21–26. 173. Durvasula VS, De R, Baguley DM, Moffat DA. Laser excision of glomus tympanicum tumours: Long-term results. Eur Arch Oto-rhino-laryngol. 2005;262:325–327. 174. Molony NC, Salto-Tellez M, Grant WE. KTP laser assisted excision of glomus tympanicum. J Laryngol Otol. 1998;112:956–958. 175. Robinson PJ, Grant HR, Bown SG. NdYAG laser treatment of a glomus tympanicum tumour. J Laryngol Otol. 1993;107:236– 237.

556

Goldstein et al.

176. O’Leary MJ, Shelton C, Giddings NA, et al. Glomus tympanicum tumors: A clinical perspective. Laryngoscope. 1991;101:1038–1043. 177. Jackson CG, Welling DB, Chironis P, et al. Glomus tympanicum tumors: Contemporary concepts in conservation surgery. Laryngoscope. 1989;99:875–884. 178. Gottfried ON, Liu JK, Couldwell WT. Comparison of radiosurgery and conventional surgery for the treatment of glomus jugulare tumors. Neurosurg Focus. 2004;17:E4. 179. Jackson CG, McGrew BM, Forest JA, et al. Lateral skull base surgery for glomus tumors: Long-term control. Otol Neurotol. 2001;22:377–382. 180. Murphy TP, Brackmann DE. Effects of preoperative embolization on glomus jugulare tumors. Laryngoscope. 1989;99:1244– 1247. 181. Briner HR, Linder TE, Pauw B, et al. Long-term results of surgery for temporal bone paragangliomas. Laryngoscope. 1999;109:577–583. 182. Manolidis S, Jackson CG, Von Doersten PG, et al. Lateral skull base surgery: The otology group experience. Skull Base Surg. 1997;7:129–137. 183. Green JD, Jr., Brackmann DE, Nguyen CD, et al. Surgical management of previously untreated glomus jugulare tumors. Laryngoscope. 1994;104:917–921. 184. Whitfield PC, Grey P, Hardy DG, et al. The surgical management of patients with glomus tumours of the skull base. Br J Neurosurg. 1996;10:343–350. 185. Tran BA, Huy P, Chao PZ, et al. Long-term oncological results in 47 cases of jugular paraganglioma surgery with special emphasis on the facial nerve issue. J Laryngol Otol. 2001;115:981–987. 186. Selesnick SH, Abraham MT, Carew JF. Rerouting of the intratemporal facial nerve: An analysis of the literature. Am J Otol. 1996;17:793–805; discussion 806–809. 187. Cece JA, Lawson W, Biller HF, et al. Complications in the management of large glomus jugulare tumors. Laryngoscope. 1987;97:152–157. 188. Spector GJ, Compagno J, Perez CA, et al. Glomus jugulare tumors: Effects of radiotherapy. Cancer. 1975;35:1316–1321. 189. Suit HD, Gallager HS. Intact tumor cells in irradiated tissue. Arch Pathol. 1964;78:648–651. 190. Jordan JA, Roland PS, McManus C, et al. Stereotastic radiosurgery for glomus jugulare tumors. Laryngoscope. 2000;110:35–38. 191. Kim JA, Elkon D, Lim ML, et al. Optimum dose of radiotherapy for chemodectomas of the middle ear. Int J Radiat Oncol Biol Phys. 1980;6:815–819. 192. Evenson LJ, Mendenhall WM, Parsons JT, et al. Radiotherapy in the management of chemodectomas of the carotid body and glomus vagale. Head Neck. 1998;20:609–613. 193. Mendenhall WM, Parsons JT, Stringer SP, et al. Radiotherapy in the management of temporal bone chemodectoma. Skull Base Surg. 1995;5:83–91. 194. Schild SE, Foote RL, Buskirk SJ, et al. Results of radiotherapy for chemodectomas. Mayo Clin Proc. 1992;67:537–540. 195. Springate SC, Weichselbaum RR. Radiation or surgery for chemodectoma of the temporal bone: A review of local control and complications. Head Neck. 1990;12:303–307. 196. Lybeert ML, van Andel JG, Eijkenboom WM, et al. Radiotherapy of paragangliomas. Clin Otolaryngol Allied Sci. 1984;9:105–109. 197. Powell S, Peters N, Harmer C. Chemodectoma of the head and neck: Results of treatment in 84 patients. Int J Radiat Oncol Biol Phys. 1992;22:919–924.

198. Valdagni R, Amichetti M. Radiation therapy of carotid body tumors. Am J Clin Oncol. 1990;13:45–48. 199. Mendenhall WM, Million RR, Parsons JT, et al. Chemodectoma of the carotid body and ganglion nodosum treated with radiation therapy. Int J Radiat Oncol Biol Phys. 1986;12:2175– 2178. 200. Mitchell DC, Clyne CA. Chemodectomas of the neck: The response to radiotherapy. Br J Surg. 1985;72:903–905. 201. Pemberton LS, Swindell R, Sykes AJ. Radical radiotherapy alone for glomus jugulare and tympanicum tumours. Oncol Rep. 2005;14:1631–1633. 202. Cole JM. Glomus jugulare tumor. Laryngoscope. 1977;87:1244– 1258. 203. Simko TG, Griffin TW, Gerdes AJ, et al. The role of radiation therapy in the treatment of glomus jugulare tumors. Cancer. 1978;42:104–106. 204. Cole JM, Beiler D. Long-term results of treatment for glomus jugulare and glomus vagale tumors with radiotherapy. Laryngoscope. 1994;104:1461–1465. 205. Foote RL, Coffey RJ, Gorman DA, et al. Stereotactic radiosurgery for glomus jugulare tumors: A preliminary report. Int J Radiat Oncol Biol Phys. 1997;38:491–495. 206. Liscak R, Vladyka V, Simonova G, et al. Leksell gamma knife radiosurgery of the tumor glomus jugulare and tympanicum. Stereotact Funct Neurosurg. 1998;70(Suppl 1):152–160. 207. Liscak R, Vladyka V, Wowra B, et al. Gamma knife radiosurgery of the glomus jugulare tumour—early multicentre experience. Acta Neurochir (Wien). 1999;141:1141–1146. 208. Eustacchio S, Trummer M, Unger F, et al. The role of Gamma knife radiosurgery in the management of glomus jugular tumours. Acta Neurochir Suppl. 2002;84:91–97. 209. Foote RL, Pollock BE, Gorman DA, et al. Glomus jugulare tumor: Tumor control and complications after stereotactic radiosurgery. Head Neck. 2002;24:332–338; discussion 338–339. 210. Lim M, Gibbs IC, Adler JR, et al. The efficacy of linear accelerator stereotactic radiosurgery in treating glomus jugulare tumors. Technol Cancer Res Treat. 2003;2(3):261–265. 211. Maarouf M, Voges J, Landwehr P, et al. Stereotactic linear accelerater-based radiosurgery for the treatment of patients with glomus jugulare tumors. Cancer. 2003;97:1093–1098. 212. Sheehan J, Kondziolka D, Flickinger J, et al. Gamma knife surgery for glomus jugulare tumors: An intermediate report on efficacy and safety. J Neurosurg. 2005;102(Suppl):241–246. 213. Krych AJ, Foote RL, Brown PD, et al. Long-term results of irradiation for paraganglioma. Int J Radiat Oncol Biol Phys. 2006;65:1063–1066. 214. Mendenhall WM, Amdur RJ, Hinerman RW, et al. Radiotherapy and radiosurgery for skull base tumors. Otolaryngol Clin North Am. 2001;34:1065–1077, viii. 215. Konefal JB, Pilepich MV, Spector GJ, et al. Radiation therapy in the treatment of chemodectomas. Laryngoscope. 1987;97:1331– 1335. 216. Verniers DA, Keus RB, Schouwenburg PF, et al. Radiation therapy, an important mode of treatment for head and neck chemodectomas. Eur J Cancer. 1992;28A:1028–1033. 217. Lalwani AK, Jackler RK, Gutin PH. Lethal fibrosarcoma complicating radiation therapy for benign glomus jugulare tumor. Am J Otol. 1993;14:398–402. 218. Elshaikh MA, Mahmoud-Ahmed AS, Kinney SE, et al. Recurrent head-and-neck chemodectomas: A comparison of surgical and radiotherapeutic results. Int J Radiat Oncol Biol Phys. 2002;52:953–956.

39 Pituitary Adenomas Mark Hornyak and William T. Couldwell

hormone. Although tumor cells may stain positively for a particular hormone on immunohistochemical analysis, they may not secrete functional hormone. Thus, a somatotrophic tumor may not cause acromegaly. Tumors that do not cause a clinical syndrome are called either silent or nonfunctional adenomas (3). Tumors containing cells that do not stain positively for any adenohypophyseal hormones and do not produce any clinical or biochemical evidence of secretion are termed nullcell tumors. The distribution of endocrinological tumor types can be found in Table 1. Large tumors can grow out of the sella turcica and cause pressure on delicate nervous structures, most commonly the optic chiasm, causing the classically described bitemporal hemianopsia. Microadenomas are defined as measuring less than 1 cm in their greatest dimension, whereas macroadenomas are greater than or equal to 1 cm. Some authors have augmented the classification with the term “mesoadenoma (4),” which has been used to describe adenomas measuring between 1 cm and 2 cm, reserving the term “macroadenoma” for even larger tumors. Most patients who are evaluated by a neurosurgeon harbor tumors larger than 1 cm, but the size of the tumor is often dependent on the endocrinological classification (Table 2). Microscopically, pituitary adenomas appear as monotonous sheets of cells with rounded nuclei and indistinct cytoplasmic borders (5). They are usually without conspicuous architectural features, but the monotony is often interrupted by delicate septae of connective tissue and by small perivascular spaces. Typically, the rounded nuclei contain delicate chromatin and small, peppery nucleoli (Fig. 1). Multinucleation and pleomorphism are not uncommon, but mitotic figures are unusual and frank anaplasia is rare (5). Tumors with significant anaplasia are often called atypical adenomas. The bland adenoma may be difficult to differentiate from normal pituitary gland, because cytology may be similar. The normal gland has a distinct acinar pattern, sinusoidal vascular spaces, and a mixed cellularity of acidophilic, basophilic, and chromophobic cell types (Fig. 2). The mixed population may be more easily appreciated on Orange G staining, and the architecture of the acini is better seen with a reticulin stain [Fig. 2(B)]. These features of normal pituitary tissue contrast with the larger, and more irregular, lobules of the pituitary adenoma that typically contain one cell type. Historically, tumors were classified on the basis of cytochromic staining characteristics with hematoxylin and eosin (H&E) stain—acidophilic adenomas were assumed to produce GH, basophilic adenomas were thought to secrete ACTH, and chromophobic adenomas were considered endocrinologically inactive (3). Currently, immunohistochemical staining is employed to determine the hormonal content of adenoma cells. Thus, these tumors are identified as producing one, or sometimes more (6), of the adenohypophyseal

INTRODUCTION The pituitary gland usually measures 16 µg/dL) do not need postoperative supplementation. If the integrity of this system is in doubt, perioperative stress–dose steroids followed by physiological replacement should be given until the HPA axis can be tested (71).

RADIATION Surgery for pituitary adenomas is safe and effective, and it remains the primary treatment of most pituitary adenomas; however, when tumors are large and invasive, surgery is not as effective. Recurrence rates for large, invasive, hormonally active tumors can be as high as 93% (72), and radical resection carries a risk of permanent cranial neuropathy. Radiation therapy has long been used to treat residual or recurrent pituitary adenomas. Conventional external beam radiation therapy (XRT) has had good results in terms of controlling tumor growth. This therapy can be administered either as an adjunct to resection or as a primary therapy. In either case, the use of XRT has growth control rates of 82% to 97% (73–83). The efficacy of XRT in normalizing hormone hypersecretion is considerably lower, with control rates of 38% to 83% (73,74,80,81,84). Typical treatment dose is 45 to 50 Gy divided over 25 to 30 fractions. Hypopituitarism is common after XRT, with approximately 50% of patients requiring hormone replacement 20 years after treatment (73). Some authors believe that all patients will experience hypopituitarism after pituitary irradiation given long enough follow-up. XRT carries a low risk of other complications, including cerebral radiation damage, cranial neuropathy including visual deterioration, cerebrovascular injury and stroke, and secondary malignancy (85).

566

Hornyak and Couldwell

Figure 12 SRS treatment plan. Axial (A), coronal (B), and sagittal (C) demonstrations of a treatment plan for radiosurgery to be delivered to the left cavernous sinus. The isocenter is denoted as a “+.” Isodose lines demonstrate the high dose (80%) delivered to the tumor (broad orange line), while the optic apparatus and pituitary gland are relatively spared (

Comprehensive Management of Skull Base Tumors

Endoscopic Skull Base Surgery: A Comprehensive Guide with Illustrative Cases

Management of Pituitary Tumors

Diagnosis and Management of Pituitary Tumors

Diagnosis and Management of Pituitary Tumors

Differential Diagnosis of Temporal Bone and Lateral Skull Base Lesions

Distributed Data-base Management Systems

Comprehensive Stress Management

Skull

Renal and Adrenal Tumors: Biology and Management

Renal and Adrenal Tumors: Biology and Management

Surgery of Spinal Tumors

Surgery of Spinal Tumors

Surgery of Spinal Tumors

Comprehensive Management of Chronic Obstructive Pulmonary Disease

Skull Of Shirzad Mir

Skull Full of Kisses

Vertebral Tumors

Desmoid Tumors

Tumors of the Chest: Biology, Diagnosis and Management

The Derelict of Skull

Surgery in Multimodal Management of Solid Tumors (Updates in Surgery)

Desmoid Tumors

Imaging of Soft Tissue Tumors

Clinical Pathology of Urological Tumors

Atlas of Skull Base Surgery (The Encyclopedia of Visual Medicine Series)

Skull Duggery

Skull Moon

Skull Duggery

Skull Moon

Comprehensive Management of Skull Base Tumors

Endoscopic Skull Base Surgery: A Comprehensive Guide with Illustrative Cases

Management of Pituitary Tumors

Diagnosis and Management of Pituitary Tumors

Diagnosis and Management of Pituitary Tumors

Differential Diagnosis of Temporal Bone and Lateral Skull Base Lesions

Distributed Data-base Management Systems

Comprehensive Stress Management

Skull

Renal and Adrenal Tumors: Biology and Management

Renal and Adrenal Tumors: Biology and Management

Surgery of Spinal Tumors

Surgery of Spinal Tumors

Surgery of Spinal Tumors

Comprehensive Management of Chronic Obstructive Pulmonary Disease

Skull Of Shirzad Mir

Skull Full of Kisses

Vertebral Tumors

Desmoid Tumors

Tumors of the Chest: Biology, Diagnosis and Management

The Derelict of Skull

Surgery in Multimodal Management of Solid Tumors (Updates in Surgery)

Desmoid Tumors

Imaging of Soft Tissue Tumors

Clinical Pathology of Urological Tumors

Atlas of Skull Base Surgery (The Encyclopedia of Visual Medicine Series)

Skull Duggery

Skull Moon

Skull Duggery

Skull Moon

Recommend Documents