Encyclopedia of Diagnostic Imaging

www.dbebooks.com - Free Books & magazines Encyclopedia of Diagnostic Imaging A LBERT L. B AERT (Ed.) Encyclopedia o...

Author: Albert L. Baert | Albert L. Baert

117 downloads 2601 Views 56MB 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

www.dbebooks.com - Free Books & magazines

Encyclopedia of Diagnostic Imaging

A LBERT L. B AERT (Ed.)


Volume 1 A–K

With 1334 Figures and 141 Tables

A LBERT L. B AERT (Ed.)


Volume 2 L–Z

With 1334 Figures and 141 Tables

Editorial Board THOMAS ALBRECHT

THOMAS LINK

JEAN-PHILIPPE BARRAUD

VASILIS NTZIACHRISTOS

CARLO BARTOLOZZI

SILVIA OBENAUER

MARIE-FRANCE BELLIN

PETER REIMER

ALFREDO BLANDINO

PHILIPPE DEMAEREL

JEAN-MICHEL BRUEL

MICHAEL RICCABONA

RICHARD FOTTER

ALBERT

UWE HABERKORN

HENRIK THOMSEN

HANS-ULRICH KAUCZOR

DIERK VORWERK

RAHEL KUBIK-HUCH

JAN WILMINK

DE

ROOS

Editor: Albert L. Baert Department of Radiology University Hospital Gasthuisberg Herestraat 49 B-3000 Leuven Belgium

A C.I.P. Catalog record for this book is available from the Library of Congress ISBN: 978-3-540-35278-5 This publication is available also as: Electronic publication under ISBN 978-3-540-35280-8 and Print and electronic bundle under ISBN 978-3-540-35860-2 This work is subject to copyright. All rights are reserved, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in other ways, and storage in data banks. Duplication of this publication or parts thereof is only permitted under the provisions of the German Copyright Law of September 9, 1965, in its current version, and permission for use must always be obtained from Springer-Verlag. Violations are liable for prosecution under the German Copyright Law. # Springer-Verlag Berlin Heidelberg New York 2008 The use of registered names, trademarks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. Product liability: The publishers cannot guarantee the accuracy of any information about the application of operative techniques and medications contained in this book. In every individual case the user must check such information by consulting the relevant literature. Springer is part of Springer Science+Business Media springer.com Printed on acid-free paper SPIN: 11008750

2109 — 5 4 3 2 1 0

Preface This compact “Encyclopedia of Diagnostic Imaging” is conceived as a two-volume printed work but will also be available online. It aims to provide basic but up-to-date information on all aspects of the very large field of medical imaging. The state of the art character of the encyclopedia is particularly evident in such topics as molecular imaging, magnetic resonance, and contrast media. Around 4000 entries are arranged in alphabetical order with extensive cross-referencing between them. They have been written by internationally recognised leading experts in the field. The encyclopedia should be used as a quick reference book. It can be recommended to medical specialists outside the discipline of radiology, to scientists involved in clinical medical research, but also to general radiologists, radiologists in training, students in medicine, radiographers and interested laypeople, particularly persons from industry and business dealing with medical imaging. For the completion of this encyclopedia, I am very much indebted to Mr. Andrew Spencer from Springer Verlag who provided excellent technical support during the preparation of this work. A.L. BAERT Editor

List of Contributors E RIC ACHTEN

F ISCHMANN A RNE

University of Ghent Ghent Belgium [email protected]

Department of Diagnostic Radiology, University of Tu¨bingen Tu¨bingen Germany [email protected]

F EDERICA AGOSTA Neuroimaging Research Unit, Department of Neurology Scientific Institute and University Ospedale San Raffaele Via Olgettina Italy

T HOMAS A LBRECHT Department of Radiology and Nuclear Medicine, Campus Benjamin Franklin Charite´ - Universita¨tsmedizin Berlin Berlin Germany [email protected]

K. A LFKE Neuroradiology UK SH Kiel Germany [email protected]

C RISTINA G UTIE´ RREZ A LONSO University Hospital Miguel Servet Department of Radiology Zaragoza Spain

K. A NDERS University of Erlangen-Nu¨rnberg Institute of Radiology, Gynaecological Radiology, Erlangen Germany

S UZANNE E. A NDERSON

R OSEMARY A RTHUR Department of Pediatric Radiology, Leeds General Infirmary Clarendon Wing Leeds UK [email protected]

D EWI A SHI Department of Radiology, Faculty of Medicine University of Indonesia Cipto Mangunkusumo General Hospital Salemba Jakarta Pusat Indonesia [email protected]

A LESSIO AUCI Department of Diagnostic and Interventional Radiology University of Pisa Pisa Italy [email protected]

K LAUS K. G AST Department of Radiology Mainz University Hospital Mainz Germany [email protected]

F. E. AVNI

University Hospital of Bern Inselspital Bern Switzerland [email protected]

Department of Medical Imaging University Clinics of Brussels, Erasme Hospital Brussels Belgium [email protected]

G U¨ NTHER A NTES

W. H. B ACKES

Department of Radiology, Klinikum Kempton-Oberallgaü Kempton Germany [email protected]

Department of Radiology, University Hospital Maastricht Maastricht The Netherlands

P INAR B ALCI

LUIS C. A PESTEGUI´ A Hospital Virgen del Cashino Pamplona Spain [email protected]

Dokuz Eylu¨l University School of Medicine Department of Radiology Izmir Turkey [email protected]

R EMIDE A RKUN

DANIELLE B ALERIAUX

Department of Radiology Ege University, School of Medicine Izmir Turkey [email protected]

Clinique de Neuroradiologie, Hoˆpital Erasme Universite´ Libre de Bruxelles Brussels Belgium [email protected]

viii

List of Contributors

R OLAND B ARES

H ARALD B ECHER

University of Tuebingen Tuebingen Germany [email protected]

Oxford University, John Radcliffe Hospital Oxford UK [email protected]

J O¨ RG B ARKHAUSEN Department of Diagnostic and Interventional Radiology and Neuroradiology University Hospital Essen Essen Germany [email protected]

J EAN -P HILIPPE B ARRAUD

C HRISTOPH R B ECKER Ludwig-Maximilians-Universita¨t Munich Germany [email protected]

E DWIN J. R.

VAN

B EEK

Guerbet Roissy France [email protected]

Department of Radiology, Carver College of Medicine, C-751 GH, Iowa City USA [email protected]

S UE B ARTER

G. L. B EETS

Cambridge Breast Unit, Addenbrookes Hospital Cambridge UK [email protected]

C ARLO B ARTOLOZZI Department of Diagnostic and Interventional Radiology University of Pisa Italy

Department of Radiology University of Trieste Trieste Italy

R. G. H. B EETS -TAN Department of Surgery University Hospital Maastricht Maastricht Netherlands

VALENTINA B ATTAGLIA Diagnostic and Interventional Radiology University of Pisa Pisa Italy

J AN S. B AUER Institut fu¨r Ro¨ntgendiagnostik, Technical University Munich Munich Germany [email protected]

J AN S. B AUER Department of Radiology San Francisco USA

W. B AUTZ University of Erlangen-Nu¨rnberg Institute of Radiology, Gynaecological Radiology Erlangen Germany

C E´ DRIC

DE

B AZELAIRE

Service de Radiologie Hoˆpital Saint-Louis Paris France

M ASSIMO B AZZOCCHI Department of Diagnostic and Interventional Radiology University of Udine Udine Italy [email protected]

M ANUEL B ELGRANO Department of Radiology University of Trieste Trieste Italy

M ARIE -F RANCE B ELLIN Department of Radiology, University Hospital Paul Brousse AP-HP University Paris-Sud 11 Villejuif France [email protected]

M ICHELE B ERTOLOTTO Department of Radiology University of Trieste Trieste Italy [email protected]

J U¨ RGEN B IEDERER German Cancer Research Center Heidelberg Germany [email protected]

A LFREDO B LANDINO Department of Radiological Science, University of Messina Messina Italy [email protected]


ix

J OHAN G. B LICKMAN

U LRIKE B REHMER

Department of Radiology UMC St Radboud Nijmegen The Netherlands [email protected]

Department of Diagnostic Imaging University Children’s Hospital Zu¨rich Switzerland [email protected]

M ARTIN J. K. B LOMLEY

C HRISTOPH B REMER

Hammersmith and Charing Cross Hospital NHS Trust & Imperial College London UK

Department of Clinical Radiology & Interdisciplinary Center for Clinical Research (1ZKF), University of Muenster Muenster Germany [email protected]

K ARIN B OCK University of Marburg Marburg Germany [email protected]

A NDREAS B OCKISCH Klinik fur Nuklearmedizin, Universitaetsklinikum Essen Essen Germany

C ARLA B OETES Department of Radiology, Faculty of Medicine University of Indonesia, Cipto Mangunkusumo General Hospital Jakarta Pusat Indonesia

A LEXEI B OGDANOV Departments of Radiology and Cell Biology University of Massachusetts Medical School Worcester USA [email protected]

N ICOLAE B OLOG Institute of Diagnostic Radiology University Hospital Zurich Switzerland

B RUNO B ONNEMAIN Villeparisis France [email protected]

P IERO B ORASCHI 2nd Department of Radiology University Hospital of Pisa Pisa Italy [email protected]

A LEXANDRA B ORGES Instituto Portugueˆs de oncologia Francisco Gentil Centro Regional Oncolo´gico de Lisboa 1093, Lisboa Portugal [email protected]

J ENS B REMERICH Department of Radiology University Hospital Basel Basel Switzerland [email protected]

PAULINE B RICE Hoˆpital de jour d’He´matologie Hoˆpital Saint-Louis Paris France

E LIAS N. B ROUNTZOS Department of Radiology Athens University Medical School, Athans Greece [email protected]

PATRICK A. B ROUWER Leids Universitair Centrum Leiden The Netherlands [email protected]

J EAN M ICHEL B RUEL

Service d’Imagerie Me´dicale Hoˆpital St. E´loi CHU de Montpellier Montpellier France [email protected]

L AURENT B RUNEREAU Poˆle d’imagerie me´dicale CHU de Tours France [email protected]

J-N. B RUNETON Service d’Imagerie Me´dicale Diagnostique et Interventionelle, Hopital de I’Archet Nice France

O RLA B UCKLEY Department of Radiology Tallaght Hospital Dublin Ireland

x


C ARLOS H. B UITRAGO -T ELLEZ

M. C ASSART

University of Basel, Head Radiological Institute, Zofingen Hospital Switzerland [email protected]

Department of Medical Imaging University Clinics of Brussels, Erasme Hospital Brussels Belgium

D OROTHY B ULAS

M AURICIO C ASTILLO

Division of Diagnostic Imaging and Radiology, Children’s National Medical Center, The George Washington University School of Medicine and Health Sciences Washington DC USA [email protected]

W ILLIAM H. B USH , J R . Department of Radiology University of Washington Seattle USA

J OSE C ACERES Department of Radiology, HGU Vall d’Hebron, Universitat Autonoma Barcelona Spain [email protected]

F ILIPPO C ADEMARTIRI Department of Radiology Erasmus Medical Center Rotterdam The Netherlands

C ARLA C APPELLI Department of Diagnostic and Interventional Radiology University of Pisa Pisa Italy [email protected]

G ABRIELE C APRONI Department of Diagnostic and Interventional Radiology University of Pisa Pisa Italy

DAVIDE C ARAMELLA Department of Diagnostic and Interventional Radiology University of Pisa Pisa Italy [email protected]

N ICHOLAS R. C ARROLL Cambridge University Hospitals, NHS Foundation Trust Cambridge UK [email protected]

F EDERICA C ARROZZO Department of Radiological Sciences Policlinico Umberto I-University of Rome La Sapienza, Rome Italy

Department of Radiology University of North Carolina at Chapel Hill Chapel Hill USA

S TEFANO C ERNIC Department of Radiology University of Trieste Trieste Italy

F RANCESCA C ERRI Diagnostic and Interventional Radiology University of Pisa Pisa Italy

A. G. C HALMERS Consultant Radiologists Leeds General Infirmary Leeds UK

C ARMEL T. C HAN Department of Radiology, Molecular Imaging Program at Stanford (MIPS) and Bio-X Program Stanford University School of Medicine Stanford USA [email protected]

F REDERIQUE C HAPON Department of Neuroradiology Hopital Timone Adultes France

N ATALIE C HARNLEY Wolfson Molecular Imaging Centre The University of Manchester Manchester UK

J.-F. C HATEIL Department of Pediatric Imaging Hospital Pellegrin Bordeaux France

S IMON R. C HERRY Department of Biomedical Engineering, Center for Molecular and Genomic Imaging University of California—Davis Davis USA [email protected]


P. C HEVALLIER

JM C ORREAS

Service d’Imagerie Me´dicale Diagnostique et Interventionelle, Hopital de I’Archet Nice France [email protected]

University of Paris V Paris France [email protected]

C HIARA F RANCHINI W K ‘K LING ’ C HONG

Hammersmith and Charing Cross Hospital NHS Trust & Imperial College London UK

Pediatric Neuroradiologist Great Ormond Street Hospital for Children London UK [email protected]

C. C HOSSEGROS Department of Maxillofacial surgery CHU La Timone Marseille France

M ALIK C HOULI Hoˆpital de Biceˆtre Paris France

M.A. C HRESTIAN Department of Pathology CHU La Timone Marseille France [email protected]

S TEFANO C IRILLO Institute for Cancer Research and Treatment Turin Italy [email protected]

M ICHEL C LAUDON Department of Radiology Childrens Hospital, Nancy University Hospital Nancy France [email protected]

R ICHARD H. C OHAN University of Michigan Michigan USA

A RON C OHEN

xi

DAVID O. C OSGROVE

M ARIA A SSUNTA C OVA Department of Radiology University of Trieste Trieste Italy

L AURA C ROCETTI Department of Diagnostic and Interventional Radiology University of Pisa Pisa Italy [email protected]

H EIKE E. DALDRUP-L INK Department of Radiology, Pediatric Radiology section and Director, Contrast agent Research Group University of California in San Francisco USA [email protected]

K ASSA DARGE Department of Pediatric Radiology Institute of Radiodiagnostic, Bavarian Julius-MaximiliansUniversity Wuerzburg Wuerzburg Germany [email protected]

M ARK DAVIES MRI Centre, Royal Orthopedic Hospital NHS Trust Birmingham UK [email protected]

A LBERT

DE

R OOS

Department of Radiology, C2-S Leiden The Netherlands [email protected]

Department of Urology University Hospital Zurich Zurich Switzerland [email protected]

K AREL D EBLAERE

X AVIER C OMBAZ

E D EKEISER

University of Ghent Ghent Belgium

Department of Neuroradiology Hopital Timone Adultes France

Necker Hoˆpital, Service de Radiologie Paris France

G IUSEPPE C OMO

A NNA D ELL’ACQUA

Department of Diagnostic and Interventional Radiology University of Udine Udine Italy

Department of Radiology Giannina Gaslini Hospital for Sick Children Genova Italy

xii


P HILIPPE D EMAEREL

F RANCESCAMARIA D ONATI

Department of Radiology, University Hospitals Leuven, Belgium [email protected]

2nd Department of Radiology University Hospital of Pisa Pisa Italy

P HILIPPE D EMAEREL Department of Radiology University Hospital K.U.Leuven Leuven Belgium [email protected]

D R H UBERT J OHN J OHN R A¨TO T. S TREBEL Department of Urology University Hospital Zurich Zurich Switzerland [email protected]

A LBAN D ENYS Centre Hospitalies Universitire Vandois Lausanne, Lausanne Switzerland [email protected]

LORENZO E. D ERCHI Centre Hospitalies Universitire Vandois Lausanne, Lausanne Switzerland

M ALONE D ERMOT

V ERONICA D ONOGHUE Radiology Department Children’s University Hospital Dublin 1 Ireland [email protected]

VALERIA D’O VIDIO Department of Clinical Sciences, Gastroenteroly Unit, Policlinico Umberto I University of Rome “La Sapienza” Italy

A NTONIOS D REVELEGAS Radiology Department Aristoteles University of Thessaloniki Thessaloniki Greece [email protected]

VOLKER D UDA University of Marburg Marburg Germany [email protected]

R OBERT E CKERSLEY

University of Genova Larva R. Benzi, 8, I-16122 Genova Italy [email protected]

Imaging Sciences Department, Faculty of Medicine Imperial College London London UK [email protected]

A NTONIA D IMITRAKOPOULOU -S TRAUSS

M ICHAEL E ISENHUT

Medical PET Group-Biological Imaging (E0601) Clinical Cooperation Unit Nuclear Medicine, German Cancer Research Center Heidelberg Germany [email protected]

O LLE E KBERG

H. D ITTMANN University of Tuebingen Tuebingen Germany

S ELCUK D OGAN Istanbul University Department of Radiology Istanbul Turkey

J IRI D OLINA Department of Gastroenterology University Hospital Brno Brno Czech Republic [email protected]

Abteilung Radiopharmazeutische Chemie Deutsches Krebsforschungszentrum Heidelberg Germany Department of Clinical Sciences/Medical Radiology, Lund University, Malmo¨ University Hospital Malmo¨ Sweden [email protected]

J AMES H. E LLIS Department of Radiology University of Michigan Ann Arbor, Michigan USA [email protected]

N EVRA E LMAS

¨ niversitesı` Tip Fakultesi Ege U Izmin Turkey [email protected]


S. M. E. E NGELEN

F LAVIO F ORRER

Department of Surgery University Hospital Maastricht Maastricht Netherlands

Department of Nuclear Medicine Erasmus Medical Centre Rotterdam The Netherlands [email protected]

C HRISTOPH E NGELKE Department of Radiology, Klinikum rechts der Isar, Technical University Munich, Munich Germany

G IOVANNI C ARLO E TTORRE Istituto di Radiologia Universita` degli Studi di Catania Catania Italy [email protected]

F. FACON Department of Otorhinolaryngology CHU La Timone Marseille France

K ANTARCI FATIH Istanbul University Department of Radiology Istanbul Turkey

R OSSELLA FATTORI Department of Radiology, Cardiovascular Unit University Hospital S. Orsola Bologna Italy [email protected]

K ATHERINE F ERRARA Department of Biomedical Engineering, Center for Molecular and Genomic Imaging University of California Davis USA

P IM J

DE

F EYTER

Department of Radiology Erasmus Medical Center Rotterdam The Netherlands

M ASSIMO F ILIPPI Neuroimaging Research Unit, Dept. of Neurology, Scientific Institute and University of Ospedale San Raffaele Via Olgettina Italy [email protected]

A RNE F ISCHMANN Department of Diagnostic Radiology University of Tu¨bingen Tu¨bingen Germany [email protected]

xiii

R OSEMARIE F ORSTNER Department of Radiology Paracelsus Private Medical University Salzburger Landeskliniken Salzburg Austria [email protected]

R ICHARD F OTTER Department of Radiology Division of Pediatric Radiology, Universiy Hospital Graz Graz Austria [email protected]

C HIARA F RANCHINI Department of Diagnostic and Interventional Radiology University of Pisa Pisa Italy [email protected]

A LAN H. F REEMAN Consultant Radiologist Cambridge University Hospitals NHS Trust Cambridge UK [email protected]

J ACQUES F RIJA Service de Radiologie Hoˆpital Saint-Louis Paris France

J URGEN J. F U¨ TTERER Department of Radiology University Medical Centre Nijmegen 6500HB Nijmegen The Netherlands [email protected]

M ASSIMO G ALLUCCI Department of Neuroradiology University of L’Aquila Italy [email protected]

S ANJIV S AM G AMBHIR Department of Radiology, Molecular Imaging Program at Stanford (MIPS) and Bio-X Program Stanford University School of Medicine Stanford USA [email protected]

V IOLA G ARBARINO Department of Radiological Sciences, Policlinico Umberto I University of Rome “La Sapienza” Rome Italy

xiv


R OBERTO M ARTI´ NEZ G ARCI´ A

C LAUDIO G RANATA

University Hospital Miguel Servet, Department of Radiology Zaragoza Spain [email protected]

Department of Radiology Giannina Gaslini Hospital for Sick Children Genova Italy

C RISTIANA G ASPARINI Department of Radiology University of Trieste Trieste Italy

F RANCESCA G RANATA University of Messina Messina Italy [email protected]

J. Y. G AUVRIT

H OLGER G REESS

Neuroradiology Department Salengro Hospital Lille France [email protected]

Department of Radiology University Erlangen-Nu¨rnberg Erlangen Germany

H EINRICH K. G EISS Qualita¨tsmanagement-Hygiene-Konzern Bad Neustadt/Saale Germany [email protected]

L ILLI G EWORSKI Charite´-Universita¨tsmedizin Berlin Germany [email protected]

G INO G HIGI University of Bologna Bologna Italy [email protected]

J ONATHAN G ILLARD Cambridge University Hospitals Foundation Trust Cambridge UK [email protected]

N ADINE J G IRARD Department of Neuroradiology Hopital Timone Adultes France [email protected]

C HRISTIAN G REIS Bracco ALTANA Pharma GmbH Klinische Entwicklung Ultraschallkontrastmittel, Konstanz Germany [email protected]

P HILIPPE A. G RENIER Hoˆpital Pitie´-Salpeˆtrie`ne Paris France [email protected]

N ICOLAS G RENIER Service d’Imagerie Diagnostique et The´rapeutique de l’Adulte Groupe Hospitalier Pellegrin Bordeaux France [email protected]

N ORBERT G RITZMANN KH Barmherzige Bru¨der Salzburg Salzburg Austria [email protected]

C HRISTIAN G LASER

LUDOVICO L A G RUTTA

Institut fu¨r Klinische Radiologie der LMU Mu¨nchen GH Mu¨nchen Germany [email protected]


F ERGUS V. G LEESON

F. Z. G UEDDARI

Consultant Radiologist Department of Radiology, Churchill Hospital, Headington Oxford UK [email protected]

Service de Radiologie Centrale, CHN Rabat Morocco [email protected]

N ESLI¨ HAN G U¨ LTAS¸ LI

G IUSEPPE G UGLIELMI

Clinique de Neuroradiologie, Hoˆpital Erasme Universite´ Libre de Bruxelles Brussels Belgium [email protected]

University of Foggia, Foggia, Department of Radiology, Scientific Institute Hospital San Giovanni Rotondo Italy [email protected]


A LEXANDER R. G UIMARAES

P ETER H AUFF

Center for Molecular Imaging Research Massachusetts General Hospital Charlestown USA [email protected]

Research Laboratories Bayer Schering Pharma Berlin Germany [email protected]

G. L. G UIT

S TIJN W. T. P. J. H EIJMINK

Radiology Department Kennemer Gasthuis Haarlem The Netherlands

S ERAP G ULTEKIN Section of Neuroradiology & MRI Center Department of Radiology, Gazi University School of Medicine Besevler, Ankara 06510 Turkey [email protected]

PATRICK H AAGE Department of Diagnostic and Interventional Radiology HELIOS Klinikum Wuppertal, University Hospital Witten/ Herdecke Wuppertal Germany [email protected]

U WE H ABERKORN Radiologische Klinik und Poliklinik Universita¨tsklinikum Heidelberg Heidelberg Germany [email protected]

Department of Radiology, Radboud University Nijmegen Medical Centre Geert Grooteplein 10 The Netherlands

G. H EINZ -P EER Department of Radiology Medical University Vienna Austria [email protected]

O LIVIER H ELENON Neckert Paris France

T HOMAS H ELMBERGER Clinic of Radiology and Nuclear medicine University Hospitals Schleswig-Holstein Campus Luebeck Germany [email protected]

K. G. H ERING

Klinik fur Nuklearmedizin, Universitaetsklinikum Essen Essen Germany

Beratender Arzt fu¨r Diagnostische Radiologie bei arbeits – und umweltbedingten Erkrankungen Knappschaftskrankenhaus Dortmund Germany [email protected]

DAVID M. H ANSELL

K LAUS -P ETER H ERMANN

M ONIA H AMAMI

Department of Radiology Royal Brompton Hospital London UK [email protected]

K. M. H ARRIS Consultant Radiologists Leeds General Infirmary Leeds UK

R OBERT P. H ARTMAN

Gesellschaft Wissenschaftliche Datenverarbeitung mbH Go¨ttingen Germany [email protected]

B EATRIZ I ZQUIERDO H ERNA´ NDEZ University Hospital Miguel Servet, Department of Radiology, Zaragoza Spain

F RANC H. H ETZER

Department of Diagnostic Radiology Mayo Clinic Rochester MN USA

Clinic of Visceral and Transplant Surgery University Hospital Zurich Switzerland

C HRISTOPHER J. H ARVEY

C LAUS P ETER H EUSSEL

Hammersmith and Charing Cross Hospital NHS Trust & Imperial College London UK [email protected]

Diagnostic and Interventional Radiology Thoraxklinik Heidelberg Heidelberg Germany [email protected]

xv

xvi


M ELANIE H IORNS

J UDY I LLES

Radiology Department, Great Ormond Street Hospital for Children London UK [email protected]

Stanford Center for Biomedical Ethics Stanford University Stanford USA [email protected]

O. H E´ LE´ NON

F. I MANI

Necker Hoˆpital, Service de Radiologie Paris France [email protected]

Service de Radiologie Centrale, CHN Rabat Morocco

T HILO H OELSCHER

Institut fu¨r Radiologie Charite´ Campus Mitte Berlin Germany

Department of Radiology University of California San Diego USA [email protected]

P. A. M. H OFMAN Department of Radiology, University Hospital Maastricht The Netherlands [email protected]

Y VONNE L. H OOGEVEEN Department of Radiology, Faculty of Medicine University of Indonesia, Cipto Mangunkusumo General Hospital Jakarta Pusat Indonesia

B ENJAMIN Y. H UANG Department of Radiology University of North Carolina at Chapel Hill Chapel Hill USA [email protected]

T. A. H UISMAN Johns Hopkins University Baltimore USA [email protected]

R OBERT W. H URST Radiology, Neurology, and Neurosurgery University of Pennsylvania Philadelphia USA [email protected]

F RANCO I AFRATE

A HI S EMA I SSEVER

S IMON A J ACKSON Imaging Directorate Derriford Hospital Plymouth UK [email protected]

J ARL A˚ . J AKOBSEN Department of Radiology Rikshospitalet Oslo Norway

O. J ANSEN Neuroradiology UK SH Kiel Germany

H OHMANN J OACHIM Universita¨tsspital Basel Institut fu¨r Radiologie Basel Switzerland [email protected]

F RANCIS J OFFRE CHU Toulouse France [email protected]

H UBERT J OHN Klinik Hirslanden Zentrum fu¨r Urologie, 8008 Zurich Switzerland [email protected]

Director of Unit of CT and MR, University of Rome “La Sapienza” Polo Pontino, I.C.O.T. Latina Italy [email protected]

M ARION D E J ONG

J EAN -M ARC I DE´ E

N ICO

Guerbet Research Division Aulnay-sous-Bois France [email protected]

ErasmusMC Rotterdam The Netherlands [email protected]

Department of Nuclear Medicine Erasmus Medical Centre Rotterdam The Netherlands DE

J ONG


H ANS -U LRICH K AUCZOR

B ODO K RESS

Department of Radiology German Cancer Research Center Heidelberg Germany [email protected]

Department of Neurology University of Heidelberg Medical Section Heidelberg Germany

E RIC

Department of Radiology Medical University of Vienna, Vienna General Hospital Vienna Austria [email protected]

DE

K ERVILER

Service de Radiologie Hoˆpital Saint-Louis Paris France [email protected]

E RIC DE K ERVILER B ERNARD K ING Department of Diagnostic Radiology Mayo Clinic Rochester MN USA

M ATTHEW P. K IRSCHEN Stanford Center for Biomedical Ethics Stanford University Stanford USA

PATRICK R. K NU¨ SEL Institut fu¨r Radiologie Kontonsspital Baden Baden Switzerland [email protected]

TAKAHASHI KOJI Department of Radiology, Asahikawa Medical college, Asahikawa Japan taka1019@asahikawa_med.ac.jp

E LISA E. KONOFAGOU Department of Biomedical Engineering and Radiology Columbia University New York USA [email protected]

F RANCK KOSKAS Department of Radiology University Paris-Sud 11, Paul Brousse Hospital Villejuif France [email protected]

S. I. KOUKOURAKI Department of Nuclear Medicine University, Iraklion Crete Greece [email protected]

S ILVIA KOVACS Department of Radiology University Hospital

C HRISTIAN R. K RESTAN

G ABRIEL P. K RESTIN Department of Radiology Erasmus Medical Center Rotterdam The Netherlands

M ARTIN K RIX Bracco ALTANA Pharma GmbH Clinical Research Konstanz Germany [email protected]

G ABRIELE A. K ROMBACH Department of Radiology University Hospital Aachen Germany [email protected]

G ERRIT K RUPSKI -B ERDIEN Universita¨tsklinikum Hamburg-Eppendorf Hamburg Germany [email protected]

S ABRINA KO¨ SLING Martin-Luther-Universita¨t Halle-Wittenberg Klinik fu¨r Diagnostische Radiologie Halle Germany [email protected]

B ARBARA KO¨ TZ Wolfson Molecular Imaging Centre The University of Manchester Manchester UK [email protected]

R AHEL A. K UBIK -H UCH Institute of Radiology, Kantonsspital Badan Badan Switzerland [email protected]

RU¨ DIGER

VON

K UMMER

Universita¨tsklinikum Carl Gustav Carns Dresden Germany [email protected]

xvii

xviii


PASCAL L ACOMBE

T HOMAS M. L INK

Department of Radiology Hoˆpital Ambroise Pare´ Universite´ Paris Ile-de-France Ouest Boulogne-Billancourt France

Department of Radiology University of California San Francisco USA [email protected]

A NDREA L AGHI Director of Unit of CT and MR, University of Rome “La Sapienza” Polo Pontino, I.C.O.T. Latina Italy [email protected]

K ARL LUDWIG Department of Diagnostic Radiology Orthopedic University Hospital Heidelberg Heidelberg Germany [email protected]

M. J. L AHAYE

P INA LUIS

Department of Surgery University Hospital Maastricht Maastricht Netherlands

Clıńica Universitaria de Navarra, Navarra Spain [email protected]

T HOMAS C. L AUENSTEIN Department of Radiology The Emory Clinic Atlanta USA [email protected]

H. P. L EDERMANN Department of Radiology University Hospital Basel, Institute of Diagnostic Radiology Basel Switzerland [email protected]

M ICHAEL L ELL Department of Radiology University Erlangen-Nu¨rnberg Erlangen Germany [email protected]

R ICCARDO L ENCIONI Department of Diagnostic and Interventional Radiology University of Pisa Pisa Italy [email protected]

A DRIAN K. P. L IM Hammersmith and Charing Cross Hospital NHS Trust & Imperial College London UK

N ICOLA L IMBUCCI Department of Neuroradiology University of L’Aquila Italy

G. M. L INGG Department of Radiology Sana Rheumazentrum Rheinland Pfalz Bad Kreuznach Germany [email protected]

C LEMENS LO¨WIK Endocrinology Research Laboratory and Molecular Imaging Leiden University Medical Center The Netherlands [email protected]

F RANCESCA M ACCIONI Department of Radiological Sciences Policlinico Umberto I-University of Rome La Sapienza, Rome Italy [email protected]

F REDERIK M AES Universitaire Ziekenhuizen Lenven Belgium [email protected]

D EAN D. T. M AGLINTE Indiana University School of Medicine Indianapolis USA [email protected]

B RINDER M AHON Queen Elizabeth Hospital Birmingham UK

PATRIZIA M ALAGUTTI Department of Radiology Erasmus Medical Center Rotterdam The Netherlands

G EERT M ALEUX Department of Radiology University Hospitals Leuven Leuven Belgium [email protected]

DAVID M ALFAIR UCSF San Francisco USA [email protected]


xix

A DA M ANDIC

C OLM M C M AHON

Department of Radiology University of Trieste Trieste Italy

Department of Radiology, St Vincent’s University Hospital Dublin Ireland

DAMIEN M ANDRY

W. B OB M EIJBOOM

Department of Radiology Childrens Hospital, Nancy University Hospital Nancy France


VASSILIOS M ANIATIS

LUIS H. R OS M ENDOZA

CT and MRI Department HYGEIA Hospital Eritherou stvrou 4, GR-15123, Maroussi Athens Greece [email protected]

G IULIANO M ARIANI Department of Diagnostic and Interventional Radiology and of Nuclear Medicine University of Pisa Italy

K ATHARINA M ARTEN Department of Radiology, Klinikum rechts der Isar, Technical University Munich, Munich Germany [email protected]

F. M ARTINELLI

University Hospital Miguel Servet Department of Radiology Zaragoza Spain [email protected]

Y VES M ENU Hoˆpital de Biceˆtre Paris France [email protected]

S AMUEL M ERRAN Neckert Paris France

S VEN M ICHEL

Radiodiagnostic Unit University of Florence Careggi Hospital Florence Italy

Radiology Kantonsspital Baden, 5404 Baden Switzerland [email protected]

O LIVIER M ATHIEU

WALTER M IER

Service de Radiologie Hoˆpital Saint-Louis Paris France

Abteilung Radiopharmazeutische Chemie Deutsches Krebsforschungszentrum Heidelberg Germany

C. M ATOS

PAOLA M ILILLO

Department of Medical Imaging University Clinics of Brussels, Erasme Hospital Brussels Belgium

Istituto di Radiologia Universita` degli Studi di Catania Catania Italy

J ULIAN M ATTHEWS

FABIO M INUTOLI

Wolfson Molecular Imaging Centre The University of Manchester Manchester UK [email protected]

Department of Radiological Sciences University of Messina, Messina Italy [email protected]

S ALVATORE M AZZEO Divisions of Diagnostic and Interventional Radiology University of Pisa Pisa Italy

Department of Radiology University Hospital Schleswig-Holstein, Campus Kiel Kiel Germany [email protected]

S ILVIO M AZZIOTTI

N ICO R M OLLET

Department of Radiological Sciences, University of Messina, Messina Italy [email protected]


S TEFAN M U¨ LLER-H U¨ LSBECK

xx


O M ONNET

C ORINNA E. E. N IERHOFF

Department of Radiology CHU La Timone Marseille France

Department of Diagnostic Radiology Orthopedic University Hospital Heidelberg Heidelberg Germany

S ANDRA M OORE

R. J. N IJENHUIS

NYU Medical Center New York USA [email protected]

A NDERS N ILSSON

S AMEH K. M ORCOS Department of Diagnostic Imaging Northern General Hospital, Sheffield Teaching Hospitals NHS Trust Sheffield UK

R OBERT M ORGAN St George’s NHS Trust London UK [email protected]

W. B. M ORRISON Department of Radiology Thomas Jefferson University Hospital Philadelphia USA

Department of Radiology, University Hospital Maastricht The Netherlands Department of Radiology, Ultrasound University, Uppsala University Hospital Uppsala Sweden [email protected]

VASILIS N TZIACHRISTOS Harvard Medical School Harvard University HMS/MGH, Center for Molecular Imaging Research Charlestown USA [email protected]

S ILVIA O BENAUER Department of Radiology Georg-August-University of Goettingen Goettingen Germany [email protected]

A LAN E. O ESTREICH G. M OULIN Department of Radiology CHU La Timone Marseille France

TAKASHI N AKAMURA Department of Radiology and Cancer Biology Nagasaki University School of Dentistry Nagasaki Japan [email protected]

L. N APOLITANO Radiodiagnostic Unit University of Florence Careggi Hospital Florence Italy [email protected]

E MANUELE N ERI Diagnostic and Interventional Radiology University of Pisa Italy [email protected]

T ONY N ICHOLSON Leeds Teaching Hospitals NHS Trust Leeds UK [email protected]

Cincinnati Children’s Hospital Medical Center Cincinnati, Ohio USA [email protected]

M ONICA O LAR Poˆle d’imagerie me´dicale CHU de Tours France

R OBERT O LSZEWSKI Department of Cardiology Military Medical Institute Warsaw Poland

PATRICK O MOUMI Poˆle d’imagerie me´dicale CHU de Tours France

P HILIPPE O TAL CHU Toulouse France [email protected]

R. H. O YEN Department of Radiology University Hospitals Gasthuisberg Leuven Belgium [email protected]


E NVER O ZER

P IETRO PAVLICA

Department of Otolaryngology, Head and Neck Surgery University Medical Center, The Ohio State University Columbus USA

University of Genova Largo R. Benzi, 8, I-16122 Genova Italy

B ERNARD P. PAELINCK

R ONALD P EETERS

University Hospital Antwerp Antwerp Belgium [email protected]

Department of Radiology University Hospital

A LESSANDRO PALUMBO Department of Radiology Erasmus Medical Center Rotterdam The Netherlands

I GNAZIO PANDOLFO Department of Radiological Science University of Messina Italy [email protected]

A MALIA PAONESSA

J EAN -P IERRE P ELAGE Department of Radiology Hoˆpital Ambroise Pare´ Universite´ Paris Ile-de-France Ouest Boulogne-Billancourt France [email protected]

G. P ELLICANO` Radiodiagnostic Unit University of Florence Careggi Hospital Florence Italy

M ARZIO P ERRI

Department of Neuroradiology University of L’Aquila Italy

Department of Diagnostic and Interventional Radiology University of Pisa Italy [email protected]

J. PARIS

DARJA B ABNIK P ESKAR

Department of Otorhinolaryngology CHU La Timone Marseille France

University Medical Center Ljubljana Ljubljana Slovenia

S ARA PARLANTI

Medisch Centrum Haaglanden, Netherlands [email protected]

Department of Radiological Sciences, Policlinico Umberto I University of Rome “La Sapienza” Italy

Z OLTAN PATAY Department of Radiological Sciences St. Jude Children’s Research Hospital Memphis USA [email protected]

A NNE PATERSON

W ILCO P EUL

C ATHERINE M. P HAN Hoˆpital de Biceˆtre Service de Radiologie BROCA Le Kremlin-Biceˆtre France [email protected]

F RANK P ILLEUL

Royal Belfast Hospital for Sick Children Belfast UK [email protected]

Gastro-intestinal and Emergency Imaging Department, Hoˆpital Universitaire Edouard Herriot, Lyon France [email protected]

J ACOB PATIJN

LUIS P INA

Pain Management and Research Centre, Department of Anaesthesiology University Medical Centre Maastricht Maastricht The Netherlands [email protected]

Department of Radiology Clıńica Universitaria de Navarra Navarra Spain [email protected]

P IETRO PAVLICA

Pharmacovigilance Department Guerbet Roissy CDG France [email protected]

Department of Radiology Hospital M. Malpighi Bologna Italy

E MMANUELLE P INE` S

xxi

xxii


B RIAN W. P OGUE

PATRICK R EINARTZ

Thayer School of Engineering Dartmouth College Hanover USA [email protected]

Department of Nuclear Medicine, University Hospital Aachen, Aachen Germany [email protected]

P ETER P OKIESER

R APHAELLE R ENARD -P ENNA

Medical University of Vienna, Medical Media Services, Department of Radiology Vienna Austria

PANOS K. P RASSOPOULOS Department of Radiology University Hospital of Alexandronpolis, Medical School of Thace Alexandronpolis Greece [email protected]

PAT P RICE Wolfson Molecular Imaging Centre The University of Manchester Manchester UK

F RANCESCA P UGLIESE Department of Radiology Erasmus Medical Center Rotterdam The Netherlands

N AGMI R. Q URESHI Fellow in Thoracic Radiology [email protected]

E. W. R ADUE

Service de Radiologie Hospital Pitie` Salpetrie`re, Assistance Publique des hospitaux de Paris Universite´ Pierre et Marie Curie Paris, France [email protected]

S HELLEY R ENOWDEN Department of Neuroradiology University of Bristol Bristol UK [email protected]

M. R ICCABONA Division of Pediatric Radiology Medical University of Graz, University Hospital Graz Austria [email protected]

A SHLEY R OBERTS University Hospital of Wales Cardiff UK [email protected]

Department of Neuroradiology University Hospital Basel Switzerland [email protected]

M ARIA A. R OCCA

T HOMAS R AND

L AURENCE R OCHER

Allgemeines Krankenhaus-Universita¨tskliniken Vienna Austria [email protected]

F INN R ASMUSSEN Aarhus University Hospital Aarhus Denmark [email protected]

M ARGRIT R EICHEL Referenzzentrumsleiterin Mammographie-Screening Wiesbaden Germany [email protected]

P ETER R EIMER Department of Radiology Klinikum Karlsruhe, Academic Teaching Hospital of the University of Freiburg Karlsruhe Germany [email protected]

MRI Research Group Scientific Institute Fondazione Don Gnocchi Milan Italy Hoˆpital de Biceˆtre Paris France

M ATHIEU H. R ODALLEC Department of Radiology Fondation Hoˆpital Saint-Joseph Paris France

J USTUS E. R OOS Department of Radiology Stanford University Medical Center Stanford USA [email protected]

R OSEMARY A RTHUR Department of Pediatric Radiology, Leeds General Infirmary Clarendon Wing Leeds UK [email protected]


S ANDRA J. R OSENBAUM -K RUMME

C ORNELIA S CHAEFER-P ROKOP

Klinik fu¨r Nuklearmedizin, Universitaetsklinikum Essen Germany

AMC Amsterdam The Netherlands [email protected]

A NDREA R OSSI Department of Pediatric Neuroradiology G. Gaslini Children’s Research Hospital Genoa Italy [email protected]

C ATHERINE R OY Department of Radiology B University Hospital of Strasbourg, Hoˆpital Civil 1 Strasbourg France [email protected]

G IUSEPPE R UNZA Department of Radiology Erasmus Medical Center Rotterdam The Netherlands

M ARY R UTHERFORD Robert Steiner MR Unit Imaging Science Dept MRC Clinical Sciences Centre Imperial College Hammersmith Campus London UK [email protected]

N ILOUFAR S ADEGHI Department of Radiology, Erasme Hospital, Brussels Belgium [email protected]

C AROLINE S AGE Department of Radiology University Hospital

M. A. S AHRAIAN Department of Neurology Sina Hospital Hassan abad Square Tehran Iran

S IMONETTA S ALEMI Department of Diagnostic and Interventional Radiology University of Pisa Pisa Italy

K UMARESAN S ANDRASEGARAN

xxiii

M ARTINA S CHARITZER Medical University of Vienna, Medical Media Services, Department of Radiology Vienna Austria

B ARRY S CHENK Leids Universitair Centrum Leiden Leiden The Netherlands

A. M. D E S CHEPPER Leiden University Medical Centre Leiden The Netherlands [email protected] [email protected]

M. S CHA¨ FERS Department of Nuclear Medicine University of Mu¨nster Mu¨nster Germany [email protected]

M ARC S CHMITTER Department of Prosthodontic, Department of Neurology University of Heidelberg Medical Section Heidelberg Germany

U. S CHNO¨ CKEL Department of Nuclear Medicine University of Mu¨nster Mu¨nster Germany

O TMAR S CHOBER Department of Nuclear Medicine, Coordinator Collaborative Research Center Molecular Cardiovascular Imaging (SFB 656-MoBil), University Hospital Mu¨nster Mu¨nster Germany [email protected]

M ARIA S CHODER Medizinische Universitat Wien Vienna Austria [email protected]

C. S CHORN

Indiana University School of Medicine Indianapolis USA

Department of Radiology Sana Rheumazentrum Rheinland Pfalz Bad Kreuznach Germany

M AREK J. S ASIADEK

R. S CHULZ -W ENDTLAND

Department of General Radiology, Interventional Radiology and Neuroradiology, Wroclaw University Wroclaw Poland [email protected]

University of Erlangen-Nu¨rnberg, Institute of Radiology Gynaecological Radiology Erlangen Germany [email protected]

xxiv


DANIEL S COFFINGS

E RICH S ORANTIN

Cambridge University Hospitals Foundation Trust, Cambridge UK

Department of Radiology, Division of Pediatric Radiology University Hospital Graz Graz Austria [email protected]

K URUGOGLU S EBUH Istanbul University Department of Radiology Istanbul Turkey

A SHLEY S. S HAW Consultant Radiologist Cambridge University Hospitals NHS Trust Cambridge UK

J OHN S HIRLEY

G UNDULA S TAATZ Department of Radiology, Division of Pediatric Radiology University of Erlangen-Nu¨rnberg Erlangen Germany [email protected]

F ULVIO S TACUL

Imaging Directorate Derriford Hospital Plymouth UK [email protected]

Department of Radiology University of Trieste Trieste Italy [email protected]

P HIL S HORVON

LYNNE S TEINBACH

Central Middlessex hospital London UK [email protected]

Department of Radiology University of California San Francisco USA [email protected]

F RANCESCA S ILIQUINI Department of Radiological Sciences, Policlinico Umberto I University of Rome “La Sapienza” Italy [email protected]

A KIF S IRIKCI Radiology Department Gaziantep University, School of Medicine Gaziantep Turkey [email protected]

D OMINIQUE S IRINELLI Poˆle d’imagerie me´dicale CHU de Tours France

M ARC S IROL Department of Cardiology Lariboisie`re Hospital Paris France [email protected]

G ILES S TEVENSON McMaster University Cowichan Hospital Duncan British Columbia Canada [email protected]

M. P. M. S TOKKEL Department of Nuclear Medicine LUMC Leiden The Netherlands [email protected]

N EIL S TOODLEY North Bristol NHS Trust Bristol UK [email protected]

A.

VAN

S TRATEN

University Medical Center Ljubljana Ljubljana Slovenia [email protected]

Radiology Department Kennemer Gasthuis Haarlem The Netherlands [email protected]

S AT S OMERS

R A¨TO T. S TREBEL

Department of Radiology McMaster University, West Hamilton Ontario Canada [email protected]

Department of Urology University Hospital Zurich Zurich Switzerland [email protected]

E VELINE S MRTNIK


M A E UGENIA G UILLE´ N S UBIRA´ N

M AJDA M. T HURNHER

University Hospital Miguel Servet Department of Radiology Zaragoza Spain

Universita¨tsklinik fu¨r Radiodiagnostik Medizinische Universita¨t Wien Vienna Austria [email protected]

M ISA S UMI Department of Radiology and Cancer Biology Nagasaki University School of Dentistry Nagasaki Japan [email protected]

S TEFAN S UNAERT Department of Radiology University Hospital Leuven The Netherlands [email protected]

N ICOLA S VERZELLATI

PAOLO T OMA Department of Radiology Giannina Gaslini Hospital for Sick Children Genova Italy [email protected]

B ERND T OMBACH Department of Clinical Radiology, University of Mu¨nster Institut fu¨r Klinische Radiologie, Universita¨tsklinikum Mu¨nster Mu¨nster Germany [email protected]

Department of Clinical Sciences, Section of Radiology University of Parma Italy [email protected]

W ILLIAM C T ORREGGIANI

KOJI TAKAHASHI

PAOLO T ORTORI -D ONATI

Department of Radiology Asahikawa Medical College Asahikawa Japan [email protected]

Department of Radiology Tallaght Hospital Dublin Ireland [email protected] Department of Pediatric Neuroradiology G. Gaslini Children’s Research Hospital Genoa Italy

T HOMAS C. T REUMANN E. T URGUT TALI Section of Neuroradiology & MRI Center Department of Radiology, Gazi University School of Medicine Besevler Ankara 06510 Turkey [email protected]

PATRICE TAOUREL Service d’Imagerie Me´dicale, Hoˆpital Lapeyronie Montpellier France [email protected]

I. K AARE T ESDAL Universita¨tsklinikum Mannheim Institut fu¨r Radiologie und Nuklearmedizin Klinikum Friedrichshafen Friedrichshafen Germany [email protected]

xxv

Kantonspital Luzern Luzern Switzerland [email protected]

E LENA T RINCIA Department of Radiology University of Trieste Trieste Italy [email protected]

F RANCESCA T URINI Diagnostic and Interventional Radiology University of Pisa Pisa Italy

M ICHAEL U DER Universita˜tklinikum Erlangen Erlangen Germany [email protected]

H ENRIK S. T HOMSEN

KORMAN U GUR

Department of Diagnostic Radiology Copenhagen University Hospital at Herlev Herlev Denmark [email protected]

Istanbul University Department of Radiology Istanbul Turkey [email protected]

xxvi


VOLKMAR U HLENDORF

D IERK VORWERK

ETH Zu¨rich Institut fu¨r Pharmazeutische Wissenschaften Zu¨rich Switzerland [email protected]

Klinikum Ingolstadt GmbH Direktor des Institutes fu¨r diagnostische und interventionelle Radiologie Ingolstadt Germany [email protected]

M ARTIN U NTERWEGER

R OLF VOSSHENRICH

Diagnostic Radiology Cantonal Hospital Baden Baden Switzerland [email protected]

Radiologen-Gemeinschaftspraxis MRT im Friederikenstift Hannover Germany [email protected]

PAOLA VAGLI

University Medical Center Groningen Groningen Netherlands [email protected]

Department of Diagnostic and Interventional Radiology University of Pisa Italy

V LASTIMIL VALEK Department of Radiology University Hospital Brno Brno Czech Republic [email protected]

E LS M. VANHERREWEGHE Department of Radiology Ghent University Hospital Ghent Belgium

F ILIP M. VANHOENACKER Department of Radiology University Hospital Antwerp Belgium Antwerp [email protected]

A NNICK V IEILLEFOND Neckert Paris France

C LAUDIO V IGNALI Department of Diagnostic and Interventional Radiology University of Pisa, Pisa Italy

G EERT M. V ILLEIRS Genitourinary Radiology Ghent University Hospital Ghent Belgium [email protected]

C ARMELA V ISALLI Department of Radiologic Sciences, University of Messina Messina Italy [email protected]

PATRICK V ROOMEN

P IETER WAER Dept of Radiology University Hospitals K.U. Leuven Leuven Belgium

M. WARMUTH -M ETZ Department of Neuroradiology University of Wu¨rzburg Wu¨rzburg Germany [email protected]

J UDITH A. W. W EBB Diagnostic Radiology Department The London Clinic St Bartholomew’s Hospital London UK [email protected]

J OHANNES W EBER Section of Neuroradiology University of Freiburg Medical School Freiburg Germany [email protected]

S. W ECKBACH Institut fu¨r Klinische Radiologie der LMU Mu¨nchen GH Mu¨nchen Germany

R. E. W EIJERS Department of Radiology University Hospital Maastricht Maastricht The Netherlands [email protected]

D OMINIK W EISHAUPT Institute of Diagnostic Radiology University Hospital Zurich Switzerland [email protected]


R ALPH W EISSLEDER

I NDRA YOUSRY

Center for Molecular Imaging Research Massachusetts General Hospital Charlestown USA

Institute of Neurology London UK

J AN T. W ILMINK Radiology Department University Hospital Maastricht Maastricht The Netherlands [email protected]

TAREK A. YOUSRY Institute of Neurology London UK [email protected]

G UIDO W ILMS

J OHANNES Z ACHERL

Universitaire Ziekenhuizen Leuven Leuven Belgium [email protected]

Medical University of Vienna, Medical Media Services, Department of Radiology Vienna Austria

K LAUS WOERTLER Department of Radiology, Klinikum rechts der Isar Technical University Munich Munich Germany [email protected]

xxvii

J OHANNES Z ENK Department of Head and Neck Surgery University Erlangen-Nu¨rnberg Erlangen Germany

A. W UTTGE - HANNIG

M ARC Z INS

Radiologie-Strahlentherapie-Nuklearmedizin Munich Germany [email protected]

Department of Radiology Fondation Hoˆpital Saint-Joseph Paris, France [email protected]

A

ABBI and Site-Select Devices, Breast R. S CHULZ -W ENDTLAND, K. A NDERS , W. B AUTZ University of Erlangen-Nu¨rnberg, Institute of Radiology, Gynaecological Radiology, Erlangen [email protected]

Indications An advanced breast biopsy instrumentation (ABBI)/siteselect system may be used for stereotactically aimed biopsy in lesions smaller than 5 mm to avoid or to reduce sampling errors in the following settings: . Histological assessment of suspicious lesions seen at mammography only (BI-RADS IV); . Presurgical confirmation of malignancy in suspicious lesions seen at mammography only (RADS V); . Histological assessment in the differential diagnosis of mastopathic changes versus ductal carcinoma in situ (DCIS) (suspicious calcifications; BI-RADS IV/V).

Contraindications Severe coagulopathy as well as allergic reactions to local anesthetics constitute absolute contraindications to using the ABBI/site-select system.

Technology Both the ABBI (advanced breast biopsy instrumentation; Firma Lorad) and the site-select system (Firma Ethicon Endo-Surgery, Breast Care) support digital stereotactic excisional biopsy. Computerized digital mammography (10 line pairs/mm, monitor matrix of 10241024 pixel) is used to guide stereotactic excision with oscillating knives to perform en-bloc resection of mammographic lesions (BI-RADS IV/V) with patients in prone position (biopsy tables provided by Lorad and Fischer).

For stereotactic excision, the patient’s breast is positioned in a table opening, thereby being accessible from under the table. Following disinfection of the biopsy area, the stereotactic equipment is calibrated and the target area for the subsequent intervention and stereo views (±15˚) is determined by means of orthogonal mammography. The breast tissue is fixed by the compression pads, and color marks are used to display positioning as well as localization and biopsy route on the patient’s skin. As a matter of principle, oncological as well as surgical considerations—transection route in planned subsequent segmental resection—should be taken into account in suspicious lesions (BI-RADS IV/V). Biopsy is performed following local anesthesia and cutaneous incision. Biopsy needles can be chosen according to the size of the target lesion (diameter 5/10/ 15 or 20 mm). If the 20-mm rotational-scalpel cannula is used, conventional surgical hemostasis has to be applied, whereas when using smaller biopsy needles, local compression for about 30 min is sufficient. The biopsy is performed after digital stereotactic localization of the equivocal lesion (BI-RADS IV/V) and placement of a t-lock needle. After additional infiltration with local anesthetics and enlargement of the stab incision, the biopsy cannula is slowly pushed forward according to the depth of the lesion. The oscillating knife is advanced 15 mm beyond the calculated location of the target, and subsequently the entire lesion is excised. With the help of an electrocoagulation wire, the tissue bloc is displaced in situ. Digital reevaluation views are used to document complete excision, and a Mikromark clip (Biopsis Medical) is placed in the resulting cavity to mark the biopsy area for follow-up studies. The biopsy material is thereafter marked with thread to facilitate the surgeon’s and pathologist’s orientations. For monitoring reasons, an X-ray of the biopsy tissue is also made (Fig. 1a–d). The tissue is deep-frozen and then paraffin-cut microscopic work-up is performed (time needed: approximately 24 h) according to the “European Guidelines for Quality Assurance.” Further proceedings are determined depending on the histological findings and the concordance or discrepancy between imaging findings and histology.

2

ABBI and Site-Select Devices, Breast

ABBI and Site-Select Devices, Breast. Figure 1 (a) Positioning and compression of the patient’s breast in a table opening before digital stereotactic localization and excision biopsy. (b) After digital stereotactic localization, desinfection of the biopsy area, local anesthesia and cutaneous incision, the 20 mm cannula is placed. (c) Whole stereotactic/radiological procedure: Scout view, stereotactic localization (+/− 15 degree), placing of the 20mm cannula at the microcalcifications and the t-lock needle. (d) Preparation radiogram with t-lock needle of the lesion of Fig. 1c: High grade DCIS.

Aberrations of Normal Development and Involution (ANDI)

As soon as the biopsy procedure is finished, the patient is put in supine position for thorough hemostasis using electrocoagulation and compression. The cutaneous defect is sutured with absorbable thread. In addition, to avoid secondary bleeding and hematoma, a compression bandage is provided. In benign findings, follow-up mammography should be performed 6 and 12 months after biopsy; in case of progression of suspicious findings, surgical clarification is mandatory.

2. 3. 4. 5. 6. 7. 8.

Duda VF, Schulz-Wendtland R (2004) (Hrsg) Mammadiagnostik. Springer-Verlag, Berlin, Heidelberg, New York Fischer U (2003) (Hrsg) Mammographiebefund nach BI-RADS. Thieme-Verlag, Stuttgart Heywang-Ko¨brunner SH, Schreer J (2003) (Hrsg) Bildgebende Mammadiagnostik. Thieme-Verlag, Stuttgart Lanyi M (2003) (Hrsg) Brustkrankheiten im Mammogramm. Springer-Verlag, Berlin, Heidelberg, New York Parker SH, Jobe WE (1993) (Hrsg) Percutaneous Breast Biopsy. Raven Press, New York Sickles EA (2001) (Hrsg) Breast Imaging. Lippincott Williams & Wilkins, Baltimore Tabar L, Tot T, Dean PB (2003) (Hrsg) Breast Cancer. ThiemeVerlag, Stuttgart

Results The sensitivity and specificity as quoted in the current literature approach or even reach 100%.

Adverse Events

Abdominal Cramps ▶Acute Abdomen, Genital Causes

Minor pain, hemorrhage, or vasovagal reactions may occur. . Pain may be minimized by profound local anesthesia; since the breast parenchyma itself is obviously less algesic than skin, thorough dermal anesthesia in particular should be of concern. . Stab incisions usually entail minor bleeding; to avoid larger hematoma, broad compression is applied (30-min manual compression, compression bandage). . Vasovagal reactions seem to occur more frequently if mammographic intervention is performed in seated rather than recumbent patients. Individual support and suitable premises may help to prevent these reactions.

Abdominal Injury ▶Trauma Hepatobiliary

Abdominal Trauma ▶Trauma Hepatobiliary

Outlook Stereotactic biopsy is approved and established for diagnostic purposes only. In the future, questions will arise about whether small carcinoma (2.5 cm) are more often symptomatic because of their mass effect. The most widely used grading scale assessing the clinical severity of SAH is the five-step Hunt and Hess scale. The Fisher grade describes the amount of blood on a noncontrast computed tomography (CT) study of the head. Both scales correlate with patient outcome. Delayed sequelae of SAH are hydrocephalus (10–20% of cases), vasospasm, brain infarction (due to direct pressure from space-occupying lesion, vessel compression from brain shift, severe diffuse vasospasm, or systemic hypotension, decreased cardiac output, hypoxia), hemorrhage, and seizures (15%).

Imaging Basically, three major modalities are used to for the diagnosis and follow-up of intracranial aneurysms. CT angiography (CTA), today used as a thin-section bolustracking technique with an excellent spatial submillimeter resolution, magnetic resonance angiography (MRA) either as time of flight (TOF) or as a contrast-enhanced technique

A

76

Aneurysm, Intracranial

(CE-MRA), and finally catheter angiography. MRA or CTA is the preferred initial methods for evaluation of unruptured intracranial aneurysms, whereas digital subtraction angiography (DSA) is the preferred modality in patients with an acute SAH. The ability of CT to reveal SAH in the acute phase is approximately 100%, and therefore CT is the standard method of choice (Fig. 1). Currently, CT involves multidetector techniques (MDCT) that allow fast scanning covering large field of views in submillimeter spatial resolution. CTA using a bolus-tracking technique after intravenous injection of contrast agent in combination with MDCT has the potential to even detect small aneurysms (Fig. 2). CTA and nonenhanced CT (NECT) may show calcifications in the neck of an aneurysm, which provides important information for the surgeon.

DSA still remains the gold standard for detecting and assessing the size, morphology, and location of an intracranial aneurysm. Three-dimensional rotational angiography contributes fundamentally in the understanding of aneurysm anatomy, even in complex situations. This technique allows accurate depiction of the aneurysm morphology and helps in the planning of treatment strategies. Generally, the aim of DSA in SAH is to identify the presence or absence of any aneurysms, to delineate the relationship of an aneurysm to its parent vessel and adjacent penetrating branches, to assess the collateral circulation, to depict vasospasm, and to determine treatment modalities. In approximately 15% of patients with SAH, no aneurysm is found. An important contributor of this group is nonaneurysmal perimesencephalic SAH, where bleeding on a CTscan or MR image is localized anterior to the brain stem and adjacent areas such as the interpeduncular fossa and ambient cisterns. Aneurysm appearance in magnetic resonance imaging (MRI) is highly variable and may be quite complex. The signal depends on the presence, direction, and flow rate, as well as the presence of clot, fibrosis, and calcification within the aneurysm itself. Currently, the two major MRA techniques in the depiction of cerebral aneurysms are TOF or contrast-enhanced (CE-)MRA. TOF is generally used as a multislab three-dimensional technique. This sequence delineates the parent artery and depicts the size and orientation of an aneurysm dome and neck and is well suited for follow-up examinations after coiling. Aneurysms smaller than 3 mm may be difficult to detect, because TOF is a flow-dependant technique and flow in small aneurysms is low.

Treatment Aneurysm, Intracranial. Figure 1 NECT of acute SAH. Basal cisterns show hyperdense acute bleeding along vessel course.

The two treatment methods are surgery (clipping or wrapping) or endovascular coiling.

Aneurysm, Intracranial. Figure 2 (a, b) CTA of an AcomA aneurysm. Three-dimensional MIP reconstruction (a), and transverse (b) source image of a CTA, all depicting the pathology.


77

A

Aneurysm, Intracranial. Figure 3 (a–c) DSA of an AcomA aneurysm that was treated endovascularly using coils (b). Complete obliteration is shown on conventional DSA and TOF-MRA images (c).

Aneurysm, Intracranial. Figure 4 (a–d) DSA of an aneurysm of MCA bifurcation. Incomplete coiling (b, arrow) was spontaneously abolished in the control DSA after 6 months (c). Residual flow in the cavity of the MCA aneurysm is shown on a TOF source image as central white spot (d).

78


Aneurysm, Intracranial. Figure 5 (a–e) Broad based aneurysm of the basilar tip in DSA (a) and three-dimensional reconstruction (b) from a rotational angiography series. Both posterior communication arteries (PCAs) are embraced in the neck of the aneurysm. Y-shaped implementation of two intracranial stents, reaching from the basilar trunk to the PCA on both sides (c, d). Complete coiling of the basilar tip aneurysm obtaining patency of the right and left PCA (e).


The goal of surgical treatment is usually to place a clip across the neck of the aneurysm to exclude the aneurysm from the circulation without occluding brain-supplying vessels. MRI-compatible clips are manufactured in various types, shapes, sizes, and lengths. When the aneurysm cannot be clipped, wrapping is another choice that aims to protect the aneurysm sac to prevent bleeding. Wrapping can be performed with cotton or muslin, with muscle, or with plastic or other polymer. The operative morbidity and mortality associated with clipping depends on whether the aneurysm has ruptured; surgery of ruptured aneurysms is more difficult and therefore morbidity is higher. The risk of surgery for unruptured aneurysms is estimated at 4–10.9% morbidity and 1–3% mortality. During the past 15 years, endovascular methods have been developed and refined to treat intracranial aneurysms. Initially, endovascular balloon occlusion of a feeding artery was performed. This procedure was soon followed by direct obliteration of the aneurysmal lumen, first by detachable balloons and later by microcoils, first described by Guglielmi and colleagues (Figs. 3 and 4). They used detachable platinum microcoils that were placed in intracranial aneurysms. These days, coiling has become the primary treatment modality for aneurysms in many centers. It can be used to treat most aneurysms. Former limitations, such as aneurysms with wide necks or complex morphologies and high rates of recurrence secondary to coil compaction, have been addressed with complex shaped coils, balloon (“remodeling”) and stent technology, and biologically active coils (Fig. 5). The purpose of the coil is to induce thrombosis at the site of deployment via electrothrombosis. Newer

79

biologically active coils are coated with various substances to enhance permanency of the thrombus within the coiled aneurysm by permitting a denser packing or engendering a tissue response at the neck of the aneurysm that decreases blood flow into the aneurysm and subsequent recanalization. For the embolization procedure, a guiding catheter is placed in the cervical internal carotid or VA. Microcatheters of varying sizes can then be navigated into the aneurysm cavity using road-mapping technique. Coils of decreasing sizes are delivered into the aneurysm cavity and electrolytically detached. This process is continued until maximal angiographic obliteration of the aneurysm cavity is achieved. Estimated risks associated with coiling based on several large studies are on the order of 3.7–5.3% morbidity and 1.1–1.5% mortality. One of the major drawbacks associated with coiling is that, over time, the coils can compact, leading to reopening or recanalization of the aneurysm. A reopened aneurysm has a certain risk of hemorrhage that is believed to be low. Newer technologic advances such as biologically active coils and stents designed to prevent recanalization are currently developed and in clinical use. Severe spasms can be treated with intraarterially administered nimodipine or papaverine (Fig. 6). Increased safety of coiling over clipping was demonstrated in the International Subarachnoid Aneurysm Trial (ISAT). The study was stopped prematurely after a planned interim analysis found a 23.7% rate of dependency or death in the coiling cohort versus a 30.6% rate in the clipping cohort. Continued follow-up from the ISAT study has recently been published: the original results were confirmed, with a higher rate of seizures in the

Aneurysm, Intracranial. Figure 6 (a–c) Severe vasospasms of the right A1, M1, and distal ICA (a). Partially resolved spasms after superselective intra-arterial administration of papaverine and nimodipine (b). Percutaneous transfemoral angioplasty (PTA) was finally performed in the A1 and M1 segments (c), resulting in restoration of approximate original vessel diameters.

A

80

Aneurysm, Splenic Artery

clipping group and a slightly higher rate of rebleeding in the coiling group. Ultimately, the decision to clip or coil should be made on an individual basis and may often involve difficultto-quantify variables such as patient interest in one technique over the other or the experience or availability of physician operators.

Aneurysmal Bone Cyst ABC is a nonneoplastic bone lesion of uncertain origin capable of marked expansion and rapid bone destruction and characterized by multiple blood-filled channels or spaces separated by fibrous septa. ▶Neoplasm-Like Lesions, Bone

Bibliography 1.

2.

3.

4.

Molyneux A, Kerr R, Stratton I et al (2002) International subarachnoid aneurysm trial (ISAT) of neurosurgical clipping versus endovascular coiling in 2143 patients with ruptured intracranial aneurysms: a randomised trial. Lancet 360:1267–1274 Wanke I, Egelhof T, Dorfler A et al (2003) [intracranial aneurysms: Pathogenesis, rupture risk, treatment options]. Rofo Fortschr Geb Rontgenstr Neuen Bildgeb Verfahr 175:1064–1070 Wiebers DO, Whisnant JP, Huston J et al (2003) Unruptured intracranial aneurysms: natural history, clinical outcome, and risks of surgical and endovascular treatment. Lancet 362:103–110 ISUIA (1998) Unruptured intracranial aneurysms—risk of rupture and risks of surgical intervention. International study of unruptured intracranial aneurysms investigators. N Engl J Med 339:1725–1733

Aneurysms, Cerebral JY G AUVRIT Neuroradiology Department Salengro Hospital Lille France [email protected]

Definitions

Aneurysm, Splenic Artery The splenic artery is the third most common site of intraabdominal aneurysm formation after the abdominal aorta and iliac arteries, representing approximately 60% of all visceral arterial aneurysms. It is usually caused by atherosclerosis; however, it may also have congenital or infective origins, and a significant relation with pregnancy, multiparity, and portal hypertension have been reported. The majority of splenic artery aneurysms are single and saccular in shape, and are located in the middle and distal parts of the splenic artery. In most cases, splenic artery aneurysm is asymptomatic although rupture of the aneurysm may occur in 5–10% of cases leading to massive intraperitoneal hemorrhage. On US examination, splenic artery aneurysm appears as a hypoechoic mass in the left upper part of the abdomen with an arterial waveform at Doppler evaluation. On CT scans, splenic artery aneurysm appears as welldefined low-density mass with or without calcifications and an intense enhancement within the residual patent lumen is observed after the administration of intravenous contrast medium. Moreover, both CT angiography and MR angiography allow an accurate evaluation of the aneurysmal site and the size, while selective splenic angiography and subsequent embolization may be considered in lesions larger than 2 cm. ▶Transplantation, Hepatic

Abnormal outpouching of an intracranial blood vessel wall. Aneurysms develop during the course of life. They are almost never found in neonates and they are also rare in children (1). The exact prevalence of unruptured aneurysms is unknown but autopsy series estimate the prevalence at 1.3 to 7.9%, the prevalence tending to increase with age. The morphological appearance is saccular or fusiform, the most common type being the saccular (‘berry’) aneurysm, constituting about 90% of all intracranial aneurysms. Fusiform aneurysms account for between 5 and 10% of cases. The latter are characterised by the absence of a defined neck, circumferential involvement of the parent artery and a longish course. Aneurysms are located intradurally or extradurally. Intradural aneurysms most often involve the bifurcation of vessels at the base of the brain, especially the anterior and middle cerebral artery, distal internal carotid artery and the vertebrobasilar trunk. Anterior circulation aneurysms constitute about 85% of cases. They can occur as solitary (70%) or multiple aneurysms (30%). While the majority of aneurysms are small in size (6 mm) but lower rates for smaller aneurysms, with detection rates averaging 60%. However, the combined evaluation of axial source images and MIP reconstructions increases the sensitivity of detection in comparison with MIP reconstructions alone (95 versus 75%). MR angiography is safe and reasonably sensitive (90%), which makes it well suited to screening people at risk of intravascular aneurysms (Fig. 2). Another excellent indication for MR angiography is the follow-up of patients who have undergone endovascular treatment. It provides an effective way of detecting aneurysm recanalisation requiring retreatment. DSA is the method of reference, but this technique is invasive, limiting its use in clinical practice. Contrast-enhanced (CE) MR angiography, the reference technique for extracranial carotid disease, can be performed to image intracranial vessels and to follow-up aneurysms treated with Guglielmi detachable coils (4).

Digital Subtraction Angiography DSA is the gold standard and is usually performed for the detection of a cerebral aneurysm because of its inherent high-spatial resolution. Due the frequency of multiple aneurysms, complete four-vessel angiography is necessary. DSA can localize the lesion, reveal aneurysm shape and geometry, determine the presence of multiple aneurysms, define vascular anatomy and collateral circulation and assess the presence and degree of vasospasm. Using rotational angiography, multiple oblique views are obtained as the source for 3D reconstructions, thereby further increasing the performance of DSA (Fig. 3). Several 2D working views are performed after the 3D reconstructions and are used during the surgical or endovascular procedure (5). However, catheter angiography is not an entirely innocuous procedure. A systematic review found a transient or permanent neurological complication rate of 1.8%. The aneurysm may re-rupture during the procedure, as occurs in 1 to 2% of cases.

References 1.

2.

3.

4.

5.

Rinkel GJ, Djibuti M, Algra A et al (1998) Prevalence and risk of rupture of intracranial aneurysms: a systematic review. Stroke 29:251–256 Wiebers DO, Whisnant JP, Huston J 3rd et al (2003) Unruptured intracranial aneurysms: natural history, clinical outcome, and risks of surgical and endovascular treatment. Lancet 362:103–110 Teksam M, McKinney A, Casey S et al (2004) Multi-section CT angiography for detection of cerebral aneurysms. AJNR Am J Neuroradiol 25:1485–1492 Farb RI, Nag S, Scott JN et al (2005) Surveillance of intracranial aneurysms treated with detachable coils: a comparison of MRA techniques. Neuroradiology 507–515 Anxionnat R, Bracard S, Ducrocq X et al (2001) Intracranial aneurysms: clinical value of 3D digital subtraction angiography in the therapeutic decision and endovascular treatment. Radiology 218:799–808

84

Angel Wings Sign, Spinnaker Sign

Angel Wings Sign, Spinnaker Sign In younger persons, the thymus may be outlined by air and this finding is specific for pneumomediastinum. These signs have been suggested as descriptive terms. ▶Pneumomediastinum

Angina Pectoris Angina pectoris is the name for a clinical syndrome due to myocardial ischemia experienced as pain or tension in the middle of the chest. Radiation into the jaw and, especially, the left arm may occur. ▶Ischemic Heart Disease, Nuclear Medicine

elements, and fat. It can occur as a solitary mass or as multiple lesions in the case of tuberous sclerosis. This tumor usually presents a combination of fat and soft tissue. Imaging features of angiomyolipoma are related to the distribution and amount of fat and to the relative proportion of its three histological components. ▶Lipomatous Neoplasms, Hepatic

Angioplasty Or percutaneous transluminal angioplasty (PTA) is the dilation of a vessel with the use of a balloon catheter. ▶Stroke, Interventional Radiology

Angiosarcoma Angiodysplasia ▶Hepatic Sarcoma Acquired vascular malformation of the bowel wall. ▶Vascular Disorders of the Gastrointestinal Tract

Angiosarcoma, Hepatic Angiogenesis The formation of new blood vessels. ▶Perfusion, Neoplasms

Angiomyolipoma ▶Lipomatous Neoplasms, Hepatic

Angiomyolipoma, Hepatic Very uncommon benign mesenchymal tumor composed of a variable amount of proliferating blood vessels, muscle

Primary malignancy of the liver occurring in adults, with a male predominance, arising from vascular endothelial cells of the liver. This form although rare represents the most common primary mesenchymal malignancy of the liver. Tumor onset seems to be related to chronic exposure to toxic agents such as inorganic arsenic and vinyl chloride or to long-term irradiation with thorium oxide. In some cases an association with hemocromatosis, von Recklingausen’s disease, alcoholic cirrhosis and anabolic, steroid intake has been described. Tumor size may vary from a few millimeters to several centimeters. The presence of internal hemorrhage determines the redbrown appearance of the nodules. Larger tumors are usually not capsulated and may contain cystic areas with blood debris filling. This tumor is composed of malignant endothelial cells organized to form vessels that may range from abortive or cavernous forms to structured, frequently dilated sinusoids. Early metastatic spreading to lungs and spleen is common. ▶Hepatic Sarcoma

Anomalies of the Cerebral Commissures

Aniridia

85

Annular Fissures A

Congential absence of the retina of the eye. ▶Neoplasms, Kidney, Childhood

▶Degenerative Conditions, Spine

Annular Pancreas Anismus Anismus is an abnormal activity of pelvic floor musculature that results in an outlet obstruction characterized by difficulties in rectal evacuation. ▶Pelvic Floor Dysfunction, Anorectal Manifestations

Anismus - spastic pelvic floor syndrome - diskinetic puborectalis muscle - pelvic floor dyssynergia ▶Pelvic Floor Dysfunction, Anorectal Manifestations

Ankylosis Loss of motion until complete immobility of a joint is referred to as ankylosis and can be due to alterations around (“false”) or in the joint itself (“true”). When there is bony bridging across the joint space and complete immobility, it is termed “osseous” or “complete.” Ankylosis can also be applied to an osseous junction between neighboring bones. ▶Dish

The annular pancreas represents a congenital anomaly characterized by a ring of normal pancreatic tissue that arises from the head of the pancreas encircling the descending portion of the duodenum. ▶Congenital Abnormalities, Pancreatic ▶Congenital Anomalies of the Pancreas

Annular Tears The tears in the annulus fibrosus caused by aging, acute or repetitive trauma or overloading. There are three types of fissures: concentric, transverse, and radial. Annular fissures are present in almost all individuals over 40 and in virtually all bulging disks. Therefore, they could be considered as paraphysiological, however some of them result in disk herniation. Annular tears can be visualized on both sagittal and axial planes on T2-weighted and postcontrast T1-weighted MR images as bands of increased signal (high intensity zones – HIZ) in the annulus fibrosus. ▶Degenerative Conditions of the Spine

Anomalies of Cortical Development ▶Gyration Disorders, Cerebral

Annexin V A 36-kDa protein, which binds in the presence of Ca2+ to externalized phosphatidyl serine (PS). ▶Apoptosis

Anomalies of the Cerebral Commissures ▶Congenital Malformations, Cerebrum

86

Anomalous Termination of Bile Ducts

Anomalous Termination of Bile Ducts Anomalous termination of the hepatic ducts into the gallbladder or anomalous end of the common bile duct into the pylorus, stomach, pancreatic duct. In the second case the refluxes of gastric contain may appear. ▶Congenital Malformations, Liver and Biliary Tract

Anophthalmia ▶Congenital Malformations, Orbit

3. Imperforate anus: Terminal bowel ends blindly and no fistula exists. 4. Anal or rectal stenosis: Mildest form, represents cases of incomplete anal or rectal atresia. Wingspread workshop classification of anorectal malformations (ARM) (2). Basically, high, intermediate, and low groups are distinguished as well as male and female subgroups: 1. ▶High ARM: The blind pouch ends above the sling of the hypotrophic puborectalis muscle. There can be associated rectobulbar fistulas in males or rectovaginal fistulas in females. 2. ▶Intermediate ARM: The rectal pouch enters the sling of the puborectalis muscle; depending on the gender, either rectobulbar or rectovaginal fistulas can exist. 3. ▶Low ARM: The rectum passes through a welldeveloped puborectalis muscle sling. The aboral ending is an anocutaneous fistula in males, whereas in females this is represented by an anocutaneous or anovestibular fistula.

Department of Radiology, Division of Pediatric Radiology, University Hospital Graz, Graz, Austria [email protected]

Female cloacae were placed in a separate group, because they may be considered high, intermediate, or low depending on the length of the common channel. Regarding outcome it is clear that results are best in the low ARM group, where associated malformations occur less frequently. In the other forms, results are less satisfying and malformations as well as urinary tract disorders are more frequent (2).

Synonyms

Associated Anomalies

Anal atresia; Imperforate anus; Rectal atresia—most of them are variants of anorectal malformations and therefore should not be used as synonyms

Associated malformations can be found in about 20–70% of patients with ARMs (2). Ho¨llwarth et al studied these associated malformations in 75 patients, finding an overall incidence of 72% (details are given in Table 1).

Anorectal Malformation E RICH S ORANTIN

Definition Several definitions exist, the two most frequent are listed later. Gans classification (1): 1. Rectal atresia: The anus is open and a variable segment of the rectum superior to the anus is atretic, no fistula is present. 2. Ectopic anus: Most common abnormality of the anorectal segment. It occurs when the terminal bowel fails to descend normally, resulting in a lack of communication with the anus and in an abnormal bowel opening via a fistula (perineal, vestibule, vagina, urethra, bladder, or cloaca).

Anorectal Malformation. Table 1 malformations in 75 children

ARM associated

Anatomical System

Total in %

Skeleton Urinary tract Cardiac system Intestinal tract Cerebral Genitalia Others

46.6 41.3 18.6 18.6 13.3 12.0 22.6

Source: Ho¨llwarth ME, Sorantin E (2001) Urinary problems associated with imperforate anus. In: Fotter R (ed) Pediatric Uroradiology, Chapter 6, Springer, Berlin, pp 105–110

Anorectal Malformation

87

A

Anorectal Malformation. Figure 1 Male, 1 day old, imperforate anus and VACTERL association, plain films. (a) Left upper extremity—radial aplasia is shown. (b) Abdomen—sacral aplasia, severe vertebral malformation as well as a right-sided pelvic rib is presented.

As can be seen, frequently more than one organ system is involved. Boemers created the acronym ▶ARGUS for children suffering from the association of AnoRectal, GenitoUrinary, and Sacral anomalies (3). Another acronym is the so-called VATER or VACTERL association, combining Vertebral anomalies, Anorectal atresia, Cardiac anomalies, Tracheoesophageal fistula, Esophageal atresia, Renal dysplasia, and Limb malformation (Fig. 1) (2). Vertebral anomalies occur more frequently in the lumbar and sacrococcygeal spine. Partial or complete sacral agenesis is part of the caudal regression syndrome and is a strong indicator of neurogenic bladder dysfunction. Imperforate anus is also included in the Currarino triad, where it is found in combination with sacral dysplasia and an anterior sacral mass (e.g., teratoma, anterior meningomyelocele). As a general rule, associated malformations occur more frequently in the high ARM or intermediate ARM groups than in the low ARM group.

Urologic Problems The outcome of ARM patients is determined by the associated malformations, especially by the structural and functional anomalies of the urinary tract, as well as complications of surgical procedures (2).

Structural In males, rectourethral fistulas can be detected in about 80% of patients in the high ARM or intermediate ARM

Anorectal Malformation. Figure 2 Male, 8 months old, imperforate anus with rectourethral fistula, loopography, last image hold, lateral projection. Loopography is performed by instillation of the contrast medium into an existing colostomy. Solid, white arrow points to the rectourethral fistula, contrast medium is accumulating within the bladder (nonsolid white arrow).

groups, whereas rectovesical fistulas are only present in 8% (Fig. 2). Renal dysplasia, renal agenesis, renal ectopia, duplications, and hydronephrosis are the most common anomalies of the upper urinary tract. In addition, 47% of

88

Anorectal Malformation

patients in the high ARM group and 35% in the low ARM group exhibit a vesicoureteric reflux, caused be either malformations of the vesicoureteric junctions or more commonly by neurogenic bladder dysfunction (4). In the lower urinary tract hypospadias, epispadias and exstrophy, urethral diverticula, valves, strictures, or duplications can be found (2).

Functional The most important functional disorder is represented by any kind of neurogenic bladder, which is either related to the surgical reconstruction procedure or inherent to the associated sacral malformation. This was confirmed by Boemers et al, who investigated ARM patients with/ without sacral anomalies urodynamically. Patients with a normal-appearing sacrum or only minor dysplasia will exhibit a normal bladder and sphincter function in 98% of cases. In contrast, almost all patients with major sacral dysplasia will suffer from neurogenic bladder.

Embryology/Pathology Simply speaking, ARM are thought to represent the failure of the hindgut to descend properly.

Clinical Presentation Abdominal distension. No evidence of an anal opening on physical examination. Incidence is about 1 in 5,000 live births.

Imaging Recently, Boemers et al published guidelines for diagnostic screening and initial management in children with ARM (“ARGUS” protocol) (3). 1. Sacral anomalies: Plain films of the spine including the sacrum in two projections. Spinal ultrasound is mandatory in almost all patients where any kind of sacral dysplasia is found. Special care must be taken in patients with suspected Currarino triad not to overlook a presacral mass such as an anterior myelomeningocele. Magnetic resonance imaging (MRI) is indicated whenever spinal ultrasound depicts an abnormality. This MRI investigation should also serve as a baseline study for the future, where spinal ultrasound can no longer be performed due to increasing vertebral ossification (usually after the first half year, and definitely after the first year of life).

2. Urinary tract: Renal ultrasound and pelvic ultrasound should be performed as soon as possible. In the first days of the life it should be noted that due to the physiological oliguria in neonates, dilatations of the collecting system can be missed. Special attention should be given to the bladder. Pathological bladder wall thickness, trabeculation, a dilated posterior urethra in combination with frequent bladder neck openings and closings point to neurogenic bladder dysfunction with uninhibited detrusor contractions (3). Voiding cystourethrography (VCU) is recommended in all patients with upper urinary tract dilatation. Furthermore, VCU should be performed on male patients without a perineal bowel opening in order to detect a rectourethral fistula (2, 3). Functional assessment of the lower urinary tract by urodynamics is mandatory in all patients with sacral dysplasia during the first 3 months of life (3). Whenever VCU is necessary, a modified VCU technique, or better still video urodynamics (VUD), should be used (5). In boys with a rectourethral fistula and no perineal opening, for VUD the abdominal pressure can be recorded either by using an existing colostomy or by placing a microtip catheter in the stomach (2). Routine intravenous urography (IVU) can now be replaced in many cases by ultrasound and isotope studies. However, there is still a role for IVU whenever detailed anatomical information is necessary as well as in the postoperative care of urologic procedures. In the future, there will also be a role for MR urography. 3. Genitals: Ultrasound provides an excellent overview of the uterus and ovaries in newborn girls. A genitogram has to be performed on all newborn girls with persistent cloaca (3). 4. CNS: Cranial ultrasound is part of the neonatal workup whenever a malformation is discovered; MRI should be performed electively. 5. Heart: Chest films are taken routinely in the preoperative phase. Whenever there is a cardiac abnormality or a heart murmur further work-up is indicated.

Nuclear Medicine As already mentioned, renal isotope studies can be used to examine split renal function and urine drainage.

Diagnosis The diagnosis is made clinically. The associated malformations are diagnosed by imaging and further clinical

Aorta and Large Vessel Disease, MRI

work-up such as endoscopy. In case of complex syndromes, genetic counseling is mandatory. ▶EI Tract, Pediatric, Congenital Malformations


Bibliography

R OSSELLA FATTORI

1.

Department of Radiology, Cardiovascular Unit, University Hospital S. Orsola, Bologna, Italy [email protected]

2.

3.

4.

5.

Berrocal T, Lamas M, Gutierrez J et al (1999) Congenital anomalies of the small intestine, Colon and rectum. Radiographics 19:1219–1236 Ho¨llwarth ME, Sorantin E (2001) Urinary problems associated with imperforate anus. In: Fotter R (ed) Pediatric Uroradiology, Chapter 6, Springer, Berlin, pp 105–110 Boemers T, Beek F, Bax N (1999) Guidelines for the urological screening and initial management of lower urinary tract dysfunction in children with anorectal malformations the ARGUS protocol. BJU Int 83:662–671 Ralph D, Woodhouse C, Ransley P (1992) The management of the neuropathic bladder in adolescents with imperforate anus. J Urol 148:366–368 Sorantin E, Lindbichler F, Fotter R (2005) Imaging in Children, 2nd edn, vol 1, Chapter 4.16, Elsevier Churchill Livingstone, Edinburgh, pp 863–869

Anorectal Malformation ▶GI Tract, Paediatric, Congenital Malformations

Anterior Mediastinum The narrow region between the pericardium and the sternum containing the thymus or its remnants, some lymph nodes and vessels and branches of the internal thoracic artery. ▶Neoplasms of the Chest in Childhood

Antibodies Proteins used by the immune system to identify and neutralize foreign objects. Each antibody recognizes a specific binding site, called an antigen. Five different isotypes are known: IgA, IgD, IgE, IgG, and IgM. For imaging and therapeutic purposes in medicine, IgG is the relevant isotype. ▶Receptor Studies, Neoplasms

89

Magnetic Resonance Imaging Techniques for Studying the Thoracic Aorta Spin-Echo MRI Spin-echo (SE) T1-weighted imaging provides the best anatomic detail of the aortic wall and of pathologic conditions, such as atheromatous plaques, intimal flaps, and intramural hemorrhage, and is still the basis of any aortic study (1), whereas T2-weighted images (TR = 2-3 R-R’; TE, 80–100 msec) can be used in tissue characterization of the aortic wall or blood components. A superior black-blood effect is achieved by using preparatory pulses such as presaturation, dephasing gradients, and preinversion, with one or more additional RF pulses outside the plane to suppress the signal intensity of inflowing blood and to nullify the blood signal (black-blood SE sequences).

Gradient-Echo MRI Gradient-echo (GE) techniques provide dynamic and functional information, although with fewer details of the vessel wall. The bright signal of the blood pool on GE images results from flow-related enhancement obtained by applying radio-frequency pulses to saturate a volume of tissue. With a short TR (4–8 msec) and low flip angle (20–30˚), maximal signal is emitted by blood flowing in the voxel, with a high degree of temporal resolution throughout the cardiac cycle (up to 20–25 frames).

Flow Mapping Accurate quantitative information on blood flow is obtained from modified GE sequences with parameter reconstruction from the phase rather than the amplitude of the MR signal. Vector mapping has been used to describe flow patterns in different aortic diseases (including hypertension, aneurysms, dissection, Marfan syndrome, and coarctation). MR maps of flow velocity are obtained twodimensionally, which is particularly important in profiles of nonuniform flow, such as that in the great vessels (2).

A

90


MR Angiography

Intramural Hematoma

Three-dimensional (3D) contrast-enhanced MR angiography (MRA) constitutes the most common angiographic method for evaluation of the great vessels (3). The technique relies on the contrast-induced T1-shortening effects of the contrast medium, and saturation problems with slow flow or turbulence-induced signal voids are avoided. During the short intravascular phase, the paramagnetic contrast agent provides signal in the arterial or venous system, enhancing the vessel-to-background contrast-to-noise ratio irrespective of flow patterns and velocity. Pulsatility artifacts are minimized, even in the ascending aorta and without electrocardiographic gating. With the support of maximum intensity projection (MIP) images and of 3D multiplanar reformation (MPR), this technique delineates all the morphologic details of the aorta and its side branches in any plane in a 3D format.

Intramural hematoma (IMH) derives from spontaneous rupture of the aortic vasa vasorum of the media layer as the initiating process, which is confined in the aortic wall without intimal tear. This results in a circumferentially oriented blood-containing space seen on tomographic imaging studies. IMH may occur spontaneously or as a consequence of penetrating aortic ulcer in intrinsically diseased media. T1-weighted images reveal a crescent-shaped area of abnormal signal intensity within the aortic wall. In the acute phase (0–7 days after the onset of symptoms), oxyhemoglobin shows intermediate signal intensity on T1 SE images and high signal on T2 SE, whereas in the subacute phase (>8 days) methemoglobin shows high signal intensity in both T1 and T2 SE images (Fig. 1). The progression of IMH to overt dissection and rupture has been reported in 32% of cases, in particular with involvement of ascending aorta.

Acquired Aortic Disease Aortic Dissection Aortic dissection is characterized by a laceration of the aortic intima and inner layer of the aortic media that allows blood to course through a false lumen in the outer third of the media extending from the ascending aorta to the iliac arteries (type A) or involving the descending aortic segments exclusively (type B). Using a magnetic resonance imaging (MRI) SE blackblood sequence in the axial plane, the intimal flap is detected as a straight linear image inside the aortic lumen. The true lumen can be differentiated from the false by the anatomic features and flow pattern: the true lumen shows a signal void, whereas the false lumen has higher signal intensity. In addition, visualization of remnants of the dissected media as cobwebs adjacent to the outer wall of the lumen may help identify the false lumen. A detailed anatomic map of aortic dissection must indicate the type and extension of dissection and also distinguish the origin and perfusion of branch vessels (arch branches, celiac, superior mesenteric, renal arteries, and coronary arteries) from the true or false channels. High signal intensity of effusion (pericardial, periaortic, or pleural) indicates a bloody component and is considered a sign of impending rupture of the ascending aorta into the pericardial space or of the descending aorta into the mediastinal or retroperitoneal space. In stable patients, adjunctive GE sequences or phase contrast images can be instrumental in identifying aortic insufficiency in type A dissection, entry, or re-entry sites as well as in differentiating slow flow from thrombus in the false lumen. Gadolinium-enhanced 3D MRA enhances visualization of the intimal flap and its relationship with aortic vessels in a 3D format.

Aortic Ulcer Aortic ulcer is characterized by rupture of the atheromatous plaque, disrupting the internal elastic lamina. The MRI diagnosis of aortic ulcer is based on visualization of a crater-like ulcer located in the aortic wall. Mural thickening with high or intermediate signal intensity on SE sequences may indicate extension of the ulcer into the media and formation of an intramural hematoma. MRA is particularly suitable for depicting aortic ulcers along with the irregular aortic wall profile seen in diffuse atherosclerotic involvement. The aortic ulcer is easily recognized as a contrast-filled outpouching of variable extent with jagged edges, which may result even in large pseudoaneurysm. The disadvantage of MRI with respect to computed tomography (CT) is failure to visualize dislodgment of the intimal calcifications that are frequently observed in aortic ulcers.

Aortic Trauma A long examination time as well as difficult access to the patient has been considered the main limitation of MRI in acute aortic pathology. The development of fast MRI techniques has shortened the examination time to a few minutes, thus MRI can be used even in critically ill patients. On SE images in the sagittal plane, a longitudinal visualization of the thoracic aorta makes it possible to distinguish a partial lesion (a tear limited to the anterior or the posterior wall) from a lesion encompassing the entire aortic circumference. The presence of periadventitial hematoma and/or pleural and mediastinal hemorrhagic effusion may also be considered a sign of instability (Fig. 2). In the same sequence used to evaluate the aortic


91

A

Aorta and Large Vessel Disease, MRI. Figure 1 MRI study of type A intramural hematoma involving the ascending and descending aorta performed few days (a, b) and two months (c) after symptoms onset. Black-blood spin-echo T1 image (a) showing a high signal intensity of the hematoma, which increases in black-blood spin-echo T2 (b), due to the presence of methemoglobin in the subacute phase. Two months later the hematoma is completely reabsorbed (c).

Aorta and Large Vessel Disease, MRI. Figure 2 Traumatic aortic injury before (a) and after endovascular stent-graft treatment (b) SE sagittal MR image of the traumatic injury providing a diverticular aneurysm with mediastinal and periaortic hematoma along the descending aortic segment. MRA after stent-graft treatment (b) the lesion is covered by the stent-graft; the metallic springs are visible like artifacts.

lesion, without the need for any additional time, the wide field of view of MRI provides a comprehensive evaluation of chest trauma, such as lung contusion and edema, pleural effusion, and rib fractures.

Thoracic Aortic Aneurysms MRI is effective in identifying and characterizing thoracic and abdominal aortic aneurysms. SE sequences are helpful in evaluating alterations of the aortic wall and periaortic

92


space. Periaortic hematoma and areas of high signal intensity within the thrombus may indicate instability of the aneurysm and are well depicted on SE images. Atherosclerotic lesions are visualized as areas of increased thickness with high signal intensity and irregular profiles. With fat suppression technique, the outer wall of the aneurysm can be easily distinguished on MR images by the periadventitial fat tissue, hence the aneurysm diameter can be accurately measured. The high level of reproducibility of MRI measurements ensures optimal reliability in monitoring expansion rate. MRA may play an important role in preoperative evaluation. Contrastenhanced 3D MRA can provide precise topographic information about the extent of an aneurysm and its relationship to the aortic branches. The capability of contrast MRA to visualize the artery of Adamkiewicz represents an important advance in planning the surgical repair of a thoracic aneurysm.

Aortitis Contrast-enhanced T1- and T2-weighted SE MRI has been shown to be highly effective in the evaluation of Takayasu arteritis even in the early phases, providing important information on the activity of the disease. Active inflammatory disease appears as variable thickening of the aortic wall, enhanced after gadolinium administration. The chronic, quiescent stage is characterized by extensive perivascular fibrosis without contrast enhancement. MRA can replace invasive angiography in the study of aortic and branch vessel stenosis, and avoidance of angiography is highly desirable in patients affected by aortitis because it carries a risk of pseudoaneurysm formation at the site of arterial puncture.

Congenital Aortic Disease Aortic Arch Anomalies Aortic arch anomalies result either from abnormal regression of an embryonic arch that normally remains patent or from persistent patency of a structure that normally regresses during fetal life. Usually, SE MRI in axial or coronal and sagittal planes and thin slice thickness provides excellent visualization of the vascular structure and their relationship with mediastinal organs, whereas MRA is useful for defining the complex anatomy of the arch and supraaortic vessels (4). An aberrant right subclavian artery (arteria lusoria) is the most common type of vascular anomaly. The right subclavian artery may also arise from an outpouching, known as the diverticulum of Kommerell, which represents persistence of the most distal portion of the embryonic right arch. On MRI SE or GE sequences, the aberrant right

subclavian artery is defined as a tubular structure crossing the mediastinum from right to left, behind the trachea and the esophagus, in an oblique ascending direction. An extrinsic compression of the esophagus, as typically shown on barium swallow, can also be seen on axial and sagittal SE MR images. A right aortic arch (0.1% of population) passes to the right of the trachea and may descend either to the right or the left of the thoracic spine. Two types of right aortic arch are described: right aortic arch with mirror image brachiocephalic branching, and right aortic arch with aberrant left subclavian artery. A double aortic arch is characterized by the presence of both a left and a right aortic arch; these arise from a branching of the ascending aorta, pass on both sides of the trachea and esophagus, and join posteriorly to form the descending aorta, which may lie to the right or left of the vertebral column. The luminal size of the two arches in relation to each other varies considerably, and one of them (usually the left) may be partially or completely atretic. The double aortic arch is a vascular ring that can produce severe symptoms if it compresses the trachea and esophagus. Cervical arch is a rare anomaly in which the aortic arch extends into the soft tissues of the neck before turning down on itself, forming the descending aorta.

Aortic Coarctation Coarctation is a common congenital anomaly in which an abnormal plication of the tunica media of the posterior aortic wall proximal to the ligamentum arteriosum causes a fibrous ridge to form, which protrudes into the aorta and causes an obstructive lesion. The stenotic segment can be focal (aortic coarctation), diffuse (hypoplastic aortic isthmus), or complete (aortic arch interruption). MRI axial and sagittal SE sequences are useful for quantify morphologic indexes of coarctation expressed as the ratio of diameter at the isthmus and above the diaphragm. However, although detection of anatomic narrowing of the aorta establishes the diagnosis of coarctation, an assessment of its clinical significance depends on determining its hemodynamic effects. GE MRI has been applied to evaluate flow turbulence across the coarctation; the severity of coarctation is quantified on the basis of the length of flow void. Further functional information can be provided by MR flow mapping, which can define the severity of the stenosis by measuring velocity jets at the level of coarctation and mean flow deceleration in the descending aorta. With this technique, it is possible to predict the coarctation severity with good sensitivity and specificity (95 and 82%, respectively) compared to catheter angiography. Flow mapping is also able to quantify the flow pattern and volume of collateral flow in the descending

Aortic Dissection

93

A

Aorta and Large Vessel Disease, MRI. Figure 3 MRA of aortic coarctation. The stenotic isthmic aortic segment as well as the collateral circulation throughout the intercostal arteries is well visible both in MIP (a) and VR reconstruction (b).

aorta, which are other important parameters of the severity of coarctation, and this information may be crucial in the choice of surgical strategy. Finally, MRA provides excellent visualization of the stenotic segment and of the collateral circulation (Fig. 3).

Aneurysms of the Valsalva Sinus Aneurysm of the Valsalva sinus is a rare congenital anomaly of the structural layers of the aortic wall and is characterized by the absence of the medial layer. This abnormality is usually limited to one of the Valsalva sinuses, most frequently the right coronary one. Because of the absence of the elastic components of the medial layer, the Valsalva sinus is asymmetrically dilated, even in the neonate. The aortic abnormality is visible on SE and MRA images. Rupture of sinus aneurysm is usually into the right atrium and creates a left-to-right shunt, which can be visualized by GE sequences.

Bibliography 1. 2. 3.

4.

Reddy GP, Higgins CB (2000) MR imaging of the thoracic aorta. Magn Reson Imaging Clin N Am 8(1):1–15 Debatin JF, Hany TF (1998) MR-based assessment of vascular morphology and function. Eur Radiol 8:528–539 Krinsky G, Rofsky N, Flyer M et al (1996) Gadolinium-enhanced three dimensional MR angiography of acquired arch vessels disease. Am J Roentgenol 167:981–987 Russo V, Renzulli M, La Palombara C et al (2005) Congenital diseases of the thoracic aorta. Role of MRI and MRA. Eur Radiol 26:1–9

Aorta, Grafts, and Prostheses The aorta is the largest artery in the body. A graft is used in vascular procedures to patch blood vessels, to line aneurysms, or to bypass blood across blocked blood vessels. A prosthesis is an artificial device used to mimic another device or part of the body. Prostheses are often placed within a part of the body. A stent graft is an example of a prosthesis. Prostheses are often used during interventional procedures. ▶Aneurysm, Aortic and Thoracic

Aortic Aneurysms An aortic aneurysm occurs when the aorta dilates to more than one and a half times its normal diameter. ▶Aneurysm, Aortic and Thoracic

Aortic Dissection Separation of the layers within the aortic wall. ▶Dissection, Aortic, Thoracic

94

Aortic Obstruction

Aortic Obstruction ▶Stenosis, Aortic, Abdominal

Aortic Syndrome ▶Stenosis, Aortic, Abdominal

Apert Syndrome Apert syndrome (acrocephalosyndactyly type I) is characterised by bilateral coronal synostosis with brachycephaly, widened metopic and sagittal sutures, hypertelorism, shallow orbits with proptosis, maxillary hypoplasia with downturned mouth and severe symmetric syndactylism of the hands and feet. The central nervous system may demonstrate megalocephaly, giral abnormalities, hypoplastic white matter, heterotopic gray matter, frontal encephalocele, corpus callosal agenesis and/ or ventriculomegaly. Choanal stenosis, cleft palate, cervical spine fusion (usually C5 and C6), otitis media, Eustachian tube dysfunction and conductive hearing loss are common. Concurrent problems with the cardiovascular, genitourinary, gastrointestinal and respiratory systems may contribute to morbidity. There may be ankylosis of the elbows, hips and shoulders. The stylohyoid ligament calcifies in 38–88% of patients. ▶Congenital Malformations, Nose and Paranasal Sinus

Apocrine Metaplasia Histologic lesion that characterizes areas of cystic change in the breast. ▶Fibrocystic Disease, Breast

Apophyseal Joint Arthritis Apophyseal joint arthritis is seen in rheumatic disease. In spondylarthropathies the lumbar spine is most commonly

affected, whereas in rheumatoid arthritis it is virtually confined to the cervical spine. In spondylarthropathies, late stages of apophyseal joint arthritis often show postarthritic ankylosis, in contrast to rheumatoid arthritis, in which postarthritic degenerative osteoarthritis is common. ▶Rheumatoid Arthritis ▶Spondyloarthropathies, Seronegative

Apoptosis M ICHAEL E ISENHUT, WALTER M IER , U WE H ABERKORN Abteilung Radiopharmazeutische Chemie Deutsches krebsforschungszentrum Heidelberg, Germany [email protected]

Definition The cell biological phenomenon termed apoptosis (programmed cell death) represents an energy-dependent elimination process for cells that have been injured, infected, or immunologically recognized as harmful or superfluous. This mechanism controls the fate and life expectancy of cells that play important roles in a large number of disorders such as myocardial infarction and neurodegenerative diseases or in transplanted organs. Apoptosis is also observed in cancer tissue, which might be enhanced after treatment with cytostatic drugs or ionizing radiation. This phenomenon was already described in the nineteenth century, but was termed “apoptosis” in 1972 by Kerr and colleagues: “A basic biological phenomenon with wide ranging implications in tissue kinetics” (1). Apoptosis is initiated by a series of events including receptor-mediated activation of the death-inducing signaling complex (DISC). As illustrated in Fig. 1, the main triggers of DISC are ▶FasL (CD95L), TRAIL, or tumor necrosis factor alpha (TNFa). Once activated through an autocatalytic cleavage/activation mechanism, the destruction of the cell starts off in the DISC releasing caspase 8, which triggers the activation of other downstream ▶caspases, the so-called executioner caspases. The 15 caspases known to date belong to the family of aspartylspecific cysteine proteases that are not only essential for the starting event in the DISC but also for the execution of apoptosis. Substrates of downstream-acting caspases are molecules involved in DNA repair, ribonucleoproteins, signaling molecules, structural proteins, and oncoproteins. Cleavage of structural proteins accounts for some of

Apoptosis

95

A

Apoptosis. Figure 1 Overview of the various pathways described in the Definition section.

the massive morphological changes such as membrane blebbing, nuclear fragmentation, and the formation of apoptotic bodies during apoptosis. However, caspase activation does not necessarily lead to apoptosis, but can occur even without inducing cell death. Cytostatic drugs are less well characterized with regard to the apoptosis-inducing mechanism. They act as DNAdamaging agents, antimetabolites, mitotic inhibitors, nucleotide analogs, or inhibitors of topoisomerases. Some drug-induced DNA damage is sensed by p53, suggesting that p53 may be considered a master switch for apoptosis. It is also suggested that anticancer drugs trigger CD95L/ CD95, TRAIL, and TNFa expression through p53 stimulation (2). The same should hold for the cytotoxic action of ionizing radiation, which injures the DNA by forming double-strand breaks. In this context, it is necessary to note that the hypothesis about tumor response to radiation is determined not only by tumor cell phenotype but also by microvascular sensitivity, because endothelial apoptosis regulates angiogenesis-dependent tumor growth (Garcia-Barros, et al, 2003). Another initiator of apoptosis is granzyme B, which assisted by the transmembrane pore-forming protein ▶perforin, penetrates into the cell where it shares the primary specificity of caspases to cleave at the carboxyl terminal of aspartate residues in their substrates. ▶Granzyme B

together with perforin is released by cytotoxic T cells upon specific interaction with the target cell and belongs, therefore, to the cellular immune response machinery. Mitochondria-associated proteins of the ▶Bcl-2 family play a major role in apoptosis induced by anticancer drugs and radiation therapy. While Bax and Bak promote cell death, Bcl-2 and Bcl-xL act against it. A third group of indirect promoters represent Bad, Bid, Bik, and Bim, so-called BH3-only proteins, docking to an extended hydrophobic groove on the prolife Bcl-2-like proteins, thereby abrogating the antiapoptotic functions of, for example, Bcl-xL. This leads to an imbalance in favor of Bax and Bak, which induce mitochondrial membrane permeabilization releasing among others ▶cytochrome c from the mitochondrion. Cytochrome c binds to apoptosis protease activating factor (Apaf-1) and aggregates to the heptameric apoptosome. Procaspase 9 dimerizes at the apoptosome scaffold, thereby activating itself. Active caspase 9 in turn activates downstream caspases 3, 6, and 7, the executioner caspases, leading to the vast degradation of the proteome. The mitochondria-involving intrinsic pathway of apoptosis is also active in the type II extrinsic pathway induced by the DISC. In this case, caspase 8 originating from the DISC cleaves Bid into its truncated form which then translocates into the mitochondrion where it is involved in the release of apoptogenic factors (vide supra).

96

Apoptosis

Pathology/Histopathology In normal tissues, cell proliferation and cell death are balanced. If this balance is disturbed by a variety of stimuli, too much or too little apoptosis is observed. Blood precursor cells, for example, need colony-stimulating factor (CSF) or direct contact with stromal cells to survive and differentiate. In the absence of these stimuli, the cells are eliminated via programmed cell death. The same happens to hormone-dependent organs, which develop atrophy through the absence of hormones. On the cellular level, various mechanisms of apoptosis lead to well-defined morphological changes in the cell. These include nuclear membrane breakdown, cytoskeletal reorganization, plasma-membrane blebbing, and loss of cell adhesion. The nucleus shrinks with condensation of chromatin, followed by disintegration of nuclear residues. This effect can be visualized by electrophoresis indicating a ladder pattern of fragmented DNA. Under the microscope, some of these effects might be visible. An often used indicator for DNA fragments is the terminal transferase dUTP nick end labeling (▶TUNEL Assay). This assay relies on the formation of nicks in DNA that can be identified by terminal transferase, an enzyme that catalyzes the addition of dUTPs that are secondarily labeled with a marker. In combination with immunocytochemistry or flow cytometry, this assay reproducibly labels cells in the final stages of apoptosis. Other indicators are artificial substrates for activated caspases that become fluorescent after cleavage and fluorescent ▶annexin V that binds in the presence of Ca2+ with externalized ▶phosphatidyl serine (PS). Processes leading to the activation of caspases have been described earlier, and the apoptosis-induced presentation of PS is a result of flippase inactivation responsible for the asymmetric lipid distribution in the double-layered cell membrane. In addition, scramblase promoting externalization is activated. This occurs during blood coagulation, thrombosis, and apoptotic cell death (vide infra).

Clinical Presentation The following diseases involve apoptosis: cancer, myocardial infarction (partly), viral infections, hematopoietic disorders, inflammation, and neurodegenerative disease. A prominent example of neoplastic disease is the t(14;18) translocation of follicular lymphoma, which driven by the Ig heavy chain promoter leads to overexpression of the antiapoptotic player Bcl-2. The resulting resistance to chemotherapy leads to a poorer prognosis. In myocardial infarction, an increased programmed cell death rate has been reported. Besides necrosis, apoptosis occurs in border areas of ischemia.

Viral infections may cause activation of cytotoxic T cells, recognizing viral peptides in combination with MHC class II molecules on the surface of infected cells. Consequently, cytotoxic T cells release granzyme B/porin and present CD95L on their plasma membrane, both triggering the proteolytic-driven death machinery (Fig. 1). Viruses have, however, developed molecular mechanisms to escape this lethal fate, producing serious clinical problems. Diseases related to hematopoietic disorders that are attributed to apoptosis may arise in the absence of growth factors. These factors are needed to develop differentiated blood cells. Thus, chronic hematological disorders, such as anaplastic anemia and myelodysplastic syndromes, may partly be caused by the activation of cell death genes. In chronic inflammatory reactions, excess lymphocytes are removed by apoptosis. This occurs through the CD95L/CD95 system. Insufficiencies in this mechanism can lead to autoimmune diseases such as lupus erythematosus. Here, circulating soluble CD95 competes with cell-bound CD95, thus preventing the necessary death signal. Loss of neurons may be induced by oxidative stress, excitatory toxicity, calcium toxicity, and survival factor deficiency. Direct evidence for an apoptosis-mediated cell death of neurons was found in Alzheimer disease. It could be demonstrated that b-amyloid induces apoptosis in cultured central nervous system neurons. Other diseases such as retinitis pigmentosa and amyotrophic lateral sclerosis are related to genetic alterations leading to apoptosis-mediated degeneration of functional cells. All mechanistic aspects on which the various diseases rely are briefly described in the “Definition” section.

Imaging Visualization of tissues with apoptotic cells is not possible with conventional radiography, computed tomography (CT), and ultrasound. Merely secondary effects on organs affected by this process can be seen, which are morphological shrinking and/or changes in the signal intensity induced by alterations of mass density. Apart from radionuclide imaging, magnetic resonance spectroscopy (MRS) is currently the clinically available method for noninvasive in vivo detection and quantification of apoptosis. The changes in lipid structure and fluidity of the cell membrane that take place during the apoptotic process generate a number of small molecules (e.g., cytoplasmic lipid bodies, choline metabolites) that can be directly monitored by water-suppressed lipid 1 H-MRS techniques (3).

Apoptosis

Nuclear Medicine As already mentioned in the previous section, radiolabeled compounds are the preferred agents for imaging apoptosis. This is due to the comparatively tiny mass amounts of carrier molecules necessary to obtain ample signal intensities. For the detection of apoptosis, radiolabeled annexin V, a human 36-kDa protein, is currently the most promising agent for detecting apoptosis in vivo (3). This agent allows the imaging of an early event in apoptosis, involving the externalization of phosphatidyl serine from the inner to the outer leaflet of the plasma membrane. The high affinity of annexin V for cells with exposed phosphatidyl serine (Kd < 10–10 M) is the basis of detecting apoptosis in vivo. This phenomenon was originally exploited for the detection of apoptotic cells by flow cytometry, using the binding of fluorescein isothiocyanate-labeled annexin V to phosphatidyl serine. The success of this method suggested the replacement of the fluorescent tag by a radioactive tracer. Consequently, 99mTc-labeled annexin V was used several years later to image apoptosis in animals. An increased accumulation was found in Jurkat cells where programmed cell death was initiated by growth factor

97

deprivation, anti-CD95 antibody, and doxorubicin treatment. In addition, anti-CD95-treated mice showed a threefold rise in hepatic 99mTc-annexin V accumulation in response to severe liver damage with histological evidence of apoptosis. Increased uptake was also detected in animal models using the acute rejection of transplanted heterotopic cardiac allografts or transplanted murine B cell lymphomas treated with cyclophosphamide. Radiolabeled annexin V is, therefore, a biological marker which gives, for the first time, direct information about the cytotoxic therapy response of tumors. Several 99mTc-labeled annexin V compounds have been studied in humans, differing in bifunctional chelators. The most successful agent is hydrazinonicotinamideconjugated annexin V, which binds 99mTc efficiently at very low molar concentrations. The investigation of 99m Tc-HYNIC-annexin V for human application revealed favorable biodistribution characteristics. In addition, 99m Tc-HYNIC-annexin V is easily obtained by a kit formulated preparation (Figs. 2 and 3). Other targets for receiving information about the biological fate of tissue undergoing apoptosis are cysteine proteases of the caspase family. As described earlier, these enzymes play a key role during apoptosis. Besides

Apoptosis. Figure 2 Perpendicular views of annexin A5. (a) Each structural domain is colored differently; the calcium ions are red (taken from http://www.lbpa.ens-cachan.fr/bentley/index.html). 99mTc complex of hydrazinonicotinamide-conjugated annexin V (b).

Apoptosis. Figure 3 Images of an untreated patient with a lung tumor showing intense uptake of [99mTc]HYNIC-annexin V. CT left, scintigram middle, fusion image right. The intensity of uptake may increase after chemo- or radiation therapy. The intraindividual change of uptake correlates with therapy response.

A

98

Apparent Diffusion Coefficient (ADC)

irreversible inhibitors, synthetic caspase substrates have been investigated for their potential as apoptosis-selective imaging agents. Although selective uptake in apoptotic cells was observed, imaging could not be demonstrated with these agents until now.

Diagnosis Clinical imaging studies with 99mTc-labeled annexin V have demonstrated the feasibility of delineating cell death in acute myocardial infarction, in tumors with a high apoptotic index, and in response to antitumor chemotherapy. Most anticancer agents act by inducing apoptosis in sensitive tumor cells. Hence, in many types of cancers, a significant increase of apoptosis after chemotherapy correlates with tumor chemosensitivity. Theoretically, a reliable evaluation of apoptotic changes, after chemotherapy to baseline, may provide valuable insights into the treatment prospect of cancers. The tumors that were investigated to date were squamous head and neck carcinomas, non-small-cell lung cancer, small-cell lung cancer, breast cancer, lymphoma, and sarcoma. Increased localization of 99mTc-labeled annexin V within 1 to 3 days of chemotherapy has been noted in some, but not all, subjects with these tumors. Most subjects with increased 99mTc annexin V uptake after the first course of chemotherapy have shown objective clinical responses. It is suggested that increased posttreatment 99mTc annexin uptake is associated with improved time to progression of disease and survival time (4).

Bibliography 1.

2. 3.

4. 5.

Kerr JFR, Wyllie AH, Currie AR (1972) Apoptosis: a basic biological phenomenon with wide ranging implications in tissue kinetics. Br J Cancer 26:239 Krammer PH (1999) CD95(APO-1/Fas)-mediated apoptosis: live and let die. Adv Immunol 71:163–210 Lahorte CMM, Vanderheyden J-L, Steinmetz N et al (2004) Apoptosis-detecting radioligands: current state of the art and future perspectives. Eur J Nucl Med Mol Imaging 31:887–919 Green AM, Steinmetz ND (2002) Monitoring apoptosis in real time. Cancer J Mar–Apr 8(2):82–92 Garcia-Barros M, Paris F Cordon-Cardo et al (2002) Tumor response to Radiotheraphy Regulated by Endothelial Cell Apoptosis. Science 300: 1155–1159

Apparent Diffusion Coefficient (ADC)

ADCs of the tissues vary according to microstructure and physiologic state of the tissues. ▶Lymphadenopathies, Head and Neck

Appendicitis PATRICE TAOUREL Service d’Imagerie Me´dicale Hoˆpital Lapeyronie, 371 avenue du Doyen Gaston Giraud, 34295 MONTPELLIER CEDEX 5. [email protected]

Definition Acute appendicitis is the inflammation of the appendix, which is a long intestinal diverticulum arising from the cecum approximately 3 cm below the ileocecal wall. It constitutes the most common cause of acute abdominal pain requiring surgical intervention in the Western world.

Pathology The primary pathogenic event in the majority of patients with acute appendicitis is luminal obstruction, which may result from fecaliths, lymphoid hyperplasia, and more rarely foreign bodies, parasites, and both primary and metastatic tumors. Fecaliths which result from the inspissation of fecal material and inorganic salts within the appendiceal lumen constitute the most common cause of appendiceal obstruction. Once appendiceal obstruction occurs, the continued secretion of mucus results in elevated intraluminal pressure and luminal distention. Increased intraluminal pressures may lead to venous engorgement, arterial compromise, and tissue ischemia and may result in appendiceal perforation. The appendiceal perforation is associated with a localized peritonitis since the terminal ileum, cecum, and omentum are generally able to “wall off” the inflammation or more rarely it is associated with a generalized peritonitis. Perforation is a relatively common complication of appendicitis with a median incidence of 20%, and it constitutes the major factor of morbidity and mortality in appendicitis.

Clinical Presentation ADCs are obtained by calculating and measuring signal intensity in a series of diffusion-weighted MR images.

The clinical presentation of patients with appendicitis depends on the location of the appendix, on the pathologic

Appendicitis

state of the inflamed appendix, and on the age and sex of the patient. Although the base of the appendix arises from the posteromedial wall of the cecum, the appendix may lie in a retrocecal, subcecal, retroileal, preileal, pelvic, or subhepatic site. Consequently, this variability in location may greatly influence the clinical presentation in patients with suspicion of appendicitis and the differential diagnoses discussed in such patients. The most accurate clinical signs of appendicitis are the presence of right lower quadrant pain, rigidity, and migration of the initial periumbilical pain to the right lower quadrant. The most reliable sign of perforation is a patient’s temperature being higher than 38.5˚C. By using only clinical diagnosis, the mean false-negative appendicectomy rate is approximately 20%; in women of childbearing age it is higher, because symptoms of acute gynecologic conditions such as pelvic inflammatory disease may have a similar manifestation.

99

A

Appendicitis. Figure 1 US of an appendicitis.

Imaging Modality Ultrasound Ultrasound (US) in well-trained hands is highly accurate in the diagnosis of appendicitis. The US approach to the right lower quadrant includes graded compression with slow and gentle maintained pressure, by using a highfrequency linear or curvilinear probe. In women, when examination via the suprapubic approach is inconclusive, endovaginal US should be added, because it may reveal a gynecologic explanation for the symptoms and it may identify the appendix in a pelvic location. The diagnosis of appendicitis is based on the identification of the inflamed appendix as a blind-ended, tubular structure with a laminated wall that arises from the base of the cecum. The appendix is aperistaltic and noncompressible. A threshold of 6 mm in anteroposterior diameter of the appendix under compression is the best finding of appendicitis (Fig. 1), with both a high negative predictive value and a high positive predictive value. When the appendix is identified, the evaluation of a periappendicular finding does not improve the accuracy of US. However, inflammatory changes in the perienteric fat are often the first and more obvious finding at US examination. Inflamed fat appears as a badly limited echogenic mass that separates the inflamed appendix from the surrounding gut and other organs. Secondary appendicular findings such as appendicoliths seen as bright, echogenic foci with clean distal acoustic shadowing (Fig. 2), Doppler enhancement of the appendicular wall, or absence of gas in the appendicular lumen are ancillary signs that may be useful in equivocal cases. There are two main pitfalls in the diagnosis of appendicitis. The first is the misinterpretation of the terminal ileum as the appendix, which may lead to over

Appendicitis. Figure 2 US of appendicitis with appendicolith.

diagnosis of appendicitis. In contrast to the appendix, the terminal ileum does not attach to the base of the cecum, is not blind-ended, shows frequent peristaltic activity, and is usually oval in cross section. The second pitfall is the nonvisualization of the normal appendix, which is classically considered as a major weakness of US in the exploration of patients with suspected appendicitis. However, technological advances combined with increased radiologist experience have dramatically improved the US visualization of the normal appendix, which is now identified in more than two-thirds of cases in welltrained teams. US diagnosis of appendicitis after perforation can be difficult since the distended appendix may no longer be visualized at US examination. Perforated appendicitis must be considered in patients without a history of appendicectomy when a collection is identified in the right lower quadrant.

Computed Tomography Computed tomography (CT) is a highly accurate and effective cross-sectional imaging technique for the diagnosis and staging of appendicitis.

100

Appendicitis

CT protocols used in patients with suspected appendicitis differ considerably with regard to the anatomic area to be included in the scan and the use of intravenously, orally, and rectally administered contrast material. Most teams use thin slices without intravenous contrast material and only administer contrast material in equivocal cases when the appendix is not identified such as in patients with mild appendicitis, patients with a paucity of mesenteric fat, or patients with perforated appendicitis. Some teams have promoted a more invasive protocol which involves scanning after the rapid administration of colonic contrast material to fill the cecum and the appendiceal lumen. The main finding of appendicitis is the identification of an inflamed appendix, which is dilated measuring more than 7–8 mm in diameter, with a circumferential and symmetric wall thickening, better demonstrated after intravenously administered contrast material with an enhancement of the thickened appendiceal wall. Reformatting (Fig. 3) and cine mode are very helpful in identifying the appendix on its entire length. Periappendiceal inflammation is present in 98% of patients with acute appendicitis with linear fat stranding, local fascial thickening, or clouding of the peri-ileal fat. Perforated appendicitis is usually accompanied by pericecal phlegmon or abscess. Although a pericecal phlegmon or abscess is strongly suggestive of appendicitis, these are nonspecific findings that may be seen in other

diseases entities. The most specific finding of perforated appendicitis is the identification of extraluminal air, which, however, may be seen in other causes of bowel perforation in the right lower quadrant such as cecal diverticulitis or Crohn’s disease, and the identification of an appendicolith within a appendiceal abscess or phlegmon.

Magnetic Resonance Imaging The indication of magnetic resonance imaging (MRI) in patients with suspicion of appendicitis is only limited to pregnant women with the same signs on CT, a poorer spatial resolution, and limitations in the identification of the extraluminal air.

Diagnosis There is not a univocal strategy in patients with suspicion of appendicitis. The elaboration of a triage strategy needs to address two questions: (i) which patients need imaging? and (ii) which imaging modality should be used: US versus CT? Although the use of imaging in clinically equivocal cases of appendicitis is universally admitted, the systematic use of imaging in patients with a clinical suspicion of appendicitis is still controversial. Several studies have been

Appendicitis. Figure 3 CT of appendicitis with oblique reformating.

ARCO

published in which the medical and financial implications of the systematic use of imaging have been assessed. It has been shown that the use of imaging improved patient care both by averting unnecessary appendicectomy and by averting delays before necessary medical or surgical treatment. The cost analysis demonstrated that routine use of imaging was cost-effective because savings achieved by eliminating unnecessary surgery and in-hospital observation outweighed the cost of performing routine imaging of appendicitis. The choice between US and CT depends on the institution’s preference and on the available experience. In most teams, particularly in the USA, CT is considered as the modality of choice. It has the advantage of being more sensitive in demonstrating a normal appendix and in excluding acute appendicitis. It is more accurate in staging periappendiceal inflammation and its extent and in differentiating periappendiceal phlegmon from abscess. The accurate staging of acute appendicitis is of importance for the management of patients, with the option of using an antibiotic therapy or a percutaneous drainage before surgery in complicated appendicitis and of triaging patients appropriately for laparoscopic versus open surgery since the laparoscopic approach may not be optimal in patients with complicated appendicitis. Moreover, CT is more useful in providing alternative diagnoses such as a Crohn’s disease,

101

cecal diverticulitis, right ischemic colitis, sigmoid diverticulitis with the sigmoid loop located in the right lower quadrant, or a urologic condition. Conversely, US is more accurate in diagnosing a gynecologic condition and is equally accurate in diagnosing mesenteric lymphadenitis, which is the main differential diagnosis of appendicitis. In our experience, the patient’s age, sex, and body habitus as well as the clinical presentation are important influencing factors and we use the following diagnostic triage both in the decision of imaging and in the choice between US and CT (Fig. 4).

Bibliography 1. 2.

Birnbaum BA, Wilson SR (2000) Appendicitis at the millenium. Radiology 215:337–348 Kessler N, Cyteval C, Gallix B et al (2004) Appendicitis; evaluation of sensitivity, specificity and predictive values of Us, Doppler US, and laboratory findings. Radiology 230:472–478

APUD Amine precursor uptake and decarboxylation—Paracrine cells of which argentaffin cells are an example. ▶Neoplasms of the Chest in Childhood

Archnoid Cyst ▶Cysts, Cerebral and Cervical, Childhood

Architectural Distortion Disruption of the normal breast architecture with no definite mammographically visible mass ▶Carcinoma, Breast, Imaging Mammography, Primary Signs ▶Sclerosing Adenosis, Breast

ARCO Appendicitis. Figure 4 Diagnostic algorythm for patients with suspicion of appendicitis.

▶Association Recherche Circulation Osseous

A

102

ARGUS

ARGUS Acronym for children suffering from the association of AnoRectal, GenitoUrinary, and Sacral anomalies. ▶Anorectal Malformation

Arm Claudication

US, while selective hepatic angiography and following embolization may be useful in symptomatic cases. ▶Vascular Disorders, Hepatic

Arteriohepatic Dysplasia Synonym for Alagille syndrome. ▶Congenital Malformations, Liver and Biliary Tract ▶Congenital Malformations, Bile Ducts

▶Claudication, Brachial

Arterioportal, Fistula ARPKD and ADPKD Autosomal recessive or dominant polycystic kidney disease: An inherited progressive condition that may manifest in different ages and with varying expression, potentially deleterious to renal function. ▶Cystic Renal Disease, Childhood

Arterial Ischemic Stroke ▶Stroke, Children

Arterio-venous Malformation, Hepatic Hepatic arterio-venous malformation (AVM) is a rare vascular disorder characterized by direct arterial connection to the venous drainage system within the liver. Arterio-venous fistulae may be congenital (vascular malformations in Rendu–Osler disease) or acquired. Iatrogenic causes (percutaneous liver biopsy, percutaneous liver tumor ablation procedures), trauma, and cirrhosis are mostly involved for acquired fistulae. The condition results in a high-flow, low-resistance shunt that can cause high-output cardiac failure and hydrops. Diagnosis is usually achieved by means of color Doppler

Arterioportal fistulas consist of a communication between the hepatic artery and the portal venous system. Arterioportal shunts may be intra-hepatic or extra-hepatic and acquired or congenital. The most common causes of acquired arterioportal fistulas are cirrhosis and hepatic neoplasms, blunt or penetrating trauma, iatrogenic injuries, rupture of an aneurysm of the hepatic artery. Congenital arterioportal fistulas can be associated with Rendu–Osler disease, hereditary haemorrhagic telangiectasia, Ehlers– Danlos syndrome. Arterioportal shunts may be minute or large. Large arterioportal fistulas themselves may cause portal hypertension and high-output heart failure. Doppler US usually allows an accurate diagnosis. Enlargement of the hepatic artery and dilatation of the involved portal venous segment can be observed at US examination. Doppler US shows pulsatile hepatofugal flow in the portal vein. Both contrast-enhanced CT and contrast-enhanced MR imaging may demonstrate marked enhancement of the main portal vein, segmental branches or major tributaries, with attenuation or signal intensity approaching that of the aorta on arterial-phase images. Arteriography can be performed to confirm the diagnosis and embolization may be considered. ▶Portal Hypertension

Arteriovenous Malformation A mass consisting of tightly packed arteries, capillaries, and veins. ▶Congenital Malformations, Vascular, Brain ▶Stroke, Interventional Radiology ▶Vascular Disorders of the Gastrointestinal Tract

Asplenia

Arteriovenous Shunting

103

Ascites A

Pathological vascular channels between arteries and veins, bypassing the capillaries with high-volume blood flow; common in tumors. ▶Contrast Media, Ultrasound, Applications in Kidney Tumor

An accumulation of transudate, exudate, or chyle in the peritoneal cavity. ▶Peritoneal Collections

Ascites, Pancreatic Arthritis Inflammation of a joint. With the use of bone scintigraphy, the floridity of arthritis or response to treatment may be judged early. ▶Bone Scintigraphy

Ascites is an abnormal accumulation of fluid in the abdomen. Pancreatic ascites develops when a pseudocyst bursts releasing pancreatic juices. Abdominal paracentesis is the most rapid and perhaps the most cost-effective method of diagnosing the pancreatic ascites. ▶Pancreatitis, Acute

Aseptic Bone Necrosis Arthritis of the Costotransversal and Costovertebral Joints Arthritis of the costotransversal and costovertebral occurs in ankylosing spondylitis and is the main cause of respiratory movement restriction. In X-ray imaging, it is most often occult. Magnetic resonance imaging, however, shows typical arthritic features: adjacent bone marrow and soft tissue edema, contrast material enhancement of the synovial membrane, and bony destruction or ankylosis. ▶Spondyloarthropathies, Seronegative

Asbestos Related Diffuse Pleural Thickening According to the International Labour Organisation (ILO), chest radiograph criteria for defining diffuse pleural thickening includes; pleural thickening greater than 5 mm at any site, bilateral pleural thickening involving at least 25% of the chest or 50% if unilateral, or obliteration of the costophrenic angle. ▶Pleural Plaques

▶Osteonecrosis, Adults

Aspergilloma Colonization with aspergillus of a previous TB cavity. ▶Tuberculosis, Lung

Aspergillosis Cerebral aspergillosis usually occurs after hematogenous dissemination from an extracerebral focus, or is a result of contiguous spread of the infection from the paranasal sinuses. ▶Infection, Opportunistic, Brain

Asplenia The absence of the spleen is a rare condition usually associated with other congenital malformations, especially cardiovascular and pulmonary anomalies. Asplenia

104

Asplenia Syndrome

may be seen in association with abdominal situs ambiguous and has been classically called asplenia syndrome. The asplenia syndrome is most frequently encountered in males and is associated with severe cyanotic congenital heart diseases. The typical anatomic features of classic asplenia syndrome are trilobed lungs with bilateral minor fissures and eparterial bronchi, bilateral systemic atria, midline liver, absent spleen, and variable location of the stomach. Scintigraphy is the standard examination in documenting the absence of the spleen. ▶Congenital Abnormalities, Splenic

Asplenia Syndrome Asplenia syndrome is characterized by asplenia, or absence of the spleen, associated with abdominal situs ambiguous. The syndrome is most frequently encountered in males and is associated with severe cyanotic congenital heart diseases. The typical anatomic features of classic asplenia syndrome are trilobar lungs with bilateral minor fissures and epiarterial bronchi, bilateral systemic atria, midline liver, absent spleen, and variable location of the stomach. ▶Splenic Anomalies

Asplenia, Polysplenia Syndromes ▶Congenital Malformations, Splenic

Asymmetric Density An asymmetry of the glandular breast tissue, visible mammographically on two views. Asymmetry is common but should be evaluated as uncommonly there may be an underlying mass or architectural distortion. ▶Carcinoma, Breast, Imaging Mammography, Primary Signs

Atelectasis C ORNELIA S CHAEFER-P ROKOP AMC, Amsterdam [email protected]

Synonyms The word atelectasis is of Greek origin and means “lack of stretch.” “Loss of volume” and “collapse” are used synonymously (caveat: collapse may refer to the complete atelectasis of one lobe)

Definition Collapse or decrease in the volume of a lung or a portion of the lung. The term refers to the region and amount of lung that is collapsed or to the underlying pathophysiology.

Association Recherche Circulation Osseous (ARCO) Pathology/Histopathology ARCO is an international organization and published the widely accepted staging criteria for osteonecrosis: committee on terminology and classification. ▶Osteonecrosis in Adults

Asthma ▶Airway Disease

There are various pathological mechanisms causing atelectasis: 1. Postobstruction or resorption atelectasis: the intraalveolar air is resorbed distal to a bronchial obstruction. The latter may be endobronchial (e.g., neoplasm, mucoid impaction, foreign body) or extrabronchial (e.g., lymphadenopathy). A mucoid or fluid bronchogram, seen on computed tomography (CT) as a low-density branching structure, should prompt the search for a central obstructing lesion (Fig. 1).

Atelectasis

105

A

Atelectasis. Figure 1 Patient with obstruction (resorption) atelectasis due to a large central mucous plug. Note the bronchi filled with mucus next to the contrasted arteries.

(Fig. 2). The lung collapse seen with a pneumothorax is also described as relaxation atelectasis. 3. Adhesive, nonobstructive atelectasis (or microatelectasis) caused by surfactant deficiency and seen postoperatively or in acute respiratory disease syndrome (ARDS). 4. Cicatrization atelectasis, focal or more diffuse collapse resulting from fibrosis or scarring. It is frequently associated with bronchiectasis and also seen in granulomatous diseases.

Atelectasis. Figure 2 Patient with pleural empyema (note pleural split sign and air inclusions after pleurodesis) with a compression atelectasis of the lung parenchyma.

2. Passive or compression atelectasis: pulmonary collapse as a result of a space-occupying lesion within the pleural space, such as pneumothorax or pleural effusion, or within the lung, such as bulla or mass

A rounded atelectasis (synonym: folded lung, atelectatic pseudotumor) represents a compression atelectasis seen after resorption of a pleural exudate and reactive fibrosis formation. It is most frequently seen in asbestos-related pleural disease, but is also seen in association with any (exudative) pleural effusion. A platelike atelectasis (synonym: discoid atelectasis) is a linear or planar opacity representing a portion of the lung with decreased volume, usually seen in lower lung zones.

Clinical Presentation Physical changes and symptoms depend on the size of the atelectasis and on the underlying and accompanying respiratory diseases. In most cases there will be no clinical findings at all. In patients with ▶right middle lobe atelectasis, the upper and lower lobe will fill the space occupied by the collapsed lung, and breath sounds can be

106

Atheromatous RAS

Atelectasis. Figure 3 Patient with a central perihilar tumor and upper lobe atelectasis: the curvilinear delineation of the atelectasis of the upper lobe indicates a central tumor as underlying disease (▶Golden S-sign).

normal. Atelectasis of the lower lobe may result in absent breath sounds. If atelectasis is caused by intrabronchial obstruction (e.g., tumor, mucous plugging), crackles distal to the obstruction may be heard (Fig. 3).

Diagnosis

Imaging

Bibliography

The collapse of one or multiple complete lung lobes causes distinct imaging features based on the volume loss on one side, the displacement of fissures and elevation of the ipsilateral diaphragm (juxtaphrenic peak), the mediastinal and hilar structures, and the compensatory overinflation of the remaining lung. On CT images, atelectasis shows a strong and homogeneous enhancement after intravenous injection of contrast media. This represents a valuable finding for differentiating atelectasis from a tumor or pneumonia, both showing a less intense and more inhomogeneous enhancement. Postobstruction or resorption atelectasis: A mucoid or fluid bronchogram, seen on CT as low-density branching structures, should prompt the search for a central obstructing lesion. A rounded atelectasis describes a peripherally located, rounded or wedge-shaped atelectasis (sometimes mimicking a bronchogenic neoplasm) that is always adjacent to pleural thickening. As opposed to a tumor, it demonstrates the characteristic homogeneous enhancement after contrast medium injection at CT; vessels and bronchi located more centrally are crowded and course in a characteristic curvilinear fashion like a “comet tail.”

The diagnosis of atelectasis is based on typical imaging findings. In most cases the type of atelectasis and the underlying disease can also be determined.

1. 2.

3. 4.

Proto AV (1996) Lobar collapse: basic concepts. Eur Radiol 23:9 Stark P, Leung A (1996) Effects of lobar atelectasis on the distribution of pleural effusion and pneumothorax. J Thorac Imaging 11:145 Ashizawa K, Hayashi K, Aso N et al (2001) Lobar atelectasis: diagnostic pitfalls on chest radiography. Br J Radiol 74:89 Raasch BN, Heitzman ER, Carsky EW et al (1984) Computed tomographic study of bronchopulmonary collapse. Radiographics 4:195

Atheromatous RAS RAS secondary to the presence of atheromatous plaque(s) in the arterial wall. ▶Stenosis, Artery, Renal

Atherosclerotic Nephropathy A complex entity. It is an important cause of end-stage renal failure, a consequence of the association of multiple

Autoimmune Pancreatitis

factors, including decreased renal blood flow, intrarenal atherosclerotic arterial disease, atheroembolism, diabetes, increased oxidative stress, medullary hypoxia, endothelial dysfunction, and inflammation. ▶Hypertension, Renal

Athyroid The physiological manifestation where there is a complete absence of thyroid function or hormone production. ▶Congenital malformations, Thyroid, and Functional Disorders

ATN ▶Tubular Necrosis, Kidney, Acute

Atypical Ductal Hyperplasia Breast lesion with some of the architectural and cytologic features of low-grade ductal carcinoma in situ in part of the affected areas. ▶Radial Scar, Breast

107

Autogenous Dialysis Fistula A A surgically created, direct communication in between an artery and a corresponding superficial vein allowing increased blood flow through the vein in order to provide luminal dilatation and thickening of the vein wall. ▶Fistula, Hemodialysis

Autoimmune Hepatitis Chronic inflammatory disease of the liver due to a cell-mediated immune response against the body’s own liver. The histologic and clinical features are virtually indistinguishable from chronic viral hepatitis. Women are more often affected than men (8:1). The cause is unknown, but a genetic predisposition or an acute liver infection has been advocated as possible factor. Autoimmune hepatitis can progress to cirrhosis. It can also sometimes occur as acute hepatitis. The diagnosis is achieved with a combination of clinical and laboratory findings. Various specific autoantibodies are routinely used. However, the definite diagnosis of autoimmune hepatitis always requires a liver biopsy. Imaging has a poor role in the diagnostic work-up, but is used to follow the progression to cirrhosis and to detect complications such as portal hypertension and development of hepatocellular carcinoma. The findings at ultrasound, computed tomography, and magnetic resonance do not differ from those observed in other forms of chronic hepatitis. ▶Hepatitis

Atypical Lobular Hyperplasia Proliferation of lobular epithelium with some of the features of lobular carcinoma in situ. ▶Hyperplasia, breast

Auditory Ossicles The middle ear ossicles are the malleus, incus, and stapes. ▶Temporal Bone, Inflammatory Diseases, Acute, Chronic

Autoimmune Pancreatitis Autoimmune pancreatitis (AIP) is a unusual type of chronic pancreatitis having an underlying autoimmunity mechanism. It is reported to represent about the 5% of cases of chronic pancreatitis, but its real incidence and prevalence are not well known. AIP is a recently defined disease, which appears to be a unique clinical entity characterized by peculiar pathologic and imaging findings, laboratory data and effectiveness of

108

Avascular Necrosis

corticosteroid therapy. In 2002, the Japan Pancreas Society proposed diagnostic criteria of AIP, including findings of imaging studies (diffuse narrowing of the main pancreatic duct with diffuse enlargement of the pancreas), laboratory data (elevated levels of serum IgG or the presence of autoantibodies) and histopathological findings (fibrotic changes with lymphocyte and plasma cell infiltration). Association with other autoimmune diseases (as Sjogren syndrome, primary sclerosing cholangitis and biliary cirrhosis,) has been reported in about 50% of patients. AIP can be classified in diffuse (more common) and focal type, which usually involves the pancreatic head and/or the uncinate process. Pathological findings include diffuse lymphoplasmacytic infiltration, variable degree of parenchymal atrophy and fibrotic changes of the gland and contiguous soft tissues. Sometimes the fibro-inflammatory process involves also the intrapancreatic tract of the common bile duct and small venous vessels. Clinical features are non-specific and include slight upper abdominal pain, obstructive jaundice, weight loss and easy fatigability. Laboratory tests may show an increased levels of IgG (especially the IgG4 subtype) and the presence of autoantibodies (anti-nuclear antibodies, anti-carbonic anhydrase anti-bodies, etc.) pointing out the autoimmune origin of this disease. Serum levels of amylase, lipase and bilirubin may be increased. US examination shows a homogeneous, hypoechoic, diffuse or focal enlargement of the pancreas, sometimes associated with dilation of the intra-hepatic and common bile ducts. After US contrast-media administration, the involved pancreatic parenchyma shows from mild to moderate enhancement, which is thought to be directly related to the inflammatory involvement and inversely related to the degree of fibrosis. On CT, the diffuse type of AIP appears as a uniform swelling of the pancreas with well-defined outline. A hypodense rim-like capsule surrounding partially or completely the pancreas and showing a delayed postcontrast enhancement can be observed; this finding is related to fibro-inflammatory changes involving the peripancreatic adipose tissue and it is characteristic of AIP. In the focal type, CT examination depicts a mass-like lesion usually located in the head and/or in the uncinate process. Before contrast-medium administration, this focal enlargement typically has smooth margins and it appears homogeneous and iso- or hypodense with respect to the surrounding pancreas. On contrast-enhanced images, the lesion usually shows a hypodense appearance in pancreatic phases with a delayed enhancement. Dilation of the main pancreatic duct and biliary tree may be depicted. Sometimes wall thickening of the distal common bile

duct showing enhancement is present. Parenchymal calcifications, pseudocysts, vascular invasion, inflammatory involvement of the mesentery, the anterior pararenal fascia and the lateroconal fascia are typically absent. MR examination can demonstrate a homogeneous, focal or diffuse pancreatic enlargement, which generally is hyper-isointense in T2-weighted images and hypointense in T1-weighted sequences compared with the liver. In gadolinium-enhanced T1-weighted images, the involved pancreatic parenchyma shows a homogeneous enhancement. A capsule-like rim surrounding the pancreas may appear as a hypointense peripheral band on both T1 and T2-weighted images and may show delayed enhancement. MR cholangiopancreatography (MRCP) and endoscopic retrograde cholangiopancreatography (ERCP) often show diffuse or focal narrowing of the pancreatic duct. Corticosteroid therapy can determine resolution of clinical manifestation, normalization of pancreatic function, and complete or partial regression of morphologic pancreatic abnormalities. However, in the absence of an appropriate clinical setting, the focal form of AIP can be extremely difficult to differentiate from pancreatic cancer on the basis of imaging features alone and biopsy or surgical resection may be necessary. ▶Pancreatitis, Chronic

Avascular Necrosis Ischemic death of the cellular elements of bone and marrow. Ischemic necrosis is a term usually reserved for epiphyseal and subarticular involvement. Bone infarcts describe a similar process occurring in the metaphysis and epiphysis. Etiologies include intraluminal occlusion (thromboembolic, Caisson’s disease), vessel abnormality (vasculitis, radiation, disruption secondary to trauma), or compression from external mechanical pressure (fatty deposition secondary to hypercortisolism, Gaucher’s). MRI and scintigraphy are sensitive early in the disease. Early MRI changes include T2-prolongation or decreased areas of enhancement. The most characteristic MRI abnormality is the double line sign with an inner border of T2 high signal surrounded by a low signal border. Radiographic abnormalities occur late in the course of disease and are characterized by patchy lucent areas of sclerosis, lucency, and osseous collapse. ▶Fractures, Peripheral Skeleton ▶Juvenile Osteonecrosis ▶Osteonecrosis, Adults ▶Osteonecrosis, Childhood

Azotemia

Avulsion

109

Axillary Node Dissection (AND) A

A pulling away of a secondary growth center bone or cartilage by tendon, ligament, or muscle. ▶Fractures, Bone, Childhood

Axillary Lymphadenopathy Enlargement of lymph nodes in the axilla ▶Carcinoma Breast, Imaging Mammography, Secondary Signs

Standard surgical removal of the whole axillary lymph nodes, levels I and II. ▶Breast Conserving Therapy

Azotemia Elevation of the blood urea nitrogen (BUN) and serum creatinine levels. ▶Glomerulonephritis

B

B-cell Lymphoma Malignant transformation of B-lymphocytes. Various subtypes can occur, depending on the moment of transformation during the B-cell life cycle. Represents the largest subset of non-Hodgkin lymphoma and is the most frequent cause of mammary lymphomatous involvement. ▶Lymphoma, Breast

splenomegaly in the absence of intra-hepatic or extrahepatic obstruction. The syndrome is characterized by signs of portal hypertension. Signs of hyper-splenism, with anaemia, leukopenia and thrombocytopenia, are often present. The advanced stage of Banti’s disease may be complicated by upper gastrointestinal haemorrhages. ▶Portal Hypertension

Barium Meal Background Radiation Ionizing radiation is part of the environment. Eighty percent of background radiation comes from natural sources including radon gas, cosmic rays, the earth’s crust, and radioisotopes in the human body and within the food chain. Industrial sources account for approximately 1% and medical sources of radiation for the remaining 19%. The average background radiation dose to the world’s population is 2.4 mSv/year. ▶Radiation Issues in Childhood

A radiographic examination of the upper GI tract that may be tailored to focus on peptic ulcer disease, on gastric cancer detection, or on evaluation of the gastro-oesophageal reflux disease. ▶Gastroesophageal Reflux in Adult Patients: Clinical Presentations, Complications, and Imaging

Barium, Indications and Contraindications B RUNO B ONNEMAIN

Backwash Ileitis This term indicates the rare involvement of the distal ileum in UC, which may occur only when the disease involves the entire colon (pancolitis). Whenever present, the differential diagnosis with Crohn’s disease can be difficult. ▶Colitis, Ulcerative

Banti’s Syndrome Banti’s syndrome, or non-cirrhotic idiopathic portal hypertension, is a condition characterized by congestive

Villeparisis, France [email protected]

Definition It was rapidly recognized after the discovery of X-rays that a radio-opaque product would be needed to opacify the alimentary tract. A number of agents were tried until Bachem and Gu¨nther, in Germany, first used barium sulfate in 1910, and this has remained the agent of choice ever since. Various formulations have been commercialized to take into account the region of interest, the site of administration (oral versus rectal route), the mode of administration (single-versus double-contrast examination), and the type of equipment used (conventional versus CT). Barium sulfate (or iodinated contrast agents)

112

Barium Interactions

are administered orally to evaluate abnormalities of the mouth, pharynx, oesophagus, stomach, duodenum, and proximal small intestine. These contrast media are also administered rectally for retrograde examinations of the colon and distal small intestine. A specific, antegrade examination of small intestine may also be performed following ingestion (small bowel follow through) or direct-infusion into the distal duodenum via an intubation tube (small bowel enteroclysis).

Indications As a general statement, the indications mentioned concern radiographic visualization of the gastrointestinal tract. The approved indications may vary from country to country, and according to the product final formulation. 1. Basic formulation (oral or rectal route): Barium sulfate is indicated for conventional radiological examination of the oesophagus, stomach, duodenum, and colon in double contrast or complete filling, as well as small bowel follow-through. 2. Formulation for colon: Barium sulfate is indicated for opacification of the colon for radiological explorations by the double-contrast technique or with barium filling. 3. Formulation for CT (oral route): Barium sulfate is indicated for opacification of the gastrointestinal tract in CT examination. 4. Oral pastes of barium sulfate are indicated for opacification of upper gastrointestinal tract: pharynx, hypopharynx, and oesophagus during radiological explorations. 5. Specific double-contrast formulations of barium sulfate (high-density products) are indicated for opacification of the oesophagus, stomach, and duodenum for doublecontrast radiological examination.

Contraindications Barium sulfate—when used properly—is a remarkably safe contrast agent. There are, however, certain contraindications, both absolute and relative, which need to be considered (the wording can vary from one country to another). These include: 1. 2. 3. 4. 5.

Leakage into the pleural or peritoneal spaces Leakage into the mediastinum Possible pulmonary aspiration Known gastric perforation Barium given orally in suspected distal large bowel obstruction.

6. Suspected or known intestinal perforation or obstruction. 7. Toxic megacolon 8. Hypersensitivity to barium sulfate or any component of the formulation 9. Within 6 days of large forceps or “hot” colonic biopsy or snare polypectomy. In addition, for Barium pastes, esophageal atresia is considered as a contraindication. Finally, caution should be exercised in patients with preexisting constipation due to the risk of barium stercoroma, particularly in elderly subjects. Since some formulations contain potassium sorbate, the potassium intake should be taken into account for patients with a low-potassium diet. Patients with gastrointestinal reflux should be monitored closely for any intensification of this symptom when barium sulfate is given orally. In conclusion, barium sulfate is helpful in radiology for gastrointestinal examinations and the benefit–risk ratio is clearly favorable if properly used.

Barium Interactions J EAN -M ARC I DE´ E Guerbet Research Division, Aulnay-sous-Bois, France [email protected]

Definition To date, no extensive studies have been reported in the literature about a deleterious interaction of barium sulfate with drugs, chemical agents, or ions. In fact, two situations should be considered: (a) potential deleterious interactions of barium sulfate with drugs, (b) interactions aimed at improving the imaging efficacy of the gastrointestinal tract. Although the latter situation refers to a rather broad definition of “interaction,” we believe it is worth mentioning here.

Interactions Side Effects Resulting from Interactions of Barium Sulfate with Drugs Barium sulfate shows negligible absorption from the gastrointestinal tract following either oral or rectal administration (1), and, consequently, is not metabolized by the liver. Therefore, traditional interactions involving

Barium, Presentations

the group of cytochrome P450 enzymes or other pathways cannot be expected in this case. Barium sulfate suspensions can be considered as unstable adsorption complexes that have the potential to bind many chemical substances, including drugs. Consequently, this may interfere with the intestinal absorption of orally-administered drugs (1). In such cases, the barium coating would lead to the inhibition of the drug absorption by the gastrointestinal tract, thus constituting a “mechanical interaction” which, in turn, would reduce or completely inhibit the drug-induced effects. However, the probability for simultaneous administration of barium and a drug is low and, to the best of our knowledge, no clinical studies have been published so far. Nevertheless, clinicians should be aware of such potential risk and take all necessary measures to prevent it.

2.

3.

4.

5.

113

Freeman A (1999) Contrast agents and the gastrointestinal tract. In: Dawson P, Cosgrove DO and Grainger RG (eds) Textbook of Contrast Media. ISIS Medical Media, Oxford, pp 181–189 Van Damme W and Gunawan OH (1981) The use of N-acetylcystein in double-contrast radiography of the colon: a clinical trial. J Belge Radiol 64:171–173 Cittadini Jr, Gallo A, De Cicco E et al (2001) Magnesium ions and barium coating of colonic mucosa: is it a simple question of viscosity? Eur Radiol 11:1429–1532 Goei R, Nix M, Kessels AH et al (1995) Use of antispasmodic drugs in double contrast barium enema examination: glucagon or buscopan? Clin Radiol 50:553–557

Barium, Presentations J EAN -P HILIPPE B ARRAUD

Interactions with Compounds to Improve Imaging Efficacy Among numerous additives, gums such as tragacanth are used to promote adherence of the barium sulfate to the mucosa. Because different manufacturers use different additive agents, these may be incompatible. Consequently, barium sulfate suspensions should not be mixed together during the same examination, to avoid flocculation (2). An excess in mucus can be detrimental to the barium sulfate coating. Schematically, two categories of methods have been used to control the local production of mucus and consequently improve the quality of barium coating: (a) systemic administration of drugs such as histamine H2 receptor antagonists or N-acetylcysteine and (b) topical treatment with mucolytic drugs. The efficacy of histamine H2 receptor antagonists is an object of debate. The mucolytic drug N-acetylcysteine has not been found to improve the quality of double-contrast barium meal whereas double-contrast barium enema has been shown to be of better quality in patients receiving this agent (3). The divalent cation magnesium has been shown to improve the quality of barium coating during doublecontrast enema, an effect that cannot be attributed to the increase in viscosity (4). Finally, antispasmodic drugs (hyoscine-N-butylbromide or glucagon) are widely used in barium enema examinations to minimize patient discomfort and to improve the quality of the examination (5).

Bibliography 1.

Swanson DP and Halpert RD (1990) Gastrointestinal contrast media: barium sulfate and water-soluble iodinated agents. In: Swanson DP, Chilton HM, and Thrall JH (eds) Pharmaceuticals in Medical Imaging. Radiopaque Contrast Media. Radiopharmaceuticals. Enhancement Agents for Magnetic Resonance Imaging and Ultrasound. Macmillan Publishing Co. Inc., New York, pp 155–183

Guerbet, Roissy, France [email protected]

Definition Manufacturers have developed a large number of barium preparations to optimize the quality and facilitate the use of barium sulfate contrast media. Their availability is subject to the locally prevailing regulations. The following information has been obtained from a selection of manufacturers who kindly accepted to send the author the relevant data.

Characteristics Paste Most pastes are thick containing more than 100% barium sulfate (w/v). They are mainly used to opacify the esophagus, the rectum, and the sigmoid colon. Manufacturers add functional excipients to obtain the characteristics of the paste and flavoring agents to improve their palatability and therefore patient compliance in explorations of the esophagus. The following list is representative of excipients mentioned by manufacturers. It is not exhaustive, because of the large number of manufacturers present worldwide. Functional excipients: xanthan gum, natural gums, potassium sorbate, sorbitol, glycerol, saccharin sodium, cellulose, carboxymethylcellulose, simethicone, polyoxyethylene glycol monooleate, polyoxyethylene glyceryl monooleate, citric acid, sulfuric acid, sodium methylparabenzoate, sodium propylparabenzoate. Flavoring agents: strawberry, vanilla, caramel.

B

114

Barium, Presentations

Examples of brand names and composition: Pastes are usually available in tubes. Evacupaste 100: 100% w/v, that is 70 g of barium sulfate per 100 ml of paste or 56% w/w. Microtrast: 154 g of barium sulfate per 100 ml, that is 154% w/v, or 70% w/w, that is 70 g of barium sulfate per 100 g of paste. Prontobario Esofago: 113% w/v. Esopho-Cat: 3% w/v

Powder Barium sulfate contrast media in powder form contain mainly barium sulfate powder. Excipients are selected by manufacturers for the various preparations depending on the administration route (oral or rectal) and the respective indications. The final preparation of the contrast media requires addition of water, which usually has to be mixed vigorously before use. There are two main families of barium contrast media in powder form: colon products and upper gastrointestinal (GI) tract products. Colon powder products contain from 92 to 99% w/w barium sulfate. The list and the choice of excipients stem from the need of creating homogeneous films for double contrast studies or homogeneous volumes of barium sulfate for repletion studies: natural gums, sodium citrate, citric acid, sorbitol, ethyl maltol, simethicone, polyoxyethylene glyceryl monooleate, polypropylene glycol, bentonite, titanium dioxide, sodium carmellose, sodium carragenine, carboxymethylcellulose. These products are often available in soft or semirigid enema bags to which water is added for constitution of the corresponding contrast media. The final concentration of barium sulfate is higher for double contrast barium enemas than for repletion studies. Examples of brand names and composition: E-Z-Paque: 95% w/w, that is 95 g barium sulfate per 100 g of product. Micropaque colon: 92% w/w. Prontobario colon: 94% w/w. SOL-O-PAKE: 99% w/w. Upper GI tract powder products contain from 81 to 98% barium sulfate powder w/w. Excipients: natural gum, citric acid, sorbitol, simethicone, polyoxyethylene glyceryl monooleate, polyethylene glycol oleate, polypropylene glycol, bentonite, saccharin sodium, sodium carmellose, sodium citrate, sorbitol. Flavoring agents are selected to ensure the maximum compliance of patients, who have to drink the constituted product: strawberry, vanilla, caramel, marshmallow.

High-density (HD) products allow preparation of contrast media containing up to 250% w/v barium sulfate for double contrast examination of the stomach. These products are usually available in single-dose beakers to which water is added depending on the final concentration. Examples of brand names and composition: E-Z-Paque: 95% w/w, that is 95 g barium sulfate per 100 g of product. Digibar 190: 97% w/w. E-Z-HD: 98% w/w Prontobario HD: 98% w/w. Entero Vu: 81%w/w.

Suspension Barium sulfate suspensions in water are the most widely used preparations, because they are ready or nearly ready to use. Medium- and HD suspensions are used for conventional radiological GI tract imaging. Suspensions contain from 13 to 210% barium sulfate (w/v), depending on the main indication of each product. The concentration of the product is often adjusted by the user by addition of water to obtain the optimal barium sulfate concentration for a given examination. Excipients: natural gums, xanthan gum, cellulose, potassium sorbate, sorbitol, ethyl maltol, sodium citrate, citric acid, simethicone, saccharin sodium, carmellose sodium, magnesium aluminum silicate, methylcellulose, sodium carboxy methylcellulose, polyoxyethylene glyceryl monooleate, potassium chloride, sodium citrate, sulfuric acid, acetic acid, hydrochloric acid, sodium methylparabenzoate, sodium propylparabenzoate Flavoring agents: strawberry, vanilla, caramel, lemon cream, blueberry, orange, apple. Examples of brand names and composition: Liquid E-Z-Paque: 60% w/v, that is 60 g barium sulfate per 100 ml of suspension. Entero-H: 80% w/v. Liquid Polibar: 100% w/v Liquid Polibar Plus: 105% w/v Maxibar: 210% w/v Micropaque: 100% w/v Prontobario 60%: 60% w/v. Lower-density suspensions have been designed for delineation of the GI tract during computed tomography (CT) examinations. Their barium sulfate concentration ranges from 0.1 to 5% w/v. Some preparations are ready to use. Other preparations must be diluted with water to reach the appropriate radiological density measured in Hounsfield units.

Battered Child Syndrome

Excipients: natural gum, xanthan gum, sodium carragenine, potassium sorbate, sorbitol, sodium citrate, citric acid, benzoic acid, pectin, simethicone, saccharin sodium, polyoxyethylene glyceryl monooleate, sodium methylparabenzoate.

Flavoring Agents: Vanilla, Caramel, Blueberry, Banana, Apricot, Apple

115

may occur following a uterine rupture or placental abruption. They are associated with the later development of cerebral palsy. The severity of the impairment relates to the extent of the lesions. ▶Hypoxic, Ischaemic Brain Injury

Basedow Goiter

Examples of brand names and composition: E-Z-Cat: 4.9% w/v, that is 4.9 g barium sulfate per 100 ml of suspension. Microcat: 5% w/v Micropaque CT/Scanner: 5% w/v. Ready-Cat 2: 1.2% w/v Volumen: 0.1% w/v

Barrett’s Oesophagus Progressive columnar metaplasia of the distal oesophagus due to long-standing gastro-oesophageal reflux and reflux oesophagitis. The risk of developing oesophageal adenocarcinoma is greatly increased in the presence of these changes. ▶Neoplasms Oesophagus ▶Reflux, Gastroesophageal in Adults

Basal Cell Nevus Syndrome

▶Thyroid Autoimmune Diseases

Basedow’s Disease

▶Thyroid Autoimmune Diseases

Battered Child Syndrome A LAN E. O ESTREICH Cincinnati Children’s Hospital Medical Center, Cincinnati, Ohio, USA [email protected]

Synonyms Child abuse; Nonaccidental trauma; Shaken baby

Genetic disorder, also known as Gorlin’s or Gorlin–Goltz syndrome, characterized by the development of multiple odontogenic keratocysts in association with a symptom complex including skeletal (bifid ribs, synostosis of ribs, kiphoscoliosis, vertebral fusion, polydactyly, hypertelorism, relative frontal bossing and prognathism, midface hypoplasia), cutaneous (multiple basal cell carcinomas), ophthalmologic, neurologic, sexual abnormalities, and ectopic calcifications. ▶Neoplasms, Odontogenic

Basal Ganglia and Thalamic Lesions

Definitions Injury inflicted on a child, most typically in the first year of life, by one or more adults or older children. The trauma results from anger, frustration, aggression, misguided discipline, or occasionally ignorance. Sexual abuse can occur throughout childhood. Neglect or emotional deprivation can also occur at any age. The radiologic findings of child abuse may be simulated by a large variety of diseases, conditions, or situations. It is required of healthcare workers to report the suspicion of child abuse to appropriate authorities.

Pathology/Histopathology The most common site of injury in neonates with HIE. The hallmark of an acute hypoxic-ischaemic event such as

If a battered child dies, forensic autopsy is directed initially toward investigating intracranial, intraabdominal,

B

116


and genital trauma. Photographic documentation of surface lesions is also pertinent, including any recognizable teeth patterns (also in the living battered child). Radiology postmortem can contribute to identifying fractures. Histopathology has value in evaluating the duration of reparative response to fracture. Since periosteal reaction and callus are not visible on radiographs (except on the lung-side of ribs) before about 10 days, uncalcified hemorrhage and early callus may be documented on pathology investigation. As a sequel to abusive injury affecting growth plate and metaphyses, columns or tubes of unossified (formerly physeal) cartilage may persist into the metaphysis during further bone growth. The length of that column on pathology section is a measure of timing since a traumatic event. Pancreatic trauma from abuse may lead to long bone infarction, which may be evident histologically earlier than radiographically. The corner fractures of abuse tend to be Salter II fractures through the junction of the straight bone bark (metaphyseal collar) with the curved bone cortex, and then extend into the physis. Occult rib fractures are often quite posterior and might be sought at autopsy. An important gross and microscopic finding to distinguish Menkes syndrome (congenital copper deficiency) from abuse in infant boys is ▶pili torti, the characteristic kinky hair, which is also brittle. Emergency physicians and pediatricians (and knowledgeable pediatric radiologists) should be alert for this microscopic finding and know its microscopic appearance, lest abuse be diagnosed instead. The appearance is particularly well illustrated in a recent textbook (1). On autopsy, the highly tortuous arteries of this disease would also be evident. Do not do diagnostic angiography injections in boys known to have Menkes. Histology in suspected battering may reveal osteoporosis, hyperparathyroidism, or osteopetrosis as alternate diagnoses for multiple fractures.

in the investigation, especially in interviewing the parents, caretakers, and other pertinent individuals. Criminal justice persons can also be involved and helpful in looking into a suspected diagnosis. The child with abuse fractures is most often under one year of age. An abused child may evidence behavior of fear or avoidance of the parent/caretaker, and the parent/ caretaker may evidence inappropriate behavior toward the child; an alert radiology technologist should communicate any opinion about such interactions promptly to the radiologist. A starved child may eat and drink excessively if permitted in the medical setting. A child may favor a fractured part or limb. The presence of blue sclerae would hint at the alternate diagnosis of osteogenesis imperfecta. Kinky brittle hair in a boy would reflect Menkes disease (Fig. 1). A rachitic “rosary” at the anterior ends of ribs and swollen wrists and knees may signify rickets, usually with secondary hyperparathyroidism. A bogus diagnosis of “temporary brittle bone disease of infancy” or similar names has been proposed to avoid the diagnosis of battered child in some legal cases (2). It is not a medically supported diagnosis. Sometimes, bone densitometry using inappropriate adult standards had been used for documentation. The entire past radiologic record should be rapidly available, to check for evidence of prior trauma or

Clinical Presentation The battered child is generally brought to medical attention in one of the four ways: (1) a person in authority suspects something is amiss; (2) a parent or caretaker seeks medical assistance, often not acknowledging that abuse may have occurred; or (3) a clinical or radiographic finding serendipitously leads a physician or other healthcare worker to suspect the condition. The most extreme presentation is (4) child’s death. Whenever the question of abuse is raised, two things are necessary to do: (1) a thorough clinical and radiologic investigation for findings of abuse and (2) a thorough consideration of alternate (nonabuse) explanations for the findings. Specialized pediatric social workers can play a major role

Battered Child Syndrome. Figure 1 Menkes kinky hair syndrome in a 4-month-old boy with characteristic hair and the diagnostic microscopic pili torti of a hair strand. (From Canepa G, Maroteaux P, Pietrogrande V (2001) Menkes Disease. Dysmorphic-Syndromes and Constitutional Diseases of the Skeleton. Piccin, Padova, pp 1049–1053.)


suspicion of battering. Abnormal eyegrounds should imply a rapid need for head CT, looking for intracranial hemorrhage. Once abuse is suspected, a complete accurately annotated skeletal radiographic survey should be obtained and reviewed straightaway by a radiologist to look for other, occult, fractures. Other children in the same household or nursery as an abused child are also candidates for having been battered. Unfortunately, one presentation of the battered child is the dead child.

Imaging The complete skeletal survey for possible abuse should include at least one view of every bone. A radiologist should immediately review the first set of radiographs, so that orthogonal (90˚ to the original view) images can be added for any area that is suspicious, but not definite, for fracture. For ribs, an appropriate oblique should be requested instead. Confirmation of proper image labeling (name, date, or side of body) at the time of first review is important; especially should eventual testimony become necessary. Important when looking for fractures is appropriate magnification of the images, either routinely or when a site is suspected. In particular, viewing of the small tubular bones is improved. With film, a handheld magnifying glass of at least 2 is useful; with PACS, as much magnification as needed should be employed. An alternative diagnostic method to seek occult fractures is coronal whole-body STIR MRI, which also may reveal soft tissue injuries. One should record an estimated bone age for the subject and look carefully for metabolic bone disease (is the ▶lamina dura around teeth well-defined? are zones of provisional calcification sharply different in density from adjoining bone? or uncalcified?) Do not forget to

117

look for stomach distention, free gas in the abdomen, or other visceral abnormality. On the long bones, the normal ▶step-off (Fig. 2), between the straight metaphyseal collar, generally 1–3 mm long, and the curved metaphyseal shape beyond it toward the shaft is not to be mistaken for fracture; images centered at that metaphysis may improve the differentiation. Carefully check for contour changes in diaphyses that might reflect old, healed fractures. Ultrasound can also document cortical breaks and periosteal reaction from abuse fractures. If the question of intracranial hemorrhage arises because of abnormal eyegrounds, behavior, or history, a CT of the head should be rapidly obtained. On follow-up, a head MRI may document hemorrhage and better define injury to the underlying brain. The abdomen may require emergency ultrasound, CT, or contrast fluoroscopy for clinical questions as well. Since it takes 10 days for periosteal reaction to occur after trauma, even in infants, a repeat skeletal survey after 10–14 days is in order if diagnosis is not certain. Depending on the initial level of suspicion and social circumstances, a bone scan may be employed without waiting those 10 days. It is always prudent to include technologists’ observation of parent (caretaker)–child interaction or unusual parental behavior within the radiology report. The radiologists’ overall impression of the probability or possibility of battering belongs in the formal report.

Nuclear Medicine Bone scanning is helpful both in confirming if a lesion suspected of being a fracture on skeletal radiographic survey is indeed one, and in revealing possible additional fractures. Because physes and other growth plates are

Battered Child Syndrome. Figure 2 Do not mistake the normal step-off at the junction of metaphyseal collar with the curved portion of the metaphysis. A normal step-off at distal fibula of a 1 1/2-year-old child with a diagrammatic representation. (The latter adapted from Oestreich AE, Ahmad BS (1992) Skeletal Radiol 21:283–286(4).)

B

118


normally hot on bone scanning, and because many abuse fractures are localized in the metaphyses at the epiphyses, results may be equivocal at those sites, especially in subjects less than 2 years old. Careful, and perhaps quantitative, comparison of the same site of the contralateral limb may help. Carefully observe also the small tubular bones of hands and feet, vertebral bodies and spinous processes, and especially the ribs for occult fracture activity. Healed fractures can show deformity on radiographs without abnormal activity on nuclear images. Pancreatic scanning may contribute to evaluation of injury of that organ, but ultrasound is more easily performed and often diagnostic, as is CT, which is more frequently done in the United States.

Diagnosis Always be alert for evidence of child battering. Once abuse is suspected be skeptical that another reason may account for the findings. Do not miss abuse and do not overcall abuse. Many skeletal survey findings raise the question of abuse: metaphyseal corner fractures, bucket-handle metaphyseal fractures, fractures of different ages, spiral or transverse shaft fractures in the first year of life, fractures not compatible with the stated history, fractures of small tubular bones in infancy, fractures of ribs, in particular of the first rib, fractures of unusual sites (acromion, clavicle, other scapular, for example), sometimes delay in bone age, occasionally “leukemic” metaphyseal bands, multiple skull fractures, unexplained fractures of vertebral bodies or spinous processes, and otherwise unexplained long bone infarcts (the latter secondary to traumatic pancreatitis). A fracture less than 10 days old will not show callus or periosteal reaction; a fracture more than 10 days old will. The normal physiologic periosteal reaction of infants about 1–6 months of age should not be mistaken for fracture callus, even if the bone is fractured. Note that the portions of ribs adjacent to lung will show soft tissue periosteal reaction before 10 days because it will be soft tissue interfacing with air. Since metaphyseal corner fractures are a characteristic feature of abuse, it is necessary to differentiate them from the normal step-off between the 1–3 mm long straight metaphyseal collar (first described by Laval-Jeantet) adjacent to the physes and the monotonically curved remainder of the metaphysis and diaphysis that has a periosteum. The bucket handle appearance at the metaphyseal region is a corner fracture, perhaps with periosteal reaction, viewed at a different angle than merely perpendicular to the incident X-ray beam. In the differential diagnosis of child abuse, remember that birth trauma simulates most of the findings, with the exception of fractures of different ages. Similarly, rescue

trauma, to pull a baby out of danger, may also simulate abuse, again except for fractures of different ages. A simulator of suspicious first rib fracture is the hox gene variation in which a cervical rib seemingly articulates with an upward process of the subjacent rib; the articulation interface simulates fracture (Fig. 3). Typical toddler fractures once a child begins to walk should not be suspicious, nor should a simple linear skull fracture with appropriate history and no signs of intracranial hemorrhage. Another pitfall would be to mistake the result of emergency intraosseous vascular access radiologic defect in the tibia for battering fracture; history is critical (3). Seek evidence of metabolic or other generalized bone disease with increased fracturability. Most osteogenesis imperfecta patients show osteoporosis and multiple (more than 7–10) wormian intrasutural bones of the skull. Multiple ▶wormian bones and increased bone fracturing are also seen in the rare pycnodysostosis (with broad mandibular angles) and the uncommon Menkes syndrome—congenital copper deficiency, the latter only in boys. If bones are abnormally dense, at least in portions, consider osteopetrosis. Hyperparathyroidism shows loss of the lamina dura around erupted and unerupted teeth; most cases of hyperparathyroidism are secondary to rickets, which should also be manifest on the survey. Scurvy is manifested by generalized osteoporosis, vigorous callus, and Salter I epiphyseolysis, especially at the knee. Periosteal reaction of hypervitaminosis A has an unusual distribution of ulnas, fibulas, and fifth metatarsals, and skull sutures may be separated. Do not forget leukemia as a cause of unexplained fractures, often with osteoporosis or leukemic lucent metaphyseal bands due to generalized illness.

Battered Child Syndrome. Figure 3 Cervical rib simulating abuse. Seven month old girl being examined for possible abuse. The cervical rib nearly synostosing with a process extending upward from the subjacent rib should not be called a rib fracture; it is a hox gene abnormality of rib segmentation. (From Oestreich AE (1996) Radiology 199:582(5).)

Benign Central Venous Thrombosis

Careful perusal of imaging and medical records of an infant may give evidence of prior suspicious episodes. It may even reveal otherwise inexplicable multiple visits for medical care with unusual findings, which would be suspicious of Munchhausen-by-proxy syndrome, in which a parent imposes symptoms or signs upon the child.

Bibliography 1.

2. 3.

4.

5.

Canepa G, Maroteaux P, Pietrogrande V (2001) Menkes Disease. Dysmorphic-Syndromes and Constitutional Diseases of the Skeleton. Piccin, Padova, pp 1049–1053 Mendelson KL (2005) Critical review of temporary brittle bone disease. Pediatr Radiol 35:1036–1040 Harty MP, Kao SC (2002) Intraosseous vascular access defect: fracture mimic in the skeletal survey for child abuse. Pediatr Radiol 32:188–190 Oestreich AE, Ahmad BS (1992) The periphysis and its effect on the metaphysis: I. Definition and normal radiographic pattern. Skeletal Radiol 21:283–286 Oestreich AE (1996) Cervical rib simulating fracture of the first rib in suspected child abuse. Radiology 199:582

Bcl-2 Family Antiapoptotic family members (Bcl-2 and Bcl- xL) and a subgroup of proapoptotic proteins (Bax and Bak) are structurally similar and form a balancing equilibrium in normal cells. ▶Apoptosis

Beckwith Wiedeman syndrome Complex syndrome that is mainly characterized by gigantism and visceromegaly. Craniofacial dysmorphism, macroglossia, omphalocele, and tumors of various organs (liver, kidneys, pancreas, adrenals, SNC, and so on) have been reported. At the level of the pancreas, cysts, pancreatoblastoma, and nesidioblastosis have all been reported in association with the syndrome. ▶Congenital Anomalies of the Pancreas

119

Bed-Side Imaging As severly sick patients, particularly those on ECMO, are extremely difficult to transport, imaging is nearly entirely performed at the patients’ bed-side using mobile devices.

Bell’s Palsy Isolated, sudden, unilateral, peripheral facial nerve paralysis caused by a viral-induced neuronitis. The induced inflammatory response leads to compression of the nerve against the rigid walls of the fallopian canal, ischemia and axonal degeneration HSV is the most commonly implicated agent thought to be in a latent state in the geniculate ganglion, reactivated by non-specific triggering events. It is the most common cause of FNP, with a reported annual incidence of 15 to 40:10,000 people. ▶Facial Nerve Palsy

Benign and Malignant Nodes in the Neck ▶Lymphadenopathies, Head and Neck

Benign and Malignant Ovarian Tumors ▶Masses, Ovarian

Benign Bone Tumors ▶Neoplasms, Bone, Benign

Bed Rest Conservative measure aimed at reducing disc herniation by axial unloading of the spine. ▶Conservative Therapy for Lumbosacral Radicular Syndrome

Benign Central Venous Thrombosis Thrombosis on the basis of a benign etiology, often hemodialysis related. Nowadays also increasingly

B

120

Benign Peripheral Sheath Tumours

observed in patients with permanent central venous access catheters and pacemakers. ▶Occlusion, Venous Central, Benign

Benign Peripheral Sheath Tumours Benign tumours originated from the peripheral nerves or soft tissues. This term includes schwannomas, neurofibromas and perineuromas. ▶Breast, Benign Tumours

Benign Prostatic Hypertrophy ▶Hyperplasia Benign Prostate

Benign Teratoma

extrahepatic bile duct adenomas are tubular. They are composed of intestinal-type glands and are lined by pseudostratified columnar epithelium appearing as intraluminal, isoechoic relative to liver parenchyma, nonshadowing masses on US images. On direct MR or MR cholangiography images a biliary adenoma appears as a polypoid filling defect within the bile duct with lobular (in the case of tubular adenoma) or cauliflower-like (in the papillary type) contour. ▶Neoplasms, Bile Ducts

Bile Duct Cyst When evaluating MR mammography, morphologic and dynamic criteria should be taken into account as classified by the MRM-BI-RADS. ▶Congenital Malformations, Bile Ducts

Bile Duct Cystadenocarcinomas

▶Teratoma, Ovaries, Mature, Ovalar

Bezoar A ball or clump of swallowed foreign matter, most often hair or vegetable fibrous material, and usually retained within the stomach although they can occur at any point in the gastro-intestinal tract. ▶GI tract, Pediatric, Foreign Bodies

BI-RADS MRM When evaluating MR mammography, morphologic and dynamic criteria should be taken into account as classified by the MRM-BI-RADS. ▶MR Mammography

Bile Duct Adenomas, Extrahepatic Rare benign tumor usually located in the common bile duct and the common hepatic duct. The majority of

Biliary cystadenocarcinoma is a rare cystic malignant tumor arising from bile duct epithelium. It typically occurs in middled-age women and causes no symptoms until it grows to a huge mass compromising adjacent anatomical structures. Since local or metastatic spread is rare, patients are usually referred for surgery. It usually appears as a large cystic lesion with intralesional septa on US, CT, and MR images, although there are no specific imaging features allowing a reliable differentiation of biliary cystadenoma from cystadenocarcinoma; the demonstration of mural nodules is suggestive of a malignant form. However, the diagnosis is usually obtained by histological analysis of the resected cystic mass. ▶Neoplasms, Bile Ducts

Bile Duct Cystadenomas Biliary cystadenomas are uncommon unilocular or multilocular benign cystic neoplasms that may occur within the liver, extrahepatic biliary tree, or gallbladder. Biliary cystadenomas range in size from 3 to 40 cm. On US they appear as unilocular or multilocular cystic lesions with intralesional septa containing low-level echoes from blood products, mucin, or proteinaceous fluid. On CT a

Biliary Anatomy

large, low-attenuating mass with lobulated margins and irregular walls with fibrous septa enhancing after contrast medium injection is usually observed. On MR imaging the tumor appears as a uniloculated or multiloculated cystic mass. The MR signal intensity of biliary cystadenoma may vary on both T1- and T2-weigthed images, depending on the content of the cystic fluid. ▶Neoplasms, Bile Ducts

Bile Duct Obstruction ▶Occlusion, Bile Ducts

121

tissue. Passage of bile into the bloodstream is due to an increased biliary-venous gradient and occurs during the resorption of the biliohematoma. Clinically, bilhemia is characterized by the rapid development of jaundice with a marked increased in the total serum bilirubin level. When a considerable amount of bile enters the bloodstream, it may cause an embolism in the lungs and kidneys. If untreated, bilhemia has a high mortality. Endoscopic retrograde or direct cholangiography is the imaging modality of choice for the demonstration and localization of a biliovenous fistula. Magnetic resonance cholangiography and biliary scintigraphy may also be useful, whereas arteriography is usually nondiagnostic. Currently, interventional radiological procedures, particularly percutaneous biliary stenting, are frequently employed for the treatement of posttraumatic bilhemia. ▶Trauma Hepatobiliary

Bile Duct Papillomatosis Biliary Anatomy Biliary papillomatosis is a rare disorder characterized by multiple and recurrent papillary adenomas in the biliary tract. The extrahepatic bile ducts are involved in most cases. The most common imaging feature is intra- and extrahepatic dilatation. Complete surgical excision of biliary papillomatosis is difficult and local recurrence is common. Papillomatosis has a greater potential for malignant transformation than solitary adenoma. ▶Neoplasms, Bile Ducts

DAVIDE C ARAMELLA , PAOLA VAGLI , C ARLO B ARTOLOZZI Department of Diagnostic and Interventional Radiology, University of Pisa, Italy [email protected]

Synonyms Diagnostic imaging of the biliary tree (bile ducts; gallbladder)

Bile Duct Tumors Definition Bile duct tumors include both benign (bile duct adenoma, bile duct cystadenomas) and malignant (bile duct cystadenocarcinomas, cholangiocellular carcinoma) neoplasms arising from the bile duct epithelium. ▶Neoplasms, Bile Ducts

Modality of depiction of the normal biliary structures by means of diagnostic imaging.

Characteristics Gross Anatomy

Bilhemia Bilhemia is a rare clinical entity usually caused by a pathologic communication between intrahepatic bile ducts and the hepatic venous system following the formation of an extensive hematoma within necrotic

The right and the left hepatic ducts are formed by the junction of the intrahepatic bile ducts, and these join at the porta hepatis to form the common hepatic duct. The common hepatic duct, about 2.5 cm below its origin, is connected with an acute angle to the gallbladder by means of the cystic duct, forming the common bile duct or choledochal duct. The cystic duct runs almost parallel to

B

122

Biliary Anatomy

the common duct and usually joins it at the inferior aspect of the porta hepatis or sometimes at a lower level. The choledochal duct extends for about 7 cm from the cystic duct to the duodenal papilla, running downward in the right free edge of the gastrohepatic ligament or lesser omentum (on the left of the Winslow foramen), with the hepatic artery to its left and the portal vein behind it. Inferiorly, its course is firstly located behind the duodenum and then in a groove of the head of pancreas before entering the second part of the duodenum. At the duodenal papilla it joins to the pancreatic duct, although occasionally they remain separate. Branches from the hepatic artery provide arterial supply to the biliary tree, while venous drainage is by means of the portal venous system. The gallbladder lies at the inferior edge of the right hepatic lobe (in the gallbladder fossa), lateral to the quadrate lobe (segment IV) in relationship to the superior border of the proximal transverse colon. It is also connected to the anterolateral aspect of the gastric antrum, duodenal bulb, and proximal descending duodenum. The gallbladder is held against the bottom surface of the liver by peritoneum, and is surrounded by fat. Its size is about 7 to 10 cm in length and it has a capacity of approximately 30 mL, although it may vary in size and shape often according to the patient’s stature. It is subdivided into three portions: the fundus, body, and neck. The neck is S-shaped and connects with the cystic duct within which lies the spiral valve of Heister. The fundus is directed downward, forward, and to the right and it comes into close relationship with the anterior abdominal wall. The body is directed upward and backward to the left, and near the porta hepatis is continuous with the neck. The gallbladder wall is structured into three layers and is composed of the mucosa, the fibromuscular layer, and the serosal layer. The mucosa presents small sulci that combine, generating a honeycomb aspect. Arterial supply is performed by means of the cystic artery that is a branch of the hepatic artery. Veins from the gallbladder join the tributaries of the portal and hepatic veins. The gallbladder and biliary ducts are both reached by parasympathetic and sympathetic nerves (1).

Ultrasound The larger intrahepatic biliary ducts may be seen arborizing into the hepatic parenchyma and can be differentiated from the intrahepatic veins that have no obvious walls. Distinction between biliary ducts and the other vessels is performed thanks to their typical branching pattern and the presence of Doppler signal. The normal choledochal duct is hard to demonstrate for its small caliber (0.5–1.1 cm) and oblique course. In sagittal sections it can be seen anteriorly to the inferior vena cava and crossing in front of the portal vein. A dilated choledochal duct may be

distinguished from the portal vein by means of serial scans demonstrating the communication of the portal vein with the confluence with the mesenteric and splenic veins and the presence of the posterior “dip” of the choledochal duct as it passes downward behind the first part of the duodenum. In transverse scans the small, circular, welldemarcated, echo-free ring produced by the choledochal duct may be recognized medial to the gallbladder, lateral to the inferior vena cava, and in front of the portal vein. Also, transverse scans at the level of the major axis of the pancreas may show the distal end of the choledochal duct on the posterior aspect of the head of the pancreas with the gastroduodenal artery placed anteriorly. Size and position determine the capability to visualize the gallbladder. Normal bile is anechoic allowing us to distinguish the gallbladder from the normal acoustic texture of the liver parenchyma. A normal, physiologically dilated gallbladder should be demonstrated on longitudinal, transverse, and oblique scanning in fasting patients. Intercostal scans in the transverse plane are of particular value for patients in whom the liver lies deeply under the costal margin. The gallbladder shows many variations in location and size. It is usually placed anterior to the right kidney and it can also have different relationships with the liver (subhepatic, intrahepatic, or mesenteric gallbladder). It appears as a well-demarcated, smooth-walled, pear-shaped, and fully transonic area, lying obliquely on the inferior surface of the right lobe of the liver in sagittal scans. The fundus presents a rounded contour, while the body has a semilunar shape and diminishes to a varying degree as it passes upward and backward to form the gallbladder neck. In this region the presence of spiral valves may be revealed in cases of dilatation. The gallbladder neck invariably lies immediately anterior and inferior to the portal vein. It may be difficult to visualize because of the presence of an acoustic shadow at this level (probably due to the collagen arrangement of the spiral valve) that should not be mistaken with stones. On axial sections with the transducer angled cranially, it is possible to show the pear shape of the gallbladder sectioned along its long axis lying anterior to the right kidney (1). Ultrasound (US) shows a threelayered gallbladder wall consisting of a strongly reflective outer layer, a minimally reflective inner layer, and an anechoic layer between. The wall thickness is usually less than 2 mm.

Computed Tomography The normal intrahepatic biliary radicles are not seen on computed tomography (CT) scans, whereas the normal common hepatic or the common bile duct can be seen as a water-dense structure, varying in diameter from 6 to 8 mm. The common hepatic duct lies anterolateral to the portal vein in the region of the liver hilus, while the caudal

Biliary Anatomy

end of the common bile duct lies within the head of the pancreas, medial with respect to the second duodenal portion. Unlike with the hepatic vessels, the intravenous administration of iodinated contrast medium determines no elevation in the attenuation value of the near waterdense biliary tree. However, a better depiction of mild dilatation of the biliary tree may be acquired after contrast medium administration, obtaining parenchymal enhancement. On CT scans the gallbladder is shown as an oval or elliptic sac of near-water density lying in a fossa on the inferior surface of the liver. When its walls are contracted it may be difficult to localize and can appear ill defined. The gallbladder neck is directed superomedially, while the fundus projects inferolaterally. In more caudal scans, the gallbladder assumes a more anterior position and occasionally may extend beyond the inferior margin of the liver near the second duodenal portion. The thickness of normal gallbladder walls is nearly imperceptible. Recent advances in three-dimensional CT (single and multirow CT) in terms of thin section, single breath hold acquisitions, and threedimensional and multiplanar reformation techniques suggest the possibility of CT ▶cholangiography for an elective visualization of the biliary tree after the intravenous or oral administration of cholangiographic iodinated contrast medium. Because of its high contrast resolution, CT enables visualization of the bile ducts that are also opacified with oral cholecystographic contrast medium before its concentration in the gallbladder (2).

Cholangiography This technique was firstly introduced in 1954 as intravenous cholangiography for the noninvasive visualization of the common bile duct. It consists in intravenous injection of a dedicated iodinated contrast medium that is excreted by hepatocytes allowing the opacification of biliary ducts. Therefore, to obtain an adequate opacification of the biliary tree, liver function must be reasonably good. Gallbladder visualization was not included in cholangiography studies and in the preUS/CT era this structure was visualized by means of oral ▶cholecystography, now considered obsolete. Intravenous cholangiography has been replaced by direct cholangiography, representing a way of direct visualization of the biliary tract (intra- and/or extrahepatic), by means of introduction of iodinated contrast medium in the biliary system that is outlined on radiographs. This technique includes different methods: . ▶Percutaneous transhepatic cholangiography (PTC) (percutaneous catheterization of an intrahepatic bile duct) . ▶Endoscopic retrograde cholangiopancreatography (ERCP) (endoscopic catheterization of the common bile duct via Vater’s papilla)

123

. Perioperative cholangiography (catheterization of the cystic duct during surgical procedures) . Trans-Kehr cholangiography (injection of iodinated contrast medium directly into the surgical T-tube). Cholangiography allows the whole course of the biliary tree to be depicted with high fidelity and is usually performed to assess the presence, location, extension, and cause of biliary obstruction, bile extravasation in case of biliary leakage, and less frequently for evaluating congenital malformations of the biliary tract. Today, cholangiography is usually a part of interventional or endoscopic procedures and has been replaced by noninvasive methods [US, CT, and ▶magnetic resonance cholangiopancreatography (MRCP)] in the diagnostic work-up of biliary tree diseases.

Magnetic Resonance Cholangiography and Magnetic Resonance Cholangiopancreatography Magnetic resonance cholangiography (MR cholangiography) was introduced in 1991 by Wallner and has emerged as an accurate, noninvasive alternative to diagnostic ERCP in the evaluation of diseases of the biliary tract. MR cholangiography is performed with the use of heavily T2-weighted images obtained with different pulse sequences (gradient echo sequences with the steady-state free precession, two-dimensional and three-dimensional heavily T2-weighted fast spin echo sequences, half Fourier rapid acquisition with relaxation enhancement RARE sequences) without the injection of contrast medium that demonstrate the fluid-containing bile ducts as highsignal-intensity structures. Stationary fluids, including bile and pancreatic secretions, have high signal intensity whereas solid organs have low signal intensity. This combination of imaging characteristics provides optimal contrast between the hyperintense signal of bile and the hypointense signal of the background. This technique also allows visualization of the pancreatic ducts generally after the administration of a secretive agent (secretin) and is then called MRCP. MRCP combines the benefits of projectional and cross-sectional imaging techniques providing an overview of the entire biliary and pancreatic ductal system by means of data acquisition and image reconstruction rendered on the coronal plane in a conventional cholangiography fashion allowing high-quality imaging comparable to that of ERCP. A comparable diagnostic accuracy with respect to ERCP in the diagnosis of choledocholithiasis is reported by various authors for MRCP. MR cholangiography may help to establish the diagnosis of malignant obstruction and is useful in evaluating those patients in whom ERCP was unsuccessful or incomplete. Moreover, MR cholangiography plays a crucial role in evaluating postsurgical

B

124

Biliary Anatomy

biliary tract alterations and can demonstrate a variety of congenital anomalies of the biliary tract (aberrant ducts, choledochal cysts, Caroli disease, pancreas divisum). In addition, intentional or incidental imaging of the gallbladder with MRCP can be used to identify calculi or help determine the presence and extent of neoplastic disease (3).

Image Processing Biliary tract examinations using rapid three-dimensional volumetric techniques can be performed with threedimensional CT (single and multirow CT) and MR cholangiopancreatography. CT techniques are especially useful when MR is unavailable or contraindicated, or when the quality of MRCP images is suboptimal (CT cholangiography has a better spatial resolution, hence its clearer depiction of small ducts). However, to date, the use of radiation, the low availability of biliary contrast medium, and the possible adverse reactions to iodinated contrast agents have all had a negative impact on a largescale diffusion of the technique. Integration of volumetric acquisition and three-dimensional surface and volume rendering techniques facilitates the study of the biliary tract in a noninvasive way. One of the advantages is the possibility to process these volumetric data sets so to generate three-dimensional reconstructions, which can be obtained with both external and endoluminal views of the organ anatomy. Maximum intensity projection (MIP), minimum intensity projection (Min IP), shaded surface display (SSD), and volume rendering (VR) algorithms provide external views of the pancreaticobiliary tract. Endoluminal views can be obtained with ▶virtual endoscopy (VE). MIP and SSD provide a global map of the pancreaticobiliary tree anatomy that can be helpful in the interpretation of two-dimensional MRCP images (particularly postsurgical anatomy, anatomic variants of the pancreaticobiliary ductal system, and large biliary neoplasms). However, various authors have stressed the limitations of such algorithms in the depiction of small intraductal pathology, and particularly in the detection of small calculi in the common bile duct. VR is the latest development in three-dimensional imaging of the anatomy of the pancreaticobiliary tree, allowing radiologists to display simultaneously different anatomical structures imaged within a single volume. VR has the advantage of providing a clear roadmap of the entire biliary tree, but its real role in demonstrating pathological changes deserves further clinical evaluation. Virtual simulation of fiberoptic endoscopy can be obtained with a new software tool based on surface or volume rendering techniques, called virtual cholangioscopy, when specifically applied to the study of the biliary system. Endoluminal views of the

pancreatic and bile ducts can be obtained by rendering CT or MRI data sets. The application of the VE surface algorithm to three-dimensional CT data sets with proper setting of the thresholds allows rendering of the interface pixels only, i.e., those pixels on the border between endoluminal fluid (bile, pancreatic secretions) and surrounding soft tissue that remains undisplayed. Segmentation of MRCP data sets has been made feasible by the high contrast difference between the bile ducts (brightest voxels) and the surrounding tissue (darkest voxels). Navigation sequences can be simulated through the common bile duct, common hepatic duct, left and right hepatic ducts, intrahepatic branches, pancreatic duct, cystic duct, and gallbladder. All these anatomical details appear as tubular structures, with a smooth internal surface. The confluence between these branches can also be represented. In most cases three-dimensional imaging does not add quantitative information to MRCP, but the main advantage over crosssectional anatomy is represented by a different rendering of data, more familiar to the human eye. Constant integration between two-dimensional source images and three-dimensional reconstructions, both with external and endoluminal views, represents an added value to the diagnostic potentialities of MRCP in the evaluation of biliary tract diseases (4).

Nuclear Medicine Hepatobiliary scintigraphy is a physiological test that evaluates hepatic bile formation, excretion, and biliary tree patency. Images are acquired in a dynamic fashion for up to 4 h after the intravenous administration of a technetium-labeled iminodiacetic acid (IDA) derivative that is an organic anion moving from the plasma into bile like bilirubin. Two 99mTc-IDA agents (disofenin and mebrofenin) are widely used showing a homogeneous liver uptake and an excellent visualization of the gallbladder and the cystic and common bile duct within 60 min of intravenous tracer injection. Following intravenous injection, these organic anion tracers are rapidly bound to plasma protein, circulate to the liver, dissociate from their binding proteins, and are taken up by hepatocytes (peak hepatocyte uptake at 10 min). After processing within the hepatocyte through the ligandin system and possible glucuroconjugation (as occurs for bilirubin), these agents are rapidly secreted into the bile providing excellent visualization of the biliary tract (90% visualization of the common bile duct and gallbladder at 30 min and 100% at 60 min after tracer injection). Patients undergoing cholescintigraphy should have fasted for a minimum of 2–4 h before tracer injection. Anterior images of the liver and biliary tract are routinely obtained at 5 and then at 10–15-min intervals for 1 h after tracer

Biliary Tuberculosis

injection or in dynamic acquisition. Intravenous administration of morphine sulfate can be performed in cases of nonvisualization of the gallbladder when adequate hepatic uptake and excretion into the bowel is documented. Morphine induces contraction of the sphincter of Oddi raising intrabiliary pressure and filling the gallbladder if the cystic duct is patent, followed by imaging at 30 and 45 min. Alternatively, delayed images at 3–4 h can be obtained to confirm the presence or absence of cystic duct patency with similar accuracy. This test may be nondiagnostic in patients with liver failure and intrahepatic cholestasis of any cause because of the inability to conjugate and excrete the radiotracer.

Bibliography 1.

2.

3.

4.

Taylor KJW, Rosenfield AT (1979) Ultrasound liver and biliary tract. In: Margulis AR, Burhenne HJ (eds) Alimentary Tract Radiology. Mosby, St. Louis, pp 183–197 Chopra S, Chintapalli KN, Ramakrishna K et al (2000) Helical CT cholangiography with oral cholecystographic contrast material. Radiology 214(2):596–601 Fulcher AS, Turner MA, Capps GW (1999) MR Cholangiography: technical advances and clinical applications. Radiographics 19 (1):25–44 Boraschi P, Lodovigi S, Campori G et al (2002) Biliary tract. In: Caramella D, Bartolozzi C (eds) 3D Image Processing. Springer, Berlin Heidelberg, pp 223–232

Biliary Atresia Most frequent cause of persisting neonatal cholestasis, characterized by obliteration or discontinuity of the extrahepatic biliary system (extrahepatic biliary atresia). If untreated secondary biliary cirrhosis develops. ▶Congenital Malformations, Liver and Biliary Tract ▶Congenital Malformations, Bile Ducts

Biliary Hamartoma Bile duct hamartomas are small clusters of slightly dilated bile ducts surrounded by a fibrous tissue. They do not communicate with the biliary tree. At pathologic analysis, they appear as grayish-white nodular lesions 0.1–1.5 cm in diameter scattered throughout the liver parenchyma. Bile duct hamartomas are rather common and usually represent an incidental finding at imaging examinations. At ultrasound biliary hamartomas may present as either hypoechoic or anechoic small nodules; in rare cases they have a hyperechoic appearance. CTshows hypodense small hepatic nodules. At MR the lesions are hypointense

125

on T1-weighted images and markedly hyperintense on T2-weighted images. Von Meyemburg complexes: Biliary hamartoma. ▶Cystic-Like Lesions, Hepatic

B Biliary Hypoplasia Reduction of the caliber of the intra and extrahepatic ducts or both resulting in an inadequate biliary drainage. The degree of the damage depends on the severity of the hypoplasia. ▶Congenital Malformations, Liver and Biliary Tract

Biliary Lithiasis ▶Gallstones

Biliary Stones Particulate, sand-like matter that forms from precipitation of solutes in bile. The clinical course is extremely variable. Possible outcomes are complete resolution, progress to gallstone formation, and complications, including intermittent abdominal pain, acute pancreatitis, and acute cholecystitis. Risk factors associated with biliary sludge are rapid weight loss, pregnancy, ceftriaxone therapy, and transplantation. At US biliary sludge is easily diagnosed: it presents as echoic sediment with or without acoustic shadowing in the dependent portion of the gallbladder and shifts slowly with positioning. ▶Gallstones

Biliary Tuberculosis Tuberculosis infection of the biliary tract is extremely rare. The presence of a high concentration of bile acids, a competent sphincter of Oddi, a rapid passage of bacilli through the duodenum and presence of pancreatic secretions represent protective factors against bile infection by Mycobacterium tuberculosis. The possible etiopathogenesis include: ascending or descending infection through the biliary tract, direct infection from a neighbouring focus,

126

Biliary–respiratory Congenital Communication

hematogenous and lymphatic spread. Tuberculous cholecystitis is usually associated with gallstones. In fact, bacilli form a nidus for the stone formation. Biliary tuberculosis has nonspecific clinical and radiological features and therefore its preoperative diagnosis is very challenging. The definite diagnosis is usually based on intraoperative and histopathological findings: evidence of caseating granulomatous inflammation with bile cytology revealing Mycobacterium tuberculosis is confirmatory. Most of the cases present with a clinical setting consistent with cholecystitis, mainly right upper abdominal pain, nausea, vomiting, and low-grade fever. In such cases US usually reveals features of acute or chronic cholecystitis. Bile duct dilatation may be also present. In most cases, it is due to compression by enlarged tuberculous lymph nodes. In some instances, biliary infection may produce multiple irregular biliary strictures. In these cases, the visualization at US and CT of periductal ill-defined soft tissue material may lead to the wrong diagnosis of cholangiocarcinoma. ▶Occlusion, Bile Ducts

Biliary–respiratory Congenital Communication Rare congenital abnormality where an anomalous tract connects the tracheobronchial tree with the biliary tree. The common sign is foamy bile sputum. ▶Congenital Malformations, Liver and Biliary Tract

Biliary-Vascular Fistula Biliary–vascular fistula is an abnormal communication between the biliary tree and the intrahepatic vascular system. Biliary–vascular fistulas are rare and usually result from interventional radiology procedures such as percutaneous liver biopsies and biliary stenting. They include both arteriobiliary and venous- or portobiliary communications, and two clinical entities, such as hemobilia and bilhemia, have been distinguished. In fact, the direction of flow through the pathologic biliary–vascular communication depends on the pressure gradient between the two systems (vascular and biliary). Because blood pressure generally exceeds pressure in the bile ducts, hemobilia is, by far, the more common problem. Flow in the opposite direction, bilhemia occurs in the rare patient in whom the normal pressure gradient is inverted, directing bile into the hepatic veins or the portal vein. ▶Trauma Hepatobiliary

Billroth’s Gastrectomy Type I Removal of lower portion of stomach (pylorus) with end to end anastomosis of the remaining stomach with the duodenum. ▶Stomach and Duodenum in Adults Postoperative

Billroth’s Gastrectomy Type II Subtotal excision of the stomach with closure of the proximal end of the duodenum and side-to-side anastomosis of the jejunum to the remaining portion of the stomach. ▶Stomach and Duodenum in Adults Postoperative

Biloma Biloma is a perihepatic or intrahepatic bile collection resulting from damage of the biliary tree and bile leakage. As a result of the slow rate of leaking, a biloma may take weeks or months to develop after trauma; hence, it usually is diagnosed by using follow-up scans. Clinical manifestations depend on the location and size of the lesion. At ultrasound, computed tomography (CT), and magnetic resonance (MR) imaging, biloma usually appears as a rounded or ellipsoid, fairly welldefined, loculated cystic mass localized in close proximity to the liver and bile duct. Hepatobiliary scintigraphy and MR cholangiography enable demonstration of the precise site of bile leakage; provide useful information in cases of suspicious biliary leakage. Although minor biloma usually resolves spontaneously, large or increasing biloma may require US- or CT-guided percutaneous drainage. ▶Trauma Hepatobiliary

Bioluminescence C LEMENS LO¨WIK , E RIC K AIJZEL Endocrinology Research Laboratory and Molecular Imaging, Leiden University Medical Center, The Netherlands [email protected]

Bioluminescence

Definition Bioluminescence refers to the enzymatic generation of visible light by living organisms. The most commonly used bioluminescent reporter gene for research purposes has been luciferase from the North American firefly Photinus pyralis. The enzyme catalyzes the transformation of its substrate D-luciferin into oxyluciferin in an oxygen and ATP-dependent process, leading to the emission of photons (Fig. 1). Biological sources of light (bioluminescence) have sufficient intensity to cross animal tissues provided that the endogenous light has a wavelength >500 nm. Above this wavelength, tissue absorption of photons decreases. The firefly luciferase/luciferin reaction emits photons at wavelengths from 500 to 620 nm and is therefore suitable for external detection. Apart from fulfilling this criterion, another important finding was that the substrate luciferin, after intravenous or peritoneal injection into animals, diffuses within a few minutes throughout the entire body and is rapidly taken up by cells (1). Luciferin, which is not immunogenic, also crosses the blood–brain barrier, although

127

this takes more time (10–15 min); therefore, luciferase can be used as a reporter in brain-related research. Bioluminescence imaging (BLI) of luciferase reporters provides a relatively simple, robust, cost-effective, and extremely sensitive means to image fundamental biological processes in vivo due to exceptionally high signal-to-noise levels. Apart from firefly luciferase, other luciferases with matching substrates have been found useful for molecular imaging, including Renilla luciferase, green or red click beetle (Pyrophorus plagiophthalamus) luciferases, and Gaussia luciferase. Renilla and Gaussia luciferases use a different substrate, coelerentarazine, and emit mainly blue light. The emission spectrum is very broad, and a part also exceeds the wavelength of 550 nm, making it suitable for BLI. Renilla and Firefly luciferase can also be used in combination as dual reporters for BLI (2).

Imaging Recent technical advances for imaging weak visible light sources using cooled charged coupled device (CCCD)

Bioluminescence. Figure 1 Biological principle of bioluminescent imaging of luciferase expression. (a) Firefly luciferase converts the substrate luciferin into oxyluciferin and light. Tissue-specific transcription factors bind to regulatory sequences cloned directly adjacent to the luciferase gene, resulting in tissue-specific expression of luciferase, which can be visualized by injection of luciferin and detection of the emitted light by cooled charged coupled device cameras. (b) Image of a transgenic (ERE-luc) mouse during proestrus expressing the luciferase gene under the control of an estrogen-responsive element. Note the intense staining of the ovaries, which subsides during estrous. Reprinted from de Boer, Blitterswijk, Lowik. Biomaterials 27 (2006) with permission from Elsevier.

B

128

Bioluminescence

cameras, peltier-cooled detectors, and microplate channel intensifiers allow detection of bioluminescent emission from inside the tissues of an animal (Table 1). These technical developments have made it possible to monitor gene expression in the living animal via a lucreporter gene linked to specific promoters or to follow in real time the fate of luciferase (luc)-transfected tumor cells or immune cells injected in living animals (Fig. 1). One of the many advantages of this methodology is that it is noninvasive and therefore allows investigations in the same animal at different time points. Consecutive analysis of the same animal means that fewer animals are needed for each study and that experimental uncertainties arising from interanimal variations are greatly reduced. This

technology is extremely useful in many research areas such as cancer research, as it focuses on detecting primary tumor growth and on micro- and macrometastatic tumor spread (Fig. 2).

Cell Trafficking BLI can also be used to follow in real time the fate of luciferase-transfected cells injected in living animals. This application is widely applied in cancer research to follow not only tumor progression but also the metastatic process by which tumor cells metastasize to distant tissues or organs. It is also used to follow the migration and fate

Bioluminescence. Table 1 Several currently available BLI systems Company Xenogen Berthold Hamamatsu + Improvision Roper Scientic Biospace Kodak

System IVIS NightOwl VIM Camera Model C2400-47 ChemiPro PhotoImager In vivo FX

Technology Liquid nitrogen-cooled CCD camera Peltier-cooled CCD camera Intensified CCD camera Cryogenic-cooled CCD camera Intensified CCD camera Thermoelectric CCD camera

Bioluminescence. Figure 2 The Arguello nude mouse model for bone metastasis (a) Injection of 100,000 human MDA-MB-231 cancer cells into the left cardiac ventricle of immunodeficient (nu/nu) mice will lead to bone/bone marrow metastases (From Arguello F, Baggs RB, Frantz CN (1988) A murine model of experimental metastasis to bone and bone marrow. Cancer Res 48:6876–6881). (b) Intracardiac injection of MDA-MB231 cells stably expressing luciferase will lead to metastases of luc-expressing cancer cells, of which progression can be noninvasively monitored using bioluminescent imaging. Reproduced with permission from Lowik, Cecchini, Maggi, Pluijm. Ernst Schering Res Found Workshop 49 (2005).

Bioluminescence

of immune cells or transplanted stem cells, such as neuronal stem cells in the brain after ischemic stroke or embryonic stem cells in heart regeneration.

Gene Expression BLI makes it possible to monitor gene expression in living animals via a luciferase reporter gene linked to cell-, tissue-, differentiation- or other process-specific promoters. These constructs can be either transfected into cells that can be transplanted into animals or be used to make transgenic gene-reporter mice.

Transgenic Gene-Reporter Mice Using luciferase-based specific gene reporters, it is possible to transfect embryonic stem cells with such a construct and make transgenic gene-reporter mice. For instance, we generated mice transgenic for the luciferase gene under the control of an estrogen-responsive element (ERE-luc)n (3). Luciferase activity in individual mice could be followed throughout the estrous cycle, and a peak of luciferase activity was observed during proestrus in reproductive tissues (Fig. 1). Using ERE-luc mice, we were able to monitor estrogen receptor activity dynamics

129

in a tissue-specific and quantitative way, demonstrating the power of BLI. Recent advances in small animal imaging instrumentation, molecular genetics, and reporter gene design have yielded the ability to integrate imageable reporters like luciferase into various transgenic mouse models that resemble human diseases. (Some examples are listed in Table 2).

Bioluminescent Tomography The current BLI systems use photographic principles to capture light emitted from the animals at sites where luciferase is expressed. Photons are detected at the surface of the animal in two dimensions, also known as planar imaging, the simplest technique for detecting optical reporter molecules in vivo. Photon attenuation, however, is strongly nonlinear as a function of depth and of the optical heterogeneity of tissue, which hampers signal quantification. Planar imaging is further complicated by the inability to resolve depth and by tissue scattering and absorption, which limits spatial resolution. A way to circumvent the problems of 2D planar BLI would be a combination of multiview angle imaging (bioluminescence tomography) with a priori information on tissue heterogeneity. This can now be achieved using the newly

Bioluminescence. Table 2 Examples of transgenic BLI reporters for models of human disease Name Transgenic mice hOC-luc BMP-4 colI(1)-luc BSP-luc VEGFR2-luc VEGF-luc PSA-luc Epx* or Epo-luc Saa1-luc IL-2-luc TNF-luc iNOS-luc NFkB-luc Ho1-luc or Hmox1-luc SOD1-luc Cyp3A4-luc RIP-luc Retn-luc GFAP-luc ER-DEVD-luc

Regulatory Element

Application

Osteocalcin promoter Bone morphogenetic protein-4 Bone-specific enhancer of the mouse collagen (α1) I gene BSP promoter VEGFR2 gene promoter VEGF gene promoter PSA promoter Epo promoter Serum amyloid A-1 promoter Interleukin-2 TNF promoter iNOS gene promoter 3 NFkB site from the Igk light chain promoter Heme oxygenase-1promoter Superoxide dismutase promoter Cytochrome p450 isoform 3A promoter Rat insulin gene promoter Resistin promoter GFAP-promoter Caspase cleavage sequence

Bone repair and development Bone repair and development Matrix deposition Matrix deposition Angiogenesis Angiogenesis Oncology Inflammation Inflammation Inflammation Inflammation Inflammation Inflammation Drug metabolism/toxicology Drug metabolism/toxicology Drug metabolism/toxicology Metabolic disease Metabolic disease Neurology Apoptosis

B

130

Bioluminescence Imaging

Bioluminescence. Figure 3 Using Xenogen IVIS 3D Living Image Software, reconstructions of a RC21-luc expressing tumor in the right kidney and its metastases in the urogenital area are displayed as voxels (dots) with a red–black color bar, red indicating the highest intensity. Cross-sectional planes and depth locations are shown on the left.

developed 3D BLI systems from Xenogen (Fig. 3). If these 3D images are combined with 3D spatial (tissue) information obtained by computed tomography or magnetic resonance imaging, algorithms can be developed to more precisely correct for photon scattering and absorption due to tissue heterogeneity and depth.

antiangiogenic therapy, and when successful can be a first step toward clinical application.

Bibliography 1.

2.

Diagnosis Noninvasive in vivo BLI is a powerful tool in small animal models of human biology and disease. BLI is perfectly suited to monitor gene expression in transgenic reporter mice and to detect and follow small numbers of cells noninvasively. In cancer, BLI enables researchers to follow tumor progression and metastasis and can help identify in vivo molecular targets of cancer and their metastases. The application of BLI in combination with new animal models for cancer will allow us to study very rapidly and conveniently the efficacy of new therapeutic approaches such as gene therapy, stem cell therapy, and

3. 4.

Contag CH, Spilman SD, Contag PR et al (1997) Visualizing gene expression in living mammals using a bioluminescent reporter. Photochem Photobiol 66:523–531 Bhaumik S, Gambhir SS (2002) Optical imaging of Renilla luciferase reporter gene expression in living mice. Proc Natl Acad Sci USA 99:377–382 Ciana P, Raviscioni M, Mussi P et al (2003) In vivo imaging of transcriptionally active estrogen receptors. Nat Med 9:82–86 Arguello F, Baggs RB, Frantz CN (1988) A murine model of experimental metastasis to bone and bone marrow. Cancer Res 48:6876–6881

Bioluminescence Imaging ▶Optical Imaging

BI-RADS, Lexicon

Biomarker Biomarkers are tissue or plasma indicators of a specific biological process. ▶PET in Drug Discovery Imaging

Biphasic Barium Study A modification of the double-contrast technique whereby an additional quantity of dilute barium is given toward the end of the examination, and further films are obtained of the compressible parts of the stomach and duodenum. It shows some lesions better than by double-contrast alone. ▶Ulcer Peptic

Biphasic Magnetic Resonance Contrast Agents This term refers to substances that exhibit contrary signal intensities on T1-weighted and T2-weighted images. Water is a typical biphasic contrast agent, showing a low signal on T1-weighted and a high signal on T2-weighted magnetic resonance images. ▶Contrast Media, MRI, Oral Agents

131

tool to reduce limitations in the terminology of mammographic reports. The ▶BI-RADS lexicon includes illustrations of each feature described, a section on auditing a mammography practice, and sample reports. It is intended to standardize the terminology in mammographic reports, assessments of findings, and recommendations of actions to be taken. The relationship between an assessment and management recommendations has implications for clinical care, teaching, and evaluation of the screening interpretations of radiologists (1–3). Different features of masses and calcifications indicate the categorization. Standardized lexicons for breast ultrasound (US-BI-RADS) and breast magnetic resonance imaging (MRM-BI-RADS) are also available (4–6).

BI-RADS Lexicon for Mammography according to the ACR (1) The BI-RADS lexicon describes four classes of breast parenchymal density (Table 1, Fig. 1). The following features are distinguished in the BIRADS lexicon:

Masses A mass is defined as a lesion seen in two different projections. If a lesion is seen in only one projection, it should be called a density. Masses are further classified by shape, margin, and density (Fig. 2):

Shape

BI-RADS, Lexicon

The shape of a mass can be round, oval, lobular, or irregular (Fig. 2a).

Margin S ILVIA O BENAUER Department of Radiology, Georg-August-University of Goettingen, Goettingen, Germany [email protected]

The margin of a mass can be described as circumscribed (well defined or sharply defined), microlobulated (undulated with short cycles), obscured (hidden by superimposed adjacent tissue), indistinct (ill-defined), or spiculated (Fig. 2b).

Synonyms Breast Imaging Reporting and Data System (BI-RADS); classification of the American College of Radiology (ACR)

BI-RADS, Lexicon. Table 1 Classification of breast tissue density according to the BI-RADS lexicon (ACR, American College of Radiology) ACR

Definition The Breast Imaging Reporting and Data System (BIRADS) of the American College of Radiology (ACR) is a

1 2 3 4

Description Almost fat Fibroglandular Heterogeneously dense Dense

Diagnostic Accuracy Very high High Limitation Limitation

B

132

BI-RADS, Lexicon

BI-RADS, Lexicon. Figure 1 Different types of breast tissue density according to the American College of Radiology (ACR). ▶ACR type 1 = fatty, type 2 = fibroglandular, type 3 = heterogeneously dense, type 4 = extremely dense breast tissue. From Obenauer S, Hermann KP, Grabbe E (2005). Applications and literature review of the BI-RADS classification. Eur Radiol 15:1027–1036.

The density of a mass can be high, equal (isodense), low, or fat containing (Fig. 2c).

. Pleomorphic or heterogeneous calcifications (granular) . Fine linear or fine linear branching (casting) calcifications

Calcifications

The distribution of microcalcifications is shown in Fig. 4.

Density

Different types of calcifications are distinguished, and the distribution of the microcalcifications is described (Fig. 3).

Types of calcifications (Fig. 3) Typically benign calcifications (Fig. 3a) . . . . . . . . . . .

Skin calcifications (dermal) Vascular calcifications Coarse or popcorn-like calcifications Large rodlike calcifications Round calcifications Lucent-centered calcifications “Eggshell” or “rim” calcifications Milk or calcium calcifications Suture calcifications Dystrophic calcifications Punctuate calcifications

Calcifications of intermediate concern (Fig. 3b) . Amorphous or indistinct calcifications Calcifications with a higher probability of malignancy (Fig. 3c)

Architectural Distortion The normal architecture is distorted with no definite mass visible. Architectural distortion can also be an associated finding. Special cases include tubular density or a solitary dilated duct, intramammary lymph nodes, asymmetric breast tissue, and focal asymmetric density.

Associated Findings The associated findings are used with masses or calcifications or alone when no other abnormality is present: . . . . . .

Skin retraction Nipple retraction Skin or trabecular thickening Skin lesion Axillary adenopathy Architectural distortion

BI-RADS, Lexicon

133

B

BI-RADS, Lexicon. Figure 2 Shape (A), margin (B), and density (C) of masses according to the BI-RADS lexicon. Shape: round/oval (Aa), lobular (Ab), irregular (Ac); margin: well-defined (Ba), ill-defined (Bb), obscured (Bc), spiculated (Bd); density: fat-containing (Ca), isodense (Cb), high density (Cc). From Obenauer S, Hermann KP, Grabbe E (2005). Applications and literature review of the BI-RADS classification. Eur Radiol 15:1027–1036.

Location of Lesion

US-BI-RADS (4)

The location of a lesion should be expressed by its

Features in ultrasound according to the BI-RADS classification are given in Table 3.

. Side (left, right, or both) . Location (according to the face of the clock and subareolar, central, or axillary) . Depth (anterior, middle, or posterior)

BI-RADS Categories After a clear description of the findings according to the above-mentioned parameters, a report with the categorization of a mammogram into the BI-RADS classification is necessary for implementing a suggestion for the next course of action (Table 2).

MRM-BI-RADS (5, 6) Features in MRM according to the BI-RADS classification are summarized in Table 4 (Figs. 5–8). The time–signal intensity curve shows an initial rise, which could be slow (100%). The postinitial curve course could be persistent (>10% further increase), a plateau (between 10% and +10%), or a washout (>10% decrease; Table 4, Figs. 9, 10).

134

Birth Asphyxia

BI-RADS, Lexicon. Figure 3 Types of microcalcifications according to the BI-RADS lexicon. Typically benign findings (A): vascular (Aa), coarse or popcorn-like (Ab), large rodlike (Ac), round (Ad), “eggshell” (Ae), lucent-centered (Af), suture (Ag), and milk or calcium (Ah) calcifications; amorphous calcifications (B); pleomorphic (C) and fine linear branching (D) calcifications.

References 1.

2. 3.

4.

5.

6.

American College of Radiology (ACR) (1998) Illustrated Breast Imaging Reporting and Data System (BI-RADS). 3rd edn. American College of Radiology, Reston, VA Liberman L, Menell JH (2002) Breast imaging reporting and data system (BI-RADS). Radiol Clin N Am 40:409–430 Obenauer S, Hermann KP, Grabbe E (2005) Applications and literature review of the BI-RADS classification. Eur Radiol 15:1027–1036 Mendelson EB, Berg WA, Merritt CRB (2001) Toward a standardized breast ultrasound lexicon, BI-RADS: Ultrasound. Semin Roentgenol 3:217–225 Orel SG, Schnall MD (2001) MR imaging of the breast for the detection, diagnosis, and staging of breast cancer. Radiology 220: 13–30 Ikeda DM, Hylton NM, Kinkel K et al (2001) Development, standardization, and testing of a lexicon for reporting contrastenhanced breast magnetic resonance imaging studies. J Magn Reson Imaging 13:889–895

Birth Asphyxia ▶Hypoxic, Ischaemic Brain Injury

Birth Canal The birth canal consists not only of the bony structure of the pelvis but also of the soft tissue, which can be accessed by magnetic resonance imaging but is not included in pelvimetry. ▶Magnetic Resonance Pelvimetry

Birth Canal

135

B

BI-RADS, Lexicon. Figure 4 Distribution of microcalcifications according to the BI-RADS lexicon: diffuse (a), regional (b), linear (c), segmental (d), cluster of microcalcifications (e). From Obenauer S, Hermann KP, Grabbe E (2005). Applications and literature review of the BI-RADS classification. Eur Radiol 15:1027–1036.

BI-RADS, Lexicon. Table 2 American College of Radiology BI-RADS categories for mammography with probability of malignancy and recommendations Category 0 1 2 3 4 5 6

Finding Needs additional imaging evaluation Negative Benign Probably benign Suspicious abnormality Highly suggestive of malignancy Histologically proven malignancy

Probability of Malignancy –

Recommendation

0% 0% 1mm = 1 1 cyst = 1 Present = 1 Slight = 1

Pronounced = 2

Slight = 1

Pronounced = 2

Totally involved = 2 Joint space ≤ 1mm = 2 > 1 cyst = 2

B

146

Blood Diseases

Bleeding Disorders, Osteoarticular. Figure 2 Magnetic resonance imaging findings in hemophilic arthropathy. T1-weighted spin echo (SE) (first row) and T2*-weighted gradiant echo (GRE) (second row) images of three different hemophilic children with arthropathy of the ankle show (a, b) thickening and low signal intensity of the synovium (arrowheads) at the anterior and posterior joint recesses, (c, d) marked synovial proliferation (arrowheads) and hemosiderosis with capsular expansion and consecutive pressure erosion (arrow) at the talar neck, and (e, f) synovial proliferation associated with diffuse damage of articular cartilage and subchondral cyst formation (arrowhead).

surrounding capsule is usually depicted as a peripheral rim of low signal intensity on all pulse sequences, owing to the presence of fibrosis and hemosiderin deposits.

3.

Nuclear Medicine

5.

In the past, skeletal scintigraphy was used to evaluate hemophilic arthropathy as well as hemophilic pseudotumor, but it has been replaced by radiography and MR imaging as more specific methods.

Diagnosis Hemophilia is diagnosed by laboratory testing of factors VIII and IX, by which the type and severity of the disease can be determined. Genetic testing can uncover carriers and individuals with mild hemophilia.

4.

Resnick D (1995) Bleeding disorders. In: Resnick D (ed) Diagnosis of Bone and Joint Disorders. 3rd edn. Saunders, Philadelphia, pp 2295–2322 Stein H, Duthie RB (1981) The pathogenesis of haemophilic arthropathy. J Bone Joint Surg Br 63:601–609 Yulish BS, Lieberman JM, Strandjord SE et al (1987) Hemophilic arthropathy: assessment with MR imaging. Radiology 164:159–762

Blood Diseases ▶Hemoglobinopathies, Skeletal Manifestations

References 1.

2.

Park JS, Ryu KN (2004) Hemophilic pseudotumor involving the musculoskeletal system: spectrum of radiologic findings. AJR Am J Roentgenol 183:55–61 Pettersson H, Ahlberg A, Nilsson IM (1980) A radiologic classification of hemophilic arthropathy. Clin Orthop 149:153–159

Blood Flow ▶Perfusion, Neoplasms

Blunt Abdominal injury

147

Blood Pool Contrast Agent A contrast agent which, after intravascular injection, is freely distributed in the whole blood volume without leaving the vascular compartment. A blood pool contrast agent is transported with the blood flow, allowing its use as tracer for the inflow, distribution and outflow of blood (wash-in/wash-out kinetics) within a certain organ. ▶Contrast Media, Ultrasound, Commercial Products

Blood Urea Nitrogen

Bleeding Disorders, Osteoarticular. Figure 3 Intramuscular hemophilic pseudotumor of the thigh in a 41-year-old patient. T1-weighted SE image shows a huge intramuscular mass with inhomogeneous predominantly hyperintense signal surrounded by a capsule of low signal intensity. Note also arthropathic alterations of the right hip joint with synovial proliferations. Deformity of the right proximal femur was caused by previous femoral neck fracture.

Blood urea nitrogen measures the amount of urea nitrogen, a waste product of protein metabolism, in the blood. Urea is formed by the liver and carried by the blood to the kidneys for excretion. A test measuring how much urea nitrogen remains in the blood can be used as a test of renal function. However, there are many factors besides renal disease that can cause BUN alterations, including protein breakdown, hydration status, and liver failure. Reference values for BUN in adults (values may differ slightly from laboratory to laboratory) are 7–20 mg/100 mL. Men may have values slightly higher than women. ▶Tubular Necrosis, Kidney, Acute

Bloodgood’s disease Blood Oxygenation Level Dependent (BOLD) Contrast The contrast which is studied in functional magnetic resonance imaging studies and which is dependent of the amount of deoxyhaemoglobin in the blood. ▶Brain, Functional Imaging

Blood Pool Agents Contrast agents (for ultrasound and magnetic resonance imaging) that are confined to the intravascular fluid compartment and do not leak into the interstitium. ▶Contrast Media, Ultrasound, Applications in Kidney Tumor

▶Fibrocystic Disease, Breast

Blue Bloaters Patients with emphysema who suffer from severe hypoxemia and hypercapnia and have peripheral edema due to right heart failure, but only mild dyspnea. ▶Emphysema and Bulla

Blunt Abdominal injury ▶Trauma, Hepatobiliary

B

148

Blunt Abdominal Trauma

Blunt Abdominal Trauma ▶Trauma, Hepatobiliary

Blunt Hepatic Injury ▶Trauma, Hepatobiliary

carpal or tarsal bones and sesamoids (including patella), and (d) the state of fusion of the physes between epiphyses and metaphyses, and how narrow is the remaining cartilage space between them if not yet fused. Generally, bone age is expressed, or implied, in statistical terms, reflecting numerical deviation from an estimated mean. For example, in our use of the ▶Greulich–Pyle method, “advanced” or “delayed” bone age represents values more than two standard deviations from the expected mean. Technically, each represents a rejection of the null hypothesis that the bone age is within normal limits with 96% confidence. When chronologic bone age is unknown, the bone age serves as a best guess at the subject’s true age.

Pathology/Histopathology

Bochdalek Hernia maldevelopment of the diaphragm resulting in a defect that is characteristically posterior and left sided. ▶Hernia, Diaphragm, Congenital

Bone Age A LAN E. O ESTREICH Cincinnati Children’s Hospital Medical Center, Cincinnati, Ohio, USA [email protected]

Synonym (Estimation of) skeletal maturation (by imaging)

Definitions Bone age estimation methods all relate the appearance of a portion of the subject’s ossified skeleton to a set of standards, usually by visual comparison. Bone age is a measure of enchondral bone development (with the exception of the use of the metaphyseal width, which is the result of membranous ossification from the one-celllayer thick bone bark around portions of the physis and metaphysis). The elements observed in the bone age process consist of (a) are the growth centers in question ossified yet? (b) relative size of growth centers once they are ossified, in relation to the adjacent metaphysis, (c) shapes of growth centers, including nonepiphyseal growth centers, such as

Many causes of disturbed bone age have histopathologic or at least gross pathology manifestations. Abnormality of the pituitary, thyroid, adrenals, and ovaries are major examples, although biochemical and radiologic investigation may obviate biopsy in some circumstances. Conditions of high local vascularity, such as juvenile idiopathic arthritis and hemophilic hemorrhage, have biochemical and radiologic manifestations, as well as those amenable to pathology/histopathology. In achondroplasia, in the first decade of childhood, the slow enchondral growth results in more of the growth cartilage surrounding growth centers to not be ossified as soon as would be expected in normals. The result is a delayed bone age. In long-term arthritis, this same cartilage is diminished and ossification increases compared to normals.

Clinical Presentation The child with delayed bone age is quite often of short stature. The child with advanced bone age often shows signs of sexual precocity. Growth hormone deficient children have normal features despite being proportionately small; hypothyroidism has many other clinical features, including coarse skin and slow reflexes; girls with low estrogen may be tall, as they continue growing for a longer time than their peers, and at the same time have a delay in pubertal features. A malnourished or chronically ill child may also have delayed bone age.

Imaging For the Greulich–Pyle (1) and several other methods, a frontal PA image of the left hand and wrist is obtained and analyzed. In certain circumstances, the right hand may be

Bone Age

149

more appropriate (such as unilateral hypoplastic or overgrown left hand). For the ▶Sontag method (2), several images are obtained of the left side of the body to show countable growth centers of upper and lower extremity. We use the Sontag method in the first year of life, and in the second year we use it if hand and wrist show no more than capitate and hamate ossified. Standards are available for the knee (AP and lateral images). The Risser method of iliac crest ossification requires a full view of at least one iliac crest. The Tanner method (3) requires numerical measurement of many centers, a rather tedious task. In addition to counting, measuring, or comparing growth centers to standards, one should also evaluate bone age images for metabolic bone disease (rickets), dysplasia (such as achondroplasia), or arthritis (juvenile idiopathic, for example Fig. 1)), that might affect maturation, as well as features of any other concomitant condition. The broad terminal tufts of Turner syndrome; the large distal epiphyses, cone epiphyses of small tubular bones, and prominent metacarpal pseudoepiphyses of cleidocranial dysplasia (Fig. 2); and the broad distal thumb phalanges of Rubinstein–Taybi syndrome are examples of findings that may initially suggest diagnoses from bone age studies. Methods have been proposed using ultrasound images of the hand and wrist as nonradiation methods for obtaining a bone age—I do not feel they have adequate precision at this time. Measurement of lateral thoracic subcutaneous fat in newborns provides a reasonable estimate of gestational age in the absence of hypoglycemia (4); it exceeds the normal standards in infants of diabetic mothers and in nesidioblastosis.

B

Nuclear Medicine The high activity of physes ceases when the epiphysis fuses to the metaphysis. Quantification of the degree of fusion by activity on scan is possible. Similarly, the onset of ossification in a cartilaginous growth center is presaged by “preossification center” activity. Endocrine causes of abnormal bone age can be investigated by nuclear imaging. They include tumors and nodules of thyroid, pituitary, adrenals, ovaries, and testes. Nuclear medicine is also helpful in locating functioning ectopic thyroid, especially at the base of the tongue, when the customary gland site does not show tissue or activity.

Diagnosis Visual comparison of a subject’s bone age image to standards such as Greulich–Pyle (1) requires knowledge of the gender, so that the gender-specific standard images

Bone Age. Figure 1 Juvenile idiopathic arthritis with selective acceleration of maturation. The carpal bones are both overdeveloped and crowded and several metacarpal heads are large for age. The proximal lateral corner of the trapezium is squared off, as shown in the enlargement below. (From Oestreich AE, Crawford AH (1985) Atlas of Pediatric Orthopedic Radiology. Thieme Verlag, Stuttgart, p 159(6)).

be used. However, it is preferable not to know the chronologic age before doing the comparisons, so that, being blinded to that age, one has a better chance not to

150

Bone Infarction

hypovascular situations. Children with Perthes disease seem to have a relatively delayed hand and wrist bone age. Advanced bone age is seen in many premature puberty states, in Marshall syndrome, hypothalamic/pituitary tumors; perhaps, overdose of growth hormone treatment, and locally from hypervascular states, including juvenile idiopathic arthritis (Fig. 1), hemophilia hemorrhage (clearing of which takes vascularity), and, quite dramatically, in neonatal onset (or infant onset) multisystem inflammatory disease (NOMID). In growth hormone deficiency/hypopituitary/Laron short stature, bones are normally shaped despite the delayed bone age. In hypothyroidism, bones (like the children themselves) are malformed, with epiphyseal and carpal and tarsal irregularity earlier in bone age, and metaphyseal irregularity later in bone age. The hypogonadal Turner syndrome, like most decreased sex hormone conditions, shows bone age delay principally in the second decade. Turner syndrome additionally shows characteristic drumstick distal finger phalanges and a recognizable abnormality in mandible shape (5). In the estrogen deficiency of very high performance athletic teenage girls, bones tend to be long as they continue to grow. Achondroplasia has dysplastic short bones in a rhizomelic pattern. Bone Age. Figure 2 Cleidocranial dysplasia. Seven year old child. Characteristic prominent pseudoepiphysis at the proximal 2nd metacarpal, cone epiphyses of ring and little finger middle phalanges, and oversized epiphyses for the distal phalanges of thumb and other fingers. (From Oestreich AE, Crawford AH (1985) Atlas of Pediatric Orthopedic Radiology. Thieme Verlag, Stuttgart, p 129(6)).

Bibliography 1.

2.

3.

let one’s expectations skew the choice of bone age. It is preferable to interpolate bone age between two standards rather than merely selecting the closest. Be sure to understand that bone ages chosen are compared to the mean expected bone age, not the chronologic age itself. The carpal bones seem much more variable than other hand and distal forearm centers. Thus, they should not be used, after about 18 months of age, as strong measures of bone age. Discordance between “carpal” and “finger” bone age is quite common. My guess is that usual and unusual childhood illnesses affect the carpals disproportionately. Left-to-right discordant bone age suggests disuse of the less mature side, such as associated with brachial plexus palsy, or hypervascularity of the more mature side, such as from diffuse unilateral arthritis. Healing of fractures increases maturation of adjacent growth centers as well. Delayed bone age may reflect chronic illness or malnutrition; occasionally neglect or abuse; endocrine diseases (thyroid, pituitary, adrenal, or gonad); and locally in bones affected with melorheostosis or other

4. 5. 6.

Greulich WW, Pyle SI (1959) Radiographic Atlas of Skeletal Development of the Hand and Wrist. 2nd edn.Stanford University Press, Stanford Keats TE, Sistrom C (2001) Skeletal Maturation: Method of Sontag, Snell, and Anderson. Atlas of Radiographic Measurement. 7th edn. Mosby, St. Louis, p 317 Tanner JM, Whitehouse RH, Marshal WA et al (1983) Assessment of Skeletal Maturity and Prediction of Adult Height (TW2 method). 2nd edn.Academic Press, London Kuhns LR, Berger PE et al (1974) Fat thickness in the newborn infant of a diabetic mother. Radiology 111:665–671 Rzymski K, Kosowicz (1975) The skull in gonadal dysgenesis. A roentgenometric study. Clin Radiol 26:379–384 Oestreich AE, Crawford AH (1985) Atlas of Pediatric Orthopedic Radiology, Thieme Verlag, Stuttgart

Bone Infarction Ischemic osteonecrosis in the nonweight-bearing regions of bones (metaphyseal and diaphyseal location). This is often an incidental finding and usually does not lead to articular destruction. Most commonly affected are the distal femur and the proximal tibia. ▶Osteonecrosis, Adults ▶Osteonecrosis, Childhood

Bone Scintigraphy

Bone Infection Bone infection is in daily practice most frequently caused by bacteria. Invasion and colonization of bone can occur by three principal routes: 1. Spread from a contiguous source as seen in chronic foot ulcers in diabetic patients. 2. Hematogenous spread as typically seen in children and in spondylodiscitis of adults. 3. Direct implantation may occur after accidents with open fractures, direct punctures or after bone surgery. Bone infection can be separated into osteomyelitis (infection of cortex and bone marrow) and osteitis (infection of cortical bone only). Depending on the duration of bone infection acute or chronic forms can be distinguished which may reveal typical appearances on plain film: Acute osteomyelitis leads to ill defined permeative bone destruction with periosteal reaction and surrounding soft tissue swelling. Chronic osteomyelitis on the other hand may reveal intraosseous abscesses, bony fistula, relatively demarcated bony destruction combined with reactive bone proliferation. ▶Osteomyelitis

151

Bone Scintigraphy A NDREAS B OCKISCH , M ONIA H AMAMI , S ANDRA J. R OSENBAUM -K RUMME Klinik fur Nuklearmedizin, Universitaetsklinikum Essen, Germany [email protected]

Synonyms Bone scan; Skeletal scintigraphy

Definition Bone scintigraphy is a diagnostic tool in the evaluation of the skeletal system. Depending on bone metabolism, especially turnover and blood flow, bone will incorporate certain radioactive tracers.


Bone Marrow Infections ▶Osteomyelitis, Neonates, Childhood

Bone Metastases Bone metastases may either be osteoblastic, osteolytic, or have mixed characteristics. Because bone scintigraphy is sensitive to bone buildup, it visualizes osteoblastic metastases very well and is also sensitive to mixed lesions. ▶Bone Scintigraphy ▶Metastases, Skeletal

Bone Scan ▶Bone Scintigraphy

Physiologically there is a balance of osteoblastic and osteolytic activity. Many metabolic disturbances lead to enhancement of the osteoblastic branch. Bone scintigraphy images osteoblastic activity and allows semiquantitative measurement of it. Therefore, diseases with osteoblastic reactions—especially focal—are the domain of bone scintigraphy. Systemic increased bone turnover may also be detected. For basic measuring reasons, the method is insensitive concerning lytic lesions. One of the most dominant applications is the search for ▶bone metastases (osteoblastic or mixed), which is performed as a static single-phase investigation. Bone scintigraphy is very sensitive to bone buildup—not only due to metastases but also due to active benign diseases such as traumata, active ▶arthritis, and benign bone tumors. Rarely, extraosseous accumulation occurs. The experienced nuclear medicine physician is aware of these possible disturbances that need to be differentiated from osseous accumulation, such as by the help of tomography (single photon emission computed tomography, or ▶SPECT). Because bone scintigraphy reveals changes in bone metabolism more than changes in bone structure, bone scintigraphy complements rather than replaces plain radiography (X-rays). The changes noted on bone scintigraphy usually precede the changes noted on radiographs because the bone metabolism usually changes before the bone structure changes.

B

152

Bone Scintigraphy

Clinical Presentation Lower back pain: The typical indication is lower back pain with negative radiographs. Due to its high sensitivity for osteoblastic reactions, bone scintigraphy can detect intravertebral arthritis or infractions very early. Using sectional imaging (SPECT) is advantageous. Osteomyelitis: In contrast to morphology-based imaging, bone scintigraphy is not or is only slightly influenced by the residues of lapsed changes, but it is very sensitive to present developments. Thus, bone scintigraphy is suitable for differentiating between an acute episode of chronic osteomyelitis and a chronic state, for example. A negative bone scan excludes active osteomyelitis. Newborns or nurslings are the only exceptions in which osteomyelitis may present with indifferent or even reduced activity in the bone scan [2]. Traumata: A bone scan reliably detects the increased osteoblastic activity after fracture or severe injury below the level of the fracture. Depending on the site, there is a calm period of some hours up to a maximum of 7 days in which the bone scan may be negative (or may not be). A high-activity period follows, and finally the increase in bone turnover abates over many weeks and months. In the beginning, the blood pool activity is increased. Considering all these facts, the age of a fracture may be estimated. Applications include identification of (more) acute fractures in the presence of multiple fractures in osteoporosis and determination of the age of a fracture for forensic reasons (such as insurance questions) or suspicion of child abuse. Bone scintigraphy can also give valuable information about the course of posttraumatic healing. Abnormally prolonged repair may be detected after fracture or surgery (implants), and pseudoarthrosis may be characterized as hypo- or hypermetabolic.

Imaging

Bone Scintigraphy. Figure 1 Essentially normal bone scintigram. The investigation is performed both in ventral view (right) and dorsal view (left). Due to the significantly higher thickness of absorbing body tissue between the photon-emitting source (here, the bone) and the gamma camera, detection of structures from the distant side of the body overprojection is negligible. There is no focal abnormality. The kidneys as well as the urinary bladder are contrasted because of the excreted radiotracer. For the same reason, contamination with radioactive urine may be observed inguinally, as in this case (small spot left anterior). Note that the rips are expected to be well separated and it should also be feasible to identify the bodies of the lumbar spine individually, depending on their angulation to the gamma camera collimator.

Refer to Figs. 1, 2, and 3 for imaging examples.

Nuclear Medicine Bone scintigraphy is one of the most frequently applied nuclear medicine procedures. It follows the metabolic bone turnover and allows semiquantitation of osteoblastic activity. In the 1950s, 18F was used as tracer, which was abandoned in the late 1970s when 99mTc became available (3) and 99mTc-labeled phosphonates for bone scintigraphy were developed, evaluated, and finally established. Because fluoride intensively takes part in bone turnover, 18F is an identical tracer and from biological aspects is ideal.

However, its emission energy is not suitable for gamma camera application. Today, 18F has had a revival in its use with positron emission tomography (PET) scanning. 18 F–PET investigations may be referred to as ace bone scintigraphy and may be considered especially in those cases in which improved spatial resolution is needed. The principle of bone scintigraphy—the increased accumulation of suitable radionuclides—is also used for therapeutic purposes. Bone pain due to osteoblastic metastases can be significantly reduced, and bone metastases may be shrunk and small ones even cured using suitable beta-emitting nuclides.

Bone Scintigraphy

153

B

Bone Scintigraphy. Figure 2 Osteoblastic bone metastases in a patient with breast cancer. Whereas the extended involvement of the right side of the pelvis and of the right shoulder was known from X-ray, the additional lesions in the left shoulder and the left side of the pelvis, multiple rips, and some vertebral bodies were diagnosed by the bone scintigraphy demonstrated here.

Principles of Bone Scintigraphy The term “bone scintigraphy” is associated with gamma camera imaging and the application of 99mTc-labeled ▶polyphosphonates. These phosphonates also take part in the bone buildup. In the course of that process, the technetium is separated and remains in an insolvable form at the point of ossification. 99mTc-diphosphonopropandicarbone-acid (DPD), 99mTc-hydroxy-ethylene diphosphonate (HDP), and 99mTc-methylene diphosphonate (MDP) are frequently used. MDP, DPD, and HM-DPD have basically similar characteristics but somewhat different kinetics. During the time period of investigation—typically up to a maximum of 4 h, and in special cases up to about 24 h— no relevant redistribution of the deposited activity occurs. In adults of standard weight, 500–750 MBq 99mTclabeled polyphosphonates are applied. Depending on the clinical question, the investigation may be performed using one-, two-, three-, or, rarely, four-phase technique.

Bone Scintigraphy. Figure 3 Patient with psoriasis arthropathy. (a) Dynamic investigation over 60 s. Each sequential image was recorded over 3 s. (b) Early scintigram (blood pool) and (c) late scintigram of the hands. Diffuse hyperperfusion is seen in the right hand compared with the left one, which is supposedly also hyperperfused—however, to a lesser extent. The blood pool image shows some joints as hyperactive, which also represents increased osteoblastic activity. In addition, other joints are positive in the late image only. In arthritis, the combination of positive early- and late-phase scintigram is suggestive for acute positive late phase only for chronic disease.

In the first phase, radionuclide angiography, dynamic imaging is performed of the area of interest with an image frequency of, for example, 2/s for creation of time–activity curves. For visual analysis, these images are summed up for an integral time of 2–3 s each. The second phase is an early static image taken between about 4 and 10 min as limits. It reflects the blood volume as the tracer is still predominantly dissolved in the blood; however, specific bone turnover also occurs, which might influence the result in spots of very high turnover such as the epiphysis in children. The blood pool investigation is often restricted to one body region. If a whole-body blood pool investigation is considered, the investigation has to be performed with high speed in order to meet the above-mentioned time limitations.

154

99mTc Bone Scintigraphy

The third phase is literally the bone scintigraphy. At that time, about half of the activity has been renally excreted. The remaining activity has mostly accumulated in the bone proportional to the osteoblastic activity. The minor part of the activity remains in the blood pool or is nonspecifically distributed extracellularly. Especially in patients with low absolute kidney function, the targetto-background ratio may be unfavorable, resulting in images of impaired quality. In this case, further late or very late (24 h) imaging may overcome the problem. The fourth phase is seldom needed and has its value in differentiating between specific accumulations in the diseased bone and increased soft tissue activity, such as in ulcera in diabetic foot syndrome. In the mineral phase (phase 3 or 4), the investigation may be performed in cross-sectional imaging technique (SPECT) in addition to planar [3]. The three-dimensional information from SPECT is especially useful when investigating more complex bony structures such as the head or the pelvis, mostly leading to increased sensitivity and specificity. Further improvement may be achieved by correlation with additional morphologic images or by performing a SPECT/ computed tomography investigation [4].

these investigations, at least a two-phase investigation is mandatory. Due to the comparatively low radiation exposure (typically 4–5 mSv for adults), bone scintigraphy is readily used, as well as in nonmalignant diseases and for children. In the latter case, the activity to apply is chosen according to European Association of Nuclear Medicine recommendations. The nuclear medicine modality allows scanning of the whole skeleton within less than half an hour and with low radiation exposure, which is independent of the extent of the investigation. Bone scintigraphy is performed either to detect disease, to characterize the disease or its clinical course, or to define the extent of disease (such as in a search for metastases of osteomyelitis or fractures in a multitrauma patient). Often, more than one of the indications applies.

References 1.

2. 3.

Diagnosis

4.

Bone scintigraphy is a quick, reliable, and cost-efficient modality to check the bone status in malignant diseases that frequently present with osteoblastic or mixed bone metastases, including breast carcinoma, lung cancer, prostatic cancer, and gastrointestinal cancer [5]. To evaluate individual diagnostic and therapeutic regimens, screening for bone metastases may be an important component; however, for lack of space, this will not be discussed here. In primary bone tumors, bone scintigraphy is performed using a three-phase technique. Although today, bone scintigraphy for primary diagnosis of bone tumors has taken a back seat, it might help to make a differentiated diagnosis in uncertain cases. Malignant tumors often show up intensively in all of the three phases. In prediagnosed malignant and benign bone tumors that are known to have a multilocal appearance, bone scintigraphy is used to clarify the extent of disease as the whole-body state is quickly acquired. In a given entity of tumors, the intensity of bone turnover is expected to correlate positively with the degree of malignancy. This experience may be used for directed biopsies, such as in fibrous dysplasia or Paget’s disease. As a functional modality, bone scintigraphy is well suited for follow-up and for early detection of response to therapy [1]. In this context, the blood pool changes have been found to be of special value. A couple of benign diseases apart from benign primary bone tumors are indications for bone scintigraphy. In

5.

Knop J, Montz R (1983) Bone scintigraphy in patients with osteogenic sarcoma. Cooperative Osteogenic Sarcoma Study Group COSS 80. J Cancer Res Clin Oncol 106: Suppl: 49–50 Schauwecker DS (1992) The scintigraphic diagnosis of osteomyelitis. AJR 158:9–18 Subramanian GJG, McAffee RE, O’Hara M et al (1971) Mehter: 99m Tc- polyphosphate PP46: A new radiopharmaceutical for skeletal imaging. J Nucl Med 12:399–400 Romer W, Nomayr A, Uder M et al (2006) SPECT-guided CT for evaluating foci of increased bone metabolism classified as indeterminate on SPECT in cancer patients. J Nucl Med 47(7):1102–1106. Wang K, Allen L, Fung E et al (2005) Bone scintigraphy in common tumours with osteolytic components. Clin Nucl Med 30(10):655–671

99mTc Bone Scintigraphy Radioisotope bone scintigraphy is the standard wholebody imaging technique for screening bone metastases. It depicts the tracer accumulation at any skeletal site with an elevated rate of bone turnover as seen in neoplasia, trauma, arthropathy, or inflammation. ▶Metastases, Skeletal

Bony Destructions Bony destructions in a marginal location of the joint surface is, together with cartilage destruction, one of the three radiologic key symptoms of arthritis (the other two are synovial soft tissue swelling and collateral phenomenon). ▶Rheumatoid Arthritis

Brachial Ischemia, Chronic

BOOP Bronchiolitis obliterans organizing pneumonia. Proliferative bronchiolitis with granulation tissue in the bronchiolar lumen of the distal airways. ▶Bronchitis and Bronchiolitis in Childhood

155

BPD ▶Dysplasia, Bronchopulmonary

BR14 Borderline Tumor Borderline ovarian tumor is a subtype of epithelial ovarian cancer and is classified as a tumor of low malignant potential. Compared to invasive ovarian cancer, it occurs in younger women and has a better prognosis. ▶Masses, ovarian

Botryoid Rounded swellings resembling a bunch of grapes. ▶Rhabdomyosarcoma

Microbubble US contrast agent used with low mechanical index. The bubbles consist of perfluorobutane and are stabilized by a phospholipid shell. The agent has a liver-specific late phase. Clinical development stopped (2003). Manufacturer: Bracco SPA, Italy. ▶Contrast Media, Ultrasound, Hepatic

Brachial Ischemia, Chronic E LIAS N. B ROUNTZOS Department of Radiology, Athens University, Medical School, Athans, Greece [email protected]

Boutonniere Deformity Definition Boutonniere deformity is a typical deformity in late-stage rheumatoid arthritis with flexion of the proximal interphalangeal joint and hyperextension of the distal interphalangeal joint. ▶Rheumatoid Arthritis

Long-standing reduction in the blood perfusion of the upper extremity as a result of an occlusive arterial lesion. In contrast to the acute upper extremity ischemia, chronic ischemia has a more subtle presentation.

Vascular Anatomy

Bouveret’s Syndrome Duodenal obstruction due to a gallstone migrated through a cholecystoduodenal fistula. ▶Cholecystitis

Bowel Disease, Inflammatory This term usually indicates both UC and Crohn’s disease. ▶Colitis, Ulcerative ▶Crotin’s Disease

See also Fig. 1. The main artery supplying the upper extremity is the subclavian artery. The subclavian artery originates from the innominate artery on the right and from the aorta on the left. It is defined distally by the lateral border of the first rib, where it becomes the axillary artery. Branches of the subclavian artery are the vertebral artery, the internal mammary artery, the thyrocervical trunk, and the costocervical trunk. The axillary artery lies between the lateral border of the first rib and the lateral margin of the teres minor muscle, where it becomes the brachial artery. Its branches are the superior thoracic artery, the lateral thoracic artery, the thoracoacromial artery, the subscapular artery, and the circumflex humeral artery. The branches of the subclavian and the axillary

B

156


Brachial Ischemia, Chronic. Figure 1 Maximum intensity (MIP) reconstruction from a gadolinium enhanced 3-D magnetic resonance angiography (MRA) of a normal subject in the RAO projection depicts normal upper extremity arterial anatomy. Vert: vertebral artery; CCA: common carotid artery; IMA: internal mammary artery; Thyr: thyrocervical trunk; Cost: costocervical trunk; Lat thor: lateral thoracic artery; Thoracrom: thoracoacromial artery.

arteries serve as a rich network of collateral circulation in any case of subclavian or axillary artery occlusion. The brachial artery ends to its bifurcation in the proximity of the radial head. One of the most consistent branches of the brachial artery is the deep brachial artery, which supplies the muscles of the posterior aspect of the upper arm. The radial and ulnar arteries arise from the brachial artery, the interosseous artery arising from the ulnar artery. A rich collateral network is formed around the elbow from the anastomoses between the recurrent ulnar and radial interosseous arteries and the brachial collaterals. The main arterial supply to the hand is at its volar side. The dorsal side contains veins. The superficial palmar arch is primarily formed by the ulnar artery. The deep palmar arch is formed by the radial artery. The paired proper digital arteries supply the fingers (1). The extensive collateral supply at the shoulder and the elbow explains the absence of symptoms in many patients with chronic segmental arterial occlusion of the subclavian, the axillary, or the brachial artery.

Pathology Arterial occlusion may be a result of a fixed lesion, or a result of vasospasm; both mechanisms are not mutually exclusive and may be combined. In 60% of patients with the primarily vasospastic Raynaud’s syndrome an underlying disorder is diagnosed producing fixed occlusive lesions of the palmar and digital arteries (2). Different

etiologies apply for the large arteries (aortic arch to the wrist), and to the small arteries (distal to the wrist) (Table 1). Atheroclerotic disease is more often diagnosed when it is located at the origins of the great arch vessels, e.g., the innominate artery, and the subclavian artery, because at these locations it is more likely to cause symptoms. Innominate artery and subclavian artery occlusive disease occurs in relatively younger patients than other types of atherosclerotic disease. The left subclavian artery is more commonly affected compared to the right side. The prevertebral part of the subclavian artery is usually involved. Atherosclerotic disease in more distal locations is less likely to cause ischemic symptoms. Involvement of the small arteries of the fingers is found in smokers, diabetics, and patients with end stage renal disease (1). Trauma is more likely a cause of acute upper limb ischemia, but in some patients treated conservatively, symptoms of chronic ischemia develop later (3). Arterial complications of the thoracic outlet syndrome (TOS) are rare compared to the neurologic involvement (less than 5% of the cases). The subclavian artery is compressed in the scalene triangle, or the costo-clavicular space. The initial lesion is an arterial stenosis, but later a poststenotic dilatation is formed not unusually containing mural thrombus. This may result in peripheral microembolization, usually of the thumb or index finger. Takayasu’s arteritis affects proximal vessels such as the innominate and subclavian arteries. A more distal involvement of the axillary and proximal brachial artery is suggestive of giant cell arteritis. Connective tissue disorders include Systemic lupus erythematosus (SEL), Scleroderma,


Brachial Ischemia, Chronic. Table 1

157

Etiology of chronic upper extremity ischemia

Large arteries 1. Atherosclerosis 2. Trauma a. Blunt, penetrating b. Iatrogenic 3. Arterial thoracic outlet syndrome (compression by cervical rib, scalenus muscle, etc.) 4. Embolization associated with thoracic outlet syndrome 5. Arteritis a. Takayasu’s b. Giant cell c. Buerger’s disease 6. Fibromuscular disease Small arteries 1. Raynaud’s syndrome 2. Atherosclerosis 3. Connective tissue disease (Scleroderma, Rheumatoid arthritis, SEL, Mixed connective tissue disorder) 4. Myeloproliferative disease (Essential thrombocytosis, polycythemia vera, Chronic Myeloid Leukemia, Myelofibrosis) 5. Hypercoagulable states (Antithrombin III, protein C, protein S deficiencies, Antiphospholipid syndrome, heparin antibodies, etc) 6. Cold injury 7. Buerger’s disease 8. Occupational vascular problems a. Vibration-induced white finger b. Hypothenar hammer syndrome c. Athletic injuries 9. Chronic renal failure

Rheumatoid arthritis, Sjoegren syndrome, Henoch-Shoenlein purpura, Dermatomyositis, Wegener granulomatosis, and Polyarteritis nodosa (PAN). This heterogeneous group of maladies is characterized by tapering of the ulnar, radial, and digital arteries with segmental occlusions and frequently superimposed vasospasm. Arterial lesions are the result of endothelial damage, fibrinoid thickening, and intimal hyperplasia leading to obliterative endarteritis of the digital arteries and caused by antigen–antibody complex deposition (4). Fibromuscular dysplasia (FMD) of the upper extremity is more commonly located in the subclavian, the axillary, and the brachial arteries. Buerger’s disease of the upper limb is found in two-thirds of the patients suffering from this disease. Raynaud’s syndrome (RS) is the most common cause of upper limb ischemia. It is characterized by episodic attacks of vasospasm caused by closure of the small arteries of the most distal parts of the extremity in response to cold or emotional stimuli. With arterial closure the hand becomes white, then with relaxation of the spasm turns blue due to cyanosis, and finally with reactive hyperemia becomes red. In a large proportion of patients, an underlying disorder can be diagnosed causing fixed arterial lesions. Prolonged professional use of vibratory tools has been associated with symptoms similar to Raynaud’s syndrome. This condition is termed “Vibration-

induced white finger.” Patients with the “Hypothenar hammer syndrome” are professionals who use the palm of the hand for pushing, pounding, or twisting (mechanics, carpenters, etc.). Typical symptoms are blanching of the fingers, coolness, paresthesias, more or less similar to Raynaud’s syndrome. Similar conditions can develop in athletes using their hands in the same fashion (handball players).

Clinical Presentation Occlusive arterial disease of the upper extremity arteries is less often symptomatic compared to that of the legs, because less muscular workload falls in arms, and the arterial collateral network in the upper extremity is very extensive to compensate for localized arterial occlusions (1). When symptomatic, chronic ischemia usually presents as arm claudication. This is more usual with proximal atherosclerotic lesions. Other symptoms include rest pain and gangrene, but those occur much less frequently and almost invariably in the fingers. Similar clinical presentation occurs in patients with Takayasu’s disease or giant cell arteritis with proximal lesions. However, these patients similarly to those with connective tissue disorders are generally younger than the athero-

B

158


sclerotic patient and may also have other symptoms from the underlying disease. Patients with arterial involvement of TOS complain of numb hand or tingling associated with activity that compresses the arteries within the thoracic outlet space. In some instances, the initial presentation is with symptoms of microembolization of the digital arteries or the palmar arch. Symptoms include Raynaud’s syndrome, paresthesias, cold sensation, or pain. Symptoms of microembolization may last long before a major embolic complication causing acute ischemia takes place. At physical examination, a pulsatile mass may be palpated in the supraclavicular fossa, but it does not always represent the subclavian aneurysm itself, but rather the underlying bony abnormality pushing the artery upward (1). The presenting symptom in patients with FMD maybe distal embolization. RS is characterized by episodic ischemic attacks involving the hands and fingers after exposure to cold or emotional stimuli.

Imaging Duplex or Color Doppler Sonography The origins of the innominate and the subclavian arteries can be evaluated by color Doppler sonography (CDS) but often not without difficulty because the vessels are situated deep in the thorax. In these cases, transesophageal sonography is more suitable. More distally the vessels become superficial and can be imaged by high-frequency probes. A complete examination of all upper-extremity arteries is accurate but time-consuming.

CT Angiography This modality is excellent suited for the evaluation of the arch and the origins of the upper extremity arteries. Excellent imaging of the arterial anatomy of the upper limb can be achieved using CT angiography (CTA) with postprocessing, although the resolution does not allow for the evaluation of the digital arteries. A noncontrast scan is initially obtained followed by the contrast scan using thin collimation. Coverage should include the proximal neck to the hand.

Magnetic Resonance Angiography The origins of the subclavian and innominate arteries and their proximal parts to the shoulder can be imaged with coronal 3-D gadolinium enhanced acquisitions. High quality images of the hand arteries comparable to the conventional invasive angiography are now possible in less than 5 min with the use of dedicated surface coils that provide high spatial resolution (4).

Conventional Angiography It is still considered the gold standard for the imaging of the upper limb arterial anatomy; everything from the aortic arch to the fingers should be imaged, otherwise significant pathology may be missed. For the imaging of the aortic arch and the origins of the arch vessels digital subtraction angiography (DSA) is performed in the left anterior oblique (LAO) projection with a pigtail catheter positioned in the ascending aorta; imaging in the right anterior oblique (RAO) projection depicts the origins of the right subclavian and common carotid arteries. Imaging of the upper extremity arteries is performed with selective catheterization: the subclavian and axillary arteries are imaged with the tip of the catheter in the subclavian artery; the brachial artery is imaged with the tip of the catheter in the axillary artery; the forearm and hand arteries are imaged with the tip of the catheter in the brachial artery. Magnification views are used for detailed depiction. To avoid vasospasm causing diagnostic pitfalls the hand should be kept warm or the injection of vasodilators is used.

Diagnosis In patients with atherosclerotic disease of the proximal upper extremity arteries angiography typically depicts a localized lesion, concentric or eccentric with or without calcification. Lesions at the origin of the right subclavian artery may involve also the origin of the right carotid artery (1). Not unusually, multiple arch arteries are involved with stenoses and occlusions. Angiography in Atherosclerotic disease of the distal upper extremity arteries depicts segmental occlusions of the palmar or digital arteries. In patients with TOS with arterial involvement chest radiograph, CT and MRI may reveal the underlying structural abnormality, such as a cervical rib, or acquired anomalies of the clavicle or the first rib. Imaging with DSA or other modalities is recommended to evaluate the subclavian artery for stenosis, aneurysm and the existence of mural thrombus, and to look for distal embolization. DSA in combination with evocative maneuvers such as 90˚ abduction and external rotation have been used to reveal the arterial compression; however, these findings are encountered in 50% of normal subjects. Ultrasonography and CTA/MRA better evaluate the vessel lumen, wall, and the surrounding tissues. In patients with arteritides, CTA and magnetic resonance angiography (MRA) characteristically depict wall thickening that enhances after gadolinium injection. Takayasu’s arteritis affects proximal vessels such as the innominate and subclavian arteries. A more distal involvement of the axillary and proximal brachial artery


is suggestive of giant cell arteritis. In addition to imaging, the diagnosis is established on the basis of clinical history and laboratory tests. In patients with connective tissue disorders angiographic findings include tapered occlusions without intraluminal filling defects (as opposed to embolization). Angiography is important for the diagnosis of PAN with 89% sensitivity and 90% specificity; without this tool the diagnosis is possible in only 20–30% of the cases whenever tissue pathology yields diagnostic results (4). Angiographic findings of PAN include digital artery occlusions as a common finding with other vasculitides, but the depiction of digital aneurysms is suspicious of PAN. In patients suspected to suffer from FMD, DSA is essential for the diagnosis depicting the characteristic beaded appearance in the case of medial fibroplasia. Because bilateral disease is often present both arms should be imaged. In Buerger’s disease angiographic findings of the hand and digital arteries include occlusions and characteristic corkscrew collaterals from the perineural arteries and vasa vasorum (1). In Raynaud’s syndrome (RS) hand arteriography before and after ice water immersion has been used in the past for the establishment of the diagnosis. After exposure to cold prominent vasospasm of the palmar and digital arteries is a feature of the RS. Today, the diagnosis of RS is based instead on angiography on Nielsen’s test and digital plethysmography (2). In patients with Vibration-induced white finger, angiography depicts multiple permanent digital arterial occlusions in these patients. The extent of the involvement is proportional to the duration of the causative activity. Angiographic findings of the Hypothenar hammer

159

syndrome include occlusion or aneurysm of the ulnar artery, adjacent to the hook of the hamate bone, and occlusions of digital arteries from microembolization.

Interventional Treatment Treatment of atheromatous lesions of the origins of the innominate and subclavian arteries is warranted in symptomatic patients. Surgical treatment includes transthoracic endarterectomy or bypass graft from the ascending aorta. More often the carotid artery or the contralateral axillary artery are used as inflow vessels for the construction of carotid-subclavian or axillo-axillary bypasses. The results of these operations are excellent with good long-term patency. Because of its minimally invasive nature percutaneous angioplasty with or without stent placement has gained widespread acceptance and is advocated as first line therapy for short isolated lesions. Meticulous technique is required during ostial right subclavian artery stent placement treatment, so to avoid inadvertent occlusion of the right carotid artery. Stent placement is not recommended across the ostium of a patent vertebral artery (Fig. 2). The immediate results are excellent with rare complications; although information on long-term results is scarce, patency appears to be comparable to open surgery. Interventional treatment is also successful in proximal lesions of Takayasu’s arteritis, provided the patient is also treated medically (5). The management of the TOS is surgical and consists of thoracic outlet decompression with resection of the cervical rib or abnormal first rib, or other underlying bone anomaly, and scalenectomy; arterial reconstruction

Brachial Ischemia, Chronic. Figure 2 (a) DSA in the LAO projection depicts occlusion of the left subclavian artery. (b) DSA in the LAO projection after percutaneous placement of a balloon expandable stent; note that the stent does not extend to the vertebral artery ostium.

B

160

Brachial Plexopathy

is mandatory in the presence of an aneurysm; embolectomy or sometimes venous bypass maybe required. As an alternative to aneurysmatectomy exclusion of the aneurysm with a stent-graft can be used (1).

Brachial Plexopathy ▶Trauma, Birth

Brachial Plexus Injury ▶Trauma, Birth

Brachial Vein Obstruction ▶Thrombosis, Venous, Brachial

Brachial Vein Occlusion ▶Thrombosis, Venous, Brachial

Brachialgia ▶Clinical Presentation Compression

of

Spinal

Nerve

Brain P HILIPPE D EMAEREL Department of Radiology, University Hospital K.U.Leuven, Leuven, Belgium [email protected]

Root

Neuroradiology deals with imaging of the central nervous system. The imaging techniques include ultrasound, computed tomography (CT), magnetic resonance imaging (MRI), X-ray myelography, and digital subtraction angiography (DSA). The advent of MRI in the early 1980s has radically changed the radiological approach in patients with neurological symptoms. MRI provides higher resolution and tissue contrast than CT. MRI has also resulted in a decrease in the number of more invasive diagnostic techniques such as DSA and X-ray myelography. This section only deals with neuroimaging of the brain. MRI has replaced CT for several clinical applications either based on the type of suspected pathology (e.g., multiple sclerosis) or based on the expected area of involvement (e.g., brain stem, cerebellum, and pituitary gland). However, there is poor evidence in the literature that MRI has a therapeutic impact or that it changes patient outcome. Quantitative assessments of the clinical effect of MRI in large-case series or well-controlled comparison trials are still missing. Due to its superior image quality, the frequent discovery of incidental abnormalities of unknown significance, particularly in the older population, should be borne in mind. Knowledge of normal aging signs on MRI is important in order to avoid unnecessary additional (possibly invasive) diagnostic or even therapeutic procedures. Apart from morphological imaging, MRI also has the potential of imaging brain function (e.g., spectroscopy, BOLD imaging, perfusion imaging, and diffusion tensor imaging). In the last decade, therapeutic or interventional endovascular procedures have become the modality of choice in the treatment of cerebral aneurysms. However, other pathologies such as arteriovenous malformations, arteriovenous fistulas, and arterial stenoses are also being treated by endovascular procedures. Cranial imaging encompasses imaging of the skull, the meninges, the cerebrospinal fluid-containing spaces, the vascular structures, the cranial nerves, and the brain substance (white matter and gray matter). Pathology can be seen in each of these anatomical compartments. Diagnostic neuroradiology can, to some extent artificially, be divided into an adult and a pediatric subspecialty. In textbooks, the different entities are usually classified into congenital, traumatic, vascular, tumoral, degenerative, metabolic, and infectious pathology. Pediatric neuroradiology is often discussed separately because the type of pathology encountered is often different from adult diseases. The role of MRI in pediatric neurological disease is even more prominent than in adult neurological diseases. The often subtle congenital cerebral malformations and the more frequent occurrence

Brain

of cerebellar pathology are two examples of the prominent role of MRI in pediatric neuroimaging. Although there seems to be an emerging role for diffusion-weighted imaging in cranial trauma, CT remains the modality of choice in daily practice. The close proximity of the CT suite to the emergency department and the need for close monitoring in acute trauma patients are in favor of CT. Bone fractures as well as fresh blood are clearly visible on CT images. MRI is superior in detecting diffuse axonal injuries. Plain skull X-ray is no longer routinely indicated in cranial trauma. Stroke is a common cause of death in the Western world. Stroke imaging is currently one of the “hot” topics in neuroradiology. The search for an optimal selection of patients who can be considered candidates for (intraarterial) thrombolysis is still ongoing. Different MR techniques play a role in the work-up of a stroke patient. Diffusion and perfusion imaging are used to demonstrate the presence of a “penumbra” and to assess the viability of this tissue at risk. An area of restricted diffusion on an apparent diffusion coefficient map needs to be correlated to the perfusion defect. When the area on perfusion appears larger than on diffusion imaging, the area of difference corresponds to the penumbra. MR angiography (MRA) or CT angiography can demonstrate the occlusion of a major artery. Some authors will argue that the rapid accessibility of CT for excluding a fresh hemorrhage is sufficient to decide on thrombolytic or conservative treatment, but diffusion-weighted MRI is certainly more sensitive in detecting recent ischemia. MRI is clearly superior to CT in detecting chronic bleeding. Simultaneously, research is being performed on plaque imaging in the carotid and vertebral arteries. MRA and CT angiography each have advantages in assessing a stenosis. The advantage of CT is the ability to demonstrate calcification in the plaque. But the advantages of MRI include the absence of ionizing radiation and not having to use iodinated contrast agents. Several systematic reviews have been published and according to the observations, there seems to be a consensus that MRA or CT angiography in combination with ultrasound can replace DSA in the majority of patients. Stroke can also be caused by a venous thrombosis or by arterial dissection. Etiologies of venous thrombosis include oral contraceptives (in combination with smoking), pregnancy, and hypercoagulable states. Venous thrombosis can involve the superficial or the deep venous system. Causes of a nontraumatic dissection include fibromuscular dysplasia, hypertension, and oral contraceptives. In both pathologies, MRI is currently the modality of choice to demonstrate the abnormalities, which will lead to the appropriate treatment. The incidence of aneurysms in the population is 2–7% and the risk of bleeding is 2% per year. About 10% of all intracranial aneurysms are multiple. Aneurysms arising

161

from the vessels forming the circle of Willis account for more than 90% of all intracranial aneurysms. MRA is accepted as a screening modality in relatives of patients with aneurysms. Following endovascular treatment, MRA can be used in the follow-up, but occasionally an additional DSA examination may be necessary. MRA at 3 Tesla (T) currently provides excellent visualization of the large and medium-sized arteries (Fig. 1). Arteriovenous malformations are rare congenital lesions and usually present in patients older than 40 years. The risk of bleeding is 2% per year, but several factors may increase this risk (e.g., venous aneurysms, deep venous drainage. In the work-up of a patient with an arteriovenous malformation, DSA remains mandatory. Dural arteriovenous malformation and fistula usually develop secondary to a venous thrombosis. They usually occur in the posterior fossa, tentorium, or in the cavernous sinus. These malformations are frequently missed on CT and MRI and only DSA will demonstrate the fistula. Cavernous and venous malformations are more benign lesions with a small to nonexisting risk of a limited bleeding. MRI has rapidly been accepted as the most important preoperative technique in brain tumors. The multiplanar imaging capability is the main advantage of MRI in

Brain. Figure 1 MR angiogram of the circle of Willis at 3 T. Lateral (a) and cranial (b) view.

B

162

Brain

visualizing the extent of a tumor. Usually, intracranial tumors are classified as intra-axial (parenchymal) tumors and extra-axial (dural/meningeal) tumors. The most common dural tumor is the meningioma. Gliomas (including the World Health Organization grade 2 and 3 astrocytoma as well as glioblastoma multiforme) are the most common type of intrinsic brain tumors. Up to 30% of brain tumors are metastases from a distant primary tumor. The general rule is that MRI will depict more metastases than CT, particularly when it concerns small lesions (Fig. 2). MRI is also more sensitive in the detection of carcinomatous meningitis. With the onset of newer techniques such as MR spectroscopy, perfusion imaging, and diffusion tensor imaging, our insight into tumors is steadily improving. MR spectroscopy may give an idea of the degree of malignancy in a noninvasive way. Perfusion imaging can assess the vascularity of a tumor without further need for DSA. Functional MRI (using BOLD imaging) will help to locate the tumor in relationship to the surrounding

functional cortex, and diffusion tensor imaging will demonstrate the displacement or invasion of white matter tracts by the tumor. Following treatment, MRI is the modality of choice in the follow-up of cerebral pathology. An accurate assessment of residual tumor tissue or the follow-up after chemo- and/or radiotherapy is best performed using MRI. The success of this examination will depend on the patient’s cooperation in the immediate posttherapeutic stage. The increasing use of high field strengths and 3-D imaging sequences has led to more clinical research on morphometry in neurological disease, for example, Alzheimer’s disease, Huntington’s disease, and partial complex seizures. It has been suggested that early stages of Alzheimer’s disease are best distinguished from normal aging by measuring the volume of the hippocampus and/ or temporal lobe. Patients with longer duration of disease were best distinguished by pathology of the medial

Brain. Figure 2 Axial contrast-enhanced CT (a), fluid-attenuated inversion recovery (b), T2-weighted (c), and Gd-enhanced magnetization transfer T1-weighted (d) MR images in a patient with brain metastases from a lung carcinoma. Note that the metastases are only visible on the Gd-enhanced T1-weighted images (d).

Brain

temporal lobe and anterior cingulate gyrus. This appears to be consistent with the progressive clinical presentation of Alzheimer’s disease. In multiple sclerosis (MS), MRI was rapidly recognized to be far superior to CT. For the clinical diagnosis of MS, the sensitivity of MRI exceeds that of all noninvasive clinical tests. Recently, MRI has been included in the criteria for clinically definite MS. Other MRI techniques with increased specificity to the more destructive aspects of MS pathology, such as magnetization transfer MRI, diffusion-weighted MRI, and proton magnetic resonance spectroscopy, have recently been applied to MS cognitive studies. In toxic and infectious diseases, MRI often plays a crucial role in the difficult diagnostic work-up along with the history and laboratory tests. The frequent involvement of deep brain structures, the brain stem, and the cerebellum in toxic and metabolic disorders is far better assessed by MRI than by CT. Staphylococcal brain abscesses can be diagnosed by demonstrating a high signal on diffusion-weighted images, but occasionally necrotic metastases may look similar. Herpes simplex meningoencephalitis will be seen earlier on MRI than on CT. Inflammatory disease of the vessel wall is called vasculitis and can be the result of an infectious etiology (bacterial, viral, tuberculous) or noninfectious etiology (polyarteritis nodosa, granulomatous angiitis). In older patients the differential diagnosis with atheromatosis can be difficult. MRA can demonstrate the segmental narrowings, but when MRA is normal or when only one narrowing is seen, catheter angiography will still be necessary and ultimately biopsy confirmation is needed. MRI does not play a role in viral meningitis but is often indicated in patients with acquired immunodeficiency syndrome who are prone to opportunistic infections.

163

Toxoplasmosis and cryptococcosis are two common opportunistic infections. The differential diagnosis with lymphoma is particularly important. The use of dedicated MR techniques in imaging pathology of the cavernous sinus and pituitary gland has resulted in a clear preference of MRI over CT. The most common pituitary tumor is the adenoma. Most microadenomas (5 cm in diameter. Women with well-differentiated DCIS 1 cm in diameter or comedo-type DCIS treated with BCT and in all patients with close margins (5 mm, seen in MRI only. 2. Presurgical proof of malignancy in highly suspicious lesions >5 mm (BI-RADS V) that are seen in MRI only.

3. 4. 5. 6.

Contraindications Pacemakers and some metal implants constitute contraindications for MRI.

7. 8.

Fischer U (2003) (Hrsg) Mammographiebefund nach BI-RADSTM. Thieme-Verlag, Stuttgart Heywang-Ko¨brunner SH, Schreer J (2003) (Hrsg) Bildgebende Mammadiagnostik. Thieme-Verlag, Stuttgart Lanyi M (2003) (Hrsg) Brustkrankheiten im Mammogramm. Springer-Verlag, Berlin, Heidelberg, New York Parker SH, Jobe WE (1993) (Hrsg) Percutaneous Breast Biopsy. Raven Press, New York Sickles EA (2001) (Hrsg) Breast Imaging. Lippincott Williams & Wilkins, Baltimore Tabar L, Tot T, Dean PB (2003) (Hrsg) Breast Cancer. ThiemeVerlag, Stuttgart

Technique Different devices for MR-based biopsy of the breast use one or more plastic pads for tissue compression. Through several fenestrations within those compression pads, biopsy needles or markers can be placed following exact three-dimensional localization. In bore magnet or closed systems, real-time control of interventions is impossible; they must be performed outside of the bore instead. Provided that MR-compatible equipment is available, conventional core-cut biopsy as well as vacuum intervention is feasible and is performed as described earlier. In discordant findings, repeated MR-guided intervention or open biopsy following MR-guided lesion marking is mandatory. Follow-up MRI is recommended 6 months postintervention. In cases of lesion progression, surgery (for definite clarification) is obligatory (Fig. 4).

Breast, Cysts Fluid-filled structures derived from the terminal duct lobular unit. ▶Fibrocystic Disease, Breast

Breast, Digital Imaging K LAUS -P ETER H ERMANN Gesellschaft Wissenschaftliche Datenverarbeitung mbH, Go¨ttingen, Germany [email protected]

Results The literature reports sensitivity and specificity values of up to 100% for MRI-guided punch or vacuum biopsy.

Bibliography 1.

2.

American College of Radiology (ACR) (2004) (Hrsg) Breast Imaging—Reporting and Data System (BI-RADSTM). Reston, Philadelphia Duda VF, Schulz-Wendtland R (2004) (Hrsg) Mammadiagnostik. Springer-Verlag, Berlin, Heidelberg, New York

Traditionally, mammography has been produced as analog images with screen-film combinations. Digital imaging technology has now been introduced in diagnostic radiology, but for many years no adequate digital technique was available for mammography. There has been much debate about spatial resolution in digital mammography. It was assumed that digital mammography would require spatial resolution similar to that of

Breast, Digital Imaging

screen-film mammography (SFM). However, it has been shown that other features of digital systems could compensate for the lower spatial resolution, leading to better detection of microcalcifications. One characteristic is the detective quantum efficiency (DQE). The DQE is dependent on the radiation dose and the local frequency, and it enables objective comparison between different imaging systems on the basis of image quality and dose efficiency. In digital mammography, contrast can be changed in postprocessing by windowing and leveling. The further advantage of digital systems is the high dynamic range and the linear relationship between dose at the detector and signal intensity, as opposed to the sigmoid relationship between optical density and dose in screen-film systems. Digital image processing technology can display overexposed or slightly underexposed images with normal image quality. The main advantage of any digital imaging system is the separation of image acquisition, processing, and display, allowing optimization of each of these steps. It is hoped that this will enable digital mammography to outperform SFM. In addition, advanced applications such as computer-assisted detection/diagnosis can be easily applied to the digital mammogram.

Digital Systems Various equipment manufacturers have adopted different approaches to digital mammography. Photostimulable phosphor computed radiography was the first digital imaging system to be used for mammography. The imaging plate is contained within a cassette, which can be used in a standard mammography machine without modification. After exposure, the cassette containing the imaging plate is inserted into a reader. The imaging plate is then scanned by a laser. The emitted light is detected by a photomultiplier system, and the resultant electrical signal is digitized. Different high-resolution storage phosphor plates are commercially available (FCR 5000MA, FujiFilm, Tokyo, Japan; CR 850, Kodak, Rochester, NY, USA; Regius 190, Konica-Minolta, Tokyo, Japan; ADC Compact Plus, Agfa, Mortsel, Belgium). Other systems combine digital storage phosphor plates with the direct magnification technique and an X-ray tube with a very small focal spot. This system, however, has failed clinical use. Systems using charge-coupled devices (CCDs) have been used for stereotactic preoperative localizations and percutaneous biopsies. However, CCD chips could not be used for full-field breast imaging due to limitations of the field of view. The SenoScan Full-Field Digital Mammography System (Fischer Imaging Corporation, Denver, CO, USA) uses an array of four phosphor-coupled CCDs in a rectangular configuration. This gives an overall detector

187

size of approximately 1 22 cm. The long axis of the detector is aligned perpendicular to the patient’s body. The X-ray beam is collimated to fit the 1-cm width of the detector. The detector then scans over a distance of 30 cm from left to right synchronously with the X-ray beam. This results in an image size of 22 30 cm. The first online system for digital mammography with US Food and Drug Administration approval was the Senographe 2000D (GE Medical Systems, Waukesha, WI, USA). The 19 23-cm2 detector is based on a semiconductor layer of amorphous silicon with a CsI phosphor layer (CsI/a-Si). The pixel size is 100 mm. The detectors used in the Selenia (Hologic, Bedford, MA, USA) and Novation (Siemens, Erlangen, Germany) systems are based on a semiconductor layer of amorphous selenium, which allows direct conversion from absorbed X-ray into electronic charges. The detector is 24 29 cm2, and the pixel size is 70 mm. Further selenium-based systems are being manufacturing by Instrumentarium Imaging (Tuusula, Finland, currently with GE Medical Systems), IMS (Bologna, Italy), and others. The individual counting of each interacting X-ray photon is the technology for a direct converting digital detector, such as the MicroDose system (Sectra Medical Systems, Linko¨ping, Sweden). Digital mammograms can be printed onto film via a laser printer or presented as a soft-copy display on a monitor. Images are typically displayed for soft-copy reporting on two monitor workstations, with each monitor having a resolution of 2,048 2,650 pixels. Soft-copy displays provide the reader with much greater freedom to change the appearance of the image, and the workstation enables questionable areas to be quickly evaluated. Windowing can be adjusted manually by mouse click. Measurement and annotation tools are available. Digital mammography is likely to improve patient throughput. Further new applications include the ability to use picture archiving and communication systems with the possibility of teleradiology. Furthermore, digital tomosynthesis and contrast digital mammography are two new innovative features that are possible with digital mammography.


4. 5.

Pisano ED, Yaffe MJ (2005) Digital mammography. Radiology 234:353–362 Noel A, Thibault F (2004) Digital detectors for mammography: the technical challenges. Eur Radiol 14:1990–1998 Freer TW, Ulissey MJ (2001) Screening mammography with computer-aided detection: a prospective study of 12,860 patients in a community breast center. Radiology 220:781–786 Niklason LT, Bradles TC, Niklason LE et al (1997) Digital tomosynthesis in breast imaging. Radiology 205:399–406 Jong RA, Yaffe MJ, Skarpathiotakis M et al (2003) Contrastenhanced digital mammography: initial clinical experience. Radiology 228:842–850

B

188

Breast, Fibrosis

Breast, Fibrosis Fibrous proliferation of the mammary stroma. ▶Fibrocystic Disease, Breast

Breast, Infection LUIS P INA Department of Radiology Clıńica Universitaria de Navarra, Navarra, Spain [email protected]

Synonyms Infection of the breast; Mastitis

Definition ▶Mastitis Puerperal, Acute is a bacterial infection of the lactiferous ducts occurring during pregnancy and lactation, usually due to Staphyllococcus aureus (1–3). Acute nonpuerperal mastitis is seen outside the lactation period, although more rarely. Any acute mastitis may evolve to subacute or chronic mastitis if treatment is not adequate. Abscesses and fistulae may be complications (4). Granulomatous conditions, such as ▶granulomatous lobular mastitis, foreign body granulomas, tuberculosis, fungal infections, sarcoidosis, autoimmune diseases, and parasitic infections, may also affect the breast.

Pathology Acute puerperal or nonpuerperal mastitis consists of an acute inflammation of the breast that may be accompanied by focal necrosis. This entity may evolve to organized chronic abscesses and fistulae. Granulomatous lobular mastitis is an inflammation involving the breast lobule with noncaseating granulomata and microabscesses. Microscopic examination in sarcoidosis reveals epithelioid granulomas in the mammary parenchyma, with multinucleated Langhans giant cells that form asteroid or Schaumann bodies. Tuberculous mastitis consists of granulomatous lesions with caseous necrosis. Specific parasitic infections, such as filariasis, may lead to a granulomatous reaction with eosinophilic infiltration and adult filarial worms.

Clinical Presentation Breast mastitis is less frequent nowadays than several years ago. Most cases occur in women between ages 18 and 50 years (3). Acute puerperal mastitis is characterized by pain, erythema, swelling, tenderness, and even systemic signs of infection (3) during the lactational period. If an abscess is formed, a fluctuant palpable mass with adjacent red skin may be observed. Puerperal mastitis typically occurs within 2–3 weeks after the beginning of lactation and is caused by Staphylococcus aureus. This organism is transmitted from the infant. A history of cracked nipple or skin abrasion may be found. Nonpuerperal mastitis may be central (periareolar) or peripheric (3). The former is more common and occurs in young women, especially in cigarette smokers and women with previous periductal mastitis, whereas the latter is less frequent and affects women with underlying diseases such as diabetes, rheumatoid arthritis, trauma, and steroid treatment. Central mastitis is seen as a periareolar inflammation that may or may not be associated with a retroareolar mass, nipple retraction, and purulent nipple discharge. This mastitis is usually indolent and relapsing. The pathogen may be Staphylococcus aureus but also anaerobes (Bacteroides, Peptostreptococcus, Propionilbacterium). Granulomatous lobular mastitis affects young women. The pathogen is unclear, but corynebacteria may be involved. Clinically it appears as a hard firm mass, similar to breast carcinoma (3). Tuberculosis is uncommon in our countries. The breast is involved by lymphatic spread from axillary or mediastinal nodes, and the clinical presentation is an acute abscess. Parasitic infections due to filariasis occur in tropical and semitropical regions. The parasite forms an abscess that is hard and may mimic carcinoma. Axillary nodal enlargement is often found. Breast involvement in systemic sarcoidosis is rare and may produce a firm mass that can be mistaken for carcinoma.

Imaging Mammography Acute mastitis presents as skin thickening, diffusely increased density, reticular edema, and, occasionally, well to ill-defined breast mass (due to abscess formation). Subacute or chronic mastitis may show similar findings on mammography, although scars may also occur (4, 5) (Fig. 1).

Breast, Infection

Ultrasound Ultrasonographic findings of acute mastitis are skin thickening, hyperechogenicity of fat lobules, decreased echogenicity of glandular tissue with posterior shadowing, dilated ducts, and abscess formation. ▶Breast abscesses are usually

189

round to oval lesions, with well-circumscribed or ill-defined margins. Breast abscesses may be anechoic, similar to simple cysts, but most of them are hypoechoic with fluctuating internal echoes (Figs. 2 and 3). Ultrasound is valuable for guiding percutaneous drainage of the abscess (4, 5).

Breast, Infection. Figure 1 Acute mastitis. Increased tissue density and thickening of trabeculae and skin are seen in the left image (craniocaudal view).

Breast, Infection. Figure 2 abscess.

A lobulated mass with indistinct margins is seen in the left image (craniocaudal view). Breast

B

190

Breast, Physiology

3.

4.

5.

Dixon LM, Bundred NJ (2004) Management of disorders of the ductal system and infections. In: Harris JR, Lippman ME, Morrow M (eds) et al Diseases of the Breast. Lippincott Williams & Wilkins, Philadelphia, pp 47–56 Heywang-Ko¨brunner SH, Schreer I, Dershaw DD (1997) Cysts. In: Heywang-Ko¨brunner SH, Schreer I, Dershaw DD (eds) Diagnostic Breast Imaging. Thieme, Stuttgart, pp 156–165 Cardenõsa G (2001) Stroma. In: Cardenõsa G (ed) Breast Imaging Companion. Lippincott Williams & Wilkins, Philadelphia, pp 307–354

Breast, Physiology Breast, Infection. Figure 3 Breast abscess on ultrasound. The lesion is hypoechoic, showing internal echoes and posterior acoustic enhancement. Note the hyperechogenicity of the surrounding fat due to inflammatory infiltrate.

S UE B ARTER Cambridge Breast Unit, Addenbrookes Hospital, Cambridge, UK [email protected]

Definition Magnetic Resonance Magnetic resonance imaging is not indicated as a diagnostic tool for breast inflammatory disease, except in selected cases (for instance, silicone implants or free silicone injections). Contrast enhancement may be observed because of the activity of the inflammatory process. Acute inflammation leads to a significant contrast enhancement, while it may be minimal in a chronic phase. Abscesses are seen as fluidfilled cavities with variable signal intensity in T1- and T2-weighted sequences. Characteristically, the wall of abscesses shows intense enhancement after paramagnetic contrast administration (4).

Breast tissue is responsive to hormonal changes, and an understanding of this is fundamental to the interpretation of breast imaging at particular phases of the menstrual cycle. Hormonal factors also need to be understood in relation to imaging patients before menarche, during pregnancy and lactation, after the menopause, and with hormone replacement therapy (HRT). It is also important to be familiar with the changes affecting the breast parenchyma in order to understand breast conditions that are so common they can be considered to be aberrations rather than disease. This is particularly relevant when considering fibrocystic change.

Nuclear Medicine

Characteristics

Diagnosis: The diagnosis of acute mastitis is based on the clinical findings. Ultrasonography may be performed to detect an abscess, to guide drainage of the abscess, and to evaluate the therapeutic response. If the response is inadequate, mammography and skin biopsy are important for excluding inflammatory carcinoma (4). In subacute or chronic mastitis, ultrasonography is useful to detect abscesses or fistulae. Mammography or skin biopsies are needed to rule out malignancy.

Breast Physiology during the Menstrual Cycle

Bibliography 1.

2.

Rosen PP (2001) Inflammatory and reactive Tumors. In: Rosen PP (ed) Rosen’s Breast Pathology. Lippincott Williams & Wilkins, Philadelphia, pp 29–63 Webster DJT (2000) Infections of the Breast. In: Hughes LE, Mansel RE, Webster DJT (eds) Benign Disorders and Diseases of the Breast. Concepts and Clinical Management. Saunders, London, pp 187–197

Breast morphology is profoundly affected by cyclical variations in hormonal levels during the normal menstrual cycle. Histological changes have been reported in both stroma and epithelium, and are summarized in Table 1 (1). During the proliferative phase of the menstrual cycle, increasing levels of oestrogen stimulate breast epithelial proliferation. The epithelium exhibits sprouting with increased mitotic activity, and increased nuclear density. These microscopic changes are attributed to increases in size and number of intracellular organelles, especially the Golgi apparatus, ribosomes, and mitochondria. Similarly in the luteal phase of the cycle progestogens induce epithelial changes. The breast ducts dilate and the alveolar cells differentiate into secretory cells.

3–7

8–14

15–20

21–27

27–2

I. Proliferative

II. Follicular

III. Luteal

IV. Secretory

V. Menstrual

Dense, cellular

Loose, broken, edematous

Loose, broken

Dense, cellular, collagenous

Dense, cellular

Stroma

Distended with secretion

Open with secretion

Open with some secretion

Defined

Tight

Lumen Cell types

Luminal columnar basophilic cell; intermediate pale cell; basal clear cell with hyperchromaticnucleus (myoepithelial) Luminal basophilic cell; intermediate pale cell; prominent vacuolization of basal clear cell (myoepithelial) Luminal basophilic cell; intermediate pale cell; prominent vacuolization of basal clear cell (myoepithelial) Luminal basophilic cell with scant cytoplasm

Single mainly pale eosinophilic cells

Breast physiology during the the menstrual cycle

Source: Adapted from Vogel PM, Georgiade NG, Fetter BF et al (1981) Am J Pathol 104:23–34.

Days

Phase

Breast, Infection. Table 1

Radial, around lumen



No stratification apparent Radial, around lumen

Orientation of epithelial cells

Epithelium

Absent

Absent

Absent

Present, average 4/10 per HPF Rare

Mitoses

Active apocrine secretion from luminal cell Rare

None

None

None

Active secretion

Breast, Physiology 191

B

192

Breast, Physiology

In the secretory phase, there is an increase in the size of the lobules, and ductules, and the stroma becomes loose and edematous. The basal cells show prominent vacuoles (Fig. 1). Immediately before the onset of menstruation there is a peak in mitotic activity, with lymphocytic infiltration and apoptosis following the onset of menstruation (2). Premenstrual breast fullness and tenderness can therefore be explained by increasing interlobular edema and the proliferation of ductules and acini under hormonal influence.

At parturition an immediate drop in placental hormones occurs allowing the effects of prolactin to become dominant. This causes the breast epithelial cells to change from a presecretory to a secretory state. With the establishment of lactation, there is even greater distension of the glandular lumina, with obliteration of the stroma. Large fat vacuoles are visible in the secretory cells. After pregnancy and lactation, involution occurs at a varying rate between individuals, and after a period of about 3 months the breasts return to normal (4).

Postmenopausal Changes or Involution Pregnancy and Lactation During pregnancy marked proliferation of ducts, alveoli, and lobules occurs under the influence of luteal and placental hormones. Prolactin is released progressively during pregnancy and also stimulates epithelial growth and secretion. In the first 3 to 4 weeks of pregnancy marked ductal sprouting with some branching, and lobule formation occurs mainly under the influence of estrogen. At 5 to 8 weeks breast enlargement is significant, with dilatation of superficial veins, and increasing pigmentation of the nipple/areolar complex (3). Secretory activity starts in the early stages with supranuclear vacuolation. In the second trimester, lobule formation becomes dominant under the influence of progesterone. The alveoli contain colostrum. From the second half of pregnancy onward, the breasts increase in size due to increasing dilatation of the alveoli, as well as hypertrophy of myoepithelial cells, connective tissue, and fat. (Fig. 2)

The term “involution” is used specifically to describe the changes that occur in the breast due to the menopause. These changes begin some years before the cessation of menstrual periods and may start as early as in the 30s in nulliparous women. There is a gradual decrease in the lobular architecture, involving both the stroma and epithelium. The stroma becomes dense, converting into hyaline collagen, resembling normal connective tissue. The basement membrane of the acini becomes thickened, and the epithelium atrophies and becomes flattened. The lumina become narrow with the cessation of secretions. Some acini coalesce with the formation of small cysts. These may later shrink spontaneously and be replaced by fibrous tissue, but may also continue to accumulate fluid and enlarge, presenting symptomatically. The interlobular ducts shrink and some disappear altogether (Fig. 3). The above changes are not uniform and may vary in degree from segment to segment. The stroma gradually is replaced by fat, and the breast becomes softer. This makes it more radiolucent, and hence mammographic screening becomes more sensitive.

Breast, Physiology. Figure 1 Normal breast lobule in the secretory phase showing vacuolation of basal cells. (Courtesy of Dr Sarah Pinder, Consultant Histopathologist, Addenbrookes Hospital, Cambridge.)

Breast, Physiology. Figure 2 Pregnancy change in the breast. Note prominent vacuolation and secretions in the ducts. (Courtesy of Dr Sarah Pinder, Consultant Histopathologist, Addenbrookes Hospital, Cambridge.)

Breast, Physiology

Breast, Physiology. Figure 3 Postmenopausal breast tissue showing atrophy of lobules and dense stroma. (Courtesy of Dr Sarah Pinder, Consultant Histopathologist, Addenbrookes Hospital, Cambridge.)

193

Most studies evaluating the effect of HRT on breast density report higher proportion of relatively radiodense tissue, compared to women who have never used HRT. This is likely to be due to epithelial proliferation (6, 7). Studies have shown some HRT formulations appear to have a greater effect than others depending upon which progestin is used. Cyproterone acetate and medroxyprogesterone acetate (MPA) seem to produce higher incidences of fibrocystic change and increased density on mammography. Unopposed estrogen does not increase mammographic breast density, but approximately 25% of women allocated to combined HRT containing MPA are likely to develop an increase in mammographic density within the first year of use, irrespective of whether the MPA administration is cyclical or continuous (5).

Hormonal Contraceptives Premenarche and Puberty During fetal development, the breast is derived from a modified apocrine or sweat gland. This results in a rudimentary organ, identical in boys and girls, consisting of a few simple branched ducts lying in stroma. At puberty the ducts elongate, divide, and form terminal ductal lobular units (TDLU). TDLU With the onset of puberty, the breasts change in size and shape. Morphologically, there in an increase in size due to an increase in connective tissue and fat. There is elongation and proliferation of the ducts, with budding and branching. New lobules form, and the nipple and areola alter in shape and become more pigmented. All of these changes are hormonally induced. Both estrogen and progesterone stimulate and promote growth of the breast parenchyma.

The combined oral contraceptive pill has been shown to increase breast epithelial proliferation. Whilst proliferation does not appear to differ according to the progestogen (i.e., comparing levonorgestrel with desogestrel), a positive correlation between increasing serum levonorgestrel and proliferative indices has been reported. Evidence about the effect of progestogen-only pills (POPs), depot progestogens or progestogen-releasing intrauterine systems (IUSs) on epithelial proliferation is lacking. Studies evaluating the effect of combined oral contraceptive pills (COCs) on mammographic density are contradictory (5).

Bibliography 1.

Hormone Replacement Therapy 2.

After the menopause, without hormone replacement, the breasts are usually collapsed and soft due to the decreased levels of circulating estrogen and progesterone. HRT reverses the physiological changes of breast involution and therefore benign conditions such as fibroadenoma, cysts, and mastalgia may persist after the menopause with its use. Continuous combined HRT has also been reported to increase the number and size of proliferating breast epithelial cells in tissue biopsies from areas of abnormal mammographic breast density compared with biopsies from women with no history of HRT exposure, or those who are taking unopposed oestrogen (5).

3. 4. 5.

6.

7.

Vogel PM, Georgiade NG, Fetter BF et al (1981) The correlation of histologic changes in the human breast with the menstrual cycle. Am J Pathol 104:23–34 Rudland PS, Barraclough LR, Fernig DG et al (1998) Growth and differentiation of the normal mammary gland and its tumors. Biochem Soc Symp 63:1–20 Rosen PP (2001) Rosen’s Breast Pathology. 2nd edn. Lippincott, Williams & Wilkins, Philadelphia Bland KI, Copeland EM (2004) The Breast. 3rd edn. WB Saunders, Philadelphia Marsden J, A’Hern R (2003) Progestogens and breast cancer risk: the role of hormonal contraceptives and hormone replacement therapy. J Fam Plann Reprod Health Care 29(4):185–187 Million Women Study collaborators (2003) Breast cancer and hormone-replacement therapy in the Million Women study. The Lancet 362:419–427 Million Women Study collaborators (2004) Influence of personal characteristics of individual women on sensitivity and specificity of mammography in the Million Women Study: cohort study. BMJ 329:477–479

B

194

Breast, Special Procedures

Breast, Special Procedures R. S CHULZ -W ENDTLAND, K. A NDERS , W. B AUTZ University of Erlangen-Nu¨rnberg Institute of Radiology, Gynaecological Radiology, Erlangen, Germany [email protected]

Galactography Galactography is a contrast-enhanced study of the lactiferous duct. Indication: Galactography is performed in cases of pathologic secretion, i.e., spontaneous, nonmilky (serous, flocculated, or bloody) secretion from a single duct in one breast. Milky secretion from several ducts or bilateral secretion does not constitute a proper indication. Contraindications: Inflammatory processes of the breast constitute an absolute contraindication, whereas previous reactions to contrast media constitute a relative contraindication for galactography. Adverse events/side effects: Mastitis following galactography as well as local pain in cases of paraductal contrast deposits may occur. Technique: Before galactography, secretion samples for cytologic assessment should be secured. Thorough disinfection of the nipple and the surrounding skin is then followed by careful probing of the nipple after expression of some fluid to mark the lactiferous duct in question. As soon as the probe (e.g., lymphography cannula) hits the ductal orifice, it will sink in without significant resistance. Between 0.3 and 0.5 ml of nonionic contrast medium may be injected, and subsequently orthogonal mammography (craniocaudal and mediolateral) is performed. Kindermann found merely ductal ectasia without intraductal lesions in 65% of his cases; biopsy was performed in 35%, with malignant lesions found in only 4.3% of all cases. Thus, papilloma and papillomatosis accounted for the majority of secretory findings. In cases of bloody secretion, the frequency/prevalence of malignant disease increases up to 37%. The most important findings are: . Regular ducts, . Ductal ectasia, . Filling defects/contrast stop, suspected papilloma, differential diagnosis intraductal cancer (further assessment warranted).

Ductal Puncture/Antegrade Galactography The current widespread use of radial/antiradial ultrasound approaches, oriented along or perpendicular to the lactiferous ducts, has increased the number of morphologically conspicuous ducts without pathological secretion. If further assessment is required, antegrade galactography can be indicated in these cases. Rissanen et al. introduced antegrade, sonographically guided percutaneous galactography in 1993. Technique: The conspicuous duct should be punctured with the least possible injury under sonographic guidance. If the tip of the needle can be clearly visualized within the ductal lumen, contrast agent may be carefully instilled, similar to conventional/retrograde galactography. Subsequently, two orthogonal mammographic views are acquired. Image reading criteria are as for retrograde galactography.

Presurgical Lesion Localization According to European Guidelines (EUSOMA), 70% of all palpable and impalpable lesions are to be assessed histologically before surgery. Additional labeling ahead of therapeutic operation enables the surgeon to exactly predefine oncological/surgical strategies and lesion access. Open biopsy should be undertaken in exceptional cases only. According to EUSOMA standards, the lesion marker should be placed less than 10 mm from the lesion in question in more than 80% of all labeling procedures. To be able to rely on the most exact needle placement possible is of great concern to the surgeon, who wants to perform tissue-saving yet efficient excision and needs to guarantee adequate safety margins in malignant lesions. In histologically benign lesions, open biopsy should yield tissue samples of less than 30 g in 90% of all cases. Before presurgical labeling, apart from standard craniocaudal and oblique views, an additional mediolateral projection should be acquired. Furthermore, the mediolateral projection might be helpful before stereotactic localization. If exact three-dimensional information on lesion position is available, the plausibility of the calculated target area can be verified before needle access.

Freehand Localization By means of orthogonal mammography views, the reader may localize the lesion in question and advance a localization needle toward the lesion center. Usually this results in the least possible trauma for the patient, and the shortest/direct access is also marked, which may be regarded as true advantages. However, in less-experienced

Breast, Special Procedures

examiners, the needle position might have to be revised, resulting in additional mammographic views for postinterventional monitoring. On craniocaudal and mediolateral views, the distance between the nipple and the access incision can be defined, as can the distance between skin surface and the lesion in question. According to these measurements, the needle access can be marked on the patient’s skin. Yet, it should be kept in mind that mammography is performed with tissue compression. Following intervention, the needle position should be checked in two orthogonal planes— and adjusted if necessary. As soon as correct needle position is achieved—and documented—wire placement or dye injection may be performed.

Sonographic Localization Lesion access should be chosen according to the shortest possible skin lesion distance, but it should not interfere with surgical transection pathways. However, to sufficiently visualize the needle and wire during placement, it should be kept parallel to the linear transducer. In circumscribed findings, the marker should penetrate the lesion center and the tip of the wire should be placed and fixed in the tissue distal to the lesion, the maximum tolerable distance being 1 cm. Orthogonal mammographic views following intervention are obligatory. The orientation of the wire as well as the distance between the skin and lesion, the depth of the tip of the wire, and the location of the wire relative to the lesion must be documented. Moreover, information should be given concerning the general extent of the excision and the dimensions in relation to the marker. In noncircumscribed findings—e.g., ductal/segmental changes—the following options are available: . The complete extent of the changes in question is marked with a longer wire, or . The lesion margins are tagged separately with two or more wires, whereas the transverse dimension can be indicated by color marks on the patient’s skin.

195

according to this second view. If correct needle position is achieved, wire placement or dye injection may be performed, and the exact positioning will be documented in orthogonal views. The major drawback of this method is that, as in stereotactic labeling, the wire/tag does not necessarily mark the ideal lesion access for surgery, but a rather prolonged one.

Stereotactic Localization Stereotactic lesion localization is carried out according to stereotactic biopsy as described in the previous section. Following exact needle placement, wire placement and/or dye injection is performed. Apart from stereotactic documentation planes (±15˚), additional orthogonal views (cc and ml views) should be obtained. Misplacement of just a few millimeters under compression may equal considerable misplacement following tissue decompression. It is thus recommended to advance the needle about 5 mm beyond the calculated target. If these rules are observed, stereotactic localization proves to be a very reliable method.

MRI-Guided Localization In lesions exclusively definable with magnetic resonance imaging (MRI), presurgical tagging must be carried out in MRI as well. Without additional (specialized) coils, MRIguided tagging results in marker deviation of ±1 cm from the target lesion because of breathing artifacts. By using dedicated coils, more exact marker placement can be achieved.

Galactographic Localization Lesions visible on galactographs only should be colormarked on the patient’s skin following up-to-date (if necessary, repeated) galactography.

Specimen Radiography/Specimen Ultrasound Localization by Perforated or Tagged Compression Needle Depending on the lesion site, a craniocaudal, mediolateral, or lateromedial mammogram will be obtained using a perforated compression panel. Needle placement is carried out under tissue compression and according to the lesion is coordinated via the perforated/fenestrated compression panel. In doing so, the needle should perforate the lesion to a certain extent, so that after decompression and relaxation of the tissue it still perforates or at least reaches the lesion. The second plane mammogram is then obtained. If necessary, the needle position may be adjusted

In all lesions subjected to presurgical labeling and excision, postsurgery sample X-ray (two planes) or ultrasound must be performed to assure complete removal. Standard and magnification techniques are used. Sample compression may be helpful to increase contrast and facilitate delineation of the resected lesion. In addition, the most suspicious areas (calcifications/solid lesions) should be marked for the pathologist with additional tags/needles. Microcalcifications always warrant additional magnification views, as those will usually give further information about morphology and about more, even smaller, calcifications.

B

196

Breast, Therapy Effects

Written notification must be given to the surgeon and the pathologist, about whether . . . .

The The The The

lesion lesion lesion lesion

has been removed completely; might be incomplete; is definitely incomplete; is not contained in the sample.

In lesions that could be seen and tagged on ultrasound only, additional sample ultrasound may increase confidence concerning successful excision; evaluation obeys the same standards as in sample X-rays (Fig. 7a–e).

Bibliography 1.

2. 3. 4. 5. 6. 7.

8. 9.

American College of Radiology (ACR) (2004) (Hrsg.) Breast imaging—reporting and data system (BI-RADSTM). Reston, Philadelphia Duda VF, Schulz-Wendtland R (2004) (Hrsg.) Mammadiagnostik. Springer-Verlag, Berlin, Heidelberg, New York Fischer U (2003) (Hrsg.) Mammographiebefund nach BI-RADSTM Thieme-Verlag, Stuttgart Heywang- Ko¨brunner SH, Schreer J (2003) (Hrsg.) Bildgebende Mammadiagnostik. Thieme-Verlag, Stuttgart Lanyi M (2003) (Hrsg.) Brustkrankheiten im Mammogramm. Springer-Verlag,Berlin, Heidelberg, New York Parker SH, Jobe WE (1993) (Hrsg.) Percutaneous breast biopsy. Raven Press, New York Rissanen T, Typpo T, Tikkakoshki T, Turunen J, Myllymaki T, Suramo, I (1993) Ultrasound-guided percutaneous galactography. J Clin Ultrasound Oct;21(8):497–502 Sickles EA (2001) (Hrsg.) Breast imaging. Lippincott Williams & Wilkins, Baltimore Tabar L, Tot T, Dean PB (2003) (Hrsg.) Breast cancer.Thieme-Verlag, Stuttgart

Breast, Therapy Effects P INAR B ALCI Dokuz Eyliil University School of Medicine, Department of Radiology, Inciralt-Izmir-Turkey [email protected]

Definitions Breast-conserving treatment involves surgical removal of a breast cancer, often with axillary dissection, usually followed by breast irradiation. In addition to postsurgical changes, changes following axillary node dissection and radiotherapy can also occur. The changes induced by breastconserving treatment with irradiation affect the entire breast and are superimposed on the changes at the surgical site. These changes can mimic and obscure malignancy. Depending on whether a hematoma, a seroma, or fat

necrosis is present, postsurgical changes can have variable radiologic presentations. To avoid diagnostic errors, the radiologist’s knowledge and experience, systematic interpretation, and selection of supplemental radiologic methods is an important part of following up these patients. Therapy effects are also seen following reconstruction, augmentation, and reduction of the breast. Reconstruction can be achieved by means of implants or myocutaneous flaps. Augmentation refers to enlargement of the breast for to correct congenital or acquired anisomastia or micromastia, or for cosmetic reasons. Reduction is performed to achieve symmetry in anisomastia or following mastectomy. Macromastia is a frequent indication for reduction mammoplasty. Neoadjuvant systemic ▶chemotherapy is used to enable breast-conserving surgery in patients with large primary operable breast cancers. It is important to be able to assess response to systemic therapy, both for assisting the surgeon and for prognostic purposes (1). Hormone therapy or hormone replacement therapy (HRT) may affect breast tissue. Use of HRT increases mammographic breast density in 17–73% of women. The incidence of increased density is more frequent with combined estrogen/progestin HRT than with estrogen alone. Breast density changes occur rapidly with HRT, with the greatest changes occurring in the first year. Breast tenderness is also a common side effect of HRT use, particularly at initiation of therapy. Onset of breast pain with HRT initiation is frequently associated with an increase in mammographic breast density, indicating that the onset of breast tenderness associated with initiation of HRT use may signify a proliferative response reflected mammographically as an increase in breast density. While increases in mammographic breast density and breast tenderness are associated with HRT, the method of HRT administration may alter the occurrence and degree of these effects (2). Postmenopausal hormone therapy has been associated with an increased risk of breast cancer, and an increase in mammographic breast density has been reported to occur in a significant proportion of women during such treatment (3). On the other hand, tamoxifen is a widely used antineoplastic agent, not only for early adjuvant therapy but also for prophylaxis and treatment of breast cancer. It has antiestrogenic effects in the female reproductive organs, especially breast parenchyma (4).

Pathology/Histopathology Radiation The breast may be exposed secondarily to radiation during diagnostic procedures such as ▶mammography and fluoroscopy or in the course of radiotherapy administered


to another organ, such as mediastinal radiotherapy for Hodgkin’s disease. The radiation exposure in these situations has been associated with an increased risk for subsequent development of breast carcinoma (1). Radiation of the breast for mammary carcinoma in the course of breast-conserving treatment involves levels of exposure that produce alterations in non-neoplastic as well as neoplastic tissues. Radiation-induced histologic changes must be distinguished from recurrent carcinoma in the interpretation of a post-treatment biopsy. Compared with a preradiation specimen, major changes in a normal breast are apparent in the terminal ductal units. These include the following: (1) collagenation of intralobular stroma, (2) thickening of periductal and periacinar basement membranes, (3) severe atrophy of acinar and ductular epithelium, (4) cytologic atypia of residual epithelial cells, and (5) relatively prominent acinar myoepithelial cells that seem to be preserved better than the epithelial cells. In a minority of specimens, atypical fibroblastin may also be found in the interlobular stroma. In one study, differences in radiation effects among individual patients were not correlated with radiation dose, patient age, post-treatment interval, or the use of adjuvant chemotherapy. Fat necrosis and atypia of stromal fibroblasts are more common in proximity to external boosted or radioactive implanted areas. Radiation-induced vascular changes may occur in external boost procedures. Cytologic atypia can create diagnostic problems even if one is aware of the typical appearance of radiation-induced atrophy of the breast, especially in fine-needle aspiration cytology (5).

Chemotherapy Chemotherapy effects are now most often encountered in the breast when patients with locally advanced or inflammatory carcinoma have been given high-dose systemic therapy preoperatively, as the so-called neoadjuvant therapy. The fundamental manifestation of chemotherapy effects is a decrease in tumor cellularity. The effects may be more pronounced after combined chemotherapy and ▶radiation therapy. The cells are enlarged because of increased cytoplasmic volume. The cytoplasm often contains vacuoles or eosinophilic granules. Cellular borders are typically well defined, and the cells tend to shrink away from the stroma. Some carcinoma cells show enlargement, pleomorphism and hyperchromasia of nuclei. In the most extreme situation, no residual carcinoma may be detectable in 6.7–10% cases. Residual degenerated and infarcted necrotic carcinoma may be recognized by the loss of normal staining properties and decreased architectural detail. With the passage of time, degenerated invasive carcinoma is absorbed. Healed sites may be appreciated because of residual architectural distortion characterized by fibrosis, stromal edema,

197

increased vascularity composed largely of thin-walled vessels, and a chronic inflammatory cell infiltrate. Fibrosis and atrophy of lymphoid tissue are characteristic features of chemotherapy effect at sites of metastatic carcinoma in lymph nodes (5).

Hormone Treatment Effect Hormone therapy has been used to treat mammary carcinoma and postmenopausal replacement for years. In postmenopausal women, normal breast epithelium is stimulated to proliferate, with elongation of small terminal ducts and the formation of lobules. Epithelial changes are accompanied by the accumulation of interlobular connective tissue. The histologic correlates to mammographic breast density during HRT are not fully clear, but they have been suggested to at least partly reflect increased proliferation of epithelium and stroma (5). On the other hand, the main antiproliferative effects of tamoxifen are mediated by competition with estrogen for binding to the cytoplasmic estrogen receptor, with subsequent inhibition by the tamoxifen-estrogen receptor complex of the many activities of endogenous estrogen within tumor cells. Breast parenchyma shows a decrease during tamoxifen therapy (4).

Clinical Presentation Depending on the extent of postoperative edema, the breast is regionally or diffusely dense and swollen in the early postoperative phase. Palpation has limited value in this situation. Axillary dissection or radiotherapy can lead to acute lymphedema of the breast, with swelling and peau d’orange. Radiotherapy leads to erythema, skin thickening, and swelling of the entire breast. A dry epitheliosis of the skin and an edema-induced induration may develop, as well as hyperpigmentation to a variable degree and, in large breasts, a wet epitheliosis along the inferior mammary fold. In general, erythema, edema, and skin thickening largely resolve during the first two years, but considerable variations can be observed. Scar regions may be palpable as flat plateau-like areas, or new nodular and firm areas can develop. Oil cysts are often clinically palpated as movable suspicious nodules or induration. Large dystrophic calcification, oil cysts, and oil granulomas can usually not be distinguished from recurrence. Mammary fibrosis and skin dimpling can be a cause of diagnostic problems clinically or on conventional imaging (1). HRT-induced breast tenderness typically occurs within the first six months and tend to subside with time. This correlates with the findings of a current study in which more women had breast pain in the first six months, and fewer

B

198


noted this problem at one year (2). McNicholas et al (6) found that seven of nine (78%) women who developed moderate or severe breast pain during the first year of continuous combined oral HRT use had an increase in breast density, whereas only two of 21 (10%) women who did not develop breast pain had an increase in breast density.

Imaging The major goals of imaging are early detection of recurrent breast carcinoma following breast-conserving therapy and achievement of the lowest possible rate of false-positive diagnostic biopsies. Mammography combined with clinical examination is the most important diagnostic modality. The highest accuracy can be achieved if both the preoperative and the postoperative studies are available at the time a new mammogram is obtained. Ultrasound may be useful as an adjunct in mammographically dense tissue. The selected use of contrast-enhanced magnetic resonance imaging (MRI) is important. If the evaluation is impaired owing to increased density and scars, contrastenhanced MRI permits markedly improved and earlier detection of recurrence and correct identification of scar-related fibrotic changes beginning one year after radiotherapy (1). Following breast reduction as well as after reconstruction or augmentation with transplanted autogenous tissue, the resultant scar formation and architectural changes are determined by the surgical technique. Tangential mammograms are necessary for evaluating breasts after augmentation or reconstruction with silicone prostheses. Contrast-enhanced MRI has proven to be an important supplemental method for high-risk patients because of its high sensitivity. HRT increases the density of breast parenchyma, thus rendering precocious diagnosis of small tumors more difficult. Any factor that increases breast density may decrease the sensitivity of mammography and influence the accuracy of its interpretation. Changes in mammographic breast density can be evaluated in several ways. The most practical means of evaluating large populations for changes in breast density is to use the Breast Imaging Reporting and Data System (BI-RADS) density categories because this information is typically included in standard mammography reports, bypassing the need to retrieve mammograms. Ultrasound may be useful in mammographically dense breasts (1).

Nuclear Medicine Scintimammography provides functional information by evaluating tracer uptake. Tc-99m sestamibi accumulates

preferentially in tumor cells. The sensitivity and specificity of the method are 83–97% and 70–90%, respectively, in palpable masses. FDG-positron emission tomography (PET) can also detect residual tumor, occult metastasis, or local recurrence, but it may not be suitable for mapping. It is sensitive for measuring response to systemic therapy, thus enabling tailoring of an individual’s treatment. A definite advantage of PET is its capability to provide a fast overview of the whole body and to detect unsuspected metastases (7).

Diagnosis Mammography is the most important method following breast-conserving therapy and irradiation. The following changes are seen: . Diffuse changes, including trabecular coarsening, skin thickening, and diffusely increased breast density secondary to irradiation and axillary dissection . Localized parenchymal changes of the skin and breast tissue due to surgical scarring . Localized parenchymal changes secondary to fat necrosis manifested as liponecrosis, oil cysts, or a lipophagic granuloma . Calcifications Acute changes (diffuse increased breast density, trabecular coarsening, skin thickening, etc.) can resolve slowly during the first two years after radiotherapy. Chronic edema can resolve or undergo fibrotic transformation, producing a similar mammographic pattern (Fig. 1). Localized changes may occur with scar formation following surgery or fat necrosis and its forms of transformation. For example, oil cysts are characterized by round and oval radiolucencies of fat density, small capsules, and egg-shell calcifications. Lipophagic granuloma generally presents as a newly developing mass with an irregular outline. Dystrophic calcifications frequently occur in conjunction with therapy-induced cell and tissue necrosis; these are large, elongated, coarse amorphous calcifications and ringlike, egg-shell forms, scattered dystrophic microcalcifications, and fine punctate calcifications at the site of tumorectomy. Magnification mammography is necessary for an exact analysis. On the other hand, recurrences can become mammographically visible as a nodular or ill-defined mass, enlarging scar, microcalcifications, or diffusely increased density. After the completion of radiation, a baseline mammogram of the treated breast should be done in 3–6 months, followed by a bilateral mammogram after 12 months. In the United States, yearly mammograms are suggested if no suspicion exists. If the cancer contained


199

B

Breast, Therapy Effects. Figure 1 (a) Mammogram (left) in MLO position following breast-conserving therapy showing skin thickening, trabeculary coarsening, metallic clips, and scar formation at the operation site. (b) Mammogram (right) in MLO position shows normal BI-RADS I breast pattern of the same patient. (c) Ultrasound of both breasts showing skin thickening and lymphatic dilatation on the left. Right side reflects normal echo pattern.

microcalcifications, a postoperative mammogram is obtained before radiotherapy. In the acute stage, sonography evaluates varying degrees of radiation-induced skin thickening. The edemarelated echogenicity increases in the subcutaneous space and decreases in the parenchyma, leading to a loss of the normal echo pattern. It can be helpful to distinguish the mammographically visible or palpable hematoma and a simple cyst from a mass. Sonography can detect a recurrence if the breast tissue is very dense. Within the postoperative 12 months, MRI may indicate false-positive calls because of patchy and focal enhancement. Beginning about 12 months after radiotherapy, supplemental MRI can detect very small

recurrent tumors in dense or even irregularly structured tissue with great sensitivity (1). Because of the partial mammographic visualization of the tissue surrounding the prosthesis, mammography plays a limited role in evaluating any scar formation or detecting recurrent disease. The role of mammography is restricted to detecting suspicious microcalcifications. Rarely, a ruptured implant may be detected by mammography. When performed after breast reconstruction with autogenous tissue transfer, mammographic evaluation of areas with dense scar or muscle tissue is limited. If reduction mammoplasty is done, breast changes may change depending on surgical techniques. The most common findings are parenchymal redistribution, elevation of the nipple,

200


calcifications, oil cysts, retroareolar fibrotic bands, and skin thickening (Fig. 2). Sonography is suitable for evaluating the tissue surrounding the prosthesis. Its accuracy is limited because of scar formation. MRI has shown that recurrences around implants and small prosthetic defects can be detected earlier with MRI than with other modalities (1). Preoperative chemotherapy has provided for downsizing and downstaging of breast cancer. In neoadjuvant systemic therapy, conventional techniques of assessing response (clinical examination, X-ray mammography, and ▶breast ultrasound) rely on changes in tumor size, which are often delayed and do not always correlate with pathologic response (8). Improved patient prognosis is a function of pathologic tumor response and residual microscopic disease. These methods are inefficient for monitoring tumor change due to the delay in change noted by clinical examination or current imaging techniques when carcinomas undergo preoperative therapy. The tumors do not necessarily contract in a uniform, spherical way. Pockets of microscopic nests of viable tumor cells may remain scattered throughout the tumor bed. The benefits of enhanced MRI have opened a window for providing evidence of tumor response and residual disease. MRI prediction has been found to be more accurate than clinical examination or mammography in assessing complete response, partial response, or no response. Rieber et al (9) noted that responders had a flattening of the Gd-DTPA uptake curve after the first

chemotherapy cycle or a complete absence of Gd-DTPA uptake after the fourth cycle. Hormone replacement has an impact on the mammographic image. A generalized increase in the extent and density of partially involuted parenchyma is possible (Fig. 3), and a new occurrence or an increase in the size of focal densities can be seen in mammographic images. Estrogen combined with progestins has a stronger association with increased breast density than with the use of estrogen alone and is more commonly observed with the use of continuous combined HRT compared with cyclic HRT. But tibolone is found to have much fewer effects on breast. Tibolone, as a tissue-specific steroid, does not have an estrogenic effect on breast cells. In older women, cysts, fibroadenomas, and other benign breast changes can develop in breasts. Cysts and fibroadenomas can enlarge and simulate a malignant process. After breastconserving treatment, breast density of healthy breast can increase unilaterally because the irradiated fibrosed breast tissue does not respond to hormones (2). Sonography is an important diagnostic method for assessing mammographically dense parenchyma. Some focal findings can be detected and interpreted only by ultrasound in these cases. The glandular tissue after hormone stimulation will generally appear homogenous and moderately hyperechoic, but variations because of breast dysplasia are possible. A simple cyst will not require further evaluation, but solid focal lesions that are not

Breast, Therapy Effects. Figure 2 Findings following reduction mammoplasty. (a) Fibrotic bands in lower quadrant and elevation of the nipple (MLO mammogram). (b) Fibrotic bands and fat necrosis in retroareolar region (CC mammogram). (c) Calcification due to fat necrosis (magnification mammogram).

BRN

201

B

Breast, Therapy Effects. Figure 3 (a) Normal breast pattern of a patient prior to ▶hormone replacement therapy (HRT). (b) Breast pattern of the same patient following two years of HRT, showing a diffuse increase in density.

definitely benign mammographically and sonographically usually require biopsy. The interruption of hormone therapy for 2–3 months may determine whether the lesion is malignant or has regressed. MRI is not indicated for diagnosing changes occurring during HRT. The resulting proliferative changes can be expected to enhance with MRI contrast agents, impairing both detection and exclusion of malignancy (1). On follow-up mammography of breast cancer patients, breast parenchyma has been shown to be decreased after tamoxifen therapy. A tamoxifen-induced breast parenchymal decrease is more significant in postmenopausal women. Because tamoxifen-induced breast parenchymal decrease is relatively less well shown on mammograms in postmenopausal women with fatty changes of the breast, mammography is the most useful method for evaluating breast parenchymal changes after tamoxifen treatment (9).

Bibliography 1. 2.

3.

4.

Heywang-Ko¨brunner SH, Dershaw D, Schreer I (2001) Diagnostic Breast Imaging. 2nd edn. Thieme, Stuttgart. pp 339–374 Harvey J, Kawakami FT, Quebe-Fehling de Palacios PI et al (2005) Hormone replacement therapy and breast density changes.Climacteric 8(2):185–193 Million Women Study Collaborators (2003) Breast cancer and hormone replacement therapy in the Million Women Study. Lancet 362:419–427 Son HJ, Oh KK (1999) Significance of follow up mammography in estimating the effect of tamoxifen in breast cancer patients who have undergone surgery. Am J Roentgenol 173:905–909

5.

6.

7.

8.

9.

Rosen PP (2001) Rosen’s breast pathology. In: Rosen PP (ed) Pathologic Effects of Therapy. Lippincott Williams & Wilkins, Philadelphia pp 887–897 McNicholas MMJ, Heneghan JP, Milner MH et al (1994) Pain and increased mammographic density in women receiving hormone replacement therapy: a prospective study. Am J Roentgenol 163:311–315 Nakamura S, Kenjo H, Nishio Tet al (2001) 3D-MR mammographyguided breast conserving surgery after neoadjuvant chemotherapy: clinical results and future perspectives with reference to FDG-PET. Breast Cancer 8(4):351–354 Beresford M, Padhani AR, Goh V et al (2005) Imaging breast cancer response during neoadjuvant systemic therapy. Expert Rev Anticancer Ther Oct 5(5):893–905 Rieber A, Zeitler H, Rosenthal H et al (1997) MRI of breast cancer: influence of chemotherapy on sensitivity. Brit J Radiol 70:452–458

Bright Liver Ultrasound finding causes a diffuse increase in echogenicity of the hepatic parenchyma, with regular, tiny, tightly packed echoes. This appearance is typically found in hepatic steatosis, but may also be found in hepatic fibrosis. ▶Steatosis, Hepatic

BRN ▶Breast Recurrent Neoplasms

202

Brodie’s Abscess

Brodie’s Abscess A Brodie’s abscess represents an intramedullary abscess formation and is a typical complication of chronic osteomyelitis, especially seen in children. Radiographs may reveal a sharply delineated radiolucent lesion surrounded by fine surrounding sclerosis. Frequent locations are the distal and proximal metaphyses of the tibia.

Bronchial Embolization Bronchial embolization consists in occluding bronchial and non-bronchial systemic supply to the lung in case of ▶Hemoptysis

Bronchiectasis Broken Bones ▶Fractures, Bone, Childhood

Permanent and irreversible dilatation of the bronchial tree as a result of airway obstruction and inflammation. ▶Cystic Fibrosis ▶Airway Disease

Bronchial Adenocarcinoma Bronchiolitis ▶Neoplasms, Chest, Childhood

Bronchial Adenomas Low-grade malignant lesions originating from the APUD system. ▶Neoplasms, Chest, Childhood

Bronchial Arteries In 90% of cases, the source of hemoptysis is the bronchial circulation. The bronchial arteries originate from the proximal portion of the descending thoracic aorta. The right bronchial artery arises from the lateral or dorsolateral aspect of the aorta, most frequently in a common trunk with an intercostal artery (intercostobronchial artery). The left bronchial artery usually originates from the anterior aspect of the thoracic aorta or the concavity of the aortic arch. A left–right bronchial artery may also be seen. ▶Hemoptysis

Bronchial Atresia ▶Congenital Malformations, Tracheobronchial Tree

Inflammation of the respiratory bronchioles due to viral infection. ▶Bronchitis and Bronchiolitis in Childhood ▶Airway Disease

Bronchiolitis Obliterans ▶Bronchitis and Bronchiolitis in Childhood

Bronchioloalveolar Carcinoma An uncommon primary malignant pulmonary neoplasm, and it accounts for 2–14% of all pulmonary malignancies. It is a subtype of adenocarcinoma with a commonly peripheral parenchymal location, no distortion of the pulmonary interstitium and neoplastic cells growth along the intra- and interlobular septa. It can have three different radiological patterns: (1) bilateral, multinodular type, (2) diffuse, infiltrative type involving a single lobe or the entire lung simulating pneumonia and (3) solitary peripheral pulmonary nodule. Growing along the septa is appreciated as crazy-paving pattern. ▶Neoplasms Pulmonary

Bronchitis and Bronchiolitis in Childhood

Bronchitis and Bronchiolitis in Childhood G UNDULA S TAATZ Department of Radiology, Division of Pediatric Radiology, University of Erlangen-Nu¨rnberg, Erlangen, Germany [email protected]

Synonyms Bronchiolitis obliterans; Tracheobronchitis

Definitions Tracheobronchitis represents viral lower respiratory tract infection with thickening of the bronchial walls and involvement of the interstitium. ▶Bronchiolitis is defined as a serious viral infection of the tracheobronchial tree typically due to parainfluenza virus, respiratory syncytial virus or adenovirus occurring in children under 2 years of age. Bronchiolitis obliterans (constrictive bronchiolitis) is characterized by a necrotising fibrotic bronchiolitis with subsequent obliteration of the bronchioles due to extensive scarring. The distal alveolar ducts and alveoli are primarily not involved. Swyer–James–McLeod syndrome, as a variant of postinfectious constrictive bronchiolitis, involves only one lung and develops usually after a pulmonary infection with subsequent alveolar destruction and obliterative bronchiolitis in infancy or early childhood. In bronchiolitis obliterans organizing pneumonia (▶BOOP) proliferative granulation tissue polyps are filling the lumens of terminal and respiratory bronchioles and with extension into alveolar ducts and distal alveoli, organizing pneumonia is associated. BOOP is a rare diagnosis in children but increasingly observed following allogeneic lung or bone marrow transplantation. Asthma is defined as a chronic inflammatory disease of the lower respiratory tract, characterized by repeated episodes of bronchiolitis, reversible airway obstruction, and tracheobronchial mucosal hyperreactivity. Asthma is divided into extrinsic (allergic or atopic), intrinsic (lateonset or nonatopic), and occupational forms. Childhood asthma is usually atopic and often associated with allergic rhinitis and atopic dermatitis.

Pathology/Histopathology In bronchiolitis, the inflammation of the bronchioles causes thickening of the bronchial wall with involvement

203

of both the mucosa and the peribronchial interstitium. The mucosal edema may lead to complete airway obstruction with atelectasis, but ▶air trapping due to partial airway obstruction is more frequent in children. Constrictive bronchiolitis, as a fibrotic bronchiolitis with destruction of the bronchioles, is characterized by histological changes varying from mild bronchiolar inflammation and scarring to concentric fibrosis with complete obliteration of the bronchioles. In BOOP, polypoid masses of granulation tissue in lumens of small airways, alveolar ducts, and some alveoli are identified histologically (1). Asthma is a process involving both central and peripheral airways and is characterized by eosinophilic infiltration and thickening of the airway wall (2). The thickened airway walls show an increased smooth muscle mass, mucous gland hypertrophy, and vascular congestion leading to a markedly reduced airway caliber. Increased amounts of mucus and inflammatory exudate block the airway passages and also cause an increased surface tension favoring airway closure (2).

Clinical Presentation In children with bronchiolitis the infection usually starts in the upper respiratory tract with rhinitis, pharyngitis, and fever. When the disease progresses to the lower respiratory tract cough, tachydyspnea, expiratory distress, chest retractions, diffuse coarse crackles, and wheezing occur. The coughs may be pertussis-like and increasing dyspnea, and cyanosis indicates hypoxia. The main differential diagnosis is asthma, but also conditions such as foreign body aspiration, cystic fibrosis, pulmonary malformations (cysts, tracheo-esophageal fistulas), vascular rings, or gastroesophageal reflux must be considered. Clinical symptoms in BOOP include persistent nonproductive cough, effort dyspnea, low-grade pyrexia, malaise, and weight loss. Pleuritic chest pain and hemoptysis are less common. Auscultation of the thorax reveals fine, dry, lung crepitations, and typically pulmonary function tests show a restrictive pattern.

Imaging Although the diagnosis of tracheobronchitis and bronchiolitis is primarily based on the clinical findings, chest radiographs are commonly obtained to differentiate viral from bacterial infection. Even in bronchiolitis obliterans, chest radiographs may be normal or show unspecific findings like hyperaeration. In childhood asthma, chest X-rays are usually obtained to rule out complications like atelectasis, pneumomediastinum, and pneumothorax. In the past years, high resolution computed tomography (HRCT) of the chest has been frequently used in

B

204

Bronchitis and Bronchiolitis in Childhood

paediatric lung disease, because this technique offers high spatial resolution and excellent evaluation of the lung parenchyma and especially the interstitium (3). Hyperpolarized gas- and molecular oxygen-enhanced MR imaging are two new techniques for high-resolution MR imaging of pulmonary airspaces, but there is only very little clinical experience with these MR techniques especially in children (4).

of the diaphragm and peribronchial infiltrates are demonstrated on the chest X-rays (Fig. 2). Atelectatic areas from mucoid plugging and pleural thickening may also be evident. Swyer–James syndrome is characterized by a unilateral hyperlucent lung with normal or reduced volume during inspiration and air trapping during expiration.

Nuclear Medicine Ventilation and perfusion scintigraphy (VQ scans) provide a relatively noninvasive evaluation of lung function and allow a quantitative assessment of the perfusion of each lung segment. Areas of abnormal aeration and air trapping may be demonstrated, however, VQ scanning gives relatively poor anatomical detail of the lungs and is not commonly used for the assessment of small airways disease in children.

Diagnosis The typical appearance of viral lower respiratory tract infection (tracheobronchitis) on chest radiographs is a symmetric parahilar peribronchial pattern, resulting from thickening of the bronchial walls with involvement of the mucosa and the interstitium. Parahilar peribronchial infiltrates are often referred as bronchopneumonia and hilar lymphadenopathy is usually associated (Fig. 1). In bronchiolitis markedly hyperinflated lungs with flattening

Bronchitis and Bronchiolitis in Childhood. Figure 1 Lower respiratory tract infection in a 3-year-old child. Hyperinflated lungs, parahilar thickening of the bronchial walls, peribronchial infiltrates, and bilateral hilar lymphadenopathy are identified on the plain chest X-ray.

Bronchitis and Bronchiolitis in Childhood. Figure 2 RSV-bronchiolitis. Markedly hyperinflated lungs with flattening of the diaphragm and peribronchial hilar infiltrates are demonstrated on the chest radiograph.

Bronchitis and Bronchiolitis in Childhood. Figure 3 HRCT of bilateral mosaic perfusion pattern in an adolescent with bronchiectasis. A large bulla is visible in the anterior right upper lobe, compressing the upper mediastinum and shifting it to the contralateral side.

Bubbles

High-resolution CT features of small airway disease include air trapping, ground glass opacifications, centrilobar nodules, bronchial wall thickening, bronchiectasis, and ▶mosaic perfusion pattern (5). Mosaic perfusion results from regional perfusion differences caused by a reduced vascularity in lucent lung areas due to secondary hypoxic vasoconstriction (Fig. 3). This pattern is most common in children with asthma, cystic fibrosis, bronchopulmonary dysplasia, and bronchiolitis obliterans. Other HRCT findings in bronchiolitis obliterans are bronchiectasis, bronchial wall thickening, and air trapping in the hyperlucent lung parenchyma. Typical HRCT findings in asthma are air-trapping, bronchial dilatation, bronchial wall thickening, mucoid impaction, and atelectasis. Cystic fibrosis has to be considered as a differential diagnosis to bronchiolitis. Besides air trapping and mosaic perfusion, common findings of cystic fibrosis on HRCT images are bronchiectasis, peribronchial thickening, mucous plugging, centrilobar nodular opacities (mucoid impaction), or cystic and bullous lung lesions.

Bibliography 1. 2. 3. 4.

5.

Epler GR, Colby TV, McLoud TC et al (1985) Bronchiolitis obliterans organizing pneumonia. N Engl J Med 312:152–158 Saetta M, Turato G (2001) Airway pathology in asthma. Eur Respir J Suppl 34:18–23 Copley SJ, Padley SP (2001) High-resolution CT of paediatric lung disease. Eur Radiol 11:2564–2575 McAdams HP, Hatabu H, Donnelly LF et al (2000) Novel techniques for MR imaging of pulmonary airspaces. Magn Reson Imaging Clin N Am 8:205–219 Franquet T, Muller NL (2003) Disorders of the small airways: highresolution computed tomographic features. Semin Respir Crit Care Med 24:437–444

Bronchogenic Cyst Foregut malformation developing from an abnormal budding of the ventral foregut (tracheobronchial tree). ▶Congenital Malformations, Tracheobronchial Tree

Broncholithiasis It is a condition in which a peribronchial calcified nodal disease erodes into or distorts an adjacent bronchus. This disorder results in airway obstruction. Histoplasmosis and tuberculosis are the most frequent causes, but other infections including actinomycosis, coccidioidomycosis and cryptococcosis, as well as silicosis, have been

205

implicated as possible etiologies. Symptoms are commonly couch, haemoptysis, recurrent episodes of fever and purulent sputum. The diagnosis is most reliably diagnosed with CT scan, which can demonstrate: a calcified endobronchial or peribronchial lymph node; bronchopulmonary complication due to obstruction, and the absence of any associated soft tissue mass. ▶Tuberculosis ▶Airway Disease

Bronchopulmonary Dysplasia Chronic lung disease, which occurs in premature infants due to the treatment with oxygen and positive pressure ventilation. ▶Dysplasia, Bronchopulmonary

Bronchopulmonary Sequestration Embryonic mass of lung tissue disconnected from the tracheobronchial tree with a blood supply from the systemic circulation. ▶Congenital Malformations, Tracheobronchial Tree

Bronchoscopy ▶Optical Imaging

Brown Tumors Osteolytic changes that can mimic secondary or primary bone tumors and are found with primary and secondary hyperparathyroidism. ▶Hyperparathyroidism

Bubbles ▶Microbubbles

B

206

Bubbly Lungs

Bubbly Lungs Typical radiographic appearance of bronchopulmonary dysplasia characterized by pseudocysts (bubbles) alternating with interstitial fibrosis and atelectasis. ▶Dysplasia, Bronchopulmonary

nonthrombotic obstruction to hepatic venous outflow (1). The obstruction may completely or partially block the hepatic veins. George Budd (1808–1882) described it in 1845, and Hans Chiari added the first pathologic description of a liver with “obliterating endophlebitis of the hepatic veins” in 1899. The syndrome most often occurs in patients with underlying thrombophilic disorders, including myeloproliferative disorders such as polycythemia vera and paroxysmal nocturnal hemoglobinuria, pregnancy, tumors, chronic inflammatory diseases, clotting disorders, and infections.

Bucket Handle Fracture Pathology/Histopathology Metaphyseal corner Salter II fracture viewed obliquely with respect to the shaft axis, looking like the handle of a bucket, and characteristic of an abuse fracture. ▶Battered Child Syndrome

Buckle (Torus) Fracture A fracture in childhood with a local bending deformity of the normal monotonically curved contour of diaphysis or metaphysis, or of the equivalent in flat bones. ▶Fractures, Bone, Childhood

Budd–Chiari Syndrome I. K AARE T ESDAL Department of Radiology and Nuclear Medicine, Klinikum, Friedrichshafen, Friedrichshafen, Germany [email protected]

Synonyms Hepatic obstruction; Nonthrombotic hepatic vein obstruction; Obliterating hepatic vein endophlebitis; Thrombotic hepatic vein obstruction

Definition Budd–Chiari syndrome is a rare condition and describes an entity of diseases characterized by thrombotic or

Obstruction of intrahepatic veins leads to hepatic congestion and hepatopathy as blood flows into, but not out of the liver. Characteristically, the caudate lobe of the liver is spared due to direct venous channels from the inferior vena cava. The blood accumulation in the liver raises the pressure in the nonoccluded hepatic veins and in the portal veins leading to portal hypertension. Hepatocellular injury results from microvascular ischemia due to congestion, and liver insufficiency result.

Clinical Presentation In the Western world, the Budd–Chiari syndrome is predominantly seen in women. Age at presentation is usually the third or fourth decade of life, although the condition may also occur in children or elderly persons. The symptoms of Budd–Chiari syndrome may begin suddenly and severely, but usually they begin gradually. Hepatomegaly, ascites, and abdominal pain characterize Budd–Chiari syndrome, but all these symptoms are nonspecific. Four main clinical variants have been described: acute liver disease, subacute liver disease, fulminant liver disease, and chronic liver failure. The most common presentation is subacute liver disease complicated by symptoms of portal hypertension and varying degrees of liver insufficiency. A model has been created using the following equation that allows for the prediction of survival of patients with BCS: 1.27 encephalopathy + 1.04 ascites + 0.72 prothrombin time + 0.004 bilirubin. Based on this model, patients can be separated into three groups with good, intermediate, and poor 5-year survivals. The parameters determining the outcome are accompanying complications such as portal vein thrombosis, thrombosis of inferior vena cava, and renal failure. (a) Acute and subacute forms: these are characterized by rapid development of abdominal pain, ascites, hepatomegaly, jaundice, and renal failure. (b) Fulminant form:

Budd–Chiari Syndrome

the patients may present with fulminant or subfulminant hepatic failure along with ascites, tender hepatomegaly, jaundice, and renal failure. This form of presentation is uncommon. (c) Chronic form: this is the most common form, and patients present with progressive ascites, and approximately 50% of patients also have renal impairment. Further severe complications in these patients are progressive liver insufficiency, hydrothorax, hepatic encephalopathy, or variceal bleeding.

Imaging As imaging modalities we use ultrasound, magnetic resonance imaging (MRI), angiography of hepatic veins,

207

and computed tomography (CT) scan to diagnose Budd–Chiari syndrome. Using ultrasound, thrombi can be visualized, and duplex sonography is the preferred mode. The sensitivity and specificity of combining different modes of ultrasound are 85–90%. MRI scanning with pulsed sequencing helps in the assessment of hepatic venous and portal blood flow, and the sensitivity and specificity are more than 90%. Angiography of hepatic veins (hepatic venography) can demonstrate thrombi or membranous webs if the hepatic veins are partially open. CT is the workhorse imaging system in most radiology departments and diagnostic centers dealing with liver diseases (Fig. 1). Contrast CT is fast, patient friendly, and has the unique ability to image all thoracic and abdominal structures. The advances in technology and

Budd–Chiari Syndrome. Figure 1 CT in a 34-year-old woman presenting symptoms of subacute Budd–Chiari syndrome and severe variceal bleeding. The arterial phase shows a hypertrophy of the liver, and an inhomogeneous parenchyma due to subtotal occlusion of the hepatic veins (venous outflow).

B

208

Bulging Disk

clinical performance using multidetector CT, enables the diagnosis of Budd–Chiari syndrome and the accompanying complications (e.g. liver cirrhosis, ascites, hydrothorax).

Diagnosis The most important diagnostic workups are the imaging modalities like duplex sonography and CT. In addition to the routine laboratory tests (usually nonspecific), the following tests should be performed to evaluate for a hypercoagulable state: Protein C activity, antithrombin, total and free protein S, activated protein C resistance, prothrombin gene G2021OA mutation, homocysteine concentration, factor V Leiden mutation, lupus anticoagulants, plasminogen, fibrinogen, and heparin antibodies. The diagnosis of bone marrow disorders like polycythemia vera, essential thrombocytosis, myeloproliferative syndrome, and myelofibrosis has to be established or excluded according to international definitions. Examination of ascitic fluid provides useful clues to the diagnosis since patients usually have high protein concentrations (>2 g/dL), but this may not be present in persons with the acute form of Budd–Chiari syndrome. A biopsy of the liver is not compulsory for the diagnosis. The typical histologic findings after liver biopsy are high-grade venous congestion and centrilobular liver cell atrophy and thrombi within the terminal hepatic venules. The extent of fibrosis and cirrhosis can be determined based on biopsy findings.

Interventional Radiological Treatment Treatment options include medical therapy, balloon dilation (PTA) of hepatic vein, portal systemic shunt surgery, transjugular intrahepatic portosystemic shunt (TIPS), and liver transplantation. Irrespective of the course of the disease, a side-to-side shunt or liver transplantation is indicated if medical treatment fails. In recent years, a number of reports of TIPS as a treatment for Budd–Chiari syndrome (BCS) have appeared.

References 1.

2.

3.

4.

Dilawari JB, Bambery P, Chawla Y et al (1994)Hepatic outflow obstruction (Budd–Chiari syndrome). Experience with 177 patients and a review of the literature. Medicine 73:21–36 Boyer TD, Haskal ZJ (2005) The role of transjugular intrahepatic portosystemic shunt in the management of portal hypertension. Hepatology 41:385–400 Tesdal IK, Filser T, Weiß Ch et al (2005) Transjugular intrahepatic portosystemic shunt: adjunctive embolotherapy of gastroesophageal collaterals in the prevention of variceal rebleeding. Radiology 236:360–367 Ro¨ssle M, Olschewski M, Siegerstetter V et al (2004) The Budd– Chiari syndrome: outcome after treatment with transjugular intrahepatic portosystemic shunt. Surgery 135:394–403

Bulging Disk A bulging disk is one in which the contour of the outer annulus extends in the axial plane beyond the edges of the disk space, over greater than 50% of the circumference of the disk and usually less than 3 mm beyond the edges of the vertebral body. ▶Herniation, Intervertebral Disk

Bulla This has been defined as a sharply demarcated area of emphysema measuring >1cm in size and being delineated by a thin (1 mm) epithelialized wall. It may be difficult to distinguish a bulla from a cyst, but bullae are usually associated with extensive emphysematous changes elsewhere in the lung. ▶Pulmonary Opacity, Cystic Pattern

BUN ▶Blood Urea Nitrogen

Byler Disease Form of progressive familial intrahepatic cholestasis pertaining to a heterogeneous group of autosomal recessive childhood cholestasis of hepatocellular origin. This form was first described in Amish kindred. It is characterized by cholestasis often arising in the neonatal period that leads to liver failure. ▶Congenital Malformations, Liver and Biliary Tract

Byler’s Disease Familial intrahepatic cholestasis, initially described in Amish descendants of Jacob Byler. The disease is

Byler’s Disease

characterized by the absence of a gene encoding for canalicular transport and bile formation. An autosomal recessive inheritance with two subtypes called respectively low GGT PFIC 1 and 2 has been demonstrated. Clinically patients present with a chronic cholestatic syndrome that

209

begins in infancy and usually progresses to cirrhosis within the first decade of life. Common features are hepatocellular cholestasis with low serum levels of gamma glutamyl transpeptidase activity. ▶Hepatic, Pediatric Tumors, Malignant

B

C

CA-125 CA-125, a glycoprotein antigen, is currently the most commonly used tumor marker for ovarian cancer. Its value lies in the assessment and monitoring of recurrent ovarian cancer. In primary ovarian cancer it is less specific, particularly in premenopausal women and in early-stage disease. ▶Carcinoma, Ovarium

Cacchi-Ricci disease ▶Medullary Sponge Kidney

Caffey Disease A LAN E. O ESTREICH Cincinnati Children’s Hospital Medical Center, Cincinnati, Ohio, USA [email protected]

Synonym Infantile cortical hyperostosis

Definitions A skeletal condition in infants and fetuses manifest by periosteal reaction, involving only a few bones, especially tubular bones and the mandible and scapula, associated with warmth and tenderness, and not due to a known specific cause. The condition is generally self-limited and heals. Because of its close simulation of the skeletal effect

of long-term use of high dose prostaglandin E, it may well be metabolic in nature; however, others (including, at one time, Caffey) have postulated a viral cause, and sometimes it occurs in several members of a family, raising the question of genetic susceptibility. As a periosteal process, it manifests only at sites of membranous bone growth.

Pathology/Histopathology The hyperostosis (periosteal reaction) of Caffey disease eventually heals into a somewhat thick cortex, which slowly returns to normal over time. The role of biopsy would be principally to exclude competing diagnoses, such as osteomyelitis or metastases from neuroblastoma, in the unusual circumstance that the diagnosis is in question. The periosteal reaction of Caffey disease will stop short of the short, straight ▶metaphyseal collar at the metaphysis, since such reaction only occurs where periosteum is present, and it cannot occur on exclusively enchondral bones (viz, tarsal or carpal bones) that do not have a periosteum.

Clinical Presentation The earliest presentation of Caffey disease is intrauterine polyhydramnios, presumably as a consequence of mandible involvement making swallowing painful for the fetus. After a baby is born, the findings are warmth, tenderness, possible swelling, and reduction in motion of the affected bones. Similar signs may be found in high dose prostaglandin E use leading to periosteal reaction (1); additionally vomiting may occur from antral narrowing in babies so treated. The Scotch terrier puppies that get a Caffey-like hyperostosis of the mandible refuse food or drink because of pain, but survive nicely if tube fed during the active disease. If periosteal reaction simulating Caffey disease is due to neuroblastoma, metastases, redness, warmth, and even tenderness may be absent (unless there is a pathological fracture).

212

Caffey Disease

Imaging Careful analysis of long bones, mandible (Fig. 1), scapula, clavicle, and ribs for periosteal reaction on plain radiographs, using magnification of the images in symptomatic areas, is the method to document Caffey disease. In high dose prostaglandin E disease, we have seen similar involvement of small tubular bones of the hand (Fig. 2); perhaps that may also occur in Caffey disease. Periosteal reaction can be shown with ultrasound or CT, but plain images suffice, especially if careful attention is directed to symptomatic sites and 10 days have elapsed since the onset of the disease. On prenatal ultrasound, evaluation for polyhydramnios is pertinent.

Nuclear Medicine

Caffey Disease. Figure 1 Right-sided mandibular periosteal reaction in Caffey disease. From Oestreich AE, Crawford AH (1985) Atlas of Pediatric Orthopedic Radiology. Thieme Verlag, Stuttgart p 104.

Since it takes about 10 days for periosteal reaction of any etiology to appear on conventional radiographs, nuclear imaging is capable of demonstrating abnormal activity earlier, as early as the first day of the process, in Caffey disease, as well as earlier than 10 days in simulating conditions, including prostaglandin E high dose, hypervitaminosis A, trauma (such as abuse), and tumor (such

Caffey Disease. Figure 2 Painful swollen forearm following 3 months of prostaglandin E therapy. The deep soft tissues are swollen (white arrows). Note periosteal reaction at metacarpal 3 (3). The metaphyseal collars (curved arrows) are spared from the periosteal reaction. The earlier cortex of the ulna (black arrow) is barely discernable. From Oestreich AE, Shownkeen H (1993) Massive prostaglandin-E periosteal reaction in an infant. Lessons to be learned. Year Book of Pediatr Radiol (Miskolc) 5:49.

Calcification, Intracranial, Neonatal

as neuroblastoma metastases; once neuroblastoma is suspected, I131MIBG is the nuclear method of choice). Caffey disease is particularly supported by increased mandible activity on bone scan.

213

Calcification, Intracranial, Neonatal V ERONICA D ONOGHUE

Diagnosis Periosteal reaction along long bones in Caffey disease stops short of the 1–3 mm metaphyseal collar (Fig. 2). For the mandible, clavicle and scapula, it may merely present as enlarged dense bone. Rib disease is best seen if the most lateral portions are involved. Other causes of diffuse periosteal reaction in infancy or later in childhood include neuroblastoma metastasis, Weismann–Netter Stuhl disease, Erdheim-Chester disease, ▶Melhem hyperostosis and hyperphosphatemia (2), as well as physiologic periosteal reaction found in a significant percentage of normal (and even abnormal) infants 1–6 months of age (3).

Bibliography 1.

2.

3.

Benz-Bohm G, Emons D et al (1984) Cortical hyperostoses after long-term prostaglandin E2 therapy [in German]. Radiologe 24:72–78 Melhem RE, Najjar SS, Khachadurian AK (1970) Cortical hyperostosis with hyperphosphatemia: a new syndrome? J Pediatr 77:986–990 Shopfner CE (1966) Periosteal bone growth in normal infants. A preliminary report. Am J Roentgenol Radium Ther Nucl Med 97:154–163

Radiology Department, Children’s University Hospital, Dublin, Ireland [email protected]

Synonym TORCH infections

Definition Pathological calcification of the newborn brain is detected on plain radiography or CT examination. The numerous causes include perinatal toxoplasmosis, others i.e. HIV, rubella, cytomegalovirus, herpes (TORCH) infections, metabolic abnormalities, vascular disease, brain infarction, tumours and ▶neurophakomatoses.

Pathology/Histopathology The pathology in neonatal intracranial calcification varies with the underlying cause. Each one is discussed in more detail elsewhere in this section.

Clinical Presentation

Calcific Periarthritis Calcifications within periarticular soft tissues frequently with a history of pain. ▶HADD

Calcific Tendinosis or Calcific Tendinotis Calcifications within tendons frequently with a history of pain. ▶HADD

Some infants with congenital infection, in particular cytomegalovirus present with a small head circumference and infants with toxoplasmosis may have hydrocephalus (1) All these infections are generalised infections and the infants may have skin rashes and haemorrhages, hepatosplenomegaly, chorioretinitis and pneumonitis. The diagnosis of tuberous sclerosis depends on detecting the characteristic cutaneous lesions and the presence of cardiac tumours such as rhabdomyomas which can be diagnosed using antenatal sonography. They can also present with renal cysts and hamartomas. In Sturge–Weber disease a facial haemangioma causing a port-wine stain in the distribution of the trigeminal nerve is characteristic. In Von–Hippel–Lindau disease there are cerebellar and retinal haemangiomas and the patients may also present with polycystic kidney disease and cysts in other organs.

C

214

Calcification, Intracranial, Neonatal

Many of these patients including those with inherited metabolic diseases and brain infarction may also present with seizures. A ▶pseudo-TORCH syndrome has been described such that patients who are microcephalic at birth have delayed motor and cognitive development, long tract signs and their work-up for congenital infections is normal. Many have affected siblings and parental consanguinity has been reported in approximately onethird of patients suggesting an autosomal recessive inheritance (3).

Imaging Intracranial calcification maybe curvilinear, linear, punctuate or clumpy. If dense it may be seen on plain radiographs but it is more easily identified with CT (Fig. 1). Ultrasonography can identify calcification particularly if it is in the periventricular regions (Fig. 2). MR imaging is a poor tool to detect calcification but it may be seen using gradient-echo sequences and also as reduced signal on T2 and increased signal on T1 sequences.

Calcification, Intracranial, Neonatal. Figure 1 CT brain on newborn infant with congenital cytomegalovirus infection. There is extensive intracranial calcification of a punctuate, clumpy and curvilinear nature.

In cytomegalovirus and less commonly rubella and toxoplasmosis infections, the calcification tends to be periventricular in location (Figs. 1 and 2). It may also involve the basal ganglia. In the other infections the location is non-specific and can occur almost anywhere in the brain. Increased echogenicity of the lenticulostriate regions on sonography, sometimes referred to as noncalcifying vasculopathy, has been reported in infants with congenital infection. This however is a non-specific finding and can also be seen in some trisomy syndromes, ischaemia and storage disorders (2). MR and CT imaging may also detect various brain developmental anomalies such as delayed myelination, lissencephaly, polymicrogyria, cerebellar hypoplasia and schizencephaly, depending on the gestational age at the time of infection. Infants with pseudo-TORCH syndrome also demonstrate calcification which is mainly periventricular in location but may also involve the basal ganglia. They also have ventricular dilatation secondary to white matter volume loss and the cerebellum and brainstem may be small. Intracranial tubers of tuberous sclerosis are seldom calcified in the newborn period. In Sturge–Weber syndrome, the calcification characteristically occurs in the cerebral cortex underlying the angiomatosis and typically has a gyriform pattern. The calcification however is seldom seen in the newborn period. In tumours and areas of infarction the calcification is non-specific.

Calcification, Intracranial, Neonatal. Figure 2 Ultrasound image in infant with congenital cytomegalovirus infection. There is extensive increased periventricular echogenicity in keeping with calcification (arrows).

Calcifications, Breast

Diagnosis If dense, intracranial calcification can be seen on plain radiography. It may also be identified on sonography particularly when periventricular in location. CT is the most sensitive imaging modality. MR imaging is poor at detecting calcification but it maybe suggested on gradient-echo sequences and also as reduced signal on T2 and increased signal on T1 sequences.

Bibliography 1. 2.

3.

Barkovich JA (ed) (2005) Pediatric Neuroimaging. 4th edn. Lippincott Williams and Wilkins, Baltimore El Ayoubi M, de Bethmann O, Monset-Couchard M (2003) Lenticulostriate echogenic vessels: clinical and sonographic study of 70 neonatal cases. Pediatr Radiol 33:697–703 Reardon W, Hockey A, Silberstein P et al (1994) Autosomal recessive congenital intrauterine infection-like syndrome of microcephaly, intracranial calcification and CNS disease. Am J Med Genet 52:58–65

Calcifications, Breast A RNE F ISCHMANN University of Tu¨bingen Department of Diagnostic Radiology, Tu¨bingen, Germany [email protected]

215

While most calcifications are easily detectable in H&E-stained sections, calcium oxalate is obscure to standard work-up and only visible in polarized light. As calcium oxalate is mainly associated to benign disease, detection is important to avoid unnecessary follow-up biopsies. Radiologic–pathologic correlation of calcifications is difficult, as up to 25% of calcifications are lost during embedding and fixation of specimens. Moreover, microcalcifications smaller than 100 are undetectable at mammography. Therefore all specimens taken for microcalcifications should undergo specimen radiography and the pathologist should thoroughly search for calcifications marked radiologically.

Clinical Presentation Due to their size and distribution, calcifications are impalpable and clinically invisible, therefore only lesions with mass effect and associated calcifications are clinically detectable. Approximately 40–50% of mammary carcinomas have calcifications that are mammographically detectable. Intraductal comedocarcinoma and infiltrative carcinoma, in particular, are frequently associated with calcifications, whereas lobular carcinomas are rarely calcified. Among the benign lesions, sclerosing adenosis is associated with calcifications with a frequency of 50%. With 58% positive predictive value in patients treated with breast-conserving therapy, calcifications are a valuable tool in the detection of recurrence.

Synonym Imaging Microcalcifications

Definition Crystals of calcium phosphate, calcium carbonate, calcium oxalate, and magnesium phosphate developing in different benign or malignant transformations of breast tissue.

Pathology/Histopathology Calcifications are usually localized in the terminal ductulolobular unit (1). In general, the various calcifications are deposited in areas where compound lipids are present. Calcifications may be psammous, granular, laminar, or amorphous. Sometimes they form circular structures referred to as Liesegang rings within cysts (2).

Mammography is the primary imaging modality in the detection and description of calcifications. The widespread use of screening mammography since the 1970s has caused a substantial increase in the incidence of calcifications. In general, detection of clustered calcifications in mammography should advocate the performance of magnification views, preferably in craniocaudal and mediolateral positioning to localize precisely the calcifications and to detect milk of calcium. Sonography is useful in showing cystic changes in fibrocystic disease. In some cases microcalcifications can be viewed directly on sonographs. Despite the characteristic appearance of calcium on magnetic resonance imaging (MRI) with low signal on T1-weighted and T2-weighted images, MRI has no practical use in the detection or classification of calcifications. This is mainly due to the small size of single

C

216

Calcifications, Breast

calcifications, which is far below the voxel size. MRI is therefore only able to detect the underlying disease. As sensitivity for in-situ carcinoma is about 80%, a normal MRI in cases of suspicious microcalcifications cannot rule out malignant disease. In fibroglandular or dense breast tissue, preoperative magnetic resonance tomography can be useful in detecting multifocal disease. This is also true for new calcifications after breast-conserving therapy, where it is difficult to distinguish liponecrotic calcifications in statu nascendi from recurrent breast cancer.

Nuclear Medicine Currently there is no known tracer for classifying calcifications. Nuclear medicine is therefore confined to cases where calcifications lead to a diagnosis of breast cancer and require further work-up.

Diagnosis Mammography is currently the only imaging modality that can detect and classify microcalcifications with sufficient diagnostic accuracy. Morphology and distribution are the most important parameters when differentiating benign from malignant calcifications, whereas the number of calcifications is of little importance. All calcifications should be classified according to the categories described in the Breast Imaging Reporting and Data System (BI-RADS) lexicon by the American College of Radiology (ACR). A summary of these descriptions is given below. Typically, benign calcifications can rule out malignancy with a high probability and are usually pathognomonic. These include: skin calcifications, vascular (Fig. 1), round, popcorn-like (fibroadenoma), large rod-like (plasma cell mastitis), lucent centered, egg-shell or rim calcifications (lipid necrosis), milk of calcium (Fig. 1), suture, and dystrophic calcifications. Coarse, heterogeneous calcifications are of intermediate concern and often represent fibroadenoma, fibrosis, or trauma. Amorphous or punctuate calcifications have a diameter of less than 0.5 mm, and are usually indistinct (Fig. 2). Pathologically they often represent psammomalike structures (1). Fine pleomorphic calcifications are of similar shape but vary in size without casting parts, which are more suspicious (3). Fine-linear or fine-linear branching calcifications are usually associated with comedocarcinoma (Fig. 3).

Calcifications, Breast. Figure 1 Multiple crescent shaped calcifications in this lateral view represent milk of calcium. These are typically benign as well as the vascular calcifications in this image.

Calcifications, Breast. Figure 2 A small group of amorphous calcifications in a segmental distribution representing ductal carcinoma in situ.

Groups can be classified into segmental, linear, multilocular, regional (>2cm2 in an area that does not suggest a ductal distribution), grouped or clustered (120 mg/dL) and effective statin therapy (high-density lipoprotein 40 mm 5 mm – ≤10 mm T1c: >10 mm – ≤20 mm T2 tumor >2 cm – ≤5 cm T3 tumor ≥5 cm T4 metastatic disease T4a infiltration of thoracic wall T4b infiltration of skin T4c infiltration of both structures T4d inflammatory carcinoma N regional lymph nodes Nx lymph node not judgeable N0 no regional lymph nodes N1 metastases in ipsilateral axillary lymph nodes, moveable N2 metastases in ipsilateral axillary lymph nodes, not moveable N3 metastases in ipsilateral lymph nodes of mammaria interna. Furthermore, peritumoral intraductal tumor components must be distinguished: PIC predominant intraductal component More than 80% intraductal component and 20% invasive component EIC extensive intraductal component Between 25 and 80% intraductal component Less than 75% invasive component

C

236

Carcinoma, Ductal, Invasive

SIC small intraductal component Less than 25% intraductal component More than 75% invasive component. Multifocality and multicentricity are discussed in the chapter about multiple carcinomas of the breast.

Clinical Presentation The patient may present with a hard, palpable mass and/ or skin, and/or nipple retraction. More details are described in the section on primary and secondary signs. However, sometimes the tumor is occult.

Imaging All descriptions of the imaging characteristics of invasive tumors should be made according to the BI-RADS classifications. Furthermore, the localization and size of the tumor must be reported with precision. The differential diagnosis includes invasive lobular carcinoma, radial scar, or scars after surgery. Some fat necrosis or abscesses could mimic an ▶invasive ductal carcinoma. Changes from a previous mammogram should be considered, such as a new density, mass, or microcalcifications.

Mammography A spiculated mass with irregular margins is a typical sign of invasive ductal carcinomas. The density of the tumor is often higher than that of the parenchyma. In some cases, the tumors also present with amorphous or pleomorphic microcalcifications. Other imaging features may be focal asymmetry or architectural distortion, and therefore changes from a previous mammogram must be interpreted carefully (Figs. 1 and 2).

Ultrasound The tumor is often characterized by an irregular, hypoechoic mass, typically more tall than wide, with a thick echogenic rim and posterior acoustic enhancement. The tumor is not comprisable.

MR Mammography Typical findings on magnetic resonance (MR) images are rim-enhancing masses, which are irregular or spiculated with heterogeneous enhancement. Signal intensity curves show a high initial contrast media uptake and a postinitial

Carcinoma, Ductal, Invasive. Figure 1 Patient with a new, palpable mass in the upper, outer quadrant of the left side. On mammography, an irregular and spiculated mass of 1.5 cm with skin retraction is seen on mediolateral oblique (a) and craniocaudal views (b). Histology revealed an invasive ductal carcinoma, IDC pT1c N0 G2.

Carcinoma, Endometrium Uteri

237

C

Carcinoma, Ductal, Invasive. Figure 2 Patient with a new, palpable mass on the right side. The tumor is difficult to see on the mammography due to the density of the parenchyma (a). On ultrasound, the tumor is characterized by an irregular, hypoechoic, 1-cm mass, more tall than wide with a thick echogenic rim and posterior acoustic enhancement (b). The tumor is not comprisable. Histology revealed an invasive ductal carcinoma.

plateau or wash-out. On water-sensitive sequences, the tumor has an intermediary signal and on T1-weighted images the tumor is hypointense (Fig. 3).

Bibliography 1.

2.

Newstead GM, Baute PB, Toth HK (1992) Invasive lobular and ductal carcinoma: mammographic findings and stage at diagnosis. Radiology 184:623–627 Boetes C, Mus RD, Holland R et al (1995) Breast tumors: comparative accuracy of MR imaging relative to mammography and US for demonstrating extent. Radiology 197:743–747

Carcinoma, Endometrium Uteri M ARTIN U NTERWEGER Diagnostic Radiology, Cantonal Hospital Baden, Baden, Switzerland [email protected]

Synonyms Endometrial carcinoma

Definition Endometrial carcinomas are the most common malignancies of the female pelvis. The mortality of this

disease has decreased in the last few years because of earlier detection and advances in tumor treatment. The peak incidence of endometrial carcinoma is around 60 years. The disease is uncommon before the age of 40 years. Risk factors include nulliparity, infertility, obesity, diabetes, and Stein–Leventhal syndrome. One important variable is prolonged stimulation of the endometrium with high-dose estrogen treatment, that is, postmenopausal hormonal replacement or oral contraception. On the other hand, endometrial carcinoma is rarely seen in patients with ovarian agenesis.

Pathology/Histopathology Endometrial carcinoma arises from the glandular component of the endometrium in the upper uterus. It may grow in a circumscribed pattern, presenting as a focal mass protruding into the uterine cavity and occasionally within an endometrial polyp. However, endometrial carcinomas can also grow diffusely, involving multiple parts of the endometrium. Approximately 90–95% of endometrium carcinomas are adenocarcinomas (Fig. 3). The remaining 10% of nonepithelial uterine cancers comprise sarcomas, mixed tumors, and secondary malignancies. Some less common epithelial varieties include mucinous, secretory, clear cell, and papillary serous carcinomas. The clear cell tumor is highly malignant and has a bad prognosis. Endometrial carcinoma typically arises from the endometrial mucosa. The depth of

238


Carcinoma, Ductal, Invasive. Figure 3 Mammography of extremely dense breast tissue (ACR type 4). Tumor is difficult to see on the right side in the inner quadrant (a). MR mammography shows an irregular and spiculated rim-enhancing mass in the right breast (b). The size of the suspected tumor is about 1 cm and it is localized in the inner upper quadrant of the right breast. On T2-weighted imaging, the tumor has intermediary signal. Signal intensity curve shows a rapid initial increase with a plateau in the delayed phase.

myometrial invasion is a very important prognostic factor and correlates with frequency of lymphatic metastasis. The most common sites of distant metastases are the lung, liver, brain, and bone.

Clinical Presentation Postmenopausal bleeding is the most important presenting symptom of endometrial carcinoma. Because there are multiple benign causes of postmenopausal bleeding, including estrogen therapy, endometrial hypertrophy,

atrophic vaginitis, and endometrial and cervical polyps, only 15% of patients with this symptom and endometrial carcinoma will be diagnosed. Less than 5% of patients with carcinoma of the endometrium are asymptomatic at the time of diagnosis. Endometrial carcinoma in premenopausal women usually presents as abnormal uterine bleeding, oligomenorrhea, or menometrorrhagia. The diagnosis of endometrial carcinoma is made histologically with fractional endocervical curettage. In most institutions, the International Federation of Gynecology and Obstetrics (FIGO) classification is applied for staging of endometrial carcinomas (Fig. 1).


239

C

Carcinoma, Endometrium Uteri. Figure 1 Staging of endometrial carcinoma according to TNM and the International Federation of Gynecology and Obstetrics (FIGO).

Around 70% of patients are diagnosed with stage I disease. The 5-year survival in this subpopulation is 76%. This survival rate decreases to 59% for stage II and 29% for stage III. Tumor grade also shows a correlation with 5-year survival.

Imaging On sonography, stage I endometrial carcinoma typically appears as widening of the hyperechogenic endometrium. The postmenopausal endometrium is atrophic and generally measures less than 3 mm. A thickness of greater than 5 mm should always be considered abnormal in women not receiving estrogen therapy. Endovaginal sonography can be useful for screening women with postmenopausal bleeding. The value of computed tomography (CT) in local staging of endometrial carcinoma is limited, since the spread of endometrial carcinomas is normally small or microscopic to the cervix, parametrium, and lymph nodes. On magnetic resonance imaging (MRI), the signal intensity of endometrial carcinomas is typically similar to that of normal endometrium on nonenhanced MR images. Tumors are generally hyperintense on T2-weighted images compared to the endometrium and isointense on T1-weighted images. The lesions can be heterogeneous because of necrosis or contents of blood products in the tumor. Large carcinomas are often seen as polypoid masses expanding the endometrial cavity. Secondary signs of small tumors include an increased thickness or lobulation of the endometrium. Endometrial carcinoma enhances variably on dynamic gadoliniumenhanced images. The enhancement is typically different from that of normal myometrium or endometrium. With

contrast-enhanced images, the detection of small tumors and differentiation between lesions and fluid or necrosis can be improved. Thus, contrast agents should be routinely applied for staging of endometrial carcinoma. Myometrial invasion is best evaluated on T2-weighted images (Fig. 2) and gadolinium-enhanced scans (Fig. 3). An intact junctional zone is present in stage IA without myometrial invasion. In stage IB, lesions infiltrate the junctional zone and the inner part of the myometrium. Invasion of more than 50% of the myometrium indicates stage IC disease. Cervical invasion (stage II) is best seen on T2-weighted or contrast-enhanced images in the sagittal plane. In stage III, the local extrauterine disease is demonstrated as extension of tumor outside the myometrium, adnexal masses, vaginal metastases, or pelvic lymphadenopathy. Rectal or bladder wall invasion of tumor (stage IVA) is suspected if the normal fat planes between these organs are interrupted. The overall accuracy of gadolinium-enhanced MRI in the staging of endometrial carcinoma is between 84 and 94%. Tumor recurrence usually depends on the type of therapy. Patients treated with radiation and surgery typically present with distant metastases. Patients treated only with surgery more often present with pelvic wall, parametrial, or vaginal apex recurrences. Most recurrent tumors will occur within 3 years after initiation of treatment. Early-stage and low-grade tumors often recur late, more than 5 years after initiation of treatment.

Diagnosis Endocervical and endometrial biopsy is the only reliable option for diagnosis.

240


Carcinoma, Endometrium Uteri. Figure 2 (a–c) Pictorial review using MR imaging and CT of the female pelvis. Endometrial carcinoma in stage IC. (a) Sagittal T2-weighted magnetic resonance (MR) image. The T2-weighted image demonstrates only an increased amount of fluid in the uterine cavity. Demarcation of the histologically proven endometrial carcinoma is missing. (b) Sagittal T1-weighted MR image immediately after intravenous contrast administration. Significantly better demarcation of the carcinoma relative to the myometrium is noted. The hypovascularized endometrial carcinoma extends to the outer layers of the myometrium. (c) Gross specimen. Note the good morphologic correlation with the contrast-enhanced MR image. (Figure 9a–c: Reprinted from Hamm, B., Kubik-Huch, R.A., Fleige, B., Eur. Radiol,. 9, 3–15, 1999).

Carcinoma, Endometrium Uteri. Figure 3 (a–c) (a) Axial T2-weighted magnetic resonance imaging (MRI) demonstrates an enlarged nonhomogenous hyperintense uterus, which was histologically proven as an endometrial carcinoma, stage III. (b) Axial T2-weighted MRI shows huge ovaries on both sides, which shows the invasion. (c) Histology of an endometrial adenocarcinoma of the uterus.

Carcinoma, Hypopharynx

Bibliography 1.

2.

3. 4. 5.

Frei KA, Kinkel K, Bonel HM et al (2000) Prediction of deep myometrial invasion in patients with endometrial cancer: clinical utility of contrast-enhanced MR imaging a meta analysis and Bayesian analysis. Radiology 216:444–449 Seki H, Takan T, Sakai K (1999) Value of dynamic MR imaging in assessing endometrial carcinoma involvement of the cervix. Am J Roentgenol., Jul 2000; 175:171–176 Kinkel K, Kaja Y, Yu KKet al (2000) Radiological staging in patients with endometrial cancer: a Meta-analysis. Radiology 1999; 212:711–718 Tavassoli FA, and Deville P (2003) Tumor of the breast and female genital organs. WHO classification of tumors Hamm B, Kubik-Huch RA, Fleige B (1999) Eur Radiol 9:3–15

Carcinoma, Gallbladder Malignant tumor arising from the gallbladder epithelium. ▶Neoplasms, Gallbladder

Carcinoma, Hypopharynx S TEFANO C IRILLO Institute for Cancer Research and Treatment, Turin, Italy [email protected]

Definition Hypopharyngeal carcinoma is a malignant neoplasm originating from the mucosal surface of the hypopharynx. The ▶hypopharynx is a musculomembranous conduit; it lies behind the larynx and extends from its junction with the oropharynx at the tip of the epiglottis (at the level of the hyoid bone) superiorly to the lower border of the cricoid cartilage inferiorly. The muscularsupporting structure is formed by the middle and inferior pharyngeal constrictor muscles. It extends inferiorly down to the cricopharyngeus muscle, where the pharynx joins with the cervical esophagus. The hypopharynx can be divided into three segments: the pyriform sinus, the postcricoid area, and the posterior pharyngeal wall.

Pathology/Histopathology The most frequent macroscopic presentation of hypopharyngeal carcinoma is an ulcerative infiltrative lesion (90%

241

of the cases), often associated with a necrotic evolution and erosion of the adjacent cartilaginous, membranous and muscular anatomic structures. The latter represent a form of barrier which hinders further diffusion to the contiguous areas. Over 95% of malignancies of the hypopharynx are squamous cell carcinomas (well differentiated or undifferentiated); the remaining 5% are squamous cell carcinoma with sarcomatous component, undifferentiated carcinomas, and verrucous and basaloid carcinoma. The hypopharyngeal carcinomas can arise in the pyriform fossa (60%), in the posterior hypopharyngeal wall (25%), and in the postcricoid area (15%). The cancer spread depends on the site of origin: when the diagnosis is made at a late stage the site of origin is often unrecognizable. Carcinomas of the pyriform sinus are rarely confined. They can extend to the supraglottis larynx involving the aryepiglottic folds, the fatty tissue of the superior paralaryngeal space, and the preepiglottic space; or they can involve the glottic plane anteriorly. Pyriform sinus tumors originating from the apex and the lateral walls often invade the thyroid cartilage and they may extend directly to the thyroid. Carcinomas of the posterior hypopharyngeal wall can spread in a craniocaudal direction (to the oropharynx or esophagus), in a circumferential direction, or deeply and posteriorly to the prevertebral muscles. Carcinomas of the postcricoid area show a submucosal spread, anteriorly to the posterior cricothyroid muscle and the cricoid cartilage, circumferentially or caudally to the esophagus (1).

Clinical Presentation Its incidence varies widely. In males, it ranges from 0.3 per 100.000 inhabitants in Iceland to 17 per 100 in France. Among women, the variability is less pronounced. Hypopharynx carcinoma represents approximately 8 and 5% of head and neck epithelial cancers in males and females, respectively, excluding tumors of the skin and the thyroid gland. Most patients who develop cancer of the hypopharynx have a history of heavy smoking and drinking. Males are about eight times more susceptible to cancer of the hypopharynx than females, In females of Irish and Scandinavian descent who present with Plummer–Vinson syndrome—characterized by esophageal webs, iron deficiency anemia, glossitis, and increased incidence of esophageal cancer—there is also an increased incidence of carcinoma of the postcricoid region. Hypopharynx carcinoma may remain asymptomatic for a long period; at presentation, the disease is often

C

242


advanced. The characteristic symptoms are sore throat, otalgia due to involvement of the Arnolds nerve, a branch of the tenth pair of the cranial nerves, and dysphagia. A “hot potato” voice or hoarseness due to vocal cord paralysis may be present. Among patients with ▶head and neck cancers, hypopharyngeal carcinomas carry a lower survival rate than cancers of other sites. There are several reasons for this poor prognosis: the hypopharynx is a “silent” area and at the time of presentation patients are often at an advanced stage of disease. Seventy-five percent of patients have lymph nodal metastases at the time of diagnosis, 36% have clinically evident adenopathies, 20–40% have distant metastases, and 4–15% show a second synchronous or metachronous tumor. Tumor size is also associated with metastases to the neck, as they occur in 50% of cases when the tumor is bigger than 4 cm, but in 85% of the cases when it is smaller than 4 cm. The overall survival of patients with hypopharyngeal carcinoma is about 40% at 5 years. The site, size, and presence or absence of neck metastases have a significant effect on the outcome; in patients with cervical metastases, there is a 20–25% risk of distant metastases within 2 years of treatment. Stage I and II posterior hypopharyngeal wall carcinomas have an excellent prognosis. In contrast, even small pyriform sinus carcinomas are notorious for metastasizing early and carry a poor prognosis. Postcricoid lesions usually present as advanced lesions with extensive paratracheal and mediastinal metastases, and have a poor prognosis (Table 1) (2).

Diagnosis The physical examination should include a thorough head and neck assessment with particular attention given to the oral cavity and the patient’s general appearance for signs of severe nutritional deficiencies. The physical examination must be associated with indirect mirror examination and direct endoscopy. The tumor must be confirmed histologically, and any other pathologic data obtained from a biopsy should be included. The most important goal of the staging endoscopy is to determine the lowermost extent of the tumor and its relation with the pyriform apex and the cervical esophagus. Esophagoscopy and biopsies should be performed after mapping of the tumor, and complete evaluation of the esophagus down to the gastroesophageal junction is mandatory, as the esophagus is the most frequent site of asymptomatic synchronous primary tumors. Hypopharyngeal carcinomas arise from the mucosa, thus they are usually visible at the surface, but their

Carcinoma, Hypopharynx. Table 1 hypopharynx carcinoma

T staging for

T1 Tumor limited to one subsite of the hypopharynx and ≤2 cm in greatest dimension T2 Tumor invading more than one subsite of the hypopharynx or an adjacent site, or measuring >2 cm but ≤4 cm in the largest diameter without fixation of the hemilarynx T3 Tumor measuring >4 cm in largest dimension or fixation of the hemilarynx T4 Tumor invading the thyroid/cricoid cartilage, hyoid bone, thyroid gland, esophagus, or central compartment soft tissue, which includes prelaryngeal strap muscles and subcutaneous fat T5 Tumor invading the prevertebral fascia, encasing the carotid artery, or involving the mediastinal structures

submucosal extension cannot be evaluated by endoscopic examination alone. Indeed, clinical and endoscopic examination may underestimate tumor extension. Integration between clinical, endoscopic, and imaging (CT/MR) data is necessary for the correct staging.

Imaging The disease extent is often endoscopically underestimated because of submucosal tumor spread that can be well visualized with computed tomography (CT) or ▶magnetic resonance imaging (MRI). CT and MRI scans are often valuable to further delineate disease extent at the primary site and in the neck. Imaging techniques may, however, underestimate the mucosal growth and overestimate its extension due to difficulty in differentiating cancer tissue from perilesional edematous reaction. Thus, integration between clinical–endoscopic examination and imaging is necessary. The role of imaging in hypopharynx carcinoma is as follows: . . . . .

Identifying the cancer Measuring its volume and dimension Evaluating deep tissue diffusion Identifying and characterizing of adenopathies Follow-up.

Conventional radiography with barium swallow has a limited role for staging, despite a reported sensitivity of 96, 87, 44% in detection of cancer of the pyriform fossa, posterior hypopharyngeal wall, and postcricoid area, respectively. ▶Multidetector CT (MDCT) is the first-choice exam for staging hypopharyngeal carcinoma: because of its rapidity, motion artifacts can be erased and scans


243

C

Carcinoma, Hypopharynx. Figure 1 Squamocellular cancer of right pyriform sinus—MDCT: extension anteriorly to paraglottic space, associated with enlargement of thyroarytenoid space.

Carcinoma, Hypopharynx. Figure 2 Squamocellular cancer of left pyriform sinus—MDCT (a) shows cancer in contiguity with the left thyroid cartilage, sclerotic. MRI (b) T1-weighted image shows low intensity of left thyroid cartilage for infiltration.

can be performed during functional and dynamic maneuvers, such as phonation or Valsalva, sometimes needed for better visualization of the apex of the pyriform sinus. MDCT allows the radiologists to make multiplanar reconstruction of excellent quality, obtaining imaging on different planes with optimal tumor visualization and delimitation of tumor volume and of the size of any associated adenopathy. Identification is possible by evaluation of a spaceoccupying mass, an area with anomalous enhancement, an obliteration of the adipose space, an asymmetric enlargement of soft tissue, and an asymmetry of the pyriform sinus. For identification, the sensitivity of MDCT (95%) is higher than that of MRI (87%). A critical point for hypopharynx carcinoma is the infiltration of the pharyngeal constrictor muscle, the laryngeal cartilages, the paraglottic space, and the prevertebral space (Fig. 1).

The accuracy of MDCT in staging hypopharynx carcinomas is 84–86.4% versus 70–87% of MRI. Although MRI permits an accurate contrast resolution of the anatomic structure, it can be biased by motion artifacts, due to the long acquisition time, especially in elderly patients with dyspnea, cough, and large expiratory excursions. MRI is necessary when MDCT cannot exclude infiltration of the laryngeal cartilages, the prevertebral muscle, and the cervical esophagus. MRI has a better sensitivity than MDCT (97 vs. 68% respectively) for assessing cartilage infiltration. Thyroid cartilage infiltration is present in most cancers arising from the apex and in 55% of those arising from the lateral walls (Fig. 2). For infiltration of the prevertebral muscles, MRI has a sensitivity of 88%, a specificity of 14–29%, and a diagnostic accuracy of 53–60% (Fig. 3). The MRI criteria useful for determining infiltration of the cervical esophagus show variable accuracy, ranging from 67 to 86%.

244

Carcinoma, Lobular, In situ, Breast

possible false-positive findings. Anatomic limitations can be partially overcome by the combination of CT– PET, which permits simultaneous acquisitions of morphological and metabolic data.

Bibliography 1. 2.

3.

4.

Carcinoma, Hypopharynx. Figure 3 Squamocellular cancer of posterior wall—MRI: T1-weighted image after i. v. gadolinium administration: thickening of posterior wall, associated with involvement of the prevertebral muscles.

The role of imaging is also to identify adenopathies that cannot be assessed by physical examination, by characterizing them and detecting possible capsular rupture. Ultrasonography is the most useful examination for assessing superficial laterocervical adenopathies, by providing an accurate morphologic and dimensional evaluation. MDCT and MRI evaluation is indicated for deep adenopathies (i.e., retrolateral pharyngeal, which represent an important prognostic factor for local recurrence and distant metastases). Another imaging objective is to identify lymph node capsular rupture, present in 23% of lymph nodes greater than 10 mm, 40% of those smaller than 20 mm, 50% of those greater 20 mm, and 70% of those greater than 30 mm. Capsular rupture is present in 25% of lymph nodes presenting with normal dimensional criteria (3, 4).

Nuclear Medicine Nuclear medicine tests are useful in the follow-up of patients who underwent surgery or radiotherapy, for whom there is a clinical suspicion of local recurrence. In this setting, FDG-positron emission tomography (PET) provides a high sensitivity (86–100%) and specificity (69–87%) (5). PET has some limitations due to its low spatial resolution and the absence of anatomic reference points, which is of particular relevance considering the complexity of the head and neck region. Moreover, uptake is at times present in physiological conditions generating

5.

American Cancer Society (2006) Cancer Facts and Figures 2006. American Cancer Society, Atlanta, GA Sturgis EM, Wei Q, Spitz MR (2004) Descriptive epidemiology and risk factors for head and neck cancer. Semin Oncol December 31(6):726–733 Keberle M, Kenn W, Hahn D (2002) Current concepts in imaging of laryngeal and hypopharyngeal cancer. Eur Radiol July 12(7):1672–1683. Epub 2002 Feb 9 Becker M, Hasso AN (1996) Imaging of malignant neoplasms of the pharynx and larynx. In: Taveras JM, Ferruci JT (eds), Radiology: Diagnosis—Imaging—Intervention Lippincott, Philadelphia, PA, pp 1–16 Sharon F. Hain (2005) Positron emission tomography in cancer of the head and neck. Brit J Oral Maxillofac Surg 43:1–6

Carcinoma, Lobular, In situ, Breast A RNE F ISCHMANN Department of Diagnostic Radiology, University of Tu¨bingen, Tu¨bingen, Germany [email protected]

Synonyms Lobular intraepithelial neoplasia (LIN); Lobular neoplasia (LN)

Definition Lobular carcinoma in situ (LCIS) is carcinoma filling and dilating the intralobular ductuli with loosely aggregated monomorphous cells without stromal invasion. Contrary to prior beliefs describing LCIS as a premalignant lesion, it is nowadays regarded as a risk factor for invasive cancer. Atypical lobular hyperplasia (ALH) is defined as a lesion that either has some, but not all, of the features of LCIS or has all the features of LCIS but involves up to only 50–75% of a lobule.

Pathology/Histopathology The cells are uniform, bland, and homogenous without mitosis and are often arranged in a linear file or planar

Carcinoma, Lobular, In situ, Breast

growth pattern. Pathology is confined to the terminal ductulolobular unit, although a pagetoid involvement of the ducts can sometimes be observed. Necrosis and microcalcifications may occur in LCIS but are rare. If less than half of one terminal ductulolobular unit is involved, the lesion is called ALH. It remains unclear whether the separation of LCIS and ALH is of clinical relevance because several features used to distinguish ALH and LCIS have no influence on prognosis (1). Similar to the classification of ductal intraepithelial neoplasia (DIN), ALH, and LCIS are classified as ▶lobular intraepithelial neoplasia (LIN) 1–3 depending on the extent of lobular involvement: 1. LIN 1 represents partial or complete filled acini without distention. 2. In LIN 2, multiple or all acini are filled and distended but to a lower degree than in LIN 3. Undermining growth into the terminal ductuli may occur. 3. LIN 3 represents extensive distention of the acini or confluent acini. If goblet cells are detected or the polymorphous form of LCIS is present, the lesion is classified as LIN 3 irrespective of extent. Coexistence of ductal and lobular neoplasia in one specimen occurs in up to 16% of cases. Cells are estrogen-receptor positive in up to 60%, a higher rate by far than in invasive carcinoma. A loss of E-cadherin enables the cells to move relatively freely in the ductulolobular system. LCIS is multicentric in 47–93% and bilateral in 30–67%. Generally, a diffuse involvement of the breast is detected. Therefore, some authors advocate assuming bilateral involvement if LCIS is detected in a specimen. A finding of lobular hyperplasia, especially with atypia, should mandate further sectioning to exclude invasive lobular carcinoma (ILC), with which it is associated in 5–16% of cases (2).

Clinical Presentation LCIS does not lead to a palpable mass and is almost always an incidental finding in a specimen retrieved for another reason. Hence, the true incidence of LCIS in the general population is unknown, as it has no clinical or mammographic manifestation. LCIS is found in about 1.1–3% of breast biopsies and in 5.7% of all breast malignancies. Although ILC is associated with contralateral carcinoma, contralateral malignancy is lower in LCIS (2%) than in DCIS (6%). Generally, LCIS is multifocal (47– 93%) and bilateral (30–67%).

245

The average age of women with LCIS is between 44 and 46, which is 10 years younger than for DCIS. A high proportion (90%) are premenopausal. LCIS is today considered to be a risk indicator for invasive breast neoplasia rather than being a true precursor lesion: 2.2–7% of patients with LCIS develop invasive carcinoma in 5 years, and 35% develop ipsilateral and 25% contralateral ILC in 20 years, which indicates a fivefold risk of breast carcinoma. A positive family history of breast carcinoma increases this risk to about 11-fold. More than 50% of these malignancies occur more than 15 years after LCIS is diagnosed, in contrast to DCIS, in which more than 90% of recurrences occur within 5 years, indicating the slow growth pattern of LCIS. Most malignancies that follow LCIS are of ductal origin, supporting the view, that LCIS is a risk factor rather than a precursor of malignant disease. One explanation of this fact would be that coexisting DCIS grows faster and becomes invasive sooner than LCIS. Mortality for LICS is low—less than 7.2% in longterm follow-up, and newer studies even report absent mortality in LCIS. Risk factors associated with LCIS are comparable to those associated with DCIS or invasive carcinoma: family history or previous history of cancer, late age of pregnancy, or nulliparity. Increasing age actually reduces the risk of subsequent carcinoma in women with LCIS.

Therapy Because of the wide dissemination of disease, early studies suggested mastectomy for DCIS: after open biopsy, up to 60% of patients were found to have residual LCIS and 6% to have residual ILC in a consecutive mastectomy, in 80% involving a different quadrant. Hence, clear margins cannot resolve the risk of recurrence. Some groups have suggested bilateral mastectomy on prophylactic grounds. Because initial mastectomy has never been shown to reduce mortality over observation alone, conservative treatment—usually close long-term follow-up—is more advisable, especially considering the relatively young age of the patients. Ample biopsy in the contralateral breast has been proposed to examine bilateral disease but is nowadays no longer acceptable in the absence of clinical or imaging criteria for biopsy. LCIS is to be viewed as a risk factor, like a family history of breast cancer, rather than as a premalignant lesion. There is no evidence that radiation therapy, chemotherapy, or axillary lymph node sampling or dissection have any therapeutic role in LCIS. Tamoxifen has reduced the risk of invasive breast carcinoma by 56% in women with LCIS and by 86% in women with ALH.

C

246

Carcinoma, Lobular, Invasive

Imaging Due to the slow growth and infiltrative pattern, neither calcifications nor a solid mass is visible in LCIS. The only mammographic sign is sometimes a parenchyma-like aggregation of breast tissue. Calcifications—usually of the powderish type—are sometimes found in adjacent breast tissue, referred to as “neighborhood calcifications.” In rare cases, LIN 3 can present with necrosis and associated microcalcifications, usually of the fine granular type (3). Because these are often associated with ILC, more aggressive treatment should be considered. Sonography does not show any specific sign. Although 30% of patients with LCIS show an unsharp region of enhancement on magnetic resonance imaging (MRI), the lesion is not distinguishable from breast parenchyma or mastopathic changes. No prospective signs of malignancy can be detected.

Nuclear Medicine At the present time, nuclear medicine has no role in the evaluation of LCIS. Diffuse heterogeneous uptake of 99m Tc-sestamibi and 99mTc-(V)DMSA has been reported. Still, the number of cases was too low to render this method more than experimental. Because of its slow growth, LCIS does not show increased uptake of 18FDG.

Carcinoma, Lobular, Invasive S ILVIA O BENAUER Department of Radiology, Georg-August-University of Goettingen, Goettingen, Germany [email protected]

Synonyms Infiltrating lobular carcinoma (ILC)

Definition About 10% of all breast cancers are invasive lobular carcinomas (ILCs). The cytologic features of the cells suggest that they arise from the lobules.

Pathology The tumor is more often multifocal or multicentric or controversial (bilaterality). The tumor metastasizes frequently to bone, the peritoneum, adrenals, gastrointestinal tract, ovary, and leptomeninges.

Diagnosis

Clinical

LCIS is usually a chance finding, detected in specimens taken for evaluation of a neighboring finding. If LCIS is found in a specimen, it must be questioned whether the clinical symptoms can be explained by the LCIS. If not, a further biopsy should be considered to rule out neighboring ILC or DCIS, especially because a rate of underestimation in ▶core needle biopsy of 31% has been reported (4). MRI can be performed to rule out invasive disease associated with LCIS.

The patient may present with a palpable mass and with negative mammography and ultrasound. Sometimes the patient reports a decrease in size of the affected breast (“▶shrinking”).


4.

Tavassoli FA (1999) Lobular neoplasia. In: Tavassoli FA (ed) Pathology of the Breast. McGraw-Hill, New York, pp 373–397 Frykberg ER (1999) Lobular carcinoma in situ of the breast. Breast J 5(5):296–303 Sapino A, Frigerio A, Peterse JL et al (2000) Mammographically detected in situ lobular carcinomas of the breast. Virchows Archiv 436(5):421–430 Elsheikh TM, Silverman JF (2005) Follow-up surgical excision is indicated when breast core needle biopsies show atypical lobular hyperplasia or lobular carcinoma in situ: a correlative study of 33 patients with review of the literature. Am J Surg Pathol 29(4):534–543

Imaging All descriptions of the imaging of the invasive tumors should be done according to the BI-RADS classifications. Furthermore, localization and size of the tumor must be reported exactly. There are large crossovers with other tumor entities in appearance (Figs. 1, 2).

Mammography ILCs are often difficult to detect mammographically due to an insidious growth pattern. Therefore, the tumors are often larger at diagnosis than other cancers and may present as palpable masses. ILC is the tumor entity with

Carcinoma, Male Breast

247

C

Carcinoma, Lobular, Invasive. Figure 1 (a) Normal mammography. (b) Mammography follow-up after nine months due to a new palpable mass on the right side. Mammography shows only a “shrinking sign” of the right breast in the follow-up. Histology: invasive lobular carcinoma.

the most diagnostic failures, especially in mammography. Most commonly, the tumor presents as a spiculated mass or mass of asymmetric density without definable margins. Architectural distortion is also possible, whereas calcifications are rare. Sometimes the tumor is seen only in one view. In addition, even when mammography shows the tumor, its extent is often underestimated.

References 1. 2.

3.

Boetes C, Veltman J, van Die L et al (2004) The role of MRI in invasive lobular carcinoma. Breast Cancer Res Treat 2773:1–7 Hilleren DJ, Andersson IT, Lindholm K et al (1991) Invasive lobular carcinoma: mammographic findings in a 10-year experience. Radiology 178:149–154 Newstead GM, Baute PB, Toth HK (1992) Invasive lobular and ductal carcinoma: mammographic findings and stage at diagnosis. Radiology 184:623–627

Ultrasound The tumor is often characterized as a vague area of shadowing without defined borders. But the tumor is often not easily recognized on ultrasound because it grows diffusely.

Carcinoma, Male Breast K ARIN B OCK

Magnetic Resonance Mammography Typical findings in magnetic resonance imaging are rim-enhancing masses that are irregular or spiculated with heterogeneous enhancement. Signal-intensity curves show a high initial contrast media uptake and a postinitial plateau or wash-out. The tumor is presented in watersensitive sequences by an intermediary signal.

University of Marburg, Marburg, Germany [email protected]

Synonyms Breast cancer, male; Breast neoplasm, male; Breast tumors, male; Tumors, male breast

248


Carcinoma, Lobular, Invasive. Figure 2 (a–d): a Mammography in mlo view without pathology (a). At the six-month follow-up, a new mass in the left upper quadrant is seen, (b) which is most visible in the spot view (c). Ultrasound shows a hypoechoic mass with irregular borders (d).

Definitions


Malignant neoplasms of the male breast that account for approximately 0.7% of all breast cancer diagnoses and less than 1% of all malignant neoplasms in males. The mean age at diagnosis for men is 67 years, which is 5 years older than the average age at diagnosis for women. Primary malignant neoplasms with an overall incidence of 0.5–3% must be distinguished from metastases to the breast of extramammary carcinomas (e.g., melanoma, lymphoma, prostate cancer).

Physiologically, the male breast consists of fat and a few strands of fibrous and connective tissue behind a small nipple with rudimentary ducts only. The range of histologic subtypes for male and female breast cancers is similar, but relative distributions differ significantly. Thus, ductal carcinoma in situ compromises approximately 10% of breast cancers in men, while the majority of cases are invasive cancers. Data obtained from more


than 2,000 male patients from the SEER cancer registry demonstrated 93.7% of male breast cancers as ductal or unclassified carcinomas, the others being papillary (2.6%), mucinous (1.8%), or lobular (1.5%). In comparison to female breast cancers, male breast cancers show higher rates of estrogen receptor expression and equal rates of progesterone receptor expression. Moreover, in contrast to women, the c-erbB2 proto-oncogene is less likely to be overexpressed (approximately 5% only).

Clinical Presentation As male breast cancer is still considered a rare disease, there is no screening for it and diagnosis is usually made clinically and in an advanced stage of disease. The most common clinical presenting symptoms in male breast cancer patients are 1. Painless subareolar lumps 2. Nipple retraction 3. Bloody nipple discharge.

Imaging The most appropriate work-up of suspicious breast findings is diagnostic mammography. As for women, the examination should consist of craniocaudal and mediolateral oblique views of each breast (see Fig. 1a, b). A small metal marker can be placed on the skin over the mass to help identify its location on mammographs. In contrast to females, supplementary mammographic views (e.g., spot

249

compression, magnification, exaggerated craniocaudal, cleavage) are rarely needed to clarify lesions. The value of digital mammography has not been evaluated yet. For evaluation of male mammograms, the breast imaging reporting and data system (BI-RADS) can be used successfully (1, 2). The sensitivity, specificity, and overall accuracy of mammography in the diagnosis of male breast cancer have been reported by some investigators to be 92%, 89–90%, and 90%, respectively (3,4). However, although there are characteristic mammographic features that allow breast cancer in men to be recognized, overlap between these features and the mammographic appearance of benign nodular lesions cannot be neglected (5). Malignant breast tumors are more often eccentric, displacing ill-defined, spiculated, or macrolobulated margins (see Figs 1a, b and 2a, b), but even well-defined margins have been described (4–6). On the other hand, benign breast tumors, such as cysts or fibroadenoma, are seldom found, due to normally rudimentary ducts and glandular tissue. Calcifications are fewer, coarser, and less frequently rod-shaped than in female breast cancer (5). Secondary features can include skin thickening as well as nipple retraction and axillary lymphadenopathy. Computer-assisted diagnosis (CAD)—Systems do not play a role in the diagnosis of male breast disease, as mammography in men is only performed when clinical findings are present. Breast ultrasound (7.5–13 MHz, linear arrays) is considered a useful adjunct without radiation hazards,

Carcinoma, Male Breast. Figure 1 (a, b) Two standard view mammographs of a man’s left breast displaying pseudogynecomastia with retroareolar breast cancer.

C

250


Carcinoma, Male Breast. Figure 2 (a, b) Magnification views (compare Figure 1a, b) displaying male breast cancer. Retroareolar irregular hyperdense lesion with convex borders, satellites toward the chest wall.

increasing the diagnostic specificity and providing additional information regarding nodal involvement, and is an image guide of choice for percutaneous procedures. The BI-RADS—Ultrasound atlas (6) improves effective use of ultrasound by defining a lexicon of terms and characterizing typical features in the sonographic assessment of lesions. Sonographic features of malignant tumors of the male breast do not differ substantially from those in female breast cancer (7). The most important features for malignant lesions being irregular shape, architectural distortion, and posterior shadowing (see Fig. 3). Other imaging techniques such as MRI are still investigational in the diagnosis of female breast cancer. Moreover, to date, there are no published data on MRI in male breast cancer. Studies on molecular imaging such as positron emission tomography (PET) or single photon emission computed tomography (SPECT) are of distinct diagnostic value in female breast cancer, but have not been tested for male breast cancer. So far, it does not seem feasible to use any of these techniques in the evaluation of clinically apparent breast findings. Nevertheless, a normal imaging evaluation should never overrule a strongly suspicious finding on physical examination.

Nuclear Medicine Not applicable in male breast disease, except for sentinel node biopsy in cases of proven breast cancer.

Carcinoma, Male Breast. Figure 3 Male breast cancer (compare Figure 1a, b and 2a, b). B-mode ultrasound (10 MHz, Siemens Sonoline Elegra 10 MHz, linear array) displaying an irregular-shaped, infiltrating mass with disturbed sound transmission as the most reliable characteristic features of malignancy.

Diagnosis After appropriate local imaging, a suspicious mass requires biopsy to confirm the diagnosis. As in the diagnosis of breast lumps in women, the method of choice should be a minimally invasive procedure, that is, core needle biopsy (≤14G) (8). Despite limited data, sentinel node biopsy seems feasible in male patients too.

Carcinoma, Multiple, Breast

Bibliography 1. 2.

3.

4.

5. 6. 7.

8.

ACR BI-RADS (2003) Mammography. 4th edn. Reston, VA Bock K, Iwinska-Zelder J, Duda VF et al (2001) Validity of the Breast Imaging—Reporting and Data System (BI-RADS) for clinical mammography in men. Fortschr Ro¨ntgenstr 173:1019–1023 Evans GF, Anthony T, Turnage RH et al (2001) The diagnostic accuracy of mammography in the evaluation of male breast disease. Am J Surg 181:96–100 Partik B, Mallek R, Rudas M et al (2001) Malignant and benign diseases of the breast in 41 male patients: mammography, sonography and pathohistological correlations. Fortschr Ro¨ntgenstr 173:1012–1018 Appelbaum AH, Evans GFF, Levy KR et al (1999) Mammographic appearances of male breast disease. Radiographics 19:559–568 ACR BI-RADS (2003) Ultrasound. 1st edn. Reston, VA ¨ stu¨n EE et al (2002) Male breast Gu¨nhan-Bilgen I, Bozkaya H, U disease: clinical, mammographic, and ultrasonographic features. Eur J Radiol 43:246–255 Janes SE, Lengyel JA, Singh S (2005) Needle core biopsy for the assessment of unilateral breast masses in men. Breast July 15 (Epub ahead of print)

251

Furthermore, the localization and size of the tumor must be reported with precision (Figs. 1 and 2).

Multifocality One or more masses are found in one quadrant, at a maximum of 2 cm away from the main tumor. The therapy of choice is quadrantectomy.

Multicentricity One or more tumors are more than 2 cm away from the main tumor, often in different quadrants. Mastectomy is usually performed. Often a tumor is seen on mammography. Because mutifocal, multicentric, or contralateral tumors have a big influence on the therapy and prognosis, we perform MR mammography preoperatively.

Mammography

Carcinoma, Multiple, Breast S ILVIA O BENAUER Department of Radiology, Georg-August-University of Goettingen, Goettingen, Germany [email protected]

A spiculated mass with irregular margins is a typical sign of invasive ductal carcinomas. The density of the tumor is often higher than that of the parenchyma. In some cases the tumors also present with amorphous or pleomorphic microcalcifications. Other imaging features may be asymmetry or architectural distortion, and therefore changes from a previous mammogram must be interpreted carefully.

Ultrasound

Definition An invasive carcinoma is a tumor with extension of tumor cells through the ductal basement membrane. About 65% of carcinomas are invasive ductal carcinomas. Multifocal tumors must be distinguished from multicentric tumors.

The tumor is often characterized by an irregular, hypoechoic mass, typically more tall than wide with a thick echogenic rim and posterior acoustic enhancement. The tumor is not comprisable.

MR mammography

The patient could present with many hard, palpable masses and/or skin, and/or nipple retraction.

Typical findings on magnetic resonance mammography (MRM) are rim-enhancing masses that are irregular or spiculated with heterogeneous enhancement. Signal intensity curves show a high initial contrast media uptake and a postinitial plateau or wash-out. On water-sensitive sequences the tumor has an intermediary signal. MRM reveals additional tumors not previously seen on mammography or ultrasound. Therefore, MRM could be performed before therapy to exclude ▶multifocality or ▶multicentricity. Studies have found that MRM reveals about 14% additional findings that were not previously seen.

Imaging

Bibliography

All descriptions of the imaging of invasive tumors should be made according to the BI-RADS classification.

2.

Pathology The tumors are classified as already described in the chapter on ductal invasive carcinomas.


1.

Fischer U (2000) Lehratlas der MR-Mammographie. Thieme Verlag, Stuttgart-New York Heywang SH, Wolf A, Pruss E et al (1989) MR imaging of the breast with Gd-DTPA: use and limitations. Radiology 171:95–103

C

252

Carcinoma, Multiple, Breast

Carcinoma, Multiple, Breast. Figure 1 Mammography of a patient with a palpable mass on the left side. Typical spiculated mass on the left side in mediolateral oblique and craniocaudal views (a). Ultrasound revealing a second mass on the left side (b) which was confirmed by MRM on a subtraction image and maximum intensity projection (c). Histology determined bifocal invasive ductal carcinoma.

Carcinoma, Other, Invasive, Breast

253

C

Carcinoma, Multiple, Breast. Figure 2 Mammography of a patient with a palpable mass in the left breast (caudal quadrant). Additional suspicious lesion in the middle of the breast (a). MRM of a multicentric carcinoma in the left breast, on subtraction image (b) and on maximum intensity projection (c).

Carcinoma, Other, Invasive, Breast S ILVIA O BENAUER Department of Radiology, Georg-August-University of Goettingen, Goettingen, Germany [email protected]

Synonyms Circumscribed carcinoma; Colloid carcinoma; Gallert carcinoma; Gelatinous carcinoma; Infiltrating papillary

carcinoma; Inflammatory carcinoma; Intracystic papillary carcinoma with invasion; Medullary carcinoma; Mucinous carcinoma; Papillary carcinoma; Tubular carcinoma

Definition Invasive carcinomas comprise medullary carcinoma, mucinous carcinoma, tubular carcinoma, papillary carcinoma, and ▶inflammatory carcinoma. Papillary carcinoma is a rare form of specified invasive ductal carcinoma (about 1–2% of all breast cancers). Medullary and mucinous carcinomas are also rare tumors ( 0.7; 92% sensitivity, 88% specificity) or a difference of RI between both kidneys larger than 0.04 (1, 3). Falsenegative RI results in the first hours after onset may be explained by initial vasoregulation and spontaneous pelvocaliceal rupture (4) (Fig. 3). A characteristic slow or absent ureteral jet in the bladder, and the twinkling artifact behind stones help in the diagnosis of stone obstruction. ▶Retrograde pyelography refers to direct ureteral contrast injection through a catheter inserted at cystoscopy. It is the most precise imaging method for diagnosing any type of ureteral pathology, but is rarely performed; mostly when obtaining material for cytology, and prior to ureterorenoscopy (URS) or the placement of double-J stents. When the insertion of double-J stents fails, therapeutically inserted percutaneous nephrostomes (PNS) can also be used for anterograde pyelography (1, 2) (Fig. 2).

Colic, Acute, Renal. Figure 1 IVU examination in a 55-year-old man with acute renal colic demonstrates (a) slightly delayed excretion on the left 5 min after 30 mL nonionic contrast injection, (b) extravasation around the nonopacified pyelon and proximal ureter after spontaneous caliceal rupture on tomography, (c) in prone position pyelon and ureter opacify 10 min postcontrast, (d) visible level of obstruction at the UVJ due to a small left ureteral calculus (arrow).

C

354

Colic, Acute, Renal

the collecting system are generated from unenhanced and enhanced CT data (Fig. 4a). MR urography is regarded as a diagnostic test of secondary reference following IVU, US, and CT (6). Two different techniques of MRI, depending on the clinical situation, are available. Static-fluid MR urograms are obtained with heavily T2-weighted turbo spin-echo sequences. Dynamic or excretory MR urograms are obtained with gadolinium contrast agent administration and T1-weighted gradient-echo sequences (Fig. 4b, c).

Diagnosis

Colic, Acute, Renal. Figure 2 (a) Small radio-opaque right-sided urinary stones detected at KUB (arrows) in a female with acute renal colic after ESWL. Percutaneous nephrostoma was inserted after unsuccessful double-J insertion. On anterograde urography these radiopaque and severely obstructing stones were demonstrated as less radiopaque than contrast. (b) Retrograde urography in 64-year old man with acute renal colic and suspected radiolucent left ureteral calculi. An atypical eccentric filling defect is delineated (arrow). IVU was inconclusive, but US confirmed a ureteral stone.

UHCT has proved to be more accurate than IVU for the diagnosis of urinary calculi and equal in demonstrating dilatation (5). It is indicated in patients with a high risk for use of contrast media, in nonfunctioning kidneys, or when urography is inconclusive. Other advantages of CT over IVU are the demonstration of other abdominal pathology. Thin section (3.5 or 5 mm) UHCT has a sensitivity of 97%, and specificity of 96% for the investigation of renal colic (1). Secondary signs typical for the acute phase are periureteral or perirenal tissue stranding and the tissue rim sign due to edema in the ureter wall around small calculi. A split bolus technique of contrast administration and scanning during the nephrographic and excretory phase provides evaluation of renal function and parenchymal alterations, and delineates the collecting system. The advent of multislice-computed tomography (MSCT) provides the ability to perform CT urography with a spatial resolution closer to that obtained at IVU. High-resolution multiplanar (MPR), maximumintensity (MIP), and average-intensity (AIP) images of

The diagnosis of an obstructive urinary stone is essential for the exclusion of acute surgical abdominal pathology (appendicitis or abdominal aortic aneurysm). Urinary tract infection, cardiac ischaemia, bowel ischaemia or obstruction, hepatic capsulitis, musculo-skeletal pain, and biliary colic, are also considered in the differential diagnosis. In patients with acute flank pain, the diagnosis of ureteral calculi can be apparent according to positive history, physical examination, and laboratory studies. In presence of hematuria, calculi are confirmed with 96% accuracy by a positive KUB. Radiological imaging is required to evaluate the location of the stone, the exact size, shape, orientation, radiolucency, all necessary for treatment planning. Kidney function and the presence of infection are assessed. Until recently, IVU was performed to evaluate stone obstruction in inconclusive cases, with a sensitivity of 95%. Recently, abdominal US supersedes IVU in patients with positive KUB. It is the test of choice if KUB is negative, since it can effectively demonstrate radiolucent stones, or exclude other causes of abdominal pain. The reported sensitivity of US for stone detection varies, the highest is 98%. US easily demonstrates larger stones within the renal pelvis and stones in the dilated ureter near the UPJ and ureterovesical junction (UVJ). In the absence of dilatation (35%), duplex Doppler US can assess an elevated RI in the affected kidney, and asymmetry of ureteral jets, thus helping to exclude falsenegative cases. It reduces the number of false-positive results by assessing nonobstructive dilatation in cases of extrarenal pelvis, UPJ obstruction, prominent vasculature, residual dilatation, vesicoureteral reflux, megacalices, papillary necrosis, pyelonephritis, full bladder, and diabetes insipidus (2). UHCT of the abdomen may be helpful as an alternative or in the clarification of equivocal US findings. UHCT of the abdomen is more sensitive and more likely to yield a diagnosis than plain radiographs and US. Management of urinary stones (1) comprises the insertion of a double-J stent in cases of uncontrollable

Colic, Acute, Renal

355

C

Colic, Acute, Renal. Figure 3 US examination in acute renal obstruction showing (a) ureteral dilatation due to calculus (arrow) 6.8 cm from the bladder (B), and (b, c) increased RI in the affected kidney with a significant RI difference of 0.07 between both kidneys. (Contributed by A. Visnar-Perovic)

Colic, Acute, Renal. Figure 4 (a) CT urography in 40-year old patient presenting with right-sided flank pain and gross hematuria was performed in a pain-free interval. The MIP image shows a calculus (arrow) within the right ureter with sufficient contrast material passage and only slight dilatation of the pelvicaliceal system. In addition, a small infundibular concrement can be appreciated on the left side. (Contributed by J. Kemper) (b, c) MRU case showing a calculus inside the left renal pelvis obstructing the UPJ in a 64-year old man with recurrent left-sided flank pain. (b) MIP from a gadolinium-enhanced T1-weighted breath-hold 3D-gradient-echo-sequence displaying the urographic overview and (c) axial standard T2-weighted TSE-image showing the signal void of the calculus (arrows). (Contributed by C. Nolte-Ernsting)

pain, complete obstruction with severe urinary infection or urosepsis, in solitary obstructed kidneys, in large stones that are considered unlikely to pass spontaneously, or in suspected ureteral strictures. Percutaneous nephrostomes

are inserted when the insertion of double-J fails. Ten to twenty percent of all urinary stones require surgical removal. Percutaneous nephrostolithotomy (PNL), URS, and ESWL successfully compete to open surgery. Most

356

Colitis, Ulcerative

urinary stones pass spontaneously in 24–48 h and do not require special treatment other than pain relief, hydration and antiemetics.

Bibliography 1.

2. 3.

4.

5.

6.

Menon M and Resnic MI (2002) Urinary lithiasis: etiology, diagnosis, and medical management. In: Campbell MF, Walsh PC, and Retik AB (eds) Campbell’s Urology. Urinary Lithiasis and Endourology. WB Saunders Company, Philadelphia, pp 3229–3305 Platt JF (1996) Urinary obstruction. Radiol Clin North Am 34 (6):1113–1129 Platt JF, Rubin JM, Ellis JH, et al (1989) Duplex Doppler US of the kidney: differentiation of obstructive from nonobstructive dilatation. Radiology 171:515–517 Roy C, Tuchmann C, Pfleger D, et al (1998) Potential role of duplex Doppler sonography in acute renal colic. J Clin Ultrasound 26:427–432 Smith RC, Verga M, McCarthy S, et al (1996) Diagnosis of acute flank pain: value of unenhanced helical CT. Am J Roentgenol 166:97–101 Nolte-Ernsting CC, Staatz G, Tacke J, et al (2003) MR urography today. Abdom Imaging 28:191–209

etiology and a chronic relapsing–remitting course. To date, the pathogenesis of UC still remains unknown, although recent studies showed an important role of environmental enteric, immune, and genetic factors. These studies suggest that UC may result from the loss of tolerance of the immunocompetent intestinal system versus the normal intestinal flora, in genetically susceptible individuals. UC is characterized by a predictable course and localization, extending continuously from the rectum to the colon, involving first the left colon, then the transverse and right colon, and very rarely the distal ileum. The different patterns of ulcerative colitis are commonly called as “ulcerative proctitis,” “ulcerative sigmoiditis,” “leftsided colitis,” or “pancolitis.” A recto-sigmoid localization is present in up to 95% of the patients. The mucosal inflammation of the terminal ileum, is rarely observed in UC, developing only in presence of a pancolitis (the so-called “▶backwash ileitis”); in those cases, UC can be hardly differentiated from Crohn’s disease (CD).


Colitis, Ulcerative F RANCESCA M ACCIONI 1 , S ARA PARLANTI 1 , F RANCESCA S ILIQUINI 1 , V IOLA G ARBARINO 1 , VALERIA D’O VIDIO 2 1

Department of Radiological Sciences, Policlinico Umberto I, University of Rome “La Sapienza”, Italy 2 Department of Clinical Sciences, Gastroenteroly Unit, Policlinico Umberto I, University of Rome “La Sapienza”, Italy [email protected]

Synonyms Ulcerative colitis; Ulcerative proctitis; Ulcerative sigmoiditis; Ulcerative pancolitis

Definitions Ulcerative colitis (UC) was described in 1859 by Sir Samuel Wilks (1824–1911), a British physician who firstly differentiated it from bacterial dysentery. The greatest incidence of chronic UC is in the second, third and fourth decades of life. The condition is more common in white women. The disease may vary greatly in its severity, course, and prognosis. It has an unknown

Cryptitis or abscesses of the crypts of Lieberku¨hn are hallmarks of ulcerative colitis. They form ulcerations that reach the lamina propria, or may produce excrescences also called as “pseudopolyps.” With the lateral extension of the ulcers, large areas of bowel wall muscle may be almost completely denuded. Microscopically, UC is characterized by moderate wall thickness, leukocyte infiltrates in the mucosal and submucosal layer, disruption of mucosal elements and aphthoid ulcers, mucosal edema, inflammatory pseudopolyps, without extra-wall lesions neither significant signs of perivisceral inflammations in the majority of cases. Fatty deposition in the submucosal layer is a common finding in long-standing disease. Macroscopically, the chronic inflammatory bowel process of UC determines loss of haustration, moderate strictures, mild diffuse colonic wall thickening, widening of the presacral space, and occasionally severe complications such as massive bleeding, toxic megacolon, tight bowel strictures, or perforation. The disease is intermittently acute and in the quiescent phase the mucosa may completely heal, but more frequently it appears atrophic with rare crypts, with distorted mucosal architecture and thickening of the lamina propria.

Clinical Presentation Three degrees of severity are distinguished as mild (about 60–80% of cases), moderate (25%), and severe or fulminant.

Colitis, Ulcerative

Eighty percent of patients have only proctitis or proctosigmoiditis in the early phases of the disease, although in 50% of them a proximal extension later occurs. Only 20% have extensive colitis at the onset of symptoms, the course of the disease can vary widely. Spontaneous remission from a flare-up occurs in 20% to 50% of the patients, although 50% to 70% have a relapse during the first year after diagnosis. The relapse rate is higher in younger patients. In acute phases, bleeding results from friable and hypervascular granulation tissue; diarrhea with urge incontinence results from damage that impairs the ability of the mucosa in reabsorbing water and sodium. If the disease is more severe, it may extend beyond the mucosa and submucosa into the muscularis mucosa (rarely to serosa) and this explains the dilation of the colon, by loss of motor tone, in cases of toxic megacolon. Severe disease is indicated by large volumes of diarrhea, weight loss, large amount of blood in the stool, high fever, elevated C-reactive protein, elevated erythrocyte sedimentation rate, low hematocrit value, and hypoalbuminemia. Approximately 65% of patients with UC have positive perinuclear antineutrophil cytoplasm antibodies (pANCA), also present in patients with primary sclerosing cholangitis. The prevalence of the extraintestinal manifestations such as arthritis, uveitis, pyoderma gangrenous, sacroilitis, spondylitis, or erythema nodosum, may vary depending on the geographic area, population, location and duration of the disease, medication, and diagnostic accuracy. Patients with ulcerative colitis have an increased risk of developing colorectal cancer. The current procedure to diminish this risk is colonoscopy surveillance and histopathological evaluation of biopsy specimens. This method is not unquestioned and is undergoing continuous evaluation.

Imaging

357

C

Colitis, Ulcerative. Figure 1 Endoscopic view of the colonic mucosa, as appears in severe ulcerative, very friable, with spontaneous bleeding. Furthermore, loss of vascularity, hemorrhage, and ulcers with fibrin and mucous are well appreciable.

strictures. These changes are continuous in the colon, and the rectum is always involved (Fig. 1). However, although ES is widely considered the first examination for the diagnosis UC, it can be contraindicated, refused or incomplete in up to 30% of patients with UC. It is an invasive examination causing patient discomfort and associated with possible complications, above all colonic perforation, whose incidence may vary according to the operator’s experience, ranging between 0.1% and 3%. Moreover, ES is absolutely contraindicated in the acute phases of severe colitis, due to the higher risk of perforation; in these phases, however, a diagnostic support is crucial to plane an effective treatment. Finally, late complications of UC, particularly tight strictures, can partially or completely prevent the passage of an endoscope. In all these cases other imaging modalities, including barium studies and ▶cross-sectional imaging, may add important information on the disease (Fig. 2).

Endoscopy Endoscopy (ES) is definitely a primary examination in the evaluation of UC. The disease is, in fact, usually confined to the colonic mucosa, which is completely accessible to endoscopy. Moreover, endoscopic biopsies, although confined to the mucosa and submucosa layer, may adequately evaluate the severity of colonic wall inflammation, which usually spares the outer muscular and serosa layers. Typical endoscopic findings include reddening of the mucosa, increased vulnerability, mucosal bleeding, irregular ulcers, pseudopolyps, granularity, loss of vascular architecture, loss of haustration, and occasionally

Conventional Radiographs Barium studies may be an alternative modality to endoscopy to assess luminal changes, particularly when performed with a double-contrast technique. Single- or double-contrast barium enema (BE) may easily visualized the typical mucosal changes of the disease. In the first phase, BE may show superficial changes related to edema and granulation tissue; occasionally the only sign of inflammation in the early phases is the blunting of the normally acute angles of the rectal valves.

358

Colitis, Ulcerative

the left colon and the 10 cm in diameter in the right colon. Colonic perforation and free peritoneal air can be easily identified at plain films in upright position as well.

Cross-Sectional Imaging: Ultrasonography, Computed tomography, and Magnetic Resonance Imaging

Colitis, Ulcerative. Figure 2 Barium studies, performed with a double contrast technique, shows lack of haustrations and tubular narrowing of the left-side colon, characterized by multiple pseudopolyps and ulcers showing a “collar-button” appearance. (Courtesy of Panzironi G, Department of Radiology, Policlinico Umberto I. Rome, Italy.)

When the disease progresses, using a double-contrast BE, ulcers seen in profile have a “collar-button” appearance. The demonstration of ulcerations by radiographic means is important because these changes indicate clinically and pathologically severe diseases. Between denudated and ulcerated areas, a large number of pseudopolyps may be observed, representing elevation of inflamed mucosa. Secondary changes can be easily seen on both single- and double-contrast BE examinations. Main signs of chronic disease are foreshortening of the colon, lack of haustrations, and tubular narrowing of the colon that gives the large bowel the appearance of a garden hose or stovepipe. Barium studies, although not useful in assessing the clinical activity of UC, can be indispensable in distinguishing between the two diseases (Goldberg et al. 1979; Gore et al. 1997). In very severe hyper acute phases, a BE may be contraindicated. In such phases a plain film of the abdomen is the best option, being crucial in the identification of the toxic megacolon. This complication is usually well identified at plain films as a severe colonic distention exceeding the 8 cm in diameter in

Cross-sectional studies may play a complementary role in the assessment of UC. Transabdominal ultrasonography (TUS) is a noninvasive cross-sectional imaging modality that can be helpful in the diagnosis of ▶inflammatory bowel disease (IBD). The main purpose of US is to exclude the involvement of the distal ileum, in order to distinguish UC from CD. High-resolution US, performed with a 7.5 MHz probe, is a well accepted modality to evaluate the wall thickening of the distal ileum in Crohn’s disease, therefore it may be extremely useful to differentiate the two diseases, whenever ES or histology are doubtful. Moreover, in more severe colitis, the mild wall thickening, particularly the left and sigmoid colon, if equal or superior to 4 mm, can be well-depicted at high-resolution US. Finally US can be useful in the investigation of biliary complications of the disease, including biliary stones, cholangitis, and cholangiocarcinoma. Usually cross-sectional modalities cannot assess the typical mucosal changes of UC, but they (CT and MRI above all) can evaluate the mild wall thickening, the increased wall vascularity associated with an active disease, as well the changes in the morphology of the colon. In some cases, cross-sectional imaging may substitute ES and BS, particularly when they are contraindicated in very severe phases. Thanks to their panoramic and multiplanar capability, both CT and MRI are usually able to distinguish between a proctitis, a left-sided colitis or a pancolitis, according to the findings observed either on axial and coronal planes. MRI Coronal or CT multiplanar reconstructions are very useful to exclude involvement of the terminal ileum, thus helping in differentiating UC from CD, similarly to US. In the evaluation of rectal disease, the mid-sagittal imaging plane offered by MRI better displays the typical widening of the recto-sacral space. Recently, MRI has been proposed to evaluate the degree of wall inflammation (disease activity) especially in the follow up of patients (thanks to lack of radiations) (Fig. 3). At MR, wall gadolinium enhancement is frequently observed in severe active UC associated marked wall thickening; on the other hand, in moderate disease a moderate wall enhancement can still be present, but

Colitis, Ulcerative

359

or 99mTc-labeled antigranulocyte antibodies. The antibody technique offers the advantage of in vivo labeling, but is less reliable than the exametazime method for imaging of colonic inflammation.

Treatment

Colitis, Ulcerative. Figure 3 Colitis, Ulcerative. The T2-weighted MR image, acquired on the axial plane, shows a mild case of ulcerative colitis, localized on rectum, demonstrating diffuse thickening and high wall T2 signal of the rectal wall, which is surrounded by abundant fat tissue proliferation, both signs of chronic inflammatory disease.

associated with lower thickening; finally, inquiescent disease, wall enhancement can be very low or absent. CT and MRI can detect complications of UC requiring surgery, such as diffuse dilatation of the colonic lumen (toxic megacolon), tight strictures, and rectal cancer. The toxic megacolon can be diagnosed with the same criteria of conventional abdominal plain films: a marked and diffuse colonic dilation (upper normal limit 5.5–6.5 cm in the transverse and left colon) with severe mucosal disease is the typical appearance.

Nuclear Medicine Radionuclide studies have a useful role in acute fulminant colitis when colonoscopy or BE study is contraindicated, similarly to CT or MRI. Radionuclide studies are also useful in depicting disease activity and the extent of disease and in monitoring the response to therapy. Promising results have been published about the clinical use of technetium 99m white blood cells (WBC) in the assessment of IBD in adults and in a small series of children. In patients with active UC, inflammation is visualized with 99mTc-exametazime-labeled leucocytes

The treatment of patients with IBD depends on knowledge of the location, extent, and activity of the inflammation. Clinical evaluation (Truelove and Witts classification or the more recent Powell–Tuck index) and ES are usually sufficient to assess the extent and severity of UC, since the inflammatory process involves the colon only, sparing the small bowel. Moreover, biopsy specimens obtained during ES, necessarily limited to the mucosa and submucosa layer, can usually detect most of the pathologic features of UC, since the inflammatory process does not extend beyond the submucosa layer. Clinical data (acute phase reactants, etc.) together with endoscopic evaluation are usually adequate to estimate the disease extent in most of patients with mild to moderate disease. In case of complicated or very severe disease, if endoscopy is limited or contraindicated, or in case doubtful cases, the diagnostic modalities above mentioned may complete the diagnostic procedure, helping to reach a final correct diagnosis. Generally, most patients with mild to moderate disease are effectively treated with drugs. Occasionally, however, the disease may be extremely severe, thus requiring urgent colectomy if it does not respond to pharmacological therapy promptly. The drugs of first choice for the treatment of an acute flare-up of UC are 5-ASA-releasing preparations and glucocorticoids. The efficacy of glucocorticoids and the beneficial effects of sulfasalazine have been known for half a century. The 5-ASA influences a wide variety of immunologic and inflammatory reactions, such as a chemotaxis of white blood cells, cytokine release, and release of reactive oxygen species. The extent of the disease influences the strategy of treatment. Left-sided colitis is better treated with rectal administration of drugs, whereas more extensive colitis requires oral or intravenous treatment, depending on the disease severity. Active distal UC should initially be treated with 5ASA. If treatment is otherwise ineffective, steroid enemas and preferably steroid foams, can be used. If the colitis involves the left colic flexure, rectal treatment is not sufficient. In patient with mild or moderate disease oral administration of 5-ASA preparation is effective.

C

360

Collateral Phenomenon

In more severe cases, or if 5-ASA fails, glucocorticoids should be used orally or intravenously and 60–100 mg of prednisolone equivalent is useful in the majority of patients. Among those with very severe colitis, however, only half of the patients respond. When adequate doses of glucocorticoids fail to improve severe UC, a colectomy should be performed, particularly in patients with fulminant colitis and toxic megacolon. Azathioprine or 6-mercaptopurine is effective in chronic active colitis. However, considering the possible side effects and the need for long term therapy this option should be weighted against colectomy and ileoanal pouch surgery. Maintenance of remission with sulfasalazine should be used and decreases the relapse rate by about 50%. Azathioprine should be used only in patients who cannot be kept in remission with 5-ASA and who are not willing or able to undergo colectomy. The most frequent indication for surgical resection in UC is severe inflammation of the large intestine and rectum that is refractory to conservative therapy, also in patients who suffer from the side effects of the medication. Toxic colitis with or without megacolon can be complicated by perforation. Operation within 3.5 days is prognostically associated with a better outcome than operation after a long, futile treatment attempt with considerable worsening of the patient’s general conditions. Three classic surgical interventions are performed in UC: conventional proctocolectomy with terminal ileostomy (without preservation of the anal sphincter), ileorectal anastomosis (only if the rectum is not extensively involved), or restorative proctocolectomy with ileo-anal pouch, which is nowadays considered the first choice procedure. In the last procedure (total colectomy with ileo-anal pouch), the rectum is completely resected whereas the anal sphincter muscles apparatus is spared; therefore there is no risk of recurrence, and at the same time the ileo-anal pouch ensures a reservoir function. Sometimes the ileal pouch may undergo to chronic wall inflammation, and the so-called “pouchitis” endoscopy is the modality of choice to evaluate the degree of wall inflammation in a pouch. Cross-sectional imaging, particularly MRI or CT, may be useful in evaluating the morphology of the pouch and possible complications.

3.

4.

5.

6. 7.

Gore RM, Ghaharemani GG, Miller FH et al (1997) Inflammatory Bowel Disease: Radiologic Diagnosis. In Syllabus of the radiological Society of North America, Categorical Course in Gastrointestinal radiology: 95–109, RSNA Publications Miner PB (2000) Clinical Features, course, laboratory findings, and complications in ulcerative colitis. In: Kirsner JB (ed) Inflammatory Bowel Disease. 5th edn. WB Saunders, Philadelphia, pp 299–304 Maccioni F (2004) In: Chapmann AH (ed) MRI of Colitis in Radiology and Imaging of the Colon. Springer-Verlag, Heidelberg, pp 201–214 Maccioni F, Colaiacomo MC, Parlanti S (2005) Ulcerative colitis: value of MR imaging. Abdom Imaging 30:online (May) Scholmerich J, Herfarth C In: Cremer M et al (eds) (1999) Ulcerative colitis in Gastroenterology and Hepatology. McGraw-Hill, New York, pp 382–392

Collateral Phenomenon Collateral phenomenon is an nonspecific finding of periarticular demineralization, which occurs, for example, due to disuse or neighborhood inflammation. Today the terms “regional osteoporosis” and “periarticular demineralization” are more common. ▶Rheumatoid Arthritis

Collimator In a radionuclide imaging device, a collimator is a block of radiation-attenuating material with one or more apertures defining the field of view and limiting the angular spread of the radiation that can reach the radiation detector assembly. ▶Scintigraphy

Colloid Carcinoma ▶Carcinoma, Other, Invasive, Breast

Bibliography 1.

2.

Goldberg HI, Caruthers SB Jr, Nelson JA et al (1979) Radiographic findings of the National Cooperative Crohn’s Disease Study. Vol 77: 925–933, Gastroenterology Gore RM (1995) Characteristic morphologic changes in chronic ulcerative colitis: Abdom Imaging. Vol 20:275–278

Coloboma ▶Congenital Malformations, Orbit

Colon, Postoperative

Colon, Postoperative A SHLEY S. S HAW 1 , A LAN H. F REEMAN 2 1

Consultant Radiologist Cambridge University Hospitals NHS Trust Cambridge, UK 2 Consultant Radiologist Cambridge University Hospitals NHS Trust Cambridge, UK

Synonyms Abdominal Surgery; Complications; Abdomen; Postoperative

Colonic resection may be performed for a number of conditions, most commonly malignancy, diverticular disease, inflammatory bowel disease (ulcerative colitis, Crohn’s disease) or ischaemic bowel. Equally, a wide variety of procedures may be performed, from short segmental resection to complete pan-proctocolectomy. When imaging patients who have undergone colonic surgery, it is important for the radiologist to establish what the underlying condition is, when and what procedures have been performed, and whether there has been any adjuvant chemo- and/or radiotherapy (1). It is only with this information to hand that one is able to identify the most likely pathology, perform the most appropriate study and interpret the images.

Paralytic Ileus Often referred to simply as ‘ileus’ (alternatively adynamic or non-obstructive ileus), paralytic ileus is atony of the intestine resulting in failure to propel the bowel contents distally. Ileus is common after any operation where the stomach, small or large bowel have been handled, although the aetiology is complex. It may be exacerbated by biochemical abnormalities (e.g. renal failure, diabetes mellitus, hypokalaemia), anti-cholinergic drugs, reflex sympathetic inhibition (e.g. retroperitoneal haematoma), sepsis or other systemic illness.

Clinical Presentation Ileus is considered normal for the first few days following colonic surgery. The patient will have little appetite, absent

361

bowel sounds on auscultation and will not pass any flatus or faeces. If this persists for greater than 4–5 days, particularly in the absence of any of the correctable causes detailed above, it may herald the presence of intraabdominal sepsis or haemorrhage and require investigation.

Imaging The key differential diagnosis is large bowel obstruction. The supine abdominal radiograph typically demonstrates distension of both small and large bowel (±gastric distension). Ileus, unlike obstruction, often demonstrates multiple loops of dilated bowel with normal calibre bowel in between. In addition, the presence of a gas filled dilated rectum is strongly suggestive of ileus. Serial films often show the degree of bowel dilatation to reduce over time. Oral contrast medium will pass slowly through the dilated bowel, being increasingly diluted by small bowel fluid and is usually unhelpful. Rectal contrast medium may be used to confirm the absence of a colonic stricture (free passage of contrast into the most distal dilated bowel loop confirms ileus). Where ileus persists, cross-sectional imaging with computed tomography (CT) may be helpful to determine whether there is a surgically remediable cause (e.g. abscess or haematoma).

Anastomotic Leak Following the resection of a segment of colon, the two free ends are usually joined together using sutures or staples. The exceptions to this are Hartmann’s procedure, complete pan-proctocolectomy and abdomino-perineal resection, where an ileostomy or colostomy is fashioned. Failure of the anastamosis occurs in a small proportion of patients, but is significantly more common when there is active inflammation or infection at the time of surgery. Segmental ischaemia or radiotherapy also contribute to friable tissue. For this reason, many surgeons may either electively perform a loop ileostomy to temporarily rest or ‘defunction’, the colon or perform the operation in two stages.

Clinical Presentation Breakdown of the surgical anastomosis usually occurs towards the end of the first post-operative week. The patient commonly presents with a fever and/or pelvic pain, indicating the presence of an underlying abscess. Preliminary investigations may demonstrate an elevation of the white blood cell count and inflammatory markers.

C

362


The presence of a pelvic abscess may lead to an ileus, development of a fistula, peritonitis or mechanical obstruction.

Imaging A supine abdominal radiograph is often the first investigation but is non-specific. It may show dilated loops of bowel from either ileus or obstruction. A gastrografin enema will delineate any ongoing leaks at the anastomosis (Fig. 1) and perhaps identify a fistula. It should be borne in mind that a proportion of leaks will seal spontaneously, but may leave the patient with a pelvic abscess. To assess the pelvis for a collection, crosssectional imaging with CT is indicated (Fig. 2). CT may be performed following administration of rectal contrast in order to help define the often complex post-operative anatomy. Whilst CT will demonstrate the anatomy, it may be difficult to differentiate haematoma from abscess, particularly when there is often a combination of the two. Where there is clinical concern, percutaneous drainage of any fluid collections is often required, both to confirm the diagnosis and treat the abscess. Failure of radiological management or the development of peritonitis will usually necessitate a repeat laparotomy.

Colon, Postoperative. Figure 1 Contrast enema demonstrating an anastomotic leak in a patient shortly after undergoing an anterior resection.

Nuclear Medicine Occasionally, it may not be possible to clearly demonstrate ongoing infection within the pelvis on crosssectional imaging, particularly where there is no discrete abscess. In these cases, particularly where there is delayed presentation, a radiolabelled white cell scan may be of use in proving the diagnosis before committing the patient to a prolonged course of antibiotics or further surgery.

Tumour Recurrence Following diagnosis of colorectal carcinoma, approximately 70% of patients will undergo ‘curative’ surgery. The reported rates of tumour recurrence vary considerably, with the liver the most common site of metastatic disease (2). Local recurrence, at or near the surgical anastomosis or within the surgical bed, is seen in approximately 10–15% of cases and is more common in rectal than colonic tumours.

Colon, Postoperative. Figure 2 CT demonstrates a cavity in the pre-sacral space containing air and fluid, indicating the presence of an abscess. The patient had recently undergone an anterior resection.

anastomosis may lead to rectal bleeding or even mechanical obstruction. However, more often there are no local symptoms and it is identified on routine post-operative surveillance, either at CT or endoscopy.


Imaging

The majority of recurrences occur within the first 2 years following surgery. Tumour recurrence at the surgical

CT is the mainstay of post-operative surveillance following resection of colorectal cancer, usually performed at


363

C

Colon, Postoperative. Figure 3 Patient with previous right hemicolectomy for Dukes B adenocarcinoma. CT does not show any abnormality, but the PET study demonstrates a focal area of recurrence at the anastomosis, confirmed at surgery.

6–12 monthly intervals for 2–3 years after surgery. This will reliably detect liver, peritoneal and nodal metastases. Recurrence at the anastomotic site will be seen as a nodule or plaque of abnormal soft tissue. The difficulty lies in differentiating such areas from post-operative fibrosis or haematoma. Previously, one could either choose to ‘watch and wait’ or attempt a biopsy of such lesions. This has been superseded in many centres by the use of functional imaging (Fig. 3).

Nuclear Medicine Positron emission tomography (PET) uses radiolabelled isotopes (principally 18fluoro deoxyglucose-FDG) to identify areas of abnormally increased metabolism within the body. This complements the anatomical imaging offered by CT, enabling differentiation of abnormal soft tissue, and often identifies hitherto unsuspected disease. Sites of recurrent colorectal cancer are detected with a sensitivity of 97% and specificity of 76%, resulting management changes in between one quarter and one third of patients (3). False positive results may occur at sites where there is active inflammation or infection, thus PET may be more difficult to interpret in the early postoperative period.

Bowel Obstruction Obstruction following colonic surgery has a number of possible causes: recurrence of disease (tumour, Crohn’s disease); adhesions; and fibrous strictures at the surgical anastomosis.

Clinical Presentation Increasing abdominal distension, abdominal pain and the absence of flatus or bowel motions are the typical presenting symptoms of bowel obstruction, with ‘tinkling’ bowel sounds on auscultation. Bowel obstruction following resection is typically a late complication and occurs at the anastamosis site with an incidence of 2–8% (1). The use of staples, rather than hand sewing the anastamosis, appears to increase the frequency of stricturing.

Imaging Anastomotic strictures may be evaluated by a number of methods, including double-contrast barium enema (DCBE), colonoscopy and CT. DCBE will allow evaluation of the mucosal surface and identify any fistulae, whereas CT permits evaluation of the extracolonic structures. Recent advances in CT colonography increasingly allow evaluation of the mucosal surface, but in cases where there is suspicion of tumour recurrence, evaluation with FDG-PET or direct visualisation with endoscopy will be required.

Crohn’s Disease Crohn’s disease may affect any part of the gastrointestinal tract, most commonly the terminal ileum and not infrequently the colon. A significant proportion of patients with Crohn’s disease will require abdominal surgery at some point during the course of their disease, either to resect a fistula or a symptomatic intestinal stricture.

364

Colonic Atresia

Clinical Presentation Following surgery for colonic Crohn’s disease, approximately one-third will relapse within the next 10 years. Most typically, this involves the neo-terminal ileum leading up to the anastomosis following an ileo-colic resection. However, recurrence may occur elsewhere. The patient will often present with abdominal pain or bleeding per rectum and has often lost weight. Fistulae may extend to the skin surface and result in severe perianal pain.

Color Doppler Ultrasound This uses standard ultrasound methods to produce a picture of a blood vessel. In addition, a computer converts the Doppler sounds into colors that are overlaid on the image of the blood vessel and that represent the speed and direction of blood flow through the vessel. ▶Breast, Therapy Effects ▶Lymphadenopathy

Imaging

Colorectal Adenomas The plain abdominal film may demonstrate mechanical obstruction, but a barium enema is the investigation of choice to evaluate the colonic mucosa. Cross-sectional imaging with CT or MRI will enable imaging of any associated abscesses and fistulae. Imaging perianal and rectal fistulae are now performed using MRI. Suppression of the signal from the pelvic fat allows the demonstration of any abnormal fluid tracks and collections, together with their relationship to the sphincters.

Bibliography 1. 2.

3.

Scardapane A, Brindicci D, Fracella MR et al (2005) Post colon surgery complications: imaging findings. Eur J Radiol 53:397–409 Faria SC, Tamm EP, Varavithya V et al (2005) Systematic approach to the analysis of cross-sectional imaging for surveillance of recurrent colorectal cancer. Eur J Radiol 53:387–396 Huebner RH, Park KC, Shepherd JE et al (2000) A meta-analysis of the literature for whole body FDG PET detection of recurrent colorectal cancer. J Nucl Med 41:1177–1189

Colorectal adenomas are benign polypoid neoplasms, pedunculated or sessile arising from the epithelial cells of the colorectum, with varying degrees of cellular atypia. Although benign, they are the direct precursors of adenocarcinomas and follow a predictable cancerous temporal course unless interrupted by treatment. ▶Neoplasms, Benign, Large Bowel

Colorectal Cancer Colorectal cancer is the most common cancer of gastrointestinal tract with approximately 150,000 new cases every year. Despite these statistics, mortality from colon cancer has decreased over the past 30 years, possibly because of earlier diagnosis through screening and better treatment modalities. ▶Neoplasm, Large Bowel, Malignant

Colonic Atresia Colovescical Fistula A congenital abnormality, usually due to an ischaemic insult in embryological development resulting in complete occlusion of the lumen of the colon. ▶GI Tract, Paediatric, Congenital Malformations

Complication of sigmoid diverticulitis presenting with frequency, UTI and pneumoturia. CT confirms diagnosis with findings of thickened bladder mucosa, fistulous track and gas in the bladder. ▶Diverticulitis

Colonoscopy Colpocystogram Colonoscopy is the “gold standard” for the detection of colonic neoplasms and the preferred colorectal cancer screening strategy. ▶Neoplasms, Benign, Large Bowel ▶Neoplasms, Large Bowel, Malignant

Invasive radiological technique to assess pelvic floor descent. Progressively replace by MRI. ▶Incontinence, Urinary

Complicated Cyst

Column of Bertin Normal variant of the renal cortex, which may simulate tumors (pseudotumor). ▶Contrast Media, Ultrasound, Applications in Kidney Tumor

Columnar Lined Oesophagus

365

Complex Cyst Cyst with internal masses, thick septations, or thick or irregular wall. ▶Cyst, Breast

Complex Partial Epilepsy (CPE), Focal Epilepsy: Temporal lobe epilepsy (TLE): TLE accounts for approximately 70% of all chronic CPE

▶Reflux, Gastroesophageal in Adults ▶Seizures, Complex, Partial

Comedomastitis ▶Duct Disease, Breast

Communicating Dissection Dissection with dual lumen and visible tears in the intimal flap. ▶Dissection, Aortic, Thoracic

Compartment Anatomy Is the concept of tumor anatomical site and definition of borders. The intracompartmental remaining tumor, which respects anatomical borders such as adjacent fascia, has a better prognosis over the tumor, which is already extracompartmental at presentation. The extracompartmental tumor is one that has already spread beyond the anatomical site of origin and requires more extensive surgery. Involvement of an adjacent joint or space such as the popliteal fossa, is extracompartmental, and may necessitate amputation, if limb salvage procedures are not possible. ▶Neoplasms, Soft Tissues, Malignant

Complex Regional Pain Syndrome Characteristic disabling joint and soft tissue abnormalities of poorly understood etiology. Also known as reflex sympathetic dystrophy, the most common theory of its pathogenesis is an injury to nerves resulting in painful afferent stimuli. These afferent stimuli are thought to activate increased sympathetic tone and other efferent discharge. Symptoms include pain, dysesthesia, swelling, and vasomotor instability. Early changes on scintigraphy are characteristic and demonstrate bilateral asymmetric increased activity in the affected juxta-articular region on all three phases of bone scanning. Decreased unilateral activity occurs in chronic disease. Radiographs may demonstrate periarticular swelling, osteopenia, and subchondral resorption. Preservation of joint space is an important feature in differentiation from an inflammatory arthritis. ▶Fractures, Peripheral Skeleton

Complex Sclerosing Lesion ▶Radial Scar, Breast

Complicated Cyst Cyst with diffuse internal echoes. ▶Cyst, Breast

C

366

Complication of Kidney Graft

Complication of Kidney Graft ▶Transplant Kidney, Complications

Compression, Extrinsic, Esophagus LUIS H. R OS M ENDOZA , M a E UGENIA G UILLE´ N S UBIRA´ N , C RISTINA G UTIE´ RREZ A LONSO University Hospital Miguel Servet, Department of Radiology, Zaragoza, Spain [email protected]

Compression, Extrinsic, Esophagus. Figure 1 Extrinsic esophageal compression related to a noncommunicating esophageal duplication cyst. The barium study (a) shows a significant compression over the esophagus with smooth contours and obtuse angles. Computed tomography (b) identifies a cystic mass, with thick walls and well-defined contours, located in close relationship to the anatomic area of the esophagus.

Synonyms Contour wall alteration; Notches; Wall displacement

Definition Morphologic alteration of the esophagus related to a neighboring mass or space-occupying lesion.

Pathology/Histopathology Pathology depends on the cause of the extrinsic compression, with the main causes classified as congenital, inflammatory, or tumoral. In congenital cases, the pathologic findings are related only to the compressive phenomena without the presence of inflammatory signs or tumoral infiltration, which might be present in the two other groups. Non-neoplastic mediastinal cysts form a group of uncommon benign lesions of congenital origin (1). Neuroenteric and duplication esophageal cysts are localized in the posterior mediastinum, and they might cause this kind of alteration (Fig. 1). Other cases present an inflammatory origin, such as retropharyngeal abscesses or mediastinal pancreatic pseudocysts. Mediastinal masses associated with chronic pancreatitis should raise suspicion for the extension of the inflammatory process to the mediastinum (Fig. 2). Tumoral masses located between the esophagus and the tracheobronchial tree, mainly related to bronchogenic

Compression, Extrinsic, Esophagus. Figure 2 Extrinsic compression of the esophagus related to a mediastinal pancreatic pseudocyst. The sagittal plane of the magnetic resonance study (a) shows a mass in the posterior mediastinum that is compressing the thoracic esophagus. The mass corresponds to a pancreatic pseudocyst. Abdominal computed tomography (b) demonstrates the inflammatory pancreatic involvement. Magnetic resonance imaging by means of the sagittal and coronal planes is a very useful tool for studying the esophagus, as it shows this structure in its major longitudinal plane.

carcinoma, can lead to esophageal infiltration and compression (2). Neurogenic tumors (neurilemmoma, paraganglioma, neuroepithelioma, and neurogenic sarcoma) might also compress the esophagus because of their frequent posterior mediastinal location. Leiomyoma, the most frequent benign tumor of the esophagus, can sometimes grow in an eccentric way, making it difficult to differentiate from an extramural mass (Fig. 3).

Compression, Extrinsic, Esophagus

4.

5. Compression, Extrinsic, Esophagus. Figure 3 Esophageal compression due to an intramural leiomyoma. The barium study (a) shows an esophageal compression that is difficult to classify as intrinsic or extrinsic by only the classic semiological criteria. The compression presents smooth contours and the marginal angles are wide open, almost obtuse. Computed tomography (b) defines a tumoral mass, located in the anatomic area of the esophagus, which is also compressing the trachea, with attenuation values very similar to those of the muscular structures.

Apart from the three groups mentioned earlier, there are other causes for esophageal extrinsic compression: 1. Vascular compression: Aberrant right subclavian artery (ARSA) is an anomaly with a reported incidence of 0.5–2%. The aberrant artery usually follows a retroesophageal course; it rarely takes an anterior course to the esophagus or trachea. Anomalies associated with ARSA are nonrecurring inferior laryngeal nerve paralysis, aortic coarctation, right-sided aortic arch, and a common origin of the carotid arteries. Other vascular causes of compression include anomalies of the aortic arch, an enlarged ascending aorta, a malpositioned descending aorta, and enlarged pulmonary arteries. Enlargement of the left atrium causes compression of the superior part of the distal esophagus, whereas global cardiomegaly can produce compression of the inferior part. 2. Lymphadenopathies: Mediastinal lymphadenopathies (from a tuberculous, metastatic, or lymphomatous origin) can also compress the esophagus and cause dysphagia (3). 3. Thyroid and parathyroid gland enlargement: Enlargement of the thyroid gland, with either a benign or malignant origin, might produce lateral displacement of the esophagus, which is well seen on the anteroposterior view of a barium swallow study. In secondary hyperparathyroidism, there is a generalized hyperplasia

6.

7.

8.

367

of parathyroid glands that, in some cases, can produce esophageal compression (4). Cervical osteophytes: These are found in approximately 20–30% of the aged population. In most of these patients bony spurs are asymptomatic, although they may be associated with neck stiffness and localized or radiating pain. However, large osteophytes that protrude from the anterior edge of the cervical vertebrae can impinge on the upper esophagus, causing dysphagia and odynophagia. Zenker’s diverticulum: This is a protrusion of the mucosal wall of the hypopharynx through the weakened muscular layer between the oblique fibers of the inferior constrictor of the pharynx and the transverse fibers of the cricopharyngeal muscle. When large, it might compress the cervical esophagus. Retropharyngeal hematoma: If this occurs, it is located just in front of the cervical spine. Trauma or anticoagulant therapy can lead to its development. Esophageal pseudomass: A narrowed sagittal diameter of the thoracic inlet is recognized as an anatomic variant that causes dysphagia because of extrinsic compression of the esophagus between the trachea and vertebral bodies, resulting in a pseudomass appearance. Pleural, lung or mediastinal scars: These can retract the esophagus to the involved side.

Clinical Presentation The most frequent symptom related to extrinsic esophageal compression is mechanical dysphagia, first for solids and in cases of advanced obstruction, for both solids and liquids. The point at which the patient experiences this symptom is useful for localizing the level and cause of the compression. Some of the illnesses responsible for the esophageal compression can present specific manifestations: . Coughing is frequent in patients with ARSA, and stridor is frequent in those with right-sided or double aortic arch. . Hiccups suggest involvement of the distal esophagus. . Zenker’s diverticulum is characterized by regurgitation, gurgling sounds, and aspiration. . Large osteophytes (greater than 10 mm) can produce aspiration. . Mediastinal masses, when large, might produce unilateral sibilants and dysphagia because they compromise both the esophagus and the tracheobronchial tree. . Thyroid and parathyroid gland alterations can present endocrine symptoms associated with esophageal manifestations (4).

C

368

Compression, Extrinsic, Esophagus

. Fever, night sweats, odynophagia, and weight loss are common manifestations of lymphoma. . Chronic pancreatitis produces chronic abdominal pain.

Imaging In some cases, conventional chest radiology might show findings that are suggestive of this condition, including intrathoracic goiter; posterior mediastinal masses; pleural, lung or mediastinal retractions; left atrium enlargement; cardiomegaly; aortic arch anomalies or other vascular alterations. Lateral neck radiographs can also be useful for showing the presence of osteophytes, localized retropharyngeal masses, or posttraumatic vertebral fractures. But the most specific semiological criteria are obtained by means of a barium swallow study (5). An intraluminal or intramural mass usually presents its center within or along the contour of the esophageal lumen; as a result, the angles produced by the esophageal wall and the lesions are acute. An extramural mass presents its center out of the contour of the lumen, with these angles being obtuse (Fig. 1a). This is called the spheroid sign. In some cases, the barium swallow shows fixation and traction of the mucosal folds with a teething appearance suggestive of parietal infiltration. By means of a barium swallow, it is also possible to differentiate between the esophageal displacement due to a mediastinal mass, which usually causes a narrowing of the lumen, and the retraction due to a neighboring scar, which produces a widening of the esophageal lumen. In general, the filling defects with smooth or slightly lobulated edges correspond to extramural masses. On some occasions, ultrasound (US) can be useful for characterizing a posterior mediastinal mass as solid or cystic or an enlargement of the thyroid gland. US-guided fine needle aspiration can be performed in selected cases. Endoscopic ultrasound (EUS) is being increasingly used as a less invasive alternative to mediastinoscopy to obtain a diagnosis of mediastinal involvement; EUSguided fine needle aspiration can provide material to establish the histologic diagnosis. Cross-sectional imaging modalities, computed tomography (CT) and magnetic resonance imaging (MRI), are also very useful tools for evaluating this situation. They confirm the presence of the extrinsic compression and allow a precise evaluation of its cause in the different spatial planes (Figs 1 and 2). MRI angiography demonstrates with great accuracy the different vascular causes of dysphagia and allows a definite evaluation of the cardiac chambers.

Nuclear Medicine Nuclear medicine techniques might be useful in studying some entities that can produce an extrinsic esophageal compression. Thyroid pathology is well defined by means of nuclear scintigraphy, and mediastinal lymphomatous masses or adenopathies might be identified by means of gallium citrate scintigraphy. The radionuclide study of choice for parathyroid location is single photon emission CT using technetium-99m sestamibi (2-methoxyisobutylisonitrile).

Diagnosis The initial diagnosis of extrinsic esophageal compression, which can be suspected when the clinical symptoms suggest it, is commonly made during a barium contrast study or during an upper digestive endoscopy that shows a mass or impression covered by normal-appearing epithelium. The first problem to solve is to differentiate between an extrinsic compression versus a lesion arising from the wall itself. The semiological criteria previously defined— spheroid sign, characteristics of the edges of the compression, teething appearance—can be the key findings to establish the differential diagnosis, although this is not always easy (Fig. 3). Some entities can present specific manifestations on the barium swallow that help determine the specific diagnosis: . ARSA appears as a characteristic posterior compression, an oblique notch “in bayonet.” . Intrathoracic goiter presents the cervicothoracic sign, compressing and displacing both trachea and esophagus. The second problem is to determine the origin (cause) of the extrinsic compression. CT and MRI are very useful tools for this because they allow not only identification of the esophageal compression but also identification of its etiology, with the possibility of a detailed evaluation in the different spatial planes. Some entities show specific findings: . Tuberculous mediastinal adenopathies show a hypodense, necrotic center. . Narrowing of the thoracic inlet diameter (normal: 6.2 cm in the sagittal plane, range 5–8.7 cm) occurs in cases of esophageal pseudomass (apparent external compression on the upper esophagus from the right side with displacement to the left and narrowing of the esophageal lumen, which is suggestive of a mass that should be excluded with CT).

Compression, Extrinsic, Stomach and Duodenum

In some cases, the diagnostic algorithm might be completed by means of a percutaneous puncture under US or CT guidance to obtain sample material for pathologic study.

Interventional Radiological Treatment Nonresectable malignant esophageal stenosis, in some cases related to parietal infiltration from a neighboring tumor, might be treated using self-expanding stents. The self-expanding plastic stent is removable, induces less hyperplasia than metal stents, and can be used to treat benign esophageal conditions.

Bibliography 1.

2.

3. 4.

5.

Trojan J, Mousset S, Caspary WF et al (2005) An infected esophageal duplication cyst in a patient with non-Hodgkin’s lymphoma mimicking persistent disease. Dis Esophagus 18(4):287–289 Simchuk EJ, Low DE (2001) Direct esophageal metastasis from a distant primary tumor is a submucosal process: a review of six cases. Dis Esophagus 14(3–4):247–250 Nagi B, Lal A, Kochhar R et al (2003) Imaging of esophageal tuberculosis. Acta Radiol 44(3)329 Weber AL, Randolph G, Aksoy FG (2000) The thyroid and parathyroid glands. CTand MR imaging and correlation with pathology and clinical findings. Radiol Clin North Am 38(5):1105–1129 Levine MS, Rubesin SE (2005) Diseases of the esophagus: diagnosis with esophagography. Radiology 237(2):414–427

Compression, Extrinsic, Stomach and Duodenum LUIS H. R OS M ENDOZA , C RISTINA G UTIE´ RREZ A LONSO, M a E UGENIA G UILLE´ N S UBIRA´ N University Hospital Miguel Servet, Department of Radiology, Zaragoza, Spain [email protected]

Synonyms Contour wall alteration; Notches; Wall displacement

Definition Extrinsic compression of the stomach and duodenum is a morphologic alteration of the gastric and duodenal contour related to a neighboring space-occupying lesion.

369

Pathology/Histopathology Pathology depends on the cause of the compression. The most frequent etiologic factors in gastric compression are liver and spleen enlargement, pancreatic pathology (tumors, cysts, and pancreatitis), gallbladder alterations, retroperitoneal lymphadenopathies, renal masses, and collections in the lesser sac. In duodenal compression, the main causes include hepatomegaly, gallbladder enlargement, coledocal compressions, right renal and adrenal masses, periportal adenopathies, and hepatic angle and transverse colon alterations (1). According to the histological characteristics, compressions can be classified as congenital (duplication cysts of the stomach or duodenum), inflammatory (acute and chronic pancreatitis, Morrison pouch abscesses), or tumoral (pancreatic adenocarcinoma, hepatic carcinoma, renal cell tumor). In the first case, the pathologic findings will be related only to the compressive phenomena without the presence of inflammatory signs or tumoral infiltration, which might be present in the two other groups. Special attention should be given to gastric duplication with its special pathologic characteristics: it can be cystic (the most frequent type), tubular, or tubulocystic. The duplication wall is close to the gastric wall, and its muscular layer fuses with the gastric one, although it rarely communicates with the gastric lumen. The duplication contains gastric epithelium (which can become ulcerated) or ectopic pancreatic epithelium (which can develop into pancreatitis).

Clinical Presentation The patient can have no symptoms, with the extrinsic compression being a casual finding. If symptoms are present, obstruction is the most frequent one, causing epigastric pain, abdominal distension, early satiety, and nausea and vomiting during or immediately after food intake. Vomiting usually alleviates the symptoms. Significant weight loss may occur (2). In low gastric and duodenal obstructions, “gastric splashing” can be found during clinical examination. Some illnesses causing extrinsic compression have special clinical manifestations: . Acute pancreatitis: Nonstop superior abdominal pain that radiates to the back. . Chronic pancreatitis: Severe abdominal pain that is relieved at the genupectoral decubitus and when the patient sits leaning to the front, together with appetite loss, steatorrhea, and diabetes. Duodenal obstruction is more frequent in cases of alcohol-related pancreatitis.

C

370


. Annular pancreas: Bilious vomiting, growth failure, abdominal pain, duodenal obstruction, pancreatitis, and obstructive jaundice. . Duodenal and gastric duplications: In neonates and infants, they produce vomiting, abdominal distention, volvulus, intussusception, and an abdominal mass.

In adults, peptic ulcer or pancreatitis of the ectopic pancreatic tissue may occur. A duplication might get infected, and if it ruptures in the peritoneal cavity, peritonitis occurs. . Duodenal diverticula: About half of the diverticula are found incidentally. When they are symptomatic, the most common clinical presentation is abdominal pain, gastrointestinal bleeding, and perforation. . Internal hernias: Might cause small bowel obstruction (closed-loop or strangulating obstruction). They can also produce chronic intermittent intestinal obstruction. . Superior mesenteric artery syndrome: Appears in cases of significant weight loss, sometimes postsurgical, and in cases of severe burns. The symptoms are a sensation of gastric fullness and abdominal distention after food intake, bilious vomiting, and colicky pain in the middle part of the abdomen, which eases in the prone decubitus and in genupectoral positions.

Imaging

Compression, Extrinsic, Stomach and Duodenum. Figure 1 Extrinsic compression of the stomach related to a pancreatic abscess. The upper gastrointestinal tract barium study (a) shows a significant extrinsic compression of the posterior gastric wall, with smooth contours and obtuse angles. Computed tomography (b) demonstrates the presence of a large pancreatic abscess as the cause of the extrinsic gastric compression. The stomach is identified by the presence of positive oral contrast within its lumen; the abscess, identified by the presence of gas bubbles, is located in close relationship to the posterior gastric wall.

Compression, Extrinsic, Stomach and Duodenum. Table 1 Body: right anterior wall Body: left posterior wall Body: posterior wall Antrum: anterolateral face Antrum: posterolateral face Antrum: posterior face (“antral pad”) Great curvature

In an abdominal plain film, displacement of the gastric luminogram can be identified in some cases. In barium contrast examination, the most common findings are a wide base compression or notch in the luminal contour of the stomach or duodenum, usually accompanied by displacement of the organ, along with a poorly defined area of low density (Fig. 1a). The location of the barium column alteration depends on the cause of the compression (Tables 1 and 2). By means of computed tomography (CT) and magnetic resonance imaging (MRI) in the different spatial planes (axial, coronal, sagittal), these same vectors of displacement can be identified, enabling evaluation not only of the extrinsic compression but also of its cause (Figs 1b and 2).

Nuclear Medicine Gastrointestinal duplications that contain gastric mucosa can be identified with technetium-99m-pertechnetate or

Causes of gastric extrinsic compression

Hepatomegaly, gastric varicose veins Splenomegaly Paraduodenal and lesser sac hernias Enlargement of the gallbladder Internal hernias through Winslow’s hiatus Pancreatic masses (tumors, cysts, pancreatitis), retroperitoneal adenopathies, left renal masses, lesser sac collections Gastric duplication


Compression, Extrinsic, Stomach and Duodenum. Table 2 First and second portions: medial displacement

Second portion: posterolateral compression Second portion: annular compression Second portion: medial wall compression (widening of the duodenal arch) Second and third portions Third portion: anterior compression Third portion: posterior compression Fourth portion: medial displacement Fourth portion: anterior compression

371

Causes of duodenal extrinsic compression

Common bile duct (lineal notch), distended gallbladder (carcinoma), hepatomegaly (caudate lobe), focal hepatic lesions, periportal adenopathies, Morrison pouch collections, colonic hepatic angle distension or neoplasms, internal hernias through Winslow’s hiatus Right renal and adrenal tumors, right renal ptosis Annular pancreas Pancreatic masses (tumors, cysts, pancreatitis), peripancreatic adenopathies (metastatic, lymphomatous), retroperitoneal masses Duodenal duplication cysts Pancreatitis (mesenterium involvement), mesenteric masses, superior mesenteric artery syndrome Retroperitoneal hematoma, aortic aneurysms Abdominal aortic aneurysms Transverse colon distension or neoplasms

technetium-99m diethyltriamine pentaacetic acid scintigraphic evaluation, visualizing the functioning ectopic gastric mucosa simultaneously with the stomach in gastric duplications and even earlier in the case of duodenal duplications (3).

Diagnosis

Compression, Extrinsic, Stomach and Duodenum. Figure 2 Extrinsic compression of the duodenum due to a cystic duodenal duplication. Computed tomography (a) shows the presence of a cystic mass with thick walls in the anatomic area of the pancreatic head. The patient had no antecedents of pancreatitis. The coronal plane of the abdominal magnetic resonance study (b) shows that the cystic mass is closely related to the duodenal walls. This and the thick walls (because congenital duplications “duplicate” all the layers of the wall of the digestive tube) are the key findings for suspected duodenal duplication cyst.

The initial diagnosis of extrinsic compression of the stomach or duodenum is commonly suspected during an upper digestive endoscopy, with the visualization of a mass or impression covered by normal-appearing epithelium, or during a barium contrast study that presents the findings previously described. The first problem to be solved is the differential diagnosis between extrinsic compression—caused by a normal or abnormal structure adjacent to the involved organ—versus a lesion arising from the wall itself. In the upper gastrointestinal tract barium study, if the largest part of the lesion projects out of the gastric or duodenal lumen, and if the angle produced by the gastric or intestinal wall and the mass is obtuse, then the cause is more likely extrinsic (Fig. 3). In addition, the notches related to an extrinsic compression are not constant, changing under the actions of gravity and breathing. Although during the upper endoscopy the finding can be evaluated by means of endosonography and, in some cases, an endosonographically guided fine needle aspiration can be performed (4), transabdominal ultrasound, CT, and MRI are more useful for defining the origin and extent of the extramural masses. The second problem is to determine the origin of the extrinsic compression. In the barium study, the location and characteristics of the extrinsic compression (tubular,

C

372


.

.

.

Compression, Extrinsic, Stomach and Duodenum. Figure 3 Extrinsic compression of the stomach related to a massive splenomegaly. The upper gastrointestinal tract barium study (a) shows the deformity of the great curvature of the stomach, which suggests an extrinsic compression depending on the spleen. A mass effect, with high radiologic density, could be identified in the splenic area. The isotopic study (b) shows the absence of the radioisotope uptake by the spleen due to a massive acute splenic infarct. Edema at this acute stage of the splenic infarct results in the splenomegaly.

serpiginous notches in the case of gastric varicose vein) can be indicative of the cause (Tables 1 and 2), but in general, the cross-sectional imaging modalities, CT and MRI, are more useful in this respect, confirming the findings of the barium study, differentiating intrinsic from extrinsic compression in doubtful cases, and showing the cause of the compression. Some entities, which are detailed later, present more diagnostic difficulties: . Gastric duplications are more frequently found in the great curvature and duodenal duplications in the anteromedial contour of the first and second portions, and they are usually noncommunicating duplications. They appear on CT as rounded lesions that present a water density and are closely related to the gastrointestinal lumen (Fig. 2). On ultrasonography they appear as cystic structures, with the internal mucosal

.

.

wall presenting a bright appearance, being the external muscular layer hypoechoic. In doubtful cases, a fineneedle aspiration biopsy under endoscopic ultrasound guidance or nuclear medicine techniques that allow the identification of gastric mucosa can be performed. The preoperative diagnosis of internal hernia can be made with barium studies, but in the presence of small bowel obstruction, CT plays an important role, allowing the identification of a sac-like mass or cluster of small bowel loops at an abnormal anatomic location, with a stretched and displaced mesenteric vascular pedicle and converging vessels at the hernial orifice. Superior mesenteric artery syndrome appears on CT as a sharpening in the middle of the third duodenal portion, with retrograde distension. On barium study there is an abrupt lineal compression with dilatation of the first and second duodenal portions and a fluctuant appearance of the barium column, as well as delayed gastric emptying. Gastric varicose veins are easy to identify on CT. They appear as serpiginous lesions that present intense enhancement after the intravenous contrast administration. Annular pancreas causes, in the barium study, a filling defect with annular morphology in the descendent portion of the duodenum at the level of the Vater ampulla, with a homogeneous dilatation of the proximal duodenum and inverted peristalsis of this area. To establish the diagnosis of this entity, endoscopic retrograde cholangiopancreatography, pancreatic CT, and MRI are very useful tools. Pancreatic neoplasms produce characteristic signs in the barium study when they have a long evolution. Enlargement of the duodenal arch and the Frostberg inverted-three sign (compression of the duodenal loop with fixation of the middle portion at the level of Vater papilla) are very well known. In cases with muscular infiltration or mucosal edema, contour spiculation and blurring of mucosal folds are produced; however, these signs are not specific to tumoral lesions, also appearing in acute and chronic pancreatitis.

Interventional Radiological Treatment In patients with advanced neoplasms that compress the stomach or duodenum, an alternative to traditional surgical palliative treatment (gastrojejunostomy and duodenojejunostomy) is gastroduodenal stent placement, which can be done under fluoroscopic guidance or with a combination of fluoroscopic and endoscopic techniques. Stent placement reduces postsurgical complications, provides better gastric emptying, and is more cost-effective than surgical palliation (5).

Congenital Abnormalities, Pancreatic

Bibliography 1.

2.

3.

4. 5.

Martıń ES, Pedrosa CS (2000) El tubo digestivo: semiologıá radiolo´gica. In: Pedrosa CS, Casanova R (eds) Diagno´stico Por Imagen. Tratado de Radiologıá Clıńica. McGraw-Hill-Interamericana de Espanã, Madrid, pp 335–378 Feldman M, Friedman LS, Sleisenger MH (2004) Enfermedades Gastrointestinales Y Hepa´ticas: fisiopatologıá, Diagno´stico Y Tratamiento. Editorial Me´dica Panamericana, Buenos Aires Kumar R, Tripathi M, Chandrashekar N et al (2005) Diagnosis of ectopic gastric mucosa using 99Tcm-pertechnetate: spectrum of scintigraphic findings. Br J Radiol 78(932):714–720 Hwang JH, Kimmey MB (2004) The incidental upper gastrointestinal subepithelial mass. Gastroenterology 126:301–307 Lopera JE, Brazzini A, Gonzales A et al (2004) Gastroduodenal stent placement: current status. Radiographics 24:1561–1573

Computed Radiography (CR) After passing through the patient, the emergent X-ray beam is detected by a photostimulable phosphor plate, forming a stable latent image. A laser then stimulates the plate, light is emitted and a digital signal produced (via a photomultiplier tube) than is converted into the image. The resultant image can be read either on laser-printed film or as soft copy on a workstation. ▶Children, Imaging Techniques

373

Congenital (or Dysgenetic) Pancreatic Cyst Congenital pancreatic cyst is an intrapancreatic cystic lesion non-communicating with the duct system and lined by a single layer of flat epithelium. The cysts are usually enclosed in a thin fibrous capsule and they are filled with a mucoid or serous fluid. They may range from microscopic lesions up to 3–5 cm in diameter. They are believed to result from anomalous development of the pancreatic ducts. Unlike cysts in the liver and kidneys, asymptomatic simple pancreatic cysts are uncommon. Most of these cysts are multiple and associated with von Hippel–Lindau disease or, rarely, with inherited polycystic kidney disease. Macroscopic pancreatic cysts are occasionally seen in patients with cystic fibrosis. At imaging, congenital cysts appear as well-defined cystic lesions with smooth walls and without intramural excrescences or septations. Associated renal and hepatic cysts may be seen. ▶Cystic Neoplasms, Pancreatic

Congenital Abnormalities, Pancreatic Computed Tomographic Colonography

H UISMAN TA 1 , C ARLA C APPELLI 2 1

Johns Hopkins University, Baltimore, USA Department of Diagnostic and Interventional Radiology, University of Pisa, Italy [email protected] [email protected]

2

CTC is a novel imaging modality for the evaluation of the colonic mucosa in which thin-section spiral CT provides high resolution two-dimensional (2D) axial images; CT data sets are edited off-line in order to produce multiplanar reconstructions (coronal and sagittal images) as well as three-dimensional (3D) modeling, including endoscopic-like views. ▶Neoplasms, Benign, Large Bowel

Synonyms Pancreatic congenital anomalies; Pancreatic congenital variants; Pancreatic developmental anomalies

Definitions

Computed Tomography Diagnostic modolity based on X-rays able to obtain crosssectional images of the human body is a non invassive manner. ▶Ischemic Heart Disease, CT

The pancreas arises from dorsal and ventral anlages that fuse at 6 weeks of gestation. The ventral anlage develops into a part of the pancreatic head and the uncinate process while the dorsal part gives rise to the pancreatic body and tail as well as a second part of the pancreatic head. Congenital variants of the pancreas are seen in

C

374


approximately 10% of children and include pancreatic agenesis, hypoplasia, ectopia, ▶pancreas divisum, and ▶annular pancreas. Complete pancreatic agenesis is extremely rare and is usually associated with other major malformations. In agenesis of the dorsal anlage only the pancreatic head develops, e.g., in combination with polysplenia syndrome. In pancreatic hypoplasia, the exocrine elements or more rarely also the endocrine elements are underdeveloped. Pancreatic ectopia is defined as pancreatic tissue in ectopic locations (e.g., stomach, duodenum, jejunum, appendix, Meckel’s diverticulum). These lesions have no contact with the normal pancreas and possess its own ductal system and blood supply. It is usually found at autopsy or as an incidental finding at laparotomy. The incidence varies between 1% and 15%. In pancreas divisum, the dorsal and ventral anlages did not fuse resulting in two separate pancreatic portions which drain separately into the duodenum: the major drainage (body and tail) is performed by the duct of Santorini through the minor papilla, while the main duct of Wirsung drains only the posterior part of the pancreatic head through the major papilla. The accessory duct of Santorini becomes the main excretory pathway of the gland. This anomaly is the most common pancreatic anomaly (4–14% of all children). The annular pancreas results from a premature fusion of the dorsal and ventral anlages. This anomaly is characterized by a ring of pancreatic tissue that encircles the second portion of the duodenum. A second embryological theory is that an annular pancreas may also result from an anomaly of the ventral bud rotation. Duodenal obstruction may occur. Incidence is increased in children with Down syndrome and other congenital anomalies.

Pathology and Histopathology While in complete pancreatic agenesis no pancreatic tissue is found at all, in partial agenesis parts of the pancreas can be identified in the pancreatic region. The histology of the pancreatic tissue is normal. In pancreatic hypoplasia, lipomatous hypertrophy of the pancreas with extensive atrophy of the exocrine acinar tissue is seen while the endocrine system is preserved; in cases of an associated endocrine hypoplasia the Langerhans cells are also atrophic. Ectopic pancreatic tissue includes all histological elements of the exocrine and the endocrine pancreas, although the ducts of the exocrine pancreatic tissue are not arranged in the normal anatomical pattern. In gastrointestinal ▶ectopic pancreas, the pancreatic tissue is

usually found in the submucosa. The ectopic pancreatic tissue may be functionally active and secreting, causing local ulceration of the mucosa. Inflammation of the ectopic pancreas with pseudocyst formation or the development of benign or malignant pancreatic tumor has also been described. Pancreas divisum occurs when the ductal systems of the ventral and dorsal pancreatic ducts fail to fuse. Grossly pancreas divisum is characterized by a completely separated pancreatic ductal system in an undivided gland. In pancreas divisum the predominant drainage is performed by the dorsal duct of Santorini through the minor papilla. In patients with pancreas divisum, signs of acute or chronic pancreatitis can be found in the dorsal duct distribution while normal pancreatic tissue is seen in the ventral duct distribution. It is suggested that the duct of Santorini and the minor papilla are too small to adequately drain the large amount of secretions which may lead to a relative outflow obstruction with resulting pancreatitis. Annular pancreas is believed to result from an early fusion of the ventral and dorsal anlages or from a failure of normal pancreatic tissue to rotate around the duodenum. The annulus represents the ventral part of the pancreas that remains fixed to the duodenum. In the extramural type, the pancreatic tissue is drained by a duct originating at the anterior surface of the duodenum, passing posteriorly around the duodenum and opening into the main pancreatic or common bile duct near the ampulla. In the intramural type, pancreatic tissue is found within the duodenal wall and small ducts drain directly into the duodenum.

Clinical Presentation Pancreatic agenesis is usually associated with severe forms of intractable neonatal diabetes mellitus. Complete pancreatic agenesis is usually characterized by severe morbidity and mortality. In pancreatic acinar hypoplasia the endocrine pancreatic function is usually intact while the exocrine cells are atrophic. The exocrine pancreatic insufficiency results in malabsorption with weight loss and chronic diarrhea. Congenital pancreatic hypoplasia should always be kept in mind when a child manifests with gross steatorrhea and responds well to pancreatic enzyme replacement therapy. ▶Shwachman–Diamond syndrome (SDS) is a rare autosomal recessive disorder characterized by exocrine pancreatic insufficiency associated with bone marrow dysfunction and skeletal anomalies (metaphyseal chondrodysplasia). In affected children the pancreatic acinar cells do not develop in utero and are replaced by fatty tissue.


Ectopic pancreas is a relatively rare entity, usually of no clinical importance and found incidentally during laparotomy or in autopsy. If present, symptoms are nonspecific, including abdominal pain, nausea, vomiting, and gastrointestinal bleeding. Specific symptoms can develop in case of complication. Pancreatitis can occur in ectopic pancreas, leading to typical symptoms and laboratory findings; the ectopic insulin secretion may cause hypoglycemia. Exocrine or endocrine tumors may develop in the ectopic pancreatic tissue. Obstructive symptoms are mainly related to the localization of the ectopic pancreas (gastrointestinal obstruction, obstructive jaundice); intestinal mechanical obstruction can be related to intussusception. Most patients having pancreas divisum are asymptomatic, although some reports have suggested a high incidence of abdominal pain and acute and chronic pancreatitis with typical clinical and laboratory findings. It has been suggested that the relative stenosis of the accessory papillary orifice, the major outflow tract for pancreatic secretions, is the cause of problems. In many instances annular pancreas is discovered incidentally during radiological imaging studies performed for other reasons. Symptoms develop when complications occur, including pancreatitis, biliary obstruction, and peptic ulcer. If the annular pancreas forms a complete ring, there may be total obstruction of the duodenum and diagnosis is rapidly evident in the neonate; if the ring is incomplete, obstruction may occur later in life or may never produce symptoms. A history of polyhydraminos and other anomalies such as intestinal malrotation, duodenal atresias, and cardiac anomalies are often associated.

Imaging Ultrasound (US) is a highly sensitive, nonionizing bedside primary imaging modality that can locate, identify, and characterize pancreatic tissue. Computed tomography (CT ) and magnetic resonance imaging (MRI) are second line imaging modalities; dynamic contrastenhancement pattern may increase sensitivity and specificity. Magnetic resonance cholangiopancreatography (MRCP) can be a noninvasive alternative for endoscopic retrograde cholangiopancretography (ERCP). In complete pancreatic agenesis, no pancreatic parenchyma can be found in the expected locations. In partial agenesis the pancreatic tissue has a normal aspect on US, CT, and MRI. Pancreatic hypoplasia can be demonstrated by imaging because of the presence of pancreatic lipomatosis with the replacement of the normal pancreatic parenchyma by fat. Ectopic pancreatic tissue is unlikely to be identified by US, CT, or MRI. Ectopic pancreas in the stomach and

375

duodenum can occasionally be identified on barium studies as submucosal masses. Their appearance is however nonspecific mimicking any kind of submucosal mass in the gastrointestinal tract. Ectopic pancreatic islands in the stomach and duodenum can display a central depression, which corresponds to the opening of a duct. Gastric locations are typically located along the greater curvature in the proximity of the pylorus, whereas in the duodenum, they are mostly found proximally between the bulb and the ampulla of Vater. Ectopic pancreas located in the gallbladder can be visualized by US, which will display hyperechoic nodular mural thickening without acoustic shadowing; however this appearance is not specific. In pancreas divisum CT and MRI can occasionally demonstrate a separate dorsal and ventral pancreatic portion divided by a thin fat plane. Definite diagnosis of pancreas divisum is obtained by MRCP or ERCP. MRCP may demonstrate a short (1–6 cm) and thin main duct of Wirsung and a larger accessory duct of Santorini which drains almost the entire pancreas from the tail to the anterior part of the head. The identification of the separated ducts can be enhanced by secretin stimulation. In rare cases, pancreas divisum is associated with a focal dilatation of the duct near the papilla, called santorinicele. Annular pancreas is usually demonstrated by barium enema studies. Asymmetric extrinsic narrowing of the lateral and medial borders of the second portion of the duodenum with normal mucosal pattern is a typical finding. On US, CT, or MRI the annular pancreas may appear as duodenal wall thickening, as an apparent enlargement of the pancreatic head or as a well-defined ring of pancreatic tissue encircling the duodenum. The presence of calcification can suggest an associated chronic pancreatitis. MRCP demonstrates the presence of the annular duct which drains into the main pancreatic duct.

Nuclear Medicine Diagnosis In patients with unexplained and persisting signs of cholangitis, pancreatitis, jaundice, recurrent abdominal pain, nausea, and vomiting, a congenital anomaly of the pancreatic or bile duct must be considered. Because clinical findings are usually nonspecific, US, CT, and MR are essential for diagnosis. In case of ectopic pancreatic tissue, diagnostic imaging can fail in detection because of the presence of obscuring complications such as pancreatitis and duodenal perforation. In such cases only the histopathological findings can demonstrate the presence of pancreatic tissue. There is however limited value for nuclear medicine imaging in congenital anomalies of the pancreas.

C

376

Congenital Adrenal Hyperplasia

Interventional Radiological Treatment

Definitions

ERCP is used for therapeutic intervention in patients with pancreas divisum. Several endoscopic and/or surgical procedures have been proposed in an attempt to improve the pancreatic outflow through the minor papilla in patients with acute recurrent pancreatitis. These include endoscopic sphincterotomy, ductal balloon dilatation, and pancreatic duct stent placement.

Anatomical Anomalies (1–4)

Bibliography 1.

2. 3. 4. 5.

Leyendecker JR, Elsayes KM, BI Gratz et al (2002) MR cholangiopancreatography: spectrum of pancreatic duct abnormalities. Am J Roentgenol 179:1465–1471 Fulcher AS, Turner MA (1999) MR pancreatography: a useful tool for evaluating pancreatic disorders. Radiographics 19:5–24 Roberts- Thomson IC (2004) Images of interest. Hepatobiliary and pancreatic: annular pancreas. J Gastroenterol Hepatol 19:464 Nijs E, Callahan MJ, Taylor GA (2005) Disorders of the pediatric pancreas: imaging features. Pediatr Radiol 35:358–373 Nagi B, Kochhar R, D Bhasin et al (2003) Endoscopic retrograde cholangiopancreatography in the evaluation of anomalous junction of the pancreaticobiliary duct and related disorders. Abdom Imaging 28:847–852

Congenital Adrenal Hyperplasia A group of inborn errors of metabolism arising from enzyme defects in the biosynthesis pathways of adrenal corticosteroids, resulting in inadequate production of glucocorticoids and mineralocorticoids and excess production of adrenal androgens. ▶Adrenogenital Syndrome ▶Ambiguous Genitalia

Congenital Anomalies of the Pancreas F.E. AVNI 1 , C. M ATOS 1 , J.-F. C HATEIL 2 1

Department of Medical Imaging, University Clinics of Brussels—Erasme, Brussels, Belgium 2 Department of Pediatric Imaging, Hospital Pellegrin, Bordeaux, France [email protected]

Synonyms Agenesis of dorsal pancreas; Annular pancreas; Congenitally short pancreas; Ectopic pancreas; Pancreatic cyst(s); Pancreas divisum; Pancreatic lipomatosis

. Pancreas divisum is the most common anatomic variant of the pancreas. Its incidence varies from 4–14% of autopsies. It is characterized by an incomplete fusion between the dorsal and ventral ducts. Only the duct of Wirsung drains into the ampulla of Vater. The duct of Santorini drains into the minor papilla. Stenosis and pancreatitis may ensue. . Annular pancreas possibly results from abnormal adherence of the ventral duct to the duodenum that results in a complete ring of pancreatic tissue around the duodenum. About 50% of the patients are present in the neonatal period with an associated duodenal obstruction. Other associated anomalies are Trisomy 21, congenital heart disease, and tracheo-esophageal fistula. . Congenitally short pancreas corresponds to an absence of the pancreatic neck, body, and tail. There is an increased risk of diabetes and associated malformations. . Ectopic Pancreas: ectopic pancreatic tissue is an aberrant rest of “normal” pancreatic tissue that can be seen in various locations (stomach, jejunum, liver, spleen, umbilicus, Meckel’s diverticulum, and so on). It may ulcerate or remain asymptomatic. . True single congenital cyst of the pancreas may be isolated or associated to ano-rectal, kidney or limbs anomalies. Intestinal duplication cysts may be sequestrated within the pancreas. Congenital anomalies resulting from systemic diseases that affect the pancreas (1–4) . Pancreatic lipomatosis is characterized by fatty infiltration of the pancreas associated with deficient secretion of pancreatic enzymes. It is a common finding in the ▶Shwachman–Diamond and ▶Johanson– Blizzard syndromes already in infants. The lipomatosis develops progressively in cystic fibrosis (CF). In CF, multiple cysts may develop in late childhood (cystosis). These congenital causes must be differentiated from acquired causes such as lipomatosis following steroid therapy. . Multiple cysts in the pancreas are seen in Von Hippel Lindau disease and in Dominant Polycystic kidney disease, but usually in adult patients. . Pancreatomegaly is observed in ▶Beckwith– Wiedemann syndrome. . Congenital tumors: congenital tumors are very rare but perinatal diagnosis of pancreatoblastoma (exocrine tumor) and insulinoma (endocrine tumor) has

Congenital Anomalies of the Pancreas

been reported. Among endocrine tumors, nesidioblastosis corresponds to ill defined tumors leading to intractable neonatal hypoglycemia. Pancreatic cysts may be acquired and appear in the course of cystic fibrosis (cystosis).

377

about the pancreatic duct and biliary tract. A normal MRCP may obviate the need for ▶ERCP. ERCP is useful whenever MRCP does not provide sufficient information or when a therapeutic maneuver is planned. In selected cases, peripancreatic venous sampling can be performed in order to localize nesidioblastosis or other ill defined pancreatic endocrine tumors (1–5).

Pathology/Histopathology The histopathological anomalies observed on the pancreas will depend upon the type of the primary lesion and complications that have occurred. For instance, inflammatory lesions or fibrosis will be observed in case of pancreatitis associated to pancreas divisum. Fatty infiltration of the gland will be observed in case of Shwachman disease. Islet cells hypertrophy or a circumscribed tumor can be identified in case of nesidioblastosis.

Clinical Presentation Congenital anomalies of the gland will be clinically evident either because of an antenatal diagnosis of pancreatic tumor (cyst) or because of an abdominal mass palpated at clinical examination. Other symptoms that orient toward a pancreatic anomaly are evidence of pancreatic insufficiency (fatty diarrhea, failure to grow, diabetes) or abdominal pain related to complications of the malformation. A pancreatic endocrine tumor should be suspected in case of seizures associated with neonatal hypoglycemia (1–4).

Nuclear Medicine Not applicable

Diagnosis The first imaging tool for evaluating the pancreas is ultrasound. The technique will allow evaluation of its size, form, and echogenicity. Its relation with the portal system will be evaluated. Whenever a mass is observed, CT or MRI should be performed. Whenever a dilatation of the pancreatic duct is observed and whenever an anomaly of the gland morphology is observed, MRCP should be performed (Figs 1 and 2). Abnormal hyperechogenicity should suggest lipomatosis (Fig. 3) and other symptoms of an associated syndrome should be looked for (1–5).

Imaging Ultrasound is the primary screening tool to evaluate the pediatric pancreas. Its size varies with age. The head and the tail increase ±1.0 cm at birth and 2 cm at 10–15 years. The body varies from 0.5 to 1 cm. The gland’s echogenicity is slightly hyperechoic to liver at birth, relatively hypoechoic in children. The width of the pancreatic duct should not measure above 1.5 mm at 1 year and 2.2 mm at 15 years. The pediatric pancreas is well seen on contrastenhanced CT. The technique is well suited to visualize pancreatic calcifications, tumors, or complication of traumatism. MRI can be used for similar indications than CT. Furthermore, the technique using special ▶MRCP sequences provides important supplementary information

Congenital Anomalies of the Pancreas. Figure 1 Annular pancreas: MRCP demonstrating the pancreatic ducts and the typical annular appearance of the pancreatic duct (arrow).

C

378

Congenital Anomalies of the Uterus

4. 5.

Lerner A, Branski D, Lebenthal E (1996) Pancreatic disease in children. Pediatr Clin North Am 43:125–156 Arcement CM, Meza MP, Arumanla S et al (2001) MRCP in the evaluation of pancreatico-biliary disease in children. Pediatr Radiol 31:92–97

Congenital Anomalies of the Uterus ▶Congential Malformations, Mullerian Duct

Congenital Anomalies of the Pancreas. Figure 2 Pancreas divisum: MRCP demonstrating separately the main pancreatic and Sontorini ducts.

Congenital Anomalies or Malformations of the Urinary Tract ▶Congential Malformations, Genitourinary Tract; Including Ureter and Urethra

Congenital Abnormalities, Splenic H UISMAN TA, C HIARA F RANCHINI Johns Hopkins University, Baltimore, USA [email protected]

Synonyms Inborn Splenic Abnormalities; Splenic Abnormalities; Splenic Malformations

Congenital

Definition Congenital Anomalies of the Pancreas. Figure 3 Pancreatic lipomatosis in a case of Shwachman syndrome. Ultrasound of the pancreas; the hyperechogenicity of the pancreatic head is striking (between arrows).

Congenital splenic anomalies include inborn anomalies of shape, location, number, and size of the spleen due to aberrant embryologic development.

Pathology and Histopathology Bibliography

Accessory Spleen

1.

The spleen is a mesodermal derivative, which first appears as a mesenchymal cell condensation inside the dorsal mesogastrium. Sometimes, additional smaller splenic condensations appear and give origin to ▶accessory spleens, representing by far the most common congenital

2. 3.

Enriquez GE, Vasquez E, Aso C (1998) Pediatric Pancreas: an overview. Eur Radiol 8:1236–1244 Nijs E, Callahan MJ, Taylor GA (2005) Disorders of the pediatric pancreas: imaging features. Pediatr Radiol 35:358–373 Synn AY, Mulvihill SJ, Fonkalsrud EW (1988) Surgical disorders of the pancreas in infancy and childhood. Am J Surg 156:201–205

Congenital Abnormalities, Splenic

abnormality of the spleen. Accessory spleens resemble the splenic structure. They may be located anywhere in the abdomen; the most common sites are near the splenic hilum and the tail of the pancreas. Other possible locations are along the splenic vessels, in the gastrosplenic and splenorenal ligaments, in the mesentery, and in the omentum. When splenectomy is performed for hypersplenism, hypertrophy of an accessory spleen may cause recurrent disease (1, 2) (Fig. 1).

Abnormal Location

Shape Abnormalities

Spleno-Gonadal Fusion and Spleno-Renal Fusion

Splenic clefts, notches, and lobules may persist in adult life as variations of the normal shape and are common (1) (Fig. 2).

379

The spleen may be seen in a variety of locations. In congenital diaphragmatic hernia, in Bochdalek diaphragmatic hernia and in eventration of the diaphragm, the spleen may be intrathoracic. If the lateral peritoneal recess is particularly deep the spleen is found posteriorly to the left kidney. The spleen may also be located in the right hypochondrium (1).

▶Spleno-gonadal fusion is a rare anomaly. Due to the close relationship between the developing spleen and

Congenital Abnormalities, Splenic. Figure 1 Accessory spleen CT study shows two small round masses near the splenic hilum (arrows). In the contrast-enhanced arterial phase, they have a strong inhomogeneous enhancement, analogous to that of the normal spleen.

Congenital Abnormalities, Splenic. Figure 2 Splenic cleft CT contrast-enhanced arterial phase. A fissure is appreciable along the posterior margin of the spleen. The smooth edges and the absence of contrast extravasation suggest a congenital splenic cleft.

C

380


left gonadal-mesonephric structures, an accessory spleen may be found attached to the left ovary or kidney or may be located within the scrotum. Spleno-gonadal fusion can be classified into two types: continuous (direct connection between the spleen and gonad) and discontinuous (no anatomic connection between ectopic splenic tissue and the principal spleen). This anomaly predominates in males. It can occur as an isolated condition or can be associated with other abnormalities, such as cryptorchismus and orofacial or limb abnormalities (1, 3).

Wandering Spleen Absence, laxity, or excessive length of ligaments of the spleen leads to an abnormal mobility of the organ. Torsion of the long vascular pedicle may occur, followed by vascular occlusion and splenic infarction (1, 4).

Spleno-gonadal fusion may manifest as a mobile and painless left scrotal mass. Other presentations include cryptorchismus, testicular torsion, and inguinal hernia. Spleno-gonadal fusion is often asymptomatic in females. In case of an intrathoracic location of the spleen, patients usually have respiratory symptoms. The ▶wandering spleen usually is symptomatic in the childhood and may present with an abdominal mass and acute, chronic, or intermittent symptoms due to torsion of the vascular pedicle. In children with asplenia and polysplenia syndromes, the clinical manifestations may be related to congenital heart disease, immune deficiency (splenic absence), or volvulus due to intestinal malrotation (1).

Imaging Accessory Spleen

Asplenia and Polysplenia The absence of the spleen (▶asplenia) or the presence of multiple small spleens (▶polysplenia) is a rare condition usually associated with other congenital malformations, especially cardiovascular anomalies. Polysplenia or asplenia may associate with abdominal situs ambiguous and are known as asplenia and polysplenia syndrome. Asplenia syndrome is most frequently encountered in males and may be associated with severe cyanotic congenital heart diseases. The typical anatomic features of asplenia syndrome are trilobed lungs with bilateral minor fissures and epiarterial bronchi, bilateral systemic atria, midline liver, and a variable location of the stomach. Polysplenia syndrome is more common in females. The typical anatomic features of classic polysplenia syndrome are bilobed lungs with bilateral hypoarterial bronchi, bilateral pulmonary atria, midline liver, and multiple spleens of variable size and number that may be located in either the left or right abdomen. In rare cases these patients have a single, lobulated spleen or even a normal spleen. Malrotation of the bowel is a frequent finding in heterotaxy syndrome (1, 5).

Splenic Atrophy Congenital ▶splenic atrophy is quite uncommon and is associated with recurrent bacterial infections (1).

Clinical Presentation Accessory spleens and splenic shape abnormalities are typically an incidental finding at imaging.

Accessory spleens may vary in number and size, usually ranging from a few millimeters to several centimeters in size. Typically, they appear as round or oval masses near the splenic hilum, but they may be found anywhere in the abdomen. Intrapancreatic accessory spleen, typically in the tail, can mimic a neoplastic mass. Imaging features are identical to those of normal splenic tissue. At computed tomography (CT), the attenuation values before and after contrast administration are identical to those of the spleen. The characteristic inhomogeneous enhancement during the arterial phase is crucial to demonstrate the nature of the mass. On magnetic resonance images (MRI), the spleen is hypointense on T1-weighted scans and hyperintense on T2-weighted scans with respect to the liver and the pancreas. The use of reticuloendothelialtargeted contrast media can confirm the splenic nature of the mass (1, 2).

Shape Abnormalities Splenic clefts, notches, and lobules are quite common findings and must be distinguished from traumatic lesions. The demonstration at US and CT of characteristic smooth edges and the absence of contrast extravasation suggests a congenital splenic cleft (1).

Spleno-Gonadal Fusion and Spleno-Renal Fusion In case of left scrotal mass, US is usually the first examination; the splenic tissue appears as a homogeneous, well-encapsulated mass with the same echotexture as the normal spleen. At Doppler US, a vascular architecture analogous to that of the spleen is seen. If spleno-gonadal


fusion is suspected, a cord-like structure connecting the spleen to the mass should be searched for. CT may be helpful to demonstrate associated splenic-renal fusion or cryptorchismus. The typical enhancement pattern confirms diagnosis. In women, spleno-gonadal fusion is usually an incidental finding at US or CT (1, 3).

Wandering Spleen On plain abdominal X-ray, the absence of the normal splenic outline in the left upper quadrant in association with a soft-tissue mass in the abdomen or pelvis may be noted. US and CT confirm the absence of the spleen in its expected location and the presence of an ectopic splenic location. Torsion of the vascular pedicle may lead to infarction. The congested or infarcted spleen may have a normal echotexture or may be hyperechogenic due to secondary hemorrhage. The spleen’s comma-shaped configuration is usually preserved. Splenomegaly with rounded edges of the organ is suggestive for torsion and is attributed to venous congestion. The lack of a flow signal on Doppler US will confirm hypoperfusion. US contrast media can increase sensitivity. Doppler spectral analysis may show a low diastolic velocity with an elevated resistive index in the splenic artery. Unenhanced CT usually shows a decreased density of the spleen; after contrast medium injection a partial or total lack of enhancement may be seen. A whirlpool appearance of the splenic vessels within the splenic hilum indicates torsion (1). Dense splenic vessels (dense artery sign), corresponding to acute thrombosis, are occasionally observed. MR can provide useful information about the precise location

381

of the wandering spleen, the viability of the splenic parenchyma and the splenic vessel anatomy (MR angiography) (4).

Polysplenia and Asplenia In polysplenia, numerous small splenic masses can be seen predominantly in the right upper quadrant at US, CT, MR, or scintigraphy. Identification of a missing spleen is more difficult than confirmation of its presence. Scintigraphy remains the gold standard (1, 5).

Nuclear Medicine 99m-Technetium sulfur colloid scintigraphy and 99mtechnetium-labeled heat-damaged red blood cells offer functional images and are highly specific for differentiating splenic tissue from other masses. Although nuclear medicine offers the most specific imaging techniques for identifying ectopic splenic tissue, CT and MR offer superior anatomic resolution (2). Scintigraphy is used to confirm the presence of normally functioning splenic tissue in cases of accessory spleens, ectopic spleen, splenogonadal and ▶spleno-renal fusion or polysplenia. Scintigraphy is also the examination of choice in documentating the absence of the spleen. However, the absence of a detectable radiotracer uptake can also occur in so-called “functional asplenia,” in which the splenic phagocytic function is markedly reduced, despite the presence of

Congenital Abnormalities, Splenic. Figure 3 Splenosis CT study demonstrates multiple masses scattered throughout the abdomen (liver surface, liver hilum, and left hypochondrium) in a patient who had a posttraumatic splenectomy several years before. The masses have regular margins. In the contrast-enhanced arterial phase, the masses typically have a strong inhomogeneous enhancement, analogous to that of the normal spleen.

C

382

Congenital Cystic Adenomatoid Malformation

splenic tissue. Functional asplenia may occur in sickle cell disease, secondary to radiation therapy and chemotherapy, secondary to tumor invasion of the spleen, in splenic anoxia or after bone marrow transplantation (1).

Bibliography

Diagnosis

3.

Ectopic splenic tissue may mimic neoplasms and lymphadenopathies. Imaging can make a definite diagnosis avoiding biopsy. Ectopic splenic tissue shows imaging features identical to those of the spleen in all imaging modalities. The enhancement pattern, especially the inhomogeneity in the arterial phase, is very specific. The combination of different MR sequences increases the confidence in diagnosis. MR reticuloendothelial-targeted contrast media can confirm the splenic nature of the mass. 99m-Technetium sulfur colloid scintigraphy and 99m-technetium-labeled heat-damaged red blood cells represent the most specific imaging techniques to confirm the presence of functioning splenic tissue. In spleno-gonadal fusion presenting with scrotal mass, US is the first examination to be performed. The homogeneity of the echotexture, the regularity of the vascular architecture and the presence of well-defined margins suggest a nonneoplastic nature of the mass. When spleno-gonadal fusion is suspected, a comparison with the US appearance of the spleen and a study directed to the visualization of a cord-like structure connecting the mass with the normal spleen should be performed. A definitive diagnosis cannot be made solely on the basis of US findings. Nuclear medicine imaging can confirm the presence of splenic areas of activity. However, surgical exploration is generally required to rule out malignancy. Nevertheless, orchiectomy can be avoided because splenic tissue can be dissected away from the tunica albuginea (3). Accurate preoperative diagnosis of wandering spleen with or without torsion represents an imaging challenge and can be made with US, CT, and MR. Perfusion and viability of the splenic parenchyma can be assessed by contrast-enhanced CT and MR, Doppler ultrasonography, and scintigraphy. Information concerning splenic perfusion and viability is important for the surgeon, especially in younger children where splenopexy instead of splenectomy is the treatment of choice in uncomplicated wandering spleen. Multiple splenic masses, usually in the abdomen, may be seen in case of multiple accessory spleens, polysplenia, and ▶splenosis (autotransplantation of splenic tissue occurring after splenic trauma or after splenectomy) (Fig. 3). History of previous splenectomy suggest the diagnosis of splenosis; in case of polysplenia, other congenital abnormalities are associated.

4.

1.

2.

5.

Paterson A, Frush DP, Donnelly LF et al (1999) A pattern-oriented approach to splenic imaging in infants and children. Radiographics 19(6):1465–1485 Miyayama S, Matsui O, Yamamoto T et al (2003) Intrapancreatic accessory spleen: evaluation by CT arteriography. Abdom Imaging 28(6):862–865 Pomara G (2004) Splenogonadal fusion: a rare extratesticular scrotal mass. Radiographics 24(2):417 Deux JF, Salomon L, Barrier A et al (2004) Acute torsion of wandering spleen: MRI findings. Am J Roentgenol 182 (6):1607–1608 Applegate KE, Goske MJ, Pierce G et al (1999) Situs revisited: imaging of the heterotaxy syndrome. Radiographics 19(4):837–852; discussion 853–854

Congenital Cystic Adenomatoid Malformation Intrapulmonary multicystic mass of pulmonary tissue with an abnormal proliferation of bronchial structures. ▶Congenital Malformations, Tracheobronchial Tree

Congenital Diaphragmatic Hernia (CDH) A congenital defect in the diaphragm that allows abdominal contents including bowel and solid abdominal organs to herniate into the chest, most commonly occurring in utero, and resulting in lung hypoplasia on the affected side (left sided in 90%). ▶Hernia, Diaphragm, Congenital ▶Contrast Media, Ultrasound, Influence of shell on Pharmacology and Acoustic Properties

Congenital Duplication Cyst Space-occupying lesion of a congenital origin that “duplicates” the different layers of the wall of the digestive tube. It can be communicating or noncommunicating. It is more frequent in the gastric great curvature and in the anteromedical contour of the first and second portions of the duodenum. It presents a water density on CT and a hypoechoic appearance on ultrasonography.

Congenital Heart Disease and Great Vessel Disease, MRI

Congenital Heart Disease and Great Vessel Disease, MRI A.

VAN

S TRATEN , G. L. G UIT, A.

DE

R OOS

Radiology Department, Kennemer Gasthuis, Haarlem, Noord-Holland, The Netherlands [email protected]

Definition Congenital heart diseases (CHD) form a wide range of cardiac malformations. The most common cyanotic congenital heart disease is Tetralogy of Fallot (TOF). Other relatively common CHD are: transposition of the great arteries, atrioventricular septal defects and various kinds of valvular diseases. The most widely used classification is based on the sequential segmental analysis in which the heart is divided into three segments: the atrial chambers, the ventricular mass and the great arteries. In this model, the first step is identification of the arrangement of the atrial chambers (situs). Then ventricular morphology and topology are determined, for example type of atrioventricular connection and morphology of atrioventricular valves. Thirdly, the morphology of the great arteries is assessed (type of ventriculoarterial connection, morphology of arterial valves and infundibulum). Associated cardiac malformations are also noted as well as the cardiac position (position of the heart within the chest and the orientation of the cardiac apex).

Pathology Most cases of CHD occur spontaneously and aetiology is multifactorial. However, a recent prospective study has shown that patients with CHD have an increased risk of having offspring with a similar CHD (1). The recurrence risk varies between 3 and 8% and is dependent on the cardiac defect and the gender of the involved parent. In general, the risk of recurrence is highest for atrioventricular defects and lowest for TOF. Further evidence of a genetic aetiology is provided by numerous reports of families comprising multiple members with CHD that follow a mendelian pattern of inheritance. This suggests a single gene defect as the specific cause of the CHD. An example of a specific disorder in which the gene defect has been elucidated is the Holt–Oram syndrome. This is an autosomal dominant disorder characterized by atrial and ventricular septal defects as wells as upper limb abnormalities. The affected gene (TBX5) is a member of the T-box family of transcription factors.

383

Clinical Presentation Patients with CHD usually present shortly after birth (e.g. transposition of the great arteries), or in (early) childhood. The clinical symptoms are usually failure to thrive and cyanosis. In some cases, patients present at a much later stage or even in adulthood, for example ‘pink Fallots’, when right ventricular outflow tract obstruction is only mild. When a patients presents with symptoms that would suggest a CHD, history is very important. In the workup of these patients, the electrocardiogram and chest film may provide further insight into the type of cardiac malformation, however, for a definite diagnosis, echocardiography or cardiac magnetic resonance (CMR) imaging is often needed. With an incidence of 0.3–0.5 per 1,000 live births, TOF is the most common cause of cyanotic CHD (Fig. 1) (2). Traditionally, corrective surgery was preceded by a palliative shunt connecting the systemic arterial circulation to the pulmonary arterial system. It was thought that these shunts would promote growth of the pulmonary arteries and total correction could be delayed until the patient was larger. Nowadays, patients undergo total repair in infancy, which may decrease the risk of late sudden death, but may increase the severity of residual pulmonary regurgitation (PR) in some patients, due to the higher number of trans-annular patch procedures needed in young infants.

Imaging The clinical role of MR imaging in diseases of the heart and great vessels is rapidly evolving. CMR has become an established non-invasive imaging modality for the assessment of various cardiac disorders, such as cardiac masses, cardiomyopathies, aortic and pericardial diseases and congenital heart diseases. Moreover, due to its accuracy

Congenital Heart Disease and Great Vessel Disease, MRI. Figure 1 Schematic view of the Fallot heart.

C

384

Congenital Heart Disease and Great Vessel Disease, MRI

Congenital Heart Disease and Great Vessel Disease, MRI. Figure 2 MR four-chamber view of an adult with TOF and RV hypertrophy due to residual pulmonary stenosis.

and reproducibility, CMR is currently considered the gold standard for quantification of ventricular volumes, function and mass (Fig. 2) (3). Comprehensive functional assessment is possible by CMR due to its capability to measure flow velocity and flow volume, which is a basic requirement to quantify lesion severity in valvular heart disease. Over the past few years, major technical advances have considerably improved acquisition speed and image quality making CMR a useful tool for the evaluation of patients with ischemic heart disease as well. Although the clinical robustness of coronary magnetic resonance angiography still needs improvement, CMR currently provides valuable information to detect reversible ischemia, myocardial infarction and residual viability. Transthoracic echocardiography is still the most commonly used technique in the non-invasive assessment of CHD, especially in young children whose small thoracic diameter provides an optimal acoustic window. After surgical intervention, however, the use of ultrasound is often restricted because of scar tissue and thoracic deformations. Unlike echocardiography, CMR provides unlimited access to the thoracic cavity and different techniques are available for detailed visualization and accurate measurement of the complex post-surgical morphology and functional status. Biventricular volumes and blood flow can be accurately measured during a single examination, allowing complete evaluation of both right and left ventricular systolic and diastolic function as well as intracardiac and vascular flow. In the seriously ill or uncooperative patient CMR imaging is often limited, and it is contraindicated in

patients with pacemakers. CMR studies are also timeconsuming and may require patient sedation. Ultrasound has been the imaging modality of choice for many years, however, recently the role of electron beam CT and multidetector CT (MDCT) has been established in the evaluation of congenital heart disease. The advantages of MDCT compared to MR imaging should be considered in selecting the optimal imaging modality for a child with CHD. Because MDCT takes less time and has fewer requirements for sedation than does MR imaging, it can be more easily performed in an unstable patient who needs intensive monitoring and care. Failure to wean the patient off the ventilator post-operatively may result from the airway compression caused by the vascular structure. The relationship between the airway and vessel can be accurately evaluated at MDCT. This information is useful for surgical management and cannot be provided by any other imaging modality. On the other hand, MDCT has several disadvantages, such as the need for iodinated contrast agents and radiation exposure, although the latter can be minimized by using a low-dose MDCT protocol.

Nuclear Medicine The role of radionuclide angiography is limited nowadays since non-invasive imaging modalities such as cardiac CMR and MDCT offer detailed information on cardiac morphology as well as biventricular function.

Diagnosis Presently, the vast majority of CHD can be surgically repaired. However, even after corrective surgery, the right ventricle (RV) may remain subject to an abnormal pressure and/or volume overload, due to longstanding residual pulmonary stenosis and/or PR. Hence, RV function has shown to be a major determinant of clinical outcome in CHD patients. With the growing number of long-term survivors amongst CHD patients, the need for accurate follow-up becomes increasingly important in clinical management. Careful monitoring of functional parameters such as cardiac function and vascular flow will help to allow early detection of the most important postoperative complications and aid in the timing of re-interventions (4). Timely detection of late complications requires adequate assessment of right ventricular size and function as well as quantification of intracardiac blood flow in these patient groups. Widely used techniques, such as echocardiography and radionuclide studies, have limitations when applied for this purpose, especially in patients with abnormal RV morphology.

Congenital Malformations, Adrenals

Interventional Radiological Management The recent implementation of fast CMR sequences along with the ability of MR imaging to acquire images in any given orientation and with high soft-tissue contrast makes this technique attractive for guiding interventional procedure. Razavi et al showed that cardiac catheterization guided by MR imaging is safe and practical in a clinical setting, allows better soft-tissue visualization, provides more pertinent physiological information, and results in lower radiation exposure than do fluoroscopically guided procedures (5). MR guidance could become the method of choice for diagnostic cardiac catheterization in patients with CHD, and an important tool in interventional cardiac catheterization and radiofrequency ablation. MR guided cardiopulmonary interventions, however, are challenging due to motion artefacts caused by the beating heart and respiration. Moreover, the tortuous anatomy of the right cardiac chambers and pulmonary arteries makes monitoring of the passage of guide wires and endovascular catheters and the deployment of stents with MR imaging more difficult. Kuehne et al showed that interactive real-time MR imaging has the potential to guide stent placement in the pulmonary valve or main pulmonary artery and measure blood flow volume in the stent lumen immediately after the intervention (6). Several other authors have reported the use of MR imaging—compatible catheters and guide wires. However, none of the investigated instruments contained properties required for complex cardiovascular interventions, such as (a) fast and reliable detection of the tip; (b) curvature of the shaft and (c) material properties such as tip flexibility, torque and tracking ability, shaft strength and flexibility. More work is needed to develop new MR imaging—compatible catheters and guide wires that are appropriate for use in cardiovascular interventions. In the future, research methods such as use of resonance circuits as fiducial markers should be investigated to provide tip and shaft detection without incorporating the risk of heating effects inherent with active catheter tracking methods.

5.

6.

Razavi R, Hill DL, Keevil SF et al (2003) Cardiac catheterisation guided by MRI in children and adults with congenital heart disease. Lancet 362:1877–1882 Kuehne T, Saeed M, Higgins CB et al (2003) Endovascular stents in pulmonary valve and artery in swine: feasibility study of MR imaging-guided deployment and postinterventional assessment. Radiology 226:475–481

C Congenital Hepato-Biliary Anomalies ▶Congenital Malformations, Liver and Biliary Tract

Congenital Lesions of the Adrenal Gland ▶Congenital Malformations, Adrenals

Congenital Lobar Emphysema Congenital progressive overdistension of a lung lobe. ▶Congenital Malformations, Tracheobronchial Tree

Congenital Lymphatic Malformation ▶Lymphangioma


Bibliography 1.

2.

3. 4.

Burn J, Brennan P, Little J et al (1998) Recurrence risks in offspring of adults with major heart defects: results from first cohort of British collaborative study. Lancet 351:311–316 Ferencz C, Rubin JD, McCarter RJ et al (1985) Congenital heart disease: prevalence at livebirth. The Baltimore-Washington Infant Study. Am J Epidemiol 121:31–36 de Roos A, Doornbos J, van der Wall EE et al (1995) Magnetic resonance of the heart and great vessels. Nat Med 1:711–713 Vliegen HW, Van Straten A, de Roos A et al (2002) Magnetic resonance imaging to assess the hemodynamic effects of pulmonary valve replacement in adults late after repair of tetralogy of Fallot. Circulation 106:1703–1707

385

R OSEMARY A RTHUR Department of Pediatric Radiology, Leeds General Infirmary, Clarendon Wing, Leeds, UK [email protected]

Synonyms Congenital lesions of the adrenal gland

386


Definition Congenital lesions of the adrenal gland arise from disordered embryogenesis, ▶inborn errors of metabolism, or the occurrence of disease processes identified in the fetus or early neonatal period.

Histology/Pathology The fetal adrenal cortex develops around 6 intrauterine weeks from coelomic mesodermal tissue and between 8 and 12 weeks it is invaded by ectodermal sympathetic cells of the neural crest to form the adrenal medulla producing catecholamines. The fetal adrenal cortex differentiates into an outer definitive zone and a larger inner fetal zone producing androgenic precursors for the placental synthesis of estriol. The fetal adrenal gland is proportionately very large, but soon after birth the fetal zone involutes disappearing by 1 year. At the same time the definitive zone of the cortex differentiates to form the glomerulosa (15%), lying peripherally and producing aldosterone, and the fasciculata (75%) and the reticularis (10%), lying next to the medulla both producing glucocorticoids and androgens. The glomerulosa becomes fully differentiated around 3 years of age, whereas the reticularis is not fully differentiated until approximately 15 years. In the adult a normal adrenal gland weighs approximately 4–5 g (1, 2).

Abnormalities in Embryogenesis A number of developmental abnormalities of the adrenal gland result from disordered embryogenesis. Most are asymptomatic and discovered incidentally, although some are associated with other developmental anomalies (2–5). Adrenal agenesis is rare and usually occurs in conjunction with ipsilateral renal agenesis. Ten percent of infants with renal agenesis will have adrenal agenesis. In these cases, the renal/adrenal agenesis is thought to be due to the failure of the mesonephric ridge to develop, whereas the majority of cases of renal agenesis (i.e., those not associated with adrenal agenesis) develop as a result of the failure of the ureteric bud to develop normally. Adrenal hypoplasia of the adrenal glands in anencephalic infants is associated with triploidy and prenatally acquired pituitary and CNS degenerative disorders due to ACTH deficiency. Hypoplasia is also seen in neonatal adrenoleukodystrophy and Zellweger’s syndrome. Accessory adrenal glands are found in up to 50% of autopsies in children, and contain both adrenal cortical and medullary tissue. They may be hormonally active becoming overproductive following adrenalectomy. They

are found around the celiac axis, along the line of the gonadal vessels and around the ovaries in girls and around the testes, epididymis, and hernial sacs in boys. Straight adrenal refers to a discoid-shaped adrenal gland that appears elongated and straight on both longitudinal and transverse imaging rather than having the normal Y, Z, or V shape. This anomaly is seen in association with renal agenesis, hypoplasia, or ectopia and is thought to result from the failure of the renal tissue to indent into the adrenal gland during early development. Horseshoe adrenal is a single mass of adrenal tissue that lies across the midline, generally posteriorly to the aorta and anteriorly to the inferior vena cava (IVC), but when associated with asplenia is always seen anteriorly to the aorta. Coexisting anomalies occur frequently including 52% asplenia, 37% neural tube defects, 29% renal anomalies, and in 3% Cornelia de Lange syndrome. Horseshoe adrenal does not occur in association with polysplenia, a differentiating feature from asplenia. Circumrenal adrenal gland is a further anatomical variant describing the finding where there is fusion of two limbs of one gland extending down and around the kidney. Adrenohepatic fusion and adrenorenal fusion are uncommon anomalies. These occur when the adrenal gland becomes incorporated within the capsule of the liver or kidney and is thought to be due to the disruption of intervening coelomic epithelium allowing these adjacent organs to fuse.

Clinical Presentation With the exception of adrenal agenesis and severe hypoplasia, these anatomical variants of fusion are associated with normal adrenal function and histologically normal tissue. Adrenal tissue in ectopias, straight adrenal, and fusion anomalies usually show the characteristic pattern of normal adrenal glands on ultrasound (Figs 1 and 2). Congenital adrenal hyperplasia is a group of inborn errors of metabolism arising from enzyme defects in the biosynthetic pathways of adrenal corticosteroids resulting in inadequate production of glucocorticoids and mineralocorticoids and the excess production of adrenal androgens. (Give link to CAH/Ambiguous genitalia essays) ▶Wolman’s disease is an inborn error of metabolism due to a deficiency of the enzyme acid esterase leading to an accumulation of triglycerides, cholesterol, and its esters within many organs particularly the inner layers of the adrenal cortex. The infants develop massive enlargement of the adrenal glands, hepatosplenomegaly, jaundice, vomiting, steatorrhea, abdominal distension, and failure to thrive.


387

C

Congenital Malformations, Adrenals. Figure 1 adrenal gland.

(a) US (longitudinal US) and (b) (transverse US), of a normal, right neonatal

Congenital Malformations, Adrenals. Figure 2 (a) US (longitudinal US) and (b) (transverse US; both same patient as Figure 1), the straight or discoid left adrenal gland associated with left renal agenesis.

The condition generally presents early in infancy and is rapidly progressive resulting in death during the first year. A number of pathological processes result in cystic adrenal lesions that need to be differentiated from simple adrenal cysts and include cystic ▶neuroblastoma, adrenal hemorrhage, adrenal abscess, epidermoid cysts, and intraabdominal extralobar pulmonary sequestration (ELPS) that may be sited within or adjacent to the adrenal glands. Careful imaging should help to differentiate these lesions from the cystic or dysplastic upper moiety of a duplex kidney. Adrenal congestion and hemorrhage occur in the perinatal period in response to perinatal asphyxia and stress. Hypoxic damage to the endothelial cells results in adrenal congestion which may be followed by hemorrhage and hemorrhagic infarction. Prolonged abdominal compression in labor, particularly in infants born to diabetic mothers, and underlying bleeding diatheses and are also considered etiological factors. Adrenal hemorrhage may be bilateral but is more commonly unilateral affecting the

right gland in 70% of cases on account of direct drainage of the right adrenal vein into the IVC. The condition is often asymptomatic, recognized as a result of a palpable kidney displaced by an enlarged adrenal gland, or may present clinically with jaundice, anemia, and rarely hypovolemic shock. If hypertension and/or impaired renal function is present, the possibility of a coexistent ipsilateral renal vein thrombosis should be considered. This occurs more commonly on the left where the adrenal vein drains into the left renal vein. Hemorrhage may track down the retroperitoneal tissues and into the scrotum causing scrotal swelling. Long-term adrenal insufficiency is unusual but may develop when more than 90% of both glands are affected. Neuroblastoma is the commonest extracranial solid pediatric neoplasm and may present prenatally or be diagnosed at birth, although more commonly it presents in the preschool years. The tumor is variable in size at presentation, and is more commonly cystic than tumors presenting later in childhood. The tumor may be

388


asymptomatic and discovered incidentally, but when large or associated with hepatomegaly it may result in early neonatal respiratory distress. Neonatal metastatic spread to the liver, skin, and bone marrow sometimes occurs, and when found in association with a localized adrenal primary is staged 4S. These 4S tumors have a favorable prognosis, in contrast to prenatal presentation with hydrops and intraspinal extension, which is associated with a poor prognosis. (Link to tumors/neuroblastoma essay) Adrenal endothelial cysts are seen infrequently in the adrenal glands and are often found incidentally during an ultrasound examination for an unrelated problem. Cysts can be huge and multiloculated, and they are generally filled with blood. They are usually unilateral but may be bilateral and generally decrease in size over time. Adrenal abscess is unusual in the neonate and is most commonly seen in association with a resolving adrenal hemorrhage following Neisseria meningitidis, Escherichia coli, b-hemolytic streptococcal, or Staphylococcus aureus septicemia.

in conjunction with areas of calcification. Lymph node enlargement and bowel wall thickening may also be noted (Fig. 3). A number of cystic abnormalities develop in the adrenal glands as discussed earlier and these may be difficult to differentiate on the basis of imaging characteristics alone. Simple adrenal cysts usually appear as simple, thin-walled cysts, whereas resolving hemorrhage and abscess will usually have a thicker wall, and the fluid may contain echoes or even a fluid–fluid level. Adrenal hemorrhage presents as an echopoor mass within a mildly enlarged adrenal and tends to become smaller and resolve over 2–4 weeks, often leaving a densely calcified involuted adrenal gland (Fig. 4). Neuroblastoma may also present as a cystic mass, and should be suspected if the wall shows some nodularity. However, the mass may be homogeneous and indistinguishable from other benign or malignant tumors. The differential diagnosis of a complex cystic/solid adrenal lesion should also include an ELPS. These are commonly left-sided and are often first seen in the second trimester on ultrasound, whereas a congenital neuroblastoma is generally first seen in the third trimester

Imaging The normal adrenal gland is Y, Z, or V shaped, closely related to the upper pole of the kidneys, and is easily seen in the neonate on ultrasound imaging using a highfrequency transducer (Fig. 1). The adrenal gland is a difficult organ to measure on account of its shape, but the thickness of the adrenal limbs should be approximately equal and less than 4 mm in width. Ultrasonically, the neonatal gland has a thin central hyperechoic stripe representing the medulla, central veins, and connective tissue with a surrounding hypoechoic cortex. With increasing age, the differentiation between the cortex and medulla disappears. The contour should be smooth or gently undulating but should not show any marked nodularity (4, 5). Imaging findings in Wolman’s disease are fairly characteristic, demonstrating a massive increase in the size of both glands that tend to retain their overall shape

Congenital Malformations, Adrenals. Figure 3 Abdominal X-ray demonstrating calcification in enlarged adrenal glands. Normal overall shape is maintained. (From Paterson A (2002) Adrenal pathology in childhood: a spectrum of disease 12:2491–2508).

Congenital Malformations, Adrenals. Figure 4 (a) Early US (b) CT, and (c) later ultrasound images of large adrenal hemorrhage, which involutes and become calcified.

Congenital Malformations, Bile Ducts

following a normal second trimester scan and is more commonly right-sided. EPLS may show increased vascularity, and the identification of a systemic arterial supply helps to clinch the diagnosis. Abdominal radiography is useful in the assessment of abdominal distension, and in Wolman’s disease it will show enlarged and densely calcified adrenal glands, whereas in neuroblastoma the calcification is more focal. In the older child a triangular, dense, calcified adrenal gland may be noted following perinatal adrenal hemorrhage. CT and MRI scanning of the neonate are rarely helpful in clearly differentiating hemorrhage from a cystic neuroblastoma or adrenal abscess, although helical CT and MRI may be useful to further delineate a mass lesion, particularly when solid and large, and to identify any metastases associated with neuroblastoma. Helical CT and MRI are also useful for the detection of aberrant vessels suggesting the presence of an ELPS. Metaiodobenzylguanidine scintigraphy is only indicated in cases of neuroblastoma.

389

Congenital Malformations, Bile Ducts F.E. AVNI , C. M ATOS , M. C ASSART Department of Medical Imaging, University Clinics of Brussels, Erasme Hospital, Brussels, Belgium [email protected]

Synonyms Alagille syndrome: Ductular paucity of intrahepatic bile ducts; Bile duct cyst; Biliary atresia; Caroli’s disease; Choledochal cyst; Choledochocele; Choledochal cyst type V; Communicating cavernous ectasia of the intrahepatic bile ducts; Extrahepatic atresia; Progressive familial cholestasis; Syndromic hepatic ductular hypoplasia; Watson-Alagille syndrome

Diagnosis

Definition

The main difficulty in diagnosis is differentiating between the various cystic, complex, or homogeneous masses. A clinical history of perinatal hypoxia should point to the possibility of adrenal congestion and/or hemorrhage, and these lesions should gradually become smaller over the first few weeks of life. Failure to involute is suggestive of more significant pathology, for example, neuroblastoma. Although the potential diagnosis of neuroblastoma is of concern in the case of both cystic and solid congenital adrenal lesions, the outlook for 4S lesions is very good and it is generally considered safe to wait several weeks, particularly in cystic lesions, to see if they involute. Percutaneous/open biopsy will be required for a definitive diagnosis in solid lesions or cystic lesions that fail to involute. Unfortunately, vanillylmandelic acid estimations, a noradrenalin metabolite, are generally unhelpful in the neonatal period.

In the infant/neonatal age group, biliary atresia (BA) is the predominant surgical malformation followed by choledochal cyst (CC) and spontaneous perforation of the common bile duct (CBD) (1, 2). BA consists of a severely complete or partial interruption of the extrahepatic biliary tree (EHBT). It occurs in 1:10,000 to 1:25,000 births. It is thought to result from a progressive in utero, inflammatory disease of the entire biliary tract. There are multiple different anatomic types depending on the extent of the inflammatory and fibrotic process. This leads to numerous variants including (partial) patency of the CBD or gallbladder but also some cystic dilatation of bile ducts (3, 4). CC consists of a spectrum of malformations of the extrahepatic and intrahepatic bile ducts. The primary anomaly is apparently related to an abnormal pancreaticobiliary junction. There is a common segment that is too long, with reflux of pancreatic juice into the CBD. It occurs in 1:100,000 births. An early classification indicated three groups of CC. Type I, fusiform or cystic dilatation of the CBD, represents 90% of the patients. Type II corresponds to a diverticulum of the CBD. Type III represents a choledochocele that protrudes into the duodenal lumen. More recently, two types have been added: type IV corresponds to the presence of multiple extrahepatic bile duct cysts, either as an isolated anomaly (IVa) or in association with intrahepatic biliary cysts (IVb; Caroli’slike disease). Type V is a cystic dilatation of the

Bibliography 1.

2. 3.

4.

5.

Miller WL (1997) The adrenal cortex and its disorders. In: Brook C, Clayton P, Brown R (eds) Brook’s Clinical Pediatric Endocrinology. Blackwell Publishing, pp 293 Paterson A (2002) Adrenal pathology in childhood: a spectrum of disease. Eur Radiol 12:2491–2508 Strouse PJ, Haller JO, Berdon WE et al (2002) Horseshoe adrenal gland in association with asplenia: presentation of six new cases and review of the literature. Pediatr Radiol 32:778–782 Daneman A (2004) The adrenal gland. In: Kuhn JP, Slovis TL, Haller JO (eds) In Caffey’s Paediatric Diagnostic Imaging. Mosby, pp 239–251 Siegel MJ (2002) Adrenal glands, pancreas and other retroperitoneal structures. In: Siegel MJ (ed) Pediatric Sonography, Raven Press, pp 475–527

C

390


intrahepatic bile ducts equivalent to Caroli’s disease. Caroli’s disease itself is a congenital cystic dilatation of the intrahepatic bile ducts that communicate with the biliary system. The disease can be associated with autosomal recessive polycystic kidney disease, hepatic fibrosis, or both; the condition is then referred as Caroli’s syndrome (1–3). Spontaneous perforation of the CBD is a very rare condition of unknown origin (ischemic lesion is a possible explanation). The perforation is always located at the junction of the cystic duct and the CBD. It may or not be associated with the very rare neonatal biliary lithiasis or biliary plug syndrome. Intrahepatic bile duct tree congenital anomalies include Byler’s disease (progressive familial cholestasis) and ▶Alagille syndrome (intrahepatic ductular paucity). The spectrum of congenital anomalies of the biliary tree also includes rare cases of agenesis, duplication, and septation of the gallbladder (anomalies that are usually without clinical consequences) (1, 2).

Pathology/Histopathology In BA, the entire EHBT is involved. The biliary ducts are atretic and hypoplastic. Early in the course of the disease, there are tiny but patent biliary structures in the perihilar tissue at the porta hepatis that disappear progressively after 4 months. Some infants may have sparing of the distal BD (patent gallbladder, cystic duct, and CBD). Some others present segmental patency with a pseudocyst-like structure at the level of the CBD or liver hilum. Microscopic evaluation of the liver reveals cholestasis, distorted bile ducts, and a distorted portal vascular system. Extensive fibrosis is also visible (1, 2). In CC, a long common biliopancreatic channel is observed with partial obstruction of the CBD. The CBD is thickened and fibrotic without a muscular layer. The mucosa is ulcerated. In the case of Caroli’s disease, pathology demonstrates saccular or fusiform ductal dilatation of the intrahepatic biliary tree (IHBT). Depending on the age of the patient, evidence of cholangitis, periductal fibrosis, and cirrhosis may be demonstrated as well.

Clinical Presentation In the neonatal period, most congenital malformations of the BT will be suspected in cases of persistent jaundice, hepatomegaly, and acholic stools. This is the case for up to 80% of patients with BA. CC is classically suspected with the association of abdominal pain, a right upper quadrant mass, and fluctuant jaundice. Antenatal diagnoses have been reported.

Other clinical symptoms are related to the (later) complications of the congenital malformations (cirrhosis, biliary stones, or rupture of varices).

Imaging Various imaging techniques provide information about the biliary tract. 1. Conventional X-ray of the spine can be done to demonstrate butterfly vertebrae (Alagille syndrome). 2. Ultrasound (US) is the primary technique used to image the biliary tract, especially in neonates. The first structure to be identified is the CBD. Yet a normal CBD is difficult to visualize; it has to be differentiated from the hepatic artery. Dilatation should be considered when the diameter is >2 mm, and CC when >7 mm. The intrahepatic bile ducts are dilated when their diameter exceeds the diameter of the parallel portal branch. Evaluation of the gallbladder (GB) is more controversial because, even in cases of BA, it may be present and normal in size (2.5–3.5 cm in length). An empty GB after a meal does not exclude the diagnosis; on the contrary, an enlarging GB after 4 h of fasting excludes BA. Ultrasound is helpful in detecting intrahepatic or extrahepatic biliary cysts. It is also important when the triangular cord sign is demonstrated; this consists of a triangular echogenic irregular structure (the fetal biliary duct) at the level of the porta hepatis. Color Doppler analysis is important to detect frequently associated vascular anomalies (preduodenal portal vein, interrupted inferior vena cava). Periportal fibrosis, portal hypertension, and polysplenia are other possible anomalies demonstrable by US (4). Magnetic resonance imaging (MRI) can be used in infants and even neonates to demonstrate a normal or abnormal biliary tract. Using T2-weighted MR cholangiography (MRCP), images of the normal or dilated biliary tract can be easily obtained as in adults, and the technique helps to exclude BA. In cases of BA, the MR equivalent of the US triangular cord sign has been described: a triangular area of high intensity at the level of the porta hepatis can be observed. The status of the liver (periportal fibrosis) and/or vascular anomalies can be analyzed using T1 (without and with gadolinium) and echogradient sequences (5). Endoscopic retrograde cholangiopancreatography (ERCP) can be performed at any age to assess the patency of the EHBTor IHBT. Its yield is greater in cases of dilated biliary ducts where it has replaced transhepatic opacification of the biliary tract. Its advantage lies in a possible therapeutic maneuver and an easier interpretation than with the imaging.


A perioperative cholangiogram is mandatory to confirm the diagnosis of BA, to determine its type and to differentiate BA with biliary cysts from CC.

391

and liver anatomy. Whenever an obstructive jaundice is confirmed, ERCP can be proposed as an alternative to surgery to precisely assess the diagnosis and eliminate any obstructive sludge or lithiasis.

Nuclear Medicine

C

Nuclear medicine scans are of utmost importance for the differential diagnosis of ▶neonatal cholestasis and a presumptive diagnosis of BA. 99m Tc-disofenin (DISIDA) has the highest hepatic extraction and the highest diagnostic yield. After proper preparation with phenobarbital (for 5 days), the examination can be performed in neonates. Lack of hepatic captation of the tracer favors hepatic insufficiency (i.e., neonatal hepatitis), and lack of excretion into the intestine on a 24-h delayed scan is highly suggestive of BA. Excretion into the intestine excludes the diagnosis (1, 2).

Diagnosis Once neonatal cholestasis or obstructive jaundice is suspected, US should be performed to evaluate the liver and biliary tract. Absence of a visible GB despite 4 h of fasting, a nondilated CBD, the presence of triangular cord sign (Fig. 1), a periportal biliary cyst, echogenic kidneys, and polysplenia are signs suggestive of BA. In equivocal cases, MRCP may provide additional information and eventually exclude the diagnosis. Nuclear medicine scanning should rapidly be performed (Fig. 2), and during surgery a cholangiogram is mandatory. If the EHBT or IHBT is dilated on US, MR imaging may be performed to better understand the biliary tract

Congenital Malformations, Bile Ducts. Figure 1 Biliary atresia in a neonate. Ultrasound demonstrating the triangular cord sign (arrow). Oblique view through the porta hepatic.

Congenital Malformations, Bile Ducts. Figure 2 Biliary atresia. DISIDA hepatic scan in a neonate. The tracer has been correctly metabolized by the liver but not excreted because no bowel has opacified, despite delayed views.

Congenital Malformations, Bile Ducts. Figure 3 Choledochal cyst type 1. Magnetic resonance cholangiopancreatographic demonstration (heavily T2-weighted sequence) of the typical cystic dilatation of the common bile duct.

392

Congenital Malformations, Bone

Both MRCP and ERCP are useful for assessing communication between intrahepatic or extrahepatic cysts and the biliary tract as observed in CC (Fig. 3) or Caroli’s disease. Again, when surgery is elected, a preoperative cholangiogram will be necessary. Whenever Caroli’s disease or syndrome is suspected, the kidneys should be assessed to confirm associated kidney disease.

Bibliography 1. 2.

3.

4.

5.

Karrer FM, Hall RJ, Stewart BA et al (1990) Congenital biliary tract disease. Surg Clin North Am 70:1403–1418 Benya EC (2002) Pancreas and biliary system: imaging of developmental anomalies and diseases unique to children. Radiol Clin North Am 40:1355–1362 Levy AD, Rohrmann CA Jr, Murakata LA et al (2002) Caroli’s disease: radiologic spectrum with pathologic correlation. Am J Roentgenol 179:1053–1057 Lee HC, Yeung CY, Chang PY et al (2000) Dilatation of the biliary tree in children: sonographic diagnosis and its clinical significance. J Ultrasound Med 19:177–182 Norton KI, Glass RB, Kogan D et al (2002) MR cholangiography in the evaluation of neonatal cholestasis: initial results. Radiology 222:687–691

Congenital Malformations, Bone A LAN E. O ESTREICH Cincinnati Children’s Hospital Medical Center, Cincinnati, Ohio, USA [email protected]

Synonym Deformity or other structural abnormality of skeletal elements present at (or before) birth

Definitions Abnormality such as deformity of skeletal elements that is present at or before birth. It may be due to a genetic cause, intrauterine position, or effects of adverse intrauterine conditions, including amniocentesis physical injury, other trauma, prenatal surgery, irradiation, infection substances including medicines ingested by the pregnant mother, or maternal metabolic disease or malnutrition. The abnormalities may affect bone, cartilage, joints, and muscles and tendons, as well as vessels, nerves, and surrounding soft tissues. Those conditions that manifest later in childhood,

but are due to causes already present at birth, may be considered as related to congenital malformations.

Pathology/Histopathology Malformations of bone do not have a specific histopathologic abnormality in general. What one sees on imaging is what the malformation would be on gross pathological examination. Associated vascular and neuromuscular changes are also generally matters of gross pathology, such as the small anterior tibial artery associated with clubfoot. One exception is congenital bony malformation in neurofibromatosis-1, in which mesodermal tissue abnormality is found.

Clinical Presentation The clinical presentations of congenital malformations are as varied as the malformations. Limp and leg length discrepancy are common to many. Some, including Down syndrome and Williams syndrome, have characteristic associated facies. Skin lesion (such as cafe´-au-lait spots), small rib cage, and fingernail abnormalities are present in some conditions.

Imaging Although malformations are initially evaluated on plain radiographs, additional modalities are appropriate for specific situations. For example, not yet ossified growth center bones of the foot may be revealed by ultrasound, computed tomography (CT), or magnetic resonance imaging (MRI). For positional relationships of the lower extremity, weight-bearing or simulated weight-bearing images are necessary. For vertical talus diagnosis and evaluation, a maximum plantar flexion lateral view is necessary. The anterior nose or anterior protrusion of the calcaneus on lateral images (1) predicts presence of calcaneonavicular coalition, which is then demonstrated nicely with the everted oblique view (of Slomann) of the foot; CT is not required. The inverted oblique view of the foot, incidentally, nicely shows accessory navicular centers of the foot, although the weight-bearing anteroposterior (AP) view that shows alignment abnormalities should also be obtained. Scoliosis evaluation requires upright images whenever possible. The flexibility of curves may be checked with supine maximal passive bending of the patient as additional views. Cervical spine evaluation for C1 or C2 malformation or for transverse ligament weakness uses flexion and extension lateral images. To evaluate for

Congenital Malformations, Bone

lateral clavicle hook (a sign of upper extremity reduction deformity or weakness); the arms should be at the side of the chest (2). To evaluate an area of upper or lower extremity for unilateral growth disturbance, comparison with the contralateral side is appropriate. In proximal focal femoral dysplasia, MRI is advisable early in life to determine the presence of cartilaginous centers and to reveal if seemingly nonconnected bony parts of the femur might be connected by cartilaginous (or fibrous) tissue. Ultrasound can yield similar information, but in less exquisite detail. For pectus excavatum, CT images allow documentation of the severity of the deformity as well as its effects on the heart and other thoracic structures. In utero imaging for malformation begins with highdetail ultrasound, including 3D imaging. Questions that arise may then be further investigated with fetal MRI. If the answer is still not given, but is important, selected radiographs of the mother’s abdomen may be considered, under close supervision by a specialized pediatric radiologist. If possible, radiographs of one twin should be avoided if the other is considered unaffected, for radiation protection considerations.

Nuclear Medicine If, in malformation, the question arises of whether a bone has capability of growth at a site, bone scanning can indicate the presence of a functioning growth plate. Furthermore, it may be of use in determining if two adjacent carpal or tarsal centers are fused to each other, and if less than normal, or no, growth plate activity occurs between them.

Diagnosis Several congenital malformations arise related to intrauterine position, including clubfoot, kyphoscoliosis of the tibia, manifestations of amniotic bands, and perhaps developmental dysplasia of the hip and ▶Poland syndrome. Clubfoot is a combination of varus of hindfoot and forefoot, equinus position of the ankle, retarded maturation/small size of tarsal bones, and hypoplasia (or absence) of the anterior tibial artery. The angulation relationship of the talus and calcaneus is observed especially on maximal dorsiflexion lateral images and simulated weight-bearing frontal images. Kyphoscoliosis of the tibia is a posteriorly and medially convex deformation, seen also in the fibula, which improves as a child gets older. Amniotic bands, from damage to a fetal part that was trapped through a gap in the amniotic cavity, cause constriction of bony and soft-tissue elements

393

C

Congenital Malformations, Bone. Figure 1 Poland syndrome. This infant had ipsilateral pectoral muscle hypoplasia. Note the absent middle phalanges 2–4 and hypoplastic middle phalanx 5, with some soft-tissue syndactyly between the middle three fingers. (From Oestreich AE, Crawford AH (1985) Atlas of Pediatric Orthopedic Radiology. Thieme Verlag, Stuttgart p 167)

and perhaps hypoplasia or aplasia distal to the (Streeter) band site. In Poland syndactyly (Fig. 1), middle phalanges of one hand are absent or hypoplastic, soft-tissue syndactyly occurs between involved fingers, ipsilateral ribs are small, the pectoral muscle is small on that side, sometimes with nipple hypoplasia, and a dextroposition of the heart sometimes accompanies left-sided disease. Dr. Josef Warkany suggested that the clenched fist repeatedly beating on the chest in utero may be the causative factor for the unusual combination of findings. Some congenital malformations may be acquired from intrauterine intervention or the intrauterine environment. For damage from amniocentesis, history of the event is important, as is the case for other trauma or surgery. Heavy metal burden from the mother, lead or bismuth, can yield dense metaphyseal and metaphysealequivalent bands in the fetal bones. Magnesium citrate therapy of the mother late in pregnancy yields instead similar lucent bands. Malformations in fetal alcohol syndrome include narrow distal phalanges. Variation in bone number or connectiveness results from homeobox (hox) gene variations. More complex genetic conditions have recognizable patterns: in cerebrocostomandibular syndrome, gaps appear within ribs and the mandible is small. The mandible is also small in Seckel’s syndrome and camptomelic dysplasia, as well as

394

Congenital Malformations, Cerebellar

in ▶Pierre-Robin sequence. In cleidocranial dysplasia, a portion or all of the clavicles are missing, wormian bones are present, the pubic symphysis is wide, pseudoepiphyses of the metacarpals and metatarsals are prominent, and vertebral bodies are biconvex. In Down syndrome, trisomy 21, the middle phalanx of the little finger is small, acetabular and iliac angles are low, the sternal manubrium has two centers in series, and the spinal canal behind the dens may be small. About 15–20% of normal children also have two manubrial centers in series. Dislocation of the proximal radius is seen in several genetic syndromes, including Williams syndrome and 4p- (WolfHirschhorn) syndrome. Many upper cervical column abnormalities are seen in ▶22q11.2 deletion, including fusions of C2 and C3, platybasia, unfused posterior arch of C1, and upswept posterior lamina and other posterior elements (the C2 swoosh) (3). Wide distal phalanges of the great toes and thumbs, wide tufts of distal phalanges, high incidence of azygos fissure, and low acetabular and iliac angles are findings in Rubinstein-Taybi syndrome. The ▶domed talus association (4) includes the ball-in-socket ankle, tarsal bone coalitions (seen on plain images in the second decade of life; on MRI detectable earlier), less than five rays of the feet, short fibula (and often tibia with absent superior tibial spines), and proximal focal femoral dysplasia, in various combinations. In this article, we were only able to cite a few of hundreds of malformations, which are increasingly being associated with specific genetic findings. In the equinus tibiocalcaneal relationship, often due to a tight heel cord, the front of the calcaneus is lower than normal in its relationship to the tibia, resembling the “q” in the front of the lower case word equinus that extends below the row of the other letters; in the calcaneus

tibiocalcaneal relationship, the front of the calcaneus is higher than normal, like the “l” of the lower case word calcaneus (Fig. 2). In a valgus relationship, the distal part heads further from the midline than normal compared to the proximal part, judged with the proximal part in anatomic position; in varus, the distal part heads closer to the midline than normal. Mnemonic: the varus proximal right femur looks like a lower case “r” and the valgus right femur looks like a lower case “l.”

Bibliography 1.

2. 3.

4.

Oestreich AE, Mize WA, Crawford AH et al (1987) The “anteater nose”; a direct sign of calcaneonavicular coalition on the lateral radiograph. J Pediatr Orthop 7:709–711 Oestreich AE (1981) The lateral clavicle hook-an acquired as well as a congenital anomaly. Pediatr Radiol 11:147–150 Ricchetti ET, States L et al (2004) Radiographic study of the upper cervical spine in the 22q11.2 deletion syndrome. J Bone Joint Surg Am 86-A:1751–1760 Pistola F, Ozonoff MB, Wintz P (1987) Ball-and-socket ankle joint. Skeletal Radiol 16:447–451

Congenital Malformations, Cerebellar P HILIPPE D EMAEREL Department of Radiology, University Hospitals, Leuven, Belgium [email protected]

Synonyms Congenital malformations; Hindbrain; Posterior fossa

Definition Congenital cerebellar malformations involve all malformations of the vermis and cerebellar hemispheres. They are classified on the basis of anatomical and embryological knowledge. Several genes, proteins, and molecules are involved in the various stages of cerebellar development, and it is expected that this knowledge will influence the classification of congenital cerebellar malformations in the future. Congenital Malformations, Bone. Figure 2 Calcaneus and equinus mnemonic diagram, by Tamar Kahane Oestreich, see text. (From Oestreich AE, Crawford AH (1985) Atlas of Pediatric Orthopedic Radiology. Thieme Verlag, Stuttgart p 115)

Pathology/Histopathology Although the anatomy of the cerebellum is straightforward, the understanding of embryology is still limited.


Approximately during the fourth week of gestation, the rhombencephalon divides into the metencephalon (pons and cerebellum) and the myelencephalon (medulla oblongata). The metencephalon will give rise to the vermis and the hemispheres, while the cavity in the rhombencephalon will expand to become the fourth ventricle. The roof of the fourth ventricle is divided into an anterior and a posterior membranous area; the posterior membranous area will ultimately communicate with the subarachnoid space at the foramen of Magendie. The posterior membranous area plays a role in several forms of cerebellar hypoplasia related to the ▶Dandy–Walker malformation. In the formation of the cerebellar cortex, there is first an outward migration of the Purkinje cells and granular cells. Later an inward migration takes place, forming the internal granular cell layer. The vermis has formed by the 16th week of gestation and the hemispheres 4–8 weeks later. The formation of the vermis is closely related to formation of the hemispheres. The foliation and fissuration occur at least partly simultaneously and are triggered by the migration of Purkinje and granular cells. These processes explain at least to some extent the malformations of fissuration and foliation.

Clinical Presentation The role of the cerebellum as a center for coordination, motor learning, and higher cognitive functions is well known. Cerebellar abnormalities will usually cause cerebellar signs and symptoms, such as ataxia, cranial nerve palsies, delayed language and speech development, eye movement disorders (nystagmus, oculomotor dyspraxia, and so on), and head/body turning attacks. Epilepsy and mild cognitive deterioration can also be part of the clinical spectrum. The Chiari I malformation appears to be asymptomatic in up to 14% of the patients in a large series. Magnetic resonance (MR) imaging is not useful in differentiating these patients from symptomatic patients. The typical symptom is occipital headache when coughing or otoneurological disturbances, but the presentation is often more ambiguous. Joubert syndrome is a separate entity within the molar tooth malformations and is characterized by episodic hyperpnea, abnormal eye movements, ataxia, and mental retardation.

Imaging MR imaging is the modality of choice in the diagnostic work-up of cerebellar malformations. Computed tomography is not suited to assess intrinsic

395

cerebellar abnormalities. Imaging in different planes is required, and if available, three-dimensional T2-weighted images can be helpful. A classification is not yet available today, but using knowledge of anatomy, embryology, and histopathology, a provisional classification can be presented. The two groups of cerebellar malformations are cerebellar hypoplasia and ▶cerebellar dysgenesis. Cerebellar hypoplasia includes the Dandy–Walker malformation. Cerebellar dysgenesis can further be divided into vermian and/or hemispheric dysgenesis. It seems inevitable that there will be some overlap between the two groups of malformations. The ▶Chiari malformations are not included in the classification and are discussed separately.

Diagnosis The imaging aspects of the common cerebellar malformations will be briefly reviewed. The Chiari I malformation is defined as a caudal herniation of the cerebellar tonsils over a distance of at least 5 mm through the foramen magnum (below a line from the basion to the opisthion on a sagittal image; Fig. 1a). When the herniation is less than 5 mm, the term “tonsillar ectopia” can be used. This malformation is likely to be the result of an abnormally small posterior fossa and/or the occurrence of craniovertebral anomalies. The Chiari II malformation is a complex entity thought to be due to a lack of expression of surface molecules that cause a persistent aperture of the posterior neuropore. The posterior fossa remains too small, and the cerebellum extends downward, upward, and around the brainstem. Not only the tonsils but parts of the vermis and even the fourth ventricle can herniate through the foramen magnum (Fig. 1b). A lumbar myelomeningocele is associated in more than 85% of cases. The Chiari III malformation consists of a Chiari II malformation with an associated cephalocele (Fig. 1c). The Dandy–Walker malformation is a typical example of vermian and/or hemispheric hypoplasia. It consists of (i) a variable degree of vermian agenesis, (ii) a dilatation of the fourth ventricle with cyst formation, and (iii) an enlarged posterior fossa (Fig. 2a, b). It is thought that a genetic mutation causes an abnormal development of the roof of the fourth ventricle. The posterior membranous area of the roof normally forms the foramen of Magendie. This does not happen in the Dandy–Walker malformation. The Dandy–Walker malformation can be associated with other congenital malformations and has been reported in several syndromes. In the differential diagnosis, one should consider the persisting Blake’s pouch, the result of a nonperforation of the foramen of

C

396


Congenital Malformations, Cerebellar. Figure 1 (a) Sagittal T2-weighted image, Chiari I malformation. Note the herniation of the cerebellar tonsils through the foramen magnum reaching the level of the endplate of C2. (b) Sagittal T2-weighted image, Chiari II malformation. Note the displacement of the fourth ventricle and the herniation of cerebellar tissue to the level of C3. There is tectal beaking and a concave lining of the clivus. The posterior fossa is small. (c) Sagittal T2-weighted image, Chiari III malformation. In addition to the typical finding of a Chiari II malformation, there is a craniocervical cephalocele.

Magendie (Fig. 2c). This malformation resembles megacisterna magna, but hydrocephalus is not seen in the latter. Megacisterna magna is asymptomatic but is sometimes considered a mild form of Dandy–Walker malformation. A late opening of the foramen of Magendie has been suggested as an underlying cause. Arachnoid cysts in the posterior fossa can resemble a megacisterna magna, but they usually cause symptoms and are often located laterally in the posterior fossa. Cerebellar dysgenesis can involve the vermis only, the hemispheres only, or both. The molar tooth malformations include several syndromes with vermian dysgenesis. Joubert syndrome is the best-known entity. A midline cleft is seen together with a peculiar shape of the fourth ventricle. These malformations are named after the molar tooth appearance of the

midbrain on axial images ( Fig. 3a, b). This is due to the horizontal course of the superior cerebellar peduncles (due to a lack of decussation of the superior cerebellar peduncles) and a narrow pontomesencephalic junction. Rhombencephalosynapsis consists of vermian agenesis or severe hypogenesis, fusion of the hypoplastic cerebellar hemispheres, and fusion of the dentate nuclei. Dilated lateral ventricles, fusion of the thalami, and absence of the septum pellucidum can be associated. The disorder is possibly induced by a genetic mutation of the Lmx 1a gene, which regulates events at the pontomesencephalic junction between the fourth and the sixth weeks of gestation. Because of this early timing, metencephalosynapsis might be a more appropriate term. The vermis fails to differentiate, while the hemispheres remain undivided.


397

C

Congenital Malformations, Cerebellar. Figure 2 (a) Sagittal T2-weighted image, Dandy–Walker malformation. The posterior fossa is enlarged. There is an upward rotation of the remaining part of the superior vermis, and there is a large posterior fossa cyst. Note the displacement of the sinus rectus and the hypogenesis of the corpus callosum. (b) Axial T2-weighted image, Dandy–Walker malformation. There is a large communication between the fourth ventricle and the posterior fossa cyst. The hypoplastic cerebellar hemispheres appear “winged outward.” (c) Sagittal T2-weighted image, persisting Blake’s pouch. There is a large communication between the fourth ventricle and the subarachnoid space, and there is supratentorial hydrocephalus. The vermis is reasonably developed.

Recently, a series of abnormalities have been reported consisting of vermian and/or hemispheric dysgenesis. When only the vermian fissures are involved, this is almost always an asymptomatic incidental finding (type 1a; Fig. 3c). Some patients also have an abnormal foliation of the anterior lobe and part of the posterior lobe of the vermis (type 1b). Hemispheric extension can occasionally be encountered. Cerebellar signs are frequently seen in this subgroup. In abnormalities of foliation and fissuration type 2, three different hemispheric malformations can be recognized: (i) cortical dysgenesis, (ii) cortical hypertrophy, and (iii) aberrant orientation of the folia. These

abnormalities can be seen together or separately. Cortical dysgenesis can be associated with small cyst-like inclusions (Fig. 3d, e). Similar abnormalities have been described in rare syndromes, including Fukuyama muscular dystrophy and Walker–Warburg syndrome, but they can also be observed in the absence of these syndromes. In 60% of the children with type 2 abnormalities, type 1b abnormalities are present, and more than 50% of these children have congenital cerebral anomalies. The pathogenesis of Lhermitte–Duclos–Cowden remains incompletely elucidated. Although the lesion is classified as a dysplastic gangliocytoma according to

398


Congenital Malformations, Cerebellar. Figure 3 (a) Axial T2-weighted image, molar tooth malformation. Note the molar tooth appearance due to the horizontal course of the hypoplastic superior cerebellar peduncles. (b) Axial T2-weighted image, molar tooth malformation. The vermian cleft is seen with the “bat wing” appearance of the fourth ventricle. (c) Coronal T2-weighted image, type 1 abnormality of fissuration. The abnormal orientation of the fissures is noted. (d) Axial T2-weighted image, type 2 abnormality of foliation and fissuration. An abnormal foliation is seen, with multiple cyst-like inclusions in the cortex. (e) Coronal T2-weighted image, type 2 abnormality of foliation and fissuration. The cortical dysgenesis is clearly visible in the left hypoplastic cerebellar hemisphere compared with the normal right hemisphere. (f) Axial T2-weighted image, Lhermitte–Duclos–Cowden disorder. A masslike lesion is observed in the right cerebellar hemisphere with a striated folial pattern.

Congenital Malformations, Cerebral (Neuro View)

the World Health Organization, there are indications that it might represent a congenital malformation. The imaging appearance is that of a masslike abnormality with, to some extent, a striated pattern of the enlarged folia (Fig. 3f).

Bibliography 1.

2. 3.

Barkovich AJ (2005) Pediatric Neuroimaging. 4th edn. Lippincott Williams & Wilkins Philadelphia, Baltimore, New York, London, Buenos Aires, Hong Kong, Sydney, Tokyo Tortori-Donati P, Rossi A (2005) Pediatric Neuroradiology. Springer, Berlin, Heidelberg, New York Demaerel P (2002) Abnormalities of cerebellar foliation and fissuration: classification, neurogenetics and clinicoradiological correlations. Neuroradiology 44:639–646

Congenital Malformations, Cerebral (Neuro View) T HIERRY T. A. 1 , V ERONICA D ONOGHUE 2 1

Johns Hopkins University, Baltimore, USA Children’s University Hospital, Dublin, Ireland [email protected] [email protected]

2

Synonyms Developmental cerebral anomalies; Disorders of neurulation

Definition Congenital cerebral malformations result from an erroneous organogenesis, histiogenesis or cytogenesis of the central nervous system (CNS). The aetiology is multifactorial and includes genetic defects, intrauterine destructive events, ischaemia, infections and environmental agents (1–3).

Pathology and Histopathology Malformations of the CNS are frequent (1:100 births) and may have a significant impact on the quality of life. Cerebral malformations encompass a wide variety of lesions ranging from tiny cortical dysplasias to extensive, highly complex pathologies. The development of the brain is a highly complex, programmed sequence of interacting processes in which

399

up to 100 billion neurons connect and communicate with each other using countless synapses. While the embryogenesis is most active during the first trimesters of gestation, maturational processes may extend into the second decade of life. The exact anatomical, maturational and functional development of the brain remains the focus of countless research projects. Classical cerebral malformations should be distinguished from acquired cerebral malformations. Classical malformations result from an erroneous development of the brain which may be genetically encoded while acquired malformations relate to secondary malformations due to a destruction of initially correctly developed brain structures. A significant overlap between these categories is present. Identical malformations may have different etiologies (e.g. infection, focal hypoperfusion or haemorrhage). In addition, identical ‘destructive events’ (e.g. focal hypoperfusion due to an arterial embolus) can result in a spectrum of pathologies. The timing of the ‘event’ in relation to the gestational age is a key feature. Moreover, an increasing number of gene defects are identified in congenital and so-called acquired cerebral malformations. Variable expressions of genetically mediated malformations may result in a spectrum of malformations. Finally, it is increasingly recognized that a predisposition for the development of malformations may be encoded in the chromosomes, a trigger is however necessary to induce the cascade of events. Cerebral malformations are typically classified into disorders of (a) organogenesis, (b) histiogenesis and (c) cytogenesis. In disorders of organogenesis an altered brain development is combined with a normal histiogenesis while in disorders of histiogenesis, the overall brain structure is normal however anomalous cells persist and continue to differentiate. Inborn errors of metabolism or leukodystrophies represent disorders of cytogenesis.

Clinical Presentation The spectrum of neurological symptoms varies widely and is often unpredictable. The extent of malformation correlates inconsistently with the severity of neurological symptoms or deficits. Large lesions may be discovered as incidental findings on neuroimaging. Neuronal and functional plasticity of the developing brain with relocation of functional centre may prevent overt neurological deficits. On the other hand, small lesions may present with significant neurological symptoms, e.g. a discrete cortical dysplasia may be causative for intractable seizures preventing normal development. Next to focal neurological deficits, many malformations result in cognitive deficits with different degrees of mental retardation.

C

400


Posterior fossa malformations may be present with ataxia. In combined brainstem and cerebellar malformations (e.g. Joubert syndrome) brainstem symptoms may accompany cerebellar symptoms. In callosal malformations and disorders of diverticulation hypothalamic– pituitary malfunction and facial malformations — ‘the face predicts the brain’ — occur.

Imaging Frequent disorders of organogenesis include posterior fossa malformations, callosal or commissural malformations. Disorders of diverticulation or cleavage disorders are less frequently encountered. Malformations of cortical development are described in the chapter on gyration disorders. Ultrasound (US) may serve as a first line imaging tool in neonates. Magnetic resonance imaging is applied when US or CT suspected complex malformations.

Posterior Fossa Malformations Chiari I Malformation. The cerebellar tonsils herniate into the upper cervical spinal canal (>5 mm below the level of the foramen magnum). The tonsils may be compressed/ deformed. Associated syringohydromyelia is seen in 20– 25% of cases. There is no increased risk for additional cerebral malformations. Chiari II Malformation. Complex malformation most probably resulting from too small a posterior fossa in patients with open neural tube defects. The spectrum of findings include varying degrees of downward displacement of the cerebellum with herniation of cerebellar structures into the upper cervical spinal canal, kinking of the upper cervical spinal cord, embracement of the brainstem by the cerebellar hemispheres, tectal plate deformation, supratentorial hydrocephalus, prominent interthalamic adhesion and fenestrations of the falx cerebri with interdigitation of the cerebral hemispheres. Frequently an associated callosal dysgenesis is observed (Fig. 1). Chiari III Malformation. Very rare malformation of the posterior fossa represented by the spectrum of findings encountered in Chiari II malformation in combination with a low occipital or high cervical meningo-encephalocele. Rhomboencephalosynapsis. Single lobed cerebellum with vermian agenesis, fusion of both cerebellar hemispheres, dentate nuclei and superior cerebellar peduncles. The cerebellar fissures typically course over the entire cerebellar surface without interruption. Additional malformations are frequent. Joubert syndrome. Congenital vermian hypoplasia or aplasia. The IV ventricle is deformed and resembles a batwing or umbrella while the brainstem resembles a molar tooth on axial imaging. Associated ocular and renal lesions are known.

Cystic Posterior Fossa Malformations Dandy Walker Malformation. Cystic dilatation of the IV ventricle in combination with varying degrees of vermian hypoplasia. The posterior fossa may be normal in size or enlarged, the torcula is frequently elevated and the vermian remnant rotated upward. The brainstem may be compressed against the clivus. The choroid plexus within the IV ventricle is usually absent, the tentorium cerebelli hypoplastic. Associated malformations are seen in up to 60% of patients, most frequently including callosal dysgenesis, polymicrogyria, cephaloceles and ventricular dilatation. The wide open IV ventricle often resembles a keyhole (Fig. 2). Dandy Walker malformations are believed to result from a combined defective development of the velum medullare anterior and posterior. Blake’s Pouch Cyst. Outpouching of the velum medullare posterior most probably resulting from an incomplete or absent rupture of the velum medullare posterior during development. The posterior fossa may be enlarged, the torcula elevated. The vermis and falx cerebelli is normally developed, the IV ventricle choroid plexus may be displaced dorsally. Associated cerebral malformations are rare. Retrocerebellar Archnoid Cyst and Mega Cisterna Magna. The differentiation between an arachnoid cyst and a mega cisterna magna can be difficult. The posterior fossa may be enlarged the cerebellar hemispheres and vermis can be displaced. The IV ventricle choroid plexus is in its normal location. Associated cerebral malformations are rare.

Callosal or Commissural Malformations In the evaluation of callosal malformations it should be noted that the normal corpus callosum (CC) has a wide variability in size and shape. In addition, it is essential to determine if the encountered callosal anomaly is a primary malformation or a secondary anomaly in shape due to e.g. an adjacent white matter ischemia with focal callosal atrophy or a callosal thinning in chronic severe hydrocephalus. The CC develops in a programmed sequence (genu, truncus, splenium, rostrum). Callosal agenesis may be partial or complete. Associated cerebral malformations are seen in 50% of cases. On sagittal images the defective CC is easily identified. In addition, the medial surface of the cerebral hemisphere reveals a radiating pattern of sulci due to the missing cingulate sulcus (no inversion of the cingulate gyrus). On coronal images, the III ventricle may extend superiorly between the hemispheres. The lateral ventricles are displaced laterally and show a trident shape due to the impression by the Probst bundles. The hippocampi may be malrotated. On axial images, the lateral ventricles show


401

C

Congenital Malformations, Cerebral (Neuro View). Figure 1 (a) Chiari II malformation: Sagittal, and coronal T2-FSE MRI reveals a small posterior fossa with herniation of the cerebellar tonsils into the upper cervical spinal canal, displacement and compression of the brainstem against the clivus and supratentorial hydrocephalus. (b) Chiari II malformation: Axial and coronal T2-FSE MRI shows a characteristic embracement of the brainstem by the cerebellar hemispheres, a prominent adhesion interthalamica and a fenestration of the falx cerebri with interdigitations of both cerebral hemispheres.

a parallel course, the occipital horns may be enlarged (Fig. 3). Occasionally, an associated interhemispheric lipoma is encountered. Callosal agenesis is frequently only one component of a more extensive commissural malformation. Consequently, in callosal anomalies, the anterior and posterior commissure as well as the fornix and hippocampi should be studied.

Disorders of Diverticulation or Cleavage Disorders These anomalies are also known as disorders of ventral induction. They include the holoprosencephaly spectrum. An increasing number of gene defects are identified in

holoprosencephaly. Basically, holoprosencephaly results from an incomplete or absent cleavage of the prosencephalon. The spectrum of malformation depends on the degree of cleavage. Alobar Holoprosencephaly. Most severe form characterized by a small holosphere, monoventricle, fused thalami, missing III ventricle, no falx or CC and no interhemispheric fissure. Frequently, associated malformations of the circle of Willis are encountered. Semilobar Holoprosencephaly. Partial cleavage of the prosencephalon. The anterior parts of the brain are fused while the posterior brain is split. The thalami are partially separated, a small III ventricle is present, rudimentary temporal horns are seen. The falx cerebri

402


Congenital Malformations, Cerebral (Neuro View). Figure 2 (a) Dandy Walker malformation: Sagittal, axial and coronal T2-FSE MRI reveal a classical Dandy Walker malformations with a cystic dilatation of the IV ventricle, in combination with a hypoplastic, upward rotated superior vermis. No IV ventricle choroid plexu is seen. The brainstem is hypoplastic. On axial images the IV ventricle mimics a key hole. In addition, a supratentorial hydrocephalus is seen. (b) Dandy Walker malformation: Sagittal and coronal T2-FSE MRI may identify a variety of coexisting malformations like e.g. a hypoplastic corpus callosum or a hypoplastic olfactory bulb.


403

C

Congenital Malformations, Cerebral (Neuro View). Figure 3 Callosal dysgenesis: Coronal US identifies the characteristic trident shape of the lateral ventricles due to callosal dysgenesis. Coronal, sagittal, and axial T2-FSE reveal the characteristic findings of a callosal dysgenesis including an enlarged, interhemispheric extending III ventricle, a radial gyral pattern of the medial brain surface, a trident shape of the lateral ventricles on coronal images, a parallel course of the lateral ventricles on axial images and T2-hypointense Probst bundles along the medial surface of the lateral ventricles. Tractograhy reconstructions based on diffusion tensor imaging confirms the anterior–posterior course of the Probst bundles.

and interhemispheric fissure is present in the dorsal parts of the brain. This is the only callosal malformation where the dorsal parts of the CC may be present without the anterior part. The optic nerves and olfactory bulbs are frequently hypoplastic. Lobar Holoprosencephaly. Near complete cleavage of the prosencephalon results in a lobar brain with hypoplastic frontal lobes and rudimentary frontal horns. The falx cerebri extends frontally, the temporal horns are formed. Septo-Optic Dysplasia. Septo-optic dysplasia is considered to be the mildest variant of lobar holoprosencephaly characterized by an absent septum pellucidum, hypoplastic optic nerves and chiasm. The olfactory bulbs are frequently hypoplastic. Associated cerebral malformations are seen in up to 60% of cases.

and degree of cerebral malformation is essential to guide treatment, may give information concerning prognosis and outcome and should be used for counselling of future pregnancies. Definite diagnosis consequently relies on high resolution triplanar neuroimaging. MRI has proven to be most reliable. With the advance of functional MRI techniques, such as diffusion tensor imaging and tractography (Fig. 3) detailed structural-functional informations can be collected. In the evaluation of malformations of the central nervous system, each radiologist should be aware that if they find one malformation they should look for additional malformations. Frequently, the most obvious lesion is just the tip of the iceberg. A proper knowledge of the complex embryology and maturation of the brain is mandatory.

Diagnosis Bibliography Ultrasound may serve as a first line imaging tool in neonates. Especially when a hydrocephalus is suspected, ultrasound is highly sensitive and may show additional gross parenchymal abnormalities next to the ventricular enlargement. A complete identification of the exact extent

1. 2. 3.

Barkovich AJ (2000) Pediatric neuroimaging, 3rd edn. Lippincott Williams & Wilkens, Philadelphia, pp 251–381 Osborn AG (2004) Diagnostic imaging: brain. Amirsys, Salt Lake City, pp I-1–4–78 Tortori-Donati P (2005) Pediatric Neuroradiology: Brain. Springer Verlag, Berlin, Heidelberg, pp 41–198

404

Congenital Malformations, Cerebrum

Congenital Malformations, Cerebrum P IETER WAER , P HILIPPE D EMAEREL Dept of Radiology, University Hospitals K.U. Leuven, Leuven, Belgium [email protected]

Synonyms Anomalies of the cerebral commissures; Cephalocele; Dorsal induction anomalies; Neuronal migration disorders; Ventral induction anomalies

Definition Depending on the time of insult, different malformations may occur during cerebral development. Often, this will be multifocal involving several brain structures. It is important to realize that most malformations can display mild, moderate, or extensive changes and that different malformations often occur simultaneously. Cortical developmental malformations are by far the most common cerebral malformations. Terminology has not always been straightforward, but nowadays a good and flexible classification for the cortical developmental malformations is available which allows the inclusion of new observations (1).

In holoprosencephaly there is a failure of cleavage into hemispheres and sometimes even a failure of separation of the telencephalon from the diencephalon. The most commonly encountered cerebral malformation is the cortical developmental malformation. The formation of the cerebral cortex begins during the seventh week of gestation with the period of neuronal and glial proliferation in the germinal matrix. The second period consists of a migration of the neurons from the germinal zone along the so-called guiding radial glial fibers. This process takes place between the eighth and twenty-fourth week of gestation. The six-layered cortex can be recognized in the sixteenth fetal week. The thickness does not exceed 4.5 mm. The period of migration is followed by cortical organization, leading to the complex threedimensional structure of the mature cortex. Up to the eighteenth week, the hemispheric surface is smooth. At this time, the callosal and parieto-occipital sulci and the calcarine fissure are visible. From 20 weeks onwards, a progressive sulcation can be observed until the thirty-second to thirty-fourth week of gestation.

Clinical Presentation Minor congenital cerebral malformations can remain asymptomatic. However, often developmental delay, mental retardation, spasticity, hypotonia, epilepsy, hemiplegia, or diplegia are seen in children with a cerebral malformation. Macro- or microcephaly can also be a sign of an abnormal development. When the cerebral malformations are part of a syndrome, extracranial anomalies can be encountered.

Pathology/Histopathology Imaging A brief review of the embryology of the brain is necessary in the understanding of the pathology of cerebral malformations (2, 3). Although the cause of cephaloceles remains not fully elucidated, it is believed they may result from a failure of neural tube closure. Other authors have suggested that they might be due to an event occurring after the closure of the neural tube. The anterior part will form the prosencephalon, mesencephalon, and rhombencephalon at around 30 days of gestation. At this time, the corpus callosum will develop from the commissural plate at the rostral end of the neural tube. The genu of the corpus callosum is formed before the truncus and the splenium. At the end, a more anterior part of the genu and the rostrum become visible. This sequence of formation will be important in analyzing partial agenesis of the corpus callosum. Remnants of the meninx primitiva give rise to the interhemispheric lipomas.

In imaging cerebral malformations, one should always obtain T1- and T2-weighted images, because the ongoing myelination may mask possible malformations during the first 2 years of life on one of these sequences. Additional three-dimensional imaging, using either a T1-weighted gradient-echo sequence or a T2-weighted sequence, will be extremely helpful in displaying the abnormalities.

Diagnosis The list of cerebral malformations is very extensive and detailed. Here, only the most common malformations are discussed. Skull and skull base defects are often referred to as “cephaloceles.” The defect can contain meninges, brain tissue, and/or cerebrospinal fluid.


The terms hypogenesis or complete agenesis are preferably used for partial or complete absence of the corpus callosum. In hypogenesis, the anterior part of the corpus callosum is usually present. The imaging findings of agenesis are quite specific. A radial pattern can be seen on sagittal images (Fig. 1). This is due to the eversion of the cingulate gyri and radial orientation of the medial hemispheric sulci from the region of the

Congenital Malformations, Cerebrum. Figure 1 Agenesis of the corpus callosum. Sagittal T1-weighted image. The typical radial pattern on a mid-sagittal image is due to the radial orientation of the medial hemispheric sulci. The structure visualized is an enlarged hippocampal commissure and does not correspond to the genes of the corpus callosum.

405

roof of the third ventricle. Colpocephaly, corresponding to the dilatation of the trigones and posterior horns of the lateral ventricles, is typically seen on axial images. The third ventricle is widened and extends superiorly. Interhemispheric cysts can be found, and different subtypes have been recognized. Interhemispheric lipomas have frequently been described in association with partial or complete absence of the corpus callosum (4). A curvilinear type of lipoma often appears to be an incidental finding, while the tubulonodular lipoma is usually associated with a more severe abnormality of the corpus callosum and tends to be symptomatic. Alobar, semilobar, and lobar holoprosencephaly have been distinguished, but significant overlap exists between the different types. The septum pellucidum and the olfactory bulb are always absent in the three subtypes. The lobar form is the least severe from the clinical point of view and the abnormalities can sometimes be very subtle. This latter form is also closely related to septo-optic dysplasia. Septo-optic dysplasia or de Morsier’s syndrome consists of hypoplasia of the optic nerves and chiasm with absence of the septum pellucidum (Fig. 2). Magnetic resonance (MR) imaging has been used to further classify patients with optic nerve hypoplasia. In 25% of the patients, no associated abnormalities could be demonstrated. The septum pellucidum was absent in 20% of the children. An ectopic neurohypophysis was observed in 15% of the patients. An associated cortical developmental anomaly was found in 15% of the patients with optic nerve hypoplasia. There was evidence of antenatal or prenatal brain injury in the remaining 25% of the children. Three groups of cortical developmental malformations are distinguished: (i) abnormalities of neuronal and glial proliferation, (ii) abnormalities of neuronal migration,

Congenital Malformations, Cerebrum. Figure 2 Septo-optic dysplasia. Sagittal T1-weighted (a) and coronal T2-weighted (b) images. The optic chiasm is hypoplastic (a). The septum pellucidum is absent (b).

C

406


and (iii) abnormalities of cortical organization. These malformations can result from several environmental factors including maternal exposure to ethanol and radiation, ischemia, or in utero infection, but an increasing number of genetic causes are being identified (5, 6). The abnormalities of neuronal proliferation occur in the early days of cortical development and they often show a high signal on T2-weighted images. Focal cortical dysplasia, first described by Taylor, consists of a focal disorganized cortex with abnormally large neurons, called balloon cells because of their histopathological appearance. On imaging the cortex may appear normal or thinner, usually with blurring of the gray–white matter interface, a useful sign for differentiating focal cortical dysplasia from polymicrogyria (Fig. 3). The high signal on T2-weighted images may be due to gliosis and/or the presence of balloon cells. Hemimegalencephaly may involve part of one hemisphere or the whole hemisphere and hemicranium.

The affected area appears strongly enlarged and has heterogeneous signals reflecting dysmyelination and abnormalities of proliferation, migration, and cortical organization. The abnormalities may undergo changes over time. The abnormalities of neuronal migration occur during the migration of the neurons from the germinal matrix to their final destination, the cortex. They typically have a signal similar to that of gray matter. Lissencephaly is also referred to as agyria-pachygyria complex. Some patients have been shown to have chromosomal mutations, the best known clinical entity being the autosomal dominant form of the Miller–Dieker syndrome. A variable involvement of the cortex can be observed, usually correlating with the severity of the symptoms. In pachygyria the cortex is thickened and the sulci are incompletely formed (Fig. 4). The junction between gray and white matter is regular as opposed to the junction in polymicrogyria.

Congenital Malformations, Cerebrum. Figure 3 Focal cortical dysplasia of Taylor. Coronal T2-weighted (a) and fluid-attenuated inversion recovery (b) images. The abnormalities in the cortex and in the subcortical white matter of the right parietal lobe are better seen on the second image (b).

Congenital Malformations, Cerebrum. Figure 4 Pachygyria. Axial T2-weighted (a, b) images. Bilateral focal pachygyria in the parietal lobe (a) and more extensive pachygyria in a 3-month-old infant (b).

Congenital Malformations, Liver and Biliary Tract

Heterotopic gray matter can be located in the subependymal region, in the subcortical region, or in between at any point along the route of migration from the subependymal region to the subcortex (Fig. 5). Heterotopia refers to normal structures in an abnormal site. Later during gestation, malformations of cortical organization can be observed. Polymicrogyria refers to an abnormal gyration consisting of an increased number of gyri. Cortical maldevelopment is due to a process affecting mainly the deep cortical layers, reducing vascularization

407

and arresting growth. Excessive proliferation of the outer layers results in the increased number of gyri. Whereas in the past it was debatable whether polymicrogyria could be diagnosed on MR imaging or only histopathologically, it is now accepted that the diagnosis can be given on imaging (Fig. 6). The inner border is sharp but irregular, which allows the differential diagnosis with pachygyria to be made, where the inner border is sharp but smooth. Bilateral perisylvian polymicrogyria is a well-recognized entity. Usually the opercular and perisylvian cortex are involved. The term schizencephaly merely refers to an abnormal cleft. The cleft is lined by a polymicrogyric cortex and extends from the subependymal region to the pia. Bilateral and unilateral schizencephaly can be observed, most often in the perisylvian area. The terms open and closed lip schizencephaly refer to the separated and adjacent walls of the cleft.

Bibliography 1.

2.

3.

Congenital Malformations, Cerebrum. Figure 5 Subependymal heterotopia. Axial T2-weighted image. Heterotopic gray matter nodules are seen in the subependymal region on the left side.

4.

5. 6.

Barkovich AJ, Kuzniecky RI, Dobyns WB et al (1996) Classification scheme for malformations of cortical development. Neuropediatrics 27:59–63 Barkovich AJ (1996) Congenital malformations of the brain and skull. In: Barkovich AJ (ed) Pediatric Neuroimaging. 2nd edn. Lippincott-Raven Publishers, Philadelphia, pp 177–276 Sidman RL, Rakic P (1982) Development of the human central nervous system. In: Haymaker W, Adams RD (eds) Histology and Histopathology of the Central Nervous System. Thomas, Springfield, pp 3–145 Vandegaer Ph, Demaerel Ph, Wilms G et al (1996) Interhemispheric lipoma with variable callosal dysgenesis: relationship between embryology, morphology, and symptomatology. Eur radiol 6:904–909 Barkovich AJ (2002) Magnetic resonance imaging: role in the understanding of cerebral malformations. Brain Dev 24:2–12 Blaser SI, Jay V (2002) Disorders of cortical formation: radiologicpathologic correlation. Neurosurg Clin N Am 13:41–62

Congenital Malformations, Liver and Biliary Tract H UISMAN TA 1 , PAOLA VAGLI 2 1

Johns Hopkins University, Baltimore, USA Department of Diagnostic and Interventional Radiology, University of Pisa, Italy [email protected] [email protected]

2

Congenital Malformations, Cerebrum. Figure 6 Polymicrogyria. Axial T2-weighted image. Numerous small abnormal gyri are seen in the left parietal lobe. Note the difference with pachygyria.

Synonyms Congenital hepato-biliary anomalies; Hepato-biliary malformations

C

408


Definition Congenital anomalies of the liver and biliary system present early in life and may result from hereditary or developmental errors.

Introduction Congenital hepato-biliary anomalies include a wide spectrum of malformations involving the liver, the biliary tree, and the gallbladder. Children may be asymptomatic or suffer from significant disease, e.g., neonatal jaundice. The most frequent congenital anomalies include ▶biliary atresia, ▶Alagille syndrome, ▶choledochal cysts, ▶Caroli’s disease, congenital hepatic fibrosis, ▶polycystic liver disease, and ▶gallbladder anomalies. In neonates with unexplained persisting jaundice hepato-biliary anomalies should be excluded.

Pathology and Histopathology Biliary Atresia Persisting neonatal jaundice results in one-third of cases from biliary atresia. Biliary atresia is characterized by a (sub-) total obliteration of the extrahepatic biliary system. Fetal cholangitis due to a malformed biliary tree is believed to be the causative. To prevent permanent hepatic damage, early surgical correction is mandatory. Biliary atresia is associated with situs inversus, congenital heart disease, vascular anomaly, trisomy 17, 18, 21, and polysplenia syndrome.

Alagille Syndrome Autosomal dominant disorder with chronic cholestasis is due to a hypoplasia of interlobular bile ducts. Associated congenital abnormalities include an abnormal facies, ocular anomalies, butterfly vertebrae, and complex cardiac malformations. Second most common cause of cholestasis in neonates. The incidence is 1 per 100,000 live births. Liver biopsy confirms diagnosis.

Choledochal Cysts Choledochal cysts encompass a spectrum of common bile duct dilatations that may present with cholestatic jaundice or recurrent cholangitis. The different forms have been classified by Todani. Choledochal cyst type I is the most frequent form (80–90% of cases) characterized by a focal dilatation of the common bile duct. Type I choledochal cysts most probably results from an anomalous proximal insertion of the pancreatic duct into the common bile duct. Reflux of pancreatic enzymes will

induce a localized cholangitis with resulting obstruction of the distal common bile duct and proximal dilatation. Type II choledochal cysts are characterized by focal diverticula of the common bile duct. Type III choledochal cysts are basically ▶choledochoceles that protrude into the duodenal lumen. Todani’s type IVA shows multiple intra- and extrahepatic cysts while type IVB shows multiple extrahepatic cystic dilatations of the common bile duct. Todani’s type V, also known as Caroli’s disease, is characterized by segmental nonobstructive dilatations of the intrahepatic bile ducts. This form is associated with stone formation, recurrent cholangitis, and hepatic abscesses. Hepatic fibrosis may develop. Caroli’s disease is frequently associated with medullary sponge kidneys, polycystic kidney disease, and nephrothisis. Portal hypertension may develop in disease progress.

Congenital Hepatic Fibrosis Autosomal recessive disorder occurs with or without associated biliary duct ectasia and infantile or adult polycystic kidney disease. Periportal fibrosis and the presence of irregularly shaped, partially cystic, dilated small bile ducts are the primary pathologic findings. During the course of the disease, hepatomegaly with signs of portal hypertension develops (1).

Polycystic Liver Disease Two forms are described: the infantile type and the adult type. The infantile type is less common and has an autosomal recessive inheritance while the adult type is autosomal dominant. Children usually present with renal disease are due to the associated polycystic kidney disease. Hepato-biliary cysts may be intrahepatic and/or peribiliary near portal triads or the hepatic hilum.

Gallbladder Anomalies The gallbladder may be congenitally absent, ectopically located, duplicated, or septated. An absent gallbladder may be an associated finding in polysplenia syndrome. Ectopically located gallbladders are seen within the left hemiabdomen (e.g., situs inversus), in the midline (e.g., asplenia syndrome), intrahepatic, within the abdominal wall, in the falciform ligament ,and in the retroperitoneal area.

Clinical Presentation Biliary Atresia Clinical symptoms are related to the neonatal cholestasis. Key findings include jaundice, dark urine, and pale stools. Hepatomegaly may present early, splenomegaly is common


on follow-up due to liver cirrhosis and portal hypertension. Conjugated hyperbilirubinemia and increased levels of alkaline phosphatase, gamma glutamyl transpeptidase and serum aminotransferases may be present.

409

intrahepatic bile ducts are usually unremarkable. Because in a minority of affected children, US will show a gallbladder, hepato-biliary scintigraphy is recommended as second imaging modality to rule out biliary atresia. Endoscopic retrograde cholangiopancreatography (ERCP) can be used to confirm diagnosis.

Alagille Syndrome The combination of a chronic cholestasis, abnormal facies, ocular abnormalities, butterfly vertebrae, and complex cardiac anomalies are highly suggestive for Alagille syndrome. Most patients if not treated by means of orthotopic liver transplantation will die before the third decade.

Choledochal Cysts In young children, the lead feature is chronic cholestatic jaundice while clinical presentation in older children may include abdominal pain, obstructive jaundice, and fever. In addition, acholic stools, hepatomegaly, and intermittent episodes of cholangitis and pancreatitis are seen. In untreated cases, recurrent ductal inflammation and biliary stasis may be complicated by hepatic abscesses, cirrhosis, portal hypertension, and cholelithiasis. Laboratory findings are nonspecific. Surgical treatment with complete excision of the cyst and creation of a biliodigestive anastomosis is recommended.

Congenital Hepatic Fibrosis Principal clinical symptoms are related to the associated renal abnormalities. Initially, liver function may remain preserved, on follow-up hepatomegaly and portal hypertension may develop. Complicating cholangitis occurs.

C Alagille Syndrome Imaging features in the newborn period are similar to those in biliary atresia but the identification of the additional components of this syndrome (hypertelorism, butterfly vertebrae, complex cardiac malformations) lead to the correct diagnosis. US of the liver can show hepatomegaly, periportal fibrosis/cirrhosis, and splenomegaly. In less severe forms, US will show a normal liver. Magnetic resonance cholangiopancreatography (MRCP) and ERCP or intraoperative cholangiography may show the patency of the extrahepatic biliary tract.

Choledochal Cysts Most choledochal cysts are easily identified by US. In addition, sludge and gallstones within the cysts can be seen. US is also helpful in diagnosing complications such as intrahepatic biliary dilatation, portal vein thrombosis, gallbladder or biliary neoplasms, pancreatitis, and hepatic abscesses. On computed tomography (CT), choledochal cysts appear as well-circumscribed hypodense cystic lesions which are separated from the gallbladder. The cyst wall can be thickened due to recurrent cholangitis. On magnetic resonance imaging (MR), choledochal cysts are T2-hyperintense (Fig 1–3). MRCP or ERCP are helpful in classifying the cysts.

Polycystic Liver Disease Most children initially present because of the associated renal disease. Clinical symptoms that are related to the liver include hepatic enlargement by the multiple cysts with compression of adjacent organs as well as the intrinsic bile ducts. In addition, cysts may rupture or hemorrhage. In late stages, hepatic fibrosis may result in portal hypertension.

Gallbladder Anomalies Gallbladder anomalies are usually asymptomatic and incidentally discovered. Clinical symptoms are related to gallstone formation (2–3).

Congenital Hepatic Fibrosis US shows an enlarged hyperechoic liver. Intrahepatic bile duct ectasia and gallbladder enlargement may be seen in case of an associated Caroli’s disease. Splenomegaly may result from portal hypertension. Color Doppler sonography should be performed to rule out cavernous transformation. CT and MR show an inhomogeneous pattern with areas of heterogeneously increased or decreased attenuation values or signal intensities.

Polycystic Liver Disease

Imaging Biliary Atresia In the majority of children, ultrasound (US) will fail to identify the gallbladder and extrahepatic bile ducts. The

US reveals multiple, variably sized hypoechoic hepatic cysts. Cysts may show an internal echo due to hemorrhage. The cysts are hypodense on CT, well circumscribed and nonenhancing. On MR, the cysts are T1-hypo and T2-hyperintense without enhancement.

410


Nuclear Medicine Biliary Atresia 99m

Tc labeled iminodiacetic acid derivates are very sensitive in diagnosing biliary atresia. In case of biliary atresia the liver will show a tracer uptake without excretion into the extrahepatic bile ducts or bowel. Conversely, evidence of intestinal excretion of the radiotracer confirms the patency of the extrahepatic biliary system. The absence of bowel activity after 24 h is an indication for liver biopsy.

Alagille Syndrome Congenital Malformations, Liver and Biliary Tract. Figure 1 US scan displays the presence of anechoic saccular dilatations of the bile ducts involving both lobes containing intraluminal bulbar protrusions and cross-bridges. The surrounding liver parenchyma shows a normal echostructure.

Typically, biliary imaging performed with 99mTc labeled iminodiacetic acid derivates, fails to show normal excretion of radioisotope into the bowel resembling biliary atresia.

Choledochal Cysts 99m

Tc labeled iminodiacetic acid derivates scintigraphy may show the absence of drainage into the intestine in case of complete obstruction of the distal bile duct or the presence of a dilated duct acting as a collecting reservoir (2–4).

Diagnosis Biliary Atresia

Congenital Malformations, Liver and Biliary Tract. Figure 2 Unenhanced CT scan shows marked intrahepatic ductal dilatation with cystic appearance (left). Enhancing central fibrovascular bundles (central dot sign) are identified in many of dilated ducts. These intraluminal dots correspond to intraluminal portal veins, indicating portal radicles surrounded by dilated intrahepatic bile ducts (right).

Diagnosis of biliary atresia emerges from recognizing clinical features and imaging findings. US represents the first choice but a normal examination does not exclude diagnosis. Neonatal hepatitis is the number one differential diagnosis. Typically in neonatal hepatitis, US will show a normal gallbladder. Scintigraphy may show a poor tracer uptake in combination with a delayed excretion into the bowel.

Alagille Syndrome Definitive diagnosis is usually achieved by liver biopsy. The differentiation from biliary atresia can be made by demonstration of an intact extrahepatic biliary tree.

Gallbladder Anomalies MRCP and US may identify and depict the whole spectrum of these anomalies. In case of an absent gallbladder an ectopic location or an atrophic gallbladder should be excluded. The spectrum of anomalies can ideally be depicted by MRCP (1–5).

Choledochal Cysts US findings are diagnostic in many patients, however, in the preoperative setting, complementary studies, such as ERCP, CT, or MR/MRCP may be helpful. Choledochal cyst may resemble hepatic cyst, hepatic abscess, pancreatic pseudocyst, or gallbladder duplications.

Congenital Malformations, Musculoskeletal System

411

C

Congenital Malformations, Liver and Biliary Tract. Figure 3 Axial T2-weighted MR image shows multiple hyperintense cystic ectasias involving the entire liver (left). MR cholangiogram shows a connection between the saccular dilated segmental bile ducts and the central biliary tree (middle). The cystic dilatations communicate with the major biliary tree, and focal narrowing of the major intrahepatic and common bile ducts are depicted by reformatted images (right).

Congenital Hepatic Fibrosis This affection should be suspected in young patients with portal hypertension of unknown origin. Evidence of nephromegaly and renal increased echogenity with polycystic changes support diagnosis. The final diagnosis is dependent on histological findings (biopsy).

Polycystic Liver Disease The presence of multiple congenital liver cysts in patients with renal disorders suggests diagnosis.

Gallbladder Anomalies These anomalies are usually an incidental finding on cross-sectional imaging. Most cases are asymptomatic. US may depict the hepato-biliary system anomaly and associated complications. MRCP provides a precise evaluation of the anatomy of the lesions (1–5).

Bibliography 1. 2.

3.

4. 5.

Zeitoun D, Brancatelli G, Colombat M et al (2004) Congenital hepatic fibrosis: CT findings in 18 adults. Radiology 231(1):109–116 Silverman FN, Kuhn JP (1993) Congenital abnormalities. In: Silverman FN (ed) Caffey’s Pediatric X-ray Diagnosis: An Integrated Imaging Approach. Mosby, St. Louis, pp 919–930 Taybi H (1979) The biliary tract in children. In: Margulis AR, Burhenne HJ (eds) Alimentary Tract Radiology. Mosby, St. Louis, pp 1504–1516 Rizzo RJ, Szucs RA, Turner MA (1995) Congenital abnormalities of the pancreas and biliary tree in adults. Radiographics 15(1):49–68 Mortele KJ, Ros PR (2001) Anatomic variants of the biliary tree: MR cholangiographic findings and clinical applications. AJR Am J Roentgenol 177(2):389–394

Congenital Malformations, Musculoskeletal System H EIKE E. DALDRUP-L INK Department of Radiology, Pediatric Radiology section and Director, Contrast agent Research Group University of California in San Francisco USA [email protected]

Congenital malformations are the leading cause of infant mortality in the United States and a major cause of morbidity and mortality throughout childhood. The children with major congenital malformations represent approximately 4% of live births with a higher rate in males than females (4.6% vs. 3.1%), and a higher rate in black children than white children (4.4% vs. 3.8%) (1). Twenty percent of infant deaths are attributed to congenital malformations, a percentage that has increased over time. Little is known about the causes of congenital malformations. Twenty percent may be due to a combination of heredity and other factors; 7.5% may be due to single-gene mutations; 6% to chromosome abnormalities; and 5% to maternal illnesses, such as diabetes, infections, or anticonvulsant drugs. Fourty to sixty percent of congenital malformations are of unknown origin. Associations between congenital malformations and environmental agents have been described for radiation

412


exposure, intrauterine infections (e.g., particular rubella), medications (e.g., thalidomide and anticonvulsants), and toxic waste dumps (e.g., Love Canal and accidents such as Three Mile Island and Seveso). In response, many states began to develop birth defects registries in order to track trends in malformation rates (1). The musculoskeletal system represents the third most common organ system involved in major congenital malformations (16%). Other potentially involved organ systems are cardiovascular (26%), genitourinary (21%), digestive and clefts (9%), CNS (5%), respiratory (3%), and others (20%). The most common congenital musculoskeletal malformations are dislocation of the hip (22%), varus deformities of the feet (20%), other limb anomalies (10%), anomalies of the skull and face (10%), reduction deformities (6%), valgus deformities of the feet (6%), other feet deformities (3%), and others (23%). A more specific overview over relatively frequent congenital malformations that involve the musculoskeletal system is given in Table 1. The following article provides a brief overview over frequent congential musculoskeletal

malformations. Due to its limited length, it cannot be comprehensive. For more detailed information, the reader should consult specific literature for the individual pathologies, mentioned later.

Congenital Malformations of the Hip ▶Hip dysplasia is the most common congenital malformation and represents an abnormal growth or development of the acetabulum, femoral head, and associated ligaments and soft tissues. It can be associated with instability and/or dislocation of the femoral head. Ultrasonography is the method of choice for the diagnosis and treatment of hip dysplasia and instability in newborns and young infants. The evaluation is typically performed by assessing the alpha angle (which outlines the superior bony acetabulum) and the beta angle (which represents the cartilaginous part of the acetabulum). A normal alpha angle should be greater than 60˚ and a normal beta angle should be less than 55˚. In older children with ossified

Congenital Malformations, Musculoskeletal System. Table 1 Relatively frequent major congenital malformations, that involve the musculoskeletal system ICD-9

Malformation

090 243 658.8 741.0 741.9 742.1 749.0 749.2 754.3 754.51 755.2 755.3 755.8 756.0 756.4 756.51 758.0 758.1 758.2 758.7 760.71 771.0 771.1 771.2

Congenital syphilis Congenital hypothyroidism Amniotic bands Spina bifida with hydrocephalus Spina bifida without hydrocephalus Microcephalus Cleft palate Cleft palate & lip Congenital dislocation of hip Talipes equinovarus Reduction deformities of upper limb Reduction deformities of lower limb Arthrogryposis multiplex congentia Craniosynostosis Chondrodystrophy Osteogenesis imperfecta Down syndrome Patau syndrome Edwards syndrome Klinefelter syndrome Fetal alcohol syndrome Congenital rubella Congenital cytomegalovirus infection Other congenital infections

Prevalence

Ratio: M/F

22.8 3.0 0.3 1.9 1.8 5.8 5.8 4.6 14.6 11.6 3.0 1.9 1.5 4.0 1.1 0.3 10.1 0.8 1.2 0.6 2.1 0.1 1.4 2.3

1.0 1.1 0.3 0.9 0.9 0.8 0.7 1.4 0.3 1.7 0.9 1.3 0.9 2.1 1.0 0.4 1.0 1.3 0.5 1.4 1.0 1.1 0.8

Source: From The New York State Department of Health Congenital Malformations Registry: //www.health.state.ny.us.


413

the capital femoral physis, and femoral anteversion. The congenital coxa vara shows a characteristic radiographic finding of a fragment of bone inferolateral to the proximal femoral physis, which represents a contained area of abnormal calcification.

Congenital Malformations of the Limbs

Congenital Malformations, Musculoskeletal System. Figure 1 Hip dysplasia: The left acetabulum is markedly shallow. There is extensive subchondral lucency and sclerosis. There is marked malformation of the left femoral epiphysis. These changes are consistent with congenital hip dysplasia and are unchanged compared to prior study. Ossific fragment is again seen superior to the left femoral head, likely avulsion injury versus heterotopic bone. Several calcific densities are present in the proximal left femoral shaft, which may represent exaggerated trabeculae versus enchondromas.

femoral heads, radiographs are obtained to profile the anterior aspect of the acetabulum (Fig. 1). If any evidence of hip subluxation is present, an additional abducted internal rotation view is added to determine the true neckshaft angle of the proximal femur. Typical measures for hip joint assessment are the Hilgenreiner line, Perkin line, Shenton Menard line, and acetabular angle (5 vessels/cm2) and peak arterial Doppler shift (>2 kHz) in hemangiomas, which is useful in distinguishing hemangiomas from other vascular malformations. In the involutional phase, the lesion decreases in size and has a reduced number of vessels. Computed tomography (CT) and magnetic resonance imaging (MRI) in the proliferative phase demonstrate a homogeneous mass with an intense, persistent, and usually lobular enhancement pattern. Involuting hemangiomas are more heterogeneous with less enhancement. MRI often demonstrates heterogeneous intermediate signal intensity on T1-weighted images. On T2-weighted images, high signal intensity is seen with a flow void in and around the mass. Hemorrhage or fat deposition is seen as high signal intensity on T1- and T2-weighted images. Angiography reveals a well-circumscribed mass with intense, persistent tissue staining in a lobular pattern and enlargement of feeding arteries. Angiography is required only if endovascular intervention is contemplated. Hemangioendothelioma: Plain abdominal radiography may show hepatomegaly and a nonspecific mass effect in the upper abdomen, with displacement of intestinal structures and occasional calcifications within the mass.

Haemangioma and Haemangioendothelioma in Childhood. Figure 1 Doppler ultrasound shows a highly vascularized lesion in the liver. (cited from Fletcher, 2004)

H The appearance of hemangioendothelioma on ultrasound is variable and nonspecific. It may be either hyperechoic or hypoechoic. Most of the tumors have well-defined borders toward the surrounding liver parenchyma. The caliber of the aorta may abruptly decrease below the level of the celiac axis because of increased hepatic arterial flow secondary to intrahepatic AV shunting. CT shows a well-defined homogeneous or heterogeneous mass, hypoattenuating relative to the normal liver parenchyma. In about 40% of cases there may be calcifications. The postcontrast enhancement pattern resembles that of an adult giant hemangioma (peripheral enhancement in the early phase and central enhancement in the delayed phase). In larger tumors this central enhancement may not occur because of fibrosis, hemorrhage, or necrosis. On MRI, low signal intensity on T1-weighted images and high signal intensity on T2-weighted images are seen. Flow voids may be observed on T2-weighted images in a tumor with AV shunting and high blood flow. The presence of hemorrhage, necrosis, and fibrosis make the tumor appear heterogeneous both on T1- and T2weighted images. Postcontrast enhancement is similar to that on CT (Figs 2 and 3). Angiography is performed only if embolization is considered either as definitive therapy or as a temporizing method prior to transplantation. Characteristically, there is prolonged pooling of contrast within the mass. Internal AV shunting with large draining veins may be seen.

Nuclear Medicine Hemangioma: Radiolabeled red cell scintigraphy. Hemangioendothelioma: Nonspecific, early uptake may be seen. Blood pool studies may show increased activity.

808

Haemangioma and Haemangioendothelioma in Childhood

Diagnosis Most hemangiomas are recognized clinically and do not require imaging. Imaging is required for atypical soft tissue masses, for deep hemangiomas with normal overlying skin, and for evaluation of deep extension of obvious hemangiomas. It is also required for hemangiomas that compromise important structures (subglottic hemangioma, periorbital lesions) or that cause highoutput cardiac failure or thrombocytopenia (Kasabach– Merritt syndrome). Differential diagnosis (Fig. 4): Vascular malformation: Haemangioma and Haemangioendothelioma in Childhood. Figure 2 On T1-weighted image, a hypointense lesion is seen in the right lobe of the liver.

. Abnormal size and/or number of vascular structures . Usually manifests as cutaneous birthmarks . No regression. The clinical manifestations of infantile hemangioendothelioma are variable. The tumor may be asymptomatic and discovered incidentally, but more often the tumor is large and manifests as hepatomegaly, abdominal distention, and/or a palpable upper abdominal mass. Highoutput failure and platelet consumption due to significant AV shunting can be present. Hematologic abnormality may be seen, including anemia and thrombocytopenia. The serum -fetoprotein level is usually normal, in contrast to hepatoblastoma. Ultrasound shows one or more solid masses with high flow in color Doppler. MRI shows a well-defined mass with an enhancement pattern resembling that of adult giant hemangioma. Differential diagnosis: Hepatoblastoma:

Haemangioma and Haemangioendothelioma in Childhood. Figure 3 On T2-weighted image, the le-

. a fetoprotein level is elevated.

Haemangioma and Haemangioendothelioma in Childhood. Figure 4 diagnosing hemangioma versus vascular malformation.

Differential diagnosis of hemangioma: algorithm for

Hemangioma, Hepatic

Hepatocellular carcinoma: .

Most common hepatic malignancy in children over 5 years of age . Rarely occurs under 3 years of age.

Bibliography 1.

2.

3.

4.

5.

6.

Pediatric Radiology, 2nd edition, The Requisites. Hans Blickman, MD, PhD, Boston University School of Medicine, Boston, MA. Mosby, 1998 Oslo (2001) Paediatric Imaging, edited by Helen Carty, Volume 7, in The Encyclopaedia of Medical Imaging. Edited by Holger Pettersson, MD, PhD, and David Allison, MD, FRCR Lund, Sweden: NICER Institute/ISIS Medical Media Fletcher BD (2004) Soft tissue neoplasms. In: Kuhn JP, Slovis TL, and Haller JO (eds) Caffey’s Pediatric Diagnostic Imaging. 10th edn. Mosby, Philadelphia, pp 2017–2018 Schlesinger AE, Parker B (2004) Tumor and tumor-like condition. In: Kuhn JP, Slovis TL, and Haller JO (eds) Caffey’s Pediatric Diagnostic Imaging. 10th edn. Mosby, Philadelphia, pp 1494–1495 ▶http://www.emedicine.com/ped. Arterial Vascular Malformations including Hemangiomas and lymphangiomas. Last updated: October 8, 2004 Roos JE, Pfiffner R, Stallmach T et al (2003) Infantile Hemangioendothelioma. Radiographics 23:1649–55

Haemangioma, Hepatic M ARZIO P ERRI , R ICCARDO L ENCIONI Department of Diagnostic and Interventional Radiology University of Pisa, Italy [email protected]

Definition Hepatic hemangiomas are benign, vascular tumors of the liver.

Epidemiology The prevalence of hemangiomas ranges from 1% to 2% in the general population, the tumor being typically discovered incidentally during evaluation of nonspecific abdominal complaints. The prevalence of hemangiomas has been overestimated in autopsy series (from 2% to as high as 20%), because of overrepresentation of elderly patients with comorbid illnesses. The majority of hemangiomas are small (200 msec) echo times, and serial dynamic gadolinium-enhanced gradient-echo T1-weighted sequences. Hemangioma appears as a homogeneous focal lesion with smooth, well-defined margins. The lesion is hypointense to liver parenchyma on T1-weighted MR images and strongly hyperintense to liver parenchyma on T2-weighted MR images (Fig. 3). The high signal intensity on heavily T2-weighted (long echo time) MR images gives to hemangiomas a consistent “light-bulb” pattern with 100% sensitivity and 92% diagnostic specificity. Dynamic contrast-enhanced MR imaging shows quite a typical perfusion pattern in hemangioma, that is, peripheral nodular enhancement in the early phase with centripetal progression to uniform or almost uniform enhancement during the portal venous and the delayed phase. Such a characteristic enhancement pattern has a sensitivity of 77–91% and a specificity of 100% for the diagnosis of hemangioma (4). However, very small (50–60 years). Symptoms due to a small tumour are indistinguishable from those of the underlying liver disease. The most common manifestations are right hypochondrial or epigastric pain, weight loss, ascitis, hepatic failure, jaundice. In the high-incidence geographical areas the tumour often develops in absence of cirrhosis in younger individuals (60 50–60 50–60 43–49 43–49 4 ng/mL) is a strong indication to proceed imaging studies and TRUS-guided biopsies. Computed tomography—With computed tomography (CT), BPH appears as a homogeneous-enhancing area with well-defined borders (Fig. 3). CT has no

H

910

Hyperplasia, Benign, Prostate

Hyperplasia, Benign, Prostate. Figure 2 Transrectal ultrasound image of the prostate in the axial plane in a 64-year-old patient. In the central gland, two large benign prostatic hyperplasia nodules are present (white arrows).

Hyperplasia, Benign, Prostate. Figure 3 Axial CT plane after intravenous contrast agent demonstrates homogeneous enhancing benign prostatic hyperplasia nodules in the central gland of the prostate (white arrows).

place in the evaluation of BPH because of low intraprostatic tissue contrast/resolution, which consequently does not allow the evaluation of the ratio of glandular to stromal tissue in the prostate. Prostate volumes can be estimated with this imaging modality. However, this imaging method is costly and has limited availability.

Magnetic resonance imaging—Magnetic resonance (MR) imaging allows high intraprostatic tissue contrast/ resolution, which can be used to estimate prostate volume and glandular to stromal ratio. MR imaging has high accuracy for determining the histologic type (8). BPH nodules appear as nodular heterogeneous (iso- to high) signal intensities lesions with a small rim of low signal intensity in the central gland (9) (Fig. 4). Positron emission tomography—Positron emission tomography (PET), using [11C]acetate, has been reported to be of clinical relevance for the diagnosis of prostate cancer (10). Normal prostate tissue or BPH can show enhanced accumulation of [11C]acetate, especially in younger men. Preliminary results indicate that careful interpretation of [11C]acetate PET images of prostate cancer is necessary because of the standardized uptake value (SUV) and the early-to-late activity ratio of the SUV for the normal prostate and for BPH overlap significantly with those for prostate cancer (11). Currently, indication for the treatment in BPH patients is based on subjective measurements of symptoms. Thus, imaging does not play a major role in the evaluation of such patients. However, the primary imaging investigation of choice is TRUS of the prostate, which may be performed either by an urologist or by a radiologist, depending on the local availability of expertise. Also TRUS is a noninvasive, cost-efficient, quick and easily available modality.

Hyperplasia, Benign, Prostate

Hyperplasia, Benign, Prostate. Figure 4 T2-weighted transverse plane of the prostate in a 63-year-old male. In the central gland of the prostate, two large benign prostatic hyperplasia nodules with low to high signal intensities are present (white arrows). Note the low signal intensity area in the left peripheral zone representing prostate carcinoma (black arrow).

International Prostate Symptom Score—Alternative to imaging modalities, the International Prostate Symptom Score (IPSS) was introduced. This score was developed to quantify and validate responses to the questions asked. A set of seven questions has been adopted worldwide and yields reproducible and quantifiable information, regarding symptoms and response to treatment. Each question allows the patient to choose one of six answers indicating increasing severity of symptoms on a scale of 0–5; the total score ranges from 0 to 35. Questions concern incomplete emptying, frequency, intermittency, urgency, weak stream, straining, and nocturia. The eighth question is known as the bother score and pertains to the patient’s perceived quality of life (QOL). Scores can range from 0 (delighted) to 6 (terrible). After calculating the total score for all eight questions, patients are classified as 0–7 (mildly symptomatic), 8–19 (moderately symptomatic), or 20–35 (severely symptomatic).

Treatment Patients with (mild) symptoms are initially treated with medical therapy. Management with hormone substitutes emerged from the discovery of a congenital form of pseudohermaphroditism secondary to dihydrotestosterone deficiency. This deficiency produced a hypoplastic prostate. Type II 5-alpha reductase is an enzyme

911

responsible for the conversion of testosterone to dihydrotestosterone. The 5-alpha reductase inhibitors block the conversion of testosterone to dihydrotestosterone, causing lower intraprostatic levels of dihydrotestosterone. This leads to inhibition of prostatic growth, apoptosis, and involution. Transurethral resection of the prostate (TURP) has long been the most common method by which obstructing prostate tissue is removed through the urethra. The indications for surgical intervention include acute urinary retention, failed voiding trials, recurrent gross hematuria, urinary tract infection, and renal insufficiency secondary to obstruction. TURP has a significant risk of morbidity (18%) and mortality (0.23%). Prostatic stents are flexible devices that can expand when put in place to improve the flow of urine past the prostate. However, their use has been associated with encrustation, pain, incontinence, and overgrowth of tissue through the stent, possibly making their removal quite difficult. Their full role and long-term effects are not fully known yet. Radical prostatectomy is now reserved for patients with very large prostates, patients with concomitant bladder stones or bladder diverticula, and patients who cannot be positioned for transurethral surgery.

Bibliography 1. McConnell JD (1998) Epidemiology, etiology, pathophysiology and diagnosis of benign prostatic hyperplasia. In: Campbell’s Urology. 7th edn. WB Saunders, Philadelphia, pp 1429–1452 2. Berry SJ, Coffey DS, Walsh PC et al (1984) The development of human benign prostatic hyperplasia with age. J Urol 132:474–479 3. Bartsch G, Rittmaster RS, Klocker H (2002) Dihydrotestosterone and the concept of 5alpha-reductase inhibition in human benign prostatic hyperplasia. World J Urol 19:413–425 4. McNeal JE (1978) Origin and evolution of benign prostatic enlargement. Invest Urol 15:340–345 5. Barry MJ, Cockett AT, Holtgrewe HL et al (1993) Relationship of symptoms of prostatism to commonly used physiological and anatomical measures of the severity of benign prostatic hyperplasia. J Urol 150:351–358 6. Jacobsen SJ, Jacobson DJ, Girman CJ et al (1997) Natural history of prostatism: risk factors for acute urinary retention. J Urol 158:481–487 7. Joorgensen JB, Jensen KME, Bille-Brahe NE et al (1986) Uroflowmetry in asymptomatic elderly males. Brit J Urol 58:390–395 8. Schiebler ML, Tomaszewski JE, Bezzi M et al (1989) Prostatic carcinoma and benign prostatic hyperplasia: correlation of high-resolution MR and histopathologic findings. Radiology 172:111–137 9. Kahn T, Burrig K, Schmitz-Drager B et al (1989) Prostatic carcinoma and benign prostatic hyperplasia: MR imaging with histopathologic correlation. Radiology 173:847–851 10. Oyama N, Akino H, Kanamaru H et al (2002) 11C-acetate PET imaging of prostate cancer. J Nucl Med 43:181–186 11. Kato T, Tsukamoto E, Kuge Y et al (2002) Accumulation of [11C]acetate in normal prostate and benign prostatic hyperplasia: comparison with prostate cancer. Eur J Nucl Med 11:1492–1495

H

912

Hyperplasia, Breast

Hyperplasia, Breast LUIS P INA Department of Radiology, Clıńica Universitaria de Navarra, Spain [email protected]

Definition Hyperplasia is the proliferation of epithelial cells lining the wall of ducts and lobules. It may involve the ducts (ductal hyperplasia) or the lobules (lobular hyperplasia). Three categories of ductal hyperplasia have been reported: ▶mild ductal hyperplasia, moderate or florid ductal hyperplasia, and atypical ductal hyperplasia (1).

Pathology/Histopathology Hyperplasia shows no specific features on gross examination (2). Mild ductal hyperplasia consists of a proliferation of more than two but less than four cell layers between basement layer and lumen, without cellular atypia (1). Moderate or florid ductal hyperplasia is defined as an intraductal epithelial proliferation greater than four cell layers, usually enough to distend the ductule and bridge the lumen. Some cellular atypia is seen (1). Atypical ductal hyperplasia is characterized by the proliferation of evenly distributed monomorphic cells with ovoid to round nuclei. Structures such as micropapillae, arcades, bridges, and others may be found, mimicking ductal carcinoma in situ. Differentiating between the two entities may be difficult. The criteria used include the involvement of only one duct or a size equal or less than 2 mm. Nevertheless, an important variability may be found between pathologists (1, 2). ▶Atypical lobular hyperplasia is composed of cells similar to those found in lobular carcinoma in situ. The differentiation may be difficult. If less than half of the acini are filled, distorted, or distended by the proliferation, the lesion is considered atypical lobular hyperplasia. If there are luminal spaces free of cells and if different types of cells are intermixed, the lesion is also considered atypical lobular hyperplasia instead of lobular carcinoma in situ (1).

Hyperplasia is an alteration associated with no specific clinical features. Most cases are microscopic in dimension and thus nonpalpable and asymptomatic (2). However, hyperplasia may be associated with other coexisting lesions, such as radial scar-complex sclerosing lesion, pseudoangiomatous stromal hyperplasia, papilloma, sclerosing adenosis, and duct ectasia. The clinical findings may be due to the accompanying lesion and present as pain, nipple discharge, or a palpable mass, usually referred to as breast thickening. Mild hyperplasia does not increase the risk of developing breast cancer. Moderate or florid hyperplasia carries a small risk (relative risk about 1.5). However, atypical ductal hyperplasia and atypical lobular hyperplasia are well-recognized risk factors for breast cancer, with a fourfold and even tenfold increased risk if there is a marked family history (1). It is controversial whether ductal hyperplasia may evolve into atypical ductal hyperplasia or low nuclear-grade ductal carcinoma in situ.

Imaging Mammography There are no characteristic findings on mammography to diagnose hyperplasia. The diagnosis is made based on biopsy of suspicious calcifications, masses, or architectural distortions (Figs. 1, 2 & 3), but the former are the most frequent (3). The calcifications associated with ductal hyperplasia, atypical ductal hyperplasia, and low nucleargrade ductal carcinoma in situ are similar, being usually round, punctate, or irregular. No linear calcifications are found due to the lack of necrotic debris.

Ultrasound There are no conclusive ultrasonographic features for diagnosing hyperplasia. If hyperplasia is associated with other lesions, the ultrasonographic appearance is usually that of the accompanying alteration (3).

Magnetic Resonance Both atypical ductal hyperplasia and atypical lobular hyperplasia may enhance after paramagnetic contrast administration.

Clinical Findings

Nuclear Medicine

Hyperplasia can affect females at virtually any age, although most women are between 35 and 60 years of age (1).

Nuclear medicine does not play a role in the diagnosis of hyperplasia.

Hyperplasia, Focal, Nodular Hepatic

913

Diagnosis

Hyperplasia, Breast. Figure 1 Round calcifications in a linear distribution. Pathology: florid ductal hyperplasia.

Hyperplasia is a histopathologic term, usually diagnosed as the result of biopsy of nonpalpable lesions such as breast calcifications. The widespread use of mammography and percutaneous biopsy techniques has led to a increasing number of cases. The management after percutaneous breast biopsy is important. A result of atypical ductal hyperplasia at 14-gauge core biopsy underestimates carcinoma in 20–50% of cases (4, 5). Vacuum-assisted devices decrease this underestimation, but underestimation persists in up to 38% of cases (4, 5). Even the complete removal of the mammographic lesion does not ensure the absence of carcinoma. Thus, a diagnosis of atypical ductal hyperplasia at percutaneous biopsy warrants surgical excision. Atypical lobular hyperplasia at percutaneous biopsy is also an indication for surgical excision. However, mild or ▶moderate ductal hyperplasia diagnosed at core needle biopsy does not need further intervention.

Bibliography 1.

2.

3.

Hyperplasia, Breast. Figure 2 Surgical specimen of calcifications of intermediate concern, rounded and flaky. Pathology: atypical ductal hyperplasia.

4. 5.

Hughes LE (2000) The relationship between clinician and pathologist in benign breast disorders. In: Hughes LE, Mansel RE, Webster DJT (eds) Benign Disorders and Diseases of the Breast. Concepts and Clinical Management. Saunders, London, pp 49–59 Rosen PP (2001) Ductal hyperplasia. Ordinary and atypical. In: Rosen PP (ed) Rosen’s Breast Pathology. Lippincott Williams & Wilkins, Philadelphia, pp 201–228 Cardenõsa G (2001) Terminal duct. Ductal hyperplasia. In: Cardenõsa G (ed) Breast Imaging Companion. Lippincott Williams & Wilkins, Philadelphia, pp 239–241 Berg WA (2004) Image-guided breast biopsy and management of high risk lesions. Radiol Clin North Am 42:935–946 Liberman L (2000) Clinical management issues in percutaneous core breast biopsy. Radiol Clin North Am 38:791–807

Hyperplasia, Focal, Nodular Hepatic R ICCARDO L ENCIONI , M ARZIO P ERRI Department of Diagnostic and Interventional Radiology University of Pisa, Pisa, Italy [email protected]

Definition Hyperplasia, Breast. Figure 3 Architectural distortion. The lesion was a complex sclerosing lesion with atypical ductal hyperplasia. In this case, the complex sclerosing lesion was the cause for the mammographic image, and the atypical hyperplasia was not visible.

Focal nodular hyperplasia (FNH) of the liver is a completely benign lesion characterized by nodular hyperplasia of hepatic parenchyma around a central scar containing an anomalous artery.

H

914


Epidemiology FNH is the second most common benign tumor of the liver, with an estimated prevalence of 0.4–0.8% in unselected autopsy series. The tumor has a female:male ratio of between 2 and 26:1 and the average age at presentation is between the ages of 35 and 50 years.

Pathogenesis FNH appears to be the result of a hyperplastic response of the hepatic parenchyma to an arterial lesion and/or portal venous ischemia. A congenital vascular malformation is suggested by the presence of a central fibrous scar containing abundant arteries with spider-like malformations and the association with other vascular abnormalities such as hepatic hemangioma and hereditary hemorrhagic telangiectasia. The finding of unbalanced expression of angiopoietin1 and angiopoietin2 genes coupled with expression of angiopoietin1 protein by the endothelial cells of dystrophic vessels, suggests a role of angiopoietin genes in FNH. Clonality studies and overexpression of important genes, involved in cell omeostasis such as Bcl-2 and transforming growth-factor alpha, support the important role for hepatocellular proliferation, in FNH. Conversely, the role of oral contraceptives in FNH is disputed. Their use has been associated with increase in size and vascularity of FNH nodes and tumor regression was observed after drug withdrawal.

Pathology and Histopathology Macroscopically, the vast majority of the patients have a single, pale tan to light brown lesion causing a central scar radiating into the liver tissue. FNH may present with classical and nonclassical forms. The latter include telangiectatic, mixed hyperplastic/adenomatous, with cytologic atypia and multiple FNH syndrome forms. Eighty percent of the patients present with the classical form, which is characterized by abnormal nodular architecture, malformed-appearing vessels, and bile duct proliferation. The majority of classical forms contain one to three macroscopic scars. Microscopically, classical FNH lesion shows nodular hyperplastic parenchyma, the nodules being completely or incompletely surrounded by fibrous septa. The central scar contains malformed vessels of various caliber, mostly large and tortuous arteries showing intimal or muscular fibrohyperplasia. Dense bile ductular proliferation accompanies the vascular structures both in the central scars and in the radiating septa with histologic

cholestasis. A mild degree of macrovascular steatosis is often present. Nonclassical forms of FNH show atypical histology and bile ductular proliferation, but lack either nodular architecture or malformed vessels.

Clinical Presentation FNH is usually an incidental finding but a few patients may have symptoms as a palpable mass or hepatomegaly. Liver chemistry is usually unaltered. In a few patients, FNH may cause slight elevations in serum GGTP levels. Serum tumor markers AFP, Ca 19.9, and CEA are invariably negative.

Imaging Ultrasound and Contrast-Enhanced Ultrasound FNH may have variable ultrasound (US) features. It usually appears as a round mass that is slightly hypoechoic or slightly hyperechoic with respect to liver parenchyma. Some lesions may be isoechoic to liver and be detected only because of vascular displacement. The lesion is frequently homogeneous. In fact, the detection of the central scar at baseline US is uncommon. Typical findings at color or power Doppler US include the presence of a central feeding artery with a stellate or spoke-wheel pattern determined by vessels running into radiating fibrous septa originating from the central scar. Doppler spectral analysis can show an intralesional pulsatile waveform with high diastolic flow and low resistive index (1). The specificity of US in the diagnosis of FNH has improved following the introduction of US contrast agents. At contrast-enhanced ultrasound (CEUS), FNH shows central vascular supply with centrifugal filling in the early arterial phase, followed by homogeneous enhancement in the late arterial phase. In the portal phase the lesion remains hyperechoic relative normal liver tissue, and becomes isoechoic in the late phase. This pattern has been observed in 85–100% of FNH (2). The central scar becomes detectable as a hypoechoic area in the portal phase of the contrast-enhanced study.

Computed Tomography FNH is usually isoattenuating or slightly hypoattenuating to surrounding liver at baseline computed tomography (CT) scanning. The detection rate of the central scar, that appears as a hypoattenuating structure, is related to the size of the lesion. It may be identified in 35% of lesions


915

Hyperplasia, Focal, Nodular Hepatic. Figure 1 Typical FNH on multiphasic spiral-CT. On unenhanced CT scan (a) a slightly hypodense nodule is depicted. At the dynamic enhanced-CT study the nodule shows homogeneous enhancement except for the central scar in the arterial phase (b) becoming isoattenuating with the surrounding liver in the portal venous phase (c).

H smaller than 3 cm in diameter and in 65% of those exceeding 3 cm (3). FNH shows strong homogeneous enhancement during the arterial phase of the contrastenhanced CT study. The central scar is typically hypoattenuating during the arterial phase. In the portal venous and delayed phases, FNH becomes isoattenuating to the hepatic parenchyma (Fig. 1). On delayed images, the central scar may become hyperattenuating because of contrast distribution within its fibrous stroma.

Magnetic Resonance Magnetic resonance (MR) imaging is the most accurate imaging method to characterize FNH. Owing to the affinity of its cells with normal hepatocytes, FNH is usually slightly hypointense or isointense with respect to normal liver parenchyma on T1-weighted images and slightly hyperintense or isointense on T2-weighted images (4). The hallmark of the lesion, the central stellate scar, is usually depicted because of its hypointensity on T1-weighted images and hyperintensity on T2-weighted images, reflecting its pathologic substratum of a vascularized connective tissue (4). At baseline MR imaging, however, the mentioned typical features, that are, homogeneous structure, isointensity to liver, and presence of the central scar, are observed in only 20–25% of the cases. Diagnostic confirmation requires a contrastenhanced MR study. This is usually performed through serial dynamic imaging following the administration of a Gadolinium chelate (5). FNH shows strong, homogeneous enhancement in the arterial phase sparing the central scar, while it becomes isointense to liver parenchyma in the portal venous and delayed phases. The central scar may show contrast uptake in the delayed phase owing to the interstitial distribution of the contrast agent. These

features have a specificity of more than 95% for the diagnosis of FNH (5). However, even with the administration of Gadolinium chelates, the central scar may not be detectable as many as 22% of the FNHs, including 80% of those smaller than 3 cm (5). Liver-specific MR contrast agent provides an alternate strategy to diagnose FNH. Owing to the affinity of its cells with the hepatocytes, FNH takes up hepatocyte-targeted agents, like normal parenchyma. These agents are then trapped within the lesion, since FNH is unable to effectively eliminate the compound via biliary excretion. Hence, it appears hyperintense to normal parenchyma on T1-weighted images (Fig. 2). Also, the central scar, that does not take up the hepatocyte-targeted agent—becomes well delineated in up to 90% of cases. This approach may enable diagnosis of 90% of the FNHs with atypical features at the baseline and conventional contrast-enhanced dynamic study (5). The diagnosis of FNH has also been achieved with the use of RES-targeted agents. Because of its rich Kupffer cell population, FNH takes up iron oxide particles, and shows marked signal intensity decrease on T2-weighted images. The central scar is usually well delineated on postcontrast images as it does not contain RES cells and therefore keeps a high signal intensity.

Nuclear Medicine 99m

Tc sulfur colloid scintigraphy has long been used to characterize FNH. In fact, up to 80% of these lesions show uptake owing to their Kupffer cell population. Unfortunately, the uptake of sulfur colloid is not highly specific. In one series, sulfur colloid studies were diagnostic in only about 20% of FNHs.

916

Hyperplastic Cystic Disease

Hyperplasia, Focal, Nodular Hepatic. Figure 2 Focal nodular hyperplasia after Mn-DPDP. On baseline T1 (a) and T2-weighted (b) images the hepatic nodule is not clearly appreciable. Conversely, on the delayed T1-weighted image after Mn-DPDP administration (c) a hyperintense, well-defined lesion is depicted in the right hepatic lobe.

Diagnosis FNH is usually incidentally detected. Diagnostic confirmation can rely solely on imaging findings, when the typical features are detected in the proper clinical setting. CT can be used to characterize lesions of medium-tolarge size, but has limitations in the diagnosis of small lesions. Although promising results have been recently reported with the use of CEUS, MR imaging is the most accurate technique to diagnose FNH. MR imaging, by means of liver-specific contrast agents, enhances the informations provided by the baseline and dynamic gadolinium-enhanced study, improving the capability in lesion characterization, especially in case of small lesions. The use of percutaneous biopsy should be restricted to cases with questionable findings. In patients with atypical FNH, a scoring system has been proposed to categorize lesions into definite FNH, possible FNH, and negative for FNH.

Hyperplastic Cystic Disease ▶Fibrocystic Disease, Breast

Hyperprostaglandin E Hyperostosis and symptoms similar to Caffey disease, but as a known sequela of high doses of prostaglandin E1 or E2, usually given to maintain patency of the ductus arteriosus. ▶Caffey Disease

Hypersensitivity Contrast Reactions Bibliography 1. 2.

3.

4. 5.

Gaiani S, Piscaglia F, Serra C et al (1999) Hemodynamics in focal nodular hyperplasia. J Hepatol 31:576 Kim MJ, Lim HK, Kim SH et al (2004) Evaluation of hepatic focal nodular hyperplasia with contrast-enhanced gray scale harmonic sonography: initial experience. J Ultrasound Med 23:297–305 Brancatelli G, Federle MP, Grazioli L et al (2001) Focal nodular hyperplasia: CT findings with emphasis on multiphasic helical CT in 78 patients. Radiology 219:61–68 Bartolozzi C, Cioni D, Donati F et al (2001) Focal liver lesions: MR imaging-pathologic correlation. Eur Radiol 11:1374–1388 Grazioli L, Morana G, Federle MP et al (2001) Focal nodular hyperplasia: morphologic and functional information from MR imaging with gadobenate dimeglumine. Radiology 221:731–739

▶Adverse Reactions, Iodinated Contrast Media, Acute, Non renal

Hypertelorism Lateral shift of the orbits. ▶Congenital Malformations, Musculoskeletal System

Hypertension, Arterial

Hypertension, Arterial M ICHAEL U DER Universita˜tklinikum Erlangen, Erlangen, Germany [email protected]

Definition Individuals with a systolic blood pressure (BP) of 140 mmHg or more or a diastolic BP of 90 mmHg or more have to be considered as hypertensive. A systolic BP of 120–139 mmHg or a diastolic BP of 80–89 mmHg has recently been defined as prehypertension, which indicates an increased risk for progression to hypertension (1).

Pathology/Pathophysiology Hypertension can be classified as either essential or secondary. Essential hypertension is the term used when no specific medical cause can be found to explain a patient’s condition. Secondary hypertension means that the high BP is secondary to other conditions, such as kidney disease or hormone-producing tumors. Renovascular hypertension (RVH) is one of the most frequent types of secondary hypertension. It results from stenosis of the main renal artery or one of its major branches. The true prevalence of RVH is unknown but it may account for 1–5% of all cases in hypertensive adults. The pathophysiology of hypertension in renal artery stenosis (RAS) was first described by Goldblatt. Reduction of arterial perfusion pressure in the stenosed kidney leads to activation of the reninangiotensin-aldosterone system (RAA) causing systemic angiotensin II-dependent vasoconstriction and aldosterone-mediated volume expansion (Fig. 1). Local RAA activation, arterial wall remodeling, and oxidative stress are responsible for maintaining the hypertension. These local paracrine and structural changes may also contribute directly to renal injury (2). In the majority of cases, RVH is caused by either atherosclerotic RAS or fibromuscular dysplasia (FMD). Atherosclerosis accounts for 70–90% of cases of RAS and usually involves the ostium and proximal one-third of the main renal artery (Fig. 2). FMD is a collection of vascular diseases that affects the intima, media, and adventitia of the middle and distal portion of the main renal artery and its major branches. It is responsible for 10–30% of cases of RAS (Fig. 3). Neurofibromatosis, Takayasu’s arteritis, and radiation are rare causes of RVH.

917

Imaging Intra-arterial digital subtraction angiography (IA-DSA) is traditionally regarded as the ‘gold standard’ in assessing renal artery anatomy. Due to its high spatial resolution, DSA is particularly useful in the early detection of FMD, in the diagnosis of intrarenal branch artery stenoses, and in kidneys with complex anatomy, including multiple accessory arteries. Disadvantages of this technique include significant interobserver variation in assessing the degree of stenosis, the absence of functional information, and the risks of contrast agent nephropathy and cholesterol emboli syndrome. Contrast arteriography is used routinely before angioplasty or surgical intervention. Ultrasonography can be used to measure renal size and detects renal asymmetry that may be due to RAS. Importantly, kidneys less than 7 cm long are unlikely to benefit from revascularization procedures. Color-coded Doppler ultrasonography can identify RAS using either peak systolic velocity in a stenosis or by poststenotic flow phenomena. Velocities of 180–200 cm/sec at the point of maximum stenosis are widely used as the cut-off point to establish at least 50% stenosis. However, this test is unsuccessful in about 10–20% of the examinations. RASs can be detected indirectly by measuring acceleration and the waveform in intrarenal arteries distal to a stenosis. Combining direct and indirect measurements may increase the diagnostic yield. The reported sensitivity and specificity of color-coded Doppler ultrasonography for diagnosis of RAS ranges from 80 to 90% in the hands of experienced investigators. A recent meta-analysis showed that the accuracy is significantly lower than that of contrastenhanced magnetic resonance angiography (CE-MRA) and multidetector computed tomography angiography (MD-CTA). It has been proposed to use the resistance index calculated from the Doppler spectrum as a predictor of therapeutic outcome. However, there are conflicting data about the value of this parameter. Three-dimensional CE-MRA is widely accepted to be the best MR imaging technique for investigation of narrowed renal arteries. CE-MRA is typically performed using a spoiled gradient echo sequence acquiring the entire 3D data set in 20–30 sec with a spatial resolution of 1.5 1.5 1.5 mm3 or better. To get information of the dynamics of blood flow, time-resolved MRA such as multiphasic MRA or time-resolved imaging of contrast kinetics has been proposed. Although these techniques are valuable for the detection of delayed flow, they suffer from relatively poor spatial resolution. Besides anatomical abnormalities, MR imaging can determine the functional significance of narrowed renal arteries. Multiple functional parameters such as renal perfusion, glomerular

H

918


Hypertension, Arterial. Figure 1 Classic Goldblatt mechanism. A stenosis of the renal artery leads to the formation of renin via a reduction in renal circulation. This renin is released via the renal vein into the systemic circulation. It converts angiotensinogen into angiotensin I. This occurs after the splitting of two amino acids to angiotensin II by angiotensin-converting enzyme. Angiotensin II leads to constriction of peripheral resistance vessels and to water retention via the release of aldosterone. Together they cause systemic arterial hypertension (according to: Uder M, Humke U (2005) Endovascular Therapy of Renal artery stenosis—Where do we stand today? Cardio Vasc Intervent Radiol 28:139–47).

Hypertension, Arterial. Figure 2 High-grade atherosclerotic stenosis of the ostium of the left main renal artery in a 53-year-old man who had severe hypertension before and after deployment of a balloon-expandable stent (according to: Uder M, Humke U (2005) Endovascular Therapy of Renal artery stenosis—Where do we stand today? Cardio Vasc Intervent Radiol 28:139–47).


919

Hypertension, Arterial. Figure 3 Fibromuscular dysplasia of the renal artery in a 43-year-old man with severe hypertension. (a) Typical changes in the distal main artery. (b) After PTA with a 6-mm balloon catheter, irregularities of the vessel still appear. However, the patient became normotonic within 3 days, because the web-like stenoses were torn apart (according to: Uder M, Humke U (2005) Endovascular Therapy of Renal artery stenosis—Where do we stand today? Cardio Vasc Intervent Radiol 28:139–47).

filtration rate (GFR), tubular concentration, and transit and oxygenation can be assessed using MRI. Although there are promising results with the new techniques, today they are not yet used in clinical routine. With the introduction of multidetector row (MD) scanners, CTA has become an alternative for evaluation of renal arteries. Sixty-four-row scanners can cover the abdominal aorta and its side branches in a few seconds with submillimeter z-axis resolution, which allows nearisotropic imaging. To provide sufficient vessel-to-background contrast, 80–150 mL of 300–400 mg I/mL nonionic contrast medium is needed. Injection speeds range from 2–6 mL/sec. CTA data sets have to be acquired in breathhold with individualized timing of contrast material bolus. The spatial resolution of MD-CTA is superior to that of CE-MRA, and has reached 0.6 0.6 0.6 m3. The method has been successfully used for the assessment of atherosclerotic RAS. Broad experience in patients with FMD is still missing. CTA is as accurate as MRA but has the disadvantage of radiation dose and radio-contrast volume.

Nuclear Medicine Captopril renography is a functional test used to detect the angiotensin II dependence of GFR. In patients with activation of the RAA system, the preadministration of 25–50 mg oral captopril delays the uptake of tracer, reduces peak uptake, prolongs parenchymal transit, and slows excretion. The result of the test can be affected by

factors such as volume depletion, concurrent medication, underlying renal dysfunction, and bilateral RAS. Therefore careful preparation of the patient is mandatory. It is widely used as a screening test for RAS, but in reality it provides little additional information than an assessment of plasma creatinine level 3–5 days after introducing the drug. However, its most useful role is probably in predicting the benefit from revascularization of a stenosed kidney. Most data are from studies of patients with RAS due to FMD, and there are few available data from patients with atherosclerotic RAS.

Diagnosis The low prevalence of renovascular disease in patients with arterial hypertension makes universal screening for RAS in all hypertensive patients inappropriate. Therefore, imaging studies should be limited to groups with a higher clinical index of suspicion. Clinical indicators for RVH are abrupt onset of hypertension before age 30 or after age 55, accelerated or malignant hypertension, hypertension refractory to multiple drugs, hypertension in the presence of diffuse atherosclerosis, presence of a systolic/diastolic epigastric bruit, hypertension with unexplained renal insufficiency, azotemia induced by an ACE inhibitor or angiotensin receptor blocker, asymmetry of kidney size, and flash pulmonary edema. In these patients a secondary screening is mandatory to exclude RAS. Although there is no general consensus on which is the best imaging method, for

H

920


the initial screening noninvasive methods such as ultrasound, MRA or CTA, or scintigraphy are widely preferred. A meta-analysis showed the superiority of both CTA and MRA over both captopril renography and ultrasonography for noninvasive detection of RAS (3). In contrast, a recent prospective multicenter comparative study including more than 400 patients casts doubts on the usefulness of CTA and MRA in patients with suspected RAS (4). The study showed that neither CTA nor MRA were able to rule out RAS in patients with a high pretest probability of the disorder. A major problem with RAS is determining whether such lesions are functionally significant. Based on animal experiments, a reduction of lumen diameter of 70% is seen as relevant. In newer studies motivated by an analogous conclusion for the evaluation of stenoses of the coronary arteries or the leg vessels, stenosis of the renal artery of more than 50% has already been used as cut-off for hemodynamical significance. However, for stenoses between 50 and 70%, data that persuasively demonstrate their relevance are missing. An angiographically proven stenosis is highly suggestive of being hemodynamically effective, if collateral vessels become visible or when arterial phase, parenchymogram, and excretion into the collecting system are delayed in comparison to the contralateral side. For stenoses based on a fibromuscular dysplasia, proof of the typical changes in the presence of renal insufficiency or hypertension is sufficient as a treatment indication. Estimation of the angiographic stenosis grade does not play a role here, because the web-like stenoses cannot be reliably evaluated angiographically for their relevance to renal blood flow.

antihypertensive medication in terms of changes in BP. However, meta-analyses of 210 patients from the randomized trials showed a moderate but significant advantage for the angioplasty as opposed to pure pharmacological therapy in change of both systolic and diastolic BP. However, the value of angioplasty or stenting in RAS for treating arterial hypertension cannot yet be definitely rated with the available data. Angioplasty has become the primary treatment for fibromuscular dysplasia of the renal arteries, with a technical success in nearly all patients and excellent longterm results. In approximately 10–20% of the patients, the procedure has to be repeated due to a relapse of hypertension by restenoses. Atherosclerotic RAS does not respond as well to PTA therapy as do dysplastic stenoses. In atherosclerotic stenoses distant to the aortic wall, PTA is the first treatment option, with success rates of 70–90%. Stents should only be used if angioplasty fails. In ostial stenoses secondary to aortic plaque, a stent is needed to achieve technical success in most cases. Of the stented patients, 20% were faced with a restenosis within 6–12 months, whereby the risk increases with smaller stent diameters. The endovascular therapy of atherosclerotic renal artery stenoses is afflicted by a complication rate of 5–36% with periprocedural deaths reported in less than 2% of cases. Most frequently observed complications are hematoma formation and puncture trauma. Cholesterol embolism was identified as a main reason for the loss of kidney function after angioplasty and stent placement.

Bibliography

Interventional Radiological Treatment Many studies show that removal of hemodynamically relevant stenoses of the renal artery can lead to an improvement in or cure of arterial hypertension (5). The success of the treatment is not dependent solely on the degree of the stenosis but also on its etiology, on the age of the patient, and the duration of hypertension, for instance. Treatment of stenoses in fibromuscular dysplasia shows cure of the hypertension in 40% of patients and improvement in 51% of patients, with cumulative success rates up to 90% over 10 years. The results for atherosclerotic RAS are less favorable. An improvement of hypertension by PTA or stenting of the renal artery can be reached in a maximum of two-third of patients. In total, 30–50% of the treated patients get a relapse of hypertension within short to midterm follow-up. Three randomized controlled trials showed no benefit of PTA over

1.

2. 3.

4.

5. 6.

Chobanian AV, Bakris GL, Black HR et al (2003) National Heart, Lung, and Blood Institute Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure; National High Blood Pressure Education Program Coordinating Committee; The Seventh Report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure: the JNC 7 report. JAMA 289:2560–2572 Dzau VJ (2001) Tissue angiotensin and pathobiology of vascular disease: a unifying hypothesis. Hypertension 37:1047–1052 Vasbinder GB, Nelemans PJ, Kessels AG et al (2001) Diagnostic tests for renal artery stenosis in patients suspected of having renovascular hypertension: a meta-analysis. Ann Intern Med 135:401–411 Vasbinder GB, Nelemans PJ, Kessels AG et al (2004) Renal Artery Diagnostic Imaging Study in Hypertension (RADISH) Study Group Accuracy of computed tomographic angiography and magnetic resonance angiography for diagnosing renal artery stenosis. Ann Intern Med 141:674–682 White CJ (2006) Catheter-based therapy for atherosclerotic renal artery stenosis. Circulation 113:1464–1473 Uder M, Humke U (2005) Endovascular Therapy of Renal artery stenosis–where do we stand today? Cardio Vasc Intervent Radiol 28:139–47

Hypertension, Pulmonary

Hypertension, Pulmonary E DWIN J. R.

VAN

B EEK

Department of Radiology, Carver College of Medicine, C-751 GH, Iowa City, USA [email protected]

921

(usually due to left heart diseases, such as mitral valve stenosis or insufficiency). The most important differentiation should be focused on eliminating thromboembolic disease and shunts, as these are treatable through medication, interventional radiological procedures or surgery (1).

Pathology/Histopathology Definition Pulmonary hypertension is defined as a mean pulmonary arterial pressure of greater than 25 mm Hg causing excessive loading of the right heart. The etiology of pulmonary hypertension is manifold. It may be caused at the precapillary level (pulmonary arterial obstruction, such as chronic thromboembolic pulmonary embolism), capillary level (often labeled as idiopathic) and postcapillary level Hypertension, Pulmonary. Table 1

Due to the heterogeneity of this group of disorders, the pathology and histopathology may differ considerably. A classification of pulmonary hypertension has been widely adopted, and the main causes are presented in the Table 1. Several of the main types are 1. Precapillary pulmonary hypertension, with its main cause of (recurrent) pulmonary embolism.

The revised World Health Organization classification of pulmonary hypertension

Group I Pulmonary arterial hypertension Idiopathic (primary) Familial Related conditions: collagen vascular disease, congenital systemic-to-pulmonary shunts, portal hypertension, HIV infection, Drugs and toxins (e.g., anorexigens, rapeseed oil, L-tryptophan, methamphetamine, and cocaine); other conditions: thyroid disorders, glycogen storage disease, Gaucher’s disease, hereditary hemorrhagic telangiectasia, hemoglobinopathies, myeloproliferative disorders, splenectomy Associated with significant venous or capillary involvement Pulmonary veno-occlusive disease Pulmonary-capillary hemangiomatosis Persistent pulmonary hypertension of the newborn Group II Pulmonary venous hypertension Left-sided atrial or ventricular hear disease Left-sided valvular heart disease Group III Pulmonary hypertension associated with hypoxemia Chronic obstructive pulmonary disease Interstitial lung disease Sleep-disordered breathing Alveolar hypoventilation disorders Chronic exposure to high altitudes Developmental abnormalities Group IV Pulmonary hypertension due to chronic thrombotic disease, embolic disease, or both Thromboembolic obstruction of proximal pulmonary arteries Thromboembolic obstruction of distal pulmonary arteries Pulmonary embolism (tumor, parasites, foreign material) Group V Miscellaneous Sarcoidosis, pulmonary Langerhans’—cell histiocytosis, lymphangiomatosis, compression of pulmonary vessels (adenopathy, tumor, fibrosing mediastinitis) Source: Simonneau G, Galie N, Rubin LJ et al (2004) Clinical classification of pulmonary hypertension. J Am Coll Cardiol 43:5S–12S

H

922


2. Capillary pulmonary hypertension, which includes media hyperplasia and the so-called idiopathic forms of pulmonary hypertension. 3. Postcapillary pulmonary hypertension, which is related to diseases that give rise to pulmonary venous hypertension. These may be primarily associated with pulmonary venous occlusive disease or with cardiac disorders, including mitral valve regurgitation, left-toright shunts and increased back-up pressures in the presence of left-sided heart failure. In all instances, pulmonary hypertension will increase the preload as well as the after-load on the right ventricle. Given sufficient time, the heart will respond by developing right ventricular hypertrophy (cor pulmonale), while the central pulmonary arteries will become dilated (a main pulmonary artery diameter of >3 cm should be considered pathological).

Clinical Presentation The clinical presentation of pulmonary hypertension depends on the underlying etiology. Gradual onset of symptoms is typical, and these often consist of (a combination of) progressive shortness of breath, palpitations, and episodes of syncope. Most patients will demonstrate a gradual decline in walking distance or inability to perform exercise. Acute presentation is rare, occurring in 5% of cases, whereas the vast majority will have experienced symptoms in excess of 6 months following onset of symptoms (2). With progressive pulmonary hypertension, the right heart will ultimately fail and right heart failure signs and symptoms will develop, including pitting edema, ascites, raised jugular vein pressure, tachycardia, enlarged pulsating liver, loud P2 sound and/or pansystolic murmur, right ventricular heave and pleural effusions. Sudden death may occur.

Imaging Chest radiography will help distinguish disorders that may mimic symptoms, such as pneumonia, tuberculosis, and heart failure. In patients with established pulmonary hypertension, the heart configuration and the main pulmonary arteries are usually (but not always!) enlarged (Fig. 1). Vascular pruning may be evident, resulting in hyperlucent areas in the lungs. Right heart catheterization is currently the reference method for measurements of pressures, pressure gradients

Hypertension, Pulmonary. Figure 1 PA chest radiography, demonstrating enlargement of the right heart with double contour due to enlargement of right atrium and large central pulmonary arteries.

and combination with stress tests helps to assess the potential response to treatment. Echocardiography will reveal right ventricular overload, consisting of right ventricular hypertrophy, wall motion abnormalities, bowing of the interventricular septum to the left and tricuspid valve regurgitation. Echocardiography is capable of giving an estimate of the pulmonary artery pressure. CT pulmonary angiography is a very useful technology for the comprehensive assessment of the pulmonary vascular system, and is also capable of evaluating the consequences of pulmonary hypertension, such as right heart dilatation and hypertrophy, bronchial artery hypertrophy and perfusion anomalies (Fig. 2). Magnetic resonance imaging (MRI) has many different techniques, which may be employed for the diagnosis of pulmonary hypertension and its etiologies. These techniques include vascular assessment, including MR perfusion and MR angiography, quantification of blood flow across the pulmonary valve and anatomical and cine imaging of the heart (Fig. 3). Although it is currently not possible to determine the exact measurement of pulmonary arterial pressure, MRI is a useful test to give an overall impression of the severity of the disease and may help to determine the precise etiology of pulmonary hypertension (3).

Nuclear Medicine Traditionally, perfusion (-ventilation) lung scintigraphy was the main diagnostic test for pulmonary vascular


923

Hypertension, Pulmonary. Figure 2 CT findings in pulmonary hypertension. (a) Pulmonary artery dilatation (notice the size of the corresponding aorta, arrow). (b) Right ventricular hypertrophy, as demonstrated by anterior wall thickening (compare this to the left ventricular wall). (c) Mosaic perfusion, as indicated by areas of high attenuation (this is where the lung is receiving increased/shunted blood flow) versus areas of low attenuation which are hypoperfused.

H

Hypertension, Pulmonary. Figure 3 MRI findings in pulmonary hypertension. (a) Patient with ventricular septal defect and Eisenmenger syndrome. Notice the high septal defect, the distended right atrium and the right ventricular hypertrophy. The perfusion scintigram of this patient is demonstrated in Figure 5. (b) Patient with chronic thromboembolic pulmonary hypertension. Notice large central pulmonary arteries, multiple caliber changes in the segmental branches, consistent with webs and stenoses after recanalization of pulmonary emboli.

disease. With the advent of CT and MRI, this technology has been largely omitted from the diagnostic work-up. Nevertheless, perfusion scintigraphy is still considered a useful adjunct to assess the severity of pulmonary hypertension and to monitor disease progression and response to therapy. Furthermore, it is exquisitely sensitive to the development or presence of left-to-right shunts and capable of determining the actual severity of these shunts (Fig. 4).

Echocardiography is a first line investigation, and will raise further suspicion of increased pressures and may demonstrate right ventricular dysfunction. The final diagnosis will be made by right heart catheterization, while non-invasive tools, such as perfusion scintigraphy, CT and MRI will assist in determining the etiology and extent of the disease process. These tools are also increasingly being developed to help monitoring disease progression and response to treatment.

Diagnosis

Interventional Radiological Treatment

The diagnosis of pulmonary hypertension is often delayed due to the insidious onset of symptoms. Once the diagnosis is considered, chest radiography and ECG findings will be abnormal in 80% of patients (2).

There is an, albeit limited, role for interventional radiological treatment. First, inferior vena cava filters may be inserted in patients who undergo surgical corrections. Second, left-to-right shunts may be treated through

924

Hypertension, Renal

Hypertension, Renal N ICOLAS G RENIER Service d’Imagerie Diagnostique et The´rapeutique de l’Adulte, Groupe Hospitalier Pellegrin, Bordeaux, France [email protected]

Synonyms Renovascular disease; Renovascular hypertension

Definitions

Hypertension, Pulmonary. Figure 4 Perfusion scintigraphic findings in different etiologies of pulmonary hypertension. (a) Patient with chronic thromboembolic pulmonary hypertension. Notice that there are multiple wedge-shaped defects, similar to that seen in acute pulmonary embolism. (b) Patient with ventricular septal defect, Eisenmenger syndrome, and right-to-left shunt. The lungs show inhomogeneous tracer uptake, with photopenic central hilar areas due to pulmonary artery dilatation. Notice that the kidneys are visible; quantification of the shunt can be performed by calculating the actual uptake as a percentage of the total injected dose.

invasive radiological procedures, including coil occlusion of arteriovenous fistulas and placement of umbrellas to cover patent foramen ovale, persistent ductus arteriosus, or small ventricular septal defects.


Simonneau G, Galie N, Rubin LJ et al (2004) Clinical classification of pulmonary hypertension. J Am Coll Cardiol 43:5S–12S Elliot C, Kiely DG (2004) Pulmonary hypertension: diagnosis and treatment. Clin Med 4:211–215 Roeleveld RJ, Marcus JT, Boonstra A et al (2005) A comparison of noninvasive MRI-based methods of estimating pulmonary artery pressure in pulmonary hypertension. J Magn Reson Imaging 22:67–72

▶Renovascular disease (RVD) is a complex entity involving renal arterial lesions, renal disease, and hypertension. The relationships between these three entities are variable from patient to patient and difficult to assess, but their severity and their association increase the patient’s cardiovascular and renal risks. ▶Renovascular hypertension (RVH) is a reversible cause of hypertension secondary to a decrease in renal perfusion pressure, which activates the renin–angiotensin system and leads to the release of renin and the production of angiotensin II (angiotensin II increases sodium reabsorption and induces postglomerular and systemic arterial vasoconstriction, causing renovascular hypertension). The prevalence of RVH is around 1% among hypertensive patients. ▶Renal artery stenosis (RAS) is the main cause of RVH, but small-vessel intrarenal vascular disease can also be responsible for it. Its prevalence is unknown. Only the stenoses that reduce the internal diameter by >60% produce a significant decrease in renal blood flow. They can be atherosclerotic (90% of cases) or dysplastic (fibromuscular dysplasia, 10% of cases), ostial or not, and be located on main or accessory arteries. Small-vessel intrarenal vascular lesions are almost exclusively atherosclerotic. The relationship between hypertension and RAS is also difficult to assess, the best test being the reversibility of hypertension after revascularization. In fact, the majority of patients with atherosclerotic RAS and hypertension have essential hypertension as suggested by the absence of benefit of revascularization. Atherosclerotic, or ischemic, nephropathy is a complex entity. It is an important cause of end-stage renal failure (ESRF), causing up to 14% of ESRF over the age of 50 years. It is a consequence of the association of multiple factors, including decreased renal blood flow (secondary to bilateral RAS or to unilateral stenosis plus contralateral occlusion or to unilateral stenosis in a solitary kidney), intrarenal

Hypertension, Renal

atherosclerotic arterial disease, atheroembolism, diabetes, increased oxidative stress, medullary hypoxia, endothelial dysfunction, inflammatory response, and proteinuria. This multiplicity of causal factors explains the great heterogeneity of renal damage. There are several arguments in favor of nonischemic factors being responsible for ▶atherosclerotic nephropathy: (i) the presence of atheroembolic disease and focal segmental glomerulosclerosis at histology; (ii) a variable functional response to revascularization, with evolutive processes independent of a reduction of renal blood flow; and (iii) the possible observation of patients with two equal-sized kidneys, unilateral RAS, and impaired renal function, or of patients with severe bilateral RAS and relatively preserved renal function.

Characteristics Clinical distinction between renovascular and essential hypertension is difficult and based on classical criteria such as, among others, the onset of hypertension before or after the age of 50 years, the absence of a family history of essential hypertension, an abdominal bruit, duration of hypertension of less than 1 year, hypokalemia, and decreased renal function (Table 1). Atherosclerotic RAS is the most common cause of RAS. The prevalence increases with age, particularly in patients with diabetes, aorto-iliac occlusive disease, Hypertension, Renal. Table 1 Clinical features suggestive of atherosclerotic renal artery stenosis 1. Hypertension • Abrupt onset before the age of 50 year or after the age of 55 year • Accelerated or malignant hypertension • Refractory to treatment with ≥3 drugs 2. Renal abnormalities • Progressive or otherwise unexplained renal insufficiency in the setting of hypertension in elderly patients • Abrupt rise in serum creatinine after ACE inhibition • Low grade proteinuria and unremarkable urinary sediment • Significant discrepancy in size between the two kidneys • Unexplained hypokalemia 3. Other findings • Abdominal or flank bruit • The presence of various atherosclerotic manifestations and/or abdominal aortic aneurysm • Unexplained left ventricular hypertrophy • Flash pulmonary edema • Severe retinopathy

925

coronary artery disease, or hypertension. Progression of the degree of stenosis occurs in approximately 51% of renal arteries 5 years after diagnosis, and 3–16% of renal arteries progress toward occlusion. But in many cases, these arterial lesions are never detected because they are asymptomatic. Most of these stenoses are located at the ostium of the renal arteries and are in fact perirenal aortic plaques extending into the ostia. Fibromuscular dysplasia (FMD) is the second cause of RAS, occurring preferentially in females between 15 and 50 years of age. It is usually located more distally than atherosclerotic RAS, on the distal two-thirds of the renal artery and its branches, and may affect the intima, media (90% of cases), or adventitia. Depending on its type, FMD is characterized by long or short (diaphragm-like) stenoses and aneurysmal dilatations. Intimal and periarterial FMD are commonly associated with progressive dissection and thrombosis, whereas medial FMD progresses in 30% of patients and is rarely associated with dissection or thrombosis. When atherosclerotic disease progresses unilaterally, the size of the ipsilateral kidney, its cortical thickness, and its blood flow decrease, whereas the serum creatinine concentration and glomerular filtration rate remain normal because of compensatory changes within the contralateral kidney. When the disease progresses bilaterally, both kidneys present with progressive shrinkage and cortical thinning and irregularities. Usually, serum creatinine concentration increases and GFR decreases when more than 50% of renal mass is lost. Renal atrophy develops in approximately 21% of patients with RAS of more than 60%. The radiological imaging techniques available today have to reach four objectives: (i) to detect and characterize the RAS in terms of nature and anatomical and hemodynamic severity, (ii) to assess the anatomical consequences of the RAS on the artery itself and on the renal parenchyma, (iii) to assess the functional and cellular consequences of the RAS on the kidney, and (iv) to identify criteria of associated renal impairment related to RVD. Because the cardiovascular risk mainly depends on the degree of hypertension, improving blood pressure control has been a major task. Today, we know that medical treatment can achieve this control without renal revascularization in many cases. Therefore, management of these patients does not always require, as before, an early diagnosis of RAS, even though recent improvements in radiological techniques allow an accurate diagnosis of RAS noninvasively.

Imaging of RAS Intraarterial digital subtraction angiography is considered the gold standard for diagnosing RAS, but it is limited

H

926

Hypertension, Renal

now to characterizing the stenosis before transluminal angioplasty during the same procedure. Doppler ultrasound has gained a major place in the detection of RAS. The examination includes measurement of both kidneys, spectral sampling of two or three interlobar or segmental arteries of each kidney, and spectral sampling of both renal arteries. Diagnosis is based on direct proximal criteria (Fig. 1a) using both focal changes of color related to acceleration (aliasing) and turbulence (perivascular artifact) and spectral waveform changes at the site of stenosis (spectral broadening and increased velocity). Once detected, stenoses are considered significant (≥60%) when the renoaortic velocity ratio (RAR) is higher than 3.5 and/or when the peak systolic velocity is ≥180 cm/sec. Intrarenal wave forms may present changes distally to severe (>75%) RAS: slowed systolic acceleration, defined by an acceleration time of systolic peak >0.07 sec and a decreased resistive index (RI). Ultrasound contrast agents may improve the feasibility of the test in patients with a high body mass index or with decreased renal function. Multislice computed tomography (CT ) angiography provides high-resolution images of renal arteries with 1 mm (or less) slice thickness. Using an 80–100 mL bolus of iodinated contrast agent and maximum intensity projections (MIP) of reformatted slices, delineation of the true lumen and its separation from calcification are enhanced. Drawbacks of CT angiography are the ionizing

radiation and the need for iodinated contrast media, a problem when renal function is impaired. Sensitivity (92%) and specificity (99%) in detecting significant stenoses are high. FMD stenoses, which are more distal than atherosclerotic stenoses, can also be better characterized using multislice CT (Fig. 1b). Magnetic resonance (MR) angiography has moved from flow-enhanced (time-of-flight or phase-contrast) sequences to 3D breath-hold T1-weighted and contrastenhanced acquisitions (Fig. 1c). Fluoroscopic guidance of arrival of the bolus in the aorta improves the reproducibility of the technique. Parallel acquisition techniques allow improved spatial resolution without increasing the acquisition time and yield an isotropic resolution of 1 mm3 in a single breath-hold. These improvements are responsible for an improved grading of stenoses, which has the same interobserver variability as conventional angiography, and better detection of accessory arteries. Its performance is excellent, with sensitivity and specificity for diagnosis of significant RAS of 88–100% and 71–99%, respectively. Comparison of techniques: A recent meta-analysis compared the validity of CTangiography, MR angiography, and ultrasound for diagnosing RAS in patients suspected of having RVH. Receiver-operating characteristic curves found that CTA and gadolinium-enhanced 3D MR angiography were significantly better than the other diagnostic tests and seemed to be preferred in patients referred for evaluation of RVH.

Hypertension, Renal. Figure 1 Diagnosis of renal artery stenosis. (a) Color flow sonography showing an aliasing phenomenon on the proximal segment of right renal artery and flow changes evocative for significant stenosis: severe spectral broadening and increased peak systolic velocity around 5 m/sec. (b) CT angiography showing an ostial stenosis of the right renal artery. (c) MR angiography showing a post-ostial stenosis of the left renal artery.

Hypertension, Renal

Anatomic and Hemodynamic Consequences of RAS Detection of RAS further requires evaluating the severity of narrowing and its consequences on renal flow, renal parenchyma, and renal function, to improve the interobserver variability and to define predictive factors of improvement after revascularization.

Renal Anatomy changes The renal anatomy changes when the renal blood flow is significantly decreased. Two parameters have been proposed to evaluate this effect in RAS: . Measurement of renal length: To be significant, a length difference of 1 cm should be considered, attesting to a hemodynamically significant RAS (Fig. 2a). If the renal length is 0.8 was associated with a poor prognosis in patients with RAS shown by an absence of improvement of hypertension, renal function or kidney survival after revascularization.

Consequences of RAS on Renal Function Glomerular Filtration Baseline semiquantitative evaluation or quantification of glomerular filtration is neither sensitive nor specific enough to detect renal dysfunction due to RAS. Blockade of the renin–angiotensin system by angiotensin-converting enzyme (ACE) inhibitors, such as captopril, is mandatory for this purpose. Administration of ACE inhibitors decreases the vasoconstrictor-stimulating effect of angiotensin II, which produces a drop in glomerular filtration on the side of the stenosis. A positive test, defined as a functional change induced by the ACE inhibitor, demonstrates the diagnosis of “functional stenoses” producing renovascular hypertension, which means a real link between

Hypertension, Renal. Figure 2 Anatomical and hemodynamical consequences of renal artery stenosis. (a) MR Angiography shows a left renal astery stenosis showing a decreased of size of the kidney. (b) The MR phase-contrast acquisition shows flow velocity–time curves on each renal artery, with a normal curve on the right and an altered flow within left renal artery. (Courtesy of Dr Stephan Schoenberg, Muenchen, Germany)

H

928

Hyperthyroidism

Hypertension, Renal. Figure 3 Functional renal consequences of renal artery stenosis. (a) Positive captopril MAG3-Tc99 scintigraphy obtained in a patient with stenosis of the left renal artery: the baseline study shows symmetrical elimination of the tracer (right of the image). After captopril (left of the image), there is an accumulation of the tracer within the left kidney whereas it remains unchanged in the right. (b, c) Dynamic Gd-enhanced MR imaging of the kidneys before (a) and after (b) captopril administration in the same patient (from top-to-bottom then left-to-right): before captopril, the tubular phase with low signal intensity within medulla (arrows) and excretion of contrast medium within renal collecting system (double arrows) are symmetrical. After captopril, the right kidney shows a normal tubular phase (arrow) and normal excretion (double arrow) whereas tubular phase is delayed on the left but then enhances with time, extending within the cortex and providing a complete low signal intensity of the left kidney; this effect is related to a severe retention of the contrast agent within the entire left kidney induced by captopril.

RAS and hypertension. This test has traditionally been coupled with scintigraphy but more recently with dynamic MR imaging (Fig. 3).

Renal Perfusion Absolute quantification of parenchymal perfusion can be assessed with diffusible (as gadolinium chelates) or purely intravascular contrast agents (as iron oxide particles). A 50% decrease in cortical perfusion (from 400 to 200 mL/ min/100 g) has been observed below tight (80%) stenoses. However, its potential role in prognosis has not been demonstrated to date.

Hyperthyroidism The physiological manifestation of increased thyroid function, hormone production or insensitivity of the end organs to thyroid hormone. ▶Congenital malformations, Thyroid, and Functional Disorders ▶Thyroid Autoimmune Diseases

Hypertrophic Arthropathy, Secondary


4.

5.

Safian RD, Textor SC (2001) Renal artery stenosis. N Engl J Med 344:431–442 Scoble JE (1999) Atherosclerotic nephropathy. Kidney Int 56: S106–S109 Vasbinder GB, Nelemans PJ, Kessels AG et al (2001) Diagnostic tests for renal artery stenosis in patients suspected of having renovascular hypertension: a meta-analysis. Ann Intern Med 135:401–411 Radermacher J, Chavan A, Bleck J et al (2001) Use of Doppler ultrasonography to predict the outcome of therapy for renal artery stenosis. N Engl J Med 344:410–417 Taylor AT, Fletcher JW, Nally JV et al (2001–2002) Society of Nuclear Medicine Procedure Guideline for Diagnosis of Renovascular Hypertension 2.0. Society of Nuclear Medicine Procedure Guidelines Manual. 82–86

Secondary hypertrophic arthropathy is one of the more common and specific paraneoplastic diseases that may present as an oligoarthritis or polyarthritis of the distal joints with clubbing, painful periostitis of the distal long bones, and noninflammatory synovial effusions. Linear ossification of the distal long bones separated by a radiolucent zone from the underlying cortex can be seen in radiographs. Usually acromegaly induced bone changes are not as painful and linear ossifications along the bones are not found. ▶Acromegaly

Hypertrophic, Osteoarthropathy

929

Hypertrophic Pyloric Stenosis Thickening of the muscle wall and mucosa in the pylorus of the stomach resulting in gastric outlet obstruction and typically occurring in young infants and resulting in classic projectile vomiting. ▶GI Tract, Pediatric, Specific Problems

Hypertrophic Pyloric Stenosis (HPS) ▶GI Tract, Pediatric, Specific Problems


Hypertrophic, Osteoarthropathy. Figure 1 Radiograph of the right distal tibia and fibula in an anteroposterior and lateral view. The patient suffered from lung cancer and severe, burning pain in both distal legs. The periosteal appositions (white ▶) have a linear appearance typically found in secondary hypertrophic osteoarthropathy. (Courtesy of Prof. Thomas M. Link, San Francisco.)

B AUER J AN S Institut fu¨r Ro¨ntgendiagnostik, Technical University Munich, Germany [email protected]

Definition Hypertrophic osteoarthropathy is characterized by ▶periostitis, arthritis, and bilateral clubbing of the digits on the hands and feet. Hypertrophic osteoarthropathy may be divided into primary and secondary forms. ▶Primary hypertrophic osteoarthropathy can be hereditary or idiopathic and is very rare (3–5% of all cases of hypertrophic osteoarthropathy). ▶Secondary hypertrophic osteoarthropathy is often associated with intrathoracic neoplastic or inflammatory conditions.

Pathology/Histopathology Periostitis with new periosteal deposition of bone is the hallmark of the disease. While the new surface is rough in primary hypertrophic osteoarthropathy, it has a more linear appearance in secondary hypertrophic osteoarthropathy (Fig. 1). In histologic cross sections, the delineation between the old and new periosteal bone is less distinct in primary compared to secondary hypertrophic osteoarthropathy, where the new bone is often less dense than the original cortex. However, if

secondary hypertrophic osteoarthropathy has a longer history, e.g. in congenital cyanotic heart disease, the findings are more similar to that of primary hypertrophic osteoarthropathy. As in primary hypertrophic osteoarthropathy, osteolysis of the digits can also be found. Articular inflammation with joint effusion is found more commonly in secondary than in primary hypertrophic osteoarthropathy. The synovial membrane presents with a mild cellular hyperplasia and a thickening of the subsynovial blood vessels. The synovial fluid is viscous and noninflammatory. The ▶clubbed digits are not a result of the periostitis, but due to an internal soft tissue edema with collagen and fiber deposition. Arterio-venous anastomosis can develop in the nail bed. Thickening of vessel walls is seen with a lymphocyte infiltration. In electron microscopy, Weibel– Palade bodies and Golgi complexes can be found indicating cellular damage.

Clinical Presentation Hypertrophic osteoarthropathy is a syndrome characterized by digital clubbing, arthritis, periostitis, swelling and thickening of the skin of distal extremities and the head (pachydermia), as well as local nervous dysfunctions like hyperhidrosis (1). Digital clubbing was first described by

H

930


Hippocrates in the 5th century BC (2). A first description of hypertrophic osteoarthropathy was given by Friedreich in 1868. Primary hypertrophic osteoarthropathy is a hereditary autosomal dominant disease with variable penetrance. It affects males about seven times more often, is more severe in males, and is more common in African Americans than in Caucasians. The onset of primary hypertrophic osteoarthropathy is typically during adolescence. The disease often starts with enlargement of the hands and feet, producing a paw-like appearance. Digital clubbing is a frequent, though not invariable feature of hypertrophic osteoarthropathy. Its presence should not be equated with the syndrome, as several other disorders show clubbing without any other features of hypertrophic osteoarthropathy. Coarsening of the skin of the face and scalp, known as cutis verticis gyrata, may be observed as well as seborrheic dermatitis and palmoplantar hyperhidrosis. The clinical manifestation patterns are variable, ranging from the full syndrome (periostitis, pachydermia, and cutis verticis gyrata) or a partial expression (no involvement of the head) to a “forme fruste” (pachydermia without periostitis) (3). Primary hypertrophic osteoarthropathy usually shows a progression of about 10 years, when it arrests spontaneously. The life expectancy is normal, but disabling can occur if the axial skeleton is involved with stiffness, kyphosis, and neurological manifestations as spinal nerve roots get compressed by new bone formations. In 1890, Bamberger and Marie independently described secondary hypertrophic osteoarthropathy as a syndrome of digital clubbing and periostitis in patients suffering from chronic heart and lung diseases (4, 5). Secondary hypertrophic osteoarthropathy shows an even distribution between males and females with no racial preference. Life expectancy is not changed by secondary hypertrophic osteoarthropathy, but it is often limited by the underlying disease. Symptoms are somewhat different from primary hypertrophic osteoarthropathy. In secondary hypertrophic osteoarthropathy, the periostitis is often associated with a deep, burning pain. Arthritis is more frequent (up to 40% of patients) and can mimic symptoms of rheumatoid arthritis when affecting small joints. Cutis verticis gyrata is observed much less frequently. However, with a long duration of secondary hypertrophic osteoarthropathy, like in congenital cyanotic heart disease, symptoms and findings can fully correlate with those observed in primary hypertrophic osteoarthropathy.

Imaging Periostitis is the characteristic feature of the disease (6). Bones most affected are tibia, fibula, radius, and ulna. Periosteal appositions usually start developing at the distal end of the diaphysis. In case of primary hypertrophic

osteoarthropathy, the periosteal appositions extend to the epiphyseal regions and appear irregular and exuberant. The cranium and pelvis can be affected as well. In more advanced stages, osteolysis can appear at the distal phalanges and ossification of ligaments can be observed. Joint capsules, menisci, and discs can also ossify. Synostosis can develop, particularly between carpal and tarsal bones, as well as at the symphysis pubis. In an advanced stage of primary hypertrophic osteoarthropathy, synovialitis may affect joints causing limited motion due to periosteal appositions near the joint. In secondary hypertrophic osteoarthropathy, the appositions are linear thin lines separated from the cortical bone. They also can have an “onion-skin”-like appearance, but are usually not as irregular as in primary hypertrophic osteoarthropathy. In a later stage of the disease, the new periosteal bone fuses with the original cortex. The epiphysis is usually not involved. Besides the tibia, fibula, radius, and ulna, periosteal appositions can also appear at the femur, humerus, metacarpalia, metatarsalia, and the phalanges. Rarely an involvement of the scapula, the clavicle, rips, spine, or skull is found. As in rheumatoid arthritis, joint effusion can be found as well as osteoporosis near the joint. In contrast to arthritis, no joint space narrowing or erosive bony lesions are seen. The digital clubbing leads to radiographically evident soft tissue swelling.

Nuclear Medicine Radionuclide bone scanning using technetium Tc99m polyphosphate can reveal the disease in an early stage, where plain films are still negative. It shows increased uptake of the tracer in the cortex of diaphysis and metaphysis. In primary hypertrophic osteoarthropathy, the epiphysis is also involved. The activity usually increases bilaterally in a linear fashion. Clubbed digits and associated synovitis can both show increased activity, particularly in an early phase of the scan.

Diagnosis and Differential Diagnosis Differential diagnosis between primary and secondary hypertrophic osteoarthropathy can be achieved based on radiographic findings (irregular, with epiphyseal involvement in primary hypertrophic osteoarthropathy), clinical presentation (e.g. painful periostitis in secondary hypertrophic osteoarthropathy), and the patient’s family history (positive in 1/3 cases of primary hypertrophic osteoarthropathy). Other differential diagnoses to consider are ▶thyroid acropachy, ▶venous stasis, hypervitaminosis A, infantile cortical hyperostosis, acromegaly, and endosteal hyperostosis. In thyroid acropachy, pretibial myxedema and a

Hypopharynx

931

H Hypertrophic, Osteoarthropathy. Figure 2 Radiograph of the right distal tibia and fibula in an anteroposterior and lateral view. The patient suffered from venous stasis with ulceration in both lower limbs. The periosteal appositions are more pronounced in the fibula (white ▶). They have an irregular appearance and are not separated from the original cortex. (Courtesy of Prof. Thomas M. Link, San Francisco.)

Hypertrophic, Osteoarthropathy. Figure 3 Radiograph of the right distal tibia and fibula in a lateral and anteroposterior view of a 59-year-old patient with pachydermoperiostitis. Multiple regions of periosteal reaction are identified involving the diaphysis, metadiaphysis, metaphysis, and epiphysis of the tibia and fibula. (Courtesy of Prof. Kenneth E. Sack, San Francisco.)

history of thyroid dysfunction are present; the typical speculated periosteal appositions have a predilection for small tubular bones. Venous stasis shows a predilection for the lower extremity with signs of phlebolitis, soft tissue swelling, and ulceration. The periosteal appositions have an undulated osseous contour and cortical thickening appears as the appositions are not well separated from the original cortex (Fig 2). In hypervitaminosis A, soft tissue nodules may be present with epiphyseal deformities. Infantile cortical hyperostosis is characterized by cranial destructions and skeletal deformities. A differentiation from acromegaly is clinically possible. In endosteal hyperostosis, digital clubbing is not evident. Other diseases, like fibrous dysplasia, Paget’s disease, fluorosis, diffuse idiopathic skeletal hyperostosis are usually easily differentiated from hypertrophic osteoarthropathy (Fig. 3).

3.

4. 5. 6.

Touraine A, Solente G, Gole L (1935) Un syndrome osteódermopathique: La pachydermie plicatureé avec pachype´riostose des extre´mite´s. Presse Me´d 43:1820–1824 ¨ ber Knochenveranderugen bei chronishen Bamberger E (1891) U Lungen und Herzkrankheiten. Z Klin Med 18:193–217 Marie P (1890) De l’osteó-arthropathie hypertrophiante pneumique. Rev Me´d 4:146–148 Resnick D (1995) Diagnosis of Bone and Joint Disorders. 3rd edn. W.B. Saunders, Philadelphia

Hypervascular Liver Lesions HCC is the most common hypervascular malignant liver lesions. Other lesions prone to chemoembolization derive from neuroendocrine tumors. ▶Chemoembolization

Bibliography 1.

2.

Martinez-Lavin M, Matucci-Cerinic M, Jajic I et al (1993) Hypertrophic osteoarthropathy: consensus on its definition, classification, assessment and diagnostic criteria. J Rheumatol 20(8):1386–1387 Hippocrates (1849) Hippocrene Books: Prognostic: Regimen in Acute Diseases, Sacred Disease, Art, Breaths, Law, Decorum, Physician, Dentition. Harvard University Press, London, UK

Hypopharynx A musculomembranous conduit that lies behind the larynx. ▶Carcinoma, Hypopharynx

932

Hypophysis

Hypophysis ▶Pituitary Gland

Hypoplasia, Pancreatic In pancreatic hypoplasia the exocrine elements, or more rarely also the endocrine elements, are underdeveloped. ▶Congenital Abnormalities, Pancreatic

Hypoxic, Ischaemic Brain Injury M ARY R UTHERFORD Robert Steiner MR Unit Imaging Science Dept, MRC Clinical Sciences Centre Imperial College Hammersmith Campus, London, USA [email protected]

Synonyms Birth asphyxia; Hypoxic-ischaemic Neonatal; Perinatal; Seizures

encephalopathy;

Hypotelorism Definitions Narrowed intercanthal area. ▶Congenital Malformations, Musculoskeletal System

Hypothenar Hammer Syndrome Symptoms occurring in patients who use the palm of the hand for pushing, pounding, or twisting (mechanics, carpenters, etc.). Typical symptoms are more or less similar to those of Raynaud’s syndrome. ▶Ischemia, Brachial

Perinatal hypoxic-ischaemic (HI) brain injury is defined here as an injury arising from a lack of blood and oxygen, birth asphyxia, occurring during labour or delivery. There may be a clear sentinel event such as a cord prolapse or uterine rupture that precedes signs of foetal distress detected as abnormalities in foetal heart rate seen on cardiotocograph with or without meconium stained liquor. These signs are usually accompanied by a low cord pH < 7.1. The diagnosis of ▶hypoxic-ischaemic encephalopathy is made when such neonates require resuscitation at birth, having low Apgar scores 2.5 cm. Abnormality of axillary lymph nodes is suggested by an increase in nodal density and loss of the

1067

fatty hilum. Although nodal size and margins are variable in both benign and malignant lesions, a pattern of spiculated nodes has been reported to result from perinodal extension from a biologically aggressive carcinoma. Microcalcifications are commonly present in the primary breast carcinoma but are rarely seen in axillary node metastases. Axillary node calcifications are usually pleomorphic and faint. Similar calcifications may also be seen with metastatic ovarian and thyroid carcinoma. Coarse calcifications may be seen in granulomatous disease such as tuberculosis and fat necrosis (Fig. 1). Tuberculosis of the breast is rare, and tuberculous axillary lymphadenitis without evidence of tuberculous mastitis is even rarer. The HIV epidemic has resulted in a higher percentage of extrapulmonary tuberculosis, and more cases of tuberculous lymphadenitis can be expected because peripheral lymph node tuberculosis is one of the more common manifestations of extrapulmonary tuberculosis. If silicone drains from breast augmentation prostheses into the axillary nodes, it may be observed as extremely dense. Mammographically, intramammary lymph nodes appear in the typical morphology of a lymph node (1, 3) and are recognized as typical morphology of lymph nodes in imaging modalities. Sonographically, asymmetric thickening and focal lobulations in the cortex, partial or complete replacement of the echogenic fatty hilum, focal or diffuse decrease of echogenity within the nodal cortex, and marginal irregularity in some metastatic cases or increased transverse diameter of the lymph node parenchyma can occur. Overall size does not have clinical significance. Several series have reported color Doppler ultrasound (US) findings that indicated malignancy: displacement of hilar vessels, focal absence of perfusion, aberrant vessels, and subcapsular vessels. The longitudinal-transverse axis ratio and the presence of a peripheral versus central flow pattern were shown in recent publications on ▶color Doppler US (Fig. 2) Different flow patterns can be depicted on US; metastatic nodes tend to show a more peripheral vasculature than benign or reactive nodes. The usual figures for accuracy and specificity range between 85 and 88%. However, inherent limitations of US are well known, including the poor accessibility of deeply located nodes, in contrast to the use of MRI or CT. Combining Doppler US with microbubble contrast agents could overcome this drawback, and contrast-enhanced US lymphography following the interstitial injection of submicron-diameter bubbles could even compete with isotope sentinel node detection, as suggested by a recent animal study (1, 4). Lymph nodes located within the breast or lower axilla may be visualized on imaging studies of the breast. Unfortunately, imaging methods are often unable to differentiate inflammatory and neoplastic nodes.

L

1068

Lymphadenopathy

Lymphadenopathy. Figure 1 (a) Mammogram in MLO position shows large lymph nodes with coarse calcifications due to tuberculosis lymphadenitis in the axillary region. Its magnification is seen in (b). (c) Doppler view of the same patient

Owing to their low invasiveness and availability, crosssectional imaging techniques have gained widespread acceptance for lymph node imaging. However, whether with CT or MRI, criteria for lymph node metastases remain limited to size assessment. Maximum short axis diameters of normal-sized nodes usually do not exceed 1 cm. However, this value may vary depending on the node’s location. Additional morphologic features suggestive of normal nodes include the presence of a fatty hilum, regular contours, and homogeneous signal or density on MRI and CT. Nevertheless, micrometastases, such as in breast cancer, are commonly reported in normal-sized nodes. Improving the spatial resolution of cross-sectional imaging and, especially, being able to identify morphologic alterations of either lymph node cortex and sinus

could allow better distinction of the involved tumor and normal nodes. Several studies have assessed whether dynamic enhancement of lymph nodes following intravenous injection of extracellular contrast agents (gadolinium-based contrast agents on MRI) could improve the identification accuracy of nodal metastases. In contrast to extracellular contrast agents, iron oxide-based contrast agents reach lymph nodes indirectly (USPIO). Benign and reactive nodes share the presence of phagocytic or mononuclear cells in contrast to massively involved metastatic nodes, which fail to show a decreased signal intensity after USPIO injection: therefore, benign nodes can be distinguished from tumor-involved nodes regardless of size criteria. USPIO thus behave as “negative” contrast agents (4).

Lymphadenopathy

1069

L

Lymphadenopathy. Figure 2 (a) Benign and malignant lymph nodes in the same image. B-mode scan shows loss of the echogenic fatty hilum, focal lobulations in the cortex, and diffuse decrease of echogenicity within the nodal cortex in metastatic lymph node on the right and a typical benign node on the left side of the figure. (b) Benign reactive lymph nodes on color Doppler ultrasound with smooth cortex, small size, and central vascularization. (c) Malignant lymph nodes on color Doppler ultrasound reflecting large size, loss of echogenic hilus, and peripheral

Lymphoscintigraphy has recently regained interest in oncology for the surgical evaluation of the sentinel node, which is defined as the first lymphatic relay in the drainage territory of a primary tumor and is thought to be the first site of metastases in cases of lymphatic spread. The technique has evolved with the use of patent blue dye and 99mTc-labeled nanocolloids, both of which make it easier to identify the sentinel node intraoperatively. As initially demonstrated for both melanoma and breast cancer, lymphoscintigraphy allows avoidance of unnecessary invasive lymph node dissection procedures by identifying the sentinel node with the guidance of a gamma probe intraoperatively. In addition, it warrants better nodal staging because skip metastases during lymph node extension of a cancer are rare. Based on several reports, the success rate of sentinel node resection exceeds 90%, and the overall accuracy exceeds 95% (4) (Fig. 3). This technique may help decrease the extent of

axillary surgery and reduce the postoperative sequelae of axillary dissection, including lymphedema.

Indirect Interstitial CT and MR Lymphography The concept of indirect interstitial CT and MR lymphography closely matches that of sentinel node mapping. Small-sized contrast agents injected intradermally can reach lymphatic vessels owing to the increased permeability of the fenestrated endothelial lining of distal lymphatic capillaries. Similarly to radio-isotopes, such agents then follow the lymphatic flow and progressively converge toward afferent nodes. However, the literature remains limited to animal studies, and clinical trials on humans still have to be conducted. Similar results can be obtained with MR following interstitial administration of iron-oxide-based contrast agents or macromolecular gadolinium-based contrast agents (4).

1070

Lymphangioleiomyomatosis

Lymphangioma. Figure 3 Sentinel lymph node imaging shows increased accumulation of tumor site and mapping of sentinel node. (a) Anterior view. (b) Lateral view.

Intervention Female patients presenting with unilateral axillary masses and normal breast on physical examination can be a diagnostic and therapeutic challenge because there are many causes of axillary masses, including benign and malignant diseases. Mammographically detected enlarged axillary lymph nodes that show any abnormal radiological features or are clinically palpable merit further investigation, including biopsy (6). For those whose diagnosis of axillary adenopathy remains a mystery, fine-needle aspiration or excisional biopsy of the axillary node is necessary to determine the cause (3). Fine-needle aspiration biopsy is very useful in this setting. Although a negative diagnosis cannot definitely exclude malignant involvement, a possible diagnosis is quite reliable and can be used to influence treatment decisions (1).

manifests as numerous lung cysts, recurrent pneumothorax, chylous effusion, and lymphadenopathy. Lymphangioleiomyomatosis predominantly affects the lungs of women of childbearing age and individuals affected by tuberous sclerosis. On chest radiography the unusual combination of a reticular pattern and hyperexpansion of the lungs is typical, and pneumothorax or pleural effusion are commonly present. Widespread cysts are the characteristic feature on computed tomography. They are multiple, thin-walled, and rounded; vessels are seen at the periphery of the cysts and generally are not displaced, as is often the case in centrilobular and bullous emphysema. The cysts do not show preferential distribution, unlike the cysts in Langerhans cell histiocytosis, which typically spare the costophrenic regions. ▶Interstitial Lung Diseases, Unknown Etiology

Bibliography 1. 2.

3.

4. 5. 6.

Heywang-Ko¨brunner SH, Dershaw D, Schreer I (2001) Diagnostic Breast Imaging. 2nd edn. Thieme, Stuttgart, pp 313–324 Walsh R, Kornguth PJ, Soo MS et al (1997) Axillary lymph nodes: mammographic, pathologic and clinical correlation. Am J Roentgenol 168:33–36 Muttarak M, Chaiwun B, Peh WCG (2004) Role of mammography in diagnosis of axillary abnormalities in women with normal breast examination. Australas Radiol 48(3):306–310 Luciani A, Itti E, Rahmouni A et al (2006) Lymph node imaging: basic principles. Eur J Radiol 2006 Feb 10; (Epub ahead of print) Golder WA (2004) Lymph node diagnosis in oncologic imaging: a dilemma still waiting to be solved. Onkologie Apr 27(2):194–199 Lim ET, O’Doherty A, Hill AD et al (2004) Pathological axillary lymph nodes detected at mammographic screening. Clin Radiol 59(1):86–91

Lymphangioma J OHAN G. B LICKMAN 1, *, D EWI A SHI 2 1,

*Department of Radiology, UMC, Nijmegen, Netherlands 2 Department of Radiology, Faculty of Medicine, University of Indonesia, Cipto Mangunkusumo General Hospital, Salemba, Jakarta Pusat, Indonesia [email protected]

Synonym

Lymphangioleiomyomatosis

Congenital lymphatic malformation

Definition A rare idiopathic disease characterized by smooth muscle cell proliferation occurring both in the lungs and in the mediastinum and retroperitoneum. In the thorax it

Lymphangiomas are uncommon hamartomatous congenital malformations of the lymphatic system that

Lymphangioma

1071

involve both the skin and subcutaneous tissues. Two major groups based on depth and size of abnormal lymph vessels are defined: Group I consists of lesions located superficially, and group II consists of lesions more deeply located (such as cavernous lymphangioma and ▶cystic hygroma). The female-to-male ratio is equal. About 50% of lymphangiomas are seen at birth, and most are evident by the time the patient is 5 years old.

Pathology/Histopathology Dilated lymph channels cause the papillary dermis to expand. These channels are more numerous in the upper dermis and often extend into the subcutis. These vessels contain smooth muscle. The lumen is filled with lymphatic fluid and often contains lymphocytes, red blood cells, and neutrophils.

Lymphangioma. Figure 1 without vascular flow.

US shows hypoechoic mass

Clinical Presentation Lymphangiomas can occur anywhere in the skin and mucous membranes. The most common sites are the head and neck, followed by the proximal extremities, the buttocks, and the trunk. Sometimes they can be found in the bowel wall, pancreas, and mesentery. Lymphangioma circumscriptum (group I) involves small clusters of vesicles measuring about 2–4 mm on the skin. These clear vesicles can vary from pink to red to black secondary to hemorrhage. This lesion can have a warty appearance and as a result can be confused with warts. Cavernous lymphangiomas typically appear as subcutaneous nodules with a rubbery consistency. The overlying skin has no lesions or changes. The area of involvement varies, ranging from lesions smaller than 1 cm in diameter to larger lesions that involve an entire limb. Cystic hygroma commonly occurs in the neck and parotid gland. It is a soft and translucent mass, with diameters larger than those of cavernous lymphangiomas.

L

Lymphangioma. Figure 2 On T2WI a hyperintense lesion is seen on the plantar side of the foot.

Magnetic resonance imaging (MRI) shows cyst characteristics ranging from fluid (hypointense on T1weighted images) to fat (hyperintense on T2-weighted images; Fig. 2).

Nuclear Medicine Imaging Nuclear medicine has no place in lymphangioma evaluation. Plain films may show a large mass displacing adjacent structures. Ultrasound (US) reveals an anechoic mass that is multilocular with fine septations. Sometimes the loculated space may be hypoechoic or hyperechoic depending on the chyle or blood content (Fig. 1). Computed tomography (CT) also demonstrates septae and loculation. The attenuation value ranges from near fat to near water, depending on the fluid content. Postcontrast, the septa may enhance.

Diagnosis The diagnosis of lymphangioma is based mainly on clinical history, physical examination, US, and conventional light microscopy. US and CT can delineate septae and cysts. Both CT and MRI can define the extension of the lesion and involvement in surrounding tissues. If the lesions contain

1072

Lymphangioma

large cystic structures, they are treated with percutaneous sclerosing. Immunohistochemical investigation for difficult cases is useful for differentiating between lymphangioma and hemangioma. Test results with factor VIII-related antigen are positive for hemangioma and negative for lymphangioma. Immunohistochemical studies for laminin show the typical multilayered basal lamina blood vessels in the case of hemangioma and discontinuous basal lamina in lymphangioma. ▶Congenital Malformations, Oral Cavity ▶Cystic Neoplasms, Pancreatic

Lymphedema Lymphatic edema of the arm, which affects 10–20% of patients who have suffered some kind of axillary surgery, especially if the patient also underwent radiation therapy and/or the lymph nodes contained cancer cells. Early signs of lymphedema are a feeling of tightness, pain, heaviness, swelling, redness, and less movement or flexibility. ▶Breast Conserving Therapy ▶Lymphopoietic System, Diseases of the

Bibliogaphy 1. 2.

3.

4.

Barnes PD (1998) Neuroimaging. In: Hans Blickman (ed) Paediatric Radiology: The Requisites. 2nd edn. Mosby St. Louis, pp 265–266 Oslo (2001)Paediatric Imaging. In: Carty H (ed) The Encyclopedia of Medical Imaging, vol 7, Publisahed by The NICER Institute, Oslo, p 317 Schlesinger AE, Parker B (2004) Tumor and tumor-like condition, In: Kuhn JP, Slovis TL, and Haller JO (eds) Caffey’s Pediatric Diagnostic Imaging. 10th edn. Mosby, Philadelphia, pp 1628–1629, p 1913 http://www.emedicine.com/derm. Lymphangioma last update November 8, 2005

Lymphangioma

Lymphoepithelial Cyst Lymphoepithelial cyst of the pancreas is extremely rare. It is a well-defined cystic lesion lined by mature keratinizing squamous epithelium surrounded by a rim of lymphoid tissue. They present predominantly in middle-aged or elderly males with non-specific symptoms. At imaging lymphoepithelial cysts show mainly a cystic appearance, but they may contain some debris as a result of their keratin content. The differential diagnosis includes all cystic lesions, but particularly dermoid cysts. ▶Cystic Neoplasms, Pancreatic

▶Cysts, Cerebral and Cervical, Childhood

Lymphoma Lymphangioma, Hepatic Convolute of dilated lymphatic channels that may contain proteinaceous material or blood. Usually solitary hepatic lesion, if multiple lesions are seen in multiple organs the condition is referred to as lymphangiomatosis. ▶Hepatic Pediatric Tumors, Benign

▶Neoplasms, Chest, Childhood

Lymphoma Hepatic, Primary Malignancy limited to the hepatic parenchyma with no evidence of further nodal or extranodal involvement. ▶Lymphoma, Hepatic

Lymphangiomatosis, Splenic Diffuse involvement of the spleen by multiple lymphangiomas. It can be part of a generalized multisystem syndrome involving the liver, bone, lung, and/or brain, and it presents during infancy, childhood, or adolescence. ▶Neoplasms, Splenic, Benign

Lymphoma Hepatic, Secondary Hepatic involvement in patients with systemic lymphoma. ▶Lymphoma, Hepatic

Lymphoma, Breast

1073

Lymphoma, Breast E LS M. VANHERREWEGHE , G EERT M. V ILLEIRS Department of Radiology, Ghent University Hospital Ghent, Belgium [email protected]

Definition Breast lymphoma is an uncommon haematologic malignancy of the breast, representing about 0.5% of all malignant lymphomas and 2% of extranodal lymphomas. In most cases, the breast is involved as part of disseminated extranodal disease, usually ▶non-Hodgkin lymphoma and very rarely Hodgkin lymphoma. Primary lymphoma of the breast, with or without axillary nodal involvement, is unusual.

Pathology/Histopathology The majority of mammary lymphomas are of the ▶B-cell type, with rare occurrences of T-cell lymphoma. Microscopically, they appear similar to extramammary lymphoma and consist of diffuse or focal (single or multiple) infiltration of the breast parenchyma with dense populations of lymphoma cells. They are fairly well circumscribed from the surrounding tissue, but usually show some marginal irregularities or indistinctness, because of their peripheral reactive lymphoid infiltrate. In contrast to primary breast carcinomas, however, they show no spiculations, since they do not elicit a desmoplastic reaction.

Clinical Presentation Most breast lymphomas present as a rapidly growing painless palpable lump that is fairly mobile and relatively soft and elastic. Skin thickening, enlargement of the breast, and systemic symptoms such as fever, weight loss and night sweats may also be present. At the time of diagnosis, bilateral disease is present in about 10% of patients and enlarged axillary lymph nodes may be associated.

Mammography The mammographic instance of breast lymphomas is non-specific and there is no correlation between the

Lymphoma, Breast. Figure 1 Breast lymphoma presenting as an oval fairly well-circumscribed mass in the upeer outer quadrant. Note that no spiculations or calcifications are present.

different subtypes of lymphoma and their mammographic features. Breast lymphomas can present as solitary or multiple, round or lobulated, fairly well-circumscribed nodules with smooth or indistinct borders, but without spiculations or calcifications (Fig. 1). In diffuse involvement, an asymmetric density with skin thickening and coarsening of the trabecular framework of the breast may be observed. Axillary nodal involvement is mammographically indistinguishable from nodal metastases of primary breast carcinoma.

Sonography Breast lymphomas may be recognized as single or multiple, round or lobulated, heterogeneously hypoechoic masses with variable through-transmission and fairly well-circumscribed margins (Fig. 2). When there is peripheral reactive lymphoid infiltration, an ill-defined echogenic rim may surround the lesion. The skin may also be thickened and the lymphomatous tissue may show decreased echogenicity (Fig. 3).

Magnetic Resonance Mammography The magnetic resonance instance of breast lymphoma is non-specific. On T1-weighted images, fairly well circumscribed hypointense masses that enhance rapidly, sometimes in a ringlike fashion, have been reported. On T2-weighted images, the masses may be ill defined or show a hyperintense halo. Magnetic resonance mammography can be helpful in determining the extent of involvement, but differentiation from primary breast carcinoma is not possible.

L

1074

Lymphoma, Genito Urinary Tract

Bibliography 1. 2.

3. 4. 5.

Lymphoma, Breast. Figure 2 Sonogram of a lobulated well-circumscribed heterogeneously hypoechoic breast lymphoma.

Meyer JE, Kopans DB, Long JC (1980) Mammographic appearance of malignant lymphoma of the breast. Radiology 135:623–626 Hugh JC, Jackson FI, Hanson J et al (1990) Primary breast lymphoma. An immunohistologic study of 20 new cases. Cancer 66:2602–2611 Giardini R, Piccolo C, Rilke F (1992) Primary non-Hodgkin’s lymphomas of the female breast. Cancer 69:725–735 Bobrow LG, Richards MA, Happerfield LC et al (1993) Breast lymphomas: a clinicopathologic review. Hum Pathol 24:274–278 Brogi E, Harris NL (1999) Lymphomas of the breast: pathology and clinical behaviour. Semin Oncol 26:357–364

Lymphoma, Genito Urinary Tract E RIC DE K ERVILER 1 , O LIVIER M ATHIEU 1 , C E´ DRIC DE B AZELAIRE 1 , PAULINE B RICE 2 , J ACQUES F RIJA 1 1

Service de Radiologie, Hoˆpital Saint-Louis, Paris, France 2 Hoˆpital de jour d’He´matologie, Hoˆpital Saint-Louis, Paris, France [email protected]

Synonyms Lymphoma, Breast. Figure 3 Cutaneous B-cell lymphoma of the breast presenting as a lobulated fairly well-circumscribed hypoechoic mass adherent to the skin.

Lymphomas of the male genital tract; Lymphomas of the urinary tract

Nuclear Medicine

Definition

PET/CT scan may be useful to detect extramammary localizations of lymphoma and to distinguish between breast involvement in extranodal lymphoma or primary mammary lymphoma.

Lymphomas are tumors of the lymphatic tissue and account for 5% of all cancers. Although, lymphomas mainly involve lymphatic organs (lymph nodes, bone marrow, thymus, spleen, and mucosa-associated lymphoid tissue), they can also involve organs systems not containing lymphoid tissue. The most frequent extranodal localizations are digestive tract, skin, bone marrow, and head and neck. Genitourinary (GU) lymphomas represent 2 to 10% of all lymphomas. The GU tract may be affected either primarily or much more frequently secondarily in disseminated lymphomas or at recurrence.

Diagnosis Since most mammary lymphomas are part of a disseminated extranodal disease, any palpable abnormality in the breast in a patient with known lymphoma should raise the suspicion of mammary involvement. The diagnosis of primary breast carcinoma is limited to those patients without evidence of extramammary lymphoma at the time of diagnosis. Since the clinical and imaging presentation of both primary and secondary lymphoma are generally non-specific, the diagnosis should be obtained through cytological smear, core biopsy or excisional biopsy.

Pathology/Histopathology Most of lymphomas involving the GU tract are aggressive non-Hodgkin’s lymphoma (mainly of B-cell origin), but low-grade lymphomas may be also seen. The appearance of a renal lesion in a patient followed up for a low-grade


1075

lymphoma suggests a transformation into a high-grade lymphoma. Usually, GU involvement is observed in patients with disseminated lymphoma, having other extranodal lesions. Primary renal lymphoma is a discussed entity, which has been reported in less than 0.7% of extranodal lymphoma series in the literature. Indeed, lymphoid tissue is not present in normal kidneys, and extrarenal lesions are frequently diagnosed at autopsy.

Clinical Presentation Patients with lymphomas of the GU tract long remain asymptomatic and the diagnosis is usually made at CT. Clinical symptoms are usually nonspecific: pain, palpable mass, or less commonly, hematuria, edema, and hypertension. Serum creatinine level is increased in 10–40% of cases. In aggressive lymphomas, B symptoms such as night sweats, fever, or weight loss may be present.

Lymphoma, Genito Urinary Tract. Figure 1 Contrastenhanced CT examination in a patient with diffuse large B-cell lymphoma shows bilateral hypoattenuating nodules and retroperitoneal lymph nodes.

Imaging Urinary Lymphomas Lymphomas of the urinary tract mainly involve kidneys, and may exhibit different patterns, depending on the mechanism of spread (1). The hematogenous spread usually results in tumor foci in the renal cortex, with progressive infiltration and formation of expansile masses displacing the collecting system, ultimately leading to complete replacement of kidney parenchyma. Infiltrative growth from retroperitoneal or perirenal preserve renal contours but invades kidney hilum and may compress excretory cavities.

Masses/Nodules Masses are one of the most common patterns of renal lymphomas. They consist of multiple bilateral nodules seen within kidney parenchyma, and usually reflect advanced disease in patients presenting for staging evaluation (Fig. 1). In some cases, lymphomas manifest as a solitary tumor simulating a renal cell carcinoma. Unique or multiple, lymphomatous nodules appear as round solid lesions in the renal parenchyma, hypoattenuating at CT, with little enhancement after contrast administration. At sonography, nodules appear hypoechoic with little vascularity. Nodules may displace calices and modify renal contours.

L

Lymphoma, Genito Urinary Tract. Figure 2 Contrastenhanced CT examination in a young patient with lymphoblastic lymphoma shows lymphomatous infiltration of both kidneys, which causes nephromegaly and heterogeneous pattern of enhancement.

Then, lymph nodes grow up and typically result in a bulky retroperitoneal mass that invades the renal hilum, and secondarily spreads toward the parenchyma (Fig. 2). As a result, the renal collecting system may be obstructed, leading to renal failure. Before initiating therapy, placing a double-J ureteral catheter allows decreasing serum creatinine level and protecting kidneys from chemotherapy toxicity.

Contiguous Retroperitoneal Extension

Infiltrative Disease

Contiguous retroperitoneal involvement is also a common pattern, observed in 25–30% of patients (2). Initially, lymphoma is confined to retroperitoneal lymph nodes.

The infiltrative form of the disease is due to lymphomatous proliferation within the interstitium of the kidney. It is observed in 20% of patients, is almost always bilateral,

1076


Lymphomas involving the bladder are exceptional. They usually result from a contiguous pelvic extension in patients with aggressive lymphomas and bulky iliac lymph nodes or pelvic masses. They usually appear as a defect into the bladder resulting from the tumor bulging. Few reports of lymphoma involving the urethra have been published.

Genital Lymphomas

Lymphoma, Genito Urinary Tract. Figure 3 Contrastenhanced CT examination in a patient with diffuse large Bcell lymphoma shows a large mass in the right retroperitoneum with extension into the right kidney. The heterogeneous pattern of enhancement in the kidneys suggests also an infiltrative pattern.

but may induce subtle changes, which may be misdiagnosed in the early stages of the disease. The main pattern of lymphomatous infiltration is enlarged kidneys with preserved contours (Fig. 3). Kidneys enhance only slightly after contrast injection due to impaired renal function. In some cases, contrast injection delineates more focal areas of infiltrations presenting as poorly defined hypoattenuating foci (3).

Perirenal Infiltration Perirenal infiltration is rare in an isolated form without parenchymal involvement or retroperitoneal lymph nodes. It may appear as a thickening of the Gerota fascia, a renal sinus infiltration, or perirenal masses (2). The diagnosis may be difficult if the patient does not receive contrast material. However, after contrast, a sharp delineation between tumor and normal kidney is clearly seen since lesions usually enhance much less than normal kidney parenchyma. When present, this pattern is highly suggestive of lymphoma.

Excretory Cavities Involvement Lymphomas involving the excretory cavities alone are very rare. They are more frequently seen at relapse than at initial diagnosis. The diagnosis ureteral lymphomas can be made on a follow-up CT, or in a patient presenting with hydronephrosis. The typical pattern is a circumferential thickening of the ureteric wall, either segmental or total, sometimes associated with an infiltration of the surrounding fat.

Genital lymphomas in male patients are mainly represented by testicle lymphoma. Lymphomas involving the prostate, the epididymis and the seminal vesicles are much less frequent.

Testicles Primary testicular lymphomas are rare and represent 0.5–2% of all lymphomas. They are more frequently observed in elderly patients. Testicular lymphomas are usually aggressive lymphomas of diffuse large B-cell type. In the largest published series, the diagnosis was obtained by orchidectomy in 95% of cases, and the disease was at a localized stage in 79% of cases (4). Lesions are unilateral in 90% of cases, bilateral in 10%, and are best seen by sonography. Testicular lymphomas are hypoechoic relative to the surrounding parenchyma in about 95% of cases. The frequency of disseminated lymphomas in patients with testicular lymphomas depends on the examinations performed, and ranges between 13 and 66% in the literature. This kind of lymphoma is usually considered as a lymphoma having a pejorative outcome, with 50% extranodal relapse at 2 years, mainly in the central nervous system.

Prostate Prostate lymphomas are very rare, and represent less than 0.1% of all lymphomas. They are usually diagnosed in elderly patients or on prostatectomy specimens. As opposed to prostate cancer, any part of the prostate may be involved. Clinical symptoms are the same as those seen in adenomas. Imaging studies detect a mass but the prostate specific antigen remains normal. Diagnosis relies on biopsy.

Lymphoma in Immunocompromised Patients Immunocompromised patients may develop lymphoma more often than healthy subjects.

Acquired Immunodeficiency Syndrome It is estimated that 10% of HIV positive patients will develop lymphoma. Lymphomas in AIDS patients are

Lymphoma, Hepatic

usually high-grade lymphomas (diffuse large B-cell lymphomas, B-cell immunoblastic lymphomas, and Burkitt lymphomas), presenting at stage IV with multiple extranodal lesions. Among these, kidney is frequently involved. The clinical course is more aggressive, and the disease is both more extensive and less responsive to chemotherapy than in general population.

Bibliography

Kidney Transplant Recipient

5.

Lymphoma is a serious complication following solid organ transplantation. The risk for lymphoma during the first posttransplant year is 20- and 120-fold higher after kidney or heart transplantation, respectively, than that in the general population. Approximately 2% of renal transplant recipient will develop lymphoma. Whereas literature reports agree that the majority of lymphomas occur during the first few months after transplantation, little is known about the long-term risk for lymphoma development. Posttransplant non-Hodgkin lymphomas differ from lymphomas in the general population in histopathologic findings, higher extranodal involvement, a more aggressive clinical course, poorer response to conventional therapies, and poorer outcomes like AIDS-related lymphomas. The renal transplant itself may be involved, mainly in the nodular form.

Diagnosis Apart from the testicular lymphoma, which is often localized, the diagnosis of lymphoma in general is evoked on a suggestive distribution of lesions at CT that is multiple lymph nodes associated with extranodal lesions. However, in patients with one renal nodule and sentinel retroperitoneal lymph nodes, the question of a renal cancer raises. The final diagnosis of lymphoma with appropriate subtyping is based on histopathology only. The progresses made in image-guided needle biopsy techniques and immunochemistry studies make possible to obtain a definite diagnosis of lymphoma in up to 93% of cases (5). This minimally invasive procedure is safe, easy to perform in the kidneys and reliable. In order to optimize tissue sampling and allow immunochemistry and cytogenetic studies, a coaxial technique is highly recommended. The use of large-cutting needles is possible since lesions are usually hypovascularized. Therefore, open surgery may be avoided as first diagnostic procedure in the management of patients with urinary lymphoma, and should now be fortuitous in case of an equivocal mass or a complication. In patients with testicular masses, the diagnostic approach still remains surgical biopsy or orchidectomy.

1. 2. 3. 4.

1077

Hartman DS, David CJ, Goldman Jr et al (1982) Renal lymphoma: radiologic-pathologic correlation of 21 cases. Radiology 144:759–766 Cohan RH, Dunnick NR, Leder RA et al (1990) Computed tomography of renal lymphoma. J Comput Assist Tomogr 14:933–938 Urban BA, Fishman EK (2000) Renal lymphoma: CT patterns with emphasis on helical CT. Radiographics 20:197–212 Zucca E, Conconi A, Mughal TI et al (2003) Patterns of outcome and prognostic factors in primary large-cell lymphoma of the testis in a survey by the International extranodal lymphoma study group. J Clin Oncol 21:20–27 de Kerviler E, Guermazi A, Zagdanski AM et al (2000) Image-guided core-needle biopsy in patients with suspected or recurrent lymphomas. Cancer 89:647–652

Lymphoma, Hepatic C ARLO B ARTOLOZZI , DAVIDE C ARAMELLA , PAOLA VAGLI Department of Diagnostic and Interventional Radiology, University of Pisa, Pisa, Italy [email protected]

L Synonyms Hepatic lymphoproliferative disease

Definition Hepatic lymphoma is considered to be primary when the malignancy is limited to the hepatic parenchyma and there is no evidence of further nodal or extranodal involvement. Hepatic involvement in patients with systemic lymphoma is defined as ▶secondary hepatic lymphoma.

Pathology/Histopathology Primary lymphoma of the liver, first documented in 1965, is extremely rare, representing only 0.4% of all extranodal lymphomas. However, the liver is often a secondary site of lymphomatous involvement, occurring in approximately half of the patients with non-Hodgkin’s lymphoma and in 60% of patients with Hodgkin’s disease at autopsy (1, 2). ▶Primary hepatic lymphoma typically occurs during the fifth decade of life and has a male predominance. It is more common among immunocompromised patients, and a strong association has been identified with chronic hepatitis C.

1078

Lymphoma, Hepatic

▶AIDS-related lymphomas represent a nosological entity whose characteristic features resemble those of lymphomas seen in organ transplant recipients. Extranodal disease is more frequent, with marked predilection for the brain and the gastrointestinal tract. The liver is the second most common site of abdominal involvement after the bowel. AIDS-related lymphomas are highly aggressive tumors with poorly differentiated histological subtypes, most frequently non-Hodgkin’s types, and are associated with a poor prognosis. Patients are typically diagnosed with advanced stages of disease (3). The interrelationship between chronic hepatitis C and primary hepatic lymphoma remains unclear, although an increased incidence of hepatic lymphomas has been reported in patients affected by chronic hepatitis C. It has been documented that the hepatitis C virus may promote the clonal expansion of B-cells as occurs in associated cryoglobulinemia; additionally, viral cRNA has been detected within extrahepatic lymphoma tissue. While Hodgkin’s disease, non-Hodgkin’s lymphoma, and various leukemias may secondarily involve the liver, virtually all primary hepatic lymphomas are nonHodgkin’s types. Most cases of primary lymphoma of the liver are of intermediate or high grade according to the classification of the working formulation for clinical usage. Diffuse large cell lymphoma is the most commonly encountered histological subtype, and most cases are B-cell lymphomas. The rare Hodgkin’s lymphoma is generally of Reed–Stemberg cell type. Macroscopically the primary hepatic lymphoma usually occurs as a single large lobulated mass involving both hepatic lobes, and less often as multiple lesions. Conversely, Hodgkin’s disease occurs more frequently as miliary lesions than masses. In both Hodgkin’s and non-Hodgkin’s lymphoma, initial involvement is seen in the portal areas, where the lymphoid tissue is mostly localized. A few cases of primary hepatic mucosa-associated lymphoid tissue (MALT) lymphomas have been described in the literature. These lymphomas have a good prognosis and are low-grade B-cell lymphomas, occurring in a variety of extranodal sites but rarely as primary hepatic lymphomas. Different from other MALT lymphomas, these are not associated with autoimmune disorders. Typically, primary ▶hepatic MALT lymphomas appear as lymphoid infiltrates invading the liver in a serpiginous fashion, with nodular entrapment of normal liver (4). Secondary hepatic involvement by malignant lymphoma (both Hodgkin’s and non-Hodgkin’s types) often appears diffusely infiltrative. Distinct nodular lesions are exceedingly uncommon in Hodgkin’s disease but occur in half of the patients with non-Hodgkin’s lymphoma. However, the nodule size is usually small, less than 1 cm (5).

Clinical Presentation The most common early signs and symptoms are epigastric and right upper quadrant pain or discomfort and hepatomegaly without splenomegaly. Fever, weight loss, and night sweats may be present. Jaundice and ascites are uncommon. Liver function tests are usually normal except for elevated lactic dehydrogenase and alkaline phosphatase levels. Alpha-fetoprotein and carcinoembryonic antigen levels are invariably normal. In immunocompromised patients the disease demonstrates an aggressive course.

Imaging Ultrasound (US) displays primary liver lymphoma as a large hypoechoic mass, sometimes anechoic with a cystic-like appearance and typically with homogeneous echo structure and regular margins. In the diffuse forms, the echogenicity and architecture may be both normal and subverted. Computed tomography (CT) findings of hepatic lymphoma consist of a large, lobulated, usually homogeneous, sharply marginated, solitary mass that is hypodense on nonenhanced scan and appears demarcated and lightly enhancing after contrast medium administration (Fig. 1). Low enhancement is due to tumoral hypovascularity. The degree of necrosis and the presence of calcifications can be variable. Typically, lymphomas do not infiltrate surrounding structures such as the vascular vessels, which are dislocated instead by mass effect (2). On magnetic resonance imaging (MR) the tumor usually appears homogeneously hypointense on T1-weighted images, isointense to the spleen in T2-weighted images (because of its rich cellularity), and minimally enhanced on early postgadolinium spoiled gradient echo images (Fig. 2). AIDS-related lymphomas may show atypical imaging patterns such as a rim enhancement or an inhomogeneous appearance, simulating a complex mass. The appearance of primary liver lymphoma is in contrast to secondary involvement, which is diffusely infiltrative or micronodular (Fig. 3) in most cases and is frequently associated with the presence of retroperitoneal lymphadenopathy. When discrete lesions are present, they appear similar to those of primary hepatic lymphoma (5).

Nuclear Medicine Imaging using radionuclide 67Ga has proved useful for evaluating posttreatment lymphoma residual masses,

Lymphoma, Hepatic

1079

Lymphoma, Hepatic. Figure 1 Primary hepatic lymphoma. Computed tomography scans shows a large mass involving most of the left hepatic lobe, causing compression but not infiltration of the surrounding structures (hepatic vessels). The lesion appears hypodense both in the arterial (left) and portal (right) venous phases.

L

Lymphoma, Hepatic. Figure 2 Magnetic resonance study of the same lesion of in Fig. 1: T1-weighted image shows a homogeneous hypointense mass (upper left). T2-weighted image at the same level shows hyperintensity of the mass that is almost isointense to fat (upper right). After Gd–DTPA administration, the nodule reveals moderate enhancement in the arterial phase (lower left), with a hypointense pattern in the delayed phase (lower right).

although the presence of normal bowel mucosal uptake and a relatively poor spatial resolution reduce the diagnostic accuracy of this technique. Fluorodeoxyglucose positron emission tomography (FDG-PET) imaging is a tomographic technique that allows precise localization of active glucose metabolism, suggesting potential sites of active disease.

Diagnosis Imaging features of primary liver lymphoma are not specific when using a single modality. On the contrary,

the integration of findings obtained with different imaging techniques may suggest the diagnosis. In the differential diagnosis, a solitary metastasis must be considered; however, a unique large metastasis is very unusual in the absence of a known primary tumor. Classic hypervascular hepatocellular carcinoma, focal nodular hyperplasia, and hepatic hemangioma can be excluded by contrast-enhanced CT if the typical hypodensity of lymphoma is demonstrated. The combination of clinical history, liver findings, and the presence of other supportive findings such as retroperitoneal lymphadenopathy can lead to the correct differentiation between primary hepatic lymphoma and other liver diseases.

1080

Lymphoma, Pancreatic

Lymphoma, Hepatic. Figure 3 Secondary lymphomatous hepatic involvement. Ultrasound image shows an aspecific pattern consisting of multiple hypoechoic nodular areas of infiltration of the liver parenchyma.

Finally, the appearance of a normal biliary tree and the absence of a late enhancement on dynamic MRI allows the diagnosis of cholangiocarcinoma to be ruled out (2).


4.

5.

Ryan J, Straus DJ, Lange C et al (1988) Primary lymphoma of the liver. Cancer 61(2):370–375 Sanders LM, Botet JF, Straus DJ et al (1989) CT of primary lymphoma of the liver. Am J Roentgenol 152(5):973–976 Rizzi EB, Schinina V, Cristofaro M et al (2001) Non-Hodgkin’s lymphoma of the liver in patients with AIDS: sonographic, CT, and MRI findings. J Clin Ultrasound 29(3):125–129 Kirk CM, Lewin D, Lazarchick J (1999) Primary hepatic B-cell lymphoma of mucosa-associated lymphoid tissue. Arch Pathol Lab Med 123(8):716–719 Gazelle GS, Lee MJ, Hahn PF et al (1994) US, CT, and MRI of primary and secondary liver lymphoma. J Comput Assist Tomogr 18(3):412–415

Although representing classic symptoms of nodal nonHodgkin’s lymphoma, fever and night sweats are rare in patients with primary pancreatic involvement. Jaundice represents an infrequent finding, even with large lesions involving the pancreatic head. On sonography, primary ▶pancreatic lymphoma usually appears as a homogeneous hypoechoic mass with enlarged peripancreatic lymph-nodes. At CT, pancreatic lymphoma generally appears as a larger, bulky tumor and is often seen as a homogeneous mass. Enhancement after administration of contrast medium is usually poor. Two distinct CT patterns have been described, including focal and circumscribed masses and diffuse enlargement of the gland. The diffuse infiltrating pattern may mimic the imaging findings of pancreatitis with gland enlargement and irregular infiltration of the peripancreatic fat. Dilatation of Wirsung’s duct is usually mild. Encasement of the peripancreatic vessels may occur, but vascular obstruction is uncommon despite the presence of a large tumor, a helpful distinguishing feature from other malignant tumors. The presence of associated enlarged peripancreatic lymph nodes also favors the diagnosis of lymphoma. At MR examination, lymphoma appears as a low-signalintensity homogeneous mass within the pancreas on T1-weighted images with poor enhancement after gadolinium administration. On T2-weighted images, the lesion shows a more heterogeneous appearance. Only mild pancreatic ductal dilatation is usually visible on MRCP. Despite the rarity of primary pancreatic lymphoma, it is important to differentiate this entity from adenocarcinoma as the management of these two conditions is different. If the imaging findings are suggestive of lymphoma, a definitive diagnosis must be achieved by percutaneous or endoscopic biopsy averting unnecessary surgery, because lymphomas are generally treated by chemotherapy. ▶Carcinoma, Pancreatic

Lymphomas of the Male Genital Tract Lymphoma, Pancreatic Lymphoma (predominantly the non-Hodgkin B cell type) involves the pancreas secondarily in approximately 30% of patients with widespread disease. Primary lymphoma of the pancreas is uncommon, representing 1–5% of cases of extranodal lymphomas. The most common clinical findings are abdominal pain and weight loss.

▶Lymphoma, Genito Urinary Tract

Lymphomas of the Urinary Tract ▶Lymphoma, Genito Urinary Tract

Lymphopoietic System, Diseases of the

Lymphopoietic System, Diseases of the O TMAR S CHOBER Department of Nuclear Medicine, Coordinator Collaborative Research Center, Molecular Cardiovascular Imaging (SFB 656-MoBil), University Hospital Mu¨nster, Albert-Schweitzer-Str. 33, D48149 Mu¨nster, Germany [email protected]

1081

lymph node stations. Admittedly, the latter techniques stand out for their more detailed representation, but they are more complicated and more invasive, because the lymphatic vessels must be cannulated directly. The nodi lymphatici iliacae internae and mammariae internae are not amenable to radiological imaging but only to scintigraphic visualization after interstitial injection of the tracer in the surrounding connective tissue. In the case of suspected lymph node metastasis, ultrasound and computed tomography or combined 18 F-FDG-PET-CT should be performed in order to determine the size, shape, and metabolic activity of the lymph nodes (cf. oncology).

Synonyms Nuclear Medicine Lymphedema; Lymph node metastasis

Definition 1. Lymphedema is defined as an increased accumulation of lymphatic fluid in the extravascular compartment. 2. Lymph node metastasis is defined as regional or distant lymphogenic spread of malignant solid tumors.

Pathology/Histopathology 1. Lymphedema may be due to an injury or insufficiency of lymph vessels, an increased venous pressure (e.g., deep vein thrombosis, right heart failure) or a decreased intravascular oncotic pressure (e.g., albumin deficiency). 2. Infiltration and/or destruction of lymph nodes by malignant solid tumor cells.

Clinical Presentation 1. Lymphedema initially presents with swelling of the legs which is often painful and extends above the legs in severe cases. 2. Lymph node metastasis may often show painless, hardened, and immobile lymph node enlargement.

Imaging The procedure of lymphoscintigraphy using indirect interstitial injections offers advantages compared with the radiological procedures of lymphatic angiography and lymphadenography with contrast medium for the visualization of the lymphatic drainage system as well as the

Using lymphoscintigraphy, the first priority is to visualize the lymphatic drainage system in the surroundings of malignant tumors. In this way the possible metastasizing routes of the particular tumor can be determined (1, 2). As a rule, colloidal particles are used in lymphoscintigraphy, for example, denaturated albumin. Owing to their size, from 20 to 30 nm, these particles are removed via the lymphatic drainage system and retained and phagocyted in the reticuloendothelial system (RES) of the lymph nodes. The colloidal particles are labeled with 20–40 MBq 99m Tc and injected intra- or subcutaneously, mostly in several smaller portions, into the surrounding tissue of the tumor. Direct injection into the tumor should be avoided so as not to risk a transfer of tumor cells into healthy tissues. Acquisitions are performed immediately after injection in order to visualize the primary lymphatic routes. Preoperatively, it may be necessary to draw these routes on the skin of the patient. Static late acquisitions, mostly after several hours, serve to register the lymph node stations, whose knowledge is necessary for the surgical planning. The primary lymph node or the primary lymph node stations, which are a first filter station in the lymphatic drainage system of the tumor, are called the “sentinel lymph nodes” (SLNs). In patients with malignant melanoma, breast cancer, and tumors in the small pelvis—like the cervix, vagina, prostate, bladder, and rectum—for the preoperative visualization of the draining lymph vessels and the regional lymph nodes, the SLN should be visualized scintigraphically, because the surgical procedure of lymphadenectomy is particularly dependent on the histology of this lymph node (Fig. 1). The reliability of the detection of the SLN is further increased through the intraoperative application of a gamma detector. Therefore, unnecessary radical lymphadenectomies can be prevented. The targeted removal of the primary lymph node means that the therapeutic

L

1082

Lymphopoietic System, Diseases of the

by mobilization of the extremities. During the first 2 h after the injection, planar scintigrams of the extremities are acquired and further acquisitions are made after an endurance test (e.g., 20 min climbing stairs).

Diagnosis

Lymphopoietic system, diseases of the. Figure 1 Malignant melanoma. After intradermal injection of 99mTc-nanocolloid in the area of a median thoracic malignant melanoma, a sentinel lymph node (SLN) is detected in the right axilla. For improved localization, the body contours are delineated using a cobalt marker.

perspectives remain preserved for many patients in an early tumor stage, despite the fact that lymphogenic metastatic spread has already occurred.

1. Lymphedema can be diagnosed by inspection and palpation. In order to determine the cause of the lymphedema, venous insufficiency, right heart failure, and albumin deficiency have to be excluded. An injury or insufficiency of lymph vessels can be determined by lymphoscintigraphy. 2. Lymph node metastasis is often diagnosed by inspection, palpation, and primarily morphologic imaging procedures. The exact anatomical position of the sentinal lymph node (SLN) can be determined using lymphoscintigraphy.

Bibliography 1.

Lymphedema Lymphoscintigraphy is also applied for the assessment of a primary or secondary lymphedema (3–5). In the case of an injury to the lymph vessels or a valve insufficiency, a diffuse outflow of lymphatic fluid occurs in the surrounding tissue (“dermal backflow”). As an indication of a proximal lymphatic vessel obstruction, collaterals between the distal lymphatic vessels can be detected. A subcutaneous injection of 20 to 40 MBq of a 99mTc-labeled microcolloid is made in the first or second interdigital space of the hands or feet. The removal of the tracer can be increased

2.

3.

4.

5.

Belhocine TZ, Scott AM, Even-Sapir E et al (2006) Role of nuclear medicine in the management of cutaneous malignant melanoma. J Nucl Med 47:957–967 Thompson JF, Uren RF (2005) Lymphatic mapping in management of patients with primary cutaneous melanoma. Lancet Oncol 6:877–885 Moshiri M, Katz DS, Boris M et al (2002) Using lymphoscintigraphy to evaluate suspected lymphedema of the extremities. Am J Roentgenol 187:405–412 Szuba A, Shin WS, Strauss HW et al (2003) The third circulation: radionuclide lymphoscinitgraphy in the evaluation of lymphedema. J Nucl Med 44:43–57 Weissleder H, Weissleder R (1988) Lymphedema: evaluation of qualitative and quantitative lymphoscintigraphy in 238 patients. Radiology 167:729–735

M

M-mode Echocardiography Motion (M) mode display displays the grey-scale level along the length of the ultrasound beam during time. ▶Ischemic Heart Disease, Ultrasound

Macklin Effect The pathophysiology of pneumomediastinum in severe blunt trauma is a three-step process: blunt traumatic alveolar ruptures, air dissection along bronchovascular sheaths, and spread of this blunt pulmonary interstitial emphysema into the mediastinum. A pneumomediastinum is therefore not necessarily related to direct rupture of the tracheobronchial tree. ▶Chest Trauma

occasional invasion of the skull base. The patient presents either with endocrine disease or with neurological/ ophthalmologic signs. On magnetic resonance imaging these tumors can be solid or cystic, and they may contain hemorrhagic or necrotic areas. ▶Pituitary Gland

Magnetic Resonance Imaging Technique characterized by high contrast resolution using the signals generated in the tissues in response to a magnetic field, produced by the instrument and converted by a computer into images of body structures: the acquisition is possible on different planes. ▶Carcinoma, Hypopharynx ▶Neoplasms, Benign and Malignant, Larynx ▶Osteonecrosis, Adults ▶Lymphadenopathy

MacMahon–Tannhauser Syndrome Pericholangiolitic biliary cirrhosis, xanthomatous biliary cirrhosis. Congenital hypoplasia of intralobular bile duct resulting in intrahepatic cholestasis and primary biliary cirrhosis associated with peculiar features such as hepatic ostheoartropathy. ▶Congenital Malformations, Liver and Biliary Tract

Macroadenoma Pituitary gland tumor of at least 1 cm located in the sella turcica with a variable suprasellar extension and

Magnetic Resonance Cholangiopancreatography A noninvasive technique for the evaluation of the biliary tract. It is performed with the use of heavily T2-weighted images obtained with different pulse sequences without the injection of contrast medium that demonstrates stationary fluids, including bile and pancreatic secretions, as having high signal intensity whereas solid organs show a low signal intensity. This technique also allows the visualization of the pancreatic ducts generally after the administration of a secretive agent (secretin). ▶Biliary Anatomy ▶Congenital Malformations, Bile Ducts

1084

Magnetic Resonance Imaging, Activatable Imaging Agents

Magnetic Resonance Imaging, Activatable Imaging Agents R ALPH W EISSLEDER , A LEXANDER R. G UIMARAES Center for Molecular Imaging Research, Massachusetts General Hospital, Charlestown, USA [email protected]

Magnetic resonance imaging (MRI) continues to be one of the diagnostic modalities of choice of producing noninvasive, high spatial resolution, multiplanar imaging of various disease states. It has recently become possible to expand MRI to obtain molecular information using genetic reporter systems (1) (see Chapter on genetic reporters). An alternative strategy has been to use targeted (see Chapter on targeted imaging agents) and activatable imaging agents. The different forms of activation strategies include: (i) imaging agents that change magnetic properties after specific enzymatic reactions (i.e., through cleavage of sugars or other molecules affecting water diffusion to paramagnetic metals [e.g., LacZ (2)]; (ii) imaging agents that amplify their signals due to oligomerization and protein binding [e.g., MPO (3)]; and (iii) nanoparticlebased magnetic relaxation switches (MRSW) capable of sensing DNA–DNA, protein–protein, and small molecule interactions (4).

Agents Based on Cleavage Louie and Meade et al have recently demonstrated an example of a novel “smart” contrast probe that remains silent until activated by a specific transgene product (2). By using bacterial b-galactosidase (LacZ), because it is easily assayed, and not expressed in most mammals, they developed a contrast agent that was associated with a substrate for LacZ, galactopyranose. They covalently linked galactopyranose to a chelated paramagnetic Gd3+, in a way that precluded access of water protons to the Gd3+ atom. This interaction with the LacZ substrate cleaved the galactopyranose, and thus allowed access of water protons to Gd3+ which shortened T1 and thus produced contrast. With intravenous mRNA to allow bgalactosidase activity within organisms, injection of this novel smart probe, produced visualization of those cells involved with LacZ activity. Using this technique, the authors measured an approximate 50% drop in T1, 100% increase in relaxivity with cleavage of the galactopyranose (2). Similar strategies have been employed using calcium activation, pH sensitive strategies, and through changes in viscosity through protein binding.

Agents Based on Oligomerization Chen et al have recently synthesized an activatable paramagnetic MR imaging agent that targets the enzyme myeloperoxidase (MPO), which is an enzyme that is a marker of inflammation (3). This class of contrast imaging agents is highly sensitive to MPO activity, and have been shown to image in vivo sites of MPO activity that reflect inflammation. The imaging agent based on a covalent conjugate of Gd-DOTA and serotonin (3-(2aminoethyl)-5-hydroxyindole). When the imaging agent was exposed to MPO, there was a resultant increase in molecular mass, and consequently increase in MR signal, with concomitant delayed clearance from tissue. Furthermore, MPO-mediated oxidation products bind to and cross-link with plasma proteins causing more local accumulation. The association with proteins also further contributes to the increase in MR signal due to R1 increase. The summation of these factors causes an approximate fivefold increase in R1, and thus increased contrast-to-noise (3). MPO is implicated in multiple disease states including Alzheimer’s disease, multiple sclerosis, lung cancer, and leukemia, and this class of agent, because of its capability of detecting sites of active inflammation in infection or injury, may lend further insight into these disease processes. One exciting possibility is in vulnerable plaque imaging, given the role that inflammation plays in atherosclerosis and plaque vulnerability. Furthermore, there is now an accumulating body of evidence that MPO plays a central role in this process, because MPO is found in high abundance in vulnerable plaques and in culprit lesions. Recent clinical trials have found that serum MPO levels predict cardiovascular risk in patients with acute coronary syndrome and chest pain. Furthermore, MPOdeficient humans appear to have decreased risk for cardiovascular diseases. Additional agents using this principle include dopa-DOTA and DPTA-bis-amide based agents.

Superparamagnetic-Based Activatable Reporter Systems Magnetic nanoparticles (MNP) offer another means of achieving signal amplification through exploitation of their spatial arrangement. Through specific molecular interaction (e.g., DNA–DNA or small molecule–protein interactions) they can be induced to assemble into larger nanoassemblies and the superparamagnetic iron-oxide cores of the nanoparticles become more efficient at dephasing the spins of water protons, that is, at enhancing the spin–spin relaxation rate (1/T2). Both dispersed (40 nm) and clustered or nanoassembled state (200 nm) remain in

Magnetic Resonance Imaging, Genetic Reporter Systems

Magnetic Resonance Imaging, Activatable Imaging Agents. Table 1 Target DNA mRNA Protein

Summary of MRSW sensors developed to date

Target

Telomeres GFP GFP CA-125 Avidin Anti-avidin-Ab Enzyme activity BamH1 Dam-methylase MboI DpnI Caspase-3 Renin Trypsin MMP-2 AKT Telomerase Peroxidases Small molecule detection Enantiomeric Impurity Organism Phage Herpes simplex virus (HSV) Adenovirus-5 Bacterial mRNA

1085

MRSW sensor (CCCTAA)3-CLIO CLIO-ATTTGCCGGTGT; TCAAGTCGCACA-CLIO Anti-GFP-av-CLIO Anti-CA125-av-CLIO Biotin-CLIO Avidin-CLIO CLIO-TTA-CGC-CTAGG-ATC-CTC; AAT-GCG-GGATCC-TAC-GAG-CLIO Methylated BamH1 MRSW sensor Methylated BamH1 MRSW sensor Methylated BamH1 MRSW sensor Av-CLIO; Biotin-GDEVDG-CLIO RK(Btn)IHPFHLVIHTK(Btn)R; av-CLIO Btn-(G)4RRRR(G)3K(Btn) or Btn-GPARLAIK(Btn); av-CLIO Btn-GGPLGVRGK(Btn); av-CLIO CLIO-CGGKGSGSGRPRTSSFAEG; CLIO-antiphosphopeptide CLIO-AATCCCAATCCC; AATCCCAATCCC-CLIO CLIO-phenol; CLIO-tyrosines CLIO-D-Phenylalanine Anti-fd bacteriophage-av-CLIO Anti-glycoproteinD(HSV-1)-av-CLIO; anti-HSV1-av-CLIO Anti-adenovirus-5-av-CLIO CLIO-GTCGTCAACTAC; CACTGAACAACA-CLIO

solution and the conversion between the two states can be induced repeatedly (4). It has also been shown that this is a reversible process, and thus these nanoassemblies can be disassembled and returned to their original dispersed state by methods including heat, enzyme cleavage, pH alteration, and disulfide bond reduction (4). This reversibility provides a powerful medium of distinguishing subtle changes in the environment for which they are applied, and have thus been termed MRSW (4). This technology has been used to measure multiple different molecular reversible molecular interactions at extremely low contrast agent concentration (fM). Concentrations as low as 0.5 fM have been observed for DNA and proteins, whereas targets like viruses, with large, high multivalence, allow for detection of as few as five viral particles per 10 mL of herpes simplex virus-1 and adenovirus-5 (4). We have exploited this observation and demonstrated MRSW assays are homogenous assays that can sense a wide variety of molecular interactions with high sensitivity and specificity with no sample preparation (Table 1). At the concentrations used (10 h in mice, and >24 h in humans) as well as lymphotrophic components (6). Other derivatives of iron-oxide MNP involve ironoxide crystals (3–10 nm in diameter) coated with lipid and polyethylene glycol (PEG), which can reduce binding of plasma proteins and phagocytosis and clearance by macrophage. The most ubiquitous, and widely clinically used MNP offer a superparamagnetic iron-oxide core with a dextran or dextran derivative coating (Feredex®, Resovist®, Combidex®, and ferumoxytol) (5). Depending on the size and surface characteristics of these surface coats, different clearance mechanisms exist (5, 7).

Clinical Applications of Molecular MR Imaging Using these techniques, advances have been made in contrast agent development in clinical and preclinical

M

1090

Magnetic Resonance Imaging, Molecular MR Imaging

Magnetic Resonance Imaging, Molecular MR Imaging. Table 1 and their specific clinical or preclinical application

List of targeted and nonspecific MRI based contrast agents

Disease

Imaging agent

Application

Cancer

Iron oxide (ferumoxide) Iron oxide (ferumoxtran, ferumoxytol) Integrin avB3 Iron oxide Iron oxide Beta-amyloid targeted iron oxide Fibrin (Gd-labeled) Iron oxide (ferumoxtran) Integrin avB3/VCAM Gadofluorine

Liver cancer Nodal staging/angiogenesis Angiogenesis Inflammation Inflammation Alzheimer’s disease Acute/subacute thrombi Macrophages/inflammation Angiogenesis Lipid-rich plaques

Infection Inflammatory diseases (e.g., arthritis) Neurological Thrombosis Atherosclerosis

models including earlier detection of disease, drug discovery and development, and biomedical research, including the elucidation of enzymatic and cellular events in inflammation, angiogenesis, apoptosis, stem cell trafficking, and atherosclerotic disease. Table 1 illustrates various targeted, and nonspecific MRI based contrast agents and their specific clinical or preclinical application (8). Although a detailed listing of all applications of MR molecular imaging is beyond the scope of this chapter, we highlight the application of MRI within cancer and atherosclerotic disease.

Molecular Imaging of Cancer Cancer Staging MR molecular imaging has made marked strides in cancer staging, in specific, lymph node metastases (6). As with cancer detection, extant imaging approaches rely on computed tomography (CT), conventional radiography, and MRI for cancer staging, and these imaging approaches, in their conventional form, lack the sensitivity and specificity to distinguish potential metastatic involvement of lymph nodes that are less than 1 cm. in short axis. In addition, small lymph nodes (90%, with specificities approaching 100% (6). In later more recent work, Weissleder and Harisinghani have expanded upon this work to semiautomate data analysis routines in lymphotropic magnetic resonance imaging (LMRI). With this semiautomated approach, the combination of two variables was determined to increase the sensitivity and specificity of LMRI with MNP to 94.3% and 93.5%, respectively, with positive and negative predictive values 93.5% and 98%. Although, gadolinium-based approaches have been applied to lymph node imaging, these have not reached clinical utility.

Angiogenesis Imaging The imaging of neovascular content is currently one of the most intensely studied tumor pathways because of the rapid development of novel strategies to inhibit neovascular growth in cancer. Angiogenesis is a prominent feature of invasive cancers and is associated with adverse prognosis (9). Toward that end, recent increased interest has focused upon the rapid development and implementation of drugs to inhibit neovascular growth (angiogenesis inhibitors). Multiple variants of these inhibitors continue to be developed and are in various phases of ongoing clinical trials. Documentation of the degree of angiogenesis in primary and metastatic tumors is also important for treatment planning and assessing the efficacy of angiogenesis inhibitors, including low-dose cytostatic therapies.

Magnetic Resonance Imaging, Molecular MR Imaging

Current biomarkers of angiogenesis include serum markers (e.g., VEGF, FGF), microvessel density (MVD) quantification, tissue markers (e.g., histologic staining for CD31, FVIII, Ulex lectin) and imaging (e.g., nuclear imaging, CT imaging with dynamic contrast or MR imaging). MVD is a correlative, surrogate marker of angiogenesis, and has been shown to be of prognostic significance in various malignant therapies. Concurrent with recent advances in the development of angiogenic inhibitors, has been technological improvements in imaging angiogenesis, so as to provide a noninvasive, surrogate marker of anti-angiogenic efficacy. Imaging angiogenesis with MRI has been focused into three different arenas: (i) functional imaging using dynamic tracking of contrast administration; (ii) steady state blood volume determinations of neovascular density; and (iii) specific molecular markers of angiogenesis. Functional dynamic contrast enhanced MRI (DCEMRI) approaches rely on the “leaky” nature of angiogenic blood vessels associated with malignancy. By using a small contrast agent, dynamic imaging is used and with the aid of compartmental modeling, various parameters are derived including permeability surface area product (a.k.a Ktrans), and fractional plasma volume. DCE-MRI has been applied clinically to many cancer systems including breast, brain, prostate, and renal cell, with interesting results, but confounded by high variance. By imaging primary neoplasms, before and following the administration of MNP, a parametric map of tumor blood volume can be calculated by the change in T2* within the primary neoplasm. By correlating tumor blood volume to known blood volume of muscle, one can derive a vascular volume fraction (VVF). We refer you to the original description of this technique for further elaboration of the mathematical details of the technique (10). In animal models it has been shown to have high accuracy when compared to other surrogate markers of angiogenesis (e.g., microvessel density, radiolabeled blood volume measurements). In recent xenograft treatment trials, results suggested that VVF measurements are earlier and more reliable indicators of successful antiangiogenic treatment regimen than is tumor volume.

MR Molecular Imaging of Atherosclerosis The noninvasive assessment and identification of highrisk atherosclerotic lesions and thus, those patients deemed “high-risk” for a cardiovascular, or cerebrovascular events, remains an important focus of clinical research. One goal of molecular imaging has been to provide functional information about those biologic processes specific to atherosclerosis (8), and the pathophysiology of vulnerable plaque including inflammation, apoptosis and angiogenesis.

1091

Activated macrophages have been found to play a key role in the pathophysiology of atherosclerosis and their presence is specific for high-risk lesions. Imaging of macrophages may offer a window into separating and identifying those patients and plaques at high risk for disruption. Iron-oxide MNP accumulate within macrophages, have been found to accumulate within human atherosclerotic macrophages and are therefore preferentially found in macrophage-rich carotid plaques. After intravenous injection of magnetic nanoparticles, there was uptake and enhancement of carotid plaques on T2* sensitive gradient echo techniques as compared to baseline MRI. Histopathologic analyses of carotid endarterectomy specimens confirmed iron localized within these regions. Specific targets have been exploited as well. An example is VCAM-1, which is a critical component of the leukocyte-endothelial adhesion cascade. By using this protein and binding to a CLIO—Cy5.5 backbone (a crosslinked iron-oxide nanoparticle that provides a platform for novel magneto-fluoro MR contrast agent development in animals), MRI and fluorescent imaging demonstrated high affinity for endothelial cells expressing VCAM-1 but low affinity for macrophages in vitro and in vivo in both neoplastic and atherosclerotic models. In addition to VCAM-1, the CLIO platform has been used to image selectively E-selectin in mouse xenograft models of Lewis lung carcinoma by attaching E-selectin-binding peptide to CLIO (Cy5.5) nanoparticle, as well as apoptosis through attachment of CLIO (Cy5.5) to Annexin V.

Bibliography 1. Gillies RJ, Morse DL (2005) In vivo magnetic resonance spectroscopy in cancer. Annu Rev Biomed Eng 7:287–326 2. Ross BD et al (2003) Evaluation of cancer therapy using diffusion magnetic resonance imaging. Mol Cancer Ther 2(6):581–587 3. Aime S et al (2002) Insights into the use of paramagnetic Gd(III) complexes in MR-molecular imaging investigations. J Magn Reson Imaging 16(4):394–406 4. Artemov D (2003) Molecular magnetic resonance imaging with targeted contrast agents. J Cell Biochem 90(3):518–524 5. Weissleder R et al (1990) Ultrasmall superparamagnetic iron oxide: characterization of a new class of contrast agents for MR imaging. Radiology 175(2):489–493 6. Harisinghani M et al (2003) Noninvasive detection of clinically occult lymph-node metastases in prostate cancer. N Engl J Med 348 (25):2491–2499 7. Libby P (2002) Inflammation in atherosclerosis. Nature 420 (6917):868–874 8. Jaffer FA, Weissleder R (2005) Molecular imaging in the clinical arena. JAMA 293(7):855–862 9. Folkman J, Cole P, Zimmerman S (1966) Tumors behavior in isolated perfused organs: in vitro growth and metastasis of biopsy material in rabbit thyroid and canine intestinal segment. Ann Surg 164:491–502 10. Bremer C et al (2003) Steady-state blood volume measurements in experimental tumors with different angiogenic burdens a study in mice. Radiology 226(1):214–220

M

1092

Magnetic Resonance Imaging, MR-Targeted Imaging Agents

Magnetic Resonance Imaging, MR-Targeted Imaging Agents

(T1-based)] and superparamagnetic [e.g., iron oxide magnetic nanoparticles (MNP)]. For nontargeted MR contrast agents see Chapter (MRI—molecular imaging). Table 1 is a list of many targets for which imaging agents have been developed.

R ALPH W EISSLEDER , A LEXANDER R. G UIMARAES Center for Molecular Imaging Research, Massachusetts General Hospital, Charlestown, USA [email protected]

Exploiting T1 and T2 contrast mechanisms have dominated the design of novel magnetic resonance imaging (MRI) agents targeted for various clinical and preclinical molecular imaging applications. Multiple such targets have been exploited using paramagnetic [e.g., Gd

Paramagnetic-Targeted Imaging Agents Targeted paramagnetic imaging agent strategies can be categorized based on the size of the molecule being developed. Examples of small molecule ( others LTB (femur, tibia), ribs, skull > others

Typical locations within bone

LTB: Meta(dia)physis LTB: Metaphysis, diaphysis Tibia, fibula (occasionally) Diaphysis LTB (humerus, femur), calcaneus, LTB: Metaphysis, pelvis > others diaphysis LTB, spine > others LTB: Metaphysis spine: posterior elements Skull, femur, spine, pelvis, ribs > others LTB: Diaphysis, metaphysis

Neoplasm-Like Lesions, Bone

1195

Neoplasm-Like Lesions, Bone. Figure 1 Tumorlike lesions of bone: radiographic findings. (a) Nonossifying fibroma of the femur: anteroposterior radiograph shows a lobulated, cortically based lesion with sclerotic margins, trabeculations, and a shell-like periosteal reaction. (b) Monostotic fibrous dysplasia of the femur: anteroposterior radiograph shows a large, well-defined lesion of ground glass density bordered by a thick sclerotic rim and extending from the femoral neck into the proximal diaphysis. (c) Osteofibrous dysplasia of the tibia: lateral radiograph shows a well-defined lucency with sclerotic borders and trabeculations at the typical site of involvement within the anterior cortex of the midshaft of the tibia. Note the slightly increased tibial bowing. (d) Simple bone cyst of the humerus: anteroposterior radiograph demonstrates a centrally located, well-defined lytic lesion with a delicate sclerotic margin and slight expansion of the lateral cortex. (e) Aneurysmal bone cyst of the tibia: anteroposterior radiograph shows an eccentrically located lytic lesion exhibiting well-defined borders, trabeculations, and marked expansion with formation of a thin neocortex. (f) Eosinophilic granuloma (Langerhans cell histiocytosis) of the femur: anteroposterior radiograph reveals a medullary osteolytic lesion within the diaphysis that causes endosteal erosion of the medial cortex and a lamellated periosteal reaction.

the lesion and therefore can help to establish the diagnosis in anatomically complex areas such as the pelvis. Aneurysmal bone cyst (ABC) on radiographs usually appears as an expansile osteolytic lesion regardless of its location within the skeleton. In long tubular bones, the most common features are well-defined margins with or without a sclerotic rim, eccentric location, marked expansion with formation of a thin neocortex, and trabeculations: the so-called “soap bubble appearance” (Fig. 1). However, the lesion might also present with more aggressive findings and therefore can not always confidently be diagnosed with radiography alone. Because MR imaging has the potential to reflect the pathologic anatomy of ABC independent from its location, it represents the most informative additional imaging modality with a view to differential diagnosis (Fig. 3). Primary ABCs are exclusively composed of multiple cystic spaces which contain noncoagulated blood and blood breakdown products which due to sedimentation effects often cause fluid levels particularly on T2-weighted images. Most lesions are surrounded by a rim of low signal intensity and typically demonstrate marked enhancement of cyst walls and internal septations with intravenous contrast administration. The presence of solid components does

Neoplasm-Like Lesions, Bone. Figure 2 Fibrous dysplasia of the mandible: CT. Sagittal CT reformation image shows a well-demarcated, slightly expansile lesion with ground glass density and some irregular ossifications. Note the distortion of adjacent teeth.

not generally exclude the diagnosis of primary ABC, but should alert to the most important differential diagnoses, teleangiectatic osteosarcoma, and ABC secondary to

N

1196

Neoplasm-Like Lesions, Bone

Neoplasm-Like Lesions, Bone. Figure 3 Aneurysmal bone cyst of the tibia: MR imaging. (a) Sagittal T1-weighted SE, (b) fat-suppressed T2-weighted TSE, and (c) contrast-enhanced fat-suppressed T1-weighted SE images show and internally septated lesion with multiple cystic cavities bordered by a rim of low signal intensity. Note the hyperintense cystic compartments at the bottom of the lesion on T1-weighted image, fluid levels on T2-weighted image, and enhancement of cyst walls and septations with contrast administration. Edema of adjacent bone marrow and soft tissue, as seen in this case, is not an unusual finding in aneurysmal bone cysts.

another bone lesion, such as giant cell tumor or osteoblastoma. Eosinophilic granuloma as a solitary lesion represents the most common manifestation of Langerhans cell histiocytosis (60–80% of cases). Further osseous lesions are detected in only 10–20% of patients who initially present with a solitary granuloma. The radiographic features of eosinophilic granuloma vary with its skeletal location. Lesions of the skull and those located in flat bones often appear purely osteolytic and sharply delineated. The presence of a radiodense focus within a lytic lesion of the cranium has been described as a “button sequestrum.” In long tubular bones, eosinophilic granuloma in most cases arises as a medullary radiolucency with central location and various degrees of endosteal erosion of the cortex (Fig. 1). Active lesions are often accompanied by a simple lamellar or an “onion skin”-like periosteal reaction. Spinal lesions typically cause destruction of vertebral bodies with consecutive collapse and flattening (“vertebra plana”), most often observed in the thoracic and lumbar spine. MR imaging usually shows an intramedullary lesion of low T1- and high T2-weighted signal intensity which is associated with more or less extensive edema of bone marrow, periosteum, and adjacent soft tissues. A rim of low signal intensity at the periphery of the main lesion seen on STIR images has been reported to represent a possible indicator of initial healing. In general, the imaging appearance of eosinophilic granuloma is rather unspecific with subacute

osteomyelitis, osteoblastoma, and Ewing’s sarcoma representing the main differential diagnoses. Whole body MR imaging is more sensitive than skeletal scintigraphy and radiography in detection of osseous manifestations of Langerhans cell histiocytosis and therefore, should be used as the screening method of choice for documentation of the presence and extent of skeletal involvement.

Nuclear Medicine Skeletal scintigraphy is not very helpful in the specific diagnosis of solitary tumorlike lesions of bone, but can add information in cases of polyostotic involvement (e.g., in fibrous dysplasia).

Diagnosis Nonossifying fibroma, fibrous dysplasia and simple bone cyst in most cases can sufficiently be diagnosed by means of radiography and do not require biopsy. The diagnosis of osteofibrous dysplasia should be verified by histology in order to exclude adamantinoma. With a view to their possible differential diagnoses histologic verification is also necessary in ABC and eosinophilic granuloma. The final diagnosis should be established by a combination of radiographic and histologic findings.

Neoplasms, Benign and Malignant, Larynx

Bibliography 1.

2.

3.

4. 5.

Fletcher CDM, Unni KK, Mertens F (eds) (2002) Pathology and genetics of tumours of soft tissue and bone. In: World Health Organization Classification of Tumours. IARC Press, Lyon Forest M, Amouroux J (1997) Pathology of pseudotumoral lesions. In: Forest M, Tomeno B, Vanel D (eds) Orthopedic Surgical Pathology: Diagnosis of Tumors and Pseudotumoral Lesions of Bone and Joints. Churchill Livingstone, Edinburgh, pp 519–670 Resnick D (1995) Tumors and tumor-like diseases: imaging and pathology of specific lesions. In: Resnick D (ed) Diagnosis of Bone and Joint Disorders. 3rd edn. WB Saunders, Philadelphia, pp 3628– 3938 Conway WF, Hayes CW (1993) Miscellaneous lesions of bone. Radiol Clin North Am 31:339–358 Woertler K (2003) Benign bone tumors and tumor-like lesions: value of cross-sectional imaging. Eur Radiol 13:1820–1835

1197

The larynx lies in the visceral space of the neck, in front of the esophagus, is lined by prismatic epithelium, and is suspended by muscular and cartilaginous structures (thyroid cartilage, cricoid cartilage, and arytenoids). The organ is divided into three anatomical regions: supraglottis, glottis, and subglottis. An important anatomical structure is the prelaryngeal space, the deepest space of the larynx, which is mainly occupied by fat tissue, and extends vertically between the thyroid cartilage and the laryngeal lumen; at the level of the vocal folds this space is named paraglottic. The supraglottic portion of the prelaryngeal space is in communication with the pre-epiglottic space. Neoplasms can be benign, such as polyps, papillomas, hemangioma, paragangliomas, and chondromas, or malignant, including squamocellular or nonsquamocellular carcinoma.

Neoplasms of the Nasopharynx Pathology/Histopathology ▶Neoplasms, Nasopharynx

Neoplasms of the Rhinopharynx ▶Neoplasms, Nasopharynx

Neoplasms, Benign and Malignant, Larynx S TEFANO C IRILLO Institute for Cancer Research and Treatment, Turin, Italy [email protected]

Definitions Laryngeal neoplasms, either benign or malignant, originate from the mucosa and surrounding cartilaginous or connective structures of any part of the ▶larynx. The larynx is shaped like a pyramidal frustum joining the upper aerodigestive tract with the trachea. Its upper base faces the tongue and its lower base continues with the trachea. It is primarily an organ of phonation, but also an important regulator of respiration, and it prevents aspiration during swallowing.

The following rare varieties of benign tumors can arise in the larynx. Chondroma: Usually 2 cm or less, it arises posteriorly from the cricoid cartilage and projects anteriorly, often causing partial airway obstruction. This benign tumor may evolve into a chondrosarcoma. Hemangioma: In infants it is a sessile, poorly circumscribed subglottic mass, often associated with obstructive symptoms. In adults, it is uncommon and usually supraglottic. Paraganglioma: It commonly involves the aryepiglottic fold. It is more frequent (70%) in males and has a malignant behavior in 3–25% of cases. Papilloma: It is a warty outgrowth of the laryngeal surface epithelium; in children it is usually multiple, most commonly located on true vocal cords, false cords, epiglottis, subglottic area, and rarely within the tracheobronchial tree. Recurrence is common, at times years after excision or destruction, and may be massive and rapidly growing, leading to airway compromise. In adults, mostly male, it is often solitary and recurrences frequently exhibit dysplasia. Vocal cord polyp: It is more common in heavy smokers or singers, due to inflammation, allergy, or immunologic causes. It almost never evolves into malignancy. It appears as a smooth, round, 1–3-mm growth on the true vocal cords, often on their anterior third. Nonsquamous cell carcinomas: They are malignant carcinomas including adenoid cystic carcinoma, angiosarcoma, chondrosarcoma, liposarcoma, lymphoma, melanoma, and rhabdomyosarcoma. These nonsquamous tumors typically grow beneath the laryngeal mucosa and are therefore more difficult to diagnose clinically than squa-

N

1198


mous cell carcinomas. Cross-sectional imaging findings are important in defining the disease extent and in directing the clinician to the optimal transmucosal biopsy site. Squamous cell carcinoma: This malignant carcinoma is the most common laryngeal neoplasm accounting for more than 90% of all malignant lesions; the most frequent histopathological type is the nondifferentiated type, while other varieties such as basaloid, papillary, spindle cell, and verrucous types are very rare. Squamous cell carcinomas originate in 30% of cases from the supraglottic larynx. Most frequently, in 30–40% of the cases, they arise from the junction between the false cord and the epiglottis. Epilaryngeal cancers spread to the pharynx (base of the tongue, pyriform sinus, hypopharynx wall); vestibular cancers spread both along the surface and deeply in the pre-epiglottic and paraglottic spaces (Table 1). The true vocal cords are the most common site of origin of laryngeal carcinomas. The anterior portion of the true vocal cord is the most common location of squamous cell carcinoma. Anteriorly, the tumor may extend to the anterior commissure, where it may involve the contralateral true vocal cord. Advanced lesions arising from the posterior third of the cord may extend posteriorly to involve the cricoarytenoid joint and the interarytenoid region. Tumors may extend inferiorly to involve the subglottic region (Table 2). Neoplasms, Benign and Malignant, Larynx. Table 1 T staging for supraglottic carcinoma T1

Tumor in one supraglottic subsite with normal cord mobility T2 Tumor invading the mucosa in more than one supraglottic subsite without laryngeal fixation T3 Endolaryngeal tumor with fixed vocal cord ± invasion of the postcricoid area or pre-epiglottic tissues T4a Tumor invading through to the thyroid cartilage ± other extra laryngeal tissues (resectable), e.g., trachea, cervical soft tissues, strap muscles, thyroid, esophagus T4b Tumor invading the prevertebral space, encasing the carotid artery, or invading the mediastinal structures (unresectable)

Neoplasms, Benign and Malignant, Larynx. Table 2 T staging for carcinoma of true vocal cords T1 Limited to cords with normal mobility T2 Spreading (supra ± subglottis) with impaired vocal cord mobility T3 Endolaryngeal tumor with fixed vocal cord ± invasion of the postcricoid area or pre-epiglottic tissues T4 Limited to larynx with fixed vocal cord

Subglottic carcinomas are rare and account for only 5% of all laryngeal carcinomas. The subglottic region is more commonly involved by the direct extension of a glottic or supraglottic carcinoma than by tumors elsewhere. When present, these lesions are characteristically circumferential and often extend to involve the undersurface of the true vocal cords. They have a tendency to early invasion of the cricoid cartilage and to extension through the cricothyroid membrane (Table 3). The most important prognostic factors for laryngeal cancers include tumor volume and T stage and N stage. The presence of adenopathy at the time of diagnosis is the most important factor for predicting lymph nodal recurrence (3). The survival rate for ▶head and neck squamocellular carcinomas decreases to 50% in patients with homolateral adenopathy, 75% in those with contralateral adenopathy, and even further when capsular rupture of the lymph nodes is present. The first station of lymph drainage from the supraglottic larynx is at the level of the subdigastric and the middle anterior cervical nodes; the second station is at the level of the lower anterior cervical nodes. The glottic larynx contains only a few nodes, and nodal spread occurs only when the primary tumor extends to the supraglottis or subglottis. The first station of lymph drainage from the subglottic larynx is at the level of the lower anterior cervical nodes, the paratracheal nodes, and the supraclavicular nodes. Glottic and subglottic tumors metastasize to the ipsilateral lymph nodes, but supraglottic tumors often spread to nodes on both sides of the neck.

Clinical Presentation Laryngeal cancer is the most common cancer of the upper aerodigestive tract (25% of the cases) and represents 4–5% of all neoplasms. The incidence of laryngeal tumors is correlated with smoking, alcohol, and poor oral hygiene. Laryngeal cancers account for approximately 1.2% of all newly diagnosed cancers; their incidence ranges from Neoplasms, Benign and Malignant, Larynx. Table 3 T staging for subglottic carcinomas T1 T2

Tumor limited to the subglottis Tumor extending to the vocal cords with normal or impaired mobility T3 Tumor limited to the larynx with fixed vocal cord T4a Tumor invading through to the thyroid cartilage ± other extra laryngeal tissues (resectable), e.g., trachea, cervical soft tissues, strap muscles, thyroid, esophagus T4b Tumor invading the prevertebral space, encasing the carotid artery, or invading the mediastinal structures (unresectable)


2.5 to 17.1 per 100,000 inhabitants in males, and from 0.1 to 1.3 per 100,000 in females in European countries; the peak incidence occurs at 50–60 years of age (1)(2). The symptoms of laryngeal tumors largely depend on their size and location. Common symptoms for supraglottic tumors include mild odynophagia, mild dysphagia, and mass sensation. Uncommon symptoms include severe dysphagia and ear pain. For glottic and subglottic tumors, the most common presenting symptom is hoarseness of the voice. Uncommon symptoms include odynophagia, ear pain, thyroid cartilage pain, and airways obstruction. All the above symptoms are, however, not specific for a malignant lesion.

Diagnosis The physical examination must include a systematic assessment of the patient’s general health condition, searching for signs of associated conditions and metastatic disease. The neck and supraclavicular fossa are palpated to search for cervical adenopathy, other masses, and laryngeal crepitus. The oral cavity and oropharynx are examined under direct endoscopic vision. The larynx should also be examined by indirect laryngoscopy after anesthetic spray administration. If there is a suspicion of cancer, biopsy is usually performed during direct laryngoscopy with the patient under general anesthesia; vital coloring is used to better identify the lesion. The examination should be extended to the whole upper aerodigestive tract and to the esophagus to exclude a second neoplastic lesion. In cases where cancer is confirmed histologically it is very important that extension be evaluated accurately, in order to plan the most appropriate treatment. Laryngeal tumors arise from the mucosa, thus they are usually visible on the surface, but their submucosal spread may not be accurately evaluated by endoscopic examination alone. Indeed, clinical and endoscopic examinations underestimate tumor extension in 45–50% of cases. Integration between clinical data, endoscopic data, and imaging [computed tomography (CT)/▶magnetic resonance imaging (MR)] is necessary for correct staging. Clinical staging is sufficient only for glottic cancer without involvement of the anterior commissure or with preserved vocal cordal motility, or exophytic cancer of the epiglottis and false vocal cord.

MRI can both assess the depth of invasion extension, which must be accurately evaluated in the following regions: a. For supraglottic cancers: . Infiltration of the base of the tongue . Diffusion to the pre-epiglottic space . Diffusion to the superior prelaryngeal space . Extension to the pyriform sinus . Extension to the glottis plane . Involvement of the cartilage b. For glottic cancers: . Diffusion to the paraglottic space . Diffusion to the cartilage . Involvement of the anterior commissure . Vertical diffusion (supraglottic or subglottic) The imaging criteria used for tumor involvement are abnormal contrast medium enhancement, soft tissue thickening, presence of a bulky mass, infiltration of fatty tissue even without distortion of the surrounding soft tissues, or a combination of these signs. Imaging can underestimate the mucosal extension and overestimate the deep extension in the presence of edema, which may not be differentiated from neoplastic tissue. ▶Multidetector CT (MDCT) is the first-choice examination for staging laryngeal neoplasms. The short scanning time of this technique limits the possibility of motion artifacts caused by swallowing and makes it possible to perform scans during functional and dynamic maneuvers, such as phonation or Valsalva, useful for identifying the tumor, and evaluating the degree of extension of the malignant tissue (Figs 1 and 2).

Imaging The possibility that patients may benefit from conservative surgery or radiotherapy depends on the accuracy of the pretreatment assessment of cancer extension. CT and

1199

Neoplasms, Benign and Malignant, Larynx. Figure 1 Epiglottic cancer, MDCT – nodular mass in paraglottic superior space.

N

1200


Neoplasms, Benign and Malignant, Larynx. Figure 2 Dynamic maneuver during laryngeal scans, MDCT: (a) abduction and (b) adduction of vocal folds.

MDCT allows one to obtain excellent multiplanar reconstructions, with optimal tumor visualization and accurate measurement of its volume and adenopathy diameter (Fig. 3). With MRI, a high contrast resolution can be obtained between the various anatomical structures, but images are often deteriorated by motion artifact caused by the long acquisition time, especially in patients in advanced stages who have dyspnea, cough, and large respiratory excursions. If conservative surgery is planned, MRI is a complementary test to MDCT when the latter cannot exclude with certainty infiltration of the base of the tongue and of the cartilage. False-positive results, however, are inevitable with both imaging modalities, due to inflammation.

Neoplasms, Benign and Malignant, Larynx. Figure 3 Anterior commissure cancer, MDCT – lesion on (a) axial plane and on (b) sagittal reconstruction.

MDCT and MRI have the same accuracy in the evaluation of diffusion to the epiglottic space (90–95%) and to the glottis plane. Sensitivity for assessing involvement of the paraglottic space is 93% for MRI and 97% for MDCT, while specificity is lower, being 50–76% for either test due to the difficulty in differentiating edema from neoplastic tissue. MDCT and MRI may not provide sufficient diagnostic accuracy in evaluating involvement of the pyriform sinus: this may help explain the high rate of hypopharyngeal recurrences. Involvement of the anterior commissure is considered if a soft tissue with a thickness of more than 1 mm is present (4,5).

Neoplasms, Benign, Large Bowel

Nuclear Medicine Tests Nuclear medicine tests are useful in the follow-up of patients who underwent surgery or radiotherapy, for whom there is a clinical suspicion of local recurrence. In this setting, FDG-positron emission tomography (PET) provides a high sensitivity (86–100%) and specificity (69–87%) (5). PET has some limitations due to its low spatial resolution and to the absence of anatomic reference points, which is of particular relevance considering the complexity of the head and neck region. Moreover, uptake is at times present in physiological conditions generating possible false-positive findings. Anatomic limitations can be partially overcome by the combination of CT-PET, which permits simultaneous acquisitions of morphological and metabolic data.

References 1. 2. 3.

4.

5.

Sturgis EM, Wei Q, Spitz MR (2004) Descriptive epidemiology and risk factors for head and neck cancer. Semin Oncol Dec; 31(6):726–733 American Cancer Society: Cancer Facts and Figures 2006. Atlanta, Ga: American Cancer Society, 2006. Thawley SE, Panje WR, Batsakis JG et al (1999) Comprehensive Management of Head and Neck Tumors. 2nd edn. Philadelphia, Pa: WB Saunders Becker M, Hasso AN (1996) Imaging of malignant neoplasms of the pharynx and larynx. In: Taveras JM, Ferruci JT (eds) Radiology: Diagnosis—Imaging—Intervention. Lippincott, Philadelphia, PA, pp 1–16 Zba¨ren P, Becker M, Laeng H (1997) Staging of laryngeal cancer: endoscopy, computed tomography and magnetic resonance imaging versus histopathology. Eur Arch Otolaryngol 254(1):117–122

Neoplasms, Benign, Large Bowel A NDREA L AGHI , F RANCO I AFRATE Unit of CT and MR University of Rome “La Sapienza”, Polo Pontino, I.C.O.T. Latina, Italy [email protected] [email protected]

Definitions A ▶polyp is any circumscribed protrusion of normal mucosa into the colonic lumen, whether originating from the mucosa or situated submucosally. According to Morson (1) there are two main groups of polyps projecting into colonic lumen: . Nonneoplastic polyps (including hyperplastic, inflammatory, juvenile, or hamartomatous lesions). . Neoplastic (or adenomatous) polyps.

1201

Since most of the colorectal tumors (90–95%) arise from sporadic adenomas (polyps) and most of the remaining are accounted for by several hereditary cancer syndrome, a quick and right diagnosis as well as an appropriate treatment is extremely important in the prevention of colonic cancer.

Pathology/Histopathology Nonneoplastic polyps include hyperplastic, inflammatory, juvenile, or hamartomatous lesions, which lack dysplastic features. Hamartomatous polyps are the most common form of polyps in children. They are usually seen in the 4th to 6th year of life in many syndromic forms. The incidence in adulthood is 0.5–3.3% of all endoscopically removed polyps. More than 80% are usually found in the rectum. These polyps can vary in size between 0.3 and 0.5 cm and they are usually pedunculated, even if occasionally can be sessile. Peutz–Jeghers Syndrome is an autosomally inherited syndrome due to mutations in LKB1 gene, identified in 50–60% of cases. This condition is characterized by hamartomatous polyps of smooth muscle through the GI tract and mucocutaneous pigmentation. Affected individuals have hamartomatous polyps in the small and large bowel and characteristic mucocutaneous brown to black melanin spots most commonly observed in the perioral area. The size of the polyp is 0.3 to 5 cm; they are commonly pedunculated, but occasionally they can be sessile. They cannot be distinguished from adenomas during endoscopy, and should be removed if possible. Risk of gastrointestinal and other malignancies is slightly increased. Juvenile Polyp Syndrome (JPS) is cystic dilationsof glandular structures in the lamina propria without malignant potential. This syndrome is an uncommon condition that tends to be sporadic although 20–50% has a family history showing autosomal dominant inheritance (2). JPS grows from the lining of the bowel and originates in the tissues supporting that lining. It does not arise from the colonocytes (colonic lining cells) but from the tissues underneath the lining cells. Juvenile polyps are relatively common in children even if by adulthood 50–200 polyps may be present, commonly in the rectosigmoid junction. Juvenile polyposis may or may not be familial. Familial juvenile polyposis has been associated with mutations in two genes: SMAD4 on chromosome 18 and PTEN on chromosome 10. There is a small but significant risk of cancer in patients with juvenile polyposis, mostly due to development of adenomatous tissue in juvenile polyps. Hyperplastic polyps are lesions characterized by an epithelial dismaturityand hyperplasia caused either by a disturbed maturation process of cell in crypts without a known cause or either by hypertrophy of the crypts due to excessive epithelial cells. These polyps can be seen very

N

1202


frequently in 75% of subjects aged over 40 years and are usually multiple. Most are often tiny lesions ranging in size between 2 and 5 mm. They are the most common colorectal neoplasms (10 times > adenomas). They have no malignant potential and they are often so small that may be missed even on endoscopic examination. ▶Colorectal adenomas are benign neoplasms, pedunculated or sessile arising from the epithelial cells of the colorectum, withvarying degrees of cellular atypia. Although benign, they are the direct precursors of adenocarcinomas and follow a predictable cancerous temporal course unless interrupted by treatment. There are three histological types:

inactivation of tumor suppressor genes. The K-ras oncogene is described in 9% of small adenomas, 58% of adenomas larger than 1 cm, and 46% of colorectal carcinomas. Inactivation of tumor suppressor genes on arms 5q, 18q, and 17p is thought to be essential in tumorigenesis. As mutated in 30–60% of persons with sporadic adenomas and adenocarcinomas, the APC gene, on 5q, has an important role in adenoma formation.

. tubular adenomas . villous adenomas . tubulo-villous adenomas

They may present with bleeding, prolapse, and rarely intussusception, but most commonly they are asymptomatic. Colonic polypectomy has simplified the management of this condition. They may, however, cause painless rectal bleeding or bleeding not apparent to the naked eye. Obstruction can occur if they are large. Usually the colorectal polyps are the ones causing symptoms, such as bleeding, diarrhea, abdominal cramps, and anemia. In such cases surgery is performed, choosing between the same operations offered to patients with familial adenomatous polyposis: colectomy and ileorectal anastomosis, proctocolectomy and pouch, or proctocolectomy and ileostomy.

According to world health organization (WHO) criteria, villous adenomas are composed of greater than 80% villous architecture. Tubular adenomas are encountered most frequently (80–86%). Tubulovillous adenomas are encountered less frequently (8–16%), and villous adenomas are encountered least frequently (5%). It has been shown that the removal of polyps by ▶colonoscopy reduces the risk of getting colon cancer significantly. Malignant potential is determined by the size with 1% risk of adenocarcinoma if lesser than 1 cm in diameter; 10–50% risk if greater than 2 cm as well as by histologic type—greatest malignant potential for villous adenoma; least for tubular. This distribution of adenomatous polyps is paralleled by the distribution of colonic cancer. They may look like a wart when small and when they grow they may appear like a cherry on a stem or fig. There may be single or multiple polyps. The incidence of polyps increases with age, particularly above 40 years of age (M:F = 2:1). The cumulative risk of cancer developing in an unremoved polyp is 2.5% at 5 years, 8% at 10 years, and 24% at 20 years after the diagnosis. The probability of any singular polyp becoming cancerous is dependent on its gross appearance, histologic features, and size. The risk of cancer is much higher in sessile villous adenomas than in pedunculated tubular adenomas (3). Cancer is found in 40% of villous adenomas, as compared to 15% in tubular adenomas. The good news is that 65% of adenomas are tubular, with villous adenomas accounting for only 10% of adenomatous polyps. Adenomas are believed to have an abnormal process of cell proliferation and apoptosis. Clinical, autoptical, and epidemiological studies provide evidence of adenoma-to-carcinoma progression. The mean age of adenoma diagnosis is 10 years earlier than with carcinoma, and progression to carcinoma takes a minimum of 4 years. Molecular genetic studies also describe an adenomato-carcinoma sequence through accumulation of lesions in a variety of genes, with activation of oncogenes and

Clinical Presentations

Imaging A double contrast barium enema (DCBE) is given in order to perform an X-ray examination of the large intestines. Pictures are taken after rectal instillation of barium sulfate (a radiopaque contrast medium) (Fig. 1). DCBE is generally accepted as less sensitive than colonoscopy for detecting polyps or CRC. Enthusiasm for the double contrast barium enema has declined in recent years in favor of colonoscopy, despite its lower cost and more recently in favor of ▶computed tomographic colonography (CTC). The reason for this decrease in use as a diagnostic tool lies in the reduced sensitivity of this test in detecting polyps of lesser than 1 cm, in detecting polyps in areas where a single lumen is not detectable (i.e., sigmoid, rectosigmoid, hepatic, and splenic flexures) and patient comfort and compliance issues. If a polyp is greater than 1 cm, the diagnostic accuracy of this technique has been reported as up to 95%; for polyps smaller than 1 cm well-performed air-contrast barium enema has a sensitivity of 61%. A false-positive rate of 5–10% is found because of improper cleaning of the bowel. Diverticulosis or redundant bowel can result in a false-negative rate of 10%, especially in the rectosigmoid. The accuracy


1203

Neoplasms, Benign, Large Bowel. Figure 1 Polypoid lesion in a 56 year old female with familial history of adenomatous polyposis. Close-up view of an overhead image from a double-contrast enema examination shows a 1.5 cm well-defined polypoid lesion (arrow) of the colon.

of the procedure also can have an element of operator dependence. CTC well known even as Virtual Colonoscopy (VC) is a novel imaging modality for the evaluation of the colonic mucosa in which thin-section spiral CT provides high resolution two-dimensional (2D) axial images; (Fig. 2a) CT data sets are edited off-line in order to produce multiplanar reconstructions (coronal and sagittal images) as well as three-dimensional (3D) modeling, including endoscopic-like views (Fig. 2b). Reported diagnostic accuracy for polyps >5 mm exceeds that of barium enema and approaches that of conventional colonoscopy. CT colonography requires no sedation—a major factor if dealing with old, unstable patients—or screening of asymptomatic subjects who would like to be able to return to work immediately after the procedure. In contrast, colonoscopy is associated with appreciable morbidity and even mortality including significant cardiovascular effects related to sedation. The success rate of CT colonography approximates 100%, if bowel preparation and distension are optimal, whereas up to 6% of conventional colonoscopy cannot reach the caecum. CT colonography has been investigated as a technique for colon cancer screening. Although it requires a full bowel cleansing similar to that required for conventional colonoscopy, the procedure requires no sedation or analgesia, and is faster to perform than conventional colonoscopy. When a colonoscopy is not

N

Neoplasms, Benign, Large Bowel. Figure 2 (a) Polypoid lesion in a 65 year old frail and uncooperative patient with rectal bleeding. Axial CT image obtained after colonic distension with air during CT colonography exam shows a 2 cm sessile polyp (arrow) of the descending colon. (b) Endoscopic like view of the same lesion where the polypoid lesion is easily appreciated (axial image obained using Lightspeed VCT 64 detector row, GE, USA; and endoluminal image obtained using VIATRONIX, V3D, USA).

possible CTC is an acceptable alternative. However, since it is only a screening procedure, patients with positive findings require conventional colonoscopy afterward. For this reason CT colonography is generally not being offered in a diagnostic setting (i.e., in symptomatic patients) because these patients have a high probability of having an abnormality that will require a colonoscopy. CT colonography can detect overt cancers, but these are rare in asymptomatic patients. The major benefit from

1204


CT colonography and other colon cancer screening techniques results from identification and removal of cancer precursors called adenomatous polyps. This assessment reviews evidence on the effectiveness of CT colonography as an alternative to colonoscopy for the purpose of colon cancer screening.

MR Colonography High performance gradient systems together with fast MRI scanning techniques allow the acquisition of complex 3D data sets within comfortable breath hold. Whereas MR has not been able to provide the necessary spatial and temporal resolution, this technique provides some advantages as optimal soft-tissue contrast and lack of radiation exposure. Fecal tagging to obviate the need for bowel preparation can be used in MR as well as in CTC. Compared to CTC, MR colonography is still under development and promises to reach the accuracy of CTC (4).

Diagnosis Diagnosis relies on good history and physical examination of patients at risk. Rectal examination can identify cancers up to 8 cm above the dentate line that represent 20% of colorectal cancer. Fecal occult blood testing (FOBT): is a test that detects the presence of occult (detectable only by chemical means and not visible) blood in the stool. This test has been suggested as a possible adjunct to other screening modalities. Only 20–40% of patients with adenomas have positive test findings, usually resulting from distal and larger polyps. On the other hand, the estimated 30–50% sensitivity is too low to consider FOBT as an effective single screening modality in high risk individuals, particularly since a positive FOBT is unlikely to occur as a result of adenomas before malignant transformation. Flexible sigmoidoscopy: Flexible sigmoidoscopy can reach as high as the descending colon and can be done by a trained primary care physician. Sigmoidoscopy has been proven to reduce the incidence and mortality of colon cancer through early detection, however, is not an adequate method of screening in hereditary colon cancer as 2/3 of the lesions develop proximal to the splenic flexure. In these cases colonoscopy should be used. Flexible sigmoidoscopy can detect about 65–75% of polyps and 40–65% of colorectal cancers. This test, for an investment of 3–5 min, can with little discomfort reduce the likelihood of your developing colon cancer and if colon cancer is present detecting it at an early, highly curable stage.

Neoplasms, Benign, Large Bowel. Figure 3 Conventional colonoscopy of 53-year old female with familial history of adenomatous polyposis showing a 1 cm lesion of the ascending colon.

Colonoscopy: Colonoscopy remains the gold standard for visualization, biopsy, and removal of colonic polyps (Fig. 3). The removal of all polyps by colonoscopy has been demonstrated to reduce the risk of colon cancer by 76–90%. Colonoscopy is the “gold standard” for the detection of colonic neoplasms and the preferred colorectal cancer screening strategy. The incidence rate of colorectal cancer has been shown to be reduced up to 90% in subjects who had polypectomy versus patients in three reference groups, including two cohorts in which colonic polyps were not removed and one general-population registry. It can be completed in more than 95% of examinations with negligible risk. Colonoscopic screening in individuals with average risk has been found to be cost effective, and similar to cervical or breast cancer screening techniques in costeffectiveness per life-year saved. Medicare has approved the use of screening colonoscopy in average-risk beneficiaries. The evidence to support colonoscopy is derived from data showing a decreased incidence of colorectal cancer mortality in subjects who have undergone colonoscopic adenoma removal. Additionally, colonoscopic screening has been shown to have favorable cost effectiveness when compared to other screening strategies.

Bibliography 1. 2.

Morson B (1974) The polyp-cancer sequence in the large bowel. Proc R Soc Med 67:451–457 Desai DC, Neale KF, Talbot IC et al (1995) Juvenile polyposis. Br J Surg 82:14–17

Neoplasms, Bile Ducts

3.

4.

Jass JR, Sobin LH (eds) (1989) WHO International Histological Classification of Tumors. Hystological Typing of Intestinal Tumours. 2nd edn. Springer, Berlin Heidelberg, New York Lauenstein T, Goehde S, Rueheme S et al (2002) MR Colonography with barium-based fecal tagging: initial clinical experience. Radiology 223:248–254

Neoplasms, Bile Ducts T HOMAS H ELMBERGER 1 , C ARLO B ARTOLOZZI 2 , M ARZIO P ERRI 2 , G ABRIELE C APRONI 2 1

Clinic of Radiology and Nuclearmedicine, University Hospitals Schleswig-Holstein, Campus Luebeck, Luebeck, Germany 2 Department of Diagnostic and Interventional Radiology, University of Pisa, Pisa, Italy [email protected] [email protected]

Synonyms Cholangiocarcinoma, CC; Cholangiocellular carcinoma, CCC; Klatskin tumor

Definition and Classification Cholangiocellular carcinoma (CCC) is a malignant tumor arising from the bile duct epithelium and comprises 15–25% of all liver and biliary tract malignancies. The tumors may arise at any part and from any component of the bile duct epithelium, ranging from the terminal ductules (canals of Hering) to the ampulla of Vater as well as at the peribiliary glands (intramural and extramural). According to the site of origin, cholangiocarcinoma can be differentiated into intrahepatic/peripheral, hilar (Klatskin tumor), and extrahepatic. Intrahepatic CCC includes tumors originating from small biliary intrahepatic ductules and is considered as a peripheral, mass-forming tumor. Hilar CCCs arise from one of the hepatic ducts or the bifurcation of the common hepatic duct and are classified, according to the Bismuth classification, into four different types: type I involves the main hepatic duct below the bifurcation, type II affects the main hepatic duct bifurcation, type III involves segmental ducts beyond the primary hepatic duct bifurcation in one liver lobe (type IIIa: right lobe, type IIIb: left lobe), and type IV involves segmental ducts in both liver lobes. Finally, CCCs originating from the bile duct below the bifurcation of the right and left hepatic ducts are classified as extrahepatic (1).

1205

Pathology The gross appearance of intrahepatic CCC is of a gray-white mass with irregular margins. The mass may be solitary or multiple with satellite nodules in its periphery. Size varies, from a few millimeters to more than 15 cm in diameter. The tumor consists typically of a huge amount of whitish fibrous tissue at cut section, especially in large lesions with central necroses. Calcifications and hemorrhages are rare. Fingerlike extensions along the portal triads are frequent. Similar to HCC, intrahepatic CCC has the propensity to invade small portal vessels, resulting in portal vein thrombosis. Metastases to regional lymph nodes and pulmonary and peritoneal spreading are common. Different histological forms of hilar and extrahepatic CCC (such as infiltrative, nodular, and papillary) have been described. The infiltrative type is the most common (over 70% of cases) and it appears as a sclerotic lesion with abundant fibrous tissue growing along the bile duct wall. It results in a diffuse, firm, gray-white annular thickening of the bile with complete or nearly complete obstruction of the lumen. The extent of the tumor may vary, ranging from few millimeters to several centimeters in length. A dense fibroblastic reaction may compress or encase the adjacent vascular structures. The nodular form usually consists of a small (1–2 cm in diameter), gray-white nodule causing biliary obstruction. The tumor arises within the mucosa, invades the bile duct wall, and grows outward to form a nodular, exophytic mass. The papillary CCC is a distinctive pathologic entity characterized by the presence of an intraluminal polypoid or sessile mass of the hepatic bile duct associated with partial biliary obstruction and dilatation. The tumor is usually small and is a low-grade malignancy (2).

Histopathology Irrespectively of the primary localization of the tumor, whether it is intra- or extrahepatic, more than 90% of CCCs are well to moderately differentiated adenocarcinomas that exhibit glandular or acinar structures with intracytoplasmic mucinous components. Characteristically, the malignant cells are of cuboidal or low columnar type resembling biliary epithelium. In more poorly differentiated tumors, solid cords of cells without biliary ducts may be present. The dense fibrous stroma is characteristic and may dominate the histological architecture. The tumor tends to invade lymphatics, blood vessels, perineural and periductal spaces, and portal tracts, which might be responsible for the secondary atrophy of the dependent hepatic segments particularly in peripheral CCC. Spread along the lumen of large bile ducts can be seen, especially in hilar tumors (2).

N

1206


Clinical Presentation CCC occurs more frequently in patients in the sixth decade of life. It is associated with primary sclerosing cholangitis, intrahepatic stone disease, choledochal cyst, congenital hepatic fibrosis, clonorchiasis, and exposure to Thorotrast. Clinical signs and symptoms, including jaundice, itching, clay-colored stools, dark urine, weight loss, and abdominal pain, may vary according to the various sites of origin of the tumor (1, 3). Jaundice is usually the leading sign in tumors located in the common bile duct or common hepatic duct as a sign of biliary obstruction. Abdominal pain and weight loss are the most common symptoms in intrahepatic CCC, whereas painless jaundice occurs later and is often a marker of advanced disease. Extrahepatic cholestasis is reflected in elevated levels of direct bilirubin. Levels of alkaline phosphatase and gamma-glutamyltransferase (GGT) usually rise in conjunction with bilirubin. There are no specific tumor markers for CCC, although elevation of serum carcinoembryonic antigen (CEA) and CA 19–9 is often found.

Imaging Intrahepatic CCC The most common appearance of intrahepatic CCC on ultrasound (US) is as an ill-defined, predominantly hypoechoic mass; however, US patterns may vary from homo- or heterogeneous hypo-, iso-, or hyperechogenicity. Hyperechoic foci with acoustic shadowing, suggesting calcifications, and segmental bile duct dilation in the presence of normal extrahepatic ducts may be observed.

The presence of satellite nodules is frequent and contributes to a poor prognosis of CCC. Color Doppler US allows definition of portal vein involvement, without significant usefulness in characterizing the primary tumor mass (3). On unenhanced computed tomography (CT) scans, intrahepatic CCC usually appears as a large and irregular mass hypodense or isodense relative to the normal hepatic parenchyma. The mass is round or oval, more or less well demarcated, and may demonstrate segmental biliary ductal dilatation peripheral to the tumor. After contrast medium administration, parenchymal CCC usually demonstrates a thin and incomplete rim-like enhancement during both the arterial and portal venous phases (Fig. 1). The central part of the tumor usually does not enhance during these phases, whereas there may be delayed enhancement based on the interstitial transition of extracellular iodinated contrast agents. This pattern of enhancement reflects the hypovascular, desmoplastic composition of most CCCs. Satellite nodules are frequent and variable in size. Additional ancillary CT findings include capsular retraction, intratumoral calcification, invasion of portal or hepatic veins more commonly than in HCC, and dilatation and thickening of peripheral intrahepatic biliary ducts (especially when associated with clonorchiasis). Extrahepatic spread is common and lymph node metastasis with involvement of the celiac and left gastric areas or direct invasion of the omentum is frequently detected (1, 4). The magnetic resonance (MR) appearance of intrahepatic CCC is that of a noncapsulated tumor, hypointense on T1-weighted images and slightly hyperintense on T2-weighted images. However, the signal intensity of the tumor is variable according to the amount of fibrosis, necrosis, and mucinous components within the tumor.

Neoplasms, Bile Ducts. Figure 1 Peripheral cholangiocellular carcinoma. On unenhanced CT scans (a) the lesion appears as an ill-defined and slightly hypodense mass. After contrast medium administration the lesion shows minimal peripheral enhancement during the arterial (b) and portal venous (c) phases. Peripheral biliary ducts dilatation, distal to the mass, is also evident.


1207

A central hypointensity, corresponding to fibrosis, is usually best appreciated on T2-weighted images. Dilatation of peripheral portions of the intrahepatic biliary ducts may be present and can easily be displayed with standard MR cholangiopancreatography (MRCP) sequences (heavily T2-weighted sequences) which is rarely seen in HCC. On dynamic contrast-enhanced MR studies, minimal to moderate peripheral enhancement is usually observed followed by moderate progressive, delayed filling. Pooling of contrast within the tumor, reflecting the large amount of fibrous tissue, is typically seen on delayed MR images corresponding to the known CT findings (5).

Hilar and Extrahepatic CCC The US appearance of hilar and extrahepatic CCC includes biliary duct dilatation, mass or bile duct wall thickening, and lobar atrophy with crowded, dilated ducts. Biliary duct dilatation is almost invariably present in hilar and extrahepatic CCC, and can be a useful sign in the differential diagnosis with other hepatic tumors. Typically, ▶Klatskin tumors manifest as segmental dilatation and nonunion of the right and left ducts at the porta hepatis. Although US is accurate for revealing the level of bile duct obstruction, visualization of the tumor mass is difficult and, therefore, characterization of a hilar or extrahepatic cholangiocarcinoma requires meticulous evaluation of the exact localization of the luminal alteration or ductal occlusion. The papillary and nodular forms of cholangiocarcinoma are relatively easy to see on US, appearing as a poorly defined, usually isoechogenic, nodular lesion with associated mural thickening. In contrast, infiltrating CCCs, the most common subtype, are difficult to appreciate on US, appearing only as a focal or diffuse thickening of the bile duct wall. Lobar atrophy is often extremely subtle on US images. In this case, US scans demonstrate crowded, dilated ducts within the atrophic lobe. The dilated ducts will often almost reach the liver surface, which is pathognomonic. The associated cholangiocarcinoma is often at a hilar localization with a predominant involvement of the duct which drains the dependent segment (3, 4). Hilar and extrahepatic CCCs usually appear as a focally thickened ductal wall obliterating the lumen on thin-section spiral CT scans. Following intravenous contrast medium administration, enhancement of the lesion is poor to moderate (Fig. 2). The degree of contrast enhancement probably depends on the presence of the fibroid components of the lesion. An enhancing border or pseudocapsule may be visible in well-delineated tumors. In a minority of cases there is early marked contrast uptake in the tumor, but often contrast enhancement is only noted on delayed scans (1, 4).

Neoplasms, Bile Ducts. Figure 2 Hilar cholangiocellular carcinoma. Portal venous phase spiral CT shows a large and hypodense mass with lobulated margins at the hepatic hilum, mainly in segment IV. The contrast enhancement of the tumor is peripheral and irregular. Dilatation of the left hepatic duct and peripheral biliary ducts is present.

Lobar hepatic atrophy with marked dilatation and crowding of bile ducts is usually observed on CT scans in patients with hilar CCC in addition to fuzzy invasion of adjacent liver parenchyma and of the hepatoduodenal ligament. In contrast, satellite nodules are less commonly seen in hilar CCC compared to intrahepatic CCC according to the earlier clinical manifestation of lesions situated at the hilum. Lymphatic metastases usually involve the portacaval, superior and posterior pancreatoduodenal lymph nodes. Retroperitoneal lymphadenopathy, peritoneal spread, and proximal intestinal obstruction occur in advanced stages. On MR images, hilar and extrahepatic CCCs appear hypointense to the liver on T1-weighted and moderately hyperintense on T2-weighted images, while on dynamic T1-weighted MR images acquired after gadolinium injection, minimal to moderate incomplete enhancement may be present at the tumor periphery on delayed images. In addition, the combined use of dynamic MR and MRCP depicts the primary tumor in a vast majority of cases and allows evaluation of the overall extent of the biliary tree involvement. The morphology of bile duct stricture detectable on high-resolution MRCP closely reflects the gross morphologic changes along the biliary ductal walls: papillary lesions appear as protuberant, as sessile flat or polypoid-like intraluminal lesions, while diffusely infiltrating tumors may be visualized as a diffuse narrowing of the ductal lumen (5).

N

1208


Percutaneous transhepatic cholangiography is indicated in any patient who is cholestatic with nondilated bile ducts when there is doubt about the diagnosis and in whom an endoscopic intervention is not possible.

The vast majority of HCCs are hypervascularized and therefore easily differentiated from CCCs. Nevertheless, some variants of HCC such as sclerosing, fibrolamellar, and cholangiohepatocellular carcinoma may mimic intrahepatic CCCs. Especially in the latter cases, even biopsy may be unable to resolve the diagnostic dilemma.

Nuclear Medicine 99m

Tc imminodiacetic acid (HIDA) and 99m Tc sulfur scans are the classic nuclear medicine techniques used to display hepatic tumors such as CCCs, but they are of significance only in comparison to contrast-enhanced CT and MRI. 99m Tc HIDA is excreted into the biliary ducts and may reveal the site of biliary obstruction. After injection, the common bile duct and cystic duct are usually visualized within 15 min. However, the ducts might not be visualized, even in healthy patients. 99m Tc sulfur colloid helps in localizing lesions larger than 2 cm. However, the appearances on sulfur colloid and HIDA scans are nonspecific and false-positive findings may occur as a result of the misinterpretation of benign tumors and other malignant tumors. Positron emission tomography (PET) can be used to assess the metabolism with the administration of positron-emitting radiolabel tracers such as 18F-FDG. Recent studies have demonstrated the potential role of PET. In particular, 18F-FDG-PET seems to improve the depiction of CCC superimposed on primary sclerosing cholangitis due to the additional inflammatory component and allows the detection of metastatic lymph nodes or peritoneal seeding that other modalities fail to demonstrate.

Diagnosis Intrahepatic CCC Intrahepatic CCC classically manifests as a large, welldefined hepatic mass with lobulated margins and peripheral rim enhancement on both contrast-enhanced spiral CT and dynamic MR images. Both modalities are equally effective in the detection and characterization of the tumor, enabling a correct diagnosis and staging of intrahepatic CCCs in a great number of cases (1). The differential diagnosis includes some bulky hepatic tumors of either primary or secondary origin. Due to a histologically similar appearance in metastatic adenocarcinomas, the differential diagnosis can be difficult for the pathologist. However, absence of a known primary malignancy, a relatively large size, and other ancillary findings such as bile duct dilatation and capsular retraction may support the diagnosis of CCC (4).

Hilar and Extrahepatic CCC Whenever extrahepatic CCC is suspected, the following aspects should be assessed: biliary dilatation, level of obstruction, presence of mass, focal or diffuse thickening of the bile duct wall, presence of a hepatic tumor, local lymph nodes involvement, and portal vein thrombosis. US is the primary diagnostic procedure in most cases. US is accurate in revealing the level of bile duct obstruction, but the tumor mass is detected in a few cases only (3–5). In contrast, thin-section contrast-enhanced spiral CT and MR (dynamic MR and MRCP) depict the lesion and allow correct evaluation of the tumor extent in a great number of cases. In addition, dynamic CT and MR provide an accurate evaluation of vascular involvement, of secondary liver atrophy, and of lymphadenopathy of the peripancreatic, periduodenal, periportal, celiac, and mesenteric nodes. The differential diagnosis of ductal CCC depends on its location. Lesions at the ductal confluence may be mimicked by inflammatory cholangitis (acquired immunodeficiency syndrome, sclerosing cholangitis), oriental cholangitis, benign biliary tumors, HCC, or gallbladder cancer, while ampullary or pancreatic cancer, nonshadowing stones, biliary adenomas or papillomas, blood clots, and benign strictures may simulate lesions of the distal common bile duct. Moreover, metastatic tumoral tissue in the bile ducts or adjacent nodes may mimic CCC at any anatomical level. Lobar hepatic atrophy with marked dilatation and crowding of bile ducts is highly suggestive of hilar and peripheral CCC, although long-standing biliary obstruction from surgical trauma or focal biliary obstruction can cause similar findings. Extrahepatic bile duct adenoma is a rare benign lesion of the biliary duct appearing as a polypoid filling defect within the bile duct at direct MRI or MR cholangiography. It is usually located at the common bile duct or at the common hepatic duct and is difficult to differentiate from small nodular extrahepatic or hilar CCC. However, additional findings of hepatoduodenal ligament adenopathy and tumor extension into adjacent structures are helpful diagnostic features supporting a malignant differential diagnosis (4). Moreover, since intraductal papillary CCC produces single or multiple intraductal masses, the differential diagnosis from a bile duct stone may be difficult. Furthermore, the clinical manifestations of both diseases are similar. On

Neoplasms, Bladder

US a papillary tumor does not cast a shadow like some nonshadowing stones, and on precontrast CT the tumor appears as a high- or low-attenuation soft tissue mass.

5. 6.

1209

Schneider G, Grazioli L, Saini S (2002) MRI of the Liver. Springer, Berlin, Heidelberg Lencioni R, Cioni D, Bartolozzi C (2005) Focal Liver Lesions: Detection, Characterization, Ablation. Springer-Verlag, Berlin, Heidelberg, New York

Interventional Radiology Although the prognosis of patients with CCC after surgical resection is poor, complete resection of the tumor is currently considered the best therapeutic option. Since only 20–40% of patients are suitable for surgical treatment, palliative interventional radiological and gastroenterological procedures, such as percutaneous biliary drainage and stenting, may be employed in an attempt to improve survival and life quality. Percutaneous transhepatic biliary drainage (PTBD) represents a useful palliative treatment for resolving jaundice. It is indicated as a preoperative or preinterventional decompression procedure, particularly when endoscopic treatment has failed or is not indicated. Preoperative PTBD helps to reduce the incidence of postoperative sepsis, abscess, bleeding, and renal failure. Obviously, urgent PTBD is mandatory in cases of neoplastic biliary obstruction complicated by acute cholangitis (6). Expandable metallic stents are often used in cases of biliary obstruction caused by an unresectable hilar CCC. Several studies show that the use of metallic stents is able to shorten hospitalization and to improve the quality of survival by lowering significantly the obstruction-related serum bilirubin levels. However, several early (fever, sepsis, bilioma, hemobilia, bile peritonitis, pancreatitis, cholecystitis) or late complications may occur. Late complications such as recurrent symptoms of cholangitis, cholangiohepatic abscess, or jaundice are usually related to stent occlusion by tumor ingrowth, overgrowth, and stent migration. In an attempt to prevent or to delay stent occlusion by tumor ingrowth, covered stents have been used. However, although covered devices impair tumor ingrowth, they cannot prevent neoplastic outgrowth. In addition, the long-term durability of the covered membrane is problematic.

Bibliography 1.

2. 3.

4.

Han JK, Choi BI, Kim AY et al (2002) Cholangiocarcinoma: pictorial essay of CT and cholangiographic findings. Radiographics 22 (1):173–187 Lim JH, Park CK (2004) Pathology of cholangiocarcinoma. Abdom Imaging 29(5):540–547 Bloom CM, Langer B, Wilson SR (1999) Role of US in the detection, characterization, and staging of cholangiocarcinoma. Radiographics 19:1199–1218 Lee WJ, Lim HK, Jang KM et al (2001) Radiologic spectrum of cholangiocarcinoma: emphasis on unusual manifestations and differential diagnoses. Radiographics 21Spec No:S97–S116

Neoplasms, Bladder G. H EINZ -P EER Department of Radiology, Medical University, Vienna, Austria [email protected]

Definition Carcinoma of the urinary bladder is one of the most common malignant tumors of the urinary tract and represents 2% of all malignancies. It is a disease of later life with a peak incidence during the seventh decade. Bladder cancer is three times more common in men than women and it affects twice as many Whites as Blacks (1). The most well documented risk factor of bladder cancer is cigarette smoking, to which has been attributed up to 45% of urothelial cancers (1). Although difficult to prove with certainty, occupational exposure to b-naphtylamine, paints, oils, gasoline, zinc, chromium, and rubber has been associated with 18–40% of bladder cancer cases. Other risk factors include chronic bladder infection or inflammation, pelvic irradiation, and treatment with cyclophosphamide (1). A genetic propensity to develop bladder cancer has also been observed (1). Other primary tumors, such as non-Hodgkin’s lymphoma and pheochromocytoma are very rare and may arise in the bladder wall. Secondary involvement of the bladder by lymphoma may occur, representing direct spread from grossly involved pelvic lymph nodes. Metastases may also develop within the bladder wall, the extraluminal component being more obvious than the intraluminal mass, providing a clue to the diagnosis in patients with known disseminated disease due to cancers such as malignant melanoma and breast cancer.

Pathology Most bladder tumors arise from the urothelium. Transitional cell tumors account for 95% of all primary malignant bladder lesions (2). The rest are squamous cell carcinomas, mixed transitional and squamous cell tumors, adenocarcinomas, and undifferentiated lesions

N

1210

Neoplasms, Bladder

(2). Most bladder cancers arise on the lateral walls (47%) or in the region of the trigone (21%). Adenocarcinomas, which account for 3% of all bladder tumors, usually arise in the region of the trigone, but are occasionally seen as exophytic growths arising at the bladder dome and originating from a persistent urachus. About one-third of cancers are multifocal at the time of diagnosis, and the whole bladder epithelium may undergo malignant change. Recurrences and new tumors are frequent (2). Bladder tumors show a variable pattern of growth and are classified as papillary, infiltrative, papillary and infiltrative, or nonpapillary and noninfiltrative (carcinoma in situ). Grade I lesions are well-differentiated tumors and are usually papillary, whereas grade III lesions are poorly differentiated and frequently show an infiltrating pattern of growth. Tumor grade correlates with the natural history of the disease, as those with grade III tumors invading muscle have a significantly worse prognosis than those with grade I or II superficial lesions (1). Approximately one-third of patients present with muscle invasive disease at the time of initial diagnosis (1). Tumor stage refers to the depth of invasion. The tumor/nodes/metastasis (TNM) staging system is most widely used. Historically, tumors confined to the mucosa (Ta) and lamina propria (T1) have been classified as superficial; muscle invasive tumors are classified as T2-T3. In general, there is a correlation between T-stage and the risk of recurrence, progression, and metastasis. Once the tumor has penetrated the basement membrane (T1) or muscle (T2-T4), there is an increased likelihood of distant metastases, depending on the depth of penetration (Fig. 1). The tumor may involve adjacent organs (prostate, uterus, vagina, rectum, small intestine) and extend to the pelvic side wall. Transitional cell carcinoma

(TCC) of the prostate is present in approximately 40% of men with invasive bladder cancer (1). Coexistent adenocarcinoma of the prostate is present in 40% of men undergoing cystoprostatectomy for invasive TCC (1). Synchronous upper urinary tract urothelial tumors occur in 2–5% of patients (2). Distant metastasis may occur via the lymphatic or vascular systems. The regional lymphatics include the paravesical, obturator, and external iliac lymph nodes. The most common sites of distant metastasis are liver, lung, and bone.

Clinical Presentation Patients most commonly present with painless ▶hematuria, either gross or microscopic. The bleeding may be intermittent, so one should not be lured into a sense of false security by the spontaneous disappearance of bleeding. Irritative voiding symptoms, including frequency, urgency, and dysuria, occur and are often associated with CIS. This constellation of symptoms, however, may also occur with benign entities, such as urinary tract infection, prostatism, and prostatits. Locally advanced tumors may lead to ureteric obstruction, pelvic side wall muscle invasion, or invasion of adjacent organs. These advanced tumors present with pelvic or abdominal pain and symptoms related to the urinary tract.

Imaging Excretory Urography In the past, routine imaging surveillance consisted of periodic excretory urography. In addition, excretory urography was generally performed to examine the upper urinary tract for synchronous tumors, which occur in 2.3% of patients with TCC of the urinary bladder. Excretory urography can also be used to detect other abnormalities such as stones or masses that could account for the patients’ hematuria. However, negative findings in excretory urography cannot exclude urothelial cancer.

Computed Tomography

Neoplasms, Bladder. Figure 1 Depth of tumor penetration according to TNM system of staging bladder carcinoma.

Bladder tumors appear on CT as soft tissue density lesions arising from the bladder wall. The tumor may be sessile or pedunculated, but in some cases the only abnormality seen is thickening of the bladder wall. Tumors enhance following injection of intravenous contrast medium, often to a greater degree than the normal bladder wall. In the pre-MDCT era, several limitations and difficulties associated with CT staging of bladder cancer

Neoplasms, Bladder

have been reported. Review of the literature shows that the accuracy of CT study in the staging of bladder neoplasm has varied between 68 and 85%, and it has been shown that CT has failed in staging early tumors (TIS-T3a). In patients with suspected perivesical spread and more advanced disease, however the usefulness of CT has been reported. CT has been regarded as inferior to MRI for staging bladder cancer, but as similar in evaluation of extravesical disease. Multidetector CT offers new possibilities for imaging of the urinary bladder and urinary tract. Using thin collimation, near isotropic imaging of the urinary tract is possible and provides high quality multiplanar reformations and 3D reconstructions of the organ including virtual cystoscopic views. Since the initial report of Vining et al (3), a growing number of articles on detection of bladder lesions with ▶virtual cystoscopy have been published. In general, results of these studies indicated that virtual cystoscopy allows accurate assessment of localization and morphology (pedunculated or sessile) of bladder masses. Additionally, the decrease of post processing time from 6–8 h to a few minutes allows its routine use in the clinical praxis. Song et al (4) showed that transverse and virtual views are complementary in lesion detection and characterization, and thus should always be used for accurate lesion detection. These authors also recommend that imaging in both positions is necessary for visualization of the entire mucosal surface without obscurity caused by residual urine (4). Preliminary results of various studies show excellent detection rates, including lesions 1 cm was 100%. Utilizing T2-weighted and Gd-enhanced T1-weighted MR-urography, Beer et al (7) evaluated maximum intensity projection (MIP) and virtual endoscopy in patients with suspected urogenital disease. Results of this study showed that both datasets were equal regarding virtual ureterorenoscopy. For virtual

Neoplasms, Bladder

1213

Neoplasms, Bladder. Figure 5 Pedunculated bladder carcinoma (arrow): Mobility while dual positioning. Stalk (arrowhead). No signs of bladder wall infiltration—T1 tumor.

cystoscopy, T2-weighted datasets were superior to the Gdenhanced T1-weighted sequences. In addition, 86% of endoluminal diseases were revealed on virtual endoscopy (VE) compared to 15% on MIP images. In renal pelvis evaluation MIP images were superior to VE. Negative contrast agents do not have any role in this setting (7).

Diagnosis When a clinical diagnosis of bladder tumor is suspected, the initial investigation consists of urine cytology and lower urinary tract endoscopy. The development of flexible cystourethroscopy has allowed diagnostic endoscopy to be performed safely in an office setting with improved patient comfort. It allows thorough endoscopic assessment of the entire urethra, including the prostatic urethra in men, and the entire urinary bladder. Papillary and well-circumscribed bladder tumors are easily recognized. Areas of CIS may appear only as red areas on the mucosa. After endoscopic visualization of a bladder tumor, the histopathologic diagnosis and initial treatment are achieved by transurethral resection (TUR) of the tumor. Complete endoscopic resection of the tumor using a loop diathermy electrode is attempted, unless there is obvious extravesical extension. Areas of abnormal appearing mucosa are biopsied (1). However, conventional cystoscopy may be limited by diminished visualization of areas such as the mucosa of the

bladder neck and within diverticula. In addition, conventional endoscopy proves to be technically very difficult in patients with urinary diversion, and some structures may not even be visualized. Conventional cystoscopy may be invasive, uncomfortable, time consuming, and expensive. In addition, rare complications like iatrogenic injury to the urethra and bladder as well as urinary sepsis may occur. In patients with bacteriuria, acute cystitis, urethritis, prostatitis, obstructive prostatic hypertrophy, and stricture or rupture of the urethra conventional cystoscopy may be contraindicated. The treatment and prognosis of urinary bladder carcinoma is largely determined by the depth of tumor growth and the extent of tumor metastases (1). Bladder saving treatment is used for superficial tumors (stages Ta-T1), whereas for stage T2-T3b tumors, radical cystectomy is performed. The treatment for stage T4a and T4b tumors and for metastatic disease is usually palliative radiation and chemotherapy, respectively. The accuracy of the UICC clinical staging which includes deep fractioned TUR is sufficient for superficial tumors. However, the accuracy of clinical staging for stage T2-T4b tumors is poor. In conclusion, CT- and MR-imaging including virtual endoscopy techniques of the urogenital tract have enormously improved over the past few years. Contributing to this progress is the continuing improvement of MR hard- and software as well as the advent of ▶Multidetector computed tomography (MDCT) and the ability

N

1214

Neoplasms, Bone, Benign

to acquire near isotropic datasets. Commercially available 3D workstations and the rapid evolution of supporting software also fasten the improvement. Considering the limitations and contraindications of conventional cystoscopy, CT- or MR-imaging including virtual cystoscopy seems to be an alternative in a patient suspected or known to have bladder cancer who is not a candidate for conventional cystoscopy. In addition, these techniques seem to be most appropriate in evaluation of the upper urinary tract. The high accuracy in detection of lesions spine > others LTB: diaphysis > meta-/epiphysis Spine: posterior elements Spine, flat bones, LTB > others Spine: posterior elements LTB: diaphysis > metaphysis STB of the hand, LTB (femur, LTB: (dia)metaphyses humerus, tibia) LTB (femur, humerus, tibia) > LTB: metaphysis others spine: posterior elements LTB (humerus, femur, tibia) > LTB: epiphysis (epimetaphysis), others apophysis LTB, calcaneus > others LTB: metaphysis Spine, skull > others

Spine: vertebral body

LTB > others

LTB: epi/metaphysis spine: vertebral body

Imaging Osteoid osteoma (Fig. 1) in cortical locations typically has the radiographic appearance of a round or oval radiolucency of 1.5 cm in diameter and showing a variety of radiographic features. Most lesions originating from

the medullary canal or cortex of long tubular bones display a geographic lytic pattern delineated by a sclerotic rim, often in association with expansion of bone and periosteal reactions. However, in most cases the radiographic findings are nondiagnostic. Osteoblastomas of the spine can present as expansile lesions that arise from the posterior elements, most commonly of a thoracic or lumbar vertebra, but they can also be invisible on conventional radiographs. As in osteoid osteoma, CT is therefore the preferred imaging modality for detecting the nidus, assessing the patterns of bone destruction and periosteal reaction, depicting osteogenic matrix mineralizations, and defining tumor extent before surgical resection. MR imaging usually offers more unspecific findings with various signal intensities within the nidus and prominent edema of nearby bone marrow and soft tissues. Osteochondroma (osteocartilaginous exostosis; Fig. 2) can occur as a solitary lesion or as part of a familial skeletal syndrome with autosomal dominant inheritance, termed hereditary multiple exostoses. Radiographs are usually pathognomonic, as they show one single or multiple sessile or pedunculated osseous excrescenses in continuity with the marrow and cortex of the host bone. The osseous stalk is covered by a cap of hyaline cartilage that might demonstrate calcifications on radiography and that usually ossifies with skeletal maturity. Malignant transformation into a (low-grade) chondrosarcoma is rare in patients with solitary osteochondroma, but has been reported to occur in 5–25% of patients with multiple lesions. Malignancy should be suspected if the thickness of the cartilage cap exceeds 2 cm

N

1216


Neoplasms, Bone, Benign. Figure 1 Benign bone neoplasms: osteogenic tumors. (a) and (b) Osteoid osteoma. (a) Anteroposterior radiograph shows circumscribed cortical hyperostosis at the midshaft of the femur. The nidus is hardly visible on radiography. (b) Corresponding computed tomography scan demonstrates centrally mineralized nidus within extensive periosteal new bone formation. Note associated sclerosis of adjacent medullary bone. (c) Osteoblastoma of the ilium. Radiograph shows geographic bone destruction with relatively well-defined margins, incomplete sclerotic rim, and central mineralizations. The cortex is partially destroyed.

Neoplasms, Bone, Benign. Figure 2 Benign bone neoplasms: chondrogenic tumors. (a) and (b) Osteochondroma. (a) Anteroposterior radiograph of the femur shows sessile bony outgrowth arising from the distal metadiaphysis. Note continuity of the cortex and marrow with those of tumor-bearing bone. (b) Corresponding transverse T2-weighted magnetic resonance image with fat suppression demonstrates hyperintense cartilage cap covering the osseous stalk. (c) Enchondroma. Radiograph of the third finger shows centrally located, lobulated lucency with endosteal scalloping of the cortex within the proximal phalanx. (d) Chondroblastoma. Anteroposterior radiograph demonstrates round osteolytic lesion with sclerotic rim and central mineralizations in the proximal epimetaphysis of the tibia.


in adults and 3 cm in children. MR imaging with the use of T2-weighted pulse sequences represents the most reliable method to assess cartilage cap thickness, whereas CT shows a tendency for underestimation, and ultrasound can be limited by the anatomic localization and orientation of the osteochondroma. Enchondroma (Fig. 2) of long tubular bones typically presents as a centrally located oval radiolucency with a sharply delineated and lobulated contour, various degrees of endosteal scalloping, and cartilaginous matrix mineralizations. Punctate, flocculent, or rings-and-arcs-like calicifications/ossifications represent characteristic mineralization patterns that usually allow for the radiographic diagnosis. Lesions located in the short tubular bones of the hand more often appear purely osteolytic, with more pronounced endosteal erosion or even expansion of bone. Epiphyseal or axial location, a lesion size >5–6 cm, periosteal reactions, cortical thickening, deep endosteal erosion (more than two-thirds of cortical thickness) or expansion of a major bone, cortical penetration, and enlargement of a radiolucency over time are findings that should alert to the differential diagnosis of chondrosarcoma. In this regard, CTrepresents the most valuable imaging method because it is best suited to assess the integrity of the cortex. On MR imaging, enchondromas reveal the typical morphology of a well-differentiated cartilaginous lesion with low signal intensity on T1-weighted images, very high signal intensity on T2-weighted images, and peripheral/septal enhancement following contrast administration (“rings and arcs”).

1217

However, compared with CT, MR imaging is less reliable in depicting cortical alterations. Chondroblastoma (Fig. 2) in typical cases shows the radiographic appearance of a round osteolytic lesion with thin sclerotic borders, which is eccentrically positioned within the epiphysis of the affected bone. Approximately 50% of chondroblastomas show matrix mineralizations that are detectable on conventional radiographs. Solid or lamellated periosteal reactions might be visible at the metaphysis or the diametaphyseal shaft. On MR images, chondroblastomas often show more inhomogeneous signal and a less typical enhancement pattern than other benign cartilaginous lesions. Furthermore, the lesions often cause edema of adjacent bone marrow and reactive synovitis. Intraosseous lipoma (Fig. 3) usually presents as a nonaggressive radiolucent lesion surrounded by a thin sclerotic margin. Trabeculations and osseous expansion might be evident. Central calcifications or ossifications due to liponecrosis are seen in many cases. Particularly in the proximal femur and the calcaneus, this radiographic appearance is virtually pathognomonic. CT and MR imaging can both be used to verify the lipomatous character of the lesion. Cystic areas, calcifications, and new bone formation are additional findings in lesions with partial or extensive fat necrosis (stages 2 and 3 lipomas). Hemangiomas (Fig. 3) of the spine usually involve the vertebral body and produce a coarse, vertically oriented trabecular pattern on radiography. On transverse CT or MR images, these thickened trabeculae are seen as

Neoplasms, Bone, Benign. Figure 3 Benign bone neoplasms; miscellaneous lesions. (a) Intraosseous lipoma: lateral radiograph shows the pathognomonic appearance of a calcaneal lipoma with delicate sclerotic margin and central mineralization. (b) Hemangioma: lateral radiograph shows rarefication the of bone structure of the first lumbar vertebra with a vertical trabecular pattern. Note extension into the pedicles. (c) Giant cell tumor: anteroposterior radiograph demonstrates a purely osteolytic lesion with eccentric localization in the proximal epimetaphysis of the tibia. The tumor shows well-defined intraosseous borders and destruction of the cortex with formation of a slightly expanded, incomplete neocortex.

N

1218

Neoplasms, Bone, Malignant

multiple punctate areas within the matrix of the hemangioma, termed polka-dot appearance. At extraspinal sites, hemangiomas show more variable radiographic features, although lytic lesions with slight expansion and a radiating, lattice-like or honeycomb trabecular pattern a highly suggestive of the diagnosis. Particularly in anatomically complex areas, CT can be helpful in identifying these features. The MR imaging appearance of hemangiomas varies with its composition of vascular and nonvascular components. Lesions with a predominantly fatty matrix show high signal intensity on T1-weighted images, intermediate to high signal intensity on T2-weighted images, and a loss of signal on STIR or fat-suppressed T2-weighted images. Predominance of the vascular component results in hypointensity on T1-weighted images, hyperintensity on STIR and T2weighted images, and marked contrast enhancement. Giant cell tumor (Fig. 3) typically arises as a purely osteolytic lesion with relatively well-defined margins, which in most cases lack a sclerotic rim. In long tubular bone, the majority of lesions show eccentric location within the epimetaphysel region and variable degrees of cortical violation without a significant periosteal reaction. MR imaging is best suited to demonstrate the true extent of giant cell tumors, particularly with a view to extraosseous growth, which is observed in more than 30% of cases. Many giant cell tumors arise with more or less low signal intensity on T1- as well as T2-weighted MR images, a feature that has been attributed to the presence of hemosiderin deposits and/or high collagen content. Contrast enhancement is usually marked and diffuse, except in cystic areas that might develop following hemorrhage or due to formation of a secondary aneurysmal bone cyst.

chondroblastoma, and giant cell tumor should undergo biopsy, and both radiologic and histologic findings have to be considered together to establish the final diagnosis.

Bibliography 1.

2.

3. 4. 5.

Fletcher CDM, Unni KK, Mertens F (eds) (2002) Pathology and genetics of tumours of soft tissue and bone. In: World Health Organization Classification of Tumours. IARC Press, Lyon Resnick D (1995) Tumors and tumor-like diseases: Imaging and pathology of specific lesions. In: Resnick D (ed) Diagnosis of Bone and Joint Disorders. 3rd edn. Saunders, Philadelphia, pp 3628–3938 Giudici MA, Moser RP, Kransdorf MJ (1993) Cartilaginous bone tumors. Radiol Clin North Am 31:237–259 Bloem JL, Kroon M (1993) Osseous lesions. Radiol Clin North Am 31:261–278 Woertler K (2003) Benign bone tumors and tumor-like lesions: value of cross-sectional imaging. Eur Radiol 13:1820–1835

Neoplasms, Bone, Malignant M ARK DAVIES MRI Centre, Royal Orthopaedic Hospital, Northfield, Birmingham, UK [email protected]

Synonyms Chondrosarcoma; Fibrosarcoma; Osteo sarcoma

Definition Nuclear Medicine In general, skeletal scintigraphy does not play a major role in the diagnosis of benign bone tumors. Enchondroma, intraosseous lipoma, and hemangioma can represent incidental findings on examinations with bone-seeking agents. The double-density sign is a relatively typical scintigraphic finding in patients with osteoid osteoma. The term describes a hot spot representing the nidus surrounded by an area of less increased uptake, which corresponds with the reactive zone of sclerosis.

Diagnosis Osteoid osteoma, osteochondroma, enchondroma, intraosseous lipoma, and hemangioma can usually be sufficiently diagnosed by imaging and require histologic verification only in doubtful cases. Osteoblastoma,

Primary malignant bone neoplasms/tumours arise from the osseous tissues (e.g. ▶osteosarcoma) or the bone marrow tissues (e.g. myeloma). They tend to be locally aggressive causing bone destruction and have the propensity to develop metastases to distant organs— typically the lungs.

Pathology/Histopathology Bone tumours are classified according to their tissue of origin and then into benign (see Neoplasms, bone, benign) and malignant subtypes (Table 1) (1). The incidence is less than one new case per 100,000 population per year. This compares with an incidence for tuberculosis in the developed world of approximately one per 10,000 population per year. The management and prognosis for a sarcoma depends on the histological grade. The higher the


Neoplasms, Bone, Malignant. Table 1 Classification of the most common malignant primary bone tumours by tissue of origin. Percentages show relative frequency (excluding myeloma) Osteogenic tumours Osteosarcoma • Conventional high grade • Telangiectatic • Low grade central • Secondary (radiation-induced/Paget’s disease) • Surface (parosteal/periosteal/high grade surface) Cartilage tumours Chondrosarcoma • Central (primary and secondary) • Peripheral (primary and secondary) • Dedifferentiated • Mesenchymal • Clear cell Fibrogenic and Fibrohistiocytic tumours Fibrosarcoma Malignant fibrous histiocytoma Ewing’s sarcoma/Primitive neuroectodermal tumour Ewing’s sarcoma Hematopoietic tumours Myeloma Lymphoma Notochordal tumours Chordoma Vascular tumours Angiosarcoma

35%

26%

6% 16%

8% 1%

grade, the increased likelihood after surgery of developing local recurrence and metastatic disease. In general, higher grade lesions show increased cellularity, nuclear enlargement and hyperchromasia, mitotic figures and necrosis. It is important to stress that bone tumour pathology is far from straightforward. Close radiological–histological correlation is important as a small biopsy may not show representative tissue (sampling error) or fail to reveal the highest grade component of the lesion. A multi-disciplinary approach to diagnosis is recommended and both biopsy and pathological assessment is best carried out in specialist centres (2).

Clinical Presentation Primary bone malignancies, with the exception of chondrosarcoma and myeloma, develop in children, adolescents or young adults. Pain and swelling in an otherwise fit individual are the cardinal signs. Frequently,

1219

tumours, particularly around the knee, may be mistaken for athletic injuries. The pain generally increases in severity and is typically non-mechanical in nature. Central lesions developing from the pelvis or shoulder girdle tend to present later than extremity lesions due to the relative size of the normal to neoplastic tissues. Occasionally, the tumour may present with a pathological fracture although this is more commonly seen with benign lytic bone tumours such as the simple bone cyst. Ewing’s sarcoma may present with fever, fatigue and a raised serum white cell count suggestive of osteomyelitis.

Imaging Despite newer imaging techniques, the radiograph is the preliminary and single most important imaging investigation. Early signs of a bone sarcoma and/or osteomyelitis include areas of ill-defined lysis of sclerosis of bone, periosteal new bone formation and soft tissue swelling (Fig. 1). In time, cortical destruction will occur with evidence of a soft tissue mass. The pathological process may be well established even in the presence of a normal radiograph. Up to 50% of trabecular bone must be destroyed before a discrete area of lucency can be seen radiographically. Erosion or destruction of the cortex is more readily seen. Once a bony abnormality has been detected, the next objective of imaging is to attempt to characterize the lesion and, in doing so, indicate an appropriate differential diagnosis. Subsequent imaging is directed at surgical staging if the lesion is considered likely to be a sarcoma (see later).

Diagnosis When faced with a radiograph of a potential bone sarcoma it is essential to know the age of the patient and whether there is an evidence of pre-existing bone lesions which might pre-dispose to malignant transformation (Table 2). The next important factor to note is the site of origin in the skeleton and which part of the bone is involved. Most bone tumours, benign and malignant, occur around the knee and as such little diagnostic information can be deduced by noting the affected bone in these cases. There are exceptions. Cartilage tumours of the hands and feet are common and usually benign. Adamantinoma, a low-grade sarcoma, classically involves the diaphysis of the tibia and is extremely rare at any other site. Chordoma characteristically arises from the clivus or sacrum. Both the longitudinal and transverse origin of the tumour should be noted. Malignancy is particularly rare in the epiphysis pre-skeletal fusion. Most osteosarcomas

N

1220


Neoplasms, Bone, Malignant. Figure 1 Child with early osteosarcoma. There is lysis of the lateral tibial metaphysis with minor cortical erosion.

Neoplasms, Bone, Malignant. Figure 2 Adolescent with Ewing’s sarcoma femoral diaphysis. There is permeative bone destruction, lamellar (onion skin) and spiculated periosteal reactions and soft tissue extension.

Neoplasms, Bone, Malignant. Table 2 Bone lesions associated with an increased risk of sarcomatous degeneration (modified from reference 1) High Risk Ollier’s disease and Maffucci’s syndrome Familial retinoblastoma syndrome Rothmund-Thomson syndrome Moderate Risk Hereditary multiple exostoses (diaphyseal aclasis) ( right 10%:3rd ventricle 80–90%: midline infiltrating mass near 3rd ventricle Well-circumscribed

Hypointense blood products

Heterogeneous

Hyperintense Hemorrhage Hyperintense cysts to CSF

Heterogeneous Isointense to hypointense

Hyperintense cysts

Isointense

CSF seeding

Occasional cysts +small necrotic foci CSF seeding Strong homogenous enhancement

Strong homogeneous enhancement

Variable heterogeneous mild to moderate enhancement

Meningeal spread + CSF seeding

Heterogeneous enhancement

CSF seeding

Rarely, meningeal enhancement Heterogeneous enhancement

Heterogeneous

Possibly heterogeneous Solid: isointense to hyperintense to GM Hyperintense Surrounding edema

50% enhance: solid, rim, or nodular

Hyperintense mass

Hypointense to isointense to GM

Solid: hypointense to isointense

Isointense to hyperintense Hyperintense cysts Hypointense blood products 75%: isodense to Isointense to hyIsointense to hyperdense pointense hypointense Ca2+: 25%: CSF entrapment Flow voids in tumor Ca2+: hypointense Hyperdense to GM Isointense to Isointense to hypointense to GM hyperintense to GM Large mass: Hyperintense: cysts + necrosis cysts + necrosis + Hypointense blood hemorrhage products

10–20%: disseminated hydrocephalus Intra-/periventricular large soft mass Variable density growing out of foramina Small cysts + hemorrhage Ca2+: 50%

40%: hemispheric or suprasellar

Hypodense to isodense Cysts + Ca2+ common

Hypodense to isodense Ca2+: 50–70%

Cortically based circumscribed cyst + Variable density, mural nodule or solid tumor mostly hypodense Remodeling of bone Ca2+: 35–50%

Large complex mass + surrounding edema Homogeneous or heterogeneous mass Often hemorrhage + necrosis Atypical teratoid rhabdoid Heterogeneous tumor 50%: infratentorial

Primitive neuroectodermal tumor

Ganglioglioma

Neoplasms, Brain, Supratentorial, Pediatric 1241

N

1242

Neoplasms, Chest, Childhood

Nuchal rigidity, back pain, and radicular symptoms may indicate intracranial or intraspinal cerebrospinal fluid dissemination.

be ruled out. Postoperative imaging should be performed as soon as possible, preferably within 24 h of surgery, because this allows discrimination between residual tumor and postoperative blood–brain barrier disruption.

Imaging Bibliography Various imaging techniques can be used to examine the cranial vault, including transfontanellar ultrasonography (US), computed tomography (CT), magnetic resonance imaging (MRI), and digital subtraction angiography (DSA; Table 1). US is primarily used in neonates. Sagittal and coronal images can be acquired at the bedside without the use of ionizing radiation. Midline shifts and obvious tumors may be diagnosed, but a detailed study of the neuroanatomy is limited. CT is a first-line imaging tool in children who have acute onset of neurological symptoms. MRI’s high spatial resolution, multiplanar imaging capability, ability to generate different tissue contrasts, use of contrast media, functional techniques—including diffusion-weighted imaging, perfusion-weighted imaging, and MR spectroscopy—and the lack of ionizing radiation make it ideally suited to image the child with a brain tumor (Fig. 1a, c). Because of the long acquisition times, the examination is often performed under general anesthesia, especially in young children. DSA is rarely indicated. Advanced functional MRI techniques have replaced most scintigraphy indications. It goes without saying that currently no imaging modality is capable of making a histological diagnosis. But by combining the anatomical and functional imaging information with the child’s age, the neurological symptoms, and the tumor’s location and progress over time, the histology of the lesion may be speculated with varying degrees of certainty.

1. 2. 3.

Barkovich AJ (2005) Pediatric Neuroimaging. 4th edn. Lippincott Williams & Wilkins, Philadelphia, pp 506–614 Osborn AG (2004) Diagnostic Imaging Brain. 1st edn. Amirsys Inc., Salt Lake City, pp I–6–4–104 Tortori-Donati P (2005) Pediatric Neuroradiology: Brain. Springer, Berlin, Heidelberg, pp 329–411

Neoplasms, Chest, Childhood J OHAN G. B LICKMAN 1 , D EWI A SHI 2 , Y VONNE L. H OOGEVEEN 2 , C ARLA B OETES 2 1

Department of Radiology, UMC St Radboud, Nijmegen, The Netherlands 2 Department of Radiology, Faculty of Medicine, University of Indonesia, Cipto Mangunkusumo General Hospital, Jakarta Pusat, Indonesia [email protected]

Synonyms Bronchial adenocarcinoma; Ewing’s sarcoma; Germ cell tumors; Germinomas; Histiocytosis; Inflammatory pseudotumor; Lymphoma; Neuroblastoma; Plasma cell granuloma; Primitive neuroectodermal tumors; Teratocarcinomas; Teratomas; Thymoma;

Definition Diagnosis Diagnosis relies on an accurate neurological examination. The radiologist should guide the neurologist in choosing the appropriate imaging modality and should localize the tumor with regard to the different anatomical and functional regions within the CNS. In addition, displacement, encasement, or infiltration of important, potentially threatened fiber tracts or vessels should be studied. By combining the age, location, and biological behavior of the tumor, the radiologist should be able to make the most likely diagnosis. In addition, he or she should differentiate the lesion from nonneoplastic lesions that can mimic tumors. Furthermore, before surgery, tumor seeding within the cerebrospinal canal or ventricular system should

The ▶thymus is a gland situated in the ▶anterior mediastinum. It is the source of T lymphocytes and plays an important role in cell-mediated immune responses (see Pediatric Imaging Case Review Ward, Robert J. Blackman, Hans. 2005. Elsevier health sciences). Bronchial adenocarcinomas are rare neoplasms of the tracheal bronchial tree commonly referred to as adenomas, which are low-grade malignant lesions that can metastasize. Benign ▶bronchial adenomas are extremely rare. More then half of these lesions are bronchial cancerous tumors that arise from ▶Kulchitsky’s cells of the respiratory epithelium, part of the amine precursor uptake and decarboxylation (▶APUD) system. Mucoepidermoid carcinomas account for 80% of bronchial adenomas.


Chest wall ▶Ewing’s sarcoma and primitive neuroectodermal tumors are biologically related lesions. The peak incidence is between 10 and 15 years of life, and they are more common in males. ▶Germ cell tumors include teratomas, teratocarcinomas, germinomas, and embryonal cell carcinomas, among others. Histiocytosis is a group of disorders characterized by granulomatous lesions in various organs. It occurs in all ages and there is no difference between boys and girls. The etiology of inflammatory pseudotumor of the lung is multifactorial and probably a repair phenomenon. It is the most common mass lesion of the lung in children, comprising approximately 50% of benign lung lesions in childhood. They can occur at any age; the youngest case reported is of a 12-month-old patient. Sometimes these lesions are difficult to differentiate from embryonic pulmoblastoma (pulmonary blastoma), which may have a malignant and semimalignant course. Lymphoma: In the first years of life, these lymphatic tumors are known as non-Hodgkin’s lymphoma (NHL). In the teenage years, these tumors predominantly represent Hodgkin’s lymphoma/disease (HD) and are almost always associated with lymphadenopathy (supradiaphragmatic). In the younger age group, the cells can be difficult to distinguish from acute lymphoblastic leukemia and are therefore known as “leukemia lymphoma complex.” Of all mediastinal tumors, approximately 20% consist of malignant lymphomas of the mediastinum. Together with germ cell tumors they account for 80–85% of anterior mediastinal masses. There is a male preponderance in the pediatric population. Neuroblastoma is a neurogenic tumor originating from the paravertebral sympathetic chain. Ganglioneuromas usually occur in older children and adolescents; neuroblastoma and ganglioneuroblastoma occur in the first decade of life. Thoracic neuroblastoma comprises about 15% of all cases of neuroblastoma. There is less dissemination than in the abdominal variant, with about 20% of cases presenting with metastatic disease. Additionally, there are other entities such as metastasis or the rare pulmonary or pleural sarcoma.

Pathology/Histology The thymus consists of lymphoid tissue containing dense accumulations of lymphocytes. It is surrounded by a thin, connective tissue capsula. Bronchial adenocarcinomas are histologically similar to salivary gland tumors. They can be of high- or lowgrade malignancy. Rarely, they metastasize to regional lymph nodes and their clinical history is usually one of

1243

respiratory infections, low bar consolidation, and a central or nodal mass. Ewing’s sarcoma of the chest and primitive neuroectodermal tumors are histologically different entities, but they both consist of malignant, small, round cells. Germ cell tumors can be malignant (10%) and can be associated with elevated levels of human chorionic gonadotropin or alpha-fetoprotein. Most often they are located in the anterior mediastinum. Germ cell tumors are asymptomatic in more than half of patients and they seldom occur before the second decade of life. Teratomas are the most common type, with varying degrees of differentiation. The immature types can be potentially malignant. If located in the thymus, it may be seen as an anterior mediastinal mass, sometimes with calcifications. Histiocytosis is characterized by granulomatous lesions with infiltration of large histiocytes and monocytes in varying degrees. ▶Langerhans’ cells are seen in the cytoplasm of histiocytes. Inflammatory pseudotumor is a benign tumor typically characterized by localized proliferation of mononuclear inflammatory cells, plasma cells, lymphocytes, and eosinophils. Pulmoblastoma is a mostly benign embryologic tumor that rarely turns malignant. Lymphoma varies with age and cell type. Histology determines the diagnosis of NHL, HD, and leukemia lymphoma complex. Neuroblastoma is derived from primitive ganglion cells. These may partially differentiate into cells having the appearance of immature neurons.

Presentation Thymoma can be seen in childhood in combination with myasthenia gravis. Children with bronchial adenocarcinoma present with wheezing, hemoptysis, or lobar collapse. Very seldom are they asymptomatic. Adenoid cystic carcinoma or cylindroma, a slow-growing lesion, may also present with respiratory symptoms and persistent atelectasis. The child with Ewing’s sarcoma of the chest or primitive neuroectodermal tumors presents in most cases with chest pain and/or a palpable mass and, often, pleural effusion. Patients with germ cell tumors may present with chest pain, hemoptysis, and possibly even a ruptured tumor. A minority of the patients is asymptomatic. If histiocytosis is located in the thymus, it may be seen as an anterior mediastinal mass, sometimes with calcifications. If the disease is located in the lungs, the child presents with coughing and/or dyspnea.

N

1244


Inflammatory pseudotumor: It is unclear whether there is a relationship with malignancy. In children without a known underlying malignancy, an ▶inflammatory pseudotumor is more likely than a malignant lesion. There is seldom a history of trauma or preceding infection, although some of these lesions begin as a radiologically designated “pneumonia.” Pulmoblastoma has a similar presentation. In lymphoma, clinical presentation varies with the extent of disease, particularly in its systemic manifestations. It can be silent or present with fatigue, swollen nodes, and malaise. The presentation of neuroblastoma depends on its anatomic site of origin. In the posterior mediastinum, neurogenic tumors make up the majority of the differential diagnosis in the abdomen. Fifty percent of children with intraspinal extension of a neuroblastoma are symptomatic when they present.

Imaging The normal thymus is seen in infancy as a low radioopaque structure that usually allows the pulmonary vessels to be seen through it. The so-called sail sign is due to a prominent thymus with straight borders. On conventional thorax images, the thymus becomes less noticeable after the age of 4 years; however, it is still easily seen on computed tomography (CT) and magnetic resonance (MR) images until after puberty (Fig. 1a and b). Under the age of 10 years, the thymus has an intermediate signal on T1-weighted images and is hyperintense on T2-weighted MR images. Ectopic thymus is in most cases positioned in the neck. Cervical thymus may contain cysts and cross-sectional imaging may be necessary, with a preference for MRI. It may show a mass effect on surrounding organs, contrary to the normal thymus. Thymolipoma is seen in children and young adults and is mostly discovered as an incidental finding. On CTand MR images, it is seen as a fatty mass containing solid, thymic tissue elements. The solid tissue elements enhance after administration of intravenous contrast medium. There is no invasion of adjacent structures. On CT images, bronchial adenocarcinomas are generally ovoid, soft tissue masses. Calcification is seen in about 50% of cases. Ewing’s sarcoma of the chest or primitive neuroectodermal tumor is, on CT, a solid heterogeneous mass with, especially in larger lesions, necrotic centers. In most cases there are no calcifications in the tumor (Fig. 2). There may be cortical rib destruction. On T2-weighted MR images the tumor is hyperintense and in most cases these lesions are of intermediate signal on T1-weighted images. Contrast enhancement is variable. Ultrasound (US) is

Neoplasms, Chest, Childhood. Figure 1 (a) Chest radiograph showing the thymus as a density that looks like a right, upper lobe consolidation. (b) CT of the thorax showing a normal homogeneous thymus with extension posterior to the superior vena cava (arrow).

Neoplasms, Chest, Childhood. Figure 2 Ewing’s sarcoma, extraosseous, situated in the posterior mediastinum. Large, solid lesion with a few low-attenuation lesions.

sometimes superior to CT and MRI in delineating the diaphragm. Most of the germ cell tumors can be diagnosed on CT (Fig. 3). Attenuation values are indicative of the germ cell layers and MR signal intensity may mirror this. They are usually well demarcated and have thick walls. Histiocytosis (Fig. 4): On a conventional thorax image, interstitial and nodular changes can be seen. Upper lobe


1245

Neoplasms, Chest, Childhood. Figure 3 Germinoma: a large lesion situated in the anterior mediastinum in a newborn child containing the three germ layers: fat (A), soft tissue (B), and bone (C).

Neoplasms, Chest, Childhood. Figure 5 (a) Chest radiograph of inflammatory pseudotumor with round mass in left lower lobe of the lung. (b) CT of the same inflammatory pseudotumor.

Neoplasms, Chest, Childhood. Figure 4 CT of histiocytosis of the lung. Bilateral increased interstitial markings with small nodules.

be visualized with US (Fig. 5b). They can resemble a coin lesion, possibly with calcification, an endobronchial mass, and a spicular mass with necrosis. There is varied enhancement following intravenous administration of contrast medium. Lymphadenopathy and plural effusion are rare. MRI is of no additional value.

Lymphoma involvement is at least equal to lower lobe involvement. CT scanning in an early stage of the disease may show a ground glass appearance, resulting in the typical honeycomb lung in the later phase. If the thymus is also involved, the organ is enlarged and shows cysts and calcifications on CT images. Mediastinal and hilar lymphadenopathy can also be present. The thoracic bone structures should also be evaluated for bony lesions on CT scanning. There is no role for MRI. Inflammatory pseudotumor/pulmoblastoma: A majority of the lesions occur in the lower lobe; they are approximately 3–4 cm in diameter and can about the pleura (Fig. 5a). CT is the major modality used in the evaluation of these lesions, and these masses are usually well circumscribed. If they reach the lung surface, they can

NHL: Typically, an anterior mediastinal mass is noted on chest radiographs (Fig. 6a). Pleural effusions are frequently noted. In addition, there may also be pericardial effusion. Airway compression may be noted, which might be important when subsequent CT imaging is contemplated. CT can be omitted as there is seldom any additional staging information. On the other hand, staging in the abdomen, and sometimes in the neck region, make CT the most efficient modality. HD: In contrast to NHL, HD is usually nodal (Fig. 6b). It is rare before 5 years of age and is characterized by the so-called Reed-Sternberg cell. The nodular sclerosing type accounts for 70% of all cases, with 20% having mixed cellularity. Painless adenopathy is the usual presentation; 30% of cases have hilar adenopathy. CT will demonstrate

N

1246


Neoplasms, Chest, Childhood. Figure 7 CT image of neuroblastoma. Arrow indicates calcification.

Neuroblastoma: Usually the chest radiograph shows a mass, possibly with calcifications, that may erode the posterior rib and vertebral body. This is then often verified by US; however, cross-sectional imaging is needed to detect nodal enlargement and intraspinal extension. Approximately 80% contain calcification on CT images (Fig. 7). Assessment of the enlargement of intervertebral neural foramina is best shown with MRI. To exclude or identify metastatic disease, sonography, nuclear medicine bone scan, or MIBG scintigraphy should also be performed.

Interventional Radiology There is no place for interventional radiologic treatment in these conditions; sometimes imaging my help by offering a safe, image-guided biopsy option or may enable uncomplicated drainage of secondary collections (usually using US, CT, or rarely fluoroscopy).

Bibliography

Neoplasms, Chest, Childhood. Figure 6 (a) NHL: chest radiograph of lesion situated in the anterior mediastinum, no effusion, normal ribs. (b) CT of Hodgkin’s lymphoma in the anterior mediastinum with necrotic areas (arrows). Large lymph nodes in left axillary tail. (c) Coronal thorax and abdomen image of a patient with Hodgkin’s lymphoma showing the mediastinal tumor (A), left axillary involvement (B), and pericardial effusion (C).

abnormalities in about 50% of patients with a normal chest radiograph. Pleural effusions as well as pericardial effusions occur frequently (Fig. 6c). Calcification is seldom present before treatment has been initiated.

1. Kim OH, Kim WS, Kim MJ et al (2000) Ultrasound in the diagnosis of pediatric chest disease. Radiographics 20:653–671 2. Curtis JM, Lacey D, Smyth R et al (1998) Endobronchial tumours in childhood. Eur J Radiol 29(1):11–20 3. Maziak DE, Todd TR, Keshavjee SH et al (1996) Adenoid cystic carcinoma of the airway: thirty-two-year experience. J Thorac Cardiovasc Surg 112(6):1522–1531; discussion 1531–1532 4. Dick EA, McHugh K, Kimber C et al (2000) Radiology of noncentral nervous system primitive neuroectodermal tumors: diagnostic feature and correlation with outcome. Clinical Radiology 56:206–215 5. Dulmet EM, Macchiarini P, Suc B et al (1993) Germ cell tumors of the mediastinum. A 30-year experience. Cancer 72(6):1894–1901 6. Smets A, Mortele K, de Praeter G et al (1997) Pulmonary and mediastinal lesions in children with Langerhans cell histiocytosis. Pediatr Radiol 27(11):873–876 7. Agrons, Rosado-de-Christenson ML, Kirejczyk WM et al (1998) Pulmonary inflammatory pseudotumor: radiologic features. Radiology 206(2):511–518

Neoplasms, Gallbladder

8. Hedlund GL, Navoy JF, Galliani CA et al (1999) Aggressive manifestations of inflammatory pulmonary pseudotumor in children. Pediatr Radiol 29(2):112–116 9. White KS (2001) Thoracic imaging of pediatric lymphomas. J Thoracic Imaging 16(4):224–237. Review 10. Thoracic neuroblastoma: a Pediatric Oncology Group study. GA, Adams SJ, Shochat Smith EI (1993) 28(3):372–377; discussion 377–378

1247

broad base or pedunculated with a well-defined stalk. Tubular adenomas present a typically lobular contour, whereas papillary types may have a cauliflower-like appearance. The lesion size is variable usually 2 cm). Any sizable or unusual polyp must be evaluated with endoscopy and biopsy.

Submucosal Tumours Benign submucosal tumours include lipomas, haemangiomas, neural tumours and small ▶gastrointestinal stromal tumours (GISTs). These tumours may be demonstrated by contrast radiology and have no discriminating features. A fatty mass on CT will represent a lipoma but otherwise evaluation of submucosal lesions requires EUS to establish the exact layer of origin within the bowel wall. By assessing position and morphology, EUS can indicate the probable aetiology (2) as well as categorising lesions as highly likely to be benign and safe to ignore, indeterminate and worthy of follow-up, or of concern and requiring biopsy or resection.

Malignant Tumours Gastric Cancer Gastric cancer (3) is the second commonest cancer worldwide with adenocarcinoma, accounting for more than 90% of all of malignant tumours of the stomach. Gastric cancer presents with a wide range of symptoms including dyspepsia, anorexia and weight loss. Diagnosis is usually made by upper gastrointestinal endoscopy and biopsy. On barium, certain features suggest malignancy such as thickened irregular folds and shallow eccentric ulceration but it is essential that any suspicion is confirmed with endoscopy and biopsy. Prognosis is related to tumour stage and accuracy is essential to plan the most appropriate management, predict prognosis, and reduce unnecessary surgical interventions. Tumour staging is based upon the widely accepted tumour-node-metastasis (TNM) classification: Primary tumour (T) T1: Tumour invades lamina propria or submucosa T2: Tumour invades the muscularis propria or the subserosa

1251

T2a: Tumour invades muscularis propria T2b: Tumour invades subserosa T3: Tumour penetrates the serosa (visceral peritoneum) without invading adjacent structures T4: Tumour invades adjacent structures (e.g. spleen, pancreas, transverse colon, etc.) CT has clear advantages when evaluating spread of disease outside the stomach but despite optimised multidetector techniques is inferior to EUS when assessing the layers of the gastric wall. EUS is the only imaging technique that can reliably discriminate the component layers allowing accurate prediction of T stage (Fig. 1a–d). The omental reflections around the stomach are not clearly seen with EUS, making it impossible to determine whether tumour has penetrated through the muscularis propria into the greater or lesser omenta to breach the visceral peritoneum, i.e. T2b or T3. Multiplanar and 3D techniques are considered a routine component of any multidetector CT assessment of gastroduodenal pathology and its potential spread. Coronal reformats are particularly suited for the study of gastric pathologies, both from improved staging accuracy and the improved overview afforded to the surgeons and medical oncologists. 3D-rendering techniques (virtual gastroscopy) have been advocated for gastric fold assessment. Direct invasion into adjacent structures (T4) can be difficult to establish on both CT and EUS. CT relies upon tumour contact with the contiguous organ and loss of intervening fat plane. EUS can exclude direct invasion if free movement is shown between tumour and adjacent organ. Regional lymph nodes (N) N0: N1: N2: N3:

No regional lymph node metastasis Metastasis in 1 to 6 regional lymph nodes Metastasis in 7 to 15 regional lymph nodes Metastasis in more than 15 regional lymph nodes

The regional lymph nodes are the perigastric nodes along the lesser and greater curvatures, and the nodes located along the left gastric, common hepatic, splenic and coeliac arteries. Involvement of other intra-abdominal lymph nodes, such as the hepatoduodenal, retropancreatic, mesenteric, and para-aortic, is classified as distant metastasis (M1). The CT assessment of nodal involvement relies almost entirely on size with upper limits of normal chosen to achieve a best compromise between sensitivity and specificity. Currently >6 mm maximum diameter short axis is advocated for regional nodes, rising to >8 mm for non-regional nodes. Certain morphological features such as nodal enhancement or a rounded margin suggest infiltration whereas clusters of smaller nodes should raise concern. Reported CT accuracy rates are no better than 50–70%. Lymph nodes are well seen on EUS and several morphological features correlate well with malignant

N

1252

Neoplasms, Gastroduodenal

Neoplasms, Gastroduodenal. Figure 1 Gastric cancer. 1a: EUS image of a T1 gastric cancer (*) demonstrating preservation of a normal layer of submucosa (sm) deep to the tumour, mp = muscularis propria. 1b: EUS demonstrating small volume ascites (f), thick-walled stomach (st) due to tumour infiltration, L = liver, gb = gallbladder. 1c: CT of an advanced gastric cancer (T) extending from the posterior wall of stomach (S) to liver (L) and abut the diaphragmatic crura (cr). Encasement of the coeliac axis (ca) and a retrocrural node (n) also noted. 1d: Same patient as 1c demonstrating further non-regional lymphadenopathy (n) and peritoneal tumour deposits (P).

infiltration. Hypoechoic nodes greater than 1 cm with well-defined rounded margins are likely to be malignant. Distant metastasis (M) M0: No distant metastasis M1: Distant metastasis CT provides an excellent whole body survey for the common sites of metastatic disease namely lung, liver, adrenals, ovaries, peritoneum and bone. EUS can detect low volume ascites, which may be the first indication of peritoneal spread. Magnetic resonance imaging (MRI) and positron emission tomography (PET) are not part of routine staging but have roles to play in problem-solving equivocal CT or EUS findings. PET may be helpful in detecting distant metastases and reports suggest that PET is more sensitive than CT for the detection of peritoneal disease. However, some gastric tumours such as mucinous or signet ring types typically take up fluorodeoxyglucose (FDG) poorly. For recurrent disease, PET may be useful when CT is negative or equivocal.

Lymphoma The commonest site of extranodal non-Hodgkin’s lymphoma is the stomach (approximately 25%) but lymphoma only accounts for 2–8% of all gastric malignancies. Morphology ranges from ulcers, nodules and polyps to fold enlargement and diffuse infiltration. The CT appearances of lymphoma and carcinoma can be similar. Lymphoma is often advanced at presentation with an average tumour size of 10 cm. Mural thickening is characteristically more marked and lymph nodes are larger. Mucosa-associated lymphoid tissue (MALT) lymphoma is most frequently found in the gastrointestinal tract, particularly the stomach. Gastric MALT lymphoma may regress completely after Helicobacter pylori eradication and EUS has been shown to be more sensitive than endoscopy and CT for staging and predicting tumour response. EUS can differentiate superficial from infiltrative variants which are of prognostic significance and can confirm remission or persistence of the disease during follow-up (4). Continuing gastric wall thickening, even


with negative histology, requires repeated biopsies as persistent disease is likely.

Metastases Metastases to the stomach are uncommon being found at post-mortem in only 2% of patients dying from malignancy. They may present as focal lesions (e.g. melanoma, lung, Kaposi’s sarcoma), or as diffuse linitis type infiltration (e.g. breast). Direct or transperitoneal spread from other abdomino–pelvic tumours can occur.

Gastrointestinal Stromal Tumours Most GISTs arise in the stomach (60–70%). About 10–30% of all GISTs are malignant and the risk of malignancy increases with size (>5 cm). Liver and the peritoneal cavity are the commonest sites for metastatic disease but lymph node and bony metastases are rare. Gastrointestinal bleeding is the commonest symptom but most small gastric GISTs are asymptomatic and detected incidentally. Barium studies show features of a submucosal mass—a smooth welldefined filling defect with either intact mucosa or central

1253

umbilication. Smaller GISTs are typically intramural and EUS will show them usually arising from the fourth hypoechoic layer, the muscularis propria (Fig. 2a). The vast majority have very low malignant potential but EUS can predict malignant change by identifying irregular margins, cystic spaces, and malignant-looking lymph nodes (5). With increasing size, a more exophytic morphology is seen and patients typically present with abdominal pain or a palpable mass. The CT features of larger GISTs are variable and range from uniformly solid to marked heterogeneity with solid and frankly cystic change mimicking lymphoma or sarcoma (Fig. 2b and 2c). GISTs can appear aggressive with involvement of multiple adjacent organs but in contrast to adenocarcinoma, they tend to abut and displace rather than infiltrate. Peritoneal spread can be extensive. After imatinib, both the primary mass and the peritoneal changes can display a rapid reduction in size and density with cystic changes mimicking a pseudomyxoma appearance (Fig. 2d). As most GISTs and their metastases take up 18FDG, a preoperative PET study is highly sensitive in demonstrating metastatic disease not visible on other imaging. PET can also be used to assess response to non-surgical treatment.

N

Neoplasms, Gastroduodenal. Figure 2 Gastrointestinal stromal tumours (GIST). 2a: EUS of a small gastric GIST (12 mm between +) arising within the muscularis propria (mp). 2b: CT demonstrating a large GIST (g) arising from the posterior wall of the stomach (S). 2c: CT of a gastric GIST with both solid (S) and cystic components (C). 2d: GIST following imatinib therapy. Low density peritoneal tumour mimicking pseudomyoma peritonei (*) along with a persisting solid component (S).

1254


Carcinoid

Malignant Tumour

Approximately 30% of gastrointestinal carcinoid tumours arise in stomach. The lesions can be small and submucosal, large and ulcerated, focal or multicentric. Benign and malignant variants occur. Small carcinoid tumours are best demonstrated by EUS or on CT during the arterial phase of contrast enhancement.

The majority of small bowel carcinomas arise in the duodenum or proximal jejunum. They are often advanced at presentation and symptoms include anaemia due to chronic bleeding or vomiting due to obstruction. Duodenal carcinomas are rare but show an increased incidence in patients with familial adenomatous polyposis and long-standing coeliac disease. Early tumours can be polypoidal and may be detected by barium or endoscopy. More advanced tumours show shouldered irregular stricturing. Biopsy of any suspicious lesion is essential to detect malignant change and CT is required for tumour staging.

Duodenum Tumours arising in the duodenum include carcinoma, GIST, neuroendocrine pathologies such as carcinoid and glucagonoma, schwannomas and paragangliomas. They account for 20% of all small bowel neoplasms.

Benign Tumour

Lymphoma Secondary involvement of the small bowel is more common than primary disease with approximately 40% of gastric lymphoma showing duodenal extension.

Adenoma Duodenal polyps are less common than gastric and are usually hyperplastic or adenomatous. Adenomas are the commonest symptomatic small bowel tumour, most frequently encountered in the duodenum, and present with bleeding, obstruction, or intussusception. If multiple polyps are seen, a polyposis syndrome should be considered when 60–70% of symptomatic lesions will be malignant.

Brunner’s Gland Hamartomas These are rare tumours of the duodenum, usually benign with only rare incidences of malignancy. Smaller lesions present as incidental findings during barium examination or endoscopy whereas larger lesions present with bleeding or obstruction. EUS may help characterise the lesion but histology is crucial and endoscopic snare excision or surgery is required depending on the size.

Lipoma Can be polypoidal or submucosal.

Haemangioma These are frequently symptomatic with bleeding a common presentation. Imaging features include calcified intraluminal phleboliths and the demonstration of feeding vessels.

Carcinoid The majority of duodenal carcinoids are located in the proximal duodenum, can be single or multiple, intraluminal and polypoidal, or intramural. In contrast to jejunal and ileal carcinoids, duodenal carcinoid rarely produces symptoms of carcinoid syndrome. The majority are gastrin cell tumours with one-third of these leading to Zollinger–Ellison syndrome.

Neoplasms, Gastroduodenal. Figure 3 Duodenal tumours. 3a: CT of a patient with duodenal lymphoma—note the gross wall thickening (*), splenomegaly (sp) and para-aortic lymphadenopathy (n). 3b: CT demonstrating a duodenal stricture (*) due to metastases from a squamous cell lung cancer, d = duodenum.

Neoplasms, Kidney, Childhood

Imaging may demonstrate nodular wall thickening of various degrees or an eccentric mass (Fig. 3a).

Secondary Spread to Duodenum Metastases to small bowel are found more frequently than primary neoplasms with lung and melanoma among the commoner primaries (Fig. 3b). The duodenum may be directly invaded by tumours originating in pancreas, gall bladder, bile ducts or colon, or from other tumours of the retroperitoneum such as sarcomas. High resolution curved planar reformats provided by ▶multidetector row CT can provide useful diagnostic information with respect to adjacent organ involvement. Ampullary tumours are discussed elsewhere.

Bibliography 1.

2.

3. 4.

5.

Ba-Ssalamah A, Prokop M, Uffmann M et al (2003) Dedicated multidetector CT of the stomach: spectrum of diseases. Radiographics 23:625–644 Kawamoto K, Yamada Y, Utsunomiya T et al (1997) Gastrointestinal submucosal tumours: evaluation with endoscopic ultrasound. Radiology 205:733–740 Kumano S, Murakami T, Kim T et al (2005) T Staging of gastric cancer: role of multi-detector row CT. Radiology 237:961–966 Le´vy M, Hammel P, Lamarque D et al (1997) Endoscopic ultrasonography for the initial staging and follow-up in patients with low-grade gastric lymphoma of mucosa-associated lymphoid tissue treated medically. Gastrointest Endosc 46:328–333 Palazzo L, Landi B, Cellier C et al (2000) Endosonographic features predictive of benign and malignant gastrointestinal stromal cell tumours. Gut 46:88–92

Neoplasms, Kidney, Childhood J OHAN G. B LICKMAN 1 , D EWI A SHI 2

1255

Predisposing anomalies may consist of ▶aniridia, ▶hemihypertrophy, ▶Beckwith–Wiedemann syndrome, and Drash syndrome as well as congenital genitourinary anomalies (horseshoe kidney).

Incidence 1. 2. 3. 4.

The peak incidence is between 1 and 3 years of age. The female:male ratio is equal. It is rare in the neonatal period. The incidence of bilateral tumor is up to 10%.

Pathology/Histopathology The classical triphasic Wilms’ tumor consists of blastemic, stromal, and epithelial elements. Tumors with favorable histology do not contain any anaplastic changes. In the National Wilms’ Tumor Study Group 5 (NWTS 5), the tumor stage is determined operatively in five stages. The 4-year survival rate is 94% for those patients whose most advanced lesion is stage I or stage II, and 76% for those whose most advanced lesion is stage III. In Europe, the staging system is based on the radiologic findings.

Clinical Presentation 1. Abdominal mass, occasionally with pain 2. Hematuria 3. Nonspecific sign such as fever, or hypertension

Imaging

1

Department of Radiology, Nijmegen, The Netherlands Department of Radiology, Faculty of Medicine, University of Indonesia, Cipto Mangunkusumo General Hospital, Jakarta Pusat, Indonesia [email protected]

2

Synonym Nephroblastoma

Definition Malignant tumor of the kidney usually arises from ▶nephroblasts. It is the most common genitourinary malignancy in children (95%).

Ultrasonography (US) reveals an intrarenal mass of heterogeneous echogenicity and may contain cystic components, either portions of obstructed and entrapped pyelocaliceal systems or hemorrhagic and necrotic tumor (Fig. 1). Extensions into the inferior vena cava and right atrium are characteristic routes of tumor growth and can be well visualized with sonography, computed tomography (CT), and magnetic resonance imaging (MRI). The tumor typically forms a pseudocapsule, but may invade the renal capsule spreading into the retroperitoneal space, or may grow directly into mesentery and omentum. On CT, Wilms’ tumor is generally spherical and intrarenal and contains small amounts of fat or fine calcifications. Calcifications are seen in about 9% of Wilms’ tumor. After contrast medium injection, the tumor shows heterogeneous enhancement and less enhancement than the

N

1256

Neoplasms, Kidney, Childhood

Stage I (43% of patients) II (23% of patients)

III (23% of patients)

Description Tumor is limited to the kidney and is completely resected The tumor is completely resected, but there is regional extension of the tumor (penetration of the renal sinus capsule, or extensive invasion of the soft tissue of the renal sinus). The blood vessels within the nephrectomy specimen outside the renal parenchyma, including those of the renal sinus, contain tumor Residual nonhematogenous tumor is present after surgery, and is confined to the abdomen

Neoplasms, Kidney, Childhood. Figure 1 Wilms’ tumor: US shows intrarenal mass with pelviocaliceal distortion.

(continued)

renal parenchyma. Low attenuation representing necrosis, cystic degeneration, or hemorrhage is seen (Fig. 2). On MRI, the tumor is intense on T1-weighted images (WI) and hyperintense on T2-WI. After contrast administration, Wilms’ tumor is hypointense in relation to normal renal parenchyma and may define the borders more clearly.

Nuclear Medicine Depicts bone metastasis.

Diagnosis Most often, Wilms’ tumor presents as an abdominal mass, and sometimes there is hematuria. Imaging findings can differentiate Wilms’ tumor from neuroblastoma. Rarely, Wilms’ tumor will invade surrounding tissues and even more rarely extend into the spinal canal, this being much more frequent with neuroblastoma. Clear cell sarcoma and rhabdoid tumor of the kidney are more aggressive than Wilms’ tumor, but they cannot be differentiated on imaging. A definite diagnosis of Wilms’ tumor is based on histopathology.

Neoplasms, Kidney, Childhood. Figure 2 CT shows a mass originating from the right kidney with distortion of the parenchyma.

2. Clear cell sarcoma of the kidney: This tumor is not a Wilms’ tumor variant, but it is an important primary renal tumor associated with a significantly higher rate of relapse and death than Wilms’ tumor of favorable histology findings. 3. Rhabdoid tumor of the kidney: This tumor tends to metastasize not only to the lungs, but also to the brain (10–15%). There is a distinct clinical presentation with fever, hematuria, and young age (mean 11 months) (see Table 1, Neuroblastoma).

Bibliography 1. 2.

Differential Diagnosis

3.

1. Neuroblastoma: The imaging hallmark is distortion of the renal parenchyma in Wilms’ tumor and renal displacement by neuroblastoma.

4.

Barnes PD (1998) Neuroimaging. In: Hans Blickman (ed) Pediatric Radiology: The Requisites. 2nd edn. Mosby St. Louis, pp 294–295 Carty H (2001) Paediatric Imaging. In: The Encyclopedia of Medical Imaging, Vol 7. Published by The NICER Institute, Oslo, pp 565–566 Feintein KA (2004) Renal neoplasms. In: Kuhn JP, Slovis TL, Haller JO (eds) Caffey’s Pediatric Diagnostic Imaging. 10th edn. Mosby, Philadelphia, pp 1787–1790 http://www.emedicine.com/med. Wilms’ Tumor. Last update June 2, 2005

Neoplasm, Large Bowel, Malignant

Neoplasm, Large Bowel, Malignant A NDREA L AGHI , F RANCO I AFRATE Director of Unit of CT and MR, University of Rome “La Sapienza”, Polo Pontino, I.C.O.T. Latina, Italy [email protected] [email protected]

Neoplasm, Large Bowel, Malignant. Table 1 for colorectal cancer

1257

Risk factors

Personal history of adenomas or colorectal cancer First-degree relative age 10 mm in the remainder). Reported N staging sensitivity is 79%, and specificity is 84% (2). Distant metastases from oesophageal cancer at presentation are seen in the liver in 35% and lungs in 20%. Ascites, pleural effusions or nodules in the omentum or pleura are suspicious for metastases and should be further investigated, e.g. laparoscopy. Indeed in AC, laparoscopy when performed routinely identifies metastases in up to 10%. Non-regional lymph nodes are considered to be M1 disease, particularly coeliac and cervical nodes.

Transcutaneous Ultrasound

EUS

Nuclear Medicine

EUS performed at 5–12 MHz depicts the gastrointestinal wall as a five-layered structure of alternating hyper and hypoechoic bands, which correlates well with histology.

The preceding staging strategy assumes FDG-PET is unavailable, but it is increasingly used for staging

Some centres, particularly in the Far East, use neck US to exclude lymph node metastases in SCC. US may help characterise liver lesions, and guide biopsy.

Magnetic Resonance Imaging Magnetic resonance imaging (MRI) has a limited role in the assessment of oesophageal neoplasms. Endoluminal MRI offers no advantage over EUS. MRI is occasionally used to clarify the nature of liver lesions.

N

1276

Neoplasms Oesophagus

Neoplasms Oesophagus. Figure 2 EUS of various stages of OC in three different patients. (a) EUS of the distal oesophagus (12 MHz) with mucosal thickening (between calipers). The hyperechoic sub-mucosa and hypoechoic MP deep to this are intact. AC confined to the mucosa was confirmed on resection. (b) The MP is absent between 5 and 8 o’clock but the tumour is separate from aorta at 6 o’clock. Malignant looking lymph node at 9 o’clock. T3 N1 confirmed at resection. (c) EUS of T4 OC invading the aortic wall at 6 o’clock. Treated with a stent.

Neoplasms Oesophagus. Figure 3 PET CT demonstrating supra-clavicular node rendering this patient inoperable.

18

FDG uptake in a junctional oesophageal tumour and a

Neoplasms, Oral Cavity

oesophageal cancer. In the detection of hepatic metastases in gastrointestinal tumours in general, FDG-PET is more sensitive (90%) than CT (72%) or MRI (76%). In a metaanalysis of the staging performance of FDG-PET in oesophageal cancer only, sensitivity and specificity was 67% and 97% for M1 disease and 51% and 84% for N1 disease (5). There is therefore an argument for performing FDG-PET first to exclude M1 disease, but this is limited by cost and availability. While 18FDG accumulates in over 92% of primary tumours, only limited Tstage information is obtained. If surgery is to be denied on the basis of M1a disease on FDG-PET, then ideally histological confirmation should be obtained as false positives do occur.

Diagnosis Histological diagnosis of oesophageal cancer is made by endoscopic mucosal biopsies.

Interventional Radiological Treatment Self expanding, metal mesh oesophageal stents are used to palliate dysphagia in patients with inoperable oesophageal cancer. Other indications include recurrence, tracheooesophageal fistulae, anastamotic leaks and perforation. Placement can be performed with endoscopic or fluoroscopic control or in combination. Recent technical advances include anti-reflux, anti-migration and retrievable devices. Most patients (75–90%) resume a near normal diet after stent placement, but further intervention is required in up to 60%. Survival may be increased by the use of concurrent chemoradiotherapy or endoluminal brachytherapy, although this may lead to more stent-related complications.

Bibliography 1.

2.

3.

4.

5.

Shaheen N, Ransohoff DF (2002) Gastroesophageal reflux, Barrett oesophagus, and oesophageal cancer: clinical applications. J Am Med Assoc 287:1982–1986 Weaver SR, Blackshaw GR, Lewis WG et al (2004) Comparison of special interest computed tomography, endosonography and histopathological stage of oesophageal cancer. Clin Radiol 59:499–504 Nishimura Y, Osugi H, Inoue K et al (2002) Bronchoscopic ultrasonography in the diagnosis of tracheobronchial invasion of esophageal cancer. J Ultrasound Med 21:49–58 Harris KM, Kelly S, Berry E et al (1998) Systematic review of endoscopic ultrasound in gastro-oesophageal cancer. Health Technol Assess 2(18):I–iv, 1–134 van Westreenen HL, Westerterp M, Bossuyt PMM, et al (2004). Systematic review of the staging performance of 18F-fluorodeoxyglucose positron emission tomography in esophageal cancer. J Clin Oncol 22:3805–3812

1277

Neoplasms, Oral Cavity S ABRINA KO¨ SLING Martin-Luther-Universita¨t Halle-Wittenberg, Klinik fu¨r Diagnostische Radiologie, Halle, Germany [email protected]

Definition Neoplasms are characterised by an abnormal growth of tissue, which occur within structures of the oral cavity. Malignant neoplasms show a greater degree of anaplasia and have the properties of invasion and metastasis. Neoplasms of the oral cavity are soft tissue tumours arising from the mucosal surfaces or deeper structures. Typical benign tumours of the oral cavity are papillomas and fibromas. Other benign soft tissue neoplasms as fibromatosis, lipoma, leiomyoma, rhabdomyoma or neurogenic tumours (granular cell tumour, schwannoma, neurofibroma) are less frequent at this site. Haemangiomas and lymphangiomas as congenital vascular lesions are presented in another chapter. Cysts and dermoids are developmental cystic degenerations and not true neoplasms. The known salivary gland neoplasms may arise from minor salivary glands in the oral cavity. They are discussed in the salivary gland chapter. Compared to other lesions, malignant lesions are found seldom (7%) in the oral cavity; 90% of them are squamous cell carcinomas (1). Adenocarcinomas may occur; melanomas and sarcomas are exceedingly rare.

Pathology/Histopathology Papillomas are characterised by a growth of keratinised or non-keratinised squamous epithelium with normal maturation pattern. They contain fibrovascular connective tissue and chronic inflammatory cells. In fibromas— hyperplasias of fibrous connective tissue—there is a proliferation of fibroblasts and collagen fibres. Several patterns that reflect progressive stages of fibroblastic differentiation are identified microscopically in fibromatosis. Dense or moderately dense, rather vascular, bland collagenic connective tissue with scattered chronic inflammatory cells can be observed. In lipomas, superficial or deeper adipose tissue proliferates; in leiomyomas smooth muscle—in the oral cavity usually of the blood vessels—and in rhabdomyomas skeletal muscle. The submucosal granular cell tumours are composed of diffuse sheets of large cells of either nerve or muscle origin with a cytoplasm of densely packed eosinophilic granules. In schwannomas (synonym:

N

1278


neurilemoma) there is a fibroblastic proliferation of the nerve sheath cells (Schwann’s cell). Neurofibromas consist of diffuse proliferations of perineural fibroblasts that are orientated in either a random or nodular pattern (2). Squamous cell carcinomas develop from the stratified squamous epithelium of the mucosa. Histologically, they show large polygonal or fusiform cells with atypical nuclei and increased mitotic activity. They may contain keratinised and inflammatory cells. Depending on the differentiation four degree (G1–G4) are distinguished: high, moderate, poor, undifferentiated or anaplastic. Adenocarcinomas (G1–G4) are epithelial tumours with glandular components, which may produce mucus. Carcinomas of the oral cavity are classified according to the ▶TNM criteria of ▶UICC (Tables 1 and 2). At first presentation about 60% of the patients with carcinomas of the oral cavity have nodal metastases, even if the tumours are small. The first order drainage of oral cavity carcinomas is the submental and submandibular lymph node group for anterior processes, and the jugulodigastric node for posterior lesions. From there the drainage of both regions goes mostly down into the deep cervical and spinal accessory chain. Especially carcinomas of the tongue are predisposed to bilateral affection. Melanomas are malignant neoplasms deriving from cells that are capable of forming melanin, which may occur in the mucosa of the oral cavity. The usually high malignant sarcomas are formed by the proliferation of mesodermal cells.

Clinical Presentation Neoplasms are usually painless. Benign tumours, but also adenoid cystic carcinomas grow slowly—malignant more Neoplasms, Oral Cavity. Table 1

rapidly. Due to slow growing, benign tumours are often asymptomatic and will be detected incidentally. A criterion of benignancy is that the lesion is movable against the undersurface. Depending on the localisation and size, neoplasms can cause dysphagia, dyspnoea and difficulties in speaking; if the tumour is injured bleedings might occur. Malignant epithelial tumours are often accompanied by foetor ex ore. Some tumours have a characteristic optical appearance. Papillomas are white or pink, sessile or pedunculated exophytic nodules of cauliflower-like appearance; they may occur anywhere on the oral mucosa. The firm and often exophytic fibromas, located on the buccal mucosa, tongue, lips or gingiva, are usually smaller than 2 cm. Lipomas are soft on palpation and of yellow colour. A nodular, possibly exophytic, ulcerous, hard, easily bleeding mass, which is more or less fixed and sometimes painful on palpation, is suspicious to be a carcinoma; otalgia may occur due to connection of the lingual to the facial nerve. Often enlarged lymph nodes can be palpated. Clinical signs of infiltration are a fixed tongue and trismus. Mesenchymal and neurogenic neoplasms lie submucosally without typical clinical signs.

Imaging Conventional radiographs are performed in tumour diagnostics of the dental apparatus. They do not play a role in tumour imaging of the oral cavity. CT and MRI have been proved to be valuable in the modern diagnostics of neoplasms in this region. Ultrasound is of secondary importance (4).

T-categories of oral cavity tumours (UICC)

T0: No evidence of tumour TIS: Carcinoma in situ T1: Tumour ≤20 mm T2: Tumour >20 mm, ≤40 mm T3: Tumour >40 mm T4a: Infiltration through cortical bone, in external tongue muscles, maxillary sinus or skin T4b: Infiltration of the masticator space, pterygoid process or skull base, or encasement of internal carotid artery Neoplasms, Oral Cavity. Table 2

N-Categories of oral cavity tumours (UICC)

NX: Regional lymph nodes are not assessable N0: No regional lymph node metastases N1: Metastasis(es) in solitary ipsilateral lymph node—≤30 mm N2a: Metastasis(es) in solitary ipsilateral lymph node—>30 mm, but ≤60 mm N2b: Metastasis(es) in multiple ipsilateral lymph nodes—none of them >60 mm in its largest dimension N2c: Metastasis(es) in bilateral or contralateral lymph nodes—none of them >60 mm in its largest dimension N3: Metastasis(es) in lymph nodes >60 mm in its largest dimension


Computed Tomography and Magnetic Resonance Imaging Neoplasms of the oral cavity are detected by clinical investigation. Imaging is performed to depict the full extension in larger benign tumours and especially in malignancies, first of all in carcinomas. Very small tumours with mucosal spread can be undetectable on CT and MRI. Due to the better soft tissue contrast, MRI is more suited in the delineation of tumourous lesions from surrounding structures compared to CT. Beam-hardening artefacts caused by dense bones and teeth, or by dental fillings may seriously disturb the visibility of tumours on CT. On the other side, a CT examination is performed within a few seconds with modern technique whereas a complete MRI study lasts about 30 min. In the case of less co-operative patients CT should be preferred. The general signs of tumours on CT are similar to those on MRI. Tumours show a space-occupying growth, larger ones an obliteration of fat planes. An infiltration of muscle and bone are hints for a malignancy. Cortical erosion is seen better on CT (Fig. 1a), the extent of medullary invasion better on MRI (Fig. 1b). Many tumours show a ▶contrast enhancement, which is different to surrounding structures. Larger tumours have sometimes inhomogeneities. On MRI, most tumours appear isointense to muscles on unenhanced T1-weighted images, and moderate to strong hyperintense on T2-weighted images. Except of lipomas and cystic lesions, there are no characteristic densities or signal intensities for a single neoplasm on CT or MRI (Fig. 2). In general, cancer has several modes of spread. Firstly, most tumours grow superficially within the mucosa.

1279

Mucosal spread is best evaluated by inspection. Before invasion of muscles and bones tumour spreads within spaces, along the muscular fasciae and the periosteum. Cancer may show a ▶perineural spread and lymph node metastases. Both have a negative impact on prognosis. The deep spread of a neoplasm cannot be exactly assessed by clinical examination; this is where imaging plays a role. In dependence of the primary site (floor of the mouth, tongue, oral mucosa with different sub-sites, upper and lower gingiva, hard palate), carcinomas show preferential ways of extension with increasing growth, which can be best visualised by MRI: Tongue carcinomas grow first along the intrinsic muscles deeper into the tongue, along extrinsic lingual muscles into the sublingual and submandibular space; they can reach the base of the tongue and the muscles of the floor of the mouth. Cancers in the floor of the mouth (Fig. 3) spread along the muscles, periosteum of the mandible, within the lingual septum and submandibular space; the tongue and tongue base may be involved; relative early the mandible is invaded; the Warton‘s duct can be occluded resulting in an adenitis of the sub-mandibular gland. Carcinomas of the oral mucosa and gingiva spread first into the buccal space; later they may invade the masticator space; a retromolar localisation is endangered for a perineural spread along the inferior alveolar nerve. Cancers of the hard palate break early through into the maxillary sinus, a perineural spread along the palatine nerves to the pterygopalatine ganglion may occur. In lymph node judgment, morphological criteria— size larger than 10 mm, central necrosis, indistinct nodal margin as a sign of extra-capsular spread—are used for

Neoplasms, Oral Cavity. Figure 1 T4-stage squamous cell carcinoma of the floor of the mouth in two different patients. (a) Axial high-resolution CT in bone window shows a cortical defect in the mandible (arrow). The extent within the bone marrow is not quite clear. (b) Axial T1-weighted image demonstrates a wide medullary invasion of the mandible (arrows).

N

1280


Neoplasms, Oral Cavity. Figure 2 Axial MRI of a patient in whom clinicians found an unclear mass in the tongue. (a) On T2-weighted image the lesion can be demonstrated best (arrow) and estimated in size. It is relatively sharp demarked. (b) On contrast-enhanced T1-weighted image, the tumour showed a described enhancement (arrows) and is less well demarcated. The imaging diagnosis was ‘probably benign neoplasm’. Histology revealed a regressively changed fibroma.

Neoplasms, Oral Cavity. Figure 3 Bioptically proven squamous cell carcinoma of the floor of the mouth. (a) Axial T2-weighted images visualise a hyperintense mass in the anterior part of the oral cavity, which crosses the midline (arrow). (b) On axial contrast-enhanced T1-weighted image, there is a blurred margin of the tumour (arrows), which is suspicious of infiltration of the genioglossal muscle as T4 criteria. The patient underwent radiochemotherapy.

the distinction between normal and metastatic lymph nodes in CT and MRI, but the number of false-negative and false-positive findings is still high with this approach. If enlarged lymph nodes are related to the primary disease and the corresponding lymph node drainage, the results can be improved. The major problem of both methods is the high false-negative number in normal sized lymph nodes with micro-metastases. Despite of technical

improvements in CT and MRI, this problem is not sufficiently solved at present. Imaging is part of the clinical staging. Tumour size in its greatest dimension can be determined for T1 to T3 categories and for the T4 category invasion of skeletal muscles and/or bone (Table 1). If a tumour is accompanied by inflammatory changes, imaging tends to overestimate the tumour size because both processes

Neoplasm, Oropharynx

cannot be differentiated exactly. Criteria for the single N categories are the number of affected lymph nodes, their size and if they are ipsilateral or bilateral (Table 2).

Ultrasound In the hand of experts, ultrasound examination can deliver a lot of information. It provides access to most structures of the oral cavity. Usually, tumours appear as hypoechoic space-demanding structures. Especially in the tongue, a tumour and its spread within the internal muscles can sometimes be better delineated than on CTor MRI. The method is of special relevance in the examination of major salivary glands and as duplexhigh-resolution-sonography in the assessment of lymph nodes, where malignant lymph nodes show irregular central vessels. If ultrasound is combined with fine-needle aspiration cytology, the specificity of lymph node findings can be increased significantly. A larger tumour extension into deeper spaces, bone invasion and retropharyngeal lymph node metastases cannot be diagnosed sufficiently. Therefore, sonography is not suitable as a solitary method in tumour imaging.

compared to imaging. Submucosal tumours cannot be assessed clinically with the same safety. Small, superficial and benign appearing neoplasms are excised without imaging. Imaging is needed in larger and/or malignant appearing neoplasms. Beside of histology, the knowledge about the full tumour extension is important for the therapeutic decision and patient’s prognosis: extension into deep soft tissue structures and spaces; infiltration of skeletal muscles, bone, neighbouring spaces and skin; crossing the midline of the tongue or floor of the mouth; perineural spread; relation to the common or internal carotid artery; and metastatic involvement of lymph nodes. The imaging method of first choice to get this information is MRI nowadays (5). In contraindications for MRI or non-compliant patients, CT is the alternative method. PET and sonography can provide relevant diagnostic information as complementary techniques.


Nuclear Medicine 4.

In the diagnostics of the oral cavity carcinomas, nuclear medicine techniques can provide relevant information in certain cases. Due to the lack of anatomic details, they have to be supplemented by CT or MRI in shape of coregistration or comparative investigation. Nowadays, ▶18FDG PET is the leading procedure, which gives information about the degree of glucose metabolism in different tissues. A certain level of increased uptake of FDG is considered as malignancy. Several reports confirmed an increased 18FDG uptake in primary tumours, lymph node and distant metastases of head and neck squamous cell carcinomas. However, the clinical use of 18FDG PET is discussed controversially. Higher accuracy compared to CT and MRI have been reported in lymph node assessment, detection of unknown sites in patients with cancer of unknown primary (CUP) and in differentiation of recurrent cancer from post-therapeutic changes. Nevertheless, a routine use of PET in pre-therapeutic staging of oral cavity squamous cell carcinomas cannot be recommended; in this situation PET is reserved for cases with equivocal findings by CT or MRI (5).

Diagnosis The diagnosis of oral cavity tumours is based on histopathology. The dignity of mucosal lesions and their superficial spread can be better judged by clinical methods

1281

5.

Parker SL, Tong T, Bolden S et al (1996) Cancer statistics. CA Cancer J Clin 46:5–27 http://www.dental.mu.edu/oralpath.htm Sobin LH, Wittekind Ch (eds) (2002) International Union Against Cancer. TNM Classification of Malignant Tumours. 6th edn. WileyLiss, Baltimore, MD Lenz M, Hermans R (1996) Imaging of the orapharynx and oral cavity. Eur Radiol 6:536–549 Dammann F, Horger M, Mueller-Berg M et al (2005) Rational diagnosis of squamous cell carcinoma of the head and neck region: comparative evaluation of CT, MRI, and 18FDG PET. Am J Roentgenol 185:1326–1331

Neoplasm, Oropharynx F RANCESCA G RANATA University of Messina, Messina, Italy [email protected]

Synonyms Malignant oropharyngeal tumours

Definitions Neoplasm of the oropharynx means a primary malignancy arising from the mucosa of this anatomic region. The oral portion of the upper aerodigestive tract is formed by two regions: the oral cavity and the oropharynx. The oropharynx communicates, in addition to the oral cavity, with nasopharynx, hypofarynx and larynx (1).

N

1282


The boundaries of this anatomical region is the soft palate superiorly and the hyoid bone inferiorly (2). The anterior part of oropharynx includes the posterior onethird of the tongue (posteriorly to the circumvallate papilla); the upper part is formed by the soft palate and uvula. Laterally, there are two facial arches, the anterior (palatoglossus muscle) and the posterior arch (palatopharyngeal muscle) and between them is the tonsillar fossa which harbours the palatine tonsil (1). The posterior oropharyngeal wall is bounded by upper pharyngeal constrictor muscles and is related to the second and third cervical vertebrae. Moreover, the oropharyngeal region includes the lingual tonsil which lies at the base of the tongue, more concentrated on the lateral surfaces. The lingual tonsil and the palatine tonsils jointly, as well as other collection of lymph tissue in this area, form the Waldeyer’s ring.

Pathology/Histopathology The majority of tumours involving the oropharynx are squamous cell carcinomas antigen (SCCA) which represent about 90% of the total. They arise from oropharyngeal mucosa. The incidence of SCCA of oropharynx increases in patients with a history of tobacco or alcohol abuse (1, 2). Other histological types of tumours are less common such as lymphoma, minor salivary gland tumours and other rare mesenchymal lesions (1, 2). The mesenchymal neoplasms of oropharynx are rare and include: rhabdomyoma and rhabdomyosarcoma (most common in childhood), granular cells tumours, fibromatoses, schwannomas, neurofibromas, and hemangiomas (2). In most cases, imaging techniques have little specificity and histologic diagnosis is needed.

Clinical Presentation Oropharyngeal tumours can be clinically silent. Sometimes, the symptomatology can be represented by dysphagia or slight pain even if latero-cervical lymphadenopathy is often the only finding.

Imaging The goal of modern diagnostic imaging is to establish the location, the size and the extent of oropharyngeal tumour. Moreover, it is important the pre-therapeutic evaluation of the relationship between tumoural mass and some critical structures like lingual artery and vein, carotid artery, internal jugular vein, muscles, connective tissue spaces and bone.

Computed tomography (CT) scan and magnetic resonance imaging (MRI) are the techniques of choice in the evaluation of oropharyngeal tumours.

Computed Tomography Scan Method The oropharynx should be studied during suspended respiration (2). The region of interest should be extended from the base of the skull to the upper mediastinum (aortic arch). Contrast-medium injection is mandatory to define the tumoural boundaries and metastatic lymphoadenopathies (2, 3). The axial plane is obligatory for the evaluation of oropharyngeal region. With modern multislice CT scanners, coronal and sagittal scans can be reconstructed from the volume data set (multiplanar reformation, MPR) or 3D images can be calculated with volume-rendering techniques. Coronal and sagittal images are useful to define the anatomical spread of the tumour and to reduce artefacts due to dental fillings (amalgam, gold or ceramics) (3). Bone algorithms are essential to evaluate any pathologic bone involvement.

Magnetic Resonance Imaging Method The advantages of MRI in comparison with CT scan are better tissue contrast and the absence of degradation due to dental fillings. The disadvantages are the long imaging time (20–30 min MRI versus 20–30 sec CT scan) and the subsequent difficulty for the patient to remain motionless (3). The MRI study of oropharynx is performed primarily in the axial plane. It is essential to obtain axial T1- and T2-weighted images. These images are usually obtained by FSE sequences. Scanning with contrast-medium (Gd-DTPA) is mandatory; the dosage is 0.2 mmol/kg body weight. Optional coronal and sagittal T1-weighted images before and after Gd-DTPA can be obtained, respectively to study lateral and midline tumours. To overcome the problems due to the high fat signal in FSE images, fat suppression techniques are used on both T2-weighted images and T1-weighted images after gadolinium injection (3). Alternative fat-suppressed technique is short-tau inversion recovery (STIR) without contrast-medium injection. This technique is very sensitive in tumour location. Moreover, dynamic T1-weighted GE sequences can be used after gadolinium injection to study tumour vascularisation because of their high temporal resolution (3). Disadvantages of GE sequences are represented by susceptibility artefacts at the air-bone-tissue borders and near dental fillings.


Diagnosis Imaging techniques play an essential role in the diagnosis and stadiation of oropharyngeal tumours. The location and the size of the neoplasm, the degree and direction of infiltration into neighbouring structures and the degree of regional lymph node involvement significantly orientate prognosis and therapy (2, 4). According to the TNM criteria, oropharyngeal tumours are classified: . . . .

Stage T1: 2 cm–4 cm Stage T4: Tumour involving bony structures (mandible or jaws), soft tissues of the neck or extrinsic muscles of the tongue.

Tumour Detection A malignant oropharyngeal neoplasm appears like a space-occupying growth associated with obliteration of fat planes. Direct sign of an advanced neoplasm is the infiltration of muscle and of bone structures (4). After contrast-medium injection, squamous cell carcinomas enhance in 85% of cases (4). The tumoural tissue has a density between 80 and 90 HU, different from the muscle (70 HU) and the fat tissue (100/200 HU). Moreover, the tumoural mass often shows structural inhomogeneities with intralesional necrosis. Some oropharyngeal tumours (15%) can be isodense to the muscle on CT scan. MRI presents an excellent tissue contrast which allows easily differentiation between tumour and normal structures (2–4).

1283

This peculiarity is useful in detection of small neoplasm in the oropharynx (4). In addition to general diagnostic signs, analogous to CT scan, MRI evaluation offers specific findings. In T1-weighted images, neoplastic tissue is significantly hypointense in comparison with fat tissue. Also, tumours are slightly hypontense or isointense than muscles. In T2-weighted images, tumours have high signal intensity. Peritumourous oedema may also produce high signal leading to over-estimation of tumour size. Therefore, MRI sensitivity can be improved using fat suppression techniques (FS-T2 FSE or STIR sequences). (Fig. 1). After gadolinium injection, all neoplasms show contrast-enhancement, better depicted using fat suppression techniques on T1-weighted sequences.

Spread Patterns and Lymphatic Drainage Oropharyngeal tumours present different spread patterns in relation to the site of origin (1, 2, 4). Carcinomas arising on the anterior tonsillar pillar (ATP) tend to spread along the palatoglossus muscle and its fascial attachments as far as to the tongue base. The neoplasm may spread superiorly to the soft palate and to nasopharynx. Nasopharyngeal infiltration is usually unilateral. The lymphatic drainage of ATP tumours is primarily to the submandibular and internal jugular nodes (2, 4). Isolated carcinomas of posterior tonsillar pillar (PTP) are rare and usually small. Carcinomas of the tonsil can spread anteriorly to involve the tongue base, the floor of the mouth and retromolar trigone; laterally, neoplasm can involve

Neoplasm, Oropharynx. Figure 1 SCCA of palatine tonsil. Axial FSE T2-weighted image. A small tonsillar tumour is visible on the left side (arrow). The neoplasm is substantially isointense to contiguous normal tissues. Axial fat suppression T2-weighted image. The fat saturation technique allows easier the tumour detection (arrow).

N

1284


parapharyngeal, masticator and carotid spaces; superiorly, the soft palate and the nasopharynx. (Fig. 2) Lymph node metastasis occurs primarily in the upper internal jugular or retropharyngeal nodes (2, 4). SCCA of tongue base may extend laterally to involve the mandible and medial pterygoid muscles; superiorly, to tonsillar fossae and soft palate; anteriorly to the oral tongue and floor of the mouth; inferiorly, to the vallecula, pre-epiglottic space, epiglottis or hypopharynx. (Fig. 3) Lymphatic drainage is often to internal jugular and submandibular nodes (2, 4). Tumours arising from the soft palate can spread in any direction. The tonsillar pillars and hard palate are usually involved first. Deeply, these neoplasms can involve

parapharyngeal space, the nasopharynx and the skull base. Palatal carcinomas drain first to upper internal jugular and subdigastric nodes, with subsequent involvement of lower internal jugular or retropharyngeal nodes. SCCA of the posterior oropharyngeal wall spread inferiorly into the hypopharynx and cranially into nasopharynx. They commonly spread submucosally invading the retropharyngeal fat. Lymphatic drainage is to submandibular and jugulo-digastric nodes (2, 4). For depiction of tumour spreading, sagittal and coronal images on both CT scan and MRI are well-suited. On MRI, T2-weighted FSE with fat suppression, FS T1-weighted FSE or T1-weighted GE after gadolinium injection are very sensitive in lymphadenopathy detection.

Neoplasm, Oropharynx. Figure 2 SCCA of palatine tonsil. (a, b) Axial T1 and T2-weighted images show a tonsillar tumour on the right side (arrows). The lesion is about 2 cm in diameter. (c, d) Axial and coronal Gd-DTPA fat suppression T1-weighted images. The tumoural tissue shows intense enhancement. Note the tumoural spreading to the tongue base, anteriorly (arrows), and to parapharyngeal space, laterally (arrowhead).

Neoplasms, Parathyroid

1285

Neoplasm, Oropharynx. Figure 3 SCCA of tongue base. Axial T2-weighted image. A neoplastic mass is visible on the left side of the tongue. Note the bulging of the free edge of the tongue (arrow) and the infiltration of the mylohyoid muscle (arrowhead). Coronal T1-weighted image shows ipsilateral lymphadenopathies along the course of the internal jugular chain (arrow).

Nuclear Medicine

Definitions

18-FDG positron emission tomography (PET) and PET/ CT scan fusion have useful role in the follow-up of patients with head and neck cancers to localise tumour remnants or relapses and to define radiotherapic planning of treatment.

Parathyroid glands are four small endocrine glands, exceptionally five or six, located behind the thyroid lobes; 80% of superior parathyroid glands are located at the cricothyroid junction, while inferior parathyroid glands are variably located (61% near the lower pole of the thyroid lobes and 26% in thyrothymic ligament). Incidence of ectopic parathyroid glands reported is variable (2 to 8%). Parathyroid glands regulate the calcium level through the secretion of the parathyroid hormone (PTH). Primary parathyroid gland carcinoma is extremely rare, with an estimated annual prevalence of about 0.005%, i.e. five cases of primary carcinoma/100,000 cases of primary hyperparathyroidism (HPT) or less than 1% of all cases of HPT. Parathyroid adenomas (PA), that are benign tumors, account for 85% of all cases of this disease. Most are functional and cause excessive secretion of PTH, resulting in HPT. In addition to the solitary or sporadic PA tumors, approximately 5% are associated with hereditary cancer syndromes such as multiple endocrine neoplasia (MEN).


Yates CB, Phillips CD (2001) Oral cavity and oropharynx. Curr Probl Diagn Radiol 30(2):38–59 Mukherji SK, Holliday RA (1996) Pharynx, chapter 8. In: Som PM, Curtin H (eds) Head and Neck Imaging. Mosby, St. Louis Lenz M, Grees H, Dobritz M (2000) Kersting-Sommerhoff B. Methods: MRT. Eur J Radiol 33:178–184 Lenz M, Grees H, Baum U, et al (2000) Oropharynx, oral cavity, floor of the mouth: CT and MRI. Eur J Radiol 33:203–215

Neoplasms, Parathyroid E MANUELE N ERI , S ALVATORE M AZZEO, VALENTINA B ATTAGLIA Diagnostic and Interventional Radiology, University of Pisa, Italy [email protected]

Synonyms Parathyroid adenoma; Parathyroid cancer

Pathology/Histopathology Parathyroid glands are composed of chief cells, oxyphilic cells and intermediate cells. Histologically, primary HPT can be attributed to a single adenoma in 80 to 85% of cases, hyperplasia in 15 to 20% of cases and carcinoma in 1% of cases. Double or multiple adenomas have been reported in patients with primary HPT including MEN.

N

1286


Hyperplasia is defined as an enlargement of more than two glands, whereas adenoma is traditionally considered to be a single gland disorder. The parathyroid adenoma consists of hypercellular parathyroid tissue arranged in solid sheets or cords with a trabecular pattern. Occasionally, interspersed adipocytes are seen with the adenoma. Nuclear pleomorphism may be present; however, mitoses, capsular invasion, or vascular invasion mitigate against a benign lesion. Parathyroid adenomas include the following subtypes: oncocytic adenoma, lipoadenoma, large clear cell adenoma, water-clear cell adenoma and atypical adenoma.

Clinical Presentation Most PTG tumors are functional and cause excessive secretion of PTH, resulting in HPT. HPT may be classified as a primary, secondary, or tertiary disorder. Primary HPT is caused by PA. Secondary HPT (hyperplasia) develops in response to chronic depression of serum calcium levels, generally due to renal impairment. In a minority of patients, this parathyroid hyperactivity becomes autonomous, resulting in tertiary HPT. Widespread testing of calcium in blood using automated analyzers in the clinical laboratory has led to the detection of hypercalcemia early in the course of HPT and in the majority of cases before other signs of HPT are apparent. Symptoms of HPT can be variably, aches and pains, depression, abdominal pain, nausea, vomiting, fatigue, excessive urination, confusion, muscle weakness.

Neoplasms, Parathyroid. Figure 1 lesions (arrows).

The excess calcium released by the bones leads to osteoporosis, osteomalacia and kidney stones, because of high levels of calcium excreted into the urine by the kidneys.

Imaging Diagnostic imaging of parathyroid is not essentially aimed to diagnose HPT, but to detect hyperplasia or neoplasms in the parathyroids or in ectopic glands. It is still controversial the use of imaging in preoperative evaluation; surgical series shows that the success rate of the intervention is excellent (95%) even if a preoperative imaging is missed. However, there is general agreement that imaging is required in cases with persistent or recurrent HPT after surgery. At present, ultrasound (US) and SPECT scanning with Tc-99m Sestamibi are the first imaging steps and the preferred methods to identify a pathologic parathyroid gland before surgery. US has a high sensitivity in detecting parathyroid adenomas in patients without thyroid diseases associated (90%) and lower in presence of thyreopaties (70%). Sensitivity reported of SPECT ranges between 70 and 100%. US can be useful mainly in localizing parathyroid cervical adenomas and in perithyroidal region, but it frequently fails in identifying ectopic lesions such as for parathyroid adenomas localized in the mediastinum and retro or paratracheoesophageal compartment (Fig. 1). US must be performed with high resolution linear transducers (at least 10MHz), and a critical factor influencing the accuracy of the technique is the experience of the examiner; this explains the variance of sensitivity among different authors. Therefore, the use of US can be suggested only if an experienced operator is performing the exam.

US: Typical hypoechoic pattern and shape of the lower (a) or upper (b, c) parathyroid


In case of thyroid goiter, the detection of parathyroid lesion can be more difficult. The use of color-Doppler US examination is frequently helpful to differentiate parathyroid from thyroid nodules or cervical lymphnodes on the basis of the different vascular pattern. SPECT is performed by injecting 20 to 25 mCu of technetium-99m sestamibi; images are obtained at 10–15 min then 2–3 h after the injection. Late phase is preferable for detecting parathyroid adenomas, as thyroid nodules clear uptake faster than do parathyroid neoplasms. Sestamibi is taken up by the high mitochondrial content of adenomatous parathyroid cells, greater than

1287

surrounding parenchyma. False positive cases can be caused by solid thyroid nodules (adenomas), Hurthle cell carcinoma, malignant thyroid lymph node metastases. It is proven that a combination of US and SPECT results in improved diagnostic accuracy before surgery in pre-operative diagnosis of parathyroid disease. MRI can be extremely valuable for localizing a parathyroid adenoma although the sestamibi has decreased the need for it dramatically. At best, an MRI will find less than 10% of diseased parathyroid glands; therefore the indications for getting this scan are very few. A renewed interest has received CT, with the advent of the multidetector technology. Multidetector CT allows to

N

Neoplasms, Parathyroid. Figure 2 (b) and coronal scan (c).

Multidetector CT: Retroesophageal parathyroid lesion (ectopic site). Axial (a), sagittal

Neoplasms, Parathyroid. Figure 3 Multidetector CT: The scan distinguishes hypervascular retrosternal parathyroid nodule (P) from hypodense lymph-node. Axial (a) and sagittal (b) scan.

1288

Neoplasms, Phyllodes, Breast

study at very high resolution and with optimal contrast enhancement the mediastinal area. Ectopic parathyroids can be detected as hypervascular nodules and well differentiate from lymph nodes (hypovascular) (Figs 2 and 3).

Diagnosis Diagnosis of HPT is based on two laboratory evidences, elevated serum Ca and elevated PTH (suppressed in PTH-rp induced hypercalcemia). The 2002 National Institutes of Health (NIH) consensus panel on the management of asymptomatic primary HPT (2002 NIH Panel) classified primary HPT into two categories: asymptomatic and symptomatic. Asymptomatic primary HPT accounts for 75 to 80% of all HPT cases. In both asymptomatic and symptomatic HPT, hypercalcemia in concert with an increased serum PTH concentration is the biochemical hallmark for the diagnosis of primary HPT. Other useful tests can be the dosage of albumin, phosphorous, BUN/Cr, 24-h urine Ca (r/o FHH) and bone mineral density. Imaging is essentially used for pre-operative planning or in case of recurrent HPT after surgery.

Bibliography 1. 2.

3.

4.

5.

Adami S, Marcocci C, Gatti D (2002) Epidemiology of primary hyperparathyroidism in Europe. J Bone Miner Res 17 Suppl 2:N18–N23 Mazzeo S, Caramella D, Lencioni R et al (1995) [Preoperative imaging in the detection of parathyroid tumefaction in patients with primary hyperparathyroidism. The authors’ own experience] Radiol Med (Torino) Dec 90(6):747–755. Italian Mazzeo S, Caramella D, Marcocci C et al (2000) Contrast-enhanced color Doppler ultrasonography in suspected parathyroid lesions. Acta Radiol Sep 41(5):412–416 Mazzeo S, Caramella D, Lencioni R et al (1996) Comparison among sonography, double-tracer subtraction scintigraphy, and doublephase scintigraphy in the detection of parathyroid lesions. Am J Roentgenol Jun 166(6):1465–1470 Adams BK, Devi RT, Al-Haider ZY (2004) Tc-99m sestamibi localization of an ectopic mediastinal parathyroid tumor in a patient with primary hyperparathyroidism. Clin Nucl Med Jun 29(6):388–389

Neoplasms, Phyllodes, Breast LUIS P INA Department of Radiology, Clıńica Universitaria de Navarra, Navarra, Spain [email protected]

Synonyms Cystosarcoma phyllodes; Phyllodes tumor

Definition Phyllodes tumor is a fibroepithelial tumor of the breast that exhibits a diverse range of biologic behavior (1).

Pathology The gross appearance is a round to oval well-circumscribed mass without a true capsule. Microscopically, this tumor shows a variety of appearances from those resembling ▶fibroadenomas to others like sarcomatous tumors. Many phyllodes tumors show a characteristic leaflike architecture. The epithelial component is usually single layered, but other changes, such as hyperplasia, atypical hyperplasia, or in situ carcinoma, may be seen. However, the stromal component determines their clinical behavior. Four histologic features are important for classifying this lesion (1): (i) stromal cellular atypia, (ii) mitotic activity per 10 high-power fields, (iii) stromal overgrowth, and (iv) tumor margins. Phyllodes tumors can be classified into three categories: benign, borderline, and malignant. The first type shows increased stromal cellularity with mild to moderate atypia, low mitotic rates (10 mm) is associated with a low risk of recurrence. Axillary dissection is not routinely performed since nodal involvement is very rare. Hematogenous metastases of malignant cases are uncommon, accounting for 10% of all phyllodes tumors (2). The lung is the most commonly affected organ.

Neoplasms, Phyllodes, Breast

1289

Neoplasms, Phyllodes, Breast. Figure 1 Well-circumscribed round mass. Notice the high density of the lesion. The patient, a 59-year-old woman, reported a rapid growth in the last 3 months. Pathology: ▶malignant phyllodes tumor.

Imaging Mammography

N

Mammographic features are similar to those of fibroadenomas showing as oval, round or lobulated masses with well-circumscribed margins, but occasionally indistinct margins may be seen (Figs. 1 and 2). However, adjacent fibroglandular tissue may occult the lesion. Sometimes mammography may detect coarse calcifications inside a phyllodes tumor, similar to fibroadenomas. Lesions greater than 30 mm are more likely to be malignant.

Ultrasound Phyllodes tumors appear as oval, round, or lobulated masses with well-delimited margins and acoustic enhancement, resembling fibroadenomas. However, ultrasound may detect internal cysts and heterogeneous echostructure (Fig. 3). Both findings are not specific, but very often they lead to a biopsy (3, 4).

Neoplasms, Phyllodes, Breast. Figure 2 A lobulated mass with indistinct margins. Craniocaudal view, (detail). Biopsy revealed a ▶benign phyllodes tumor.

Nuclear Medicine There are no indications for the diagnosis of phyllodes tumor.

Magnetic Resonance Phyllodes tumors present as oval, round, or lobulated wellcircumscribed masses, usually hyperintense on T2-weighted images. After contrast medium administration, these lesions enhance rapidly. The role of magnetic resonance is not well established, but it may be useful for planning conservative surgery or evaluating thoracic wall involvement (4).

Diagnosis The rapid growth of a palpable, round to oval smoothly marginated mass is the main symptom to raise the suspicion of a phyllodes tumor. However, the widespread use of mammography has led to the detection of nonpalpable

1290

Neoplasms Pulmonary

Neoplasms Pulmonary K ATHARINA M ARTEN , C HRISTOPH E NGELKE Department of Radiology, Klinikum rechts der Isar, Technical University Munich, Munich, Germany [email protected]

Synonym Pulmonary neoplasms Neoplasms, Phyllodes, Breast. Figure 3 Ultrasonography of the case presented in Fig. 2. Lobulated hypoechoic solid mass.

phyllodes tumors. The main goal is to differentiate them from other benign lesions, such as fibroadenomas (1). This differentiation is usually difficult without a biopsy. The fibroepithelial composition of phyllodes tumors is similar to fibroadenomas, and this feature makes the differentiation between both lesions difficult. Furthermore, the wide spectrum of benign to malignant cases increases this difficulty. Fine-needle aspiration usually fails to distinguish fibroadenoma and phyllodes. The cystic component may be misinterpreted as “fibrocystic change.” Core needle biopsy may suggest the diagnosis, but the differentiation with a fibroadenoma may be impossible. In most cases excisional biopsy is needed for diagnosis, especially if a rapid growth of a solid mass is detected (1, 5).

Bibliography 1.

2.

3.

4.

5.

Anderson BO, Lawton TJ, Lehman CD et al (2004) Phyllodes tumors. In: Harris JR, Lippman ME, Morrow (eds) M et al Diseases of the Breast. Lippincott Williams & Wilkins, Philadelphia, pp 991–1006 Bellocq JP, Magro G (2003) Fibroepithelial tumours. In: Tavassoli FA, Devilee P (eds) Pathology and Genetics of Tumours of the Breast and Female Genital Organs. International Agency for Cancer, Lyon, pp 99–103 Cardenõsa G (2001) Lobules. In: Cardenõsa G (ed) Breast Imaging Companion. Lippincott Williams & Wilkins, Philadelphia, pp 269–306 Heywang-Ko¨brunner SH, Schreer I, Dershaw DD (eds) (1997) Other semi-malignant and malignant tumors. Diagnostic Breast Imaging. Stuttgart, Thieme, pp 265–279 Hughes LE (2000) Fibroadenoma and related tumors. In: Hughes LE, Mansel RE, Webster DJT (eds) Benign Disorders and Diseases of the Breast. Concepts and Clinical Management. Saunders, London, pp 73–94

Definition Primary neoplasms of the lung encompass numerous distinct histological entities that arise from various pulmonary tissues and are best categorised according to the classification scheme proposed by the world health organisation (WHO). These tumours include pulmonary carcinoma, neuroendocrine neoplasms, neoplasms of the tracheobronchial glands, tracheobronchial papillomas and rarely among others, primary pulmonary lymphoma, primary pulmonary sarcoma, ▶pulmonary haemangiopericytoma and ▶malignant angioendotheliomatosis.

Pathology/Histopathology The term pulmonary carcinoma refers to tumours arising from the epithelial surface of the bronchi and alveoli, and embraces four main histological subtypes: adenocarcinoma, squamous cell carcinoma, small cell carcinoma and large cell carcinoma. Adenocarcinoma (Figs. 1, 2) is the most common pulmonary carcinoma in the United States, accounting for about 30–35% of all cases, and can be subdivided into four histological subtypes: acinar carcinoma, papillary carcinoma, solid carcinoma with mucin formation and ▶bronchioloalveolar carcinoma (BAC). Most adenocarcinomas (except BAC) are peripherally located and well circumscribed. BAC may present as solitary nodule or as parenchymal consolidation. Non-destructive growth is a characteristic feature, as tumour cells spread along the pulmonary framework, commonly forming a single layer. The surrounding interstitium is thickened by fibrous tissue or a chronic inflammatory infiltrate. Focal fibrosis and necrosis occur frequently, while cavitation is relatively rare (1–3). Squamous cell carcinomas (Fig. 3) account for about 30% of all pulmonary carcinomas. Most frequently they arise from the segmental or lobar bronchial epithelium. Early lesions may expand within the bronchial mucosa

Neoplasms Pulmonary

Neoplasms Pulmonary. Figure 1 Sixty-nine year-old woman with stage IV inoperable non-small-cell Lung cancer of the right upper lobe. Note the tumour compression of the right pulmonary artery, the invasion into the ▶superior vena cava (*) and the infracarinal mediastinum. There is right-sided pleural effusion containing malignant cells due to pleural invasion.

Neoplasms Pulmonary. Figure 2 Seventy-one year-old male smoker with incidental finding of a 1.8 cm solid nodule in the posterior left upper lobe segment displaying spiculated margins and pleural tags. Note the advanced centrilobular emphysema of the right upper lobe. The lesion was bioptically confirmed as non-small-cell lung cancer.

without invasion into the airway lumen or the peribronchial tissue. At a later stage, however, most lesions invade and obliterate the bronchial lumen as polypoid or papillary tumours with subsequent airway obstruction

1291

Neoplasms Pulmonary. Figure 3 Fifty-three year-old male smoker with cavitating nodule and solid satellite nodule in the apical right upper lobe segment displaying extensive marginal spiculations and feeding bronchus. Note the moderate centrilobular bilateral emphysema of the upper lobes. The lesions were bioptically confirmed as non-small-cell lung cancers.

expressed by distal atelectasis and obstructive pneumonitis. Large tumours often display central necrosis and erosion into a local bronchus with drainage of necrotic parts and resulting cavitation. Histologically, invasive squamous cell carcinoma is characterised by inter-cellular bridges and/or keratinisation. Tumours may be well differentiated, displaying obvious keratinisation, or are poorly differentiated, rendering differentiation from large cell or small cell carcinoma difficult. Histological variants of squamous cell carcinoma include small cell, papillary, clear cell and basaloid tumours (1, 2). Small cell carcinoma (Fig. 4) contributes to about 15–25% of all pulmonary carcinomas. Evidence suggests that it is a neuroendocrine tumour, most likely derived from undifferentiated airway epithelial cells capable of neuroendocrine differentiation. Small cell carcinoma is typically located in the vicinity of proximal airways, most commonly of lobar and main bronchi. Rarely, tumours are located in the lung periphery without obvious airway association. Early, centrally located small cell carcinoma is frequently poorly defined and spreads along submucosal and peribronchovascular connective tissues. Advanced lesions invade the adjacent lung parenchyma and obliterate underlying bronchi and vascular structures. Endobronchial growth occurs much less frequently than in squamous cell carcinoma. Histologically, tumour cells may be round or fusiform and may be grouped in clusters, sheets or rosettes; stroma is normally scanty. Necrosis is frequent; a desmoplastic reaction is commonly minimal or absent (1, 2).

N

1292

Neoplasms Pulmonary

Neoplasms Pulmonary. Figure 4 Forty-nine year-old male smoker with extensive disease from small-cell lung cancer. (a) There is complete mediastinal encasement by tumour masses. (b) The patient had massive secondary pulmonary hypertension with dilated pulmonary artery trunk due to compression of the main pulmonary arteries (arrows).

Large cell carcinoma is diagnosed in 15–20% of pulmonary carcinoma cases. Most frequently it occurs in the lung periphery and often has a bulky appearance. Cavitation is rare, but focal necrosis is a common feature. Large cell carcinoma consists of sheets of large cells containing abundant eosinophilic cytoplasm. Inter-cellular bridges, keratinisation, acinar or papillary structures are absent. A number of histological subtypes exist, e.g. large cell neuroendocrine carcinoma, clear cell carcinoma, basaloid carcinoma and lymphoepithelioma-like carcinoma, among others (1, 2). Neuroendocrine neoplasms encompass carcinoid tumours, pulmonary tumorlets and neuroendocrine cell hyperplasia. These tumours display common ultrastructural and immunohistochemical features resembling neuroendocrine cells, i.e. cells that contain neurosecretory granules and polypeptide hormones that occur in the normal tracheobronchial epithelium. Carcinoid tumours are low-grade malignancies, which are derived from the glandular or surface epithelium of the airways. Clinicopathologically, they are subdivided into typical and atypical forms. About 80–90% of cases are typical carcinoids that most frequently arise within lobar, segmental or proximal subsegmental bronchi (1). Occasionally, multiple tumours occur. Typical carcinoids are histologically well-defined tumours, which may be separated from the adjacent airway epithelium by a thin layer of connective tissue. If there is association to an airway, the epithelium is commonly intact and may display squamous metaplasia or ulceration. The overall histological pattern of atypical carcinoids closely resembles that of typical carcinoids. However, atypical carcinoids display increased mitotic activity, nuclear pleomorphism and hyperchromasia, areas of increased cellularity or necrosis.

The term neoplasms of the tracheobronchial glands refers to a group of tumours with the histological appearance of tracheobronchial mucous glands, with close resemblance of oropharyngeal salivary glands. The most common histological types are adenoid cystic carcinoma and mucoepidermoid carcinoma. Adenoid cystic carcinoma most commonly arises within the tracheal and main bronchial lumen and in about 10–15% in the lung periphery. Mucoepidermoid carcinoma develops in the trachea and bronchi (1, 2).

Clinical Presentation The clinical presentation of lung neoplasms is nonspecific, and symptoms commonly occur only at an advanced tumour stage. They may reflect local disease, particularly tracheal or bronchial obstruction (e.g. wheeze, cough, haemoptysis or obstructive pneumonitis), or originate from tumour invasion into neighbouring anatomic structures such as the mediastinum or the chest wall. At late stages distant hepatic, osseous or cerebral metastases may cause symptoms specific to their locations. ▶Paraneoplastic syndromes (see keywords) may result from hormone secretion by the tumour or from immunologic reactions. Non-specific symptoms further include fatigue and weight loss.

Imaging On chest radiography (CXR) and CT, pulmonary carcinomas commonly present as a lung nodule (lesser than 3 cm in size) or a pulmonary mass. This is particularly the case in squamous cell and adenocarcinomas, whereas small cell carcinomas may present with

Neoplasms Pulmonary

isolated hilar or mediastinal lymphadenopathy. Occasionally, obstructive pneumonitis or atelectasis may occur as isolated findings, predominantly in squamous cell and small cell carcinomas. Differentiation of the obstructing tumour from post-obstructive atelectasis may be challenging. On CXR, segmental or lobar atelectasis, consolidation and mucous plugging of bronchi distal to the obstruction are common. A specific feature is ‘Golden’s S sign’ which refers to a focal convexity of the tumour with the concave shape of the inter-lobar fissure resulting from distal atelectasis with a characteristic S-shaped configuration. On CT, the tumour displays moderate contrast enhancement (15–30 Hounsfield units (HU) for most lesions on dynamic contrast-enhanced scans), whereas areas of atelectasis enhance conspicuously (>50 HU). Peripheral carcinomas presenting as a pulmonary nodule may rarely display calcifications that are typically eccentric and result from tumoural engulfment of a preexisting granuloma. Calcifications are more common in mass lesions larger than 3 cm in size. Carcinomas commonly display irregular, spiculated margins. Smooth or lobulated margins occur less frequently. Air bronchograms or bronchiolograms may be detected after engulfment of these structures by the tumour. Cavitation of peripheral carcinomas occurs most frequently with squamous cell carcinoma and is associated with hilar or mediastinal lymphadenopathy. A very important CT feature is the air-space pattern which almost exclusively occurs in BAC, reflecting the non-destructive growth of these tumours. This pattern may be localised or diffuse ranging from ground-glass opacification to frank airspace consolidation without displacement of normal lung structures. In mixed subtype adenocarcinomas with BAC component, a prominent BAC component is likely to have a ground glass appearance, while adenocarcinomas with invasive component are likely to be solid on thin section CT. This differentiation is important because lesions with large ground glass components are less likely to have vascular invasion or lymph node metastases. Mixed subtype adenocarcinomas with BAC component may occur as solitary or multiple unilateral or bilateral nodules, or a large mass with satellite nodules within the same lobe. The lobar consolidation pattern in BAC which is commonly combined with the ‘CT angiogram sign’ in BAC refers to well-opacified vessels surrounded by lowdensity pulmonary parenchyma following contrast administration; however, it is difficult to distinguish from pneumonia and may also occur in other disorders (1, 3). Carcinoid tumours most commonly present with evidence of airway obstruction on CXR, as a large majority of these tumours (80–85%) are located within lobar or segmental bronchi (1). Complete obstruction will induce peripheral atelectasis, reflected by increased attenuation with loss of volume or precipitate obstructive

1293

pneumonitis with features of pneumonia commonly without volume loss. Recurrent infections distal to the obstruction site may result in bronchiectasis and lung abscesses formation. In cases of partial proximal airway obstruction, hypoxic vasoconstriction and volume loss of the dependent lung may suggest the presence of an endobronchial lesion and should prompt bronchoscopy. Peripherally located carcinoid tumours present as pulmonary nodules that are well-defined, homogeneous, often lobulated, round or oval in shape. Most lesions measure 1–3 cm in diameter; however, large lesions also occur, particularly in atypical carcinoid tumours. Intratumoural calcification is frequently detected on CT and may display variable patterns (1). Adenoid cystic carcinoma presents as a lobulated, polypoid or smooth endotracheal or endobronchial mass on CXR which encroaches onto the airway lumen. CT is an excellent tool for the assessment of extraluminal growth and mediastinal invasion. The radiographic features of mucoepidermoid carcinoma are variable, ranging from a solitary nodule or mass to consolidation, atelectasis or a central mass with obstructive pneumonitis. The CT findings include a mass, which is oval or lobulated in shape, and may display punctuate calcifications. With involvement of the trachea there may be an evidence of an intra-luminal nodule on CT and with larger lesions also on CXR. Tracheobronchial papillomas may be solitary or multiple and are seen on CXR and CT as nodules within the airway lumen. On CT the presence of multiple papillomas may be evident by the detection of diffuse nodular thickening of the tracheal wall. Chronic bronchial obstruction is bound to induce atelectasis, obstructive pneumonitis or bronchiectasis, while partial obstruction may as well cause the decreased perfusion and decreased attenuation pattern of the affected parenchyma. With tumour involvement of the distal airways and lung parenchyma, a nodular pattern may be evident with predilection for the perihilar and posterior lung regions. Cavitation of such nodules, with occasional air–fluid level formation, is more frequent with larger lesions (2). Primary pulmonary lymphoma may present as single or more commonly multiple pulmonary nodules or masses 2 mm to 8 cm in size, as well as unilateral or bilateral areas of consolidation. Distribution is frequently peribronchial, air bronchograms commonly occur. Less common findings on CT are thickening of the bronchovascular bundles, ground-glass opacification and septal lines (1).

Nuclear Medicine Positron-emission tomography (PET) using intravenous administration of 2-(fluorine-18)-fluoro-2-deoxy-D-

N

1294

Neoplasms, Small Bowel

gluocose (18FDG) is likely to become more important for the diagnosis of lung neoplasms in the future, particularly for the differentiation of benign from malignant pulmonary nodules larger than 8–10 mm in size. 18 FDG-PET imaging relies on physiologic glucose metabolism to diagnose lung cancer and has a high sensitivity and specificity for malignant nodules of at least 0.8 cm in diameter. However, false-positive diagnoses occur in active granulomatous infections, whereas false-negative diagnoses are commonly encountered in BAC and carcinoid tumours. Single-photon emission computed tomography (SPECT) imaging is clearly inferior to PET imaging in the differentiation of benign and malignant pulmonary nodules. Both PET and SPECT, however, play important roles in lymph node staging of lung cancer, with their sensitivities and specificities for diagnosis of malignant lymphadenopathy reaching 80%. Imaging of distant metastasis of lung cancer has also become a domain of PET-imaging, particularly in combination with CT. Octreotide-PET imaging holds promise in the diagnosis of small peripheral carcinoid tumours with ectopic ACTH production, that are difficult to detect on CT (4, 5).

Diagnosis The capacity of CXR to reveal early stage lung neoplasms is limited. Recent technologic advances with application of computer-aided diagnosis and dual-energy subtraction methods on digital CXR data may offer certain improvements on this. Latest technologies in CT, particularly thin-collimation multislice CT (MSCT) possess a very high sensitivity for the detection of pulmonary nodules, but have a limited specificity for diagnosis of lung cancer (6). Advances in scanning technique such as dynamic contrast-enhanced CT for perfusion imaging, as well as recognition and metric assessment by computer-aided diagnosis with automated nodule volumetry may enhance its ability to determine nodule growth and therefore help to diagnose malignancy at an early stage. Sputum analysis encompasses sputum cytology, sputum immunostaining and newer methods such as sputum polymerase chain reaction based assays for detecting oncogene mutations. Sputum cytology has a relatively low overall sensitivity for detection of lung cancer of 20–30%, showing moderate variation with the employed technique. The sensitivity is highest for squamous cell carcinoma and lowest for adenocarcinoma. Centrally located lesions, lesions larger than 2 cm and lower lobe lesions are best detected with sputum cytology. The false-positive rate is very low (70 years).

Acquired Cholesteatoma Individuals of all ages may develop cholesteatoma. The leading symptom is chronic otorrhea and conductive hearing loss. In more advanced stages, symptoms are due to progressing bone erosion and may include vertigo (erosion of the lateral semicircular canal) or facial palsy (erosion of the facial nerve canal). Meningitis, abscess in the temporal lobe, and lateral sinus thrombosis are possible late-stage complications.

Symptoms depend on the exact location and size of the tumor and may include dysfunction of cranial nerves V, VII, and VIII. There is a female predominance (2–3:1). The age peak is >60 years.

Most lipomas are asymptomatic. Dizziness and sensorineural hearing loss are the most common symptoms. Only in cases unresponsive to medical therapy is surgical decompression done. Because nerve fibers run through the lipoma, surgical therapy is extremely difficult.

Endolymphatic Sac Tumor Sensorineural hearing loss and tinnitus are common symptoms. The tumor occurs in patients between 20 and 80 years. Endolymphatic sac tumors have a higher incidence in patients with von Hippel–Lindau disease.

Rhabdomyosarcoma This tumor is the second most common head and neck malignancy in children and has an age peak at 4–6 years. Otalgia, bloody otorrhea, facial nerve palsy, and sensorineural hearing loss are common symptoms.

Imaging Exostosis High-resolution computed tomography (HRCT) with bone algorithm shows a circumscribed tumor that consists of dense compact bone. It has a broad base and begins deep to the isthmus within the medial aspect of the osseous internal auditory canal in the close vicinity of the annulus.

Glomus Tympanicum Due to the vascular nature of the tumor, pulsatile tinnitus is the most common symptom at initial presentation (90%). Conductive hearing loss occurs in 50% of patients and is due to blocked movement of the tympanic membrane.

Osteoma HRCT demonstrates a pedunculated tumor with both layers of normal bone (cortical shell and trabecular bone).

N

1320

Neoplasms, Temporal Bone

Malignant Neoplasm of the External Auditory Canal and Ear HRCT and MRI usually do not allow for a specific diagnosis of malignant tumors in adults. However, erosion of the underlying bone and infiltration of neighboring structures are signs of malignancy of a mass. The main role of imaging is to assess the extent of the neoplasm and infiltration of vital structures and determine the presence of metastasis. The periparotid lymph nodes represent the primary lymphatic drainage of the ear. Thus, the parotid gland and upper neck must be included in imaging studies.

Acquired Cholesteatoma HRCT shows the extent of the soft-tissue mass. Erosion of bone is usually present (Fig. 1) and allows differentiation of cholesteatoma from middle ear adenoma. On MRI, the mass is isointense to cerebrospinal fluid on T1- and T2-weighted

images. Cholesteatoma does not enhance after administration of contrast medium, but due to granulation tissue, there is enhancement at the rim of the tumor.

Glomus Tympanicum MRI is the modality of choice and demonstrates a hyperintense middle ear mass on T2-weighted images. After intravenous administration of contrast medium, the tumor strongly enhances on T1-weighted images. CT does not demonstrate erosion of the bone. The ossicles are usually spared.

Middle Ear Adenoma CT shows a mass behind the intact tympanic membrane. The mastoid is usually well pneumatized. The ossicles are encased by the tumor, but usually no bony erosion is present.

Schwannoma The CT appearance of middle ear schwannoma is nonspecific, and differentiation from paraganglioma is usually not possible. On MRI, schwannomas strongly enhance on T1-weighted images after the administration of contrast medium (Fig. 2). T2-weighted high-resolution images also show the tumor (Fig. 3).

Meningioma

Neoplasms, Temporal Bone. Figure 1 Acquired cholesteatoma. The axial high-resolution computed tomographic image shows a middle ear cavity opacification. The ossicles are eroded (arrow).

Meningiomas enhance strongly on CT and MRI after intravenous injection of contrast medium. The dural tail sign—linear enhancement along the adjacent dura—is a typical finding on MRI. Brain edema in the vicinity of the tumor is a sign of infiltration of the arachnoid. In these cases the tumor is difficult to remove, and a higher rate of recurrence must be expected.

Neoplasms, Temporal Bone. Figure 2 Acoustic schwannoma. Axial T1-weighted magnetic resonance image before (left) and after intravenous administration of contrast medium (right). The tumor is isointense to the brain before administration of contrast medium (arrow in the left image). It strongly enhances after administration of contrast medium (arrow in the right image).

Neoplasms, Thyroid, Benign and Malignant

1321

Diagnosis

Lipoma

Several tumors, such as schwannoma and lipoma, have a pathognomonic appearance on imaging. In all tumors without such a pathognomonic imaging appearance, biopsy or surgical removal and histology are essential for correct diagnosis. In patients with exostosis, otoscopy shows a circumferential stenosis of the external auditory canal. In cholesteatoma, otoscopy shows a retraction pocket or perforation of the tympanic membrane and a white mass behind the tympanic membrane. If otoscopy shows a reddish mass behind the posteroinferior part of the tympanic membrane, glomus tympanicum tumor must be considered. Before biopsy can be performed, imaging is required to rule out an aberrant internal carotid artery. This developmental anomaly causes similar clinical symptoms (pulsatile tinnitus and conductive hearing loss) and has the same appearance at otoscopy. Furthermore, glomus tympanicum must be differentiated from glomus jugulare, which arises from the jugular foramen and may extend through the posterior floor of the middle ear cavity into the middle ear, because the two tumors require different surgical approaches.

On T1-weighted MR images, the tumor has the typical high signal intensity of fatty tissue. It loses signal if fat saturation techniques area applied.

Bibliography

Neoplasms, Temporal Bone. Figure 3 Acoustic schwannoma. The T2-weighted magnetic resonance image shows a small tumor (arrow) in the internal auditory canal.

1.

Endolymphatic Sac Tumor HRCT shows a soft-tissue mass located at the posterior wall of the temporal bone, eroding the bone. The tumor matrix contains speculated calcifications. MRI shows foci of enormous enhancement within the mass after intravenous injection of contrast medium.

2.

3.

4.

Vogl T, Bruning R, Schedel H et al (1989) Paragangliomas of the jugular bulb and carotid body: MR imaging with short sequences and Gd-DTPA enhancement. Am J Roentgenol 153(3):583–587 Salzman KL, Davidson HC, Harnsberger HR et al (2001) Dumbbell schwannomas of the internal auditory canal. Am J Neuroradiol 22(7):1368–1376 Dahlen RT, Johnson CE, Harnsberger HR et al (2002) CT and MR imaging characteristics of intravestibular lipoma. Am J Neuroradiol 23(8):1413–1417 Gaeta M, Blandino A, Minutoli F et al (1999) Sudden unilateral deafness with endolymphatic sac adenocarcinoma: MRI. Neuroradiology 41(10):799–801

Rhabdomyosarcoma HRCT demonstrates a mass that causes extensive bony destruction. On T2-weighted MR images, rhabdomyosarcoma is hyperintense. On T1-weighted images, the mass is hypointense and enhances after intravenous administration of contrast medium.

Neoplasms, Thyroid, Benign and Malignant C ARMELA V ISALLI

Nuclear Medicine In patients with neoplasm of the temporal bone, nuclear medicine has only a limited role. Bone scintigraphy with technetium 99 m methylene diphosphonate may lead to detection of metastasis to the temporal bone at an early stage or show metastasis from a temporal bone neoplasm to other parts of the body.

Department of Radiologic Sciences, University of Messina, Messina, Italy [email protected]

Synonyms Benign and malignant neoplasm; Thyroid nodules; Thyroid tumors

N

1322


Definition Thyroid neoplasms include all lesions caused by abnormal thyroid tissue growth due to excessive cellular division and proliferation. They are comprehensive, including benign and malignant nodules. Nodules are palpable masses in the thyroid gland, often found incidentally on ultrasound examination. They occur in 4–7% of the population and can be solitary lesions or occur in a context of multinodular enlargement of the gland. This pathological condition is four times more common in women than in men and occurs more often in people who live in geographic areas with iodine deficiency or who are exposed to ionizing radiation.

Pathology/Histopathology In general, nodular disease of the thyroid is common; however, malignancy of the thyroid is rare. Approximately 5–10% of thyroid nodules are malignant, whereas the remainders represent a variety of benign diagnoses, including colloid nodules, degenerative cysts, hyperplasia, thyroiditis, and benign neoplasms (Table 1). A rational approach focal to management of a thyroid nodule is based on the clinician’s ability to distinguish the more common benign entities from malignancy in a highly reliable and economic way. Colloid nodules are the most common and do not have an increased risk of malignancy. Most follicular adenomas are benign; however, some may share morphological aspects with follicular carcinoma. Although carcinoma of the thyroid gland is rare, it is the most common malignancy of the endocrine system. Neoplasms, Thyroid, Benign and Malignant. Table 1 Types of thyroid nodules Adenoma

Liquid nodule Inflammatory nodule Carcinoma

Other malignancy

• • • • • • • • • • • • • • •

Macrofollicular adenoma Microfollicular adenoma Atypical adenoma Papillary adenoma Simple cyst Colloid nodule Focal thyroiditis Papillary (60–80%) Follicular (10–20%) Medullary (5–10%) Anaplastic (5%) Lymphoma (4%) Sarcoma Teratoma Metastases (2–17%)

Thyroid carcinoma usually presents as a solitary nodule. There are four forms: papillary, follicular, medullary, and anaplastic (1). Differentiated tumors (papillary or follicular) are highly treatable and usually curable. Poorly differentiated tumors (medullary and anaplastic) are aggressive, metastasize early, and have a much poorer prognosis. ▶Thyroglobulin can be used as a tumor marker for well-differentiated forms. Women are affected more often than men, and any age group can be affected, although thyroid cancer usually occurs in people between the ages of 25 and 65 years.

Papillary Thyroid Cancer Papillary thyroid cancer is the most common type of thyroid cancer, accounting for 60–80% of all thyroid malignancies. It occurs more frequently in women and presents in the 30–40-year-old age group. It is also the predominant cancer type in children with thyroid cancer and in patients who have had radiation to the head and neck. In this group the cancer tends to be multifocal, with early lymphatic spread and a poor prognosis. Papillary carcinoma is a nonencapsulated lesion that histologically consists of a stromal core and packed papillae with a small amount of colloid within follicles. Small calcified bodies, known as psammoma bodies, may be found in the stroma.

Follicular Thyroid Cancer Follicular thyroid cancer occurs in approximately 20% of patients with thyroid cancer and is more common in women above 50 years of age. A particular form of follicular cancer is Hu¨rthle cell carcinoma, characterized by large polygonal thyroid follicular cells. Follicular carcinoma is an encapsulated lesion, generally solitary and characterized by angioinvasive behavior with a consequently high frequency of distant metastases. Regional lymph node involvement has been reported to be less than 13%. When lymph nodes are involved, the prognosis is usually poor. Because this cancer is usually very well differentiated, it is difficult to distinguish from follicular adenoma. For the definitive diagnosis, demonstration of capsular or surrounding thyroid tissue invasion is necessary.

Medullary Thyroid Cancer Medullary thyroid cancer originates from the parafollicular cells (C cells), which produce the hormone calcitonin. Carcinoembryogenic antigen (CEA) is a tumoral marker produced by medullary thyroid cancer. Its prognosis is poorer than that for follicular or papillary


thyroid cancer, especially when it has metastasized beyond the thyroid gland. This cancer can develop as a “sporadic form” or as a part of the complex syndrome of ▶multiple endocrine neoplasias, type 2 (MEN 2a and MEN 2b). Medullary thyroid cancer can cause diarrhea and flushing episodes. The main sites of spread of this cancer are local lymph nodes in the neck and mediastinum, as well as liver, lung, and bone.

Anaplastic Thyroid Cancer This form of neoplasm has a very poor prognosis, with a high mortality (near 100%) due to its aggressive behavior, rapidly invading surrounding tissue and the trachea.

1323

be used to determine whether a nodule is benign or malignant, even if the presence of clinical and laboratory signs of hyperthyroidism or hypothyroidism is most likely related to a benign nodule. The serum TSH level allows assessment for thyrotoxicosis (TSH). Thyrotoxicosis may be due to a diffuse toxic goiter (Graves’ disease), toxic multinodular goiter, or single toxic nodule (Plummer’s disease). The most common cause of hypothyroidism is Hashimoto’s thyroiditis. When the TSH level is normal, nodular aspiration should be considered. Thyroglobulin levels are useful tumor markers once the diagnosis of malignancy has been assessed, but they are nonspecific with regard to differentiating a benign from a cancerous thyroid nodule (2).

Other Tumors The thyroid gland may also be the site of other tumors, including sarcoma, lymphoma, teratoma, and metastasis from other cancers, such as lung, breast, and kidney neoplasms.

Imaging After physical examination, ultrasound is the first imaging modality for assessing thyroid disorders. CT and MRI have a small role, but specific indications include staging of cancer or evaluation of the extension of mediastinal goiter (3).

Clinical Presentation At physical examination, thyroid nodules may be diffuse or localized, soft or hard, mobile or fixed, and painful or painless (2). If nodules are less than 1 cm in diameter, they are usually not palpable unless they are located in the anterior portion of the gland. Physical characters indicative of malignancy are a hard, fixed lesion; nodules greater than 4 cm; or hoarseness. After clinical examination, the diagnostic algorithm for evaluating a thyroid nodule includes the following: . Laboratory tests: thyroid stimulating hormone (TSH), triiodothyronine (T3), thyroxine (T4), antithyroid peroxidase antibodies, thyroglobulin, calcitonin . Ultrasonography . Nuclear medicine imaging (scintigraphy and positron emission tomography) . Fine-needle aspiration (FNA) . Computed tomography (CT) . Magnetic resonance imaging (MRI)

Laboratory Tests Laboratory tests are important for distinguishing different thyroid diseases, but thyroid function tests should not

Ultrasonography Ultrasonography with high-resolution probes is used to confirm the presence of a nodule, to determine whether it is a solitary nodule, and to assess any cystic component. It can also differentiate thyroid from adjacent nonthyroidal masses. Ultrasonography is sensitive for identifying a focal lesion. In fact, it detects more nodules than palpation. It is suitable for evaluating the size, location, and structure of a nodule (solid or cystic, hypoechoic, hyperechoic, with or without halo sign, microcalcifications) (Fig. 1). In a solid hypoechoic nodule, irregular borders, central microcalcifications, prevalent, intranodular vascularity, and cervical adenopathies are the main signs suggesting malingnance. Conversely, findings of hyperechoic texture, comettail artifacts from colloid, lack of blood flow in the nodule, a peripheric halo, and a smooth border. Because ultrasonography cannot confidently distinguish benign from malignant nodules, FNA is necessary for this (4). Ultrasonography can guide the biopsy and be used to determine changes in the size of nodules over time, either in the follow-up of a lesion thought to be benign or in detecting recurrent lesions in patients with thyroid cancer (3, 4).

N

1324


The predictive value of ultrasonography may significantly increase if the ultrasonographic parameters of thyroid nodules, such as microcalcifications, echogenicity, and halo sign, are combined with their vascular pattern. In fact, color Doppler and spectral analysis can demonstrate different vascular patterns as prevalent perinodular or central vascularization (Fig. 2). Many studies have shown that nodules with perinodular vascularization have a lower risk of malignancy than nodules with central vascularization.

CT and MRI CT and MRI usually have no place in the assessment of patients with thyroid nodules, but nodules may be found incidentally during examination of the neck for reasons not relating to the thyroid gland. CT and MRI do play an important role in evaluating the mediastinal extension of thyroid masses and also in assessing the spread of cancer (3).

Nuclear Medicine Scintigraphy

Neoplasms, Thyroid, Benign and Malignant. Figure 1 Longitudinal section of the right lobe of the thyroid. Nodule with hypoechoic and dishomogeneous structure, regular borders with halo sign.

Scintigraphy is still the standard method for functional imaging of the thyroid (3). The two isotopes most commonly used are 123 I and 99m Tc-pertechnetate. Thyroid scanning measures the amount of iodine (usually 99m technetium isotope) trapped within the nodule. A normal scan indicates that the iodine uptake is similar in both lobes of gland. A nodule is classified as “cold” (decreased uptake), “warm” (uptake similar to that of surrounding tissue), or “hot” (increased uptake). A large number of thyroid nodules may be cold on radionuclide scan, but only 5–15% of these are malignant. A “hot” nodule is hyperfunctional and is almost always benign. Nuclear imaging cannot reliably distinguish between benign and malignant nodules and is not required if nodules are present. FNA biopsy has replaced nuclear imaging as the initial evaluation procedure. However, in patients with a suppressed TSH level, thyroid scanning determines regional uptake or function and can be used as a secondary study (4).

Positron Emission Tomography Positron emission tomography (PET) with 18F-fluorodeoxyglucose (FDG) is an imaging technique that might improve the diagnostic accuracy, allowing identification of metabolic alterations in tumors. At present it plays only a small role in thyroid imaging but can be particularly useful to evaluate metastatic disease. FDG PET measures the body’s metabolic activity, and areas in which cancer is present will show up as brighter on the scan because the disease is more metabolically active than noncancerous cells.

Radiometabolic Therapy Neoplasms, Thyroid, Benign and Malignant. Figure 2 The nodule shows a rich periferic and intranodular vascularization at the evaluation with color Doppler.

Nuclear medicine plays a specific role in the ablative treatment of differentiated thyroid neoplasms. Ablation

Neoplasms, Urethra

therapy is performed with indications:

131

I and for the following

. To destroy the small amount of thyroid tissue remaining in the neck after surgery . To treat functional metastases . To treat patients with elevated thyroglobulin levels 131

Before and after ablation therapy, a whole-body I scan is necessary to identify metastatic disease and treatment response, respectively.

1325

Neoplasms, Urethra F RANCK KOSKAS 1 , O LIVIER H E´ LE´ NON 2 , M ARIE -F RANCE B ELLIN 1 1

Department of Radiology, University Paris-Sud 11, Paul Brousse Hospital, Villejuif, France 2 Department of Radiology, University Paris, Necker Hospital, Paris, France [email protected] [email protected]

Diagnosis For a diagnosis of thyroid neoplasm, evaluation of the patient should start with anamnesis, physical examination, and laboratory tests (TSH, T3, T4, autoantibodies). Ultrasonography should then be done to confirm the presence of a nodule or multiple nodules and their structural characteristics. If a solitary thyroid nodule is found, FNA biopsy should be performed to determine whether the nodule is malignant. FNA biopsy is believed to be the most effective method available for distinguishing between benign and malignant lesions. This modality has a sensitivity of 68–98% and a specificity of 72–100%. The accuracy of FNA biopsy in diagnosing thyroid conditions depends highly on the expertise and experience of the cytopathologist and the technical skill of the physician performing the biopsy. The results of FNA are interpreted as benign, malignant, suspicious, or indeterminate. False-negative or false-positive cases may occur when nodules are very large or very small; these errors can be minimized by using ultrasound-guided biopsy. To confirm a thyroid nodule or to obtain functional information, thyroid scintigraphy can be used. Although thyroid scanning may give a probability that a nodule is benign or malignant, it cannot truly differentiate benign from malignant nodules (5).

References 1.

2.

3. 4.

5.

Baloch ZW, Li Volsi VA (2000) Mini-symposium: endocrine pathology. Newly described tumors of the thyroid. Current Diagnostic Pathology 6:151–164 Bennedbaeck FN, Perrild H (1999) Diagnosis and treatment of the solitary thyroid nodule. Results of a European survey. Clinical Endocrinology 50:357–363 Naik KS, Bury RF (1998) Imaging the thyroid. Clinical Imaging 53:630–639 Iannuccilli JD, Cronan JJ, Monchik JM (2004) Risk for malignancy of thyroid nodules as assessed by sonographic criteria. The need for biopsy. J Ultrasound Med 23:1455–1464 McHenry CR, Slusarczyk SJ, Askari AT et al (1998) Refined use of scintigraphy in the evaluation of nodular thyroid disease. Surgery 124:656–662

Synonyms Proliferative disorders of the urethra; Tumors of the urethra

Definition Urethral neoplasms include benign and malignant tumors, the latter being either primary tumors of the urethra or metastatic tumors. They are rare entities. The urethra represents the distal part of the urinary tract. It extends from the bladder neck to the external urethral meatus.

Characteristics Urethral neoplasms are rare entities with only a few reported series in the radiology literature. Herein, we review the anatomy and histology specific to male and female urethras. We also describe the imaging features of urethral tumors.

Normal Anatomy and Histology of the Male Urethra The male urethra has a mean length of 18 cm and is subdivided into anterior and posterior portions, each of which is subdivided into two parts: 1. The posterior urethra stretches from the bladder neck to the lower edge of the urogenital triangle and includes: a. The prostatic urethra, which is 3 cm long in the young male, is the widest part of the canal. It is spindle-shaped and narrowest at its junction with the membranous urethra. On the floor of the canal is a narrow longitudinal ridge, the seminal colliculus. On each side of the ridge lies the prostatic sinus, a depressed fossa into

N

1326

Neoplasms, Urethra

which the periurethral prostate ducts empty. At the summit of the seminal colliculus (verumontanum), lies the blind pouch of the prostatic utricle surrounded by the slit-like openings of the ejaculator ducts of the seminal vesicles. b. The membranous urethra, approximately 1.5 cm long, passes through the urogenital triangle, which contains the Cowper glands embedded in the external urethral sphincter. In this tract, the urethra is surrounded by the compressor muscle of the urethra and the perineal muscles. 2. The anterior urethra, 14–15 cm long is subdivided into: a. The bulbar urethra, which is of larger diameter (1.5–2 cm), is surrounded by the bulb of the corpus spongiosum, does not have stiff fascia, and extends from the perineal area to the suspensor ligament of the penis. Many small mucous glands of Littre´ are located at its inner surface. b. The penile or pendulous urethra, has a relative uniform diameter approximately 1 cm, stretching from the penile ligament to the external urethral meatus. Before its emergence at the meatus, there is an ampullar dilatation called fossa navicularis. Glands of Littre´ are also located near the fossa navicularis but fewer than in the bulbar urethra. Because of its complex embryological origin, the urethral epithelial lining has several histological characteristics: transitional epithelium, from the bladder neck to the seminal verumontanum; then cylindrical, to the fossa navicularis, and, finally squamous epithelium, to the external meatus. These differences explain the histological diversity of urethral tumors.

urethra is located close to the anterior vaginal wall and enveloped by common musculature (the compressor muscle of the urethra).

Pathological, Clinical, and Imaging Features Clinically, these tumors are often heralded by obstructive voiding symptoms and hematuria, which clears near the end of micturition. On urethrograms, urethral tumors appear as filling defects and only the localization provides a clue to determine the origin. Biopsy is mandatory to establish the correct diagnosis.

Benign Tumors For both men and women, condylomata are found on the anterior urethra: it is the most common abnormality. Benign tumors of the urethra are very rare: they may be of epithelial or mesenchymal origin. In women, leiomyoma of the urethra is extremely rare. In men, a fibroepithelial polyp is of embryonic origin, usually originating in the prostate and projecting into the urethra. The polyp is connected to the verumontanum via a stalk, which maintains the polyp in the prostatic urethra or, sometimes, with extension through the bladder neck into the bladder (Fig. 1). MRI cannot differentiate between benign and malignant tumors.

Malignant Tumors of the Male Urethra Carcinoma of the male urethra is rare, representing less than 1% of all urological cancers, and predominantly concerns men over 50 years old. Tumors of the male urethra are classified according to their location and histological diversity.

Normal Anatomy and Histology of the Female Urethra The female urethra is about 3–4 cm long, nearly equivalent to the length of the male posterior urethra. It extends from the internal urethral meatus at the bladder neck through the urogenital triangle to the external urethral meatus, anterior to the vaginal opening. A sit courses obliquely downward and forward, it is slightly curved with the concavity directed forward and upward. Multiple tiny urethral mucous glands, called the paraurethral glands of Skene, open into the urethral canal. The same histological diversity as in males is found with transitional, cylindrical, and squamous epithelia. The outer portion of the urethra is composed of striated muscle, which, in the upper two thirds of the urethra, is primarily circular and extends proximally to blend with the bladder base. The lower portion of the

Neoplasms, Urethra. Figure 1 Voiding ▶urethrography. Polyp arising from the anterior urethra. (Reprinted with permission from Helenon et al, 2005)

Neoplasms, Urethra

The bulbomembranous urethra is the most frequent site (around 60% of cases), followed by the penile urethra (30%) and the prostatic urethra (10%). Histologically, 80% of male urethral carcinomas are squamous cell carcinomas, 15% are transitional cell carcinomas, and 5% are adenocarcinomas or undifferentiated carcinomas. The histological subtype of urethral cancers also varies according to the anatomical location. The prostatic urethra gives rise to transitional cell carcinomas in 90% of patients and squamous cell carcinomas in 10%. In the bulbomembranous, squamous cell carcinomas develop in 80% of patients, transitional cell carcinomas in 10%, and adenocarcinomas or undifferentiated carcinomas in 10%. In the penile urethra, squamous cell carcinomas represent 90% of the patients and transitional cell carcinomas the remaining 10%. Adenocarcinomas arise from the Littre´ or Cowper glands (Fig. 2). ▶Urethritis secondary to sexually transmitted infectious disease can cause urethral stricture and is considered a risk factor for malignancy. In fact, more than 50% of patients with a carcinoma of the urethra have a history of urethral stricture, which is always symptomatic and the stricture is most frequently located in the bulbomembranous urethra (most common site of urethral carcinomas).

Urethral Cancer Staging System 1. Ta—Noninvasive carcinoma 2. Tis—Carcinoma in situ 3. T1—Tumor is confined to the subepithelial connective tissue

1327

4. T2—Tumor invades the prostate, periurethral muscle or the corpus spongiosum 5. T3—Tumor invades the corpus cavernosum, the bladder neck or beyond the prostatic capsule 6. T4—Tumor invades other adjacent organs 7. N0—No regional lymph node metastasis 8. N+—Evidence of regional lymph node involvement 9. M0—No evidence of distant spread of disease 10. M+—Evidence for distant spread of disease. Male urethral carcinomas can disseminate directly to adjacent structures or metastasize to regional lymph nodes. The lymphatic vessels from the anterior urethra drain into the superficial and deep inguinal lymph nodes and, occasionally, into the external iliac lymph nodes. Tumors of the posterior urethra most commonly spread to the pelvic lymph nodes. On MR images, urethral tumors have signal intensity similar to or lower than that of the surrounding the corpus cavernosum on T1-weighted images and intermediate to low signal intensity on T2-weighted images. On contrast-enhanced MR images, the tumor usually shows mild enhancement. Sometimes, it has very high signal intensity on T2-weighted images due to associated inflammation. MRI can be especially useful for local staging (extension into the tunica albuginea or septa of the corpus cavernosum) and for detection of tumors that invade the root of the penis when physical examination is poorly accessible. Surgical excision is the primary treatment of choice: anterior urethral carcinoma usually has a better prognosis than posterior urethral carcinoma, which is often associated with extensive local invasion and distant metastases.

Malignant Tumors of Female Urethra

Neoplasms, Urethra. Figure 2 Retrograde urethrography. Multiple recurrent tumors of a transitional cell carcinoma. (Reprinted with permission from Helenon et al, 2005)

Carcinoma of the female urethra is four times more frequent than that of the male urethra but still remains an uncommon pathology (less than 0.01% of all carcinomas in women). It predominantly affects women over 50 years old with risk factors, such as chronic irritation, urinary tract infection, and proliferative lesions, such as caruncles, papillomas, adenomas, polyps, and leukoplakia of the urethra. Anterior (40% of cases) and posterior urethral cancers (60% of cases) can be distinguished. Anterior tumors are located in the distal third of the urethra and the lymphatic drainage empties into the superficial and deep inguinal nodes. Local surgical excision is the treatment of choice for these lesions. Because the posterior urethra lymphatic vessels drain into the external iliac, hypogastric and obturator nodes, local spread to the bladder neck and trigone explain

N

1328

Neovascularization

the poor prognosis: fewer than 10% of women are still alive at 5 years with a 4 cm tumor. For such advanced lesions, combined surgery, radiation therapy, and, possibly, chemotherapy is necessary. Two thirds of urethral carcinomas are squamous cell carcinomas, followed by transitional cell carcinomas (20%), adenocarcinomas (10%), undifferentiated tumors and sarcomas (8%), and melanomas (2%).

making biopsy mandatory. The aim of MRI is to help stage the malignant tumor to determine the best strategy of treatment.


Urethral Cancer Staging System 1. Ta—Noninvasive carcinoma 2. Tis—Carcinoma in situ 3. T1—Tumor is confined to the subepithelial connective tissue 4. T2—Tumor invades periurethral muscle 5. T3—Tumor invades the anterior part of vagina or bladder neck 6. T4—Tumor invades other adjacent organs 7. N0—No regional lymph node metastases 8. N+—Evidence of regional lymph node involvement 9. M0—No evidence of distant spread of disease 10. M+—Evidence of distant dissemination. On MR images, the tumor has low signal intensity on T1-weighted images and relatively high signal intensity on T2-weighted images. The size of the tumor is best evaluated on sagittal T2-weighted images. Tumors in the distal urethra may extend into the adjacent perineum and the target-like appearance of the normal urethra on axial T2-weighted images may be disrupted. MR is helpful in evaluating the size, location, and local extension of urethral tumors. Its accuracy to assess local tumor extension has been reported to be 90%.

4.

Pavlica P, Barozzi L, Menchi I (2003) Imaging of male urethra. Eur Radiol 13:1583–1596 Ryu JA, Kim B (2001) MR Imaging of the male and female urethra. Radiographics 21:169–1185 Helenon O, Ramella G, Lapray JF, et al (2005) Ure`tre. In: Imagerie de l’appareil urinaire. Flammarion (ed) 2:898–930 Kawashima A, Sandler CM, Wasserman NF, et al (2004) Imaging of urethral disease: a pictorial review. Radiographics 24:S195–S216

Neovascularization The process of formation of new vessels, such as in tumors. ▶Contrast Media, Ultrasound, Applications in Kidney Tumor

Nephritis ▶Glomerulonephritis

Nephritis, Bacterial, Acute Metastatic Tumors of the Urethra Metastatic tumors are uncommon and mainly concern the male urethra. Bladder carcinomas may spread to the anterior urethra by means of seeding during urethral instrumentation or at cystectomy. The corpus spongiosum can become involved by carcinoma spread from the spermatic cord, testis, prostate and rectum, and can be responsible for extensive urethral narrowing and irregularity. Hematogenous metastases of malignant melanoma and primary prostate, bladder, colonic, testicle or kidney cancer to the corpora cavernosa and corpus spongiosum have been reported. In conclusion, the diagnosis is based on urethrography detecting a nonspecific filling defect. MRI is unable to differentiate between benign and malignant tumors,

Severe renal infection, term initially applied to diabetic or immunocompromised patients, with prolonged clinical course and parenchymal loss. ▶Pyelonephritis, Acute

Nephritis, Interstitial, Chronic Kidney disease that primarily affects the renal interstition and tabules, characterized by tubulointerstitial fibrosis and atrophy, associated with a mononuclear cellular infiltrate. ▶Chronic Pyelonephritis

Nephrocalcinosis and Urolithiasis in Childhood

Nephro- and Urolithiasis ▶Nephrocalcinosis and Urolithiasis in Childhood

Nephroblastoma Alternative name for Wilms’ tumor, which is a cancerous tumor of the kidney in children. ▶Neoplasms, Kidney, Childhood

Nephrocalcinosis Renal lithiasis in which calcium deposits form in the renal parenchyma and result in reduced kidney function and blood in the urine. ▶Stone Disease, Urinary

Nephrocalcinosis and Urolithiasis in Childhood M. R ICCABONA , R. F OTTER Department of Radiology, Division of Pediatric Radiology, Universiy Hospital Graz, Graz, Austria [email protected]

Synonyms Nephro- and urolithiasis; Urinary tract calculus; Urinary tract stone

Definition Nephrocalcinosis (NC) is a microscopic renal tubular or interstitial calcification. Depending on the location a medullary, a cortical or a diffuse type can be differentiated. Note that in a variety of diseases urolithiasis and NC may occur simultaneously, with urolithiasis being a sequelae of the underlying disease that causes NC. Urolithiasis is defined as macroscopic calcification in the urinary collecting system. Urinary stones are composed of crystal agglomerations, sometimes mixed with proteins that form at the papillae, the epithelium or in the collecting system.

1329

Embryology and Pathogenesis Causes are infection (particularly in children), hypercalciuria, cystinuria or oxaluria (potentially familial), as well as impaired inhibitor function such as low citrate excretion, some (diuretic) drugs and various metabolic disorders (e.g. Bartter’s syndrome). Incidence varies widely with a large geographic distribution (more common in the far and middle east), with some age (typical medullary changes in pre-term newborns), gender (boys affected more often than girls), composition (uric acid stones are extremely rare in children) and race variation (less often in black people). Urinary calculi can occur in the kidney, the ureter and the urinary bladder, the latter—as a primary condition—becoming rare. NC and urolithiasis is less common in children than in adults. They should always give reason for a thorough investigation (metabolic causes, familial disease, underlying renal or urinary tract condition) (Table 1).

Clinical Presentation Nephrolithiasis and urinary tract calculi as well as NC may be clinically asymptomatic; if stones cause obstruction they present with pain, colic, (micro-) haematuria, (secondary) urinary tract infection. In nephrolithiasis clinical presentation depends on the underlying condition and the stage of the disease, ranging from asymptomatic patients to renal insufficiency and urolithiasis with all its potential symptoms.

Imaging Imaging usually relies on detailed ultrasound (US) as the initial study. There NC is seen as an increased echogenicity of the affected structure that may start with just an increased echogenicity of the cortico-medullary junction, with an inversion of the normal renal parenchymal pattern in medullary NC, an increased cortico-medullary differentiation with a very echogenic cortex in cortical NC, or a diffusely increased echogenicity with loss of differentiation in the diffuse type (Fig. 1). Using the location and amount of echogenicity changes (that correspond with the degree of the disease) US offers a grading possibility (Table 2). In long-standing disease, these areas may eventually become very echogenic and exhibit a stone-like appearance, obviating analysis of the deep renal structures. Stones—if calcified—sonographically are echogenic and cause posterior acoustic shadowing as well as (often) a twinkling artefact on colour Doppler sonography (CDS) (Fig. 2). Note that for visualisation of a calculus in the

N

1330


Nephrocalcinosis and Urolithiasis in Childhood. Table 1a Causes for ▶nephrocalcinosis and urolithiasis. Common causes of nephrocalcinosis and important differential diagnosis Nephrocalcinosis Medullary

Cortical

Differential diagnosis

Common causes ACTH therapy Adrenal insufficiency Bartter’s syndrome Bone metastases Cushing syndrome Hypercalciuria Hyperoxaluria Hyperparathyroidism Hyper-, Hypothyroidism Idiopathic hypercalcaemia Lipoidnecrosis Lesch-Nyhan syndrome Lowe’s syndrome Malignant neoplasm Medication: furosemide, dexamethasone Medullary sponge kidney long time parenteral nutrition, ascorbic acid supplementation d-RTA Tyrosinaemia Sarcoidosis, other granulomatous diseases Sickle cell disease Vitamin D, A intoxication William’s syndrome, Wilson’s disease Chronic hypercalcaemia Lipoidnecrosis Ethylene glycol intoxication Primary hyperoxaluria Sickle cell disease Acute cortical necrosis Alport syndrome Chronic glomerulonephritis Kidney transplant rejection Pyelonephritis Renal tuberculosis Renal vein thrombosis Tamm-Horsfall depositions

distal ureter a sufficiently extended urinary bladder is mandatory. Assessment of the ureteric bladder inflow jet (by CDS) can be helpful for differentiating complete from partial obstruction (Fig. 3). In a significant acute

Nephrocalcinosis and Urolithiasis in Childhood. Table 1b Metabolic disturbances associated with urolithiasis and/or nephrocalcinosis Hypercalciuria Normocalcemic Idiopathic hypercalciuria hypercalciuria d-RTA Diuretics—Furosemide Bartter’s syndrome Wilson’s disease, Lowe’s syndrome Hypercalcemic Primary hyperparathyroidism hypercalciuria Immobilisation Hyperthyroidism Hypothyroidism Cushing syndrome—ACTH therapy Adrenal insufficiency Hypervitaminosis D, A Idiopathic hypercalcaemia of childhood Hyperoxaluria Primary hyperoxaluria type I, II, III Secondary hyperoxaluria: Malabsorption syndromes Lack of intestinal Oxalobacter formigenes Short bowel syndrome Dietary Cystinuria Type I, II, III Hyperuricosuria Inborn errors of metabolism Lesch–Nyhan syndrome Glycogen-storage diseases, type I, III, V, VII Overproduction in Leukaemia Non-Hodgkin’s lymphoma High-protein diet Hypocitraturia d-RTA Idiopathic Source: Adapted from Benz-Bohm G, Hoppe D (2001) In: Fotter R (ed) Pediatric Uroradiology, Springer Berlin–Heidelberg–New York, pp 281–286, Table 17.2, p 283 and Table 17.5, p 287.

obstruction, the resistance index is asymmetrically elevated in the parenchymal arteries of the affected kidney, despite of an often only minimally dilated collecting system of a swollen, enlarged, slightly hyperechoic and hazy differentiated kidney. Non-calcified calculi may appear less echogenic and cause no or little dorsal shadowing—particularly infectious and uric acid ‘stones’, which in children are seen more commonly in combination with other pre-existing urinary tract malformations or systemic/metabolic disease.


1331

Nephrocalcinosis and Urolithiasis in Childhood. Figure 1 Nephrocalcinosis (NC)—ultrasound (US) appearance. Note the echogenic papilla and (outer) medulla, particularly the rim-like hyperechoic appearance of the cortico-medullary junction.

Nephrocalcinosis and Urolithiasis in Childhood. Table 2 Sonographic grading of medullary nephrocalcinosis (Dick et al. 1999) Grade I

Mild increase in echogenicity around the border of the medullary pyramids Grade II Mild diffuse increase in echogenicity of the entire medullary pyramids Grade III Greater, more homogeneous increase in the echogenicity of the entire medullary pyramid Source: Adapted from Benz-Bohm G, Hoppe D (2001). In: Fotter R (ed) Pediatric Uroradiology, Springer Berlin–Heidelberg–New York, pp 281–286, Table 17.3, p. 284

US is usually supplemented by an abdominal plain film focused at the kidney, ureter and bladder region (‘KUB’). Age-adapted exposure and film-screen combinations should be chosen. It demonstrates calcification projecting on the urinary tract structures. Intravenous urography (IVU) is hardly necessary in children, only in some rare cases, where US and KUB cannot sufficiently diagnose the condition it may become indicated. If applied, age-adapted contrast dose and selected number of individually timed exposures should be used for a tailored investigation to reduce radiation burden. In difficult cases or complex disease, spiral (multislice) CT is the best method for looking at urolithiasis and particularly ureteral stones. Usually un-enhanced low

N Nephrocalcinosis and Urolithiasis in Childhood. Figure 2 Twinkling artefact in urolithiasis. (a) Axial view of grey scale US depicts the echogenic calculus in the renal pelvis (but only little shadowing). (b) Twinkling colour signals on colour Doppler sonography (CDS) in the same view caused by the calculus.

dose protocols suffice. Only for differentiation of tumourlike lesions with calcifications (e.g. xanthogranulomatous pyelonephritis, tuberculosis) intravenous iodinated contrast is recommended. MR has currently no role in investigating nephrolithiasis and NC in children.

Diagnosis Diagnosis is made by imaging, with functional information important for clinical decision making, particularly in urolithiasis with colic and acute obstruction. Usually a detailed US study of a properly hydrated urinary tract at sufficiently extended bladder supplemented by a KUB-film—together with clinical and laboratory information—suffices for making the diagnosis in

1332


Nephrocalcinosis and Urolithiasis in Childhood. Table 3 Diagnostics in urolithiasis and nephrocalcinosis Patient history

Nephrocalcinosis and Urolithiasis in Childhood. Figure 3 Colour Doppler sonography (CDS) of the bladder in urolithiasis. (a) Normal ureteral bladder inflow jet from the left side demonstrated on CDS in an infant with an obstructing distal ureteral calculus of the right ureter, with no depictable urine inflow from right side. (b) Note the twinkling pre-ostial ureteral calculus in the distal right ureter just at the ostium.

infants and children and reveals all treatment relevant information (Table 3). Rarely IVU or CT may become necessary. Follow-up—particularly in NC—is usually performed by US; and often US appearance persists even when the underlying disease is treated and has diminished or ceased.

Bibliography 1. 2.

3.

Alon US (1997) Nephrocalcinosis. Pediatrics 9:160–165 Benz-Bohm G, Hoppe D (2001) Urolthiasis and nephrocalcinosis. In: Fotter R (ed) Pediatric Uroradiology, Springer Berlin– Heidelberg–New York, pp 281–286, Table17.2, p 283, Table 17.5, p 287, Causes for nephrocalcinosis and Urolithiasis Cremin B, Wiggelinkhuizen J, Bonnici F (1982) Nephrocalcinosis in children. Br J Radiol 55:413–418

Familial stone disposition, stone recurrence

Stone localisation US, abdominal X-ray, IVU, Spiral-CT Lab Electrolytes, uric acid, creatinine, urea, Mg, PO4, plasma oxalate, acid base status, AP Urine Culture, sediment 24 h urine ) Promotors and Inhibitors Renal function Clearance Diet Daily fluid intake +, meat *, milk ** Ask for Diuretics, ACTH Drugs Vitamin D, A, C overdose Allopurinol Chemotherapy Inherited Cystinuria Metabolic Primary hyperoxaluria Disorders Xanthinuria 2,8-Dihydroxyadeninuria D-RTA Dent’s disease Lesch–Nyhan syndrome Wilson’s disease Bartter’s syndrome William’s syndrome Lowe’s syndrome Chronic Malabsorption syndromes (e.g. Cystic fibrosis) Diseases Steatorrhoea Celiac disease Short bowel syndrome Immobilisation Hypercalciuria Stone analyses Infrared-spectroscopy, X-ray diffraction Differential e.g. Appendicitis diagnosis Imaging Primarily: US Supplemented by KUB Selected cases: IVU, CT e.g. complicated, unclear, XPN DD Tumour, pre-operative Rare cases: Nephrostomy + Pelography e.g. complete obstruction, bridging infected(pyonephrosis), urosepsis Source: Adapted from Benz-Bohm G, Hoppe D (2001). In: Fotter R (ed) Pediatric Uroradiology, Springer Berlin–Heidelberg–New York, pp 281–286, Table 17.4, p 286 4. 5.

Dyer RB, Chen MYM, Zagoria RJ (1998) Abnormal calcifications in the urinary tract. Radiographics 18:1405–1424 Riccabona M, Lindbichler F, Sinzig M (2002) Conventional imaging in paediatric uroradiology. Eur J Radiol 43:100–109

Nerves, Cranial

Nephrolithiasis (Renal Lithiasis) The presence of calculi in the kidney. ▶Stone Disease, Urinary

Nerves, Cranial TAREK A. YOUSRY, I NDRA YOUSRY Institute of Neurology, London, UK [email protected]

Definitions Cranial nerves are the means by which the brain receives information from, and controls the activities of the head, neck and to a lesser extent the thoracic and abdominal viscera. The cranial nerves (CN) consist of 12 pairs of which the CN III–XII are genuine peripheral nerves. The olfactory tract and the optic nerve are parts of the brain, because their fibres are medullated by oligodendrocytes (central glial cells). The other cranial nerves (CN III– XII) have segments of variable length covered by central glia. These segments are considered to be part of the brain and are known as the root entry or exit zones. These zones are important as symptomatic neurovascular contacts only occur at their site. The central segments of the motor nerves are shorter than those of the sensory nerves; the longest central segment however, is found in CN VIII. The olfactory nerves (CN I) are the only sensory cranial nerves that project directly to the cerebral cortex rather than via the thalamus. They arise from nerve cells within the olfactory mucosa that coats the superior part of the middle and lateral walls of the nasal cavity. They pass through the cribriform plate of the ethmoid bone and terminate in the olfactory bulbs (Fig. 1a). From here the olfactory tract passes posteriorly and terminates by forming three striae: the lateral, the intermediate and the medial olfactory striae. The optic nerve (CN II) is constituted by a great number of nervous fibres arising from cells in the retina which converge towards the optic disc and, pass through the sclera and choroid membranes, and as they emerge from the eyeball form the optic nerve. CN II passes through the orbital cavity, optic canal and ends at the optic chiasm (Fig. 1a). The oculomotor nerve (CN III) is a motor nerve. Its nuclei are situated at the level of the superior colliculus. The nerve emerges along the internal margin of the

1333

cerebral peduncle, passes through the interpeduncular fossa to enter the lateral wall of the cavernous sinus (Fig. 1b). It then passes through the superior orbital fissure into the orbit. It innervates the superior, inferior and medial recti as well as the levator palpebrae. CN III also carries parasympathetic fibres to the sphincter ciliary muscles of the pupil. The trochlear nerve (CN IV) is a motor nerve. Its nucleus is situated at the level of the inferior colliculus. It crosses the midline and exits at the posterior surface of the mesencephalon below the inferior colliculus (Fig. 1c). It then travels around the mesencephalon, enters the lateral wall of the cavernous sinus, passes through the superior orbital fissure into the orbit where it innervates the superior oblique muscle. The trigeminal nerve (CN V) has motor and sensory nuclei. The motor nucleus is located in the pons, the sensory nuclei extend in the brainstem from the mesencephalon to the cervical spinal cord. CN V emerges from the pons with a large sensory root and small superior and inferior motor roots (Fig. 2a). All roots enter the trigeminal cavity (Meckel’s cave) where the sensory root continues into the trigeminal ganglion and then provides three main branches. The ophthalmic nerve (V1) enters the lateral wall of the cavernous sinus, and passes through the superior orbital fissure into the orbit. The maxillary nerve (V2) enters the lateral wall of the cavernous sinus, and passes through the foramen rotundum into the pterygopalatine fossa. The mandibular nerve (V3) carries in addition to the sensory fibres the motor fibres of the trigeminal cavity. It passes through the foramen ovale. The abducens nerve (CN VI) is a motor nerve. Its nucleus is situated at the lower level of the pons. CN VI exits the brainstem at the pontomedullary junction (Fig. 2b). It then travels through the pontine cistern, enters the cavernous sinus through Dorello’s canal, and passes through the superior orbital fissure into the orbit where it innervates the lateral rectus muscle. The facial nerve (CN VII) has motor and sensory nuclei. The motor nucleus is located at the floor of the 4th ventricle. The motor fibres for the viscera (parasympathetic) arise from the superior salivatory nucleus and the lacrimal nucleus. The visceral sensory fibres terminate in the nucleus solitarius in the dorsal medulla oblongata. The motor (CN VII) and the sensory (CN VIIb, intermediate) nerve roots exit the brainstem at the pontomedullary junction and enter the internal auditory meatus (Fig. 2c). CN VII then enters the facial canal and exits the skull through the stylomastoid foramen. The parasympathetic fibres of the lacrimal nucleus and visceral sensory fibres pass as the greater petrosal nerve through the Vidian canal. The vestibulocochlear nerve (CN VIII) is a sensory nerve which consists of the cochlear and the (superior and

N

1334

Nerves, Cranial

Nerves, Cranial. Figure 1 CISS images showing the olfactory nerve (arrow, a), optic nerve (arrowhead, a), oculomotor nerve (arrow, b) and trochlear nerve (arrow, c).

inferior) vestibular nerves (Fig. 2c, d). The cochlear as well as the vestibular nuclear complex are located in the dorsolateral side of the brainstem at the level of the pontomedullary junction. The medial vestibular nucleus bulges into the 4th ventricle as the ‘vestibular area’. The nerve enters the pons at this level coming from the internal auditory canal. The glossopharyngeal nerve (CN IX) is a mixed nerve. The motor fibres arise from the ambiguous nucleus located in the medulla oblongata. The motor fibres for the viscera (parasympathetic) arise from the inferior salivatory nucleus. The visceral sensory fibres terminate in the nucleus solitarius in the dorsal medulla oblongata. CN IX emerges from the posterolateral sulcus of the medulla oblongata, traverses the premedullary cisterns and exits the skull through the anterior compartment of the foramen jugulare. The vagus nerve (CN X) is a mixed nerve. The motor fibres arise from the ambiguous nucleus in the lateral

medulla oblongata. The motor fibres for the viscera arise from the dorsal nucleus of the vagus nerve, which bulges into the 4th ventricle as the vagal trigone. The visceral sensory fibres terminate in the nucleus solitarius in the dorsal medulla oblongata. CN X emerges from the posterolateral sulcus of the medulla oblongata and exits the skull together with the nervus accessorius through the intermediate space of the foramen jugulare (Fig. 3a). The accessorius nerve (CN XI) is a motor nerve with two roots. The cranial root arises from the nucleus ambiguous and exits the medulla oblongata anterior to the posterolateral sulcus (Fig. 3b). The spinal root arises from the spinal accessory nucleus in the lower medulla and the upper five spinal cord segments. It courses between the dorsal and ventral roots of the upper cervical nerves, enters the foramen magnum and joins the cervical root in the intermediate portion of the jugular foramen. The hypoglossal nerve. (CN XII) is a motor nerve. Its nucleus is located in the posterior medulla oblongata,

Nerves, Cranial

1335

N Nerves, Cranial. Figure 2 CISS images showing the trigeminal nerve (arrow, a), abducens nerve (arrow, b), facial nerve (arrow, c) and superior vestibular nerve (arrowhead, c) and cochlear nerve (arrow, d) and inferior vestibular nerve (arrowhead, d).

Nerves, Cranial. Figure 3

CISS images showing the vagus nerve (arrow, a) and accessory nerve (arrow, b).

1336

Neuralgia

causing a bulging of the floor of the 4th ventricle, the hypoglossal trigone. The nerve rootlets emerge from the preolivary sulcus, pass through the premedullary cistern, and pass through the hypoglossal canal as one trunk.

Imaging The visualisation of the cranial nerves is a function of the thickness of the nerves and of anatomic properties of the segment that is to be identified. Anatomically the cranial nerves III–XII can be divided into the following segments: (i) intramedullary,(ii) cisternal, (iii) foraminal and (iv) extracranial segments. Nuclei. The cranial nerve nuclei are currently not accessible to conventional in-vivo MR imaging. Some nuclei however, can be identified through the bulges they cause in the floor of the 4th ventricle. These are the nuclei of CN VI (facial colliculus), CN VIII (vestibular area), CN X (vagal trigone) and CN XII (hypoglossal trigone). These bulges are important landmarks for the identification of the corresponding nuclei. Intramedullary segments. The intramedullary segments are currently not accessible to conventional in-vivo MR imaging. However, tractography has the potential to visualise those segments in the future. Cisternal segments. The visualisation of the various cranial nerves in their cisternal course is determined by, among other factors, the anatomic properties of each individual nerve such as the nerve’s diameter, the course and the amount of surrounding cerebrospinal fluid (CSF). The small diameter of some of these nerves is in many instances the main reason for the difficulties encountered in visualising them. The larger nerves can be well visualised with the regularly used T2 FSE sequences. The smaller nerves can be usually identified with that sequence, though the reliability increases with the use of sequences with a cisternographic effect (3D CISS, 3D DRIVE and FIESTA). Some nerves, for example the trochlear nerve, can only be identified when using these sequences. A problem encountered when using the CISS, 3D DRIVE and FIESTA sequences is the differentiation of nerves and vessels. This can be achieved by combining the 3D CISS sequence with a plain and an enhanced TOF sequence. The comparison of the source images with the 3D CISS images in the appropriate plane will allow the differentiation of nerves, arteries and veins. It is this combination that enabled the reliable identification of the smallest cranial nerves, CN IV. This combination of sequences also enabled for the first time the reliable identification of an arterial neurovascular contact at the root exit zone of the trochlear nerve in patients with ‘superior oblique myokymia’, thus supporting the neurovascular compression hypothesis.

The 3D CISS, 3D DRIVE and FIESTA sequences, are therefore the standard sequences to be used in the evaluation of cranial nerves. A further improvement in imaging resolution can be achieved at higher field strength such as 3T. Foraminal segments. The foraminal segments of CN V1, V2, V3, IX–XII as well as the cavernous segments of CN III, IV, V1 and VI are best visualised on MR using contrastenhanced 3D CISS, MPRAGE or TOF sequences. The evaluation of the bony structures, however, is best performed by high resolution computed tomography (HRCT).

Bibliography 1. 2.

3.

Leblanc A (2001) Encephalo-peripheral nervous system: anatomy, imaging, vascularisation. Springer Verlag, Berlin, Heidelberg Yousry I, Camelio S, Schmid UD et al (2000) Visualization of cranial nerves I–XII: value of 3D CISS and T2-weighted sequences. Eur Radiol 10:1061–1067 Yousry I, Dieterich M, Naidich TP et al (2002) Superior oblique myokymia magnetic resonance imaging support for the neurovascular compression hypothesis. Ann Neurol 51:361–368

Neuralgia ▶Spinal Nerve Roots, Clinical Syndromes

Neuroarthropathy ▶Neuropathic Joint Disease

Neuroblastoma J OHAN G. B LICKMAN 1 , D EWI A SHI 2 1

Department of Radiology, UMC St Radboud, Nijmegen, Netherlands [email protected] 2 Department of Radiology, University of Indonesia, Jakarta Pusat, Indonesia

Definition Malignant round cell tumor arising from the ▶adrenal medulla (commonest), premature neuroblasts of the embryonic neural crest and sympathetic ganglia.

Neuroblastoma

Incidence The third most common malignancy of childhood after leukemia and primary brain neoplasm. Neuroblastoma accounts for about 10% of all pediatric neoplasms and is the commonest extracranial solid tumor. In the majority of cases, it presents between 2 months and 2 years of age, but the lesion may also present in the neonatal period. The tumor is slightly more common in boys than girls, with familial incidence reported.

Pathology/Histopathology Neuroblastoma is composed of small round cell that characteristically arranged in rosettes. ▶Ganglioneuroma represents the most differentiated end of the spectrum and is benign. ▶Ganglioneuroblastomas represents an intermediate mixed group. Approximately two-thirds of the tumors are located in the abdomen, followed by tumors arising in the paravertebral sympathetic chain or the presacral soft tissue region. It can spontaneously transform into more benign ganglioneuroma(blasto)ma. The International neuroblastoma staging system (INSS) defines the significance of tumor resectability, anatomic “midline,” and lymph node involvement (Table 1).

Clinical Presentation Neuroblastoma is clinically silent until it invades adjacent structures. Symptoms may include bone pain, fever, Neuroblastoma. Table 1 The international neuroblastoma staging system (INSS) Stage I IIA IIB III IV IVS

1337

weight loss, anemia, and hypertension (10%). Two-thirds demonstrate excess of urinary catecholamine excretion and present with flushing, sweating, and irritability. Leg edema due to tumor compression of veins may occur. Paraneoplastic syndromes may occur and present with myoclonus and watery diarrhea. Horner syndrome can be a clinical presentation of a cervical tumor location. Syndromes associated with neuroblastoma include Beckwith-Wiedemann, Klippel–Feil, and fetal alcohol.

Imaging Conventional radiographs may reveal a mass that in twothird of the cases contains calcification (stippled, diffuse, amorphous). Widening of the interpedicular distance and paravertebral mass may be seen on plain abdominal X-ray. Ultrasound (US) is used as the screening modality for suspected abdominal pathology and shows the organ of origin, liver metastases, and identifies tumor encasement of vessels. The tumor is inhomogeneous on US. It has hypoechoic regions due to hemorrhage, necrosis, or cyst formation and hyperechoic regions due to calcification (Fig. 1). Mainly cystic lesions may be present particularly in neonates. CT is superior to US in defining the extent and retroperitoneal spread of the primary tumor (Fig. 2). It reveals tumor calcification in about 85% of the patients. The lesions have an attenuation equal to or less than that of the muscle. Enhancement is heterogeneous after contrast injection. MRI has advantages over CT in that extension to the vertebrae canal, bone marrow involvement, encasement of the superior mesentery artery (SMA) and other vessels can be better evaluated.

Description Localized tumor confined to the area of origin complete gross excision With or without microscopic residual Unilateral tumor with incomplete gross excision; ipsilateral and Contralateral lymph node negative microscopically Unilateral tumor with complete or incomplete gross excision with positive ipsilateral regional lymph node identifiable, but the contralateral lymph nodes are negative microscopically. Tumor crosses the midline Distant metastases Metastatic disease confined to liver, skin, and bone marrow, with the primary tumor stage 1 or 2. (Age 20 HE with sensitivity 98%, specificity 73%), while benign lesions show slower and lower enhancement. The latter appears to be a useful criterion for excluding malignancy when using a threshold value of 3 cm) with a focal discrepancy in the calibre of the bowel at the transition between obstructed and non-obstructed bowel (Fig. 2). In paralytic ileus, the small bowel is also dilated, but the right colon is usually also dilated, often tapering to normal at, or distal to, the hepatic flexure (4).

Determining the Level of Obstruction On PFA, the presence of dilated small bowel superiorly in the abdomen may indicate proximal SBO. Dilated pelvic loops more frequently imply distal SBO. In SBO, dilated small bowel loops align along the long axis of the mesentery and jejunal loops can be found in the pelvis, with ileal loops in the upper abdomen. In order to distinguish large bowel obstruction from SBO on PFA, the following features are helpful: Megibow (4) describes the systematic assessment of the bowel on CT in bowel obstruction. The bowel should be systematically evaluated in a retrograde way, beginning at the rectum, and proceeding to the caecum. After

Diagnosis Confirming the Diagnosis of Obstruction

Small bowel obstruction

The PFA is diagnostic of SBO in up to 64% of cases. Dilated loops of bowel are seen, usually within 5 h of the onset of obstruction (Fig. 1). If the dilated loops are entirely fluidfilled, then the diagnosis may be occult on PFA. Dilated bowel loops can also be seen in paralytic ileus.

Haustra absent Valvulae in jejunum Many loops Central distribution

Large bowel obstruction Haustra present No valvulae Few loops Peripheral distribution (continued)

Obstructive Uropathy in Childhood. Figure 1 PFA of SBO. Multiple dilated loops of small bowel shown.

Obstructive Uropathy in Childhood. Figure 2 SBO due to adhesions. ▶Zone of transition in distal small bowel (arrow). Note normal bowel wall enhancement (curved arrow).

O

1376

Occlusion and Subocclusion, Small Bowel in Adults

‘clearing’ the colon, the distal ileal loops should be identified and assessment should proceed proximally to the level at which the calibre of the small bowel is increased (Fig. 2). Review of the images on cine-mode and multiplanar reconstructions [particularly with multidetector-row CT (MDCT)] are helpful in this evaluation.

Diagnosis of the Cause of Obstruction Adhesions, hernias and neoplasms account for over 80% of cases. CT is the primary diagnostic tool in determining aetiology of SBO, with accuracy of 70–95% (1). In developed countries, adhesions account for 50–80% of causes. A history of prior abdominal surgery is usual, although previous peritoneal inflammation may also cause adhesions. Even in patients with known intraabdominal malignancy, SBO is due to adhesions in 21–38% (4). On CT, diagnosis of adhesions is based on an abrupt change in small bowel calibre without another cause of obstruction at the transition point. A beak-like narrowing can be seen. Although the adhesions cannot be visualised with CT, it is important to make the diagnosis of obstruction secondary to adhesions when no other cause is seen (4). Hernias account for 10–15% of cases (1). External hernias are more common than internal and include inguinal, femoral, paraumbilical, obturator, Richter, Spigelian and incisional hernias. Internal hernias include paraduodenal hernias, transmesenteric hernias, and herniation through the foramen of Winslow. These can be recognised by an abnormally positioned segment of small bowel often with protrusion of mesentery through the internal defect, with proximal bowel dilatation. With increasing use of laparoscopic surgery, port-site hernias are a cause of SBO to recognise. Laparoscopic port sites vary in diameter, generally from 5 to 10 mm. Herniation through laparoscopic fascial defects as small as 5 mm has been described. SBO due to port-site hernias is more common in the early post-operative period (5). Neoplasia is the third most common cause of SBO. Infiltration of the bowel wall may be caused by adenocarcinoma, carcinoid or metastatic carcinoma. Extrinsic masses such as peritoneal carcinomatosis can also cause obstruction. In these cases a mass is seen at the transition point. Mural mass lesions may also lead to obstruction by causing intussusception.

Strangulation is defined as CLO associated with intestinal ischaemia. In CLO (Fig. 3), a loop of bowel is occluded at two adjacent points along its course. CLO usually occurs due to constriction by adhesions or a hernia. The CT signs of CLO are . Radial distribution of small bowel loops with mesenteric vessels converging toward apex . U or C-shaped dilated bowel loop . Two adjacent collapsed or triangular loops at the site of obstruction CLO is prone to twisting of the affected bowel and mesentery leading to venous congestion with subsequent arterial ischaemia and infarction. This is termed strangulation. The CT signs of strangulation are . Circumferential mural thickening . ‘Target’ sign—concentric mural rings of different attenuation, post-intravenous contrast, a more specific form of mural thickening . ‘Whirl’ sign—corkscrew configuration of mesenteric vessels due to twisting . Increased attenuation of bowel wall . Lack of bowel wall enhancement . Air in bowel wall (intestinal pneumatosis) . Mesenteric haemorrhage . Ascites In SBO complicated by ischaemia, the bowel wall appearance depends on the balance between venous obstruction, reduced arterial perfusion and reperfusion. If reduced arterial perfusion predominates, in the absence

Assessing for Evidence of Strangulation or Mesenteric Ischaemia The presence of ischaemia in SBO increases mortality from between 5 and 8% to between 20 and 37% (4). It requires urgent surgical management. CT is 83% sensitive and 93% specific for ischaemia in SBO.

Occlusion and Subocclusion, Small Bowel in Adults. Figure 3 Closed-loop obstruction. C-shaped loop of bowel. Note congested mesentery (arrow) and poor enhancement of bowel wall (curved arrow).

Occlusion, Artery, Femoral

of significant reperfusion, then the bowel wall will be of normal thickness and many of the above signs will be absent. Therefore, CT is good at ruling in ischaemic SBO, but poor at ruling out ischaemic SBO (3).

1377

Definition An arterial obstruction of the infrainguinal arteries is an occlusion or narrowing of an arterial segment between the groin and the lower limbs.

Determining the Need for Surgery The impact of imaging on management and indications for surgery have been summarised by Taourel et al. (1).


1. 2. 3. 4.

Occlusions are complete obstructions of the infrainguinal arterial lumen and are due to thrombus formation, mainly on preexisting atherosclerotic plaque. In stenoses, the lumen is still patent but narrowed, with a diameter reduction of more than 50% causing symptoms. Alternatively, acute occlusions occur because of thrombotic emboli from the heart.

Strangulation and ischaemia (emergency) Obstruction due to a tumour Obstruction due to irreducible hernia (urgent) Obstruction due to adhesions, without ischaemia, should be managed conservatively, where possible

As stated above, if SBO is slow to resolve with conservative management, WSFT is a useful test to determine the likelihood of successful non-operative treatment (2).

Bibliography 1.

2.

3.

4. 5.

Taourel P, Kessler N, Lesnik A et al (2002) Non-traumatic abdominal emergencies: imaging of acute intestinal obstruction. Eur Radiol 12:2151–2160 Staunton M, Malone DE (2005) Can diagnostic imaging reliably predict the need for surgery in small bowel obstruction? Critically appraised topic. Can Assoc Radiol J 56:79–81 Staunton M, McNamara A, Maher MM et al (2000) The application of evidenced based medicine (EBM) to radiology: how valuable is computed tomography (CT) in the diagnosis of bowel obstruction? RSNA Conference Proceedings. Radiology 217(S)169 Megibow AJ (1994) Bowel obstruction. Evaluation with CT. Radiol Clin North Am 32:861–870 Tonouchi H, Ohmori Y, Kobayashi M, Kusunoki M (2004) Trocar site hernia. Arch Surg 139:1248–1256

Occlusion, Artery, Femoral D IERK VORWERK Klinikum Ingolstadt GmbH Institut fu¨r diagnostische und interventionelle Radiologie, Ingolstadt, Germany [email protected]

Synonyms Femoral arterial obstruction; Femoral arterial occlusion; Infrainguinal arterial obstruction; Infrainguinal arterial occlusion; Popliteal arterial obstruction; Popliteal arterial occlusion

Clinical Presentation Clinical indications for treatment include claudication, pain at rest, and nonhealing ulceration. The severity of the symptoms depends on the acuteness of the occlusion, the condition of collateral pathways, the lesion’s location, and concomitant disease such as diabetes or renal insufficiency.

Imaging Many imaging modalities allow the physician to diagnose infrainguinal artery obstruction. Color-coded duplex sonography as well as magnetic resonance angiography (MRA), computed tomography (CT), computed tomography angiography (CTA), and intra-arterial angiography are useful tools for detecting the location and extent of an iliac obstruction.

Nuclear Medicine Nuclear medicine plays no particular role in the diagnosis of aortic stenosis.

Diagnosis Diagnosis is reliably achieved by angiography or MRA. Detection by duplex sonography is a reliable tool for the femoropopliteal segment but is sometimes limited for detection and exact lesion description in the lower limbs. In stage IV (Fontaine) disease, MRA may be limited because of low arterial flow, movement artifacts, or artifacts due to superimposed venous flow.

O

1378


Interventional Radiological Treatment Endovascular Versus Surgical Treatment Endovascular therapy is known to be of low invasiveness with good technical success, achieving fair overall patency. In data taken from eight publications reporting on 1,469 procedures, the weighted average technical success of femoropopliteal endovascular interventions was 90%, the complication rate was 4.3%, and the 3-year patency rate was 51%. Stents did not improve patency, showing a patency of 58% after 3 years (1). Surgery offers acceptable results for distal reconstruction: An average 5-year patency of 80% for vein bypasses and 65–75% for expanded polytetrafluoroethylene (ePTFE) bypasses has been reported. The combined mortality and amputation risk was calculated to be about 2.2% for aortobifemoral reconstructions and 1.4% for femoropopliteal reconstructions (1).

Location of Lesion Claudication is mainly related to lesions in the aortoiliac and femoropopliteal regions. It is unlikely to be due to infrapopliteal lesions, and there is general agreement that treatment below the knee should be strictly limited to patients with critical limb ischemia, that is, stages III and IV (Fontaine) or categories 4–6 (Rutherford).

Type of Lesion The morphology of a treated lesion influences the technical outcome, follow-up results, and also the risk of treatment. Therefore, the TransAtlantic Inter-Society Consensus (TASC) document introduced a classification system that tries to categorize lesions with regard to their accessibility to either percutaneous treatment or surgery: type A lesions, which are ideal for a percutaneous approach; type B lesions, in which the percutaneous approach is still the preferred technique; type C lesions, in which a surgical approach should be preferred; and type D lesions, in which surgery is the option of choice. The TASC classification overrides older classifications because it takes into account all of the available and published techniques (including stent technology), which offer a much wider variation of treatment as well as effective tools to deal with acute complications of balloon angioplasty, such as occluding dissection and vascular rupture. If we consider percutaneous therapy as the preferred method to treat those patients presenting with mild or moderate claudication, treatment might be offered to those presenting with type A and B lesions but should be discussed in depth with patients with type C lesions, as the risk and the potential benefit of treatment are related to the underlying morphology.

In the femoropopliteal field, type A lesions are single stenoses up to 3 cm in length not involving the very proximal superficial femoral and the distal popliteal artery. Type B lesions include stenoses 3–5 cm in length, heavily calcified stenoses, multiple lesions (each up to 3 cm), and lesions with no sufficient tibial run-off (the latter are unlikely to meet the criteria of mild or moderate claudication). Type C lesions are classified as stenoses or occlusions longer than 5 cm and multiple midsize lesions (3–5 cm). Total common femoral, superficial femoral, and popliteal occlusions are classified as type D lesions. There was some dissenting discussion on the definition of type B lesions: Interventional radiologists represented by the Cardiovascular and Interventional Radiological Society of Europe wished to express their assumption that even longer lesions of up to 10 cm may be justified as being classified type B instead of type C; they claimed that the reported results were mainly due to underdeveloped techniques and instruments that have substantially improved and that no data exists comparing the efficacy of percutaneous transluminal angioplasty (PTA) versus bypass surgery for lesions between 4 cm and 10 cm. (See Figs 1 and 2) Other than in the iliac area, few femoral lesions meet the criteria for types A and B lesions, especially if limited to 5 cm in length. Thus, few patients with mild or moderate claudication due to femoropopliteal lesions will be ideal candidates for percutaneous treatment. Moreover—without limiting the importance of the TASC document, which certainly means a step forward in the joint approach to peripheral vascular disease—the morphological classification does not take into account some technical considerations that depend on the age and composition of a lesion. Particularly in femoral occlusions, the degree of organization of the occluding thrombus or the composition of the lesion with the original stenosis at the proximal and distal ends or in the middle are factors that are not very predictable but may influence the technical outcome of the intervention or its complication rate. (For instance, distal embolization might aggravate symptoms.) Other than in the iliac arteries, the liberal use of stents and stent grafts may help overcome a failed balloon angioplasty and resolve the technical outcome, but it does not achieve an improved long-term efficacy, and it may start a lifetime dependency on recurrent interventional or surgical procedures. These associated potential drawbacks have to be carefully balanced against the potential benefits and need to be discussed in depth with the patient before treatment is performed, especially in association with mild or moderate claudication. These considerations restrict the use of endovascular treatment in femoropopliteal lesions to mainly stages IIb and IIa patients presenting with type A or less pronounced type B lesions.


Occlusion, Artery, Femoral. Figure 1 Popliteal artery stenosis. (a) Short-segment stenosis at the level of the joint. The stenosis is eccentric, which sometimes indicates insuffient reaction to percutaneous transluminal angioplasty (PTA). Stenting in that region is problematic because it is in the flexure zone of the popliteal artery. If needed as a bail-out treatment, a very flexible stent is preferred. (b) After PTA, sufficient opening has been achieved, and no further treatment is necessary.

Assisting Forms of Treatment It is widely accepted that well-conducted physical exercise should precede any type of interventional treatment and that cessation of smoking is mandatory. Nevertheless, it is also true that in many institutions it is very difficult to find an infrastructure that allows instruction of state-ofthe-art physical exercise for claudicants. And as far as smoking is concerned, there is a major difference between wanting to stop and actually stopping. Moreover, even with state-of-the-art exercise, young patients will not recover completely from claudications in all activities, including sports. The process will be long and will compromise their abilities in their professional lives. Therefore, it should perhaps be discussed whether young and active patients, especially, should be held to the

1379

Occlusion, Artery, Femoral. Figure 2 Superficial femoral artery lesion. (a) Short stenosis of a small superficial femoral artery close to the adductor’s canal in a patient with stage IV disease and diabetes. (b) After percutaneous transluminal angioplasty, the arterial lumen is wider, but a dissection remains. No stent was used due to good flow, absence of collateral filling, and small arterial lumen.

axiom of “physical exercise first” or whether invasive treatment might be offered even as a first approach to this group of patients.

Treatment Options with Relation to Location and Lesion Treatment of femoropopliteal lesions in claudicants has to be seen as more critical and less liberal compared with the iliac region. The main reasons are less favorable technical success, a higher complication rate, and poorer long-term success. There are many more lesions in the femoropopliteal arteries that do not meet the criteria for suitability for endovascular treatment. On the other hand, the versatility of endoluminal techniques opens treatment options in many particular lesions, and taking clinical symptoms as the only criteria to indicate or exclude treatment is not justified because—depending on the type of lesion—a simple and limited intervention can mean considerable improvement for the patient.

O

1380


Additional Morphological Factors (not Included in the TASC Classfication) Femoropopliteal occlusions may especially become a source of complications, particularly if they occurred recently. Simple PTA may result in downward embolization of occlusion material that may aggravate the symptoms or may turn the condition into a limb-threatening situation. Even in short occlusions, PTA may be insufficient to reopen the vessel, necessitating additional treatment such as stent placement (which does not result in better patency compared with balloon angioplasty alone). Reobstruction of stents, however, is more difficult to treat compared with simple restenosis. Eccentric calcified stenosis is frequently insufficiently treated. Because stenting is a technical but not necessarily a long-standing solution to such lesions, alternative techniques such as atherectomy may be considered if available. Unfortunately, these niche techniques are difficult to place in the market because of the costs involved, and some of the well-advanced devices such as the Simpson atherectomy catheter have been withdrawn from the market. (See Fig. 1)

Techniques Balloon Angioplasty Balloon angioplasty remains the working horse in femoropopliteal lesions. Modern angiographic units allow a built-in, fairly exact measurement of the true arterial diameter, and with the use of semicompliant balloons, adaptation to the diameter is well performed. We prefer not to grossly overdilate the artery in order to avoid dissection. Dilation times of 1–3 min are preferable by using pressure gauges. Balloon lengths of 2–4 cm are mainly used. In cases of major dissection, the first step should be an additional attempt to improve the result by prolonged balloon dilatation over 4–5 min; in many cases, the result will be improved by this cost-effective and simple approach (Fig. 1).

The overall results of femoral stenting are disappointing. New developments with drug-coated stents are on the way that allow elusion of drugs, such as rapamycin (Sirolimus) or taxol, from the stent surface. Especially for rapamycin, the first results from the coronary arteries are very promising, but no valid data yet exist on their use in the femorals. Radiation in stents, primarily at the time of insertion, did not show improved patency but was followed by an increased risk of thrombosis. Afterloading might therefore be a potential tool in the treatment of stent reobstruction.

Stent Grafts Stent grafts still play a limited role in the femoropopliteal field. ePTFE-covered self-expanding stent grafts such as the Hemobahn device (Gore, Flagstaff, AZ, USA) yielded promising results in a multicenter trial even in the femoropopliteal field and stimulated the hope of offering a percutaneous alternative, especially for those patients presenting with long femoropopliteal occlusions. But there is also a risk of midterm or late rethrombosis. Below the inguinal ligament, an ePTFE covering should be used exclusively because in animal experiments it has shown much less tendency to induce neointimal growth compared with Dacron covering. Other than in extraluminal bypasses, transcovering growth of tissue has been demonstrated, probably due to the long-segment wall contact between the stent graft and the original vascular lumen (2, 3). A considerable disadvantage of stent grafts is that important collaterals frequently have to be covered by the full body of the stent graft. In the event of reocclusion, these collaterals will not be available anymore, which might cause aggravation of symptoms. This is particularly true for the popliteal artery, for which development of compensating collaterals is limited. Therefore, we favor limiting the use of stent grafts to the proximal twothirds of the superficial femoral artery, especially in claudicants.

Stent Placement The use of all kinds of stents should be limited to those cases in which balloon angioplasty in all its variations did not achieve a sufficient result. This is particularly true for occluding dissections. Other than in the iliac field, stents should not be used liberally. The stented segment should be as short as possible. Balloon-expandable stents normally allow coverage of only short segments and might therefore be preferred for those lesions. In longer segments or in parts where bending of the artery is an issue, a self-expanding stent is advantageous if a stent cannot be avoided at all.

Results Balloon Angioplasty In femoropopliteal endovascular interventions (data taken from eight publications reporting on 1,469 procedures), the weighted average technical success was 90%, the complication rate was 4.3%, and the 3-year patency rate was 51% (1). Long-term patency was positively influenced by a good outflow tract (two or three lower limb arteries), absence of diabetes, and absence of residual stenosis. The latter would


favor the use of stents, but unfortunately, there is no proof that stenting will improve overall patency. Analyzing subgroups after femoral PTA, Huninck et al found different patencies for patients with stenotic and occlusive femoral lesions and good run-off (62% vs. 48% after 5 years) as well as those with poor run-off (stenoses: 62% vs. 43% after 5 years; occlusions: 43% vs. 27% after 5 years) (4).

Stents Follow-up results from stent implantation into the femoropopliteal arteries have not yielded improved results compared with balloon angioplasty alone. In a meta-analysis, Muradin et al found a 3-year patency of 63–66% after 3 years for stents, compared with 61% for PTA of stenoses. They also found, however, that in patients with more severe disease and more severe lesions, the patients achieved a higher benefit from stenting compared with those with less severe disease (5). In a randomized trial, Cejna and colleagues found no significant difference between patients who received PTA alone and those who received stenting (6).

1381

Endoluminal radiation therapy with afterloading or beta irradiation as well as drug-eluding stents may change the overall results in the future. To date, stenting in the femoral arteries should be used as a bail-out therapy in the case of PTA failure. Failure, however, needs to be defined strictly as severe dissection refractory to prolonged balloon dilatation, antegrade dissection with increasing obstruction, or severe residual obstruction. Minor irregularities of the wall are not enough to justify stenting because treatment of restenosis is more difficult compared with treatment after PTA alone. (See Figs 3 and 4).

Stent Grafts Little data exist on the usefulness of stent grafts in the femoropopliteal arteries. In a multicenter trial using the Hemobahn endoprosthesis, Lammer and colleagues achieved a primary patency of 90% after 6 months and 79% after 12 months with 80 limbs treated. Secondary patency was 93% at 12 months after treatment (7). These encouraging results are in contrast to many single-center experiences in which endografts showed a high rate of thrombosis that was frequently due to development of stenoses adjacent to the stent graft.

O

Occlusion, Artery, Femoral. Figure 3 Infrapopliteal occlusion. (a) Occlusion of all lower limb arteries in their proximal segments in a patient with stage IV disease. (b) After mechanical passage and percutaneous transluminal angioplasty, all three arteries have been sufficiently recanalized. No peripheral embolization has occurred.

1382


Occlusion, Artery, Femoral. Figure 4 (a) A 7-cm-long occlusion of the popliteal artery in a patient with stage IV disease. (b) After subintimal recanalization and percutaneous transluminal angioplasty, sufficient reopening with stent placement.

Complications The nature and quality of complications in the femoropopliteal arteries do not differ principally from those in the aortoiliac area: dissection, perforation, and embolization of occluding material. With stents, the risk of early thrombosis was a problem in the very beginning but has since become rare with combined treatment that includes modern antiplatelet drugs. In occlusions, the risk of embolization of the occluding material is the most potentially dramatic complication. Aspiration embolectomy in combination with selective thrombolysis is the treatment option of choice. Especially in claudicants, the risk therefore needs to be well balanced with the potential benefit.

usually prescribed. After femoral stent placement or in patients with marked irregularities after PTA, heparinization may be prolonged up to 72 h, and an additional platelet inhibitor such as Clopidogrel is recommended for 4–6 weeks.

Bibliography 1.

2.

3.

Adjunctive Drug Regimen

4.

In iliac and femoral PTA in claudicants, heparinization during the intervention and for 24 h after it—either by low-molecular-weight heparin or conventional heparin— is usually sufficient. A dose of 100 mg of aspirin daily is

5.

The TASC Working Group Management of peripheral arterial disease (PAD) (2000) Transatlantic Inter-Society Consensus (TASC). J Vasc Surg 31:S1–S296 Schurmann K, Vorwerk D, Uppenkamp R et al (1997) Iliac arteries: plain and heparin-coated Dacron-covered stent-grafts compared with noncovered metal stents—an experimental study. Radiology 203:55–63 Cejna M, Virmani R, Jones R et al (2001) Biocompatibility and performance of the Wallstent and several covered stents in a sheep iliac artery model. J Vasc Interv Radiol 12:351–358 Hunink M, Wong J, Donaldson M et al (1995) Revascularization for femoropopliteal disease. A decision and cost-effectiveness analysis. JAMA 274:165–171 Muradin G, Bosch J, Stijnen T et al (2001) Balloon dilation and stent implantation for treatment of femoropopliteal arterial disease: metaanalysis. Radiology 221:137–145

Occlusion, Artery, Iliac

6. 7.

Cejna M, Schoder M, Lammer J (1999) [PTA vs. stent in femoropopliteal obstruction] Radiologe 39:144–150 Lammer J, Dake M, Bleyn J et al (2000) Peripheral arterial obstruction: prospective study of treatment with a transluminally placed self-expanding stent-graft. International Trial Study Group. Radiology 217:95–104

1383

well as magnetic resonance (MR) angiography, computed tomography (CT), CT angiography, and intra-arterial angiography are useful tools for detecting the location and extent of an iliac obstruction. Because of access limitations, angiography is preferably performed via transbrachial or contralateral access if the clinical findings suggest iliac artery obstruction.

Occlusion, Artery, Iliac Nuclear Medicine D IERK VORWERK Klinikum Ingolstadt GmbH Institut fu¨r diagnostische und interventionelle Radiologie, Ingolstadt, Germany [email protected]

Synonyms

Nuclear medicine plays no particular role in the diagnosis of aortic stenosis.

Diagnosis Diagnosis is reliably achieved by angiography or MR angiography. Detection by duplex sonography is sometimes difficult due to unfavorable anatomic conditions.

Iliac artery occlusion; Iliac artery stenosis artery

Interventional Radiological Treatment Definition An arterial obstruction of the iliac arteries is an occlusion or narrowing of an arterial segment between the aortic bifurcation and the groin, not including the common femoral artery (CFA).

Pathology/Histopathology Occlusions are complete obstructions of the iliac arterial lumen due to thrombus formation, mainly onto a preexisting atherosclerotic plaque formation. In stenoses, the lumen is still patent but narrowed, with symptoms occurring with a diameter reduction of more than 50%.

Clinical Presentation Unilateral claudication is the leading clinical symptom and has a predilection for the upper thigh. Depending on the collateral pathways, the walking distance may vary between a few meters and a few hundred meters. Severe ischemic syndromes are rare unless there is no additional involvement of the infrainguinal arteries.

Imaging Many imaging modalities allow clinicians to diagnose iliac artery obstruction. Color-coded duplex sonography as

In the iliac segment, many lesions are amenable to percutaneous treatment with an acceptable outcome. Thus, a lesion’s location and type must be taken into consideration before treatment is recommended. Whereas most lesions in the aortoiliac segment will be conducive to an endovascular approach, this is not generally true for femoropopliteal lesions. In addition, the risk of treatment is related to its location and has to be addressed before an endovascular approach is recommended.

Type of Lesion The morphology of the treated lesion will influence the technical outcome, follow-up results, and also the risk of treatment. The Transatlantic Inter-Society Consensus (TASC) document therefore introduced a classification system that tries to categorize lesions with regard to their accessibility to either percutaneous treatment or surgery: Type A lesions, for which the percutaneous approach is ideal, type B lesions, for which the percutaneous approach is still the preferred technique; type C lesions, for which a surgical approach should be preferred; and type D lesions, for which surgery is the option of choice. The TASC classification overrules older classifications because it takes into account all available and published techniques, including stent technology, which offer a much wider variation of treatment and are also effective tools for dealing with acute complications of balloon angioplasty such as occluding dissection and vascular rupture. If we consider percutaneous therapy as the preferred method to deal with patients presenting with mild or

O

1384


moderate claudications, treatment might be offered to those presenting with type A and B lesions but should be discussed in depth with patients with type C lesions, since the risk and the potential benefit of treatment will be related to the underlying morphology. For iliac lesions, single stenoses up to 3 cm in length both in the common iliac artery (CIA) and external iliac artery (EIA) are classified as type A lesions, while single stenoses of 3–10 cm (not involving the CFA), double stenoses not longer than 5 cm each, and unilateral occlusions of the CIA are classified as type B lesions (Fig. 1).

Type C lesions include bilateral long stenoses (5–10 cm in length), unilateral EIA occlusions not extending into the CFA, and unilateral EIA stenoses extending into the CFA. More advanced lesions are classified as type D. Using this classification, many iliac lesions will fall into the A and B groups, opening a potentially growing field for endovascular procedures performed on mild to moderate claudicants. Even in some type C lesions, percutaneous treatment has no major technical concerns, complication risks, or compromised outcomes. This is particularly the case for EIA occlusions not extending into the CFA. However, published data are lacking to back up this experience.

Multilevel Disease Even mild symptoms may be associated with multilevel disease (e.g., iliac stenosis and a well-collateralized femoropopliteal lesion). There is some chance that exclusive intervention in the iliac region may be sufficient to improve the clinical situation. If present, multilevel disease does not preclude treatment in these patients.

Assisting Forms of Treatment It is widely accepted that well-conducted physical exercise should precede any type of interventional treatment and that cessation of smoking is mandatory. Nevertheless, it is also true that in many institutions it is very difficult to find an infrastructure that allows teaching of state-of-theart physical exercise to claudicants, and as far as smoking is concerned, there is a major difference between willing and doing. Moreover, even with state-of-the-art exercise, young patients will not recover completely from claudications in all their activities, including sports. The process will be longer and compromise their abilities in their professional lives. Therefore, it might be discussed whether young and active patients should be vigorously kept to the axiom of “physical exercise first” or whether invasive treatment might be offered to this group of patients even as a first approach.

Treatment Options with Relation to Location and Lesion Occlusion, Artery, Iliac . Figure 1 Occlusion of the left common iliac artery. (a) Occlusion of the left common iliac artery imaged via a retrograde access. The internal iliac artery is open. (b) After passage of the occlusion, the complete occlusion is visible with a stump at the orifice. (c) After primary stenting followed by careful balloon dilation, patency is fully restored.

The aorta and the iliac arteries have been a primary field for percutaneous interventions for a long time. Easy access to the lesion, the relatively large diameter of the target vessels, and the comparably benign outcomes even with major complications contribute to the wide acceptance of percutaneous interventions in that area. Over the years, the indications have been extended and now include not only stenotic but also occlusive disease and treatment of


aneurysms. The introduction of vascular stents has especially been very helpful for overcoming major problems and offers a tool to treat major complications that otherwise remained a domain for surgical repair. Iliac occlusive disease accounts for approximately one-third of occlusive arterial disease, while two-thirds is located subinguinally. However, iliac percutaneous transluminal angioplastry (PTA) plays a major role in interventional radiological routines because it is wellestablished in most institutions and by most surgeons, many lesions are amenable to percutaneous treatment, and technical as well as clinical results are satisfying. Clinically, intermittent claudication starting in the upper thigh in combination with lower limb claudication is the leading symptom. Erectile dysfunction in men may also be present. In isolated iliac lesions, critical ischemia is rare if not combined with additional subinguinal disease. Rarely, blue-toe syndrome might be present if cholesterol embolization has occurred from an ulcerated plaque in the iliac axis. Weakened femoral pulses and a reduced ankle-arm index are simple clinical signs that can indicate iliac obstruction and that can be verified by direct or poststenotic color-coded or duplex studies. For planning a percutaneous intervention, angiography is still the most helpful procedure. Clinical indications for treatment depend on the severity of symptoms and how they limit the daily life of an individual patient. Stent placement in stenotic lesions should be indicated from a technical point of view if angioplasty remains insufficient, as defined by visibly poor outflow or major pressure gradients. Because follow-up data are now available showing that iliac stent placement is safe, a liberal approach is justified, although primary stenting of stenoses does not seem to be generally recommended because of socioeconomic reasons and potential followup problems. Furthermore, primary stenting is not beneficial over successful balloon dilation (8). Thus, primary stenting is recommended only if PTA fails or if technical requirements compromise the success of simple PTA, such as with iliac occlusions.

Iliac Artery Stenoses Although balloon angioplasty has proved to be an effective procedure in treating iliac stenoses in particular, the indication for stent placement should be restricted to lesions that are not primarily amenable to PTA alone. An inadequate postangioplasty result has been suggested as a general indication for stent placement, although the term “inadequate” remains ill-defined. Residual pressure gradients are certainly a useful way to assess the angioplasty

1385

result, but it is still unclear what the borderline gradient ultimately requiring additional intervention is. Moreover, the decision should not be made without referring both to morphologic criteria and visibly reduced flow. Long-segment stenoses with irregular surfaces, aneurysm formation, or markedly ulcerated plaques may be included in the group of complex lesions. Eccentric stenoses and ostial lesions with extension to the aortic bifurcation are known to not respond well to balloon angioplasty. A stenotic lesion may respond well to balloon inflation but may collapse after the balloon is deflated. Complications of balloon angioplasty can be treated well by stent placement. These include intramural hematoma and flow-obstructing dissections complicating PTA, which may be an acute indication for stent placement in order to maintain the vascular lumen, obviating emergency surgery. Iliac restenosis after previous PTA does not require stent placement in general since there is no proof that stenting prevents restenosis under those circumstances. However, stenting may be considered from a technical point of view in cases in which the result of balloon angioplasty remains compromised. An inadequate result after state-of-the-art balloon dilatation of the iliac stenoses is a prerequisite for secondary stent placement in stenotic lesions.

Iliac Artery Occlusions Percutaneous treatment of iliac occlusions is technically feasible. In cases of acute thrombosis, thrombolysis as an alternative to surgical thrombectomy might precede PTA of an underlying lesion. Mechanical thrombectomy via percutaneous access is still in its infancy and cannot be recommended as a routine approach because potential risks such as downward or cross-over embolization are possible, and no data are yet available to determine the overall complications of such an approach. In chronic occlusions with an occlusion time exceeding 3 months, balloon angioplasty alone, thrombolysis with subsequent balloon angioplasty, and elective stenting or mechanical passage of the occlusion followed by primary stent implantation have been described as alternative techniques. Metallic stents—and self-expandable endoprostheses in particular—offer a new concept of percutaneous revascularization in chronic iliac occlusion, which we believe is a primary indication (9). Self-expandable stents are used to cover the occluding thrombotic material, thereby preventing peripheral dislodgement, a well-known complication of percutaneous recanalization of occlusions. Indications for the use of metallic stents into arteries should always be technical. The type and morphology of

O

1386


the lesion, the outcome of balloon angioplasty, and complicated situations are important considerations, although stenting has been tested and shown to be a safe procedure. This is particularly true for treatment of restenosis. There is currently no proof that stent placement is more effective in preventing restenosis than balloon angioplasty with good technical success. Furthermore, there is no proof that in a restenosed vessel, the use of a stent would be beneficial to prevent recurrent stenosis.

Technical Considerations in the Iliac Arteries

important details that may lead to the preference of one or another type. Exact placement is mandatory to avoid major complications, especially if the stent has to placed close to the aortic bifurcation. While self-expanding Wallstents can be corrected during placement to a limited extent, balloon-expandable stents and self-expanding nitinol stents cannot undergo correction of their localization once inflation of the balloon has begun. Chronic iliac artery occlusions are primarily indicated for stent placement. We avoid predilatation and place the stent directly into the occluded segment. After stenting, careful balloon dilatation is performed to avoid dislodgement of occluding material.

Balloon Dilatation Balloon dilatation of the iliac arteries is relatively simple to perform. A retrograde transfemoral approach provides the easiest access to that type of stenotic lesions. Crossover dilatation may be performed in special indications such as double-sided stenoses in the event that both lesions should be dilated in one session, or if an external iliac stenosis extends far down into the CFA. After the diseased segment is carefully traversed, a suitable balloon is placed across the lesion, and dilatation is performed either manually or by using a pressure-monitoring syringe. The size of the balloon can be depicted either by film measurements or in digital subtraction angiography images by use of graduated catheters that allow fairly exact measurement of vessel size. There is wide agreement that both the hemodynamic relevance of a lesion as well as post-PTA success can be accurately monitored by measuring the pressure gradient across the lesion. However, there is some dispute about the level of the pressure. A mean pressure gradient of 10 mmHg and less after peripheral drug-induced vasodilation is accepted by most authors to indicate successful PTA even if the morphological result is not excellent. Some authors use a systolic gradient of 10 mmHg. There is no uniform agreement on whether PTA of simple iliac lesions requires any additional anticoagulation afterward. We regularly keep our patients on full heparinization (500–1,000 IU/h) for 12–24 h and recommend lifetime aspirin therapy (100 mg/day).

Stent Implantation If balloon angioplasty fails by morphologic and functional criteria, stent implantation can be considered in stenotic lesions. The technique depends on the type of stent used. There is no stent preference because most clinical series have shown similar results. The length and location of the lesion, the experience of the investigator, and the availability of appropriately sized stents are

Atherectomy Directional atherectomy does not play a major role in treating iliac arterial disease. This is because the ratio of introduction and working diameters in most atherectomy systems is relatively low, which requires a considerably large puncture to achieve atherectomy in larger iliac diameters of 8–12 mm. The Simpson atherectomy catheter—which is no longer available commercially— required an 11F sheath to sufficiently treat iliac lesions. Atherectomy plays a more important role in the recanalization of stent reobstruction in order to debulk stents from reobstructing neointimal tissue.

Stent Reobstruction In cases of in-stent stenosis, directional atheterectomy or balloon dilation are both recommended. If PTA is used, a balloon size according to the outer diameter of the stent in place is recommended in order to compress the neointimal intima to a maximum, especially if occurring within a self-expanding stent that does not allow overexpansion. If a balloon-expandable stent has been used, slight overdilation of the stent is recommended to gain a larger diameter despite neointimal tissue. Some authors prefer atherectomy to debulk the stent. This is achievable in smaller stents such as in the femorals, but it may require very large instruments in iliac stents. In stent occlusion, treatment is more difficult. In acute occlusions early after placement, technical problems are mainly responsible, and it is mandatory to overcome these problems to maintain long-term success. Recent thrombosis should be treated by thrombolysis followed by PTA and/or additional stent placement. Late occlusion is mainly due to reobstruction by neointima within or adjacent to the stent. There is not much published experience about how to treat complete stent occlusion at a chronic stage. Thrombolysis, atherectomy, and mechanical aspiration followed by balloon

Occlusion, Artery, Popliteal

angioplasty are possible techniques. One of the simpler techniques is the stent-in-stent technique: After traversal of the occluded stent (which is usually easy to accomplish), a stent is placed within the occluded segment, bridging it at both ends. It is then carefully dilated with a tendency to underdilation of 1–2 mm. This is considered safe in order to avoid embolization of occlusion material.

1387

Pathology/Histopathology Occlusions are complete obstructions of the infrainguinal arterial lumen and are due to thrombus formation, mainly on preexisting atherosclerotic plaque. In stenoses, the lumen is still patent but narrowed, with a diameter reduction of more than 50% causing symptoms. Alternatively, acute occlusions occur because of thrombotic emboli from the heart.

Bibliography 1.

2. 3.

4.

5.

6.

7.

8. 9.

The TASC Working Group (2000) Management of peripheral arterial disease (PAD). Transatlantic Inter-Society Consensus (TASC). J Vasc Surg 31:S1–S296 Vollmar J (1996) Rekonstruktive Chirurgie der Arterien. Thieme, Stuttgart, pp 207–214 Berger T, So¨rensen R, Konrad J et al (1986) Aortic rupture. A complication of transluminal angioplasty. Am J Roentgenol 146:373–374 Strecker E, Hogan B, Liermann D et al (1993) Iliac and femoropopliteal vascular occlusive disease treated with flexible tantalum stents. Cardiovasc Intervent Radiol 16:158–164 Dietrich EB, Santiago O, Gustafson G et al (1993) Preliminary observation on the use of the Palmaz stent in the distal portion of the abdominal aorta. Am Heart J 125:490–500 Rholl K, Van Breda A (1994) Percutaneous intervention for aortoiliac disease. In: Strandness E, Van Breda A (eds) Vascular Diseases. Churchill Livingstone, New York, pp 433–466 Long A, Gaux J, Raynaud A et al (1994) Infrarenal aortic stents. Initial clinical experience and angiographic follow-up. Cardiovasc Intervent Radiol 16:203–208 Dietrich EB (1993) Endovascular techniques for abdominal aortic occlusions. Int Angiol 12:270–280 Becker G, Katzen B, Dake M (1994) Noncoronary angioplasty. Radiology; 170:921–940

Clinical Presentation Clinical indications for treatment include claudication, pain at rest, and nonhealing ulceration. Severity of the symptoms depends on the acuteness of the occlusion, the condition of collateral pathways, the lesion’s location, and concomitant disease such as diabetes or renal insufficiency.

Imaging Many imaging modalities allow the physician to diagnose infrainguinal artery obstruction. Color-coded duplex sonography as well as magnetic resonance angiography (MRA), computed tomography, computed tomography angiography (CTA), and intraarterial angiography are useful tools to detect the location and extent of an iliac obstruction.

Nuclear Medicine

Occlusion, Artery, Popliteal D IERK VORWERK Klinikum Ingolstadt GmbH Institut fu¨r diagnostische und interventionelle Radiologie, Ingolstadt, Germany [email protected]

Synonyms

Nuclear medicine plays no particular role in the diagnosis of aortic stenosis.

Diagnosis Diagnosis is reliably achieved by angiography or MRA. Detection by duplex sonography is a reliable tool for the femoropopliteal segment but is sometimes limited for detection and exact lesion description in the lower limbs. In stage IV (Fontaine) disease, MRA may be limited because of low arterial flow, movement artifacts, or artifacts due to superimposed venous flow.

Infrainguinal arterial obstruction

Interventional Radiological Treatment Definition Endovascular versus Surgical Treatment An arterial obstruction of the infrainguinal arteries is an occlusion or a narrowing of an arterial segment between the groin and the lower limbs.

Endovascular therapy is known to be of low invasiveness with good technical success, achieving fair overall patency. In femoropopliteal endovascular interventions (data

O

1388


taken from eight publications reporting on 1,469 procedures), the weighted average technical success was 90%, the complication rate was 4.3%, and the 3-year patency rate was 51%. Stents did not improve patency, showing a patency of 58% after 3 years (1). Surgery offers acceptable results for distal reconstruction; an average 5-year patency of 80% for vein bypasses and 65–75% for expanded polytetrafluoroethylene (ePTFE) bypasses has been reported. The combined mortality and amputation risk was calculated to be about 2.2% for aortobifemoral reconstructions and 1.4% for femoropopliteal reconstructions (1).

Location of Lesion Claudication is mainly related to lesions in the aortoiliac and femoropopliteal regions. It is unlikely to be due to infrapopliteal lesions, and there is general agreement that treatment below the knee should be strictly limited to patients with critical limb ischemia, i.e., stages III and IV (Fontaine) or categories 4–6 (Rutherford).

Type of Lesion The morphology of a treated lesion influences the technical outcome, follow-up results, and also the risk of treatment. Therefore, the TransAtlantic Inter-Society Consensus (TASC) document introduced a classification system that tries to categorize lesions with regard to their accessability to either percutaneous treatment or surgery: type A lesions, which are ideal for a percutaneous approach; type B lesions, in which the percutaneous approach is still the preferred technique; type C lesions, in which a surgical approach should be preferred; and type D lesions, in which surgery is the option of choice. The TASC classification overrides older classifications because it takes into account all of the available and published techniques (including stent technology), which offer a much wider variation of treatment as well as effective tools to deal with acute complications of balloon angioplasty, such as occluding dissection and vascular rupture. If we consider percutaneous therapy as the preferred method to treat those patients presenting with mild or moderate claudication, treatment might be offered to those presenting with type A and B lesions, but should be discussed in depth with patients with type C lesions, as the risk and the potential benefit of treatment are related to the underlying morphology. In the femoropopliteal field, type A lesions are single stenoses up to 3 cm in length not involving the very proximal superficial femoral and the distal popliteal artery. Type B lesions include stenoses 3–5 cm in length, heavily calcified stenoses, multiple lesions (each up to 3 cm), and lesions with no sufficient tibial run-off (the

latter are unlikely to meet the criteria of mild or moderate claudication). Type C lesions are classified as stenoses or occlusions longer than 5 cm and multiple midsize lesions (3–5 cm). Total common femoral, superficial femoral, and popliteal occlusions are classified as type D lesions. There was some dissenting discussion on the definition of type B lesions: Interventional radiologists represented by the Cardiovascular and Interventional Radiological Society of Europe wished to express their assumption that even longer lesions of up to 10 cm may be justified as being classified type B instead of type C; they claimed that the reported results were mainly due to underdeveloped techniques and instruments that have substantially improved and that no data exists comparing the efficacy of percutaneous transluminal angioplasty (PTA) versus bypass surgery for lesions between 4 and 10 cm. Other than in the iliac area, few femoral lesions meet the criteria for types A and B lesions, especially if limited to 5 cm in length. Thus, few patients with mild or moderate claudication due to femoropopliteal lesions will be ideal candidates for percutaneous treatment. Moreover—without limiting the importance of the TASC document, which certainly means a step forward in the joint approach to peripheral vascular disease—the morphological classification does not take into account some technical considerations that depend on the age and composition of a lesion. Particularly in femoral occlusions, the degree of organization of the occluding thrombus or the composition of the lesion with the original stenosis at the proximal and distal ends or in the middle are factors that are not very predictable but may influence the technical outcome of the intervention or its complication rate (for instance, distal embolization might aggravate symptoms). Other than in the iliac arteries, the liberal use of stents and stent grafts may help overcome a failed balloon angioplasty and resolve the technical outcome, but it does not achieve an improved long-term efficacy, and it may start a lifetime dependency on recurrent interventional or surgical procedures. These associated potential drawbacks have to be carefully balanced against the potential benefits and need to be discussed in depth with the patient before treatment is performed, especially in association with mild or moderate claudication. These considerations mainly restrict the use of endovascular treatment in femoropopliteal lesions to stages IIb and IIa patients presenting with type A or lesspronounced type B lesions.

Assisting Forms of Treatment It is widely accepted that well-conducted physical exercise should precede any type of interventional treatment and that cessation of smoking is mandatory. Nevertheless, it is also true that in many institutions it is very difficult to


find an infrastructure that allows instruction of state-ofthe-art physical exercise for claudicants. And as far as smoking is concerned, there is a major difference between wanting to stop and actually stopping. Moreover, even with state-of-the-art exercise, young patients will not recover completely from claudications in all activities, including sports. The process will be long and will compromise their abilities in their professional lives. Therefore, it should perhaps be discussed whether young and active patients, especially, should be held to the axiom of “physical exercise first” or whether invasive treatment might be offered even as a first approach in this group of patients.

1389

Techniques Balloon Angioplasty Balloon angioplasty remains the working horse in femoropopliteal lesions. Modern angiographic units allow a builtin, fairly exact measurement of the true arterial diameter and by use of semicompliant balloons, adaptation to the diameter is well performed. We prefer not to grossly overdilate the artery in order to avoid dissection. Dilation times of 1–3 min are preferable by using pressure gauges. Balloon lengths of 2–4 cm are mainly used. In cases of major dissection, the first step should be an additional attempt to improve the result by prolonged balloon dilatation over 4–5 min; in many cases, the result will be improved by this costeffective and simple approach (Fig. 1).

Treatment Options with Relation to Location and Lesion Treatment of femoropopliteal lesions in claudicants has to be seen as more critical and less liberal compared with the iliac region. The main reasons are less favorable technical success, a higher complication rate, and poorer long-term success. There are many more lesions in the femoropopliteal arteries that do not meet the criteria for suitability for endovascular treatment. On the other hand, the versatility of endoluminal techniques opens treatment options in many particular lesions, and taking clinical symptoms as the only criteria to indicate or exclude treatment is not justified because depending on the type of lesion, a simple and limited intervention can mean considerable improvement for the patient.

Additional Morphological factors (not Included in the TASC Classfication) Femoropopliteal occlusions may especially become a source of complications, particularly if they occurred recently. Simple PTA may result in downward embolization of occlusion material that may aggravate the symptoms or may turn the condition into a limbthreatening situation. Even in short occlusions, PTA may be insufficient to reopen the vessel, necessitating additional treatment such as stent placement (which does not result in better patency compared with balloon angioplasty alone). Reobstruction of stents, however, is more difficult to treat compared with simple restenosis. Eccentric calcified stenosis is frequently insufficiently treated. Because stenting is a technical but not necessarily a long-standing solution to such lesions, alternative techniques such as atherectomy may be considered if available. Unfortunately, these niche techniques are difficult to place in the market because of the costs involved, and some of the well-advanced devices such as the Simpson atherectomy catheter have been withdrawn from the market.

Stent Placement The use of all kinds of stents should be limited to those cases in which balloon angioplasty in all its variations did not achieve a sufficient result. This is particularly true for occluding dissections. Other than in the iliac field, stents should not be used liberally. The stented segment should be as short as possible. Balloon-expandable stents normally allow coverage of only short segments and might therefore be preferred for those lesions. In longer segments or in parts where bending of the artery is an issue, a self-expanding stent is advantageous if a stent cannot be avoided at all. The overall results of femoral stenting are disappointing. There are new developments with drug-coated stents on the way that allow elusion of drugs, such as rapamycin (Sirolimus) or taxol, from the stent surface. Especially for rapamycin, the first results from the coronary arteries are very promising, but no valid data yet exist on their use in the femorals. Radiation in stents, primarily at the time of insertion, did not show improved patency but was followed by an increased risk of thrombosis. Afterloading might therefore be a potential tool in the treatment of stent reobstruction.

Stent Grafts Stent grafts still play a limited role in the femoropopliteal field. ePTFE-covered self-expanding stent grafts such as the Hemobahn device (Gore, Flagstaff, AZ, USA) yielded promising results in a multicenter trial even in the femoropopliteal field and stimulated the hope of offering a percutaneous alternative, especially for those patients presenting with long femoropopliteal occlusions. But there is also a risk of midterm or late rethrombosis. Below the inguinal ligament, ePTFE covering should be used exclusively because in animal experiments it has

O

1390


Occlusion, Artery, Popliteal. Figure 1 (a) A 7-cm-long occlusion of the popliteal artery in a patient with stage IV disease. (b) After subintimal recanalization and percutaneous transluminal angioplastyPTA, sufficient reopening with stent placement.

shown much less tendency to induce neointimal growth compared with Dacron covering. Other than in extraluminal bypasses, transcovering growth of tissue has been demonstrated, probably due to the long-segment wall contact between the stent graft and the original vascular lumen (2, 3). A considerable disadvantage of stent grafts is that important collaterals frequently have to be covered by the full body of the stent graft. In the event of reocclusion, these collaterals will not be available anymore, which might cause aggravation of symptoms. This is particularly true for the popliteal artery, for which development of compensating collaterals is limited. Therefore, we favor limiting the use of stent grafts to the proximal two-thirds of the superficial femoral artery, especially in claudicants.

Results Balloon Angioplasty In femoropopliteal endovascular interventions (data taken from eight publications reporting on 1,469 procedures), the weighted average technical success was 90%, the complication rate was 4.3%, and the 3-year patency rate was 51% (1).

Long-term patency is positively influenced by a good outflow tract (two or three lower limb arteries), absence of diabetes, and absence of residual stenosis. The latter would favor the use of stents, but unfortunately, there is no proof that stenting will improve overall patency. Analyzing subgroups after femoral PTA, Huninck et al. found different patencies for patients with stenotic and occlusive femoral lesions and good run-off (62% versus 48% after 5 years) as well as those with poor runoff (stenoses: 62% versus 43% after 5 years; occlusions: 43% versus 27% after 5 years) (4).

Stents Follow-up results from stent implantation into the femoropopliteal arteries have not yielded improved results compared with balloon angioplasty alone. In a meta-analysis, Muradin et al. found a 3-year patency of 63–66% after 3 years for stents, compared with 61% for PTA of stenoses. They also found, however, that in patients with more severe disease and more severe lesions, the patients achieved a higher benefit from stenting compared with those with less severe disease (5). Cejna and colleagues found no significant difference between

Occlusion, Bile Ducts

patients who received PTA alone and those who received stenting in a randomized trial (6). Endoluminal radiation therapy with afterloading or b irradiation as well as drug-eluding stents may change the overall results in the future. To date, stenting in the femoral arteries should be used as a bail-out therapy in the case of PTA failure. Failure, however, needs to be defined strictly as severe dissection refractory to prolonged balloon dilatation, antegrade dissection with increasing obstruction, or severe residual obstruction. Minor irregularities of the wall are not enough to justify stenting as treatment of restenosis is more difficult when compared with treatment after PTA alone.

References 1.

2.

3.

4.

5.

6.

Stent Grafts Little data exist on the usefulness of stent grafts in the femoropopliteal arteries. In a multicenter trial using the Hemobahn endoprosthesis, Lammer and colleagues achieved a primary patency of 90% after 6 months and 79% after 12 months with 80 limbs treated. Secondary patency was 93% at 12 months after treatment (7). These encouraging results are in contrast to many single-center experiences in which endografts showed a high rate of thrombosis that was frequently due to development of stenoses adjacent to the stent graft.

1391

7.

The TASC Working Group Management of peripheral arterial disease (PAD) (2000) Transatlantic Inter-Society Consensus (TASC). J Vasc Surg 31:S1–S296 Schurmann K, Vorwerk D, Uppenkamp R et al (1997) Iliac arteries: plain and heparin-coated Dacron-covered stent-grafts compared with noncovered metal stents – an experimental study. Radiology 203:55–63 Cejna M, Virmani R, Jones R et al (2001) Biocompatibility and performance of the Wallstent and several covered stents in a sheep iliac artery model. J Vasc Interv Radiol 12:351–358 Hunink M, Wong J, Donaldson M et al (1995) Revascularization for femoropopliteal disease. A decision and cost-effectiveness analysis. JAMA 274:165–171 Muradin G, Bosch J, Stijnen T et al (2001) Balloon dilation and stent implantation for treatment of femoropopliteal arterial disease: metaanalysis. Radiology 221:137–145 Cejna M, Schoder M, Lammer J (1999) [PTA vs. stent in femoropopliteal obstruction] Radiologe 39:144–150 Lammer J, Dake M, Bleyn J et al (2000) Peripheral arterial obstruction: prospective study of treatment with a transluminally placed self-expanding stent-graft. International Trial Study Group. Radiology 217:95–104

Occlusion, Bile Ducts T HOMAS H ELMBERGER 1 , C ARLO B ARTOLOZZI 2 , PAOLA VAGLI 2 , C HIARA F RANCHINI 2 1

Complications The nature and quality of complications in the femoropopliteal arteries do not differ principally from those in the aortoiliac area: dissection, perforation, and embolization of occluding material. With stents, the risk of early thrombosis was a problem in the very beginning but has since become rare with combined treatment that includes modern antiplatelet drugs. In occlusions, the risk of embolization of the occluding material is the most potentially dramatic complication. Aspiration embolectomy in combination with selective thrombolysis is the treatment option of choice. Especially in claudicants, the risk therefore needs to be well balanced with the potential benefit.

Adjunctive Drug Regimen In iliac and femoral PTA in claudicants, heparinization during the intervention and for 24 h after it—either by low-molecular-weight heparin or conventional heparin— is usually sufficient. A dose of 100 mg of aspirin daily is usually prescribed. After femoral stent placement or in patients with marked irregularities after PTA, heparinization may be prolonged up to 72 h, and an additional platelet inhibitor such as Clopidogrel is recommended for 4–6 weeks.

Clinic of Radiology and Nuclear medicine, University Hospitals Schleswig-Holstein, Campus Luebeck 2 Department of Diagnostic and Interventional Radiology, University of Pisa, Pisa, Italy [email protected] [email protected]

Synonyms Bile duct obstruction; Cholestasis; Obstruction of the biliary tree

Definition Any condition in which there is a complete obstacle to the biliary flow.

Pathology and Histopathology Cholestasis may be related to mechanical, cellular, and metabolic causes. Mechanical biliary obstruction may be due to intrinsic or extrinsic obstruction of bile flow, which can occur either at the intrahepatic or extrahepatic level (Table 1).

O

1392


Occlusion, Bile Ducts. Table 1 obstruction

Mechanical biliary

Intrahepatic Intrinsic Congenital cause • Biliary atresia • Alagille syndrome • Intrahepatic calculi complicating Caroli disease • Cystic fibrosis Benign cause • Intrahepatic stones • Sclerosing cholangitis • Biliary parasitosis Malignant cause • Cholangiocarcinoma extrinsic Benign and malignant tumors Extrahepatic Intrinsic Congenital anomalies • Choledochal atresia • Choledochal cysts Benign conditions • Gallstones • Parasitosis • Cholangitis • Inflammatory strictures • Postinterventional strictures • Vater’s papillary stenosis • ▶AIDS-related cholangiopathy • ▶Biliary tuberculosis • ▶Sump syndrome Malignant tumors • Cholangiocarcinoma • Vater’s papillary carcinoma extrinsic Benign causes • ▶Mirizzi syndrome • Enlarged lymph nodes • Pancreatitis • Local benign tumors and cysts Malignant conditions • Gallbladder carcinoma • Malignant lymphadenomegalies • Pancreatic head carcinoma • Stomach and colon cancer

Overall, gallstones are the most common cause of biliary obstruction. Intrahepatic intrinsic biliary obstruction can be due to: congenital affections (biliary atresia, Alagille syndrome, intrahepatic calculi complicating Caroli disease and cystic fibrosis), benign conditions (intrahepatic ductal stones, sclerosing cholangitis, and biliary parasitosis), and malignant conditions, with cholangiocarcinoma as the most important malignant cause. Extrinsic causes of intrahepatic obstruction include compression by benign and malignant tumors. Extrahepatic intrinsic biliary obstruction can be due to: congenital anomalies (choledochal atresia, choledochal cysts), benign conditions

(gallstones representing the first cause, parasitosis, cholangitis, inflammatory and postinterventional strictures, Vater’s papillary stenosis, AIDS-related cholangiopathy, biliary tuberculosis, sump syndrome), and malignant tumors (cholangiocarcinoma, and Vater’s papillary carcinoma). Extrinsic compression of extrahepatic biliary tract includes benign conditions (Mirizzi syndrome, enlarged lymph nodes, pancreatitis, local benign tumors, and cysts), and malignant conditions (gallbladder carcinoma, malignant lymphadenomegalies, pancreatic head carcinoma, stomach and colon cancers) (1).

Clinical Aspects Symptoms are determined by abnormal elevation of primary and secondary bile products in blood and the absence of the extrinsic intestinal bile function. Additionally, cholangitis may represent a concomitant complication in biliary obstruction. Common symptoms are jaundice, pale coloured stools, dark urine, itching, fever, and right upper quadrant pain. Blood tests show increase of bilirubin, alkaline phosphatase, and hepatic enzymes.

Imaging Plain radiographs are of limited value to detect abnormalities in the biliary system. Frequently, calculi are not visualized because of their low radiopacity. Biliary obstruction results in dilatation of bile ducts involving the biliary tree proximally to the obstruction; depending on the site of obstruction, the bile ducts dilatation may involve an isolated lobe or segment or both lobes. Especially in biliary stone disease, mostly both lobes present ductal dilatation. Nevertheless, the absence of bile ducts dilatation does not exclude recent or intermittent obstruction. Ultrasound (US) is the method of first choice to establish the presence and extent of bile duct dilatation (Fig. 1). While peripheral ductal branches can be visualized only if dilated, the normal common bile duct is easily displayed (2). In general, US is able to establish the level of obstruction and reveal the cause of obstruction, as intrahepatic calculi, gallbladder stones, lymphadenopathy, hepatic and pancreatic masses. The demonstration of common bile duct calculi is strongly dependent on their location within the biliary system; in fact, calculi in the lower part of the common bile duct may be obscured by duodenal gas. Computed tomography (CT) is more accurate than US in determining the specific cause and level of obstruction, but it is of limited value in detecting radiolucent bile duct stones. On the other hand, US can


Occlusion, Bile Ducts. Figure 1 Biliary dilatation. Ultrasound can easily demonstrate dilatation of both intrahepatic and extrahepatic bile ducts in mechanical biliary obstruction.

be significantly limited in the differentiation of cholangiocarcinoma from sludge accumulation since both appear hypoechoic. In these cases, the different enhancement behaviour (absence of contrast enhancement in case of sludge) allows differential diagnosis on CT. With the advent of easy applicable magnetic resonance (MR) techniques displaying fluid-filled structures (i.e., heavily T2-weighted, snap-shot techniques, “MRhydrography”), MR cholangiopancreatography (MRCP) is the reference standard for noninvasive imaging of the biliary tree. In comparison to invasive endoscopic retrograde cholangiopancreatography (ERCP), MRCP provides noninvasive detailed evaluation of the biliary tree. Since MRCP is generally combined with complementary crosssectional MR a comprehensive evaluation of in- and extrinsic factors that might affect the biliary tree is possible. With high-resolution techniques in- and extrinsic bile duct, obstructions and dilatations can be visualized by MRCP with a superior high sensitivity and specificity. Since the MR visualization of ducts is strongly correlated to the presence of fluid (bile) within the lumen, severe strictures, sludge, small stones or postoperative scars and clips may appear as short segments of signal void. In general, this condition can be clarified by additional unenhanced and contrast-enhanced T1- and T2-weighted axial (and/or multiplanar) sequences. In characterizing the features of the biliary tree, the same characteristics already known from cholangiography may be applied to MRCP. In general, stone-related obstruction can be displayed easily with the concave transition from fluidfilled ductal lumen to the stone and the usually smooth, regular outlining of the lumen. An irregular tapering of the lumen can often be seen in malignant obstructions due to an extrinsic or infiltrating mass that can be uncovered by cross-sectional imaging. Typical examples

1393

for that are tumors of the pancreatic head affecting the common bile duct alone or together with the pancreatic duct (so called “double duct sign,” in about 5% of cases of pancreatic cancer). In case of such malignancies, MR can provide a comprehensive evaluation in terms of staging for intrahepatic metastases, lymph node metastasis, peritoneal tumour spread, and vascular encasement. However, comparing MR and multidetector CT there is no significant difference in staging of biliary and pancreatic malignancies regarding recent literature. Due to the noninvasive, high diagnostic accuracy of MRCP, in most centres percutaneous (PTC) and endoscopic procedures (ERCP) are performed in selected cases for diagnostic purpose (e.g., tissue sampling by brush) and interventional procedures (e.g., balloon dilatation, stone extraction, stent placement) avoiding pure diagnostic interventions which may cause procedure related complications in up to 5% (e.g., acute pancreatitis) (1). In patients with external biliary drainage (i.e., Kehr, Bracci), trans-catheter cholangiography is performed to evaluate the placement of the catheter, the patency of the common bile duct and the transpapillar drainage.

Nuclear Medicine In patients with biliary obstruction, cholescintigraphy with 99mTc IDA (iminodiacetic acid) or analogs demonstrates a “congestion” of the radiotracer mostly associated with a normal hepatic distribution/activity in absence of delayed (>1 h after tracer injection) biliary activity.

Interventional Radiology Radiologic interventional procedures in biliary obstruction may be performed using either a percutaneous or an endoscopic approach or a combination of both (i.e., “rendevouz procedures”). Possible interventions are catheter placement, biliary stenting, balloon dilatation of biliary strictures, stone removal, and tissue sampling. Stone removal is the most common biliary intervention and generally performed by endoscopy. Usually a papillotomy is necessary, followed by stone extraction by balloon or basket instruments guided by the endoscope. Larger stones (>2 cm in diameter) are less prone to endoscopical removal and present a higher risk of perforation, bleeding, secondary pancreatitis, and cholangitis (3, 4). In the era of advanced endoscopic techniques, the use of percutaneous transhepatic radiological interventions is restricted to the treatment of intrahepatic stones and cholangitis where an internal– external drainage may be needed to relieve the biliary system. Sequentially, a stone can be removed or an

O

1394

Occlusion, Bowel in Childhood

obstruction can be treated. In cases of unsuccessful or impractical endoscopic maneuvers, stone removal, or stone fragmentation and stent placement are also percutaneously possible. Balloon dilatation can be necessary both in patients with benign and malignant biliary strictures. Stenoses are safely and minimally invasively treated by balloon dilatation. Among the benign strictures biliary-enteric anastomotic strictures respond best to dilatation whereas a temporarily support by a drainage tube may enhance the result (5). In case of recoiling, redilatation may be necessary. Malignant strictures, due to the high radial force compressing the ductal system or the intruding mass within the duct, often require the combination of balloon dilatation, temporarily drainage and stent placement. Stenting of the biliary tract is performed both in benign and malignant stenoses. There are three types of stents, which are used in biliary intervention: plastic endoprostheses (polyethylene stents), balloon-expandable metallic stents, and self-expandable metallic stents with and without covering. They can be placed either percutaneously or endoscopically. Metallic stents become strongly adherent to the bile duct and are impossible to remove. In the setting of malignancy, stents are mainly indicated for the palliation of patients with unresectable tumors. In such cases, metallic stents are preferred because they remain patent much longer than polyethylene stents and usually a single session of metal stenting maintain ductal patency for the reminding lifetime of the patients (6). Occlusion of metallic stents usually is determined by tumoral in-growth and can create the need for further interventions. For treatment of benign biliary strictures, plastic endoprostheses are preferred, since they are needed in most cases only for limited time. Nevertheless, in chronic recurrent biliary stenoses where a long-term treatment is needed, they may be occluded by biliary encrustations, while in metallic stents reactive epithelial hyperplasia may occur.

Diagnosis The diagnosis of biliary obstruction is suggested by the presence of jaundice, hypocholic stools, dark urine, itching, increase of serum levels of bilirubin, with prevalence of conjugated bilirubin, and alkaline phosphatase. Imaging has the role to confirm biliary obstruction and to establish the level and the cause of obstruction. US is performed in the initial evaluation to demonstrate bile duct dilatation, the level of obstruction and sometimes the cause of obstruction, such as stones, hepatic masses, pancreatic masses, lymphadenopathies. The main

limitations of US are the inconsistent visualization of the distal tract of the choledochal duct and the retroperitoneum. CT provides a more comprehensive examination that permits evaluation of the liver, biliary tree, pancreas, portal and retroperitoneal lymph nodes, and vascular structures, but it has a limited value in case of radiolucent stones of common bile duct. MRCP offers a noninvasive accurate assessment of the biliary tract. Associated cross-sectional images adequately demonstrate extra-ductal findings (1). Invasive techniques include ERCP and PTC. Currently they should be performed only to guide interventional procedures and cannot be recommended as diagnostic imaging techniques (2).

Bibliography 1.

2.

3. 4. 5.

6.

Soto JA, Alvarez O, Lopera JE et al (2000) Biliary obstruction: findings at MR cholangiography and cross-sectional MR imaging. Radiographics 20:353–366 Stabile Ianora AA, Memeo M, Scardapane A et al (2003) Oral contrast-enhanced three-dimensional helical-CT cholangiography: clinical applications. Eur Radiol 13:867–873 Vakil N, Everbach EC, and Knyrim K (1993) Pathogenesis and treatment of gallstones. N Engl J Med 24, 328(25)1855 Portincasa P, Moschetta A, and Palasciano G (2007) Cholesterol gallstone disease. Lancet 15, 368(9531):230–239 Vos PM, van Beek EJ, Smits NJ et al (2000) Percutaneous balloon dilatation for benign hepaticojejunostomy strictures. Abdom Imaging 25(2):134–138 Vitale GC, Larson GM, George M et al (1996) Management of malignant biliary stricture with self-expanding metallic stent. Surg Endosc 10(10):970–973

Occlusion, Bowel in Childhood M ELANIE H IORNS Radiology Department Great Ormond Street Hospital for Children London, UK [email protected]

Synonyms Small bowel obstruction (SBO); Large bowel obstruction

Definition The lumen of the small or large bowel is occluded. This may be due to extrinsic compression of the bowel, an intrinsic abnormality of the wall or lumen of the bowel, or due to a filling defect in the lumen of the bowel. It may be congenital or acquired. Complete occlusion of the bowel is described as ‘obstruction’.


Pathology/Histopathology Any of the pathologies listed in the tables may give rise to bowel obstruction and if the diagnosis is delayed this may go on to cause bowel ischaemia with necrosis and possible perforation.

Clinical Presentation The infant or child will usually present with abdominal distension, irritability, pain and vomiting, or high nasogastric aspirates if a tube is in place. The timing of the clinical presentation may be partly determined by the underlying causes: congenital causes will usually present in the first few hours or days of birth and 95% of small bowel obstruction in the perinatal period is due to an atresia of some type. Meconium ileus will present within the first 48h of life and occurs almost exclusively in

Congenital causes of bowel occlusion

Acquired causes of bowel occlusion

Stenosis (jejunal, ileal, colonic) Atresias (jejunal, ileal, colonic) ▶Meconium ileus (1) ▶Small left colon syndrome Annular pancreas Duplication cysts

Intussusception

Hirschsprung’s disease

Malrotation with midgut volvulus Postoperative adhesions Ileus Caecal volvulus Necrotising enterocolitis with secondary strictures Incarcerated hernias (inguinal, umbilical, omental (continued)

Acute complete bowel occlusion

Chronic partial bowel occlusion

Intussusception Incarcerated hernias Malrotation with midgut volvulus

Hirschsprung’s disease Caecal volvulus Inflammatory bowel disease including Crohn’s disease Post-operative adhesions

Post-operative adhesions Annular pancreas Meconium ileus Necrotising enterocolitis strictures

(continued)

1395

patients with cystic fibrosis. Fifteen to twenty per cent of cystic fibrosis patients present in this way (2). Small left colon syndrome (also termed meconium plug syndrome and functional immaturity of the colon) presents with failure to pass meconium and an increasingly dilated abdomen. Intussusception is the most common cause of obstruction in infants of 3 to 6 months. Strictures secondary to NEC most commonly occur within 1 to 6 months of the initial episode and are usually in the terminal ileum and the colon. Post-operative adhesions may occur at any time but most frequently in the first 6 months following surgery and in approximately 2% of patients who have had a laparotomy, accounting for 7% of small bowel obstruction overall.

Imaging The initial investigation is usually an abdominal radiograph, but this may be supplemented by an upper GI series in the case of a suspected high obstruction or by a contrast enema for a suspected low obstruction. Ultrasound has been used in some centres to examine the large bowel following the introduction of saline per rectum. The abdominal radiograph will show dilated gas and fluid-filled loops of bowel (Fig. 1). In the neonate and infant it may not be possible to differentiate small and large bowel obstruction and hence the level of obstruction is described as high or low in the GI tract depending on the pattern of distended loops. If only a few high central loops are seen this is termed a ‘high’ obstruction and the level of obstruction is most likely to be in the small bowel. If distended loops are seen throughout the abdomen it is likely to be a ‘low’ obstruction with the level of obstruction being in the most distal ileum or in the colon. Further investigation of a high obstruction is by an upper GI series and of a low obstruction by a contrast enema. Watersoluble contrast should be used in both cases due to the increased risk of perforation and the high probability of the patient subsequently undergoing surgery to resolve the obstruction. On a contrast enema a long filling defect of meconium in the left side of the colon is indicative of small left colon syndrome and the colon proximal to this will be dilated (Fig. 2). The neonate will usually clear the meconium plug spontaneously after the contrast enema. However, meconium ileus will demonstrate a micro-colon on contrast enema (Fig. 3). The colon will be thin and long, having not been used and dilated loops of bowel will be confined to the small bowel loops above the level of obstruction by meconium in the distal ileum. Meconium ileus is the only cause of obstruction that may be treated in the fluoroscopy room. A standard contrast enema is performed first to establish the diagnosis. Having established the diagnosis, ensuring that

O

1396


Occlusion, Bowel in Childhood. Figure 1 Abdominal X-ray showing typical appearances of a high (upper GI) obstruction. The stomach and two loops of bowel are shown to be dilated and gas filled. Bowel gas is absent below this level indicating complete obstruction. Fluid levels are present as this film was taken erect. The patient was confirmed to have jejunal atresia.

Occlusion, Bowel in Childhood. Figure 3 Contrast enema showing a micro-colon and meconium causing obstruction in the distal ileum in a neonate with meconium ileus.

standard non-ionic water soluble contrast has reached beyond the obstructing meconium to the dilated bowel loops may be enough to precipitate clearing of the meconium. However, some centres value the use of a gastrograffin® enema: if the infant is stable and appropriately fluid resuscitated, a ▶gastrograffin enema can be performed using diluted gastrograffin (for details see also entry ‘meconium ileus’). The gastrograffin should be instilled until it is observed to have reached and be filling dilated loops of bowel, therefore being above the level of the obstructing meconium. Gastrograffin is hyperosmolar and is believed to draw water into the gut thereby loosening the sticky meconium, allowing it to be passed, and relieving the obstruction.

Nuclear Medicine Has no role to play in this condition.

Diagnosis Occlusion, Bowel in Childhood. Figure 2 Contrast enema showing in ‘small left colon syndrome’ with dilated colon seen proximally and meconium plugs seen in the descending colon.

The diagnosis of obstruction is made by a combination of the clinical findings and the imaging as earlier. The underlying cause is usually confirmed at surgery.

Occlusion, Venous Central, Benign

Interventional Radiological Treatment Only for meconium ileus as described above (also see entry ‘meconium ileus’); note that sonographically guided saline reduction of meconium ileus is performed in some centres at the bedside of babies too ill to be transported to the fluoroscopy suite (3); some also recommend the use of a C-arm for fluoroscopic guidance on such occasions.

Bibliography 1.

2.

3.

Berrocal T, Lamas M, Gutieerrez J et al (1999) Congenital anomalies of the small intestine, colon, and rectum. Radiographics 19(5): 1219–1236 Agrons GA, Corse WR, Markowitz RI et al (1996) Gastrointestinal manifestations of cystic fibrosis: radiologic–pathologic correlation. Radiographics. 16(4):871–893 Riccabona M, Haim M, Kutschera J (2006) Sonografically guided reduction of meconium ileus in preterm neonates. Eur Radiol 16(1):284

Occlusion, Venous Central, Benign PATRICK H AAGE Department of Diagnostic and Interventional Radiology, University Hospital Witten/Herdecke, Wuppertal, Germany [email protected]

Synonyms Benign central venous obstruction; Benign central venous thrombosis

Definition Acute or chronic thrombotic uni- or bilateral occlusion of one or more central veins.

Pathology/Histopathology Central venous occlusion can broadly be divided into two eliciting categories: benign and malignant. Regardless of the underlying malady, early detection and treatment of complications is essential to provide adequate care for patients suffering from central venous obstruction. The most common cause for benign central venous obstruction is hemodialysis related; other benign reasons are rather

1397

uncommon, but are picking up due to the increased use of permanent central venous access catheters and implantable cardiac rhythm management devices.

Clinical Presentation Clinically these patients present with arm swelling and occasionally obvious widespread subcutaneous collateral vessels around the shoulder and thoracic aperture. Additional swelling of the face, neck, and breast may develop.

Imaging Consequently, the diagnostic and therapeutic regimen of hemodialysis related central venous obstruction will be the centre of attention and discussed herein. In the hemodialysis patient, chronic swelling of the access arm is the most indicative clinical symptom of central venous obstruction. Striking superficial collaterals veins may be observed accompanied by pain and paresthesia. In such an obvious case of impeded central venous flow, digital subtraction angiography of the fistula or graft and the complete venous outflow tract must be executed, since the central veins cannot be confidently examined with ultrasonography. Direct antegrade puncture of the access is suggested (1).

Diagnosis Since clinical diagnosis is often unreliable, imaging techniques need to be incorporated in the diagnostic process.

Interventional Radiological Treatment Interventional treatment of central venous lesions is indicated when they are impairing hemodialysis or arm swelling is painful and limiting. Reported primary patency rates in patients treated with PTA alone were 10% or less at one year with frequent restenoses (2). Primary stent implantation has clearly been shown to improve primary one-year patency rates to 56% and more, similar to those reported from surgical intervention (1, 3). Yet, due to the invasiveness of surgery for central venous obstructions, the less invasive percutaneous interventional therapy can be considered primary choice for treatment (1). Regular follow-up and reinterventions are however required to maintain patency and achieve long-term clinical success (4). Reports show that

O

1398

Occlusion, Venous Central, Benign

symptomatic central venous obstruction in dialysis patients can be treated with a high success rate through radiological intervention (5, 6).

Regarding the placement technique, stent placement should evade overlapping the ostium of a patent internal jugular vein to achieve a secure and satisfactory result, since this latter vein is important for future placement of central venous catheters. Correspondingly, a stent placed in the innominate vein if possible should not overlap the ostium of the contralateral vein; otherwise contralateral stenosis may come about and prohibit later use of the contralateral arm for access creation (1). A suitable endoprosthesis for central veins should be flexible enough to be used in curved and tortuous vessels. To avoid stent dislocation and proximal embolization, a self-expanding stent is necessary, in view of the fact that venous occlusions may undergo progressive luminal enlargement after stent placement (Fig. 1). Mechanical thrombectomy should not be regularly used as a primary therapy for dialysis-related central venous occlusions, because of the sharp angles and slim vessel walls observed in this vascular region. Furthermore, modest data are available on the application of thrombolytic agents in hemodialysis-related ▶benign central venous thrombosis. It can therefore not be recommended as a primary treatment regimen. In any case reocclusion is a frequently observed complication and is more likely to occur after thrombosis has occurred for the first time (7). The radiologist should be prepared for repeat interventions. To summarize, treatment of the hemodialysis patient population is specific due to the unusual underlying pathophysiology in dialysis outflow veins, which are exposed to much higher flow volumes. In the event of hemodialysis-related central venous occlusion, primary stent deployment is very effective in ensuring long-term vascular access for hemodialysis with superior long-term patency rates compared with percutaneous PTA alone or other therapeutic approaches.

References 1.

2. 3. 4.

Occlusion, Venous Central, Benign. Figure 1 (a) Digital subtraction angiogram demonstrates complete occlusion of the right brachiocephalic vein draining a Brescia–Cimino fistula; (b) & (c) after unsuccessful transbrachial negotiation of the occlusion, angiogram shows restoration of flow and vanishing of collateral veins after placement of a self-expanding stent and subsequent PTA via a transfemoral approach.

5.

6.

7.

Haage P, Vorwerk D, Piroth W et al (1999) Treatment of hemodialysis-related central venous stenosis or occlusion: results of primary Wallstent placement and follow-up in 50 patients. Radiology 212:175–180 Sprouse 2nd, Lesar CJ, Meier 3rd et al (2004) Percutaneous treatment of symptomatic central venous stenosis. J Vasc Surg 39:578–582 Mickley V (2001) Stent or bypass ? Treatment results in benign central venous obstruction. Zentralbl Chir 126:445–449 Oderich GS, Treiman GS, Schneider P et al (2000) Stent placement for treatment of central and peripheral venous obstruction: a longterm multi-institutional experience. J Vasc Surg 32:760–769 Dammers R, de Haan MW, Planken NR et al (2003) Central vein obstruction in hemodialysis patients: results of radiological and surgical intervention. Eur J Vasc Endovasc Surgery 26:317–321 Kovalik EC, Newman GE, Suhocki P et al (1994) Correction of central venous stenoses: use of angioplasty and vascular Wallstents. Kidney Int 45:1177–1181 Gaylord GM, Taber T (1993) Longterm hemodialysis access salvage: problems and challenges for nephrologists and interventional radiologists. J Vasc Interv Radiol 4:103–107

Occlusion, Venous Central, Malignant

Occlusion, Venous Central, Malignant PATRICK H AAGE Department of Diagnostic and Interventional Radiology University Hospital Witten, Herdecke, Wuppertal, Germany [email protected]

Synonyms Malignant central venous obstruction; Malignant central venous thrombosis

Definition Acute or chronic thrombotic uni- or bilateral occlusion of one or more central veins.

Pathology/Histopathology The central venous vasculature, in particular the superior caval venous system, may be obstructed by two types of lesions. The so called SVC syndrome is in more than 90% of cases caused by malignancy. Bronchogenic carcinoma is the most common causative malignant tumor and often leads to edema of the upper thorax, shortness of breath, cough, dysphagia, hemoptysis, and headaches. Less often, direct extension or compression due to the primary tumor or by invasion of the mediastinal lymph nodes is triggered by lymphoma, extra-thoracic tumors, mesothelioma, and lymph node metastases (1). Benign diseases causing SVC obstruction are often iatrogenic due to central venous catheters and pacemaker leads.

Clinical Presentation Clinical signs and symptoms of SVC obstruction (SVCO) are scored according to Kishi (2, Table 1). Clinical manifestations of venous obstruction can be extremely serious, requiring prompt treatment (also see ▶Benign central venous occlusion). Although the primary diagnostic suspicion can be rendered clinically, imaging is required for demonstrating the extent of the pathology.

Imaging Contrast venography for decades has been the standard of reference for benign and malignant central venous

1399

obstruction; however this procedure has its shortcomings. Venous puncture can be challenging in a swollen extremity, the procedure may cause thrombophlebitis and there is a low risk of an allergic contrast agent reaction. Ultrasonography is not reliable for detection of central venous pathologies, owing to difficult access to these vessels. In addition enlarged collateral veins and nonocclusive thrombi may cause false negative results. Sensitivity can be improved with the demonstration of normal cardiac pulsatility and respiratory phasicity within the examined vessels (read ▶Thrombosis, vein brachial).

Occlusion, Venous Central, Malignant. Table 1 Kishi scoring system for superior vena cava obstruction with the total score for signs and symptoms calculated as the sum of the highest grades in each category Signs and symptoms grade Neurologic symptoms Stupor, coma, or blackout 4 Blurry vision, headache, dizziness, or amnesia 3 Changes in mentation 2 Uneasiness 1 Laryngopharyngeal or thoracic symptoms Orthopnea or laryngeal edema 3 Stridor, hoarseness, dysphagia, glossal edema, or shortness of breath 2 Cough or pleural effusions 1 Nasal and facial signs or symptoms Lip edema, nasal stiffness, epistaxis, or rhinorrhea 2 Facial swelling 1 Venous dilatation Neck vein or arm vein distension, upper extremity swelling, or upper body plethora 1

Nowadays contrast enhanced spiral or preferably multislice computed tomography is employed to define the site of the obstruction and the presence of possible thrombosis and reveal surrounding soft tissue alterations (3). Alternatively, MRI is of comparative or even higher sensitivity and specificity in evaluating the patency of the central chest veins and may just as well hint at impending SVCO (▶Thrombosis, caval vein inferior). Another drawback of digital subtraction angiography is that it can only evaluate one single venous drainage system for each injection and other major draining vessels, for instance the internal jugular veins, may remain indeterminate. CT and MR venography more clearly depict the site and extension of the obstruction in clinically relevant venous vessels segments (4).

O

1400

Occlusion, Venous Central, Malignant

Both cross-sectional modalities provide a fast, virtually noninvasive evaluation of the central chest veins. If a percutaneous therapy is anticipated, naturally, digital subtraction venography should be carried out immediately prior to, during, and after the intervention.

Diagnosis Due to the deficiencies of a clinical diagnosis, imaging techniques need to be incorporated in the diagnostic process.

Interventional Radiological Treatment Especially in acute central venous obstruction, traditional treatment methods in malignancy have been nonoperative, such as steroids, radiation therapy, and chemotherapy. These however may require up to 4 weeks to show an effect and thus often are too time consuming (5). In benign and malignant lesions, anticoagulation alone is not efficient but may be used combined with other treatment modalities (6). Thrombolysis is ineffective in 75% of cases when the event is older than 5 days. In early chronic and chronic occlusions, thrombolysis is outright unsatisfactory (7). If a stenosis is the trigger for thrombosis, sole lysis will also be inefficient (8). In conclusion, an approach which offers urgent and rapid nonoperative relief should be the preferred treatment of choice. Admittedly, with balloon angioplasty alone early restenosis can be expected; plus, interventional success is limited because of the well-known fibrous and elastic features of venous lesions (6, 8). It however can be valuable before stenting by allowing the stent to uncomplicatedly cross relatively tight lesions. For all the above mentioned reasons percutaneous endovascular stenting of obstructive central venous lesions, which are symptomatic and caused by benign or malignant lesions, is an effective therapeutic option with acceptable patency rates and proven efficacy (9). Stenting results in a rapid and consistent relief and maintains patency throughout the life span of most patients suffering from malignant tumors. Different vascular access sites like the femoral vein, internal jugular vein, subclavian vein, and basilic vein are possible. Recanalization can be attempted with a hydrophilic-coated steerable 0.035 inch guide wire or a straight guide wire with a movable core, combined with a selective catheter. The obstruction may require predilation after safe passage through the segment, but only if presence of thrombus material can be excluded. To avoid venous rupture, which in the worst case may lead to cardiac tamponade, PTA should be performed carefully and by hand. Next, a stent which is flexible enough to allow implantation even in kinked vessels should be introduced.

Coverage of the obstructed segment is advised to be at least 1 cm free at the proximal and distal end to cover beyond the obstruction. Sometimes the placement of an additional stent may be necessary, especially if there is obstruction of both anonymous veins and the superior vena cava. In this case recanalization and revascularization of one anonymous vein lead to good clinical results and are associated with fewer complications provided that sufficient venous collaterals from left to right or vice versa are present (10). Stent size should be adapted to the diameter of the adjacent nonobstructed vessel segment. Postprocedural balloon dilatation is advised. Stent size should be at least 10% above the venous diameter. To avoid stent dislocation and central embolization, a self-adjusting, selfexpanding stent is advantageous because especially chronic venous occlusions may undergo progressive luminal enlargement after stent placement (11). Previously, many interventionalists preferred balloon expandable stents because of their flexibility and their marginal risk of migration. Current self-expanding stents however have overcome these problems of significant foreshortening and migration. Often a superimposed thrombosis can be observed which can be treated with thromboaspiration or fibrinolytic therapy before stenting (1). The significant bleeding risk in patients with corresponding contraindications must be considered though. The chance of hematoma formation and gastrointestinal and intracerebral hemorrhage must be taken into account. Nevertheless the presence of extensive thrombus may require the use of thrombolytics. The thrombolytic agent should be infused with the tip of the infusion catheter inside the thrombus at a rate of 0.02 mg tissue plasminogen activator/kg body weight/hr (12). To save time, pulse spray injection can be employed. Active thrombus removal with mechanical thrombectomy devices may be an adjunct or even alternative to intralesional thrombolysis, however it must be handled with care and expertise. The peri- and postprocedural anticoagulation for stent placement with or without additional thrombolysis is still unclear. Heparinization during stenting and postprocedural intravenous or subcutaneous heparin can be administered. Subsequent antiplatelet therapy, typically aspirin and/or clopidogrel is recommended. Thrombosis may occur after the stent insertion in up to 45% (13). Long-time anticoagulation therefore helps in avoiding clinical deterioration. To recapitulate, the efficacy and safety of stent placement in central venous occlusion of benign and malignant origin have been proven with rapid relief and less invasiveness for the often extremely ill patients. Stenting is widely accepted now; it provides fast and durable symptomatic relief and can nowadays be favored to radiation and chemotherapy or used in combination

Oesophageal Disease, Childhood

with them. If clinical symptoms are severe or worsen rapidly, stenting is indicated while surgical therapy should be reserved for patients undergoing refractory to percutaneous therapy. Repeated percutaneous intervention can prolong the cumulative patency rate. Pharmacologic/pharmacomechanical/mechanical thrombolysis may be necessary to improve the final result in case of superimposed central venous thrombosis; they should however not be employed as a single means for revascularization.

References 1. Thony F, Moro D, Witmeyer P et al (1999) Endovascular treatment of superior vena cava obstruction in patients with malignancies. Eur Radiol 9:965–971 2. Kishi K, Sonomura T, Mitsuzane K et al (1993) Self-expandable metallic stent therapy for superior vena cava syndrome: clinical observations. Radiology 189:531–535 3. Uberoi R (2006) Quality assurance guidelines for superior vena cava stenting in malignant disease.Cardiovasc Intervent Radiol 29:319–322 4. Kroencke TJ, Taupitz M, Arnold R et al (2001) Three-dimensional gadolinium-enhanced magnetic resonance venography in suspected thrombo-occlusive disease of the central chest veins. Chest 120:1570–1576 5. Nicholson AA, Ettles DF, Arnold A et al (1997) Treatment of malignant vena cava obstruction: metal stents or radiation therapy. J Vasc Interv Radiol 8:781–788 6. Schindler N, Vogelzang RL (1999) Superior vena cava syndrome: Experience with endovascular stents and surgical therapy. Surg Clin North Am 79:983–994 7. Gray BH, Olin JW, Graor RA et al (1991) Safety and efficacy of thrombolytic therapy for superior vena cava syndrome. Chest 99:54–59 8. Kee ST, Kinoshita L, Razavi MK et al (1998) Superior vena cava syndrome: Treatment with catheter-directed thrombolysis and endovascular stent placement. Radiology 206:187–193 9. Yim CD, Sane SS, Bjarnason H (2000) Superior vena cava stenting. Radiol Clin North Am 38:409–424 10. Dinkel HP, Mettke B, Schmid F et al (2003) Endovascular treatment of malignant superior vena cava syndrome: Is bilateral Wallstent placement superior to unilateral placement? J Endovasc Ther 10:788–797 11. Haage P, Vorwerk D, Piroth W et al (1999) Treatment of hemodialysis-related central venous stenosis or occlusion: results of primary Wallstent placement and follow-up in 50 patients. Radiology 212:175–180 12. Schifferdecker B, Shaw JA, Piemonte TC et al (2005) Nonmalignant superior vena cava syndrome: pathophysiology and management. Catheter Cardiovasc Interv 65:416–423 13. Kim YI, Kim KS, Ko YC et al (2004) Endovascular stenting as a first choice for the palliation of superior vena cava syndrome. J Korean Med Sci 19:519–522

Occupational Lung Diseases ▶Pneumoconioses

1401

Oesophageal Atresia A congenital abnormality in which the upper oesophagus is a blind-ending sac and is not continuous with the lower oesophagus. ▶Oesophageal Disease, Childhood ▶GI Tract, Paediatric, Congenital Malformations

Oesophageal Cancer ▶Neoplasms, Oesophagus

Oesophageal Clearance This term is used to describe the process by which the oesophagus is cleared of refluxed stomach acid. ▶Gastroesophageal Reflux in Adult Patients: Clinical Presentations, Complications, and Imaging

O Oesophageal Disease, Childhood M ELANIE H IORNS Radiology Department Great Ormond Street Hospital for Children London, UK [email protected]

Synonyms Achalasia; Gastro-oesophageal reflux; Gastro-oesophageal reflux (GOER); Oesophageal atresia; Oesophageal duplications (and other foregut duplications); Oesophageal foreign body; Oesophageal stenosis (corrosive ingestion); Oesophageal webs; Tracheo-oesophageal fistula

Definition ▶Oesophageal atresia is the congenital occlusion of the oesophagus. It has an incidence of between 1 in 3,000 to

1402


4,500 live births. Oesophageal atresia is associated with ▶tracheo-oesophageal fistula in more than 85% of cases. There are several different anatomic variations of the oesophageal atresia and the insertion of an associated tracheo-oesophageal fistula. Tracheo-oesophageal fistula is the abnormal connection between the trachea and the oesophagus and occurs in isolation only in 8% of cases. The remainder of the time it occurs in conjunction with oesophageal atresia. Oesophageal webs may be congenital or acquired and are a thin, 2–3 mm, eccentric, smooth extension of the normal oesophageal wall that can occur anywhere along the length of the oesophagus but typically is located in the anterior post-cricoid area of the proximal oesophagus. ▶Oesophageal duplications are part of a wider group of foregut duplications. Duplications may cystic or tubular and may contain gastric mucosa. They can be associated with sequestrations or congenital stenosis or atresia of the oesophagus. Oesophageal duplications may be separated from the oesophagus or may share a common wall. Duplications of the oesophagus are sometimes associated with vertebral anomalies and intraspinal cysts and often are associated with intra-abdominal intestinal duplications. ▶Oesophageal stenosis is a narrowing of the oesophagus at any point along its length and over a variable distance. A stenosis my be congenital or may be acquired such as after surgery for oesophageal atresia, secondary to acid or alkali ingestion, or to another disease process such as epidermolysis bullosa. ▶Achalasia: Abnormal dilatation and motility of the distal oesophagus with failure of relaxation of the lower oesophageal sphincter. ▶Oesophageal foreign body: An object that is ingested but does not pass freely through the oesophagus. The most common sites for hold-up are at the level of cricopharyngeus, at the aortic knuckle and at the gastrooesophageal junction. ▶Oesophageal inflammatory change: Inflammation of the mucosa of the oesophagus in response to an irritant. This in most commonly secondary to reflux or to ingested corrosives but may also be seen in Crohn’s disease, epidermolysis bullosa, pemphigoid and other systemic or dermatological diseases. ▶Gastro-oesophageal reflux: Reflux is the retrograde movement of fluid across a sphincter and in the context of the oesophagus is the movement of stomach contents up into the oesophagus.

Pathology/Histopathology Oesophageal atresia is due to posterior deviation of the tracheo-oesophageal septum leading to incomplete

separation of the oesophagus from the laryngo-tracheal tube. There is failure of recanalisation of the oesophagus in the eighth week of foetal development. There are several variations on how the fistula may connect the trachea and oesophagus. The most common is that of a blind-ending upper pouch with a fistula then connecting the distal trachea to the lower, patent, part of the trachea. A fistula that joins a patent oesophagus to a patent trachea is termed an ‘H’-shaped fistula (8%). Oesophageal atresia and tracheo-oesophageal fistula is the most common congenital malformation of the oesophagus (1). Oesophageal webs may represent incomplete recanalisation of the oesophagus during foetal life. They are made of the normal structures in the wall of the oesophagus. Oesophageal duplications: It is believed that these anomalies are due to failure of the notochord to detach from the endoderm, resulting in a persisting neurenteric canal. They often contain gastric mucosa. Oesophageal stenosis: Congenital stenosis is believed due to poor canalisation of the oesophagus in embryonic life. Acquired stenoses are usually secondary to an insult and will therefore show scar tissue. ▶Achalasia: Aetiology unknown. Oesophageal foreign body: No specific pathology. Oesophageal inflammatory change: The inflammatory change may be superficial involving only the mucosal surface or maybe transmural. Gastro-oesophageal reflux: In GOER the lower oesophageal sphincter (LES) opens inappropriately or is incompetent and there is transient complete relaxation of the sphincter. Due to a pressure difference across the LES (with negative pressure in the oesophagus during inspiration) or a transient rise in intra-abdominal pressure the stomach contents move up into the oesophagus This may result in chronic inflammation in the oesophagus and in severe cases can lead to stricture formation.

Clinical Presentation Oesophageal atresia and tracheo-oesophageal fistula: Most neonates present at birth with difficulty swallowing secretions, drooling, choking and respiratory distress as secretions and feeds spill over into the airway. In complete OA, the neonate will present with difficulty swallowing secretions and will be unable to tolerate the first feed. If a fistula is present between the trachea and oesophagus, air can still reach the gastrointestinal (GI) tract. An ▶H-fistula may present later with blue episodes or repeated chest infections and aspiration as some fluid tracks from the oesophagus through the fistula into the airway. A blind-ending upper pouch is often detected


when there is failure to pass a nasogastric tube and/or this becomes coiled in the upper oesophagus. With the increase in antenatal scanning oesophageal atresia may be detected in utero. The mother may present with polyhydramnios and on ultrasound there will be very little if any fluid present in the stomach or GI tract of the foetus. A complication of repair of oesophageal atresia is a stricture of the oesophagus and the child may present with dysphagia and regurgitation of food. Oesophageal web: The very young child will present with drooling and aspiration, and may suffer repeated chest infections. The older child will may have dysphagia and regurgitation of solids. Oesophageal duplication: Completely separated duplications can present as a mediastinal mass and may or may not be associated with symptoms. Incompletely separated duplications may give rise to dysphagia, regurgitation and rarely airway obstruction (2). Oesophageal stenosis: Dysphagia and regurgitation. Achalasia: Chronic regurgitation of undigested food, repeated aspiration pneumonia, retrosternal discomfort in the older child. Oesophageal foreign body: The patient will usually be drooling and may be dysphagic. Oesophageal inflammatory change: Presentation will depend on the underlying cause but the patient will usually have retrosternal pain. With chronic change going on to stricture formation they may present with dysphagia. Gastro-oesophageal reflux: GOER is most common in infants and young children and a minor degree of reflux in the first few months of life may even be considered within normal levels. The symptomatic infant will present with recurrent regurgitation of feeds and/or pain. In severe cases, the child may fail to gain weight or even lose weight if it is not possible to maintain adequate calorific intake. Older patients may present with retrosternal pain, food aversion or less commonly with a stricture. Alternatively, young children and patients with neurologic deficits may present with recurrent chest infections secondary to aspiration as a result of GOER.

Imaging Oesophageal atresia and tracheo-oesophageal fistula: A chest X-ray will often show an air-filled, dilated and blind-ending upper oesophageal pouch. If the passing of a nasogastric tube (Fig. 1) has been attempted this is likely to be coiled in the upper pouch (1). A Replogle tube (identified by a dashed radio-opaque marker allowing easy identification on X-ray) may have been placed in the upper oesophagus to allow aspiration of the secretions.

1403

Oesophageal Disease, Childhood. Figure 1 Chest X-ray showing a nasogastric tube coiled in a blind-ending upper pouch in a patient with oesophageal atresia.

This is a dual lumen tube which allows secretions to be continually sucked from the blind-ending upper oesophagus thereby preventing aspiration. It is used almost exclusively in this condition. The abdomen will be gasless unless there is an associated fistula, and indeed the presence of gas in the abdomen will confirm that a fistula is present, this being the only route for air to get from the airway into the GI tract. A contrast study is not required and the immediate management is usually surgical. Imaging for an H-type fistula (a fistula between a patent airway and a patent oesophagus) is by a tube oesophagram. With the patient prone and with lateral screening water-soluble contrast is injected into the oesophagus whilst the nasogastric tube is slowly withdrawn. If a fistula is present, the contrast should flow through the fistula from the oesophagus into the airway. This procedure therefore carries a significant risk of acute respiratory compromise and should only be undertaken if suction, oxygen, nursing support and appropriate resuscitation facilities are available. Stricture formation as a delayed complication of repair of oesophageal atresia is best demonstrated by upper GI (UGI) series (Fig. 2). Oesophageal web: Imaging is by upper GI series or direct visualisation on endoscopy. Oesophageal duplication: If the duplication is not completely separated it may be demonstrated on an upper GI series (Fig. 3). A closed duplication may be best demonstrated on MRI of the mediastinum, although CT and ultrasound may also be used. Ultrasound would generally use an approach through the sternal notch or between the ribs, but in very young patients a transsternal approach can sometimes be used as the sternum has not ossified.

O

1404


Oesophageal Disease, Childhood. Figure 3 Contrast outlines an oesophageal duplication with two, parallel, fluid-filled cavities opacified. Oesophageal Disease, Childhood. Figure 2 Upper GI series demonstrating a tight oesophageal stenosis at the site of previous surgery for oesophageal atresia with tracheo-oesophageal stenosis.

Oesophageal stenosis: Upper GI series will show the level and extent of the narrowing. Achalasia: An upper GI series will show the accumulated food debris in the dilated oesophagus and poor opening of the lower oesophageal sphincter with a characteristic fine jet of contrast passing through or an abrupt point (‘birds beak’) at the gastro-oesophageal junction through which no contrast passes. Oesophageal foreign body: Chest X-ray (AP and lateral) will show radio-opaque foreign bodies; an upper GI series may demonstrate the filling defect on a radiolucent foreign body. Endoscopy allows direct visualisation and simultaneous removal (see entry Foreign body, ingestion, children). Oesophageal inflammatory change: Very subtle superficial changes may not be detected by imaging and will only be identified under direct vision with endoscopy. An upper GI series using barium will demonstrate mucosal surface irregularity with an uneven appearance to the barium column (Fig. 4). If a double contrast image is achieved ulceration may be visible. Endoscopic ultrasound will show the degree of intramural thickening.

Gastro-oesophageal reflux: An upper GI series or ultrasound of the lower oesophagus are the modalities of choice. Reflux is usually self-evident in infants by the regurgitation of feeds so much of the value of the UGI is to exclude anatomical causes for the reflux such as hiatus hernia or gastric outlet obstruction and in preparation for surgery if a fundoplication is being considered. GOER is characteristically intermittent and it would be inappropriate to image the patient with prolonged screening to try and demonstrate an occasional episode. Only intermittent screening should be performed and the absence of demonstrable reflux during a UGI does not exclude the diagnosis. Ultrasound is increasingly used to demonstrate relaxation of the lower oesophageal sphincter and fluid tracking into the lower oesophagus. It has the advantage that the sphincter can be observed directly and for prolonged periods but the disadvantage that it cannot show the rest of the stomach and duodenal anatomy. An UGI is also useful for assessment of the complications of GOER such as stricture formation.

Nuclear Medicine No role to play in this condition.

Oesophageal Stenosis (corrosive ingestion)

1405

Oesophageal Duplication A congenital abnormality in which there has been duplication of part of the foregut in embryological development resulting in either a cyst or second lumen of the oesophagus. ▶Oesophageal Disease, Childhood

Oesophageal Duplications (and other foregut duplications) ▶Oesophageal Disease, Childhood

Oesophageal Dysmotility

Oesophageal Disease, Childhood. Figure 4 Upper GI series with contrast outlining an irregular and narrowed oesophagus following ingestion of caustic fluid.

Diagnosis By imaging as detailed above. GOER may also be diagnosed by monitoring with pH probe.

Interventional Radiology Image guided balloon dilatation of strictures (3), and sometimes removal of oesophageal foreign bodies may be carried out by the interventional radiologist (see also entry ‘foreign body, ingestion, children’).

This may be either abnormal contractions of the oesophagus, or a diminution or failure of normal peristalsis. It has many causes, of which the most common in industrialized countries is gastro-oesophageal reflux disease. ▶Gastroesophageal Reflux in Adult Patients: Clinical Presentations, Complications, and Imaging

Oesophageal Foreign Body ▶Oesophageal Disease, Childhood ▶GI tract, Pediatric, Foreign Bodies

Oesophageal Malignancy ▶Neoplasms, Oesophagus

Bibliography 1.

2. 3.

Berrocal T, Madrid C, Novo S et al (2004) Congenital anomalies of the tracheobronchial tree, lung, and mediastinum: embryology, radiology, and pathology. Radiographics 24(1):e17 Stern LE, Warner BW (2000) Gastrointestinal duplications. Semin Pediatr Surg 9(3):135–140 Lan LC, Wong KK, Lin SC et al (2003) Endoscopic balloon dilatation of esophageal strictures in infants and children: 17 years’ experience and a literature review. J Pediatr Surg 38(12):1712–1715

Oesophageal Stenosis (corrosive ingestion) ▶Oesophageal Disease, Childhood

O

1406

Oesophageal Tumor

Oesophageal Tumor ▶Neoplasms, Oesophagus

Oesophageal Webs ▶Oesophageal Disease, Childhood

Oesophagitis V LASTIMIL VALEK 1 , J IRI D OLINA 2

. Miscellaneous 1. Skin disorders . Dystrophic epidermolysis bullosa . Pemphigus vulgaris . Bullous pemphigoid . Benign mucous membrane pemphigoid . Stevens–Johnson syndrome . Lichen planus . Acanthosis nigricans . Darier’s disease . Leukoplakia 1. Behcet’s disease 2. Graft-versus-host disease (GVHD) 3. Crohn’s disease 4. Ulcerative colitis 5. Sarcoidosis 6. Collagen vascular diseases.


1

Department of Radiology, University Hospital Brno, Brno, Czech Republic 2 Department of Gastroenterology, University Hospital Brno, Brno, Czech Republic [email protected] [email protected]

Synonyms Esophagitis; Gastroesophageal reflux disease (GERD); Oesophagitis

Definition In the literature the term “esophagitis” (oesophagitis) is sometimes used synonymously with “reflux esophagitis” because gastroesophageal reflux is the most common cause of esophagitis. However, numerous etiologic factors may cause esophagitis: . Infectious esophagitis 1. Fungal infections 2. Viral infections (a) Herpes simplex virus (HSV) (b) Cytomegalovirus (CMV) (c) Varicella-zoster virus (VZV) 3. Bacterial infections 4. Parasitic infections . Drug-induced esophagitis . Radiation-induced esophagitis . Caustic esophagitis . Reflux esophagitis

Infectious Esophagitis In Candida esophagitis, plaques consist of heaped-up areas of necrotic epithelial debris or actual colonies of C. albicans on the esophageal mucosa; the esophagus itself as an irregular or shaggy appearance. Three viral infections cause ulcerative esophagitis: HSV, CMV, and VZV. In HSV the earliest esophageal lesions are vesicles, the centers of which slough to form discrete, circumscribed ulcers with raised edges. Routine histologic stains may show epithelial cells with occasional multinucleation, “ground-glass” nuclear staining, and intranuclear inclusions. More specific and sensitive is immunohistologic identification with monoclonal antibodies to HSV antigen and in situ hybridization for HSV nucleic acid. CMV infection occurs only in immunosuppressed patients. CMV may be activated from latency or acquired from blood-product transfusions. CMV lesions first appear as serpiginous ulcers in otherwise normal mucosa but coalesce to form giant ulcers, particularly in the distal esophagus. Infection occurs within submucosal fibroblasts and endothelial cells, not in epithelium, and is usually part of widespread visceral infection. Routine histologic stains show large cells in the submucosa bearing amphophilic intranuclear inclusions and intracytoplasmic inclusions. Immunohistology with monoclonal antibodies to CMV antigens and in situ hybridization for CMV DNA are useful for finding infected cells that are neither large nor inclusion bearing.

Drug-induced Esophagitis In drug-induced esophagitis, affected individuals ingest the medication with little or no water immediately before

Oesophagitis

going to bed. When such drugs dissolve in the esophagus, they cause mucosal injury either by creating an acid pH (analogous to a form of caustic esophagitis) or by direct irritating the epithelium.

Radiation Esophagitis A radiation dose of 4,500–6,000 rad over a 6- to 8-week period or more to the mediastinum may cause severe injury to the esophagus. Acute radiation-induced esophagitis usually occurs 2–4 weeks after the initiation of radiation therapy. The mucosa typically has a granular appearance because of edema and inflammation. Ulceration and decreased luminal distensibility are other frequent findings.

Caustic Esophagitis Caustic esophagitis causes injury similar to that from thermal burns. Liquid lye causes liquefactive necrosis, resulting in the most severe form of caustic injury to the esophagus. Alkaline agents can produce deep coagulation necrosis in minutes. Necrosis from acids tends to be more superficial. The severity and extent of esophageal injury depend on the type, concentration, and volume of the caustic agent. By 2–6 weeks, healing is well in progress, often accompanied by severe fibrosis.

Gastroesophageal Reflux Disease, Esophagitis (Reflux Esophagitis) This type encompasses reflux esophagitis and is characterized histologically by inflammatory cells and reflux changes consisting of epithelial hyperplasia without inflammation. The multiple determinant factors in the production of reflux disease include the frequency and volume of GER, the volume of gastric contents available to reflux, the potency of the reflux material, the efficacy of esophageal clearance, and the tissue resistance to injury.

Miscellaneous Esophagitis Eosinophilic esophagitis (IEE) is a chronic inflammatory disease characterized by eosinophilic infiltration of the esophagus with an increased number of intraepithelial eosinophils (more than 20 eosinophils per high-power field) in endoscopic biopsy specimens from the esophagus. Alkaline reflux esophagitis is caused by reflux of bile or pancreatic secretions into the esophagus after partial or total gastrectomy. Pemphigus and pemphigoid are nonhereditary conditions in adults, and epidermolysis bullosa dystrophica occurs as an autosomal recessive condition in children. Pemphigus, pemphigoid, and epidermolysis bullosa can produce esophagitis.

1407

Pemphigus vulgaris affects skin, mouth, and other mucous membranes with weeping bullous lesions. Histology shows acantholysis and intraepithelial bullae, and specific immunohistology should be used. Bullous pemphigoid is a chronic disease of the elderly, in whom tense, pruritic skin bullae arise. Esophageal bullae occur rarely, with sloughing of the epithelium as a cast. The histology shows subepithelial bullae and circulating antibodies to the basement membrane. Behcet’s disease is a multisystem inflammatory process characterized by oral and genital ulcers, ocular inflammation, skin lesions, and vasculitis. Esophageal lesions include ulcerations that can tunnel under the mucosa, strictures, and perforations. GVHD is an immunologic reaction against host tissues by donor lymphoid cells. The esophagus shows webs, rings, and tight strictures but often fails to show generalized desquamation apparent on endoscopy. Crohn’s disease rarely affects the esophagus but may show small aphthous ulcers, inflammatory strictures, filiform polyps, and fistulas. Granulomas can occasionally be found in endoscopic biopsy specimens.

Clinical Presentation Infectious esophagitis is seen with increasing frequency in debilitated individuals, alcoholics, diabetics, immunocompromised or transplant patients, individuals with impaired T-cell function from acquired immunodeficiency syndrome (AIDS), and patients receiving steroids, antibiotics, radiotherapy, or chemotherapy. Candida esophagitis (and infectious esophagitis generally) is usually characterized by the abrupt onset of odynophagia (painful swallowing), chest pain, or dysphagia. Clinical symptoms can occasionally be absent. Symptoms of tuberculosis esophagitis include dysphagia and chest pain, but esophageal symptoms are commonly absent. Patients with HSV esophagitis present with odynophagia, retrosternal pain, and heartburn. Infection can also start with mild symptomatology of nausea and pyrosis (heartburn) or with very severe symptoms of hematemesis. The clinical presentation of CMV esophagitis is nausea, vomiting, painful swallowing, heartburn, and/or hematemesis. Symptoms of drug-induced esophagitis are usually sudden in onset and consist of chest pain, odynophagia, or dysphagia. A typical location is the aortic arch level, where tablets are delayed because of the aortic indentation on the esophagus and the low contractile force of peristalsis. Bleeding or esophageal perforation may occur. Radiation esophagitis typically produces mild heartburn or dysphagia several weeks after the onset of treatment. Some patients may have progressive dysphagia

O

1408

Oesophagitis

due to strictures 4–8 months after completion of radiation therapy. In caustic esophagitis, the initial clinical symptoms are the rapid onset of chest pain and dysphagia. These symptoms tend to resolve in several days. Acute complications include shock, fever, respiratory distress, mediastinitis, and perforation. Late complications are related primarily to fibrosis and stricture, which may cause dysphagia several weeks after the initial injury. Symptoms of GERD esophagitis (reflux esophagitis) can be divided into typical (esophageal) symptoms (heartburn, regurgitation, dysphagia) and atypical (extraesophageal) symptoms (chronic cough, respiratory complaints, laryngeal symptoms). Dysphagia is usually due to esophageal narrowing. Some patients complain of a lump in the throat (globus). In young children the predominant reflux symptoms are regurgitation, repetitive vomiting, and failure to thrive. IEE appears in the adult population. It typically affects young men (20–40 years of age) with long-standing dysphagia and recurrent food impactions who have not responded to the usual forms of antireflux therapy. Other findings include chest pain and vomiting.

Imaging Barium examination continues to be the primary radiologic modality for evaluating patients with dysphagia, reflux symptoms, or other clinical findings of esophageal disease. The double-contrast phase optimizes the ability to detect all kinds of esophagitis, particularly reflux disease, whereas the single-contrast phase optimizes the ability to detect hiatal hernia and lower esophageal rings or strictures. Barium contrast studies are useful for evaluating mucosal surface lesions but provide little information about the extramucosal extent of disease. Computed tomography (CT) and magnetic resonance imaging (MRI), on the other hand, permit evaluation of wall thickness, mediastinal involvement, adjacent lymphadenopathy, and distant spread.

Diagnosis Infectious esophagitis: Candida esophagitis is characterized on esophagrams by plaques or a “shaggy” esophagus. Plaquelike lesions are seen as linear or irregular filling defects that tend to be oriented longitudinally and are separated by normal mucosa. The luminal contour may show fine speculations, irregularity, a cobblestone pattern, or bizarre thickened folds simulating varices. Barium may dissect beneath a pseudomembrane, causing a shaggy contour that gives the appearance of ulceration. The

esophagus tends to be atonic. Peristalsis may be feeble or incomplete. Because of muscular hypotonia, the esophagus is generally slightly dilated or normal in caliber but may show areas of moderate narrowing. Monilia organisms are a frequent companion of esophageal intramural pseudodiverticulosis. Double-contrast esophagrams have been found to have a sensitivity of 90% for detecting Candida esophagitis, primarily because of the ability to show these plaques. CT findings are nonspecific and commonly seen in various kinds of esophagitis (circumferential esophageal wall thickening of >5 mm, with relatively long segmental involvement). Enhanced scans may also depict the target sign (circumferential wall thickening and enhancing internal mucosa). Morphologic abnormalities caused by tuberculosis esophagitis are often eccentric and may show skip areas. The luminal contour may show mild irregularity, large or deep ulcers, and sinus tracts. Fistulas are common. The esophageal wall is generally thickened and the lumen often narrowed. Enlarged mediastinal nodes may displace or compress the esophagus and widen the mediastinum. In HSV, barium contrast X-ray shows multiple ulcers. Small ulcers surrounded by edema give the appearance of targets or shallow irregularities in the profile view. Diagnosis requires endoscopic brushing and biopsies. The first, presumptive diagnosis can be made in patients with esophageal symptoms and HSV infections of the mouth or nares. CMV esophagitis is characterized by one or more giant flat ulcers that are several centimeters or more in length. The ulcers may have an ovoid or diamond-shaped configuration and are often surrounded by a thin radiolucent rim of edematous mucosa. Endoscopy with multiple biopsies targeting the center of the ulcer formation is mandatory. Brushing of overlying exudates is seldom useful. Drug-induced esophagitis: Signs are luminal irregularity, frank ulceration, and luminal narrowing. Tetracycline and doxycycline are associated with the development of small shallow ulcers in the upper or middle part of the esophagus. Other drugs may cause more severe esophageal injury. Radiation esophagitis: A morphologic abnormality consists of diffuse ulceration. Late findings consist of strictures that are generally smooth with tapering margins and that rarely show irregularity or ulceration. Caustic esophagitis: In the acute stage, the examination should be initiated with water-soluble contrast medium to exclude esophageal or gastric perforation. Such studies may also reveal marked edema, spasm, and ulceration of the affected esophagus. During the first week, characteristic frank ulcerations are seen. A pseudomembrane may cause intramural trapping of barium. Long, ulcerated strictures may be observed in patients who ingested lye or

Oil Cyst

other caustic agents, and in severe cases, diffuse esophageal narrowing may reduce the thoracic esophagus to a thin, tight stricture. Cross-sectional images depict narrowing or obliteration of the esophageal lumen, and perifibrotic tissue may be observed. Reformatted CT images may be useful for demonstrating the surrounding fibrotic change to which caustic esophagitis leads. GERD esophagitis: Reflux esophagitis manifests at esophagography as finely nodular or granular relief with poorly defined radiolucencies that fade peripherally due to edema and inflammation of the mucosa. Endoscopy is the gold standard for diagnosing erosive GERD, and the Los Angeles classification for esophagitis is generally accepted as the endoscopic assessment of GERD. Most patients—approximately 60% referred for endoscopy with typical reflux symptoms—do not have erosive reflux disease. The spectrum of GERD can be subdivided into endoscopy-negative, nonerosive GERD (NERD), reflux esophagitis (GERD), and GERD with esophageal columnar metaplasia (Barrett). Further changes include irregularity of luminal contours, discrete ulcerations, transverse esophageal folds, thickened longitudinal folds, esophageal wall thickening, smooth polypoid protuberance (also known as an inflammatory esophagogastric polyp), and segmental narrowing. The ulcers can have a punctate, linear, or stellate configuration and are often associated with a surrounding halo of edematous mucosa, radiating folds, or sacculation of the adjacent wall. Predominant involvement of the distal esophagus and the presence of an associated hiatal hernia and gastroesophageal reflux should suggest the correct diagnosis. Thickened folds wider than 3 mm are best seen on mucosal relief views of the collapsed esophagus. Multiple delicate transverse folds 1–2 mm wide may also be found in patients with gastroesophageal reflux disease. Between 10 and 20% of patients with reflux esophagitis develop peptic strictures as a result of circumferential scarring of the distal esophagus. The classic appearance of a smooth, tapered area of concentric narrowing in the distal esophagus above a sliding hiatal hernia should be virtually pathognomonic of a benign peptic stricture. Most peptic strictures range from 1 to 4 cm in length and from 0.2 to 2.0 cm in width. Between 25 and 50% of patients with reflux esophagitis have abnormal esophageal motility, manifested by feeble or absent primary peristalsis associated with an increased frequency of nonperistaltic contractions. Miscellaneous esophagitis: Alkaline reflux esophagitis is characteristic by mucosal nodularity or ulceration or, in severe disease, by distal esophageal strictures that often progress rapidly in length and severity over a short period of time. During the acute stage of pemphigus, pemphigoid, and epidermolysis bullosa esophagitis, bullae may

1409

cause multiple esophageal filling defects on esophagograms, or ruptured bullae may appear as ulcerations. Repetitive insults often lead to strictures that are generally smooth but sometimes have an irregular contour. IEE manifests at esophagography as segmental esophageal strictures (sometimes ringlike—the so-called ringed esophagus at endoscopy and/or barium studies) and occasionally with diffuse esophageal narrowing, which produces a “small-caliber” esophagus.

Interventional Radiological Treatment Esophageal dilation is usually indicated for benign stenoses and strictures, which can be caused by esophagitis. The main indications are strictures after caustic and reflux esophagitis. All esophageal strictures should be carefully evaluated with esophagography or endoscopy before dilation.

Bibliography 1.

2.

3. 4.

Therasse E, Oliva VL, Lafontaine E et al (2003) Balloon dilation and stent placement for esophageal lesions: indications, methods, and results. Radiographics 23:89–105 Zimmerman SL, Levine MS, Rubesin SE et al (2005) Idiopathic eosinophilic esophagitis in adults: the ringed oesophagus. Radiology 236:159–165 Levine MS, Rubesin SE (2005) Diseases of the esophagus: diagnosis with esophagography. Radiology 237:414–427 Jang KM, Lee KS, Lee SJ et al (2002) The spectrum of benign esophageal lesions: imaging findings. Korean J Radiol 3:199–210

Oesophagogram Radiologic examination where images are obtained of the oesophagus during drinking of either barium sulphate suspension or iodine contrast medium. When an effervescent agent is used a double-contrast effect may be obtained. ▶Diverticulum, Oesophagus

Oil Cyst Conglomerate of almost entirely pure neutral fat, encapsulated by a thin, smooth, fibrous wall in which calcium deposition may occur. ▶Trauma, Breast

O

1410

OLF

OLF ▶Ossification or Calcification of Ligamenta Flava (OLF)

OLT ▶Transplantation, Liver ▶Orthotopic Liver Transplantation

Omphalocele Persistence of herniation of the abdominal contents into the base of the umbilical cord at the time of birth and associated with a high incidence of other congenital anomalies. ▶GI Tract, Paediatric, Congenital Malformations

Open and Closed Spinal Dysraphisms Etymologically, the term “dysraphism” implies defective closure of the neural tube, and should therefore be used to refer to abnormalities of primary neurulation only. However, the term has gained widespread use as a synonym to congenital spinal cord malformation. Spinal dysraphisms are categorized into open (OSD) and closed (CSD). OSDs are characterized by exposure of nervous tissue to the environment through a congenital defect of the child’s back. On the contrary, CSDs are covered by skin, although cutaneous birthmarks, such as angiomas, dimples, overgrowing hair, dyschromia, and dystrophy, are present in greater than 50% of cases. Use of the term “occult spinal dysraphisms” is discouraged as it suggests complete absence of external abnormalities, a condition that occurs only in a minority of CSDs. ▶Congenital Malformations, Spine and Spinal Cord

OPLL ▶Ossification of Posterior Longitudinal Ligament

Opportunistic Infections Opportunistic infections are infections caused by a microorganism that normally does not cause disease but becomes pathogenic in persons with impaired immune system. ▶Infection, Opportunistic, Brain

Opportunistic Screening Screening not performed in an organized or populationbased screening program, i.e., resulting from self-referral or from referral as a result of a routine medical consultation, a consultation for an unrelated condition, or on the basis of a possibly increased risk for developing breast cancer (family history or other known risk factor). ▶Screening, Breast Cancer

Optical Contrast Agents ▶Molecular Probes, Optical Probes

Optical Imaging B RIAN W. P OGUE Thayer School of Engineering, Dartmouth College, Hanover, USA [email protected]

Synonyms

Ophthalmology or Colposcopy ▶Optical Imaging

Bioluminescence imaging; Bronchoscopy; Diagnostic imaging in endoscopy; Fluorescence imaging; Laparoscopy; Microscopy; Ophthalmology or colposcopy; Optical tomography

Optical Imaging

Definition Optical imaging is a broad term which can be used to describe a large range of imaging systems, from tissue microscopy, through macroscopic imaging of tissue with endoscopic, laparoscopic, or telescopic systems. Generally, this wording is used to describe optical imaging, when compared with other imaging systems which do not use optical detection, such as X-ray, ultrasound, or magnetic resonance. Optical imaging is carried out with a light sensitive system for capturing the images. Standard optical imaging systems use a charged coupled device (CCD), for detecting the image, having many pixels and a lens or fiber optic system, which is customized for the application. Analysis of the system performance in terms of resolution and contrast are always specific to the geometry and the tissue being imaged. Imaging of tissue in vivo is largely based upon the effects of absorption and scattering of the light as it interacts with tissue, causing chromatic changes or allowing viewing of morphologic features. Some in vivo imaging relies upon tracking temporal changes or changes in response to a stimulus, but this is less common. Preclinical or ex vivo imaging, such as in pathology typically relies upon contrast from exogenously introduced agents which are specific to chemical features of the tissue. Optical imaging of contrast agents in vivo is also used in experimental and developmental work as well as a few clinical applications, looking at fluorescence imaging of tissue, where a filter is used on the image detection side to remove the excitation light and only allow the longer wavelength emission light into the imaging camera. Recent developments in biochemistry and experimental biology have introduced a large number of fluorescent proteins and bioluminescent agents that can be transfected into the DNA of cells and animals, thereby allowing optical imaging of organs or specific gene-expression. The use of optical imaging in experimental biomedical research has increased substantially due to these developmental areas.

1411

image the airways. These are typical flexible fiber bundles which allow movement through these complex structures. The video or CCD camera is usually fed to a television for viewing of the procedure in real time by the endoscopist or bronchoscopist.

Lens Coupled Systems Lens coupled imaging systems are used routinely in ophthamoscopic imaging, cervix imaging, ear imaging, as well as most experimental biology imaging systems. The commonality in this area is that if broadband light imaging is required, the lenses used must be compound lenses to avoid issues of chromatic aberration in the resulting image at the camera. Thus, significant care is taken to design and optimize the lens system and how it focuses onto the camera. Generally, lower the f-number of the objective lens and the closer the lens is to the tissue, the more light will be captured in the imaging procedure.

Microscopy Microscopy is by far the most widely used application of optical imaging in medicine, yet it is often considered in a different category because it is so specialized. Pathological analysis of biopsied tissue is the most common application here, where the tissue is fixed and stained for imaging.

O Tomographic Imaging Tomographic imaging of tissue is sometimes called more generically optical imaging, but is better described in the Optical Tomography section or Fluorescence Imaging section. The major difference between tomography and imaging is largely considered to be the acquisition of signals from below the surface, thereby allowing reconstruction or backprojection of the image below the surface.

Fiber Coupled Systems Fiber optic coupled systems are used throughout routine medical practice in imaging the interior cavities of the body, or imaging organs during surgical intervention or exploratory examination see Endoscopy, Brochoscopy or Laparoscopy. Generally, the optical fibers bundle is used to translate the image from a lens inside the body to a remote camera which is mounted on the exterior end of the device being held by the person doing the procedure. The endoscope is used in this way for imaging the digestive tract, and similarly a bronchoscope is used to

Surface Topography and Tracking Optical imaging systems are used in many different applications for surface tomographic mapping, and several commercial systems exist either using patterned light generation to measure surface topography or using stereovision cameras together with computed algorithms to create surface maps. These are used in applications where the three-dimensional topography of a tissue surface is needed to be known.

1412

Optical Probes

Optical Probes ▶Molecular Probes, Optical Probes

Optical Tomography B RIAN W. P OGUE Thayer School of Engineering, Dartmouth College, Hanover, USA [email protected]

Synonyms Diffuse imaging; Diffuse optical tomography; Frequencydomain photon migration; Near-infrared imaging; Nearinfrared tomography; Photon density wave imaging; Photon migration; Time-resolved optical tomography

Definition Optical tomography has been an active area of research study since the late 1980s, when both technological and algorithmic breakthroughs led to workable methods to understand light transport in tissue and solve the inverse image reconstruction problem. The exact tools for tomographic imaging vary widely in terms of hardware and software, but the commonality is that the method allows imaging of the interior of tissue, by recovering the interaction coefficients or chromophore and scattering parameter maps. Near-infrared light, in the range of 650 to 950 nm, is most commonly used in optical tomography as it has the lowest scattering and absorption values, thereby providing the best penetration through tissue possible. Imaging through volumes of up to 10–14 cm is possible given sufficiently designed equipment. Tomographic measurements in this wavelength range are most sensitive to the molecules that absorb light, which are hemoglobin, oxyhemoglobin, water, and lipids. Thus imaging tissue function related to blood concentration and oxygen saturation is possible with near-infrared tomography. In addition, injection of absorbing or fluorescent contrast agents is possible, providing further information about the tissue function. The latter application of imaging fluorescent agents is often called fluorescence tomography, and described in another section on Fluorescence Imaging. Imaging with optical tomography is commonly applied to imaging brain tissue for functional physiology studies, or neonatal cerebral imaging for tracking disease, or female breast cancer imaging. There are few commercially available

devices which do optical tomography, but a larger number of experimental and investigational devices have been developed in research labs (1). Tomography of tissue requires measurement of multiple cross-sectional measurements of light transmission from several different source and detector locations. When overlapping paths are measured, as with X-ray computed tomography, it is possible to mathematically reconstruct the interaction coefficient maps of the tissue interior. In the case of optical tomography, this inverse problem is not as analytically tractable as in the filtered backprojection methods used in X-ray tomography (2). The path of the light is highly scattered, and so the transport mechanism is dominated by elastic scatter. Optical tomography became more intensively studied when it was established the optical path length through a scattering medium could be measured by time-resolved or frequency-domain signals, by directly measuring the signal propagation time. Through modeling the transmission process as a diffusive transport problem, it is possible to mathematically quantify independent absorption and transport scattering coefficients (1). Existing systems typically use this approach to them compute concentration maps of hemoglobin, or relative oxygen saturation of the hemoglobin, or water and lipid fractions in the tissue, through computation of the exact absorption coefficient spectrum.

Characteristics Most systems are based upon multiple fiber optic placements upon the tissue to be imaged, and illuminating the source light into the tissue while detecting the remitted light at all other fiber points. When this is completed, the light is cycled to each of the source fibers in turn to provide multiple overlapping measurements through the tissue volume, as indicated in Fig. 1. Recovery of the remitted light can also be done with noncontact application as well, delivering the light through a scanned beam and picking up the remitted intensity with optics to fibers or charge coupled devices (CCD) (3). The type of light signals used in these systems are divided into time-resolved signals or frequency-domain signals or continuous wave signal illumination, as illustrated in Fig. 2. Each of these provide slightly different information about the tissue, but all have been shown to allow separation of absorption and scattering effects with different levels of success. It is largely agreed that time-resolved signals provide the optimal separation of the scattering and absorption coefficients, whereas frequency domain provides nearly equal ability. However, continuous wave signals are thought to be insufficient for separation of the effects of absorption versus scatter, without the use of multiple wavelengths or other

Optical Tomography

1413

Optical Tomography. Figure 1 Illustrating two common geometries for imaging with optical tomography, namely an effectively circular or thick slabs of tissue volume with a single source and multiple detectors. The source location is cycled around to make multiple overlapping projection measurements, and the banana-shaped projections are the diffuse photon paths, which represent the dominant measurement path. Inward arrows represent the sources here and outward arrows represent the detectors.

O

Optical Tomography. Figure 2 An illustration of a time-resolved measurement (a) where a short pulse of light is transmitted through a hypothetical block of scattering tissue, spreading the pulse out in time, and a frequency-domain measurement scheme (b) using a sinusoidally modulated light intensity transmitted through the tissue.

constraining information. Measurement detectors are either photomultiplier tubes for the weakest light signals, or photodiodes, diode arrays or CCDs. Multichannel detector devices have been produced in many different configurations. Diffusion theory based reconstruction requires the iterative solution to a perturbation type equation or a Newton method approach (2). Both of these require some initial guess of the tissue properties, often assumed to be the best homogeneous estimate, and then the image reconstruction process uses the Jacobian or sensitivity matrix to iteratively improve the estimate of the coefficients at each pixel within the image. The goal of the reconstruction is to minimize an objective function, which

is often calculated as the normalized square difference between the calibrated data set and the calculated diffusion theory prediction of the signal. This minimization process requires inverting the Jacobian matrix, which is an illposed problem, and therefore requires careful regularization (addition of a constant term to the diagonal of the matrix) to make it invertible. This process inherently smooths the overall resulting image, and the value of the regularization parameter is often estimated for each particular problem, often in a Levenberg–Marquardt methodology. Most current systems use multiple spectra measurements, thereby allowing recovery of the absorber concentration maps from the absorption coefficient spectra, and

1414

Optical Tomography

with a priori knowledge of the chromophore extinction coefficients. It has been shown that direct reconstruction of the chromophores can also be achieved by including the spectral fitting process within the inverse image reconstruction problem, and this process reduces the ill-posed nature of the problem, and improves the accuracy.

Applications in Preclinical Studies Small animal imaging with near-infrared tomography has been a highly used avenue, for studies in brain function and tumor imaging. Imaging of tumor characteristics in vivo has been possible both without and with a scattering coupling medium. Imaging of brain tissue function, in terms of hemoglobin and oxygen saturation has been done extensively and used for functional brain activation studies.

potential use. Point spectroscopy of hemodynamics in the brain during childbirth and in utero have both also been shown and used in studies. Measurement of cytochrome oxidase changes in oxygenation within the brain have also been studied, with limited success in this area. Brain topography systems have been studied extensively and commercialized by Hitachi as a tool to study cognitive function in humans. The difference between topography and tomography lies in the depth penetration of the light, with topography being surface measurements across large areas of the brain, without any reconstruction of the lower lying areas. This approach is simpler and more successful than tomography, and provides important information about the outer cortical layers where blood flow and oxygen consumption issues can readily be measured in response to energy demand in the brain cells. Study of cognitive function with optical topography and optical tomography is a highly active area of research and development (4).

Applications in Brain Diagnostics Applications in Breast Cancer Significant interest in optical tomography was initiated with the possibility of monitoring or measuring neonatal cerebral hemodynamics in vivo. Commercial systems are available for spectroscopy monitoring of the brain in neonates with single point transmission measurements, and the desire to extend this to a full tomographic system was the driving force for much of the development of optical tomography throughout the past 20 years. Prototype systems have been developed in several labs and are still undergoing feasibility assessment for their

Breast cancer imaging with optical tomography has been shown in many laboratories, with large clinical trials ongoing and being sponsored by the US National Cancer Institute. Commercial prototype systems have been created by a number of companies, with each having very different technological approaches and different clinical goals. Initial feasibility as a screening tool for widespread use has largely proven unsuccessful, however niche applications in screening high-risk groups, younger women or complicated tissue

Optical Tomography. Figure 3 (a) Radiographic image of a patient’s breast, showing a well-localized fibroadenoma in the upper central region of the breast (marked with an arrow). The resulting NIR images of absorption coefficient at (b) 750 nm and (c) 800 nm wavelength are shown, along with the computed images of (d) total hemoglobin and (e) oxygen saturation. The gray-scale bar units for absorption coefficient are mm−1. The units for total hemoglobin are in micromolar, and the units for oxygen saturation are in percent relative to 100% oxygenated hemoglobin.

Oral Cavity, Inflammatory Diseases

cases remains a possible avenue for use. Integration of optical tomography into MRI scanners has been demonstrated and continues under development. A set of example images from a subject with a focal fibroadenoma tumor in the breast is shown in Fig. 3 (5). The ability to image tumor properties of hemoglobin, oxygen saturation, water fraction, and scattering has undergone rapid development in the past 5 years. Direct recovery of these concentration images has been shown to improve the accuracy of the values, through multispectral fitting and reconstruction in a single step. Optical tomography has also been coupled into standard clinical imaging systems, including MRI, ultrasound, X-ray tomography, X-ray tomosynthesis, MEG, EEG and is thought to provide complementary information, predominantly related to hemoglobin, oxygen saturation and uptake and retention of injected contrast agents.


4.

5.

Delpy DT, Cope M (1997) Quantification in tissue near-infrared spectroscopy. Phil Trans R Soc Lond B 352:649–659 Arridge SR (1999) Optical tomography in medical imaging. Inverse Probl 15(2):R41–R93 Pogue BW, McBride TO, Osterberg UL et al (1999) Comparison of imaging geometries for diffuse optical tomography of tissue. Opt Express 4:270–287 Franceschini MA, Fantini S, Thomspon JH et al (2003) Hemodynamic evoked response of the sensorimotor cortex measured noninvasively with near-infrared optical imaging. Psychophysiology 40(4):548–560 Pogue BW, Poplack SP, McBride TO et al (2001) Quantitative hemoglobin tomography with diffuse near-infrared spectroscopy: pilot results in the breast. Radiology 218(1):261–33

Optical Tracers ▶Molecular Probes, Optical Probes

Oral Cavity, Inflammatory Diseases S ABRINA KO¨ SLING Martin-Luther-Universita¨t Halle-Wittenberg Klinik fu¨r Diagnostische Radiologie Halle, Germany [email protected]

Synonyms Inflammation; Inflammatory lesion

1415

Definition An inflammation is a non-specific immune response in reaction to any type of injury: pathogenic organisms, injuries, foreign bodies or ionising radiation. It may remain localized, sub-clinically and temporarily if the body’s defensive mechanisms are effective. An inflammation can be acute, primary-chronic or an acute inflammation may persist and spread by extension to become a subacute or chronic state. Inflammations of the oral mucosa are called depending on their localization as follows: . Glossitis—inflammation of the tongue . Stomatitis—inflammation of larger parts of the oral mucosa . Gingivitis—inflammation of the gum . Cheilitis—inflammation of the lips If deeper structures are involved it might be a ▶phlegmon or ▶abscess.

Pathology Regarding the severity, macroscopic appearance, pathogenic agent and extent of the inflammatory process, various kinds of inflammation in the oral cavity do occur. Aphthae are intra-epithelial cavities with a size up to 5 mm filled with serous fluid. They are caused by virus, toxic substances or medicaments. Habitual aphthae are characterized by chronical recurrent aphthae of unknown aetiology. A ▶vasculitis is the underlying lesion of Behcet’s disease—a seldom, systemic, HLA-B51 associated illness—in which recurrent aphthae of the oral cavity and genital region appear together with other organ manifestations as uvea, skin, joints and central nervous system. In herpes simplex labialis, small vesicles at the border between the dermis and mucosa of the lips appear due to herpes simplex ▶infection. If the inflammation extends on the entire oral cavity it is called herpetic stomatitis (synonym: stomatitis aphthosa). In a later stage of the disease, vesicles proceed to mucosal erosions and deeper ulcerations with red border. Ulcerative stomatitis (synonym: necrotizing ulcerative gingivitis) is an extended, severe, necrotizing, oedematogenic inflammation caused by a complex of different organisms (streptococci, spirillum, fusiform bacteria and fungi) in immune-compromised patients or bad oral hygiene. Oral herpes zoster infection is caused by the reactivation of varicella-zoster virus in the area of innervations of the maxillary, mandibular or seldom glossopharyngeal nerve. It is characterized by unilateral crops of clustered vesicles.

O

1416


Oral candidiasis (synonym: thrush), an opportunistic infection especially in immune-compromised patients, is mainly caused by Candida albicans. It is usually limited to the skin and mucous membranes. White pseudomembranes consisting of mycetes and necrotic epithelium are a characteristic finding. Acquired immunodeficiency syndrome (AIDS) favours the development of opportunistic bacterial, viral or mycotic infections including oral candidiasis and herpes simplex viral infection. Hairy leukoplakia—a column-shaped hyperkeratosis of the acanthoticaly broadened epithelium—can proceed years before the outbreak of AIDS. In a phlegmon or ▶cellulitis, there is a diffuse spread of granulocytic infiltrates within the tissue. An ▶abscess consists of a necrosis—resulting from purulent colliquation of tissue—and a membrane, which borders on vital tissue. Both are kinds of bacterial infection (streptococci, staphylococci). Most often, they occur in the floor of the mouth due to secondary involvement from other spaces: masticator space (dental infection) or submandibular space (sialolithiasis). Ludwig’s angina is an extensive bacterial infection of the floor of the mouth that always involves both the sublingual and submandibular space. It is frequently bilateral and produces gangrene or serosanguinous phlegmon, but little or no frank pus; involves connective tissue, fascia and muscle, but not glandular structures (1). In necrotizing fasciitis, there is severe or extensive phlegmon that extends into the superficial and deep fascia, producing thrombosis of the subcutaneous vessels and gangrene of the underlying tissues. The oral cavity may be involved. The oral cavity can be involved in syphilis (synonyms: lues, treponemiasis), a Treponema pallidum infection. Primary affection shows a coarse infiltrate or ulceration of deep red colour (ulcus durum); secondary affection flat the infiltrates, ulcers with red halo and whitish blur on mucosa; tertiary affection causes bleeding, necrotizing and rubbery infiltrations, the so-called gumma. Actinomycosis, caused by Actinomyces israelii, is characterized by hard, bluish-violet infiltrates—actinomyces drusen with leucocytes and foam cells—and fistulas to the skin as well as to bone (2, 3).

A reduced general condition and fever occur in ulcerative stomatitis, herpetic stomatitis, oral herpes zoster, purulent infections, Ludwig’s angina, secondary syphilis and Behcet’s disease. The fever is very high in purulent infections and subfebrile in Behcet’s disease. Swelling of lymph nodes accompanies ulcerative stomatitis, herpetic stomatitis, purulent infections, Behcet’s disease and primary syphilis. Except for syphilis, the enlarged lymph nodes are usually painful. The clinical pictures of necrotizing fasciitis which occurs seldom in the head and neck region is characterized by a fulminating onset of the disease, painful edema, erythema, warmth, tenderness and septicemia. Sialism, redness, necrotic odour, haemorrhage and swelling of the gingiva as well as the oral mucosal are seen in ulcerative stomatitis. Sialism also occurs in herpetic stomatitis and oral herpes zoster; necrotic odour in herpetic stomatitis and oral candidiasis. Some pseudomembranes in oral candidiasis are easily wiped off from the affected oral tissues and leave an erythematous, eroded or ulcerated surface, which may be tender. Phlegmon and abscess of the floor of the mouth show a reddish, oedematous mucosa; they are hard in palpation; trismus may be present (2).

Imaging Most oral cavity inflammations are diagnosed by inspection due to their typically macroscopic appearance. An indication for imaging is given in patients in whom a deeper spread, mainly an abscess, is suspected or for exclusion of a tumour in chronic inflammation as actinomycosis or ulcerative stomatitis. Ultrasound is the first performed imaging modality in acute inflammatory diseases of the oral cavity. If there are equivocal findings, CT or MRI can provide similar information. CT is clearly faster performed, less susceptible to motion artefacts, the better available technique and therefore the preferred method in adults. Considering the lack of radiation exposure, MRI should be given the preference in children.

Ultrasound Clinical Presentation Characteristic signs of nearly all oral cavity inflammations are pain, burning of the tongue and/or mucosa and dysphagia. Primary and tertiary affection in syphilis and infiltrates in actinomycosis are painless. In oral herpes zoster, the pain is of severe neuralgic character. Disturbances of tasting are typical for glossitis and oral candidiasis.

The floor of the mouth can be well investigated by ultrasound. The appearance of an abscess depends generally on its stage. In early stage, it is poorly defined and its demarcation against surrounding soft tissue is worse. With maturation it becomes more and more hypoechoic and has finally a dorsal sound-amplification. Fine echoes within the mass are suspicious of gas. If the


infection starts from the submandibular gland, the gland is painful enlarged and hypoechoic. Calculi can be identified up to a size of 3 mm. Ultrasound has limitations in the detection of mandibular involvement,

1417

the demonstration of spread into the retropharyngeal space and the visualization of phlegmonous processes.

Computed Tomography On CT, an abscess appears as an area of low density, with or without gas collection. In mature abscesses a rim enhancement is seen. Surrounding structures are usually involved: muscles are thickened, fat is oedematous with increased density, overlying fascia and skin are thickened showing an enhancement and draining lymph nodes are enlarged. Calculi in the salivary glands or major ducts, erosions of the mandibular cortex and a dental origin can be demonstrated (Figs 1 and 2). A phlegmon and an abscess cannot always be differentiated by clinical methods. Changes—as described in the neighbourhood of an abscess—without abscess formation are found on CT. In necrotizing fasciitis, CT supports the clinical diagnosis; identifies gas, the spread of the disease along vascular sheaths and into the mediastinum (4). Oral Cavity, Inflammatory Diseases. Figure 1 Axial contrast-enhanced CT of a patient with clinically suspected abscess in the floor of the mouth. CT shows an area of low density as a sign for and abscess surrounded by swollen tissue of increased contrast enhancement and prominent lymphoid tissue on the left. Calculus as a cause could be detected (arrow). The patient was treated by incision and antibiotic therapy.

Magnetic Resonance Imaging On MRI, an abscess has low T1-weighted and high T2weighted signal intensities. As in CTonly mature abscesses show a rim enhancement. Oedematous changes within the fat are not quite well visible as on CT. MR-sialography

O

Oral Cavity, Inflammatory Diseases. Figure 2 CT of a patient with dysphagia and swelling of the mouth floor for 2 days. After ultrasound a phlegmon was suspected. (a) Contrast-enhanced scan demonstrates a diffuse swelling of soft tissue in the floor of the mouth without clear abscess formation on the left. The neighbouring fat has an increased density and the fascia is thickened (arrow). (b) High-resolution scan in bone window gives a hint of the underlying dental infection (arrow). The molar was extracted and the infection treated by antibiotics.

1418

Organized Screening

has to be performed for the visualization of calculi. MRI is superior to CT in the demonstration of the medullary component, if there is an accompanying ▶osteomyelitis in the mandible.

Nuclear Medicine In diagnostics of oral cavity inflammations, nuclear medicine plays a limited role. In suspected osteomyelitis of the mandible, bone scintigraphy can demonstrate an increased tracer uptake, but MRI has the same sensitivity and gives additionally exact anatomic information.

Diagnosis The diagnosis of the oral cavity inflammation is based on anamnesis, inspection, mirror investigation and laboratory tests. In bacterial and mycotic diseases, smears are taken for identification of germs. In lues, spirochetes can be serologically proved 4 to 5 weeks after infection. Varicella-zoster-virus-titer is increased in herpes zoster. In Behcet’s disease and oral candidiasis the differential diagnostics should include a HIV-test; in the first one also a lues-serologic-test. Actinomycosis and syphilitic gummas need a histological proof. Imaging—ultrasound or CT, seldom MRI—is performed in suspected deeper inflammation to demonstrate the extent, to document gas-forming infection, to show the relationship to neighbouring bony structures and to give a support in the decision on a conservative or surgical therapy. CT is able to reveal underlying dental infection and sialolithiasis. The latter one can also be found by ultrasound.


Smoker WRK (2003) The oral cavity. In: Som PM, Curtin HD (eds) Head and Neck Imaging. St. Louis, Mosby, pp 1398–1405 Berghaus A, Rettinger G, Bo¨hme G (1996) Hals-Nasen-OhrenHeilkunde. Hippokrates, Stuttart, pp 387–396 http://www.dental.mu.edu/oralpath.htm Becker M, Zbaren P, Hermans R et al (1997) Necrotizing fasciitis of the head and neck: role of computed tomography in diagnosis and management. Radiology 202:471–476

Organized Screening Screening program organized at the locoregional or national level. With an explicit policy, it is a team

responsible for organization and for delivery of the screening services, and a structure for quality assurance. ▶Screening, Breast Cancer

Oropharyngeal Foreign Bodies ▶Foreign Bodies, Gastrointestinal

Orthotopic Liver Transplantation OLT represents the most common procedure for liver transplantation and consists in the replacement of the diseased liver by a liver coming from a cadaveric donor. ▶Transplantation, Liver

Os Carpale The postarthritic ankylosis of two or more (incomplete os carpale) or all carpal bones as a typical feature of latestage rheumatoid arthritis. ▶Rheumatoid Arthritis

Osgood–Schlatter Pattern Overgrowth and fragmentation of the secondary growth center of the anterior tibial tubercle with swelling of the inferior patellar tendon; may be an overuse syndrome or related to repeated minor trauma. ▶Osteonecrosis in Childhood

Osmotic Myelinolysis Osmotic myelinolysis, also referred to as central pontine myelinolysis when it involves the pons, is a disorder associated with chronic alcoholism and malnutrition. It is due to chronic electrolyte imbalances, most commonly hyponatremia. Osmotic myelinolysis is frequently precipitated by rapid iatrogenic sodium correction, resulting in breakdown of the blood–brain barrier and a

Osteochondrodysplasia

noninflammatory demyelination with relative preservation of neurons and their axons. The central pons is the most commonly affected site. Clinically, patients present with seizures, dysphagia, pseudobulbar palsy, dysarthria and movement disorders. ▶Toxic Disorders, Brain

Ossification of Posterior Longitudinal Ligament OPLL results from growth of lamellar bone posterior to the vertebral bodies involving (by calcification) the posterior longitudinal ligament. It is usually diagnosed in elder patients from its characteristic radiographic appearance and may lead to severe neurologic deficit due to spinal canal stenosis. ▶Dish

Ossification or Calcification of Ligamenta Flava Enthesopathy, ossification (mostly thoracolumbar spine), and calcification (often cervical spine) in (frequently) thickened ligamanta flava may contribute to the entity of OLF. There is coexistence of OLF with both DISH and OPLL. ▶Dish

Osteitis Condensans ilii A benign condition characterized by sclerosis in the iliac bones adjacent to the sacroiliac joints. The sclerosis is generally bilateral and usually asymptomatic. The adjacent sacroiliac joint is unaffected. This entity is thought to represent a stress reaction and is most commonly found in postpartum patients. ▶Fractures, Stress

Osteitis Deformans ▶Paget Disease

1419

Osteitis Pubis A noninfectious inflammatory condition involving the pubic bone and pubic symphysis that is thought to occur secondary to periosteal trauma. It occurs most commonly in postsurgical patients or in athletes. Radiographic abnormalities include sclerosis and osteolytic changes. Its imaging characteristics resemble osteomyelitis, However, it is a separate clinical entity and responds to rest and anti-inflammatories. ▶Fractures, Stress

Osteo Sarcoma ▶Neoplasms, Bone, Malignant

Osteoarthritis ‘Osteoarthritis’ and ‘osteoarthrosis’ are both used in medical terminology, however, osteoarthrosis may be the more appropriate term since inflammatory changes are not pronounced in most of the joints. The best phrase to describe degenerative changes of articulations is ‘degenerative joint disease’. ▶Degenerative Joint Disease, Peripheral Joints ▶Gout

Osteoblastoma Osteoblastoma is a benign osteoblastic tumor that differs from osteoid osteoma in having a “nidus” larger than 1.5 cm in diameter, by showing more variable histologic features, and by possessing a potential for local bone destruction and aggressiveness. ▶Neoplasms, Bone, Benign

Osteochondrodysplasia ▶Osteodysplasia ▶Dysplasia, Skeletal

O

1420

Osteochondroma

Osteochondroma Osteochondroma (osteocartilaginous exostosis) represents a benign cartilage-forming lesion that consists of a bony outgrowth covered by a cartilaginous cap. ▶Neoplasms, Bone, Benign

Osteochondrosis Dissecans Sometimes called osteochondritis dissecans, it is a traumatic osteochondral injury, usually along a convex articulating surface of a bone. ▶Osteonecrosis in Childhood

Osteoclast-Like Giant Cell Tumor, Pancreatic Osteoclast-like giant cell tumor is a rare pancreatic tumor that closely resembles giant cell tumor of bone. It is composed of undifferentiated spindle-shaped epithelial or mesenchymal cells mixed with non-neoplastic osteoclastlike giant cells. Some tumors also contain areas of ductal adenocarcinoma. At imaging, the tumor may present as a solid inhomogeneous mass or as a cystic lesion. Invasion of adjacent structures is common, but metastatic spread is found in only 50% of patients. Prognosis is more favorable than for ductal adenocarcinoma. ▶Carcinoma, Pancreatic

Osteodysplasia A LAN E. O ESTREICH Cincinnati Children’s Hospital Medical Center, Cincinnati, Ohio, USA [email protected]

Definition Osteodysplasias are well-defined abnormalities of bone growth, typically genetic in origin. They may be systematic, affecting one aspect of skeletal development, typically in proportion to growth potential of each site, or else “aleatoric,” affecting sites of an aspect of bone development in a chance distribution, with a higher likelihood of occurrence at sites of greater normal growth. [I have borrowed the term aleatoric from modern music composition in which chance plays a role (1)] The aspect of growth may be slowed, disordered, or accelerated. The aspect involved is most often exclusively enchondral in nature or membranous in nature, with only secondary changes in the other. Most, but not all, dysplasias result in short stature (dwarfism). Kindred disorders include dysostoses, in which individual aspects of growth are affected, and sequences, in which an abnormal aspect of development causes a chain of consequent effects on bone (and other tissue) development. As more genetic markers for dysplasias are determined, more conditions are being grouped into “families” of dysplasias of like genetic origin.

Pathology/Histopathology Various histologic manifestations of various dysplasias are known, some of which help explain radiographic findings. In multiple exostoses, the cartilage cap has a growth plate histologically close to the pattern of a normal physis, but somewhat more disorganized. In ▶achondroplasia, not only are the physeal and acrophyseal enchondral columns shorter and slower to progress from resting cartilage to provisional calcification, but also the columnar cells are grouped into separated clusters. The clustering and subsequent lack of ossification of the zones between the clusters, within the primary spongiosa zone of the metaphysis, may account for the enchondromatous areas in achondroplasia metaphyses and diametaphyses (Fig. 1). In Kniest disease, growth cartilage is characteristically irregular in pattern, described as Swiss cheese-like. Osteogenesis imperfecta shows osteoporosis. Cultured fibroblasts from persons with this dysplasia help characterize the condition. Osteopetrosis has denser than normal bone, similar to the hibernating bat, perhaps due to a temporary failure of osteocytic osteolysis, although most investigators favor a failure of osteoclast function as the direct cause.

Clinical Presentation Synonym Osteochondrodysplasia

Disproportionate short stature is found in many systemic dysplasias; limb length discrepancy, with limp, is frequent in aleatoric dysplasias. Fractures (and blue sclerae) are a

Osteodysplasia

Osteodysplasia. Figure 1 Example of prominent enchondral rests in the metaphyses of distal femur of a 9-year-old boy with achondroplasia. The medial and lateral margins of the femur metaphyses and diametaphyses are more concave than normal, partly as a manifestation of achondroplasia enchondral growth slowing and partly because the knees are somewhat flexed.

hallmark of osteogenesis imperfecta; teeth tend to show dentinogenesis imperfecta crowding and irregularity as well. Tall stature is seen in the ▶dolichostenomelias, notably Marfan syndrome and homocystinuria. With regard to characterization of short limbs in dysplasias, the term rhizomelic refers to shortening of limbs most pronounced proximally, such as achondroplasia; ▶mesomelic dysplasias have most of the relative shortening in the intermediate segments (radius/ulna and tibia/fibula); and acromelic refers to the greatest shortening distally. Other rather specific findings include hitchhiker thumb in diastrophic dysplasia; thumb extension beyond the fist in Marfan syndrome; early-in-life shortening, especially of the limbs and later shortening of the trunk in metatropic dysplasia; gargoyle-like facies in the mucopolysaccharidoses; and occasionally the ability to draw the shoulders together in front of the chest in cleidocranial dysplasia/dysostosis.

Imaging Skeletal surveys for osteochondrodysplasia or dysostosis need to be customized for the suspected diagnoses; once certain findings appear, other views may be required. For example, whenever platyspondyly (flatter than normal vertebral bodies) is observed, one needs to check the dens and C1 region in detail, with careful flexion and extension views as necessary, to evaluate for associated subluxability between C1 and C2. The small tubular bones of the hand and feet should be included in any survey because they may show diagnostic features key to the diagnosis. Tall patients with dolichostenomelia (such as Marfan) need

1421

good lateral chest or thorax images to evaluate for pectus excavatum or carinatum. In aleatoric disorders such as multiple exostosis or enchondromatosis, two orthogonal views of each part are needed for complete evaluation of the lesions, even if not necessary for diagnosis. In utero imaging for dysplasia or dysostosis begins with high-detail ultrasound. Questions that arise may then be further investigated with magnetic resonance imaging (MRI). If the answer is still not given, but is important, selected radiographs of the mother’s abdomen may be considered, under close supervision by a specialized pediatric radiologist. If possible, for radiation protection considerations, radiographs of one twin should be avoided if the other is considered unaffected. Abnormal bone length, shape, or a positive family history of dysplasia or dysostosis should initiate the careful prenatal evaluation for specific entities. MRI can give insight into the nature and quantity of cartilage in several dysplasias, including Kniest disease (2). MRI of the spinal cord is prudent before planned spinal surgery in children with any severely abnormal spinal curvature. Three-dimensional reconstruction of abnormal joints assists in surgical planning for specific procedures. Accurate measurement of long bone lengths requires perpendicular X-ray beam images at each end of the bone with a radiopaque ruler alongside that is not shifted between the exposures. Alternately, measurements can be made from a computed tomography (CT) scout view. Long bones for dysplasia evaluation need to be parallel to the film or screen, lest foreshortening in space be mistaken for dysplastic growth foreshortening through time.

Nuclear Medicine Multiple cartilaginous exostoses each have a cartilage cap with a growth plate (I call it “paraphysis”) that will have uptake on bone scan similar to physes. Fractures in such conditions as osteogenesis imperfecta, Ollier enchondromatosis, and osteopetrosis will be hot on bone scan. The periosteal reaction causing the thick cortices of Engelmann disease will be hot on bone scan in the active stage, as it will be in other membranous bone overactivity dysplasias or dysostoses, such as van Buchem disease. Osteopetrosis usually gives a superscan of high activity of all involved bone segments. Uptake is also increased in active fibrous dysplasia.

Diagnosis Only a few dysplasias and dysostoses will be described here.

O

1422

Osteodysplasia

Systematic dysplasias are symmetric side-to-side and generally change one rule or rate of growth, with the sites of greatest normal growth being affected the most. The prototype is achondroplasia with rhizomelic shortening; lumbar pedicles abnormally close side-to-side with the distance decreasing downward, greater than normal concavity of medial and lateral margins of metaphysis (Fig. 1), frontal bossing of the skull, petrous ridges closer to each other than normally, frequent cartilage rests in metaphyses, smooth but delayed growth centers with epiphyseal shape following the shape of abnormally concave physes, proximal ends of the femurs in infancy showing the pattern of an ice cream scoop on end, horizontal sacrum, shorter than normal metaphyseal collar, and trident (Vulcan salute) hands, among the many imaging findings. Hypochondroplasia, an allelic (due to the same gene locus) disorder, is less severe; thanatophoric dysplasia, also allelic, is more severe (demonstrating femurs resembling European telephone receivers and quite pronounced platyspondyly). Achondroplasia and its family results from slowed enchondral growth, whereas Marfan syndrome and homocystinuria have dolichostenomelic (predominant proximal segment) lengthening due to accelerated enchondral growth. The medial and lateral margins of long bone metaphyses are less concave than normal. Children with Marfan syndrome show pectus carinatum or excavatum, tortuous aortic arch, and wide lumbar canal; homocystinuria shows truncal osteoporosis. Among the aleatoric dysplasias, the findings of exostosis or enchondromas occur in a chance distribution, more likely occurring in areas of greater growth potential. The more and the larger the exostosis, the greater the impairment of longitudinal growth of an affected long bone. If one paired bone is more affected than its mate, bowing and, often, dislocations occur (Fig. 2). Exostoses of epiphyses and their equivalents are termed Trevor disease (also known as dysplasia epiphysealis hemimelica); exostoses also can occur at terminal tufts of phalanges and the nonepiphyseal ends of other short tubular bones. Osteogenesis imperfecta is systemic with regard to osteoporosis but aleatoric with regard to fractures. The multiple wormian bones also show side-toside symmetry in distribution. The weakened bones may lead to tam-o’-shanter skull (protrusio occipiti; basilar invagination) from the calvarium sinking on the cervical spine (if this deformity is severe, the clivus may run upward). Teeth are numerous and irregularly placed. The lateral clavicles may show increased upward convexity; the ribs resemble coat hangers. Prenatal sites of fracture may yield bones that look wide at birth from healing fractures. Treatment of osteogenesis imperfecta with bisphosphonates results in tell-tale nearly parallel thin

Osteodysplasia. Figure 2 Growth disparity from multiple cartilaginous exostosis/osteochondromatosis in a 10-year-old boy. The ulna has far more exostoses than the radius distally, resulting in a shorter ulna and a secondary lateral bowing of the radius. The more the exostoses, the greater the impairment of longitudinal growth. (From Oestreich AE, Crawford AH (1985) Atlas of Pediatric Orthopedic Radiology. Thieme Verlag, Stuttgart, p 27)

lines similar in shape to the growth plates in the metaphyses (and equivalent areas in secondary growth centers) that record the jolt to the skeletal system from each dose. Chondrodysplasia punctata (multiple stippled epiphyses) in infancy consists of dense dots of calcification within unossified growth cartilage; the involved bones are short, irregularly shaped, or even abnormally unossified (Fig. 3). As the child gets older, the stipples resolve, and the pattern becomes one of misshapen epiphyses and their equivalents, a pattern then called multiple epiphyseal dysplasia (Fig. 3). In metatropic dysplasia, the metaphyseal collar bone bark seems to lack its usual ability to restrict too rapid transverse growth of physis and metaphysis, so that bones resemble dumbbells with unusually broad metaphyses. The megaepiphyseal dysplasia Kniest disease shows coronal cleft vertebral bodies and delayed ossification of the (large) centers for the femoral head. Severe cervical kyphosis is seen (when lateral images are obtained) in diastrophic dysplasia and camptomelic dysplasia and is one of the cervical vertebral anomalies seen in Larsen

Osteoid Osteoma

1423

Osteodysplasia. Figure 3 From multiple stippled epiphyses to multiple epiphyseal dysplasia. (a) In infancy, one sees the multiple epiphyseal stipples of the distal carpal row as well as the (longitudinally short) first metacarpal. (b) In a 16-year-old, one sees highly dysplastic and narrow distal carpals (i.e., multiple epiphyseal dysplasia) with relatively normal proximal carpals, as well as the longitudinally short first metacarpal. [After F. Silverman. From Oestreich AE (2004) Epiphyseal dysplasias and dysostoses. In: Ferrucci JT (ed) Taveras and Ferrucci’s Radiology on CD-ROM Diagnosis Imaging Intervention, Vol 5, chapter 5]

syndrome. A small mandible with an abnormally concave undersurface is the key finding of Pierre-Robin sequence and is seen in camptomelic dysplasia, cerebrocostomandibular syndrome, and Seckel syndrome. Progressive pseudorheumatoid dysplasia has joint region changes closely resembling rheumatoid arthritis, but also has both early and unusually large os trigonum centers behind the talus, and bullet-shaped or Scheuermann-like (irregular endplates) thoracolumbar vertebral bodies unlike rheumatoid arthritis. Every normal child, incidentally, has at least one small os trigonum center that appears near the end of the first decade of life and fuses with the talus in about a year, appearing somewhat earlier in girls than in boys. Then, in the second decade, another os trigonum center appears in some 10–20% of children, which may or may not also fuse to the talus to remain as a posteriorly protruding process. Postaxial polydactyly and mesomelic short limbs are characteristic in Ellis–van Creveld syndrome. An interesting accompaniment to hand polydactyly is the ham-shaped hamate (a wider than normal hamate ossification in the form of a cured ham). Various characteristic vertebral body shapes on lateral images occur in the several spondyloepiphyseal and spondylometaphyseal dysplasias, as well as mucopolysaccharidosis dysostosis multiplex conditions. Twisted ribs and long bones are the prime feature of Melnick–Needles osteodysplasty. On in utero imaging, achondrogenesis, the second most frequent lethal skeletal dysplasia, is characterized by a lack of ossified vertebral bodies. Very short ribs, such as in short rib polydactyly syndromes, also predict neonatal demise, as do features of thanatophoric dysplasia (the best-known lethal dysplasia). The shorter the ribs in Jeune syndrome, the less likely the neonatal survival.

Bibliography 1.

2. 3.

Struble JW (1995) John Cage and the aleatoric revolution. The history of American classical music: MacDowell through minimalism. Facts on File, New York, pp 285–303 Dwek JR (2005) Kniest dysplasia: MR correlation of histologic and radiographic peculiarities. Pediatr Radiol 35:191–193 Oestreich AE (2002) Mega os trigonum in progressive pseudorheumatoid dysplasia. Pediatr Radiol, Lippincott Williams & Wilkins, Philadelphia 32:46–48

Osteofibrous Dysplasia Osteofibrous dysplasia (ossifying fibroma, Kempson–Campanacci lesion) is a benign fibroosseous lesion which mostly occurs in the tibia of young children. Histologically, the entity differs from fibrous dysplasia by containing scattered bony trabeculae that are rimmed by active osteoblasts. ▶Neoplasm-Like Lesions, Bone

Osteoid Uncalcified organic bone phase consisting of collagen fibers (approximately 94%), proteoglycans, and glycoproteins. ▶Osteomalacia

Osteoid Osteoma Osteoid osteoma is a benign bone-forming lesion of limited size that usually induces reactive new bone

O

1424

Osteoidosis

formation. The lesion itself has classically been described as the “nidus” and should, by definition, not override a maximum diameter of 1.5 cm. Larger lesions should be termed osteoblastoma. ▶Neoplasms, Bone, Benign

▶osteoid (organic bone phase) in cortical and trabecular bone. In children, this condition is known as ▶rickets and includes the defective mineralization of the cartilaginous part of the epiphyseal growth plates.


Osteoidosis ▶Osteomalacia

Osteoma The most frequent benign osseous tumor of the face and sinusal cavities, usually affecting men over 50 years of age. These tumors are commonly asymptomatic, fortuitously discovered on a radiologic examination, in the frontoethmoidal cavity or in the external auditory canal. ▶Fibro-Osseous Lesions, Facial Skeleton

Osteomalacia Osteomalacia (OM) is a pathological condition in adult bone metabolism characterized by the impaired and delayed mineralization of osteoid, causing the accumulation of osteoid and altering the mechanical properties of bone. ▶Osteomalacia

Osteomalacia A HI S EMA I SSEVER Institut fu¨r Radiologie, Charite´ Campus Mitte, Schumannstr, Berlin, Germany

Synonym ▶Osteoidosis

Definition In adult bone metabolism – after closure of the epiphyseal growth plates – ▶osteomalacia (OM) is defined to be a condition of impaired and delayed mineralization of

Physiological bone remodeling is a dynamic process of constant bone formation and resorption, respectively, carried out on the cellular basis through osteoblasts and osteoclasts. Chronologically, bone formation can be divided into two phases. In phase I the osteoblasts produce osteoid in the organic bone matrix consisting of collagen fibers, proteoglycans, and glycoproteins. In phase II the osteoblasts produce calcium, phosphate, hydroxyl, and carbonate creating a plate-like crystal known as hydroxyapatite, which mineralizes on the osteoid-forming calcified bone. Pathophysiological disturbances in phase II based on alterations of the calcium phosphate homoeostasis – in which the serum calcium X phosphorus product is low – lead to the condition of OM in adults and to rickets in children. The calcium phosphate homoeostasis is maintained by a multitude of factors, and thus the etiology of OM and rickets is very diverse. A large number of disorders are associated with OM and rickets. Nonetheless, hypovitaminosis D is considered to be the most frequent cause of OM. The primary role of vitamin D is its physiological function in opposing a decline of serum calcium. Examples of disturbances in vitamin D metabolism that may cause OM are: . Low dietary intake . Insufficient UVabsorption of the skin (reduced natural sunlight exposure, skin covering garments, dark skin complexion which leads to the absorption of only certain spectra of the UV light) . Intestinal malabsorption, e.g., after gastrectomy, bowel resection, or nontropical sprue . Accelerated vitamin D excretion by virtue of anticonvulsive drugs inducing liver enzymes . Chronic renal failure leading to a deficient hydroxylation of the liver 25(OH)-vitamin D3 into the active 1,25(OH)2-vitamin D3 form As described above, a low calcium X phosphorus product is the pathophysiological starting point for the genesis of OM and rickets, and therefore conditions leading to low phosphorus levels must also be considered: . Acquired phosphate depletion due to malnutrition, alcoholism, and use of aluminum-containing antacids which bind phosphorus intestinally . Increased renal excretion of phosphorus caused by an impaired tubular resorption (e.g., adult-onset vitamin D-resistant hypophosphatemic osteomalacia, Fanconi syndrome)

Osteomalacia

Attention should also be paid to oncogenic conditions causing OM (tumor-induced or oncogenic OM), although they occur seldom. This paraneoplastic syndrome is associated with several different neoplasms, of which the majority are of benign and mesenchymal origin, namely, bone and soft tissue tumors. These tumors release circulating factors that increase the phosphorus renal clearance consecutively causing low serum phosphorus levels, while the calcium serum concentration is normal.

Clinical Presentation The clinical presentation of OM is unspecific. Frequent complaints of patients suffering from OM are bone pain and muscle weakness. Bone pain is of diffuse and dull character. In general the symptoms begin as lower back pain and spread symmetrically into the pelvic region and the hips, or upward to the vertebral column into the rib cage and the shoulder girdle. Compression, vibration, or mere muscular activity of symptomatic regions may provoke and increase pain, in some cases leading to the adjusted pain-avoidance behavior of the patient, such as cautious walking, in extreme cases even immobilization or the fear of coughing. OM-related muscle weakness is generally localized proximal to the body trunk, e.g., around the hips involving the gluteal muscle group leading to a waddling gait in late stages of the disease. Whether muscle weakness is secondary to bone pain or of primary genesis is difficult to distinguish. In advanced stages of OM, morphological deformities of the skeleton – due to bone softening and hence reduced mechanical strength – can be observed resulting in hyperkyphosis of the thoracic vertebral column or coxa vara and a high fracture susceptibility. In children suffering from rickets, morphological deformities of the skeleton vary in their dimension according to the time of onset in relation to the phase of epiphyseal bone growth. In periods of rapid growth, rickets tends to achieve high clinical severity. Craniotabes (softening of the calvaria), late fontanelle closure, rachitic rosary (palpable bloating of the costochondral junctions), Harrison’s groove (indentation of the lower ribs at the diaphragm), bowing of the long bones (tibia, femur, radius, ulna), and thickening of the wrist, knees, and ankles (on grounds of metaphyseal widening) are the most visible clinical complications. Muscle weakness to severe muscle hypotonia is another major complication of rickets.

Imaging With conventional radiographs, computed tomography, and magnetic resonance imaging, a wide range of modalities for diagnosing OM are at hand.

1425

In OM, conventional radiographs are usually obtained first. The overall radiographic appearance of bone can be described as homogenous and fuzzy. The reason for this is the decline of contrast differences between bone marrow and calcified bone due to the increased density of unmineralized osteoid. In contrast to this rather nonspecific “ground-glass” appearance, so-called ▶Looser’s zones – found in only 5–10% of patients suffering from OM – are considered to be typical. Looser’s zones – also known as ▶milkman’s zones or ▶pseudofractures – are radiolucent bands of unmineralized osteoid usually oriented perpendicular to the surface of the bones ranging in size from a few millimeters to several centimeters. Looser’s zones are frequently found bilaterally, symmetric, at sites where larger arteries are adjacent to the bones (inner aspects of the femur, pubic rami, the lateral edge of the scapula, and the metatarsals); the hypothesis being that the pulsation of these vessels and the reduced mechanical strength of the bones cause pseudofractures of the latter which heal only insufficiently with osteoid. On computed tomography images the fuzzy “groundglass” appearance of bones affected by osteomalacia is also observed, and Looser’s zone may be identified more accurately in contrast to pathological fractures of other genesis. The relative increase of unmineralized osteoid in relation to mineralized bone or the relative decrease of mineralized to unmineralized bone can be quantified measuring the overall bone mineral density using either dual x-ray absorptiometry or quantitative computed tomography. Pseudofractures may be identified as hypointense lines or fissures on T1- and T2-weighted magnetic resonance (MR) images as well as on images acquired using short T1 inversion recovery (STIR) sequences. High signal intensity around the fracture area on T2-weighted and STIR MR images reveals an acute process, whereas isolated hypointense regions are considered to be chronic. Therefore, the clinical activity of OM may be determined using MR imaging.

Nuclear Medicine For isotope bone scanning, technetium 99m-labeled methylene diphosphonate – which preferably adsorbs onto bone surfaces undergoing new bone formation – is being used. The diagnostic strength of isotope bone scanning for metabolic bone diseases is based on its high sensitivity in detecting and localizing functional alterations before structural changes have taken place. In OM the bone scan often shows a nonspecific increased uptake of technetium 99m involving the entire skeleton. However, Looser’s zones may be identified on bone scans

O

1426

Osteomyelitis

before they are visible on conventional radiographs. For the diagnosis of oncogenic OM and the identification of the primary tumor, it has been reported that tracers selectively binding onto somatostatin receptors (octreotide) may improve the detection rate.

help describe the acuteness of fractures. If all imaging modalities and laboratory tests fail, the best – gold standard – diagnostic modality for identifying OM is the histological analysis of bone biopsies in which the accumulation of osteoid is visible as an increase in the average osteoid volume (>10%) and thickness (>15 mm).

Diagnosis References The difficulty in diagnosing OM arises from its very nonspecific clinical presentation as described above. In the population of elderly patients, diffuse bone pain and muscle weakness can be misinterpreted as rheumatological conditions, thus OM remains undiagnosed, although in most cases it is a curable diseases. Higher awareness and suspicion of osteomalacia in daily clinical routine is therefore essential. Laboratory test results are given in the form of a table. Imaging modalities for the diagnosis of OM have been described above; nonetheless, a brief summary in the context of the overall diagnostic procedure is as follows: patients presenting with unspecific symptoms and laboratory findings suggesting that OM might be the underlying cause may first undergo isotope bone scanning in order to identify typical Looser’s zones. Conventional radiographs of suspicious regions can validate the diagnosis in some cases, but one should keep in mind that radiographically visible morphological changes may occur rather late in the course of OM as compared to pathological uptake behavior on bone scans. On the other hand, if patients can localize peak pain spots, conventional radiographs may be sufficient for identifying Looser’s zone without the use of isotope bone scan. Bone mineral density measurements are rather unspecific for the diagnosis of OM because a multitude of disorders, particularly osteoporosis, are associated with a decreased bone mineral density. Besides having a higher spatial accuracy, computed tomography can be used if it is difficult to ascertain with conventional radiographs whether fractures are caused by OM or other pathological conditions. MR imaging, in addition, can

Serum concentration Ionized calcium Inorganic phosphorus 25(OH)-vitamin D3

Vitamin D deficiency

Chronic renal failure

Renal tubular disorders

Normal/↓

↓

Normal

↓

↑

↓

↓

Normal

Varies (continued)

1. 2. 3. 4.

5.

Edmister KA, Sundaram M (2002) Oncogenic osteomalacia. Semin Musculoskelet Radiol 6(3):191–196 Freyschmidt J (2003) Osteomalazie. Skeletterkrankungen 2nd edn. pp 185–195 Hain SF, Fogelman I (2002) Nuclear medicine studies in metabolic bone disease. Semi Musculoskelet Radiol 6(4):323–329 Kanberoglu K, Kantarci F, Cebi D et al (2005) Magnetic resonance imaging in osteomalacic insufficiency fractures of the pelvis. Clin Radiol 60(1):105–111 Reginato AJ, Coquia J (2003) Musculoskeletal manifestations of osteomalacia and rickets. Best Practice Res Clin Rheumatol 17(6):1063–1080

Osteomyelitis H. P. L EDERMANN 1 , W. B. M ORRISON 2 1

Department of Radiology, University hospital Basel, Basel, Switzerland 2 Department of Radiology, Thomas Jefferson University Hospital, Philadelphia, USA [email protected]

Synonym Bone infection

Definition The term osteomyelitis implies an infection of bone and bone marrow, whereas osteitis indicates contamination of the cortical bone only. Chronic osteomyelitis results most commonly after trauma or surgery and results from unsuccessful treatment of acute bone infection. A sequestrum represents a segment of necrotic bone that is separated from living bone by granulation tissue. An involucrum denotes a layer of living bone that has formed about the dead bone. An opening in the involucrum is termed a cloaca. A Brodies abscess is an intraosseous abscess, frequently seen in children caused by Staphylococcus aureus and represents a subacute or chronic infection. Chronic recurrent multifocal osteomyelitis is a rare sterile (noninfectious) disorder primarily involving

Osteomyelitis

children and adolescents and characterized by a prolonged, fluctuating course most often involving the tubular bones, the clavicle, and less often the spine.

1427

specimens in pedal infection may be unreliable due to contamination from underlying infected soft tissue. Bone biopsy cultures in osteomyelitis may be false negative in up to 50%, whereas accuracy of histopathologic diagnosis of osteomyelitis is high.

Pathology/Histology Pathology: Osseous structures can be contaminated by three principal routes (1): 1. Hematogenous spread of infection through the blood stream as in osteomyelitis of the child or spondylodiscitis. 2. Direct spread from a contiguous source of infection as seen in diabetic foot infections or decubital ulcers in paralyzed patients. 3. Direct implantation of infectious material into the bone as seen after penetrating injuries, punctures, or other surgical procedures. Hematogenous osteomyelitis of the child is most frequently located in the metaphysis of long bones. In the infant up to 1 year of age, bone infection may be complicated by epiphyseal involvement. Subperiosteal abscesses and septic arthritis may complicate osteomyelitis in older children. In adults, osteomyelitis most frequently results from direct spread from a contiguous source of infection and most of these patients have pedal osteomyelitis due to longstanding diabetes mellitus. Prevalence of posttraumatic osteomyelitis directly depends on the degree of traumatization of the involved limb (extensive soft tissue damage, necrotic bone, and presence of foreign bodies). Infection results from direct implantation of microorganism in most cases and S. aureus is the most common pathogen. Postoperative infection occurs via direct implantation, spread from a contiguous septic focus or hematogenous contamination. Histology: Bone biopsy is performed if clinical and radiologic evaluation is not conclusive, if a neoplasm is suspected or if microbial diagnosis is attempted. Definite diagnosis of osteomyelitis relies on positive culture results of causative organisms from a biopsy sample. Characteristic histologic findings of bone infection include aggregates of inflammatory cells (including neutrophils, lymphocytes, histiocytes, and plasma cells), erosions of trabecular bone and marrow changes that range from loss of normal marrow fat in acute osteomyelitis to fibrosis and reactive bone formation in chronic disease. Limitations of percutaneous and surgical bone biopsy include sampling error, false negative cultures in patients receiving antibiotics, difficulties in distinguishing other osteopathy from osteomyelitis histopathologically, and the risk to damage the bone as a result of trauma or iatrogenic infection. Culture results of percutaneous bone biopsy

Clinical Presentation The clinical manifestations of the different forms of bone infections vary considerably depending on the activity of infection. Childhood osteomyelitis is often associated with a sudden onset of high fever, a toxic state, and local signs of inflammation. Diabetic foot infection most commonly has a chronic, relatively indolent course but may quickly lead to septicemia and toxic shock. Tuberculous osteomyelitis differs from pyogenic osteomyelitis by the absence of fever and pain. Posttraumatic and postoperative infections may, in the acute and early stages, lead to exquisite focal symptoms with significant inflammation, fever, and leukocytosis. If the infection can not be cured in the acute stage, chronic recurrent osteomyelitis develops with chronic recurrent bouts of infection, abscesses, bony sequestra, and fistula.

Imaging Radiographs Radiographs are usually the first radiological examination performed if bone infection is suspected. In children, first signs of hematogenous osteomyelitis may be perceptible a few days after onset of symptoms. First radiographic signs of hematogenous osteomyelitis are a subtle swelling of the juxtacortical soft tissues, which may only be evident by direct comparison with the other extremity. Destructive metaphyseal osteolysis and periostitis become visible after 8 to 10 days. Brodies abscesses are typically seen radiographically as well-defined metaphyseal radiolucencies with sclerotic borders, “dripping” to the physeal plate. In adults, radiographic findings of osteomyelitis do not appear in adults for 10 to 14 days after infection and until 35 to 50% of the bone has been destroyed. Radiographic changes of osteomyelitis are not only delayed, but sensitivity is poor. Typical radiographical signs of acute osteomyelitis are permeative bone destruction with periosteal reaction and surrounding soft tissue swelling (Fig. 1a). Beginning cortical bone resorption can be identified as endosteal scalloping, intracortical tunneling, and poorly defined subperiosteal bony defects. Radiographs are also the basic imaging modality in posttraumatic and postoperative osteomyelitis displaying postoperative bony remodeling, sequestra, foreign bodies and osteosynthetic material, and deformities. Acute

O

1428

Osteomyelitis

a new linear periosteal reaction, osteolysis, and sequestration are suspicious of active infection. Delayed union or nonunion of fracture may also be caused by chronic infection.

Computed Tomography Computed tomography has been largely replaced by magnetic resonance (MR) imaging to evaluate osteomyelitis. CT may however be useful to evaluate the presence of sequestra and involucra in chronic osteomyelitis. Since CT offers exceptional detail of the bony architecture in a cross-sectional display, it can also be used to evaluate cortical destruction and fistula, periosteal new bone formation, and the presence of intraosseous gas (Fig. 2), all of which are less conspicuous on MR images.

Magnetic Resonance Imaging

Osteomyelitis. Figure 1 Illustration of typical radiographic signs of acute osteomyelitis. (a) Advanced osteomyelitis of the first metatarsal head in a diabetic patient with an ulcer at the plantar aspect of the first metatarsophalangeal joint. The radiograph reveals extensive permeative bone destruction of the first metatarsal head (arrow) with fragmentation (arrowhead). Also note the narrowing of the joint space due to concomitant septic arthritis. (b) Postoperative acute osteomyelitis of the distal fibula with permeative bone destruction (arrowheads) around the second most distal screw (arrow), which is loosened at the plate with a fine radiolucent rim around the screw head.

postoperative and posttraumatic osteomyelitis leads to illdefined bone destruction, which usually develops in bone directly adjacent to the metal (Fig. 1b). Chronic osteomyelitis can lead to reactive bone formation, periosteal reaction, bony fistula, intraosseous abscesses, and bony sequestra. Presence of surgical implants and posttraumatic bone remodeling may considerably complicate evaluation of chronic osteomyelitis. To evaluate activity of chronic osteomyelitis, it is very helpful to compare current and old films together as changes such as

MR imaging allows early detection of osteomyelitis in contrast to radiographs and CT. Diagnosis of osteomyelitis on MR images is based on the identification of altered bone marrow signal (Fig. 3). Infection of the marrow compartment results in loss of the normal fatty marrow signal on T1-weighted images, with edema on T2-weighted or STIR images, and enhancement on post-gadolinium T1-weighted images. MR protocols should include fatsaturated T2-weighted images and contrast-enhanced, T1-weighted fat-suppressed images (2). Identification of such marrow signal alterations away from the subchondral bone results in high sensitivity for osteomyelitis; however, other entities can alter the bone marrow signal in similar fashion, including fracture, tumor, severe inflammatory arthritis or neuropathic disease, or recent postoperative changes. The lack of bone marrow edema on STIR images excludes osteomyelitis with a specificity of 98% (2). Pedal osteomyelitis, which is by far the most frequent form of osteomyelitis in adult diabetic patients, results in over 95% of patients from contiguous spread of a soft tissue infection; the majority of these patients have some combination of adjacent skin ulceration, cellulitis, soft tissue abscess, and sinus tract. These signs are also called “secondary signs” of osteomyelitis, and can improve specificity. Ischemic tissue in pedal infection is best visualized on fat-suppressed, contrast-enhanced images (3). Osteomyelitis and abscesses may not enhance in necrotic tissue (3). In complicated posttraumatic cases, postoperative bone marrow signal alterations can persist as long as 1 year (4). MR protocols in postoperative osteomyelitis should include fast spin-echo (FSE) sequences with high bandwidth to decrease susceptibility artifacts from the

Osteomyelitis

Osteomyelitis. Figure 2 Illustration of CT findings in chronic active posttraumatic osteomyelitis. (a) Axial scan reveals a cortical defect (white arrow) in the severely remodeled cortex of the distal femur and gas inclusions (black arrow) in the marrow space. (b) Coronal reformat reveals extensive bony remodeling and a large cortical defect (white arrow) filled with enhancing tissue extending to the bone marrow. Intramedullary gas inclusions (black arrow) in the marrow are also indicative of active infection.

metal implants. Gradient echo images and frequency selective fat suppression lead to extensive metal artifacts. T1-weighted subtraction images may be used instead after gadolinium administration. An intraosseous abscess leads to typical findings of a “target” appearance with a center of low signal intensity on T1-weighted images and high signal on T2-weighted

1429

Osteomyelitis. Figure 3 Typical MR signal alterations of osteomyelitis in a diabetic patient with an ulcer dorsal to the proximal interphalangeal joint of the second toe. (a) Fat-suppressed T2-weighted image reveals hyperintense signal in the bone marrow of the proximal and middle phalanx (arrows) and interruption of the dorsal skin surface with hyperintense sinus tract (arrowhead). Note the destruction of the proximal interphalangeal joint with subluxation indicating septic arthritis. (b) Contrast-enhanced, T1-weighted fat-suppressed image with diffuse hyperintense signal in the bone marrow of the proximal and middle phalanges confirming osteomyelitis. Note the delineation of the small dorsal ulcer (arrowhead) and the linear hypointense sinus tract extending to the destroyed joint.

images with rim enhancement. Chronic osteomyelitis and sclerosing osteomyelitis are indolent processes with areas of bone necrosis and sclerosis; as a result, MR imaging may show areas of low signal on T1- and T2-weighted images. Differentiation between reactive, noninfective tissue edema with inflammation from true bacterial infection may be difficult using MR imaging in the following situations (1): (a) differentiation of secondary osteomyelitis from reactive bone marrow edema in septic arthritis, (b) differentiation of postoperative and

O

1430

Osteomyelitis, Neonates, Childhood

posttraumatic reparative signal alterations from infection (4), and (c) differentiation of pedal osteomyelitis and diabetic neuroarthropathy (see also Neuropathic Joint Disease).

3.

4.

Ledermann HP, Schweitzer ME, Morrison WB (2002) Nonenhancing tissue on MR imaging of pedal infection: characterization of necrotic tissue and associated limitations for diagnosis of osteomyelitis and abscess. Am J Roentgenol 178:215–222 Ledermann HP, Kaim A, Bongartz G et al (2000) Pitfalls and limitations of magnetic resonance imaging in chronic posttraumatic osteomyelitis. Eur Radiol 10:1815–1823

Ultrasound Ultrasound imaging can be very useful in pediatric patients due to the lack of radiation. Sonographic signs of osteomyelitis include asymmetric juxtacortical soft tissue swelling, thickening of the periosteum, and demonstration of an adjacent hypoechoic collection. In advanced cases, frank cortical destruction can be seen.

Osteomyelitis, Neonates, Childhood A LAN E. O ESTREICH

Nuclear Medicine Technetium-99m MDP three-phase bone scan has a high sensitivity for osteomyelitis and is excellent for excluding bone infection in case of a normal radiograph, but specificity is low. Increased accumulation of tracer can be seen in other conditions such as bone tumors, neuropathic osteoarthropathy, fractures, and following trauma and surgery. One advantage of Tc99m MDP studies over radiography is that it can detect osteomyelitis within 24 to 48 h of onset. Labeled white blood cell scans or Tc99m labeled monoclonal antigranulocyte antibodies raise the specificity to detect osteomyelitis, and may be helpful in excluding infection in a Charcot joint.

Diagnosis Timely diagnosis of acute osteomyelitis is best achieved by either MR imaging or scintigraphy if radiographs are normal. Sonography may be very useful in acute pediatric osteomyelitis or septic arthritis. Activity of chronic posttraumatic or postoperative osteomyelitis is often best evaluated by comparison of recent and older radiographs. In complex cases with chronic posttraumatic osteomyelitis, a multimodality approach involving specialists from different disciplines (infectious disease, radiology, orthopedic surgery) is usually required. ▶Oral Cavity, Inflammatory Diseases ▶Neoplasms, Odontogenic

Bibliography 1.

2.

Resnick D (2002) Osteomyelitis, septic arthritis and soft tissue infection: mechanisms and situations. In: Diagnosis of Bone and Joint Disorders. 4th edn. W.B. Saunders, Philadelphia, vol 3, chapter 59, pp 2377–2480 Morrison WB, Schweitzer ME, Bock GW et al (1993) Diagnosis of osteomyelitis: utility of fat-suppressed contrast-enhanced MR imaging. Radiology 189:251–257

Cincinnati Children’s Hospital Medical Center, Cincinnati, Ohio, USA [email protected]

Synonyms Bone infections; Bone marrow infections; Joint infections; Soft tissue infections

Definition Infections of individual bony parts, joints, and other supporting musculoskeletal tissue, including combinations of the three. The responsible agent may be bacterial, fungal, parasitic, viral, or other organisms. A condition resembling osteomyelitis, but without a known agent, is chronic recurrent multifocal osteomyelitis (▶CRMO), with various skeletal manifestations behaving symptomatically and radiographically similar to infection. An open fracture is associated with a break in the skin or at the nail bed. A closed fracture has no such direct communication with the outside environment.

Pathology/Histopathology According to the causative organism, the histopathology of bone (and joint and soft tissue) infection varies. Additionally, the histopathology of subacute and chronic infection is different from the acute type. Moreover, failures in body defense mechanisms vary between these underlying conditions – for example, in ▶chronic granulomatous disease of childhood, white cells can ingest bacteria and other organisms, but not digest them when the organisms do not contain hydrogen peroxide (for this reason, Ed Neuhauser called the condition “dyspeptic granulomatosis”). The physis is a relative barrier to


1431

passage of infection in bones, except in infancy or when affected by trauma. The bone bark of the metaphyseal collar is a relative barrier to pus in bone, so that breaking through cortex to form periosteal reaction generally stops at the step-off between periosteum and bone bark (in luetic bone disease, this feature is known as the Wimberger sign). In those joints that surround metaphyseal bone, infection travels readily between bone and joint. Neonates of mothers infected with syphilis who show “leukemic lines” at birth do not, at the time of neonatal x-ray imaging, have skeletal infection yet, but rather show the generalized effect of intrauterine stress (of the same nature of such lucent bands when the mother is given magnesium sulfate during late pregnancy (2)). Later in childhood syphilitic bone disease may or may not include local spirochetes. Biopsy of lesions in chronic recurrent multifocal osteomyelitis shows many plasma cells, but no microorganisms.

Clinical Presentation Cardinal signs and symptoms of osteomyelitis include fever, malaise, pain, tenderness, swelling, warmth, redness, and loss of motion. The same signs occur with septic arthritis, and most are found with pyomyositis and other soft tissue infections, so that clinically the differentiation of the sites of infection may be difficult indeed. In patients with immune deficiency and other disorders of defense mechanisms, symptoms may be more silent. Any ▶stubbed toe or stubbed finger fracture (dorsal Salter II fracture of a distal phalanx) must be suspected of being infected (3) (Fig. 1). Bones adjacent to tissue sites of tuberculosis or actinomycosis are vulnerable to becoming infected, as those infections tend to cross tissue planes. In osteopetrosis, one should be alert for jaw osteomyelitis.

Osteomyelitis, Neonates, Childhood. Figure 1 Osteomyelitis of the distal phalanx of the great toe in a boy several weeks after a stubbed toe Salter II fracture. Bone destruction and periosteal reaction have appeared. From Oestreich AE, Crawford AH (1985) Atlas of Pediatric Orthopedic Radiology, Thieme Verlag, Stuttgart p. 70.

The echogenic periosteum is separated by a hypoechoic zone from the echogenic cortex (Fig. 3). MRI can reveal edema in marrow and surrounding soft tissue, as well as show the organizing fluid elevating periosteum. A needle stick under fluoroscopic, ultrasound, or computed tomography (CT) guidance can also give a diagnosis quickly. CT can distinguish soft tissue infection from its normal surroundings, as can ultrasound and MRI. Dental films give high detail of the periapical region if infection beyond teeth is suspected. Fluid surrounding a tendon is well depicted by ultrasound in tendonitis.

Imaging Since bone local demineralization and periosteal reaction do not appear on plain radiographs for 10 days after an infection begins, it is essential that the radiologist not dismiss a “normal” study before 10 days as ruling out infection (Fig. 2). It is also wrong to give the advice of solely a return for repeat imaging before the 10 days have elapsed or even after that time – the time to diagnose and treat osteomyelitis is right away (not only after plain film abnormality is evident). Nuclear imaging, ultrasound, and magnetic resonance imaging (MRI) are particularly good for early detection of infection. With ultrasound, elevation of the periosteum may be detected within a day, far earlier than the 10 days required for plain imaging.

Nuclear Medicine Acute osteomyelitis that has caused necrosis or avascularity will be cold (less activity than normal bone) on bone scan, unlike the majority of bone infection – acute, subacute, or chronic – which will be hot. A bone scan alone may not distinguish infarction from new sterile infarction in sickle cell anemia, but gallium scanning gives relatively much stronger uptake in infection than infarction. Infection limited to joint or soft tissue should not increase activity in adjoining bone on precise nuclear images. Since it takes 10 days for bone demineralization or periosteal reaction to become visible on radiographs,

O

1432


Osteomyelitis, Neonates, Childhood. Figure 2 (a) Less than 10 days from onset of symptoms, a plain image of the great toe shows some soft tissue swelling, but the bone is normal in appearance. (b) Ten days later, destruction of the proximal medial metaphysis and metadiaphysis is evident (arrow). It is a medical error to wait until the radiograph is abnormal; other imaging or some treatment should have been initiated at the time of presentation. From Oestreich AE, Crawford AH (1985) Atlas of Pediatric Orthopedic Radiology, Thieme Verlag, Stuttgart p. 247.

Osteomyelitis, Neonates, Childhood. Figure 3 A young child with acute osteomyelitis of the proximal femur (F), with no plain image bone abnormality yet. The arrow points to periosteum elevated by the sonolucent pocket of pus external to the femoral shaft (F) on longitudinal ultrasound. The right-sided image is the normal contralateral side.

the radiologist should strongly consider bone scanning for diagnosis if symptoms have lasted less than that time. Gallium 67 citrate is generally positive in osteomyelitis by 24–48 h. Labeled white cells with indium 111 or Tc99m HMPAO are other means for seeking infection by nuclear scanning. Active chronic recurrent multifocal osteomyelitis is expected to be positive on bone scan during active phases. In Langerhans cell histiocytosis (one theory considers the disease formerly known as histiocytosis X to be of infectious origin), increased bone scan activity with a photopenic center may be seen. For spondylodiskitis in a child, gallium scan shows high activity; MRI is an alternative useful modality.

Diagnosis Perhaps the most important point to restate about radiographs of osteomyelitis is that bone demineralization and periosteal reaction are not visible until 10 days after the infection begins; so the diagnosis should not wait for that. Something else should be done (nuclear imaging, needle biopsy, ultrasound, CT, MRI, or heuristic treatment possibly guided by blood culture). Tarsal and carpal bones and the patella and other growth centers have no periosteum, so they will not have periosteal reaction. Infection around the roots of teeth leads to loss locally of the lamina dura and bone demineralization. Infection of

Osteonecrosis, Adults

1433

but sometimes CT, guidance. If fluid is obtained, it should be investigated for organisms, including tuberculosis, regardless of the gross appearance.

References 1. 2.

3.

Jurik AG (2004) Chronic recurrent multifocal osteomyelitis. Semin Musculoskelet Radiol 8:243–253 Malaeb SN, Rassi AI et al (2004) Bone mineralization in newborns whose mothers received magnesium sulphate for tocolysis of premature labour. Pediatr Radiol 34:384–386 Yellin JA, Towbin RB, Kaufman RA (1985) Stubbed finger osteomyelitis. J Trauma 25:808–809

Osteonecrosis, Adults C HRISTIAN R. K RESTAN Osteomyelitis, Neonates, Childhood. Figure 4 Four weeks after only soft tissue swelling and a positive elbow fat pad, this 6-month-old infant has full-blown osteomyelitis seldom seen today, with an involucrum (I) surrounding sequestrum (S) of the ulna and a distal cloaca evident (arrow). From Oestreich AE, Crawford AH (1985) Atlas of Pediatric Orthopedic Radiology, Thieme Verlag, Stuttgart p. 66.

Department of Radiology, Medical University of Vienna Vienna General Hospital, Austria [email protected]

Synonyms Aseptic bone necrosis; Avascular necrosis

the long bone shaft usually does not disturb the 1–3 mm metaphyseal collar, but bone demineralization and periosteal reaction begin beyond it. Untreated or unsuccessfully treated tubular bone osteomyelitis may eventually lead to involucrum (from periosteum) surrounding sequestrum (the damaged avascular bone) and discharge of pus through a radiographically evident cloaca (Fig. 4). Subacute osteomyelitis (Brodie abscess) tends to be seen as ovoid bone demineralization, longitudinally directed, near a physis or perhaps crossing it. Infection, and other fluid, in the hip or humerus is generally not perceived on plain images (until eventual local bone infection or regional demineralization occurs). Infection in the knee and ankle is easily seen (but not specific) from interfaces, the suprapatellar bursa and the ankle tear drop just adjacent to the talus, respectively. Soft tissue infections on plain images are seen as soft tissue swelling and disturbance of the local interfaces between tissues, including subcutaneous fat. Exceptionally, gas density is seen from gas-forming organisms.

Interventional Radiological Treatment Interventional pediatric radiologists may be asked to drain deep abscesses, generally done under ultrasound,

Definition Negative bacteriologic studies from well-documented cases of bone necrosis led to the use of the term “aseptic necrosis.” Subsequent observations indicated that such cases were not only aseptic but also avascular. Hence, the terms “ischemic necrosis,” “▶avascular necrosis,” and “▶bone infarction” were suggested (1). By convention, the term “bone infarction” is reserved for bone necrosis in the metaphyseal and diaphyseal regions, whereas “avascular necrosis” (ischemic necrosis) applies to the epiphyseal and subarticular regions.

Pathology/Histopathology Knowledge of the anatomy and histology of the articular and subarticular region are important for understanding the pathophysiology of osteonecrosis. In the hyaline cartilage, four different zones can be differentiated microscopically because of different orientation of the fibers. Below the three superficial zones lies the very thin zone of calcified cartilage, divided from the uncalcified zone by the so-called tide mark that represents

O

1434


the zone of new cartilage formation. The chondrocytes produce the fibers and the ground substance, which is composed of collagens (5–10%), proteoglycans (10–30%; important for osmotic pressure), and water (65–80%). At its base, adult articular hyaline cartilage is bordered by the subchondral plate (cortical endplate) with a high number of arterial vessels, capillaries, sinusoids, and venous vessels, with a declining number by 20% from adolescence until the seventh decade. While the upper (outer) zones of cartilage get their nourishment from the synovial fluid, there is some evidence that the very important zone of calcified cartilage and, definitely, the subchondral regions are fed by the subchondral vessels. The medullary bone has a dual blood supply, from the medullary arteries and periosteal arteries (2). Bony structures with a relatively high amount of cartilage surface (head of femur, talus) are very prone to problems in vascularization. Ischemic necrosis of bone and bone infarction occur in areas of predominantly fatty marrow, which has a much lower blood flow. Infarcts (ischemic necrosis) occurring within the epiphysis or in small round bones (such as the talus) are covered by compact subchondral bone and cartilage. Mineralized bone does not appear to be directly materially altered by ischemic necrosis. Bone density is unchanged within the dead bone but is usually lowered in the surrounding bone due to reactive changes (osteopenia).

Clinical Presentation The clinical symptoms depend on the anatomic location of the osteonecrosis. A small proportion of patients may even present without symptoms, but usually in ▶Association Recherche Circulation Osseous (ARCO) stage 2 avascular necrosis is associated with considerable pain. If the femoral head is affected, patients suffer from slight pain in the hip joint, groin, or buttock or may experience limping and even be unable to walk at all. The symptoms are unspecific, so many other conditions, including arthrosis, insufficiency fractures, and sciatica, have to be considered in the differential diagnosis. In patients with Ahlback’s disease the clinical onset is relatively abrupt, with pain that worsens during the night and accompanying effusion and tenderness. The clinical presentation may mimic meniscal disease or arthrosis and can only be diagnosed with ▶magnetic resonance imaging (MRI). Avascular necrosis of the lunate bone leads to localized pain and swelling of the wrist. Again, without proper imaging modalities, the diagnosis can be delayed. Early diagnosis is of utmost importance for proper treatment of the disease because failure to promptly initiate appropriate therapy can lead to bone and joint destruction with subsequent need for total joint replacement.

Imaging Standard radiographs, computed tomography (CT), and MRI are used in the diagnosis of osteonecrosis/bone infarction. The early diagnosis must be based on the visualization of changes in the soft tissue/bone marrow, which are altered on MRI. The role and relevance of high-resolution CT in detecting subchondral (micro)insufficiency fractures in early diagnosis of epiphyseal osteonecrosis is still under discussion. The only existing imaging modality to visualize necrotic marrow tissue is MRI. The combination of the inner linear, high-intensity rim with the parallel-running outer, low-intensity rim on T2-weighted or contrast-enhanced MR images is typically called the “double-line” sign and is characteristic for epiphyseal ischemic osteonecrosis with repair (3) (Fig. 1a–c). In this stage (ARCO 2), an arc-shaped subchondral lucent lesion with a thin sclerotic rim can be seen on standard radiographs. These findings are even better demonstrated with CT. CT also shows the abnormal configuration of the trabeculae resulting in irregular localized tiny defects with neighboring sclerosing areas. This ARCO stage 2 may last for weeks and months, but subchondral bony fractures (“crescent sign”) representing ARCO stage 3 may finally develop. They are best seen on standard radiographs and/ or CT (4). Finally, a complete impression fracture will occur. In ARCO stage 4, all typical signs of degenerative osteoarthritis are summarized, which includes joint deformity with flattening of the most involved part, subchondral “cystic” and sclerotic changes, formation of osteophytes, and joint space narrowing. Application of intravenous MR contrast medium is advocated by some authors, which could be helpful in unclear cases to exactly differentiate the necrotic part with no contrast enhancement from the surrounding viable tissue (5). As the most sensitive MR sequence, STIR images should be used first. If any abnormality is seen, T1-weighted fat-suppressed and T2-weighted gradient-echo (or fast spin-echo) should be added. In (medullary) bone infarction the early phase with central necrosis and surrounding reactive and usually irregularly shaped tissue, which shows significant contrast enhancement, is followed by increasing calcification. Finally, either the medullary calcification can be found as the only remaining sign, or there will be complete healing.

Nuclear Medicine Cessation of blood flow can be visualized by a three-phase bone scan, in which the first, vascular phase shows a low uptake of the radioactive marker, which is surrounded by a circular area of elevated uptake in later phases (“cold in hot


1435

Osteonecrosis, Adults. Figure 1 (a) Anteroposterior view of both hips in a 30-year-old man. Ill-defined translucency represents ARCO stage 2 of right hip necrosis. (b) Coronal magnetic resonance imaging (STIR) shows marked edema of the right femoral head. (c) Coronal magnetic resonance image (T1-SE) shows hypointense demarcation in the right femoral head; ARCO stage 2.

spot”). Nuclear medicine studies, however, have been largely replaced by MRI because of better specificity and availability and the lack of ionizing radiation with MRI (6).

Diagnosis The most preferred classification system for the staging of epiphyseal osteonecrosis was published by ARCO, in which emphasis was placed on a four-part staging system comparing radiographs, CT, bone scintigraphy, and MRI with histology and including prognostic factors (Table 1). Osteonecrosis can be differentiated in necrosis of the epiphysis (usually called osteonecrosis) and in necrosis of the metadiaphysis (usually called infarction). Spontaneous medullary bone infarcts are often incidental findings in the metaphyses of long tubular bones. They appear as peripheral rims or shells of calcifications in the humerus, femur, tibia, or fibula and must be distinguished from enchondromas. In the majority of cases, the exact etiology of osteonecrosis is unknown and hence is called idiopathic osteonecrosis. Most commonly, the femoral head or femoral condyles are involved in idiopathic epiphyseal

necrosis (Ahlback’s disease, or SONK). The differential diagnosis of idiopathic (epiphyseal) osteonecrosis includes bone marrow edema syndrome (BMES), osteomyelitis, bone tumor, and malignancies such as leukemia. BMES is a self-limiting disease with a relatively large zone of bone marrow edema and accompanying joint effusion and demineralization (“transient osteoporosis”) but without double-line sign. Secondary osteonecrosis may be due to trauma, corticosteroid therapy or hypercortisolism (such as after transplantation), hemoglobinopathies, Caisson disease (dysbaric conditions), small vessel disease (such as collagen disease), alcoholism and pancreatitis, gout and hyperurecemia with reactive vessel obstructions, Gaucher disease, irradiation/chemotherapy with direct vessel damage, and high intra-articular pressure (such as in infection and hemophilia). The prognosis of osteonecrosis depends on the size and location of the ischemia. Osteonecrosis that occurs in a weight-bearing position and involves more than 30% of the articular head (such as the femoral head) has a much worse prognosis than lesions in a nonweight-bearing position and involving less than 15% of the articular head.

O

1436


Osteonecrosis, Adults. Table 1 ARCO Classification (1992) ARCO stage 0 1

2

3

4

Histology Microosteonecrosis • Osteopenia, fractures • Microosteonecrosis • Marrow edema • Hyperemia • Osteonecrosis • Reactive changes: hyperemia, osteopenia, new bone formation

Radiographs 0 0

• • • •

Computed tomography

Bone scintigraphy

0 0 • Osteopenia • Centrally in • Insufficiencylesion low micro-fractures uptake Osteopenia Reactive sclerosis Cysts

• • • •

Osteopenia Sclerosis Cysts Asterisk sign

• Subchondral • Crescent sign • Crescent sign impression fracture in weight-bearing area • Reactive degenerative • Flattened articular changes surface • Joint space narrowing • Subchondral cyst formation • Reactive osteophytes • Subchondral sclerosis • Deformity, malpositioning, atrophy

The most common sites for posttraumatic osteonecrosis are the femoral head, the body of the talus, the humeral head, and the carpal scaphoid (Preiser’s disease). Rarely, the tarsal navicular, capitate, carpal hamate, or lunate bone is involved. The development of the osteonecrosis may take 6–36 months. The likelihood of development of osteonecrosis gets higher with the severity of the trauma. The principal blood supply of the affected bones is responsible for the risk and location of osteonecrosis. Early diagnosis is made by MRI, and sequential follow-up is with CT and standard radiographs. MRI shows persistent edema with localized zones of necrosis. The osseous flattening, collapse, and fragmentation are usually delayed for a period of 9 months and also depend on the weight-bearing load of the epiphysis. Finally, the involved bony parts may become very dense, smaller, and irregularly shaped (Fig. 3). Adult idiopathic osteonecrosis may be due to chronic overloading of the hip joint (head of femur) and the femoral condyle (Ahlback’s disease). Early diagnosis is possible with MRI and three-phase bone scan. Primary necrosis of the femoral head probably affects men more frequently than women and is usually seen between the fourth and seventh decades of life. The reported

Magnetic resonance imaging 0 • Localized edema

• High uptake • Double-line sign (reactive margin of infarct with hyperemia and new bone formation) • Single-line sign in insufficiency fractures • High uptake • Crescent sign • Cartilage fracture (loss) • High uptake, • Cartilage defects deformity • Deformity • Effusion, • Subchondral sclerosis • Osteophytes

prevalence of bilateral disease varies between 35 and 72%. The typical radiographic signs are round, oval, or triangular low densities that are neighbored by a highdensity rim, followed by the crescent sign and impression fracture (ARCO stages 1–3). Early diagnosis depends on proof of bone marrow edema (with MRI) or diminished blood supply (with three-phase bone scintigraphy). Femoral condyle osteonecrosis (Ahlback’s disease, spontaneous osteonecrosis) is a typical condylar idiopathic osteonecrosis localized in the medial condyle of the femur and occurs in patients older than 50 years and in women more than in men, but it may be also found in the lateral condyles and the tibial plateau. On standard radiographs, the first sign is a subtle flattening of the neighboring joint surface, followed by a narrow zone of increased density adjacent to the depressed osseous surface. A radiolucent area in the condyle can be detected over the ensuing weeks, becoming more sharply demarcated by time. If untreated, further depression of the bony margins and progressive sclerosis and intra-articular osseous bodies will follow. Over a period of months or years, all signs of secondary degenerative osteoarthritis will develop. Bone collapse, varus deformity, and displacement can also be noted (Fig. 2).


1437

Osteonecrosis, Adults. Figure 2 (a) Coronal magnetic resonance image (STIR) shows marked edema of the medial condyle in a 90-year-old woman (Ahlback’s disease). (b) Anteroposterior view shows no evidence of osteonecrosis in this digital radiograph.

O Osteonecrosis, Adults. Figure 3 (a) Coronal MPR of multidetector computed tomography reveals nonunion of the scaphoid in a 35-year-old woman after trauma. (b) Coronal magnetic resonance image (T1-SE after intravenous gadolinium) demonstrates no contrast enhancement in the proximal fragment (Preiser’s disease) of the scaphoid bone.

Kienbock’s disease occurs most commonly between the ages of 20 and 40 years and has a predilection for the right lunate bone. The male-to-female ratio is 2:1, with the following stages: In stage I the radiograph is normal or shows a subtle fracture. MRI visualizes a bone marrow edema, and bone scintigraphy a hot spot. In stage II, abnormalities in radiodensity with changes in size and shape are seen. MRI will reveal necrotic hypointense areas in all sequences. In stage III, the lunate bone collapses and shows a high density with irregular borders. In stage IV, signs of degenerative osteoarthritis are demonstrated. The etiology seems to be chronic trauma. In patients with short ulnae (ulnar minus variant), the loading forces on the lunate are much higher than normally. In dysbaric osteonecrosis (Caisson disease), in cases of too rapid decompression the released nitrogen may produce bubbles that act as gas-emboli. Those may

occlude vessels partially or completely. A delay of at least 6 months or years between the exposure and the onset of radiologically evident juxtaarticular or diaphyseal and metaphyseal radiodense foci/radiodense lesions is typical. Early diagnosis is possible with three-phase bone scintigraphy and whole-body MRI.


4.

Johnson LC (1964) Histogenesis of avascular necrosis. Proceedings of the Conference on Aseptic Necrosis of the Femoral Head, NIH, p 55 Brookes M, Revell WJ (1998) Blood Supply of Bone. Springer, London Sugimoto H, Okubo RS, Ohsawa T (1992) Chemical shift and the double-line sign in MRI of early femoral avascular necrosis. J Comput Assist Tomogr 16(5):727–730 Stevens K et al (2003) Subchondral fractures in osteonecrosis of the femoral head: comparison of radiography, CT, and MR imaging. Am J Roentgenol 180(2):363–368

1438

5.

6.

Osteonecrosis, Childhood

Vande Berg BE et al (1993) MR imaging of avascular necrosis and transient marrow edema of the femoral head. Radiographics 13(3):501–520 Imhof H et al (1997) Imaging of avascular necrosis of bone. Eur Radiol 7(2):180–186

Osteonecrosis, Childhood A LAN E. O ESTREICH Cincinnati Children’s Hospital Medical Center, Cincinnati, Ohio, USA [email protected]

Synonyms Avascular necrosis (Perthes disease = Legg–Calve´–Perthes disease = idiopathic avascular necrosis of the hip in childhood); Bone infarction; Chondrolysis (of cartilage)

Definitions Arterial or venous ischemia leading to cell death and/or disturbance of growth in bone. Historically, several (eponymous) variations in skeletal growth, now known not to be due to necrosis, were once so considered and are still often discussed in the context of true necrosis states. Death or disturbance of the zone of resting cartilage leads to delay, deformity, or cessation of enchondral growth; other conditions such as frostbite and ▶Kashin–Beck disease may locally impair enchondral growth as well. The conditions in infancy known as multiple stippled epiphyses may well be secondary to necrosis of growth cartilage not yet in a growth plate.

Pathology/Histopathology Clinical bone necrosis is usually more dynamic than mere death of osteocytes. The body has ways of repairing damage, so that the histopathologic pattern may be composed of both destruction and rebuilding. In Perthes disease, for example, the first step is impairment of vascular supply. The first irreversible structural change is a crumbling fracture of bone just below the lateral zone of provisional calcification, seen on frog leg radiographs. Then waves of bone necrosis appear with lack of viable osteocytes, followed by waves of overlying new bone (creeping substitution). Years later, a viable but deformed

head is usually formed, more or less congruent with its acetabulum. However, because the shape is less spherical than normal and does not fully fit in its acetabulum, secondary arthrosis tends to appear relatively early in adulthood. Because of avascularity in the femoral neck and primary spongiosa in Perthes’ disease, cartilage from the physis persists in the otherwise ossified neck, giving gouges of lucency on X-ray images. Both the head and neck of the femur widen transversely compared to normal as the disease progresses—perhaps related to increased periosteal formation in the neck from the processes of healing. An adverse event in the progress of Perthes is tethering across the physis in some patients. The greater trochanter growth is generally not impaired in Perthes, so it is relatively overgrown, leading to varus deformity of the hip, which is further exaggerated if tethering of the main physis occurs. Osteochondrosis (some still refer to it as osteochondritis) dissecans is an injury to bone and overlying cartilage, which is painful and more severe, indeed unstable, whenever joint fluid enters the space between the lesion and normal bone. Schmorl nodes are protrusions of normal vertebral disk substance through weak areas in the vertebral end plates (zones of provisional calcification). Schmorl nodes are often a component of Scheuermann disease of the spine. In frostbite and Kashin–Beck disease (1), portions or the entire involved physis or acrophysis becomes nonviable, leading to local lack of growth (and perhaps early physeal fusion). A cone epiphysis is a result of a similar loss of viability of the central (more senior) portion of a physis, whether due to injury, infection, or genetic causes.

Clinical Presentation Pain (in the groin, thigh, or referred to a knee), limping, and favoring a limb may reflect Perthes disease; however, Perthes may be clinically silent for months and discovered serendipitously on radiographs acquired for other reasons. Symptoms for avascular necrosis from other causes and at other sites are similar. In Gaucher disease, the spleen is often quite large; in sickle cell anemia, the spleen is usually small or absent. Traumatic avascular necrosis occurs after hip dislocation. For steroid-induced hip necrosis, other symptoms of steroid use could be sought. The incidence of osteochondrosis dissecans rose when teenagers were dancing The Twist, which indicates that certain challenging repetitive motions might be a predisposing factor. For frostbite, the history of severe cold exposure if available is helpful for the diagnosis (although the pattern of involvement on X-ray images is virtually diagnostic—especially if the thumb is spared by being


held in the fist during the exposure). Kashin–Beck disease is geographically limited to China, Mongolia, and Tibet, where thousands of cases have occurred.

Imaging In Perthes disease, magnetic resonance imaging (MRI) and nuclear images show avascularity earlier than the first plain image finding, which, in turn, appears first on the frog-leg view (Fig. 1). However, unilateral retarded maturation of one femoral head might be appreciated on the frontal radiograph as well. Ultrasound shows associated hip effusion easily; plain images do not. As Perthes disease progresses, computer tomography (CT) images show the current integrity of bone, extent of involvement, and nonosseous components; however, MRI is preferred because it does not use radiation and also displays the femoral head and neck and acetabulum throughout the long course of Perthes (2). ▶Osteochondrosis dissecans of the distal femur is usually situated posteriorly in the involved condyle, and thus is better seen on a notch (angled) view tangential to it. Evaluation for instability or fluid between the osteochondrosis and adjacent healthy bone could be made with MRI or a CT arthrogram. When there is doubt about whether a normal variant of ossification or an osteonecrotic bone is present, a display of the osseous and cartilaginous elements on MRI

1439

may help one decide, with overlying normal cartilage favoring the normal variant. For example, Meyer dysplasia of the hip (3) can be distinguished from Perthes disease by the finding of normal cartilage and normally vascularized bone. Ultrasound can show the findings in Osgood– Schlatter condition of the patellar tendon and anterior tibial apophysis.

Nuclear Medicine Nuclear imaging, like MRI, shows findings earlier in Perthes disease than the upper outer lucent crescent found on radiographic frog-leg images. On bone scans, the earliest finding is a cold defect in the femoral head, a larger zone indicating more extensive disease. This phase, however, is reversible without progression to irreversible definitive Perthes disease. The perthetic femoral head eventually regains scan activity in the reparative phase later in the disease evolution. One week after the onset of sterile infarction of bone, whether from sickle cell disease, Gaucher disease, pancreatitis, or other noninfectious causes, bone scans begin to show high activity (as healing begins) rather than the earlier lower than normal (cold) activity. Lack of bone scan activity is also a sign of avascularity after certain fractures, for example, in the proximal scaphoid following a transverse fracture of the scaphoid waist. The differential diagnosis between infection and infarction of bone in sickle disease can be assisted by dual scanning: in infection, gallium 67 should be much more avidly taken up than technetium in conventional bone scan.

Diagnosis

Osteonecrosis, Childhood. Figure 1 Perthes disease lucent crescent (arrowhead) of the outer femoral head on frog-leg view in an 8-year-old boy. From Oestreich AE, Crawford AH (1985) Atlas of Pediatric Orthopedic Radiology. Thieme Verlag, Stuttgart p 190.

Bones with osteonecrosis are generally denser than the nearby bones, whether because of the loss of volume putting more bone substance into a smaller space from collapse or because living bones demineralize from local irritation whereas dead bones do not. A classic example is the proximal scaphoid being denser than the distal one when its blood supply is lost from a fracture across the bone’s waist. Infarction of bone shafts can also incite callus, which is denser than the nearby bone. The early plain image changes of Perthes disease are a smaller epiphysis than contralaterally because of decreased vascularity and, on the frog-leg view alone, a slit of lucency below the outer femoral head zone of provisional calcification, representing crumbling fracture (Fig. 1). The extent of the slit reflects the severity of involvement. As the disease progresses (with or without surgery to improve position or protect the hip), the head becomes irregularly denser and longitudinally shorter,

O

1440


while the head and neck become transversely wider (and thus incompletely covered by the bony acetabular roof). Gouges of lucency may extend down the metaphysis of the neck from the physis, reflecting avascular zones in the primary spongiosa. Without ultrasound or cross-sectional imaging, one cannot determine whether effusion accompanies Perthes disease. The perthetic hip through the progress of disease becomes varus as the greater trochanter

Osteonecrosis, Childhood. Figure 2 Frostbite sequelae: distal phalanges 2 through 5 are short from physeal closure, as are middle phalanges 4 and 5. The distal (acrophyseal) ends of the middle phalanges are also irregular from enchondral damage. As is often the case, the thumb was presumably spared by being protected in the fist during the cold exposure. From Oestreich AE, Crawford AH (1985) Atlas of Pediatric Orthopedic Radiology. Thieme Verlag, Stuttgart p 169.

continues to grow nearly normally, while the femoral physis is slowed, or occasionally tethered, by premature fusion across part of the physis. Sickle cell infarction of small tubular bones, generally about 1 year of age, is painful accompanied by periosteal reaction and called “hand–foot” syndrome. This may be the presenting symptom of previously unknown sickle cell disease. The typical Lincoln log vertebral bodies in sickle cell disease result from infarction or impairment of the more central portions of the vertebral end plates. Distinguishing infarction from infection in long bones in sickle cell disease by imaging can be quite difficult; dual nuclear scanning with bone scan and gallium scan (or labeled white blood cells) may help by showing especially high activity on the latter. In frostbite, Kashin–Beck disease, and some sequelae of rat bite (1), enchondral bone growth is locally destroyed, so that portions or the entire bones do not grow (Fig. 2). Cone-shaped epiphyses, fused epiphyses, pumice-shaped carpal bones, and irregular nonepiphyseal ends of small tubular bones are manifestations. In frostbite the distribution is acral, from the fingertips proximally, because of the nature of the cold injury, often sparing the thumb if it had been protected in the fist. Distribution of enchondral damage in Kashin–Beck disease is more scattered—involvement asymmetrically of the lower extremity bones leads to length discrepancy and hence limp. The apophysis of the posterior calcaneus normally appears denser than the rest of the bone. It is no longer considered “Sever disease” unless localized symptoms occur and the nuclear scan is abnormal. Many diagnoses of “Ko¨hler disease” of the tarsal navicular are actually an overlap of multiple ossification centers rather than true osteonecrosis (Fig. 3). Close perusal of oblique and lateral images will usually solve the question. Similarly, the ossification across the closing inferior ischiopubic synchondrosis is often both vigorous and asymmetric from

Osteonecrosis, Childhood. Figure 3 Mimic of Ko¨hler disease in a 6-year-old patient. Lateral image suggests sclerosis and irregularity of the navicular; but the frontal view shows it results merely from overlap, the bone developing from three normal-density growth centers. The pain, incidentally, was lateral, not medial.

Osteoporosis

side-to-side. Unless abnormally increased bone scan activity can be shown, van Neck osteonecrosis should not be considered. Kienbo¨ck disease or lunatomalacia is a real entity, however, often associated with ulna minus. Meyer dysplasia of the hip in the early years of life is an irregular appearing ossification of the femoral head. However, it has normal vascularity on nuclear scan or MRI and uncommonly progresses to true Perthes disease.

References 1.

2.

3.

Oestreich AE (2000) Pediatric arthroses as a sequelae of enchondral damage. Examples of frostbite, Kashin–Beck disease, rat bites and other etiologies [in German]. Radiologie 40:1149–1153 Mahnken AH, Staatz G, Ihme N et al (2002) MR signal intensity characteristics in Legg–Calve–Perthes disease.Value of fat-suppressed (STIR) images and contrast-enhanced T1-weighted images. Acta Radiol 43:329–335 Schittich I (2001) MRI in the diagnosis and treatment of Perthes disease and epiphysiolysis of the head of the femur [in German]. Orthopa¨die 30:519–527

Osteopenia In childhood, a reduction of bone substance either from osteoporosis or hyperparathyroidism or, after physes fuse, also from osteomalacia. ▶Demineralization, Bone, Childhood ▶Osteoporosis

Osteophytes Bony spurs or outgrowths in the proximity of a joint, osteophytes, are found in a variety of musculoskeletal disorders, including degenerative joint disease and DISH. Presence of multiple osteophytes is referred to as osteophytosis. ▶Dish

1441

Synonyms Osteoporosis and osteopenia are precisely defined according to the World Health Organization (WHO) Criteria (see later)

Definitions Osteoporosis is defined as a disease associated with a loss of bone mass and a deterioration of bone structure, both resulting in increased bone fragility and susceptibility to fracture (1). In 2000, the definition given by the National Institutes of Health consensus development conference in 1993 was modified (2). Osteoporosis was now defined as a skeletal disorder characterized by compromised bone strength predisposing to an increased risk of fracture. Because this definition appears fairly abstract, the following statements were added. Bone strength reflects the integration of two main features: bone density and bone quality. Bone density is expressed as grams of mineral per area or volume and in any given individual is determined by peak bone mass and amount of bone loss. Bone quality refers to architecture, turnover, damage accumulation (e.g., microfractures), and mineralization. Because bone density is the parameter that can be determined best in vivo, has a high precision, and correlates well with the biomechanically determined bone strength [it explains approximately 70% of bone strength (2)], the WHO defined osteoporosis on the basis of bone mineral density (BMD) (3). A BMD that is more than 2.5 standard deviations below that of a white, young, healthy female adult reference population (T-score) is defined as osteoporosis. A BMD that is 1–2.5 standard deviations below that of the young and healthy reference population is defined as ▶osteopenia. This definition, however, was originally only established for BMD of the proximal femur determined using ▶dual-energy X-ray absorptiometry (DXA), but it has been applied to define diagnostic thresholds at other skeletal sites, such as the spine (anterior–posterior) and the distal radius, and for other technologies.


Osteoporosis T HOMAS M. L INK , J AN S. B AUER Department of Radiology San Francisco, USA [email protected]

Osteoporosis is characterized by reduced activity of osteoblasts and increased activity of osteoclasts. Because trabecular bone has a turnover up to seven times higher than the cortical shell, bone loss occurs here first. The trabecular network transforms from a platelike to a rodlike structure with thinner trabeculae and wider intertrabecular spaces. Connections between trabeculae

O

1442

Osteoporosis

are lost as osteoclasts form resorption lacunae that may have the size of thinner trabeculae. To quantify trabecular structure changes, histomorphologic parameters can be calculated from biopsies of the iliac crest, or noninvasively using high-resolution magnetic resonance imaging (MRI) or peripheral MicroCT at the distal radius, the tibia, or the calcaneus.

Clinical Presentation Osteoporosis may be undiagnosed for a long time and frequently manifests itself with insufficiency fractures. Early symptoms such as back pain and height loss are nonspecific, and a hunchback represents already advanced osteoporosis that is frequently associated with several vertebral fractures. The most severe insufficiency fractures affect the proximal femur with high associated disability and mortality. Osteoporosis-related vertebral fractures also have important health consequences for older women and men, including disability and increased mortality. The presence of one vertebral fracture increases the risk of any subsequent vertebral fracture fivefold, and 20% of women who have had a recent diagnosis of fracture will sustain a new fracture within the next 12 months. Since a number vertebral fractures do not come to clinical attention, the radiographic diagnosis is particularly important. Because these fractures can be prevented with appropriate medications, recognition and treatment of high-risk patients is warranted.

Imaging Conventional radiographs of the spine are not suited to determine bone mass in the early stage of osteoporosis

Osteoporosis. Figure 1 fractures.

because it takes a bone loss of more than 20–40% before a loss of bone mass is visualized on radiography. Morphological signs described on spine radiographs, such as a coarse trabecular structure and a framelike appearance of the vertebrae, are also not very reliable. Conventional radiographs, however, are important in diagnosing fractures and in the differential diagnosis of osteoporosis because a number of other diseases may present with bone loss and fractures. According to Genant et al (4), a vertebral deformity in T4-L4 of more than 20% of loss in height with a reduction in area of more than 10–20% is defined as a fracture. Using this fracture threshold, a semiquantitative score to grade the severity of vertebral fractures as visually determined from radiographs has been described (Fig. 1). In rare cases, osteoporosis may present with a coarse trabecular structure with thick vertical trabeculae suggestive of vertebral hemangioma. This so-called hypertrophic atrophy, however, is generalized, and the trabecular bone structure appears more coarse than in hemangioma. Important differential diagnoses in osteoporosis are osteomalacia, hyperparathyroidism, renal osteopathia, and malignant bone marrow disorders such as multiple myeloma and diffuse metastatic disease. Endplate fractures are found in Scheuermann’s disease and malignant lesions. The differential diagnosis of osteoporotic and malignant pathologic fractures may be difficult. Fractures located above the Th 7 level present with a soft tissue mass, osseous destruction, and fractures of the posterior part of the vertebrae in conventional radiographs are more likely to be malignant. Conventional radiographs of the proximal femur and the distal radius are usually obtained after a low-impact trauma with persistent symptoms in postmenopausal elderly individuals. It should be noted that osteoporotic

Spinal fracture index as defined by Genant et al (4) used to classify osteoporotic vertebral

Osteoporosis

fractures may be difficult to detect on conventional radiographs due to demineralization of the bone and are not infrequently occult. Computed tomography (CT) and MRI may be helpful in detecting occult fractures (Fig. 2), differentiating osteoporotic and malignant fractures, depicting multiple lesions, and soft tissue masses or destructive changes. Bogost et al showed that 37% of proximal femur fractures were not detected in conventional radiographs, which were demonstrated in MR scans of these patients (5). Nonenhanced T1 and STIR (respectively T2-weighted fatsaturated) sequences are recommended in patients with a high clinical suspicion of fracture but negative radiographs. Diffusion-weighted MR sequences and iron oxide contrast media in MRI have been successfully used to differentiate malignant and benign bone marrow pathology. CT is less sensitive in depicting bone marrow pathology; however, it is better suited to assess the stability of an osteolytic lesion or fracture because it directly visualizes the bony structures and demonstrates fracture lines in a detailed fashion.

Nuclear Medicine Bone scintigraphy has a limited role in the assessment of osteoporosis. The technique may be useful to detect occult fractures (though it is less sensitive than MRI), to depict multiple lesions, and to differentiate between old and new vertebral fractures.

Diagnosis Diagnosis of osteoporosis is usually made on the basis of bone density or the presence of osteoporotic fractures. Currently the most important techniques in osteodensitometry are DXA and ▶quantitative computed tomography (QCT).

1443

DXA: The principle of DXA is a dual-energy measurement that is based on the fact that radiation of distinct energies is attenuated by tissues to different extents. In both soft tissue and bone, a low-energy beam is attenuated to a greater degree than a high-energy beam is. Contrast in attenuation between bone and soft tissue is greater for the low-energy beam than for the high-energy beam, such that the attenuation profile of bone may be determined by subtracting both the low- and high-energy attenuation profiles. DXA scanners provide either pencil or fan beam techniques. Fan beam techniques are faster. The precision of DXA is high, and radiation exposure is low. BMD is most frequently determined at the spine (anteroposterior or lateral) (Fig. 3a) and at the proximal femur (Fig. 3b). Wholebody measurements as well as measurements at the distal radius and the calcaneus may also be obtained. The anteroposterior examination of the lumbar spine is a standard procedure with a precision in vivo of 1%, a radiation exposure of 1(-50) mSv (the higher dose is required for digital high-resolution images), and a fairly high accuracy (4–10%). For monitoring BMD, the precision alone, however, is not the only parameter required to assess the diagnostic performance of a technique. We also need to know the annual rate of BMD loss in normal patients using this technique as well as the least significant change between two measurements, which are 1–2% and 3–4%, respectively, for the anteroposterior spine. Using automated software, areal BMD (g/cm2) is determined, usually at L1–L4 (Fig. 3a). These projection images, however, have a number of limitations: (i) vertebrae with a larger size have a higher BMD, (ii) aortic calcification and all other soft tissue calcifications in the regions of interest (ROIs) increase BMD, and (iii) degenerative changes of the spine including osteophytes, facet sclerosis, and degenerative disc disease may also falsely increase BMD. In elderly patients with substantial degenerative changes of the lumbar spine, anteroposterior DXA of the lumbar spine may therefore not be a suitable technique. Lateral

Osteoporosis. Figure 2 Radiographically occult insufficiency fracture of the right proximal femur in a postmenopausal woman. On the radiograph of the right femur (a), no fracture is shown, whereas the coronal T1-weighted magnetic resonance image (b) clearly shows a fracture of the right femur neck.

O

1444

Osteoporosis

Osteoporosis. Figure 3 Standard techniques to measure bone mineral density: dual-energy X-ray absorptiometry images of the anteroposterior spine (a) and the proximal femur with standard regions of interest (b) as well as quantitative computed tomography of the lumbar spine (c) with an oval region of interest in the vertebral body.

DXA is influenced less by these changes because it assesses only the vertebral bodies and thus focuses more on trabecular bone. However, drawbacks of this technique are a lower precision, a higher radiation exposure, and superimposition of the pelvis and the ribs, which may limit analysis of the lumbar spine to L3. So far, anteroposterior DXA is still the standard DXA procedure to assess the lumbar spine. When analyzing DXA scans, a number of pitfalls have to be considered that may be operatordependent, such as mislabeled vertebrae, misplaced disk space markers, wrongly sized ROIs, and artifacts in the analysis region. These analysis errors are of greater magnitude than the machine’s intrinsic precision errors. DXA of the proximal femur is a particularly important examination because it is currently one of the best techniques to assess fracture risk of the hip (Fig. 3b). But the examination of the hip is more demanding than that of the spine; the proximal femur has to be positioned in a standardized fashion, and a number of ROIs have to be placed correctly. The correct location of these ROIs varies according to the manufacturer. Standard ROIs are the neck region, the trochanteric region, and the intertrochanteric region. The ROI used most frequently is the total femur. The total femur ROI consists of the neck region, the trochanteric region, and the intertrochanteric region. Ward’s triangle has an inferior precision compared with the other ROIs and is currently not used as a standard ROI. The precision for hip BMD and the annual rate of loss are lower compared with anteroposterior spine, and the least significant change is higher. As in DXA of the lumbar spine, a number of operatordependent errors may occur in the proximal femur and should be detected by the radiologist. Most of these errors are due to improper positioning of the patient and the ROIs. Correct positioning of the patient includes internal rotation of the hip with a straight femoral shaft (the lesser trochanter should not or just barely be visualized).

Correct positioning and size of the ROIs, in particular the neck box, may vary according to the manufacturer; for example, Lunar/GE systems have a standardized size of the neck box, which is placed automatically in the region of the neck with the smallest diameter. Osteoarthritis, Paget’s disease, fracture, vascular calcifications, calcific tendinitis, enostosis, and avascular necrosis of the hip are also potential sources of error. Conventional radiographs may be required if an atypical density profile is shown. If these lesions are too large or if developmental dysplasia of the hip is found, BMD has to be determined at a different site. QCT: In contrast to DXA, QCT allows a true densitometric measurement (in mg/mL) of trabecular bone, whereas DXA gives an areal BMD (in mg/cm2) that includes trabecular and cortical bone. Since the trabecular bone has a substantially higher metabolic turnover, it is more sensitive to changes in BMD (annual rate of bone loss in QCT 2–4% vs. 1% in anteroposterior DXA of the spine). On the other hand, the precision of two-dimensional (2D) QCT (but not of volumetric QCT) is lower than that of DXA (1.5–4% vs. 1%). A big advantage of QCT is that it is not as susceptible to degenerative changes of the spine as DXA is. Osteophytes and facet joint degeneration as well as soft tissue calcifications (in particular of aortic calcifications) usually do not falsely elevate the BMD in QCT. As in DXA, however, fractured or deformed vertebrae must not be used for BMD assessment because these vertebrae usually have an increased BMD. QCT may be performed with any CT system; however, a calibration phantom is required to transform the attenuation measured in Hounsfield units (HU) into BMD (mg/mL). Dedicated software improves the precision of the examination. The patient is examined lying supine on the phantom, usually with a water- or gel-filled cushion in between to avoid artifacts due to air gaps. Standard 2D QCT is performed of the lumbar spine; usually, the first to third lumbar vertebrae are analyzed

Ovarian Cancer

using a single midvertebral section that is aligned along the endplates. Volumetric QCT can be performed of the spine and the proximal femur using axial, contiguous 3-mm sections of each site. BMD data obtained by QCT are compared to an age-, sex-, and race-matched database. T-scores used for the assessment of osteoporosis according to the WHO definition have been established for DXA but not for QCT, though they may be given by the manufacturers of the software. If these T-scores are used to diagnose osteoporosis, a substantially higher number of individuals compared to DXA will be diagnosed as osteoporotic, since BMD measured with QCT shows a faster decrease with age than DXA. Researchers have therefore advocated using BMD measurements analogous to the WHO definition but with thresholds corresponding to lower T-scores. Thus, BMD values from 80–120 mg/mL have been classified as osteopenic, and BMD values below 80 mg/mL as osteoporotic (6), which corresponds to a T-score of approximately -3.0. Peripheral QCT and DXA are less frequently used and clinically of limited significance. New techniques to assess microarchitecture and macroarchitecture of bone have not been introduced into the clinical arena but may give additional information on fracture risk and have future potential. ▶Acromegaly ▶Demineralization, Bone

Bibliography 1. 2.

3.

4.

5.

6.

NIH, Conference CD (1993) Diagnosis, prophylaxis, and treatment of osteoporosis. Am J Med 94:646–650 NIH Consensus Development Panel on Osteoporosis Prevention D, and Therapy (2001) Osteoporosis prevention, diagnosis, and therapy. JAMA 285:785–795 WHO (1994) Technical report: assessment of fracture risk and its application to screening for postmenopausal osteoporosis: a report of a WHO study group. In: World Health Organization. Geneva, Switzerland Genant HK, Wu CY, Kuijk C van et al (1993) Vertebral fracture assessment using a semiquantitative technique. J Bone Miner Res 8:1137–1148 Bogost G, Lizerbram E, Crues J (1995) MR imaging in evaluation of suspected hip fracture: frequency of unsuspected bone and softtissue injury. Radiology 197:263–267 Felsenberg D, Gowin W (1999) Knochendichtemessung mit Zwei-Spektren-Methoden. Radiologe 39:186–193

Osteoporotic Vertebral Fractures According to the spinal fracture index, deformities of the vertebrae of more than 20% are defined as fractures. These are not infrequently asymptomatic but are excellent

1445

indicators for future fractures. Even if bone mineral density is not below a T-score of -2.5, the presence of fractures is indicative of osteoporosis. ▶Osteoporosis

Osteosarcoma Most frequent malignant primary bone tumour found predominantly in young adults. Typically aggressive morphology in radiological images with complex periosteal reaction and osteoblastic matrix pattern. ▶Neoplasms, Bone, Malignant

Osteosclerosis The hardening or the abnormally high density of bone. ▶Neoplasms, Odontogenic

Outlet Obstruction Syndrome Outlet obstruction syndrome, also called obstructed defecation is defined as incomplete evacuation of fecal contents from the rectum. ▶Pelvic Floor Dysfunction, Anorectal Manifestations

Outlet Obstruction Syndrome – Obstructed Defecation ▶Pelvic Floor Dysfunction, Anorectal Manifestations

Ovarian Cancer Ovarian cancer is in the majority (85%) of cases epithelial in origin. Familial evidence of ovarian cancer is the strongest risk factor for ovarian cancer. At diagnosis in more than 75% of patients with epithelial ovarian cancer, peritoneal tumor spread outside the pelvis or lymphatic metastases are detected. ▶Carcinoma, Ovarium

O

1446

Ovarian Cancer Screening

Ovarian Cancer Screening Screening with CA-125 and sonography are currently only suggested in high-risk patients. These are patients with a positive family history of ovarian cancer or BRCA1 or BRCA2 gene mutations. ▶Carcinoma, Ovarium

Ovarian Cancer Staging The most commonly used staging system of ovarian cancer is the FIGO (International Federation of Gynecologists and Obstetricians) classification system. Alternatively, ovarian cancer is staged on the basis of the TNM classification. Staging is based on the findings detected during explorative laparotomy including cytologic assessment of the peritoneum. ▶Carcinoma, Ovarium

Ovarian Teratomas Ovarian teratomas consist of a series of tumors which derive from primordial germ cells. They comprise mature teratomas, immature teratomas, and monodermal teratomas. Mature teratomas constitute the vast majority of these tumors (99%). ▶Teratoma, Ovaries, Mature, Ovalar

Ovarian Torsion Ovarian torsion is the most important complication of dermoids. It is reported in 3.2–16% of cases and is a gynecological emergency. Increasing tumor size correlates with increased risk of torsion. ▶Teratoma, Ovaries, Mature, Ovalar

Ovarian Vein Obstruction Ovarian Cancer ▶Carcinoma, Ovarium

▶Thrombosis, Vein, Ovarian

Ovarian Vein Occlusion ▶Thrombosis, Vein, Ovarian

Ovarian Metastases Ovarian metastases comprise approximately 5–15% of malignant ovarian tumors, and derive most commonly from colon, stomach, breast, and melanomas as primary cancers. Krukenberg tumors display characteristic imaging features, which include bilateral, solid ovarian tumors, often with central necrosis. Other ovarian metastases present with similar imaging findings as ovarian cancer. ▶Masses, ovarian

Ovarian Neoplasm ▶Masses, Ovarian

Oxygen–Ozone Therapy A mixture of oxygen–ozone gas is injected into the centre of the intervertebral disc at non-toxic concentrations. Ozone has a direct lytic effect on the proteoglycan molecules that form the nucleus pulposus. The resulting proteoglycan fragments have limited water-binding abilities, which leave intradiscal water molecules free to diffuse into the surrounding tissues. The resulting loss of water causes a decrease of intradiscal pressure. This causes the herniated portion of the nucleus pulposus to recede towards the centre of the disc, relieving pressure on the nerve root. ▶Percutaneous Interventions for Lumbar Radicular Syndrome

P

Paget Disease M ARK DAVIES MRI Centre, Royal Orthopedic Hospital NHS Trust Birmingham, UK [email protected]

Synonym Osteitis deformans

Definition Paget’s disease of bone named after the nineteenth century British surgeon Sir James Paget is a localised or multifocal disorder of bone characterised by abnormal bone turnover with increased osteoclastic activity and compensatory increased osteoblastic activity. It has three phases, lytic, mixed and sclerotic. The aetiology of the disease is unknown with two principal theories. One is thought to be a genetic predisposition and the other a ‘slow’ virus infection.

Pathology/Histopathology Mirroring the radiographic features, the microscopic appearances depend on the phase of the disease. In the initial osteolytic phase there is active osteoclastic bone resorption with loss of trabeculae and replacement of the marrow by highly vascular fibrous tissue. In time osteoblastic activity predominates with increased bone density and thickening of the remaining trabeculae. The disordered bone turnover produces cement lines or reversal fronts that gives the classic histological ‘mosaic pattern’. In the late ‘burnt-out’ phase bone turnover may return to an almost normal level and the degree of hypervascularity is also reduced. In this situation, the microscopic appearances show thickened trabeculated bone with a prominent mosaic pattern and restoration of the marrow.

Clinical Presentation There is increasing prevalence of Paget’s disease of bone with age. It is rare in patients under 40 years of age. There are also significant geographic/ethnic variations being relatively common in the Caucasian races of northern Europe and yet rare in blacks and Asians. In northern Europe, the incidence has been estimated to be 3% of the population over the age of 40 rising to 10% in the elderly. There is a male predominance. Studies of the past two decades, however, have suggested that Paget’s disease is disappearing. The majority of cases are an incidental finding on radiographs obtained for an unrelated clinical indication. The remainder present due to the complications of the disease (Table 1). The disease predominates in the axial skeleton-spine (75%), pelvis (60%) and proximal femur (75%). About 75% cases are polystotic and 25% monostotic. Clinical laboratory findings include a grossly elevated serum alkaline phosphatase with normal calcium and phosphorus levels. Urinary hydroxyproline levels are also raised due to increased bone tissue breakdown.

Imaging The initial osteolytic phase is the least frequent manifestation identified on radiographs with the exception of the skull vault. In the skull, there is a large area of demineralisation with a sharp line of demarcation. This appearance, known as osteoporosis circumscripta, commences at the skull base most commonly affecting the

Paget Disease. Table 1 disease of bone 1. Bone softening

Skeletal complications of Paget’s – bowing of long bones – protrusio acetabuli – basilar invagination – partial/complete transverse

2. Fractures 3. Malignant transformation ( 4 cm) (2). In addition, rectoceles are classified into those with complete evacuation and those with incomplete evacuation (Fig. 1), depending on the presence or absence of contrast material retention at the end of defecation. A clinically significant rectocele should be considered based on the following criteria: patient history, size exceeding 2 cm in sagittal diameter, retention of contrast medium, reproducibility of the patient’s symptoms, and the need for evacuation assistance.

1465

straining (Fig. 2). External rectal prolapse is an invagination of the rectal wall through the anal canal (Fig. 3) and is a clinical diagnosis. Regardless of its extension, internal rectal prolapse may involve all rectal wall layers (fullthickness prolapse; Fig. 2) or only the mucosa (mucosal prolapse) (2). Dynamic MRI is superior to conventional defecography for diagnosing internal rectal prolapse because it enables differentiation between a mucosal internal prolapse and a full-thickness internal prolapse. These two types of rectal prolapse have different treatment strategies.

Enterocele Rectal Prolapse Rectal prolapse is an invagination of the rectal wall and may be classified as internal or external (1). Internal rectal prolapse, also called intussusception, may be classified as intrarectal internal prolapse if the invagination is confined to the rectum or as intraanal internal prolapse if its apex penetrates the anal canal and remains in it during

An enterocele is an internal herniation of the peritoneal sac into the rectovaginal space below the pubococcygeal line (PCL). The PCL is used as the reference line for quantifying various abnormal findings of the pelvic floor, including quantification of the severity of an enterocele and the severity of pelvic floor descent. The PCL is defined as the line that joins the inferior border of the symphysis pubis to the last coccygeal joint on midsagittal images (Fig. 1). Enteroceles most frequently occur at the end of evacuation and can be filled with omental fat (peritoneocele), small bowel (enterocele), or sigmoid colon

P

Pelvic Floor Dysfunction, Anorectal Manifestations. Figure 1 A 63-year-old woman with fecal incontinence and diffuse pelvic pain after hysterectomy. A T1-weighted spoiled gradient-recalled echo magnetic resonance (MR) image obtained during evacuation (patient in supine position in a closed-configuration MR system) shows a moderate anterior rectocele with incomplete evacuation (small arrow). The anorectal junction is below the ▶pubococcygeal line (PCL), with formation of a moderate rectal descent (large arrow). In addition, the patient has a moderate sigmoidocele (black arrow). The bladder (Bd) is in the normal position.

Pelvic Floor Dysfunction, Anorectal Manifestations. Figure 2 A 25-year-old man with chronic constipation and incomplete evacuation. A sagittal T1-weighted spoiled gradient-recalled echo magnetic resonance image with the patient in sitting position obtained during defecation shows a full-thickness intraanal rectal prolapse (large arrow). The anorectal junction (small arrow) is below the pubococcygeal line, with formation of a moderate rectal descent.

1466

Pelvic Floor Dysfunction, Genitourinary

when the distance between the anorectal junction and PCL is more than 6 cm (5). Cystoceles are expressions of the descent of the anterior compartment, and descent of the vaginal vault (or any part of the remaining cervix in the case of hysterectomy) represents descent of the middle compartment. The descent of these compartments is quantified in a similar fashion as described for rectal descent.

References 1. 2.

3. 4.

Pelvic Floor Dysfunction, Genitourinary. Figure 3 A 49-year-old man with incomplete evacuation. A sagittal T1-weighted spoiled gradient-recalled echo magnetic resonance image obtained during defecation with the patient in sitting position shows a full-thickness external rectal prolapse (large arrow). In addition, the patient presents a moderate rectal descent (small arrow).

(sigmoidocele; Fig. 1). While most enteroceles spontaneously disappear at rest, only a few are fixed down at the rectovaginal space. The extent of an enterocele is measured 90˚ to the PCL from the lowest margin of herniation content during evacuation and is classified as small if it extends less than 3 cm below the PCL, moderate if it extends from 3 to 6 cm below this line, and large if it extends 6 cm or more below this line.

5.

Stoker J, Halligan S, and Bartram CI (2001) Pelvic floor imaging. Radiology 218:621–641 Roos JE, Weishaupt D, Wildermuth S et al (2002) Experience of 4 years with open MR defecography: pictorial review of anorectal anatomy and disease. Radiographics 22:817–832 Rociu E, Stoker J, Eijkemans MJ et al (1999) Fecal incontinence: endoanal US versus endoanal MR imaging. Radiology 212:453–458 Bolog N and Weishaupt D (2005) Dynamic MR imaging of outlet obstruction. Rom J Gastroenterol 14:293–302 Bertschinger KM, Hetzer FH, Roos JE et al (2002) Dynamic MR imaging of the pelvic floor performed with patient sitting in an open-magnet unit versus with patient supine in a closed-magnet unit. Radiology 223:501–508

Pelvic Floor Dysfunction, Genitourinary T HOMAS C. T REUMANN Kantonspital Luzern, Luzern, Switzerland [email protected]

Definition Rectal Descent Rectal descent is an excessive caudal movement of the anorectal junction during evacuation. Pelvic floor descent is a common finding in patients with fecal incontinence and outlet obstruction. Although the clinical examination and electrophysiological tests play a significant role in diagnosis, dynamic pelvic MRI provides the most accurate assessment of pelvic floor descent. Rectal descent is usually associated with descent of the middle and anterior pelvic compartments. MRI enables evaluation of all three pelvic compartments (posterior, middle, and anterior) in a manner superior to conventional defecography (4). The grading of the rectal descent is assessed with respect to the PCL (5). A rectal descent is considered small when the anorectal junction is less than 3 cm below the PCL, moderate when the distance between the anorectal junction and PCL is between 3 and 6 cm (Fig. 1), and large

Pelvic floor dysfunction describes a wide range of problems of the three organ systems passing the pelvic floor: the urinary system, the genital system, and the colorectal system. It is associated with pelvic floor weakness or with organic or functional neuromuscular disorders. Conditions may include pelvic floor pain, urinary incontinence, bowel incontinence, stool outlet obstruction, pelvic organ descent, pelvic organ prolapse, and pelvic floor dyssynergy or spasm.

Pathology Anatomically, the pelvic floor is formed by the genitourinary diaphragm and the levator ani muscle. The two parts of the levator ani muscle are the puborectalis muscle


sling and the iliococcygeus muscle. The puborectalis muscle sling originates from the posterior surface of the pubic symphysis and goes around urethra, vagina, and anus. Urethra, vagina, and anal canal pass the pelvic diaphragm within the midline. Contraction of the muscle pulls the rectum toward the symphysis and closes urethra, vagina, and rectum. Weakness of this muscle causes laxity of the genitourinary hiatus. The iliococcygeus muscle plates originate at the genitourinary diaphragm and insert at the lateral inner surfaces of the pelvis. Contraction of this muscle elevates the pelvic floor. Weakness of this muscle leads to descent of the pelvic floor. Pelvic floor dysfunction is a disease of elderly female patients. The underlying cause for incontinence and organ prolapse is pelvic floor laxity, usually resulting from compromised levator ani and puborectalis muscle function and endopelvic fascia laxity and fascia defects. Predisposing factors are multiple childbirths, vaginal delivery, menopause, age, hysterectomy, and inherited connective tissue laxity from collagen defects. The cause of urinary incontinence is not fully understood. Incontinence correlates with obesity and smoking, and only to a lesser extent with pelvic floor laxity and bladder neck hypermobility. The normal process of urination involves coordination between the bladder and pelvic floor muscles; the pelvic floor muscles relax whereas the bladder contracts. In most patients, the bladder contracts normally but the urine flow is poor due to continued tightening of the pelvic floor. The sensation of urinary frequency and urgency seem to be linked to pelvic floor laxity. The natural tendency for many patients is to strain and push the last drop of urine out so they can prolong the intervals between bathroom sessions. Unfortunately, pushing and straining appear to further damage the pelvic floor muscles and worsen the symptoms. Hence, a vicious cycle develops. Pelvic organ descent and prolapse is the result of laxity of the connective tissues in the pelvis, especially defects in the lateral fascial bands of the endopelvic fascia between the vagina and lateral pelvic wall, and defects in the rectovaginal septum. Hysterectomy seems to be a risk factor for pelvic organ descent. Organ descent develops during Valsalva maneuver, voiding, and defecation. Descent of the posterior bladder wall below the internal urethral orifice is called a ▶cystocele. Bulging of the rectum toward the posterior vaginal wall is called a ▶rectocele. An ▶enterocele is the descent of the peritoneal sac between vagina and rectum down to the level of the perineum with bulging of the posterior vaginal wall. An enterocele may follow a rectocele after defecation (Fig. 1). A cystocele Grade II leads to impression of the vaginal wall inside the vagina. A cystocele grade III everts part of

1467

Pelvic Floor Dysfunction, Genitourinary. Figure 1 Midsagittal T2-weighted MR image of the pelvis and pelvic floor during Valsalva maneuver after partial defecation. Rectocele (R) with retention of bowel content. Enterocele (E) between rectocele and posterior vaginal wall (black arrow). Partial prolapse of the posterior vaginal wall (white arrow) as result of the enterocele. S is the sigmoid colon, A is the anus, U.bl. is the urinary bladder, and Ur is the urethra.

the vaginal wall to below the introitus. Rectoceles and enteroceles are graded similarly. Descent of the uterus can be quantified by measurement of the change of the vaginal length during Valsalva. Prolapse of the uterus begins with descent of the portio below the introitus. Stool outlet obstruction may be caused by two conditions: pelvic floor laxity and pelvic floor dyssynergy. Pelvic floor laxity can lead to descent of the peritoneal sac or uterus and to compression of the rectum by these organs. A large rectocele can retain stool. Compression of a rectocele by the peritoneal sac may intussuscept the anterior rectal wall into the anal canal. Detachment of the rectum from the sacrum in pelvic floor descent may lead to intussusception of the rectum in itself. In dyssynergy, the pelvic floor contracts during pressure raise in the abdomen. In anism, the anal sphincter is spastic and hypertrophies. This may be a behavioral problem or may be part of an organic neurologic pathology, such as Hirschsprung’s disease. Damage to the nerves in the pelvis and pelvic floor can be a result of pelvic floor laxity with permanent stretching of the nerves.

P

1468


Clinical Presentation Symptoms associated with anterior compartment may be ▶stress urinary incontinence, urinary urgency with increased frequency of urination and decreased urinary flow, and sensation of incomplete urination with residual bladder volume. In genital descent, patients may present with protrusion of parts of the anterior vaginal wall by a cystocele or of the posterior vaginal wall by a rectocele or enterocele. Protrusion may go as far as complete prolapse and eversion of the vagina to the outside. Descent and prolapse may occur only during straining, or it may be fixed with permanent vaginal prolapse. If the uterus is still present, the portio may descend down to the pelvic floor, or the uterus may protrude partially or completely to outside the body. In the posterior compartment, rectal mucosa prolapse or rectal wall prolapse may develop in patients with functional disorders with permanent high-pressure defecation. Muscular spasm causes pelvic floor pain. In stool outlet obstruction by rectal descent and intussusception, the patient complains about defecation in small fractions. Large rectoceles with retention of bowel content raise the need to digitize from the vagina to empty the rectocele. In anisms or dyssynergy, defecation needs high intraabdominal pressure, and the defecation process is prolonged.

Imaging Ultrasound is usually performed by the gynecologist as part of the diagnostic work-up of stress urinary incontinence and genitourinary prolapse. It allows for the morphological and dynamic assessment of the lower urinary tract. Transvaginal ultrasound enables the detection of pathologies of the bladder and uterus. Introital ultrasound can detect urethral diverticula, periurethral masses, funneling of the urethra and cystoceles. Color Doppler may be used to document stress urinary incontinence. Translabial ultrasound detects defects in the rectovaginal septum and helps to identify rectoceles. Roentgen colpocystodefecography is the original imaging method to evaluate pelvic floor descent. It requires filling of the urinary bladder, vagina, rectum, bowel, and peritoneum with roentgen contrast medium and is associated with a high-dose radiation exposure. The method is now replaced by MRI. MRI of the pelvic floor is performed as a combination of static and dynamic sequences. Static MRI is done with T2-weighted images of the pelvis in three imaging planes. These images give an overview of the pelvic organs and

the pelvic floor muscles in the resting position. T1- or proton density images may be used for the pelvic floor muscles additionally. Dynamic MRI is performed with repetitive fast single shot T2-weighted images. Best contrast-to-noise, excellent T2 contrast, and shortest acquisition times are provided by True FISP sequences. Acquisition times of 0.5 to 0.7 sec per image can be achieved with a 1 mm in plane resolution. Parallel imaging can even shorten the imaging time. A delay of 0.1 to 0.2 sec between should be left between images to reduce signal loss by spin saturation. Alternatively, half Fourier single shot turbo spin echo (HASTE) sequences can be used, but are not as fast as True FISP. T2-weighted imaging requires filling of the vagina and the rectum with watery contrast material. Ultrasound gel is recommended. The patient should empty the bladder before going into the magnet. Normal urine production leads to sufficient opacification of the bladder during investigation (Fig. 2). A repetitive single slice in the mid sagittal plane of 10 mm thickness is acquired during pelvic floor contraction and Valsalva maneuver with defecation. For contraction series, we use 30 images per series, for Valsalva and defecation, 80 to 100 images are required. The series is repeated as often as necessary to see best contraction and relaxation and maximum descent of pelvic floor, best relaxation of levator ani and puborectalis muscle, maximum organ descent and organ prolapse and complete emptying of the rectum. The investigation is performed in supine position of the patient, with the knees slightly flexed and thighs abducted. The patient is equipped with a pampers. Studies have shown that no relevant pathology is missed in supine position compared to upright position. MR images are evaluated qualitatively and quantitatively. We recommend to use the HMO-system introduced by Comiter et al in 1999. Pelvic floor laxity is described by the length of the Hiatus line from the posterior border of the pubic symphysis to the sling of the puborectalis muscle behind the anorectal junction. Normal range is 7±1.5 cm. Pelvic floor descent is described by the position of the puborectalis muscle sling in relation to the pubococcygeal line (PCL). Normal range is between 4 and 6 cm (Fig. 3). Organ descent is seen directly on the MR images and can be graded according to the description in the pathology section. Organ descent can be measured in relation to the Hiatus line. Mobility of the bladder neck can be quantified using the rotation angle of the long axis of the urethra. The pathologic threshold is not clearly defined, but the 30˚ rotation, adapted from the ▶Q-tip test, seems to be oversensitive. A rotation of more than 90˚ or beyond the horizontal plane is definitely abnormal.


1469

Pelvic Floor Dysfunction, Genitourinary. Figure 2 Midsagittal T2-weighted MR images of the pelvis and pelvic floor during defecation. Marked anterior bulging of the anterior rectal wall, representing a rectocele (Rc), with partial prolapse of the posterior vaginal wall (a). Incomplete evacuation of the rectocele because of fold formation in the posterior rectal wall covering the anal canal (b). Descent of the peritoneal sac and sigmoid colon (S) in between the posterior vaginal wall (white arrowhead) and the anterior wall of the rectum (black arrowheads). Ur.bl. is the urinary bladder, V is the vagina, and A is the anus.

Diagnosis The diagnosis of pelvic floor dysfunction is made on the basis of symptoms, clinical findings and of the exclusion of other causes for the clinical disorder. Urinary incontinence requires urodynamic studies, a pad weigh test, and the exclusion of urinary tract infection. Stool outlet obstruction requires exclusion of drugs, slow transit constipation, dolichocolon, and neoplastic disease. Endoscopy should be performed in any colorectal disorder. Anal manometry is required to exclude Hirschsprung’s disease and anal sphincter hypertrophy, spasm, or laxity. Stool incontinence requires neurologic assessment. In any suspicion of a neurologic disorder, MRI of the spinal canal should be performed to rule out tumor or damage to the spinal cord. Palpation and visual observation of the pelvic floor, electromyography, ultrasound, and magnetic resonance imaging (MRI) during rest, contraction, Valsalva maneuver, voiding, and defecation measure different aspects of pelvic floor function. Since there is no specific test for pelvic floor dysfunction, several methods have to be combined to focus the problem.

MRI is the most helpful imaging method to illustrate pelvic floor function. It is less investigatordependent than ultrasonography, does not need ionizing radiation and provides us with high-quality images not only of the pelvic floor, but also of the pelvic organs, and generates dynamic information, playable as video sequences. It is superior to clinical investigation and ultrasound in detection of enteroceles and differentiation of rectoceles from enteroceles and combined problems. Up to 60% of enteroceles are missed by clinical investigation. In pelvic floor dysfunction and anism, MRI is the method of choice to demonstrate the problem to the patient.

Bibliography 1.

2.

Pannu HK, Kaufman HS, Cundiff GW et al (2000) Dynamic MR imaging of pelvic organ prolapse: spectrum of abnormalities. Radiographics 20:1567–1582 Comiter CV, Vasavada SP, Barbaric ZL et al (1999) Grading pelvic prolapse and pelvic floor relaxation using dynamic magnetic resonance imaging. Urology 54:454–457

P

1470

Pelvic Inflammatory Disease (PID)

Pelvimetry, Magnetic Resonance. Figure 3 HMO-system to quantify pelvic floor morphology. Midsagittal T2-weighted MR image of the pelvis and pelvic floor in resting position (a) and during Valsalva maneuver after defecation (b). The urogenital hiatus is represented by the Hiatus line (H), which is drawn from the posterior margin of the pubic symphysis (S) to the muscle sling posterior to the anorectal junction (black arrow in a). The position of the muscular pelvic floor is measured as the distance of the puborectalis muscle sling to the pubococcygeal line (PCL), a fixed line at the pelvic skeleton. The descent or prolapse of the organs is measured in relation to the Hiatus line. There is a partial prolapse of the urinary bladder (U.bl.) during Valsalva maneuver. The extent of the organ descent is represented by the Organ-Line (O). The part of the bladder descending more than 1 cm below the level of the urethra (Ur) is called a cystocele. The light arrows mark the part of the anterior vaginal wall which is prolapsing beyond the introitus by descent of the bladder. Ut is the uterus. Vag is the vagina and Sigm. is the sigmoid colon.

3.

Tunn R, Petri E (2003) Introital and transvaginal ultrasound as the main tool in the assessment of urogenital and pelvic floor dysfunction: an imaging panel and practical approach. Ultrasound Obstet Gynecol 22:205–213

Pelvimetry, Magnetic Resonance S VEN C LAUDE A NDRE M ICHEL

Pelvic Inflammatory Disease (PID) A polymicrobial infection affecting the reproductive tract in females in the reproductive years. PID is most commonly associated with sexually transmitted microorganism with a predominance of Neisseria gonorrhea and Chlamydia trachomatis. ▶Abscess TuboOvarian

Pelvimetry ▶Pelvimetry, Magnetic Resonance

Radiology, Kantonsspital Baden, 5404 Baden, Switzerland [email protected]

Synonyms Pelvimetry

Definition Magnetic resonance (MR) pelvimetry is a method for determining pelvic dimensions and narrowness of the ▶birth canal by MR imaging.

Pelvimetry, Magnetic Resonance

Characteristics MR pelvimetry is a safe way (for both the fetus and the pregnant woman) to measure the diameters of the birth canal, in particular the entrance and the outlet, to detect a mismatch between the fetal head and the birth canal. The measurements help guide the physician in deciding whether vaginal birth is possible or a cesarean section should be performed.

Clinical Presentation The Birth Canal The fetus has to pass only 15 cm through the vagina to reach the light. However, this short way can require a lot of time, pain, and energy for both of the pregnant woman and the fetus. The birth canal has several narrow areas that the fetus must pass through. The measuerments of the diameters of these key areas are performed in different planes through the pelvis. First of all, is the entrance consists of two diameters: the diameter transversa (axial) and the ▶obstetric conjugate (conjugata vera obstetrica, sagittal). If the head is too big, the fetus cannot enter the birth canal which may be indicated a persisting distended abdomen of the pregnant woman. Therefore, in the last days before birth, the abdominal size usually decreases with the descent of the fetal head into the entrance of the birth canal. Within the birth canal, the two most critical points of narrowness are the ▶interspinous (axial) and the ▶intertuberous distances (axial). Finally, the ▶sagittal outlet (sagittal) is a measure of the birth canal’s exit. Although the pelvis consists of bone, all distances can be modified by the birthing position due to the pelvis’ flexibility, which is increased during pregnancy (1). Probably the most important distance for successful vaginal delivery is the interspinous distance, which can be taken as the “point of no return.” Once the fetus passes this landmark, performing an emergency caesareansection becomes dangerous.

Why Do Pelvimetry? There are several indications for measuring the natural narrowness of the pelvis. When a baby is oversized or overweight there may be a mismatch between the birth canal and the size of the fetal head (2), preventing the fetus from entering the birth canal, so that, a cesarean section must be performed. Even more critical is when the baby becomes stuck in one of the critical inner narrow points. In these cases, the fetus may become wedged in the pelvis, putting the fetus and mother at risk for lifethreatening complications.

1471

In most instances, we can overcome these problems by measuring the baby and the size of its head by ultrasound and then perform pelvimetry early enough. However, pelvimetry is also indicated after a prolonged delivery when the mother decides to have another baby. If the pelvic measurement shows a small birth canal compared with the fetal head, a cesarean section should be planned early enough to avoid an emergency intervention. After a primary cesarean section, pelvimetry may indicate the possibility of vaginal birth with the next baby, if desired. Another indication for pelvimetry is after a previous pelvic fracture. The decision of whether birth can proceed vaginally is ultimately up to the treating obstetrician in agreement with the pregnant woman. Birth-limiting pelvic dimensions can be revealed radiographically and with recently updated published data on birth outcomes (3), the risk of a vaginal birth can be estimated.

Imaging External Pelvimetry by the Obstetrician External measurements of the pelvis can be determined very easily and allow an approximate calculation of the internal pelvic parameters of the birth canal. The most widely used formula for the obstetric conjugate is based on the external conjugate (usually 20 cm) minus 9 cm. These values are, however, very indefinite, mainly due to the different thicknesses of the underlining tissues. The pelvic form can also be assessed easily, which may be helpful in decision making prior to birth (Table 1).

Radiological Measurements In the early days, pelvimetry was performed with radiographs (lateral: Colchner-Sussmann, frontal: Martius/ Guthmann) and was often the indicator for cesarean section. But now we know that the concomitant radiation exposure, although low, may damage fetal DNA. Therefore, in pregnant women, MR pelvimetry is preferable in any case. Although conventional radiographic pelvimetry in a non-pregnant woman probably will not lead fetal damage, but the ovaries will nonetheless be exposed to significant levels of radiation. It must be noted that radiographic pelvimetry is a projection examination, and the distances have to be modified by a correction factor based on a ruler placed within the examined field. Again, the surrounding tissue may cause some error. In contrast, MR pelvimetry assesses these diameters directly and much more accurately.

MR Pelvimetry MR utilizes radio waves and a magnetic fields many times stronger even than that of the earth. There has been much

P

1472

Pelvimetry, Magnetic Resonance

Pelvimetry, Magnetic Resonance. Table 1 Forms

Pelvic forms and their risks

Measurements

Gynecoid

Expected problems during birth

Pelvic inlet diameter 12 cm, about 11 cm, pelvic outlet 10 cm Anthropoid Pelvic inlet diameter 12 cm, about 11 cm, pelvic outlet 3 joints simultaneously >6 weeks Arthritis of wrist or finger joints >6 weeks Symmetric arthritis >6 weeks Rheumatoid factor Subcutaneous rheumatoid nodules X-ray signs of rheumatoid arthritis In accordancce with the ARA criteria (1987), rheumatoid arthritis is diagonised when four of seven criteria are positive.

1629

4.

5.

Devauchelle-Pensec V, Saraux A, Alapetite S et al (2002) Diagnostic value of radiographs of the hands and feet in early rheumatoid arthritis. Joint Bone Spine 69:434–441 Van der Heijde D (2004) Quantification of radiological damage in inflammatory arthritis: rheumatoid arthritis, psoriatic arthritis and ankylosing spondylitis. Best Pract Res Clin Rheumatol 18:847–860 Vincent C, Nogueira L, Clavel C et al (2005) Autoantibodies to citrullinated proteins: ACPA. Autoimmunity 38:17–24 Reijnierse M, Dijkmans BAC, Hansen B et al (2001) Neurologic dysfunction in patients with rheumatoid arthritis of the cervical spine. Predictive value of clinical, radiographic and MR imaging parameters. Eur Radiol 11:467–473 Tehranzadeh J, Ashikyan O, Dascalos J (2004) Advanced imaging of early rheumatoid arthritis. Radiol Clin North Am 42:89–107

R Rhizomelic A shortness of the extremities most pronounced in the proximal (humerus and femur) region. An example is achondroplasia. ▶Osteodysplasia

Rickets A pathological condition in child bone metabolism before epiphyseal growth plate closure has taken place

1630

Rickets—Vitamin D Deficiency

characterized by the impaired and delayed mineralization of osteoid and the cartilaginous part of the epiphyseal growth plates. ▶Osteomalacia

Rickets—Vitamin D Deficiency ▶Vitamin Deficiency

Risser Method The Risser classification of the apophyses of the iliac crest enables estimation of the expected growth of the vertebral body and thereby estimation of the progression tendency of scoliosis. Risser classifies the progression of ossification of the apophysis of the os ileum into six grades. In grade 0 the apophyses cannot be seen, and in grade 5 the apophyses are fully fused. ▶Scoliosis

Risser Method Bone Age RICM ▶Radiographic Iodinated Contrast Media

Right Middle Lobe Syndrome An atelectasis of the middle lobe due to narrowing of the middle lobe bronchus by a neoplastic process or chronic inflammatory changes. ▶Atelectasis

Evaluation of growth remaining in the second decade of life based on appearance, then excursion medially, and finally fusion, of the superior iliac apophysis. ▶Bone Age

RNC ▶Radionuclide Cystography

RNP ▶Reflux Nephropathy

Right-to-Left Shunt A right-to-left shunt occurs when a poorly oxygenated venous systemic blood reaches the systemic arterial circulation. PAVM and congenital or acquired cardiac right-to-left shunts can lead to hypoxemia and to ▶Paradoxical Systemic Embolisms ▶Pulmonary Arteriovenous Malformations

Risk Factors The factors, both inherited and acquired through lifestyle which increase the chance of developing breast cancer. ▶Carcinoma, Breast, Demography

Romanus Lesion Romanus lesion is a subdiscal osteitis of the anterior upper corner of the vertebral body with triangular sclerosis and erosion. It is a common feature in ankylosing spondylitis and occurs most often in the thoracolumbar junction segments. ▶Spondyloarthropathies, Seronegative

Root Entry Zone A boundary zone between central and peripheral myelination of cranial nerves close to their brainstem

Ruptured Intervertebral Disk

entrance. This area, of variable length depending on the cranial nerve, is particularly vulnerable to damage. ▶Facial Nerve Palsy

1631

RS ▶Raynaud’s Syndrome

Round Pneumonia Spherical configured pneumonic consolidation mimicking a rounded intrapulmonary mass. ▶Pneumonia in Childhood

RTP ▶Radiation Therapy

Roux-en-Y Gastric Bypass Rupture of Dermoids The surgical procedure consists of forming a small gastric fundal pouch to exclude the remainder of the stomach and duodenum. A jejunal loop is anastomosed to the gastric pouch. A distal side-to-side anastomosis of the excluded jejunal limb and the antegrade-flowing jejunallimb is formed; creating a “Y” shaped intestinal junction. ▶Stomach and Duodenum in Adults Postoperative

RR ▶Regional Recurrence

Rupture is a complication of dermoids that may occur spontaneously or iatrogenically. It causes chemical peritonitis due to intraperitoneal spillage of the tumor contents. ▶Teratoma, Ovaries, Mature, Ovalar

Ruptured Intervertebral Disk ▶Hernia, Disk, Intervertebral

R

S

Sabre-Sheath Trachea A narrowing of the coronal diameter of the intrathoracic portion of the trachea. By contrast, the coronal diameter of the trachea measured above the thoracic inlet is normal. Sabre-sheath trachea is a radiographic finding that is highly suggestive of COPD. The deformity of the trachea is fixed and rigid and the tracheal rings can be densely ossified. ▶Airway Disease

Sagittal Outlet Distance on a midsagittal section from the end of the sacrum to the bottom of the inner cortex of the symphysis. ▶Magnetic Resonance Pelvimetry

Salivary Gland Inflammation ▶Salivary Glands, Inflammation, Acute Chronic

Sacrococcygeal Teratoma This is found in the region of the primitive pit and node. The primitive pit is the depression in the primitive node that connects the notochordal canal with the surface ectoderm and yolk sac. These connections are referred to as the neurenteric canal. ▶Teratoma

Sacroiliacal Joint Arthritis, Sacroiliitis Sacroiliacal joint arthritis, sacroiliitis is the key manifestation in spondathropathies, mainly in ankylosing spondylitis. The typical X-ray imaging triad is destruction, sclerosis, and bony bridging. ▶Spondyloarthropathies, Seronegative

SAE ▶Stimulated Acoustic Emission (SAE)

Salivary Glands, Inflammation, Acute, Chronic S ILVIO M AZZIOTTI Department of Radiological Sciences, University of Messina, Messina, Italy [email protected]

Synonyms Salivary gland inflammation; Sialoadenitis

Definition Sialoadenitis is a condition characterized by inflammation of one or more of the salivary glands. Inflammatory diseases are the most common illness affecting the major salivary glands. There are both acute and chronic forms. Acute inflammation of the salivary glands has only few causes, most commonly being of viral or bacterial origin. On the other hand, chronic salivary inflammation may be caused by numerous processes as infectious, systemic, autoimmune and certain neoplasms.

1634

Salivary Glands, Inflammation, Acute, Chronic

Pathology/Histopathology Acute ▶sialadenitis is a condition characterized by inflammation and swelling of one or more salivary glands. The gland may be diffusely or focally enlarged. The pathologic process is characterized by very important interstitial edema with limited infiltration of periductal and interacinar connective by granulocytes, lymphocytes, and monocytes. Moreover, acinar-ductal epithelium is swollen and desquamed. Granulocytes and round cells can be observed in the ductal lumen. During acute sialadenitis, abscess formation may take place. Predisposing factors include dehydration and excretory duct obstruction caused by stones or fibrosis. Purulent material may be expressed from the ductal orifice in most bacterial infections. In chronic inflammation, salivary glands are normal sized or smaller. In chronic sialadenitis, there are varying degrees of acinar atrophy, lymphoid infiltrate with or without germinal centers, and fibrosis. The ducts exhibit dilatation and hyperplasia of the lining epithelium with metaplasias. Extensive dilatation can result in cyst formation. Goblet cell metaplasia produces abundant mucin. The overall architecture of the gland is usually maintained.

Clinical Presentations Acute inflammation of the salivary glands is usually of viral or bacterial origin; viral and bacterial infections are the most common causes of salivary abnormalities. Mump represents not only the most common cause of parotid swelling, but also the most common viral disorder of the salivary glands. Children are most often affected with peak of incidence at approximately 4–6 years of age. Typically, there is a 2–3 week incubation period with associated fever, malaise, myalgia, and headache which may precede the observed parotid swelling. Salivary gland involvement may be seen in a wide range of other viral illnesses including those caused by cytomegalovirus, lymphocytic choriomeningitis virus, coxsackievirus A, echovirus, and parainfluenza virus type C among others. Acute suppurative sialadenitis is most commonly bacterial in origin and most often involves the parotid, and to a lesser extent, the submandibular glands. This is in part due to the fact that the serous saliva of the parotid has a lower bacteriostatic effect than the more mucous saliva of the ▶submandibular gland. The most common offending agent is Staphylococcus aureus. Other aerobic organisms isolated are Streptococcus pneumoniae, Hemophilus influenzae, and Escherichia coli. Salivary stasis secondary to either obstruction (e.g., sialoliths, ductal

strictures) or decreased production, often as a result of dehydration, is thought to be an integral component in the pathogenesis of this condition. Common clinical settings in which this entity may occur include the elderly postoperative patient after cardiothoracic or gastrointestinal surgery and more frequently the debilitated, dehydrated patient, usually with poor oral hygiene. The typical presentation involves sudden, diffuse enlargement of the gland with associated induration and tenderness. Massage of the involved gland results in the secretion of purulent material from the duct. Patients with an undiagnosed or incompletely treated acute suppurative sialadenitis can develop an intraglandular abscess. Small abscess may form and coalesce to form a larger abscess, or a solitary abscess may develop. It usually manifest as painful swelling of the salivary gland with skin reddening. Decreased salivary flow with stasis is a key factor in chronic sialadenitis. Like acute sialadenitis, this condition is more common in the ▶parotid gland. Its development is often associated with a previous episode of acute suppurative inflammation with subsequent glandular destruction. Another possibility is the recurrent parotitis of childhood which has continued into adulthood. Symptoms include intermittent swelling of the gland, often painful, that may or may not be associated with food assumption. Approximately 80% of patients experience permanent xerostomia. It is very important to differentiate the obstructive and nonobstructive disease, since their treatment and prognosis often vary significantly. The chronic obstructive disease involves the submandibular gland more frequently than the parotid gland. Conversely, the nonobstructive disorders are more common in the parotid glands. Other infectious agents, like Mycobacterium tuberculosis, atypical mycobacteria, Toxoplasma gondii, Actinomyces israeli can determine chronic sialadenistis either as an ascending infection from the oral cavity or as a part of a systemic process. Noninfectious conditions such as autoimmune disease, previous irradiation, or idiopathic causes may also lead to chronic inflammation. Sjogren Syndrome is a chronic, slowly progressive, relatively benign autoimmune disease characterized by lymphocyte-mediated destruction of the exocrine glands resulting in keratoconjunctivitis sicca and xerostomia. The disease primarily affects middle-aged women, but can be seen in all ages.

Imaging Ultrasonography. The examination should be carried out with the highest-frequency transducer possible. Usually, 5–12-MHz wide-band linear transducers are used. In acute inflammation salivary glands are enlarged and with

Salivary Glands, Inflammation, Acute, Chronic

reduced echogenicity (1–3). They may be inhomogeneous for the presence of multiple, small, oval hypoechoic areas in their content. Increased intraglandular blood flow can be depicted at US. Abscesses are hypoanhechoic lesions with posterior acoustic enhancement and blurred borders (1). Central colliquation may be distinguished as an avascular area or identified by means of moving debris. A hyperechoic halo can surround an organized abscess (1). In chronic inflammation, salivary glands are normal sized or smaller, hypoechoic and non homogeneous; usually they do not have increased blood flow at Doppler examination. Salivary glands’ US is the technique to be used in the suspect of calculosis. US features of scialolithiasis include strongly hyperechoic lines or points with distal acoustic shadowing, which represent stones. However, in chronic ductal scialolithiasis complicated by chronic or recurrent inflammation, the gland may lose its function and stones located in a nondilated duct may be difficult to demonstrate. Conventional radiographs. The plain radiographic examination is rarely used today for the evaluation of the major salivary glands. Although conventional radiographs can be obtained quickly and at low cost they are of limited clinical value since they only identify reasonably large, fairly dense calcifications (2). Sialography. Sialography is a radiographic examination of the parotid or submandibular gland that uses a positive contrast agent to demonstrate the ductal salivary anatomy. It is an invasive procedure in which radiopaque contrast medium is injected retrograde into the glands ductal system via the intraoral opening of either Wharton’s or Stensen’s duct. Conventional sialography has been used for investigating the ductal system of major salivary glands. Imaging is performed to delineate inflammatory or obstructive changes within the ductal system of parotid and/or submandibular glands (2, 3). The technique is contraindicated in the acute setting of sialoadenitis, as it could cause retrograde spreading of the infection into the gland. Sialography should be used essentially for assessing chronic sialoadenitis as well as Sjogren’s syndrome (2, 3). Computed tomography. CT examination performed very early in an acute infection, allows to demonstrate dilatation of the central ducts and enhancement of the central ducts and of Stensen’s duct. The salivary gland can appear variably enlarged and has an increased attenuation on basal scans due to the cellular infiltration (2, 3). There may be diffuse enhancement of the gland on postcontrast images, reflecting the increased vascularity. Presence and location of sialoliths can be clearly detected at CT. If present, inflammatory stranding into the overlying subcutaneous tissue is usually identifiable. CT scans obtained in the coronal plane for inflammatory

1635

conditions of the parotid and submandibular glands may be helpful for evaluating the extent of the lesion; in particular the relationship of the inflammatory mass to the floor of the mouth for submandibular lesion and the skull base for the parotid gland masses, both findings of great importance with respect to the surgical approach (2, 3). Because of the fatty attenuation that is intrinsic to the parotid glands, abscesses are particularly well seen on contrast-enhanced CT scan. Most important CT findings in chronic sialadenitis are related to changes from chronic scialolithiasis that may result in a small atrophic gland with focal intraglandular calcifications. More rarely a large ductal stone without signs of acute glandular inflammation can be seen. Magnetic resonance imaging. Magnetic resonance provides better imaging of soft tissue than CT, but it is less suitable to investigate the suspected inflammatory diseases of the salivary glands due to its inability to visualize small potential clinically significant calcifications. In acute inflammation, on MRI the interested gland is variably enlarged with a reduced visualization of both interstitial and ductal component compared to a normal gland. The gland can have either higher o lower signal intensity than normal one on T2-weighted images, depending on whether edema or cellular infiltration predominates (2, 3). The MR sialographic technique with heavily T2-weighted images for depiction of static or nearly static fluids represents a noninvasive technique for the detection of ductal abnormalities in major salivary glands (4). The most obvious advantage of this technique is the fact that images can be obtained without use of ionizing radiation and its noninvasiveness as well. No contrast material has to be injected, and images can be obtained with short acquisition times of less than 10 sec. Contralateral salivary glands can be easily visualized without further positioning of the patient and can serve as controls. Contraindications for conventional sialography such as acute sialadenitis are not contraindications for MR sialography. Ductal structures can also be visualized in cases of complete obstruction of the ductal system. Also, additional cross-sectional images of the glandular parenchyma can be obtained by using conventional T1- and T2-weighted sequences. Furthermore, MR sialography has an excellent sensitivity for visualization of edema in the salivary parenchyma, which is more difficult to diagnose with CT.

Nuclear Medicine Nuclear medicine has a limited role in the assessment of sialadenitis. Sequential scintigraphy is currently employed for assessing the functional status of all the four major salivary glands and evaluating the chronic evolution of glandular damage.

S

1636

Salivary Glands, Neoplasms

Diagnosis

Definition

Most of the salivary glands diseases are characterized only by a few distinct clinical and laboratory tests patterns. Clinical judgment with a careful history and physical examination are still considered of great relevance. However, to obtain a definite diagnosis, imaging techniques are required in most cases. Salivary glands US is the technique to be firstly used because in many cases the nature of underlying disease can be promptly suggested. However, as the imaging identification of calcification is important in the diagnosis of inflammatory process of the salivary glands, in doubtful cases at US, CT represents the examination of choice. In particular, CT is more accurate than US in detecting the presence and the number of sialoliths and in distinguishing multiple clusters of stones from single large stones. Usually, CT in this setting is best performed without administration of contrast material, since small opacified blood vessels may simulate small sialoliths. If an abscess or inflammatory process is suspected, adding enhanced scans after identifying stones on nonenhanced scans may be useful. Conventional sialography today should be used essentially for assessing chronic sialoadenitis unrelated to ▶sialolithiasis. At present, MR sialography should be preferred because of the greater sensitivity in diagnosing inflammatory diseases of the salivary glands so that incannulation of the salivary duct is not required. Finally, in case of suspected autoimmune sialadenitis, the disease could be usefully assessed by mean of needle biopsy.

Epithelial (adenomas and carcinomas) and nonepithelial tumors arising from the major and minor salivary glands. Salivary gland neoplasms represent 3–5% of all head and neck tumors, and occur most frequently (approximately 80%) in the parotid gland. The majority (70–80%) of the parotid gland tumors are benign. Tumors of the submandibular gland and sublingual glands occur less frequently, but are more likely to be malignant; approximately 40% and 70% of tumors are malignant in the submandibular and sublingual glands, respectively. On the other hand, 50% of all minor salivary gland tumors are malignant.

Bibliography 1.

2. 3. 4.

Bialek EF, Fakubowski W, Zajkowski P et al (2006) US of the major salivary glands: anatomy and spatial relationships, pathologic conditions, and pitfalls. Radiographics 26:745–763 Yousem DM, Kraut MA, Chalian AA (2000) Major salivary gland imaging. Radiology 216:19–29 Rabinov JD (2000) Imaging of salivary gland pathology. Radiol Clin North Am 38:1047–1057 Kalinowski M, Heverhagen JT, Rehberg E et al (2002) Comparative study of MR sialography and digital subtraction sialography for benign salivary gland disorders. Am J Neuroradiol 23:1485–1492

Salivary Glands, Neoplasms M ISA S UMI , TAKASHI N AKAMURA Department of Radiology and Cancer Biology, Nagasaki University School of Dentistry, Nagasaki, Japan [email protected]

Pathology/Histopathology WHO’s histological classification of salivary gland tumors includes 13 adenomas and 24 carcinomas (1). Pleomorphic adenomas are the most common benign epithelial tumors of the salivary glands. This benign tumor usually appears as an ovoid, well-circumscribed mass that may exhibit a pedunculated outgrowth, for example, from the superficial to deep lobes of the parotid gland. The lesion may be associated with cystic change, hemorrhage, and calcification. In major parts of the tumor contain glandular, ductal, or solid structures of epithelial elements. The lesion also contains mesenchymal tissues, which may be frequently associated with chondroid and fibromyxomatous tissues. Warthin’s tumor exclusively occurs in the parotid gland, typically in the tail of the gland, and is the second most common benign lesion of the parotid gland. It may occur as multiple masses within one or both parotid glands. Cystic change is among the characteristic features of the lesion. Warthin’s tumor is composed of a double layer of oncocytes that line the papillary projections, extending into the cystic spaces. Mucoepidermoid carcinoma represents approximately 30% of the malignant tumors of the salivary glands, and arises most commonly (approximately 50%) in the parotid gland and the second most commonly (approximately 45%) in the minor salivary glands. The tumors may be classified as low, intermediate, or high grade. Low-grade tumors are well-circumscribed, usually contain large cystic areas with mucinous components, and abundant goblet cells that are lined by squamous components. Low-grade tumors are rarely associated with nuclear polymorphism, mitotic features, or necrosis. In contrast, high-grade tumors are poorly circumscribed, infiltrative, and exhibit dominant proliferation of the squamous components; goblet cells are rarely observed. Adenoid cystic carcinoma represents the most common malignancy arising from the submandibular and


sublingual glands, accounting for approximately 12% and 15% of tumors of these glands, respectively. The tumor is partially encapsulated and is rarely associated with cyst formation and hemorrhage. Histopathologically, adenoid cystic carcinoma is classified into the tubular (low grade), cribriform (intermediate grade), or solid (high grade) type; occasionally, some of these types are occasionally seen in a single tumor. This tumor has a tendency to spread via the nerves. Perineural invasion is a hallmark of this tumor; it may be present distally in a nerve, and the proximal segment may appear normal (perineural skip extension). Acinic cell carcinoma represents approximately 9% of all malignant tumors of the parotid gland. The parotid gland is the most common site of origin of this tumor, and bilateral involvement is observed in 3% of patients. It is a solid mass or is partially cystic. Histologically, the acinic cell carcinoma recapitulates serous cells with bubbly basophilic cytoplasm containing zymogen granules. Salivary duct carcinoma is an extremely aggressive tumor, and histologically displays squamoid appearance, reminiscent of mucoepidermoid carcinoma or squamous cell carcinoma. Perineural spread is also common in this malignant neoplasm, and lymph node metastasis is not uncommon.

Clinical Presentation Benign tumors are usually slow growing, movable, painless masses. Malignant tumors present as firm, fixed and painless masses; in advanced stage, symptoms can include pain, otorrhoea, paraesthesia, facial nerve palsy, dysphagia, trismus, and cervical lymph node involvement. Pain or itching over the course of the facial nerve suggests a malignant parotid tumor. In addition, the combination of cranial nerve VII and V3 neuropathy suggests adenoid cystic carcinoma of the deep lobe of the parotid gland. The American Joint Committee on Cancer proposed the clinical staging of the malignant tumors of the major salivary glands chiefly based on the tumor size. Minor salivary gland tumors are staged as those arising from the corresponding anatomical sites. The anatomical region of a malignant salivary gland tumor is an important prognostic factor for local control and overall survival (3). The T-stage is an independent prognostic factor for local control. The histological type is an independent prognostic factor for distant metastasis. The T- and N-stages, and skin involvement are prognostic factors for distant metastasis and overall survival. The N-stage and facial paralysis are prognostic factors for local control. Metastasis to the regional lymph nodes is relatively rare in malignant salivary gland tumors, while it occurs in 40% of squamous cell carcinomas. Distant metastasis occurs in approximately 20% of malignant salivary gland tumors (most frequently seen in adenoid cystic carcinomas).

1637

Imaging Preoperative prediction of the malignancy or benignancy of salivary gland tumors is clinically very important, because this information strongly influences the surgical plan. In general, MR imaging is the first choice for the imaging of the salivary gland tumors. CT and sonography are alternatives to MR imaging.

Magnetic Resonance Imaging Although conventional T1-weighted MR imaging with or without gadolinium enhancement and T2-weighted MR imaging techniques have been used for the diagnosis of salivary gland tumors, many investigators have reported difficulty in the differentiation of these tumors, particularly between benign and low-grade malignant tumors. ▶High-resolution MR imaging of the salivary gland tumors using a small surface coil well delineates the detailed structures of the tumors and can provide important clues regarding their histologies and for prediction of their benign or malignant nature (Fig. 1). Pleomorphic adenomas are usually well-defined, homogeneously intermediate to hypointense on T1-weighted images, and heterogeneously hyperintense on fat-suppressed T2-weighted images (Fig. 1a). Warthin’s tumors exhibit various features ranging from hypointense to hyperintense on T1-weighted and fat-suppressed T2weighted images (Fig. 1b). Cyst formation is observed in one-third of Warthin’s tumors, resulting in hyperintense foci on fat-suppressed T2-weighted images. Hyperintensities in the cystic areas on T1-weighted images indicate the presence of proteinaceous fluids, colloidal materials, or hemorrhage. In general, high-grade malignancies have irregular, infiltrating margins (Fig. 2). Compared with low-grade malignancies and benign lesions, high-grade malignant tumors are associated with high cellularity, and thus have lower signal intensities on fat-suppressed T2-weighted images. Differential diagnosis of benign and malignant salivary gland tumors based on the descriptive characterization on conventional T1-weighted and fatsuppressed T2-weighted imaging may be further substantiated by using quantitative parameters obtained by functional MR imaging, for example, ▶apparent diffusion coefficient (ADC) maps obtained by ▶diffusionweighted MR imaging and ▶time-signal intensity curves (TICs) obtained by ▶dynamic contrast-enhanced MR imaging. The increased nuclear-to-cytoplasmic ratio and hypercellularity, both of which are frequently observed in highly malignant tumors, reduce the extracellular matrix and the diffusion space of water protons in the extracellular and

S

1638


Salivary Glands, Neoplasms. Figure 1 High-resolution MR imaging of benign salivary gland tumors using a 47-mm microscopy coil. Axial T1-weighted images (left panels), and fat-suppressed T2-weighted images (center panels), and gadolinium-enhanced T1-weighted images (right panels) show pleomorphic adenoma (a) and Warthin’s tumor (b) of parotid glands. High-resolution imaging reveals detailed architectures of salivary gland tumors. Note that multiple occurrences of Warthin’s tumor as in (b).

Salivary Glands, Neoplasms. Figure 2 MR imaging of malignant salivary gland. Axial T1-weighted images (left panels), fat-suppressed T2-weighted images (center panels), and gadolinium-enhanced, T1-weighted images (right panels) of adenoid cystic carcinoma (a) and carcinoma ex pleomorphic adenoma (b) of parotid and sublingual glands, respectively.


1639

intracellular dimensions, along with a resultant decrease in the ADCs. Therefore, the ADCs of malignant solid lesions are lower than that of benign tumors, except Warthin’s tumors. The ADCs of solid lesions of Warthin’s tumors are low and mimicked values of malignant tumors. A proliferation of the epithelial component and intense lymphoid accumulation in the stroma may have decreased the ADCs of the Warthin’s tumors. Based on the time of peak enhancement (T-peak) and the washout ratio (WR), salivary gland tumors could be categorized into four types (4). In type A, the TIC has a long T-peak, and is characteristic of pleomorphic adenomas. In type B, the TIC has a short T-peak and a high WR and is characteristic of Warthin’s tumor. In type C, the TIC has a short T-peak and a low WR and is characteristic of cancers. In type D, the TIC is flat and is suggestive of cystic or necrotic area. The combined use of the ADC maps and TIC criteria could greatly improve the diagnostic ability of MR imaging.

Sonography The sonographic examination of the salivary glands is usually performed using a high-frequency (10–14 MHz) transducer. This yields high-resolution images of superficial organs such as the parotid and submandibular glands (Fig. 3). However, sonographic images of the deeper parts of the glands may be of poor quality. This is the chief reason that sonography is not used as frequently as CT or MR imaging in examining patients with possible deep extensions of salivary gland tumors. Differentiation of benign and malignant salivary gland tumors on the basis of descriptive studies on gray-scale images is difficult. On the other hand, quantitative sonographic analysis could be employed for effective diagnosis of benign and malignant tumors of the parotid gland; the combined use of the sonographic criteria (the mean gray level and the standard deviation of the echo levels) greatly improved the diagnostic accuracy when compared with that of qualitative diagnosis (5). It is not clear which differences in tumor tissues may result in the characteristic differences in internal architectures of sonographic images of the salivary gland tumors. We often observe cystic areas, connective tissues, hyalines, and necrotic and keratinized materials in tumor tissues. Probably, these microscopic changes may result in the observed differences in echogenicity.

Salivary Glands, Neoplasms. Figure 3 Sonographic images of benign and malignant parotid gland tumors. Sonography shows well-defined tumor margins in pleomorphic adenoma (a) and Warthin’s tumor (b), while ill-defined margin in adenoid cystic carcinoma (c). Tumor echogenicity is homogeneous in pleomorphic adenoma (a), while heterogeneous in Warthin’s tumor (b) and adenoid cystic carcinoma (c).

Nuclear Medicine Nuclear imaging is most frequently employed for examination of salivary gland tumors such as Warthin’s tumors and oncocytomas, both of which highly accumulate technetium (Tc) 99m. Positron emission tomography

(PET) studies using 2-[F-18]fluoro-2-deoxy-D-glucose (FDG) are capable of demonstrating high-grade malignant tumors with increased metabolic activity. However, significant overlaps occur with benign tumors such as Warthin’s tumors.

S

1640

Salt-Losing Crisis

Diagnosis In addition to the assessment of the benign or malignant nature of salivary gland tumors, the imaging should be mainly focused toward evaluating the following points: (1) whether a tumor is confined to the parenchyma of the gland or extends outside the gland into the upper neck and skull base; (2) whether a tumor involves the facial nerve; although the precise identification is difficult even on high-resolution MR imaging, the tumor location relative to the main trunk of the facial nerve (pes anserinus) may be identified by using anatomical landmarks such as the “tragal cartilaginous pointer” (the facial nerve trunk runs 1 cm caudal and 1 cm medial to the “pointer”), the stylomastoid process (the nerve runs lateral to the process and in the fat tissues), and the vascular plain (which lies lateral to the superficial temporal, retromandibular, and anterior facial veins); (3) whether the tumor has smooth or infiltrative margins; (4) whether lymph node metastasis is absent or present; and (5) whether multiple tumors are present; multiple occurrence is observed particularly in Warthin’s tumors, pleomorphic adenomas, oncocytomas, acinic cell carcinomas, basal cell adenomas, and lymphomas.

Bibliography 1.

2.

3.

4.

5.

Barnes L, Eveson JW, Reichart P et al (2005) World health organization classification of tumors. Pathology and genetics of head and neck tumors. In: Tumours of the Salivary Glands. International Agency for Research on Cancer, Lyon, pp 209–274 American Joint Committee on Cancer (1988) Major salivary glands. AJCC Cancer Staging Handbook: AJCC Cancer Staging Manual. 5th edn. Lippincott William & Wilkins, Philadelphia, pp 57–64 Terhaard CHJ, Lubsen H, Van der Tweel I et al (2004) Salivary gland carcinoma: independent prognostic factors for locoregional control, distant metastases, and overall survival: results of the dutch head and neck oncology cooperative group. Head & Neck 26:681–693 Yabuuchi H, Fukuya T, Tajima T et al (2002) Salivary gland tumors: diagnostic value of gadolinium-enhanced dynamic MR imaging with histologic correlation. Radiology 226:345–354 Yonetsu K, Ohki M, Kumazawa S et al (2004) Parotid tumors: differentiation of benign and malignant tumors with quantitative sonographic analyses. Ultrasound Med Biol 30:567–574

Salt-Losing Crisis A deficiency of aldosterone that results in hyponatremia and hyperkalemia, leading to dehydration, hypotension, circulatory collapse, and, when severe, death. ▶Adrenogenital Syndrome

SAPHO G M L INGG , C S CHORN Department of Radiology, Sana Rheumazentrum Rheinland Pfalz, Bad Kreuznach, Germany [email protected]

Synonyms Acquired hyperostosis syndrome; Pustulotic arthroosteitis; Sternoclavicular hyperostosis

Clinical Presentation SAPHO is an association of skin and bone disease. The skin manifestation is either acne conglobata or fulminans or, more typically, psoriasis pustulosa palmaris and plantaris. Eighty percent of patients present with anterior chest wall affection, with relapsing and migrating symmetrical or asymmetrical pain and a tough swelling around the sternoclavicular joints that is often aggravated by cold exposure. There is no gender predilection. Children exhibit CRMO, whereas adults show multiforme osteitis such as sternoclavicular. The location of the osteitic focus may be quite variable, causing a variable clinical presentation. The sacroiliacal and vertebral affection is facultatively associated.

Imaging In anterior chest wall involvement, plain X-rays of the sternoclavicular joints show a varying degree of enostal and periostal sclerosis of the sternum and medial clavicular ends. Sternoclavicular arthritis with mild destruction may occur. Ligament and costal cartilage ossification is typical. In late stages, broad ossification bands from the sternum to the clavicle and upper rips obscure the normal anatomy. In early case, the plain films may show no abnormality. Sternoclavicular CT is employed. Sclerosis, bone formation, and enostal and periosteal and extrathoracal and intrathoracal (retrosternal fibroostitis) as well as sternoclavicular arthritis are seen with more conspicuity. MR imaging is not recommended because the sclerotic foci are a predominant finding and are best visualized by means of CT. MRI is, however, sensitive to the inflammatory process. Sacroiliitis is a typical feature, mainly indistinguishable from ankylosing spondylitis, but sometimes with atypical asymmetrical marked sclerosis or destruction.

Sarcoidosis, Musculoskeletal System

The affection of the spine may be similar to ankylosing spondylitis, but it presents with certain atypical features in a number of cases, such as marked vertebral sclerosis, ▶giant shining corners, and ▶bamboo-cane fragments. The latter refers to an oligosegmental sclerosing, but in these few segments, completely syndesmophytic ankylosed spine is in contrast to the otherwise normal radiologic appearance. In some cases the osteoproliferative features are heavily pronounced and the aspect is more like an atypical DISH (atypical because of, for instance, destructive and sclerotic details or ankylosis of the appendicular joints). Osteitic foci in other locations are erosive or nonerosive, multiforme, and expansile. Severe differential diagnostic problems can arise because of the resemblance to bacterial spondylitis/osteomyelitis, osteoblastic metastases, and osteogenic tumors.

Nuclear Medicine Bone scans show foci of increased osteoblastic activity (hot spots), especially in the sternoclavicular region. As a rule, in degenerative disease, which also causes sternoclavicular hot spots, the tracer accumulation is not intense enough to be seen from the back, and costal cartilage ossification does not expose tracer accumulation at all. In contrast, in SAPHO the costal cartilage ossification is an inflammatory-promoted process and does accumulate markedly. Bone scans are very sensitive but unspecific and can show foci in any location and direct the diagnostic work-up to these areas.

1641

(sternoclavicular arthritis, sternal [clavicular] sclerosing hypertrophic osteitis with marked bony apposition, ligament ossification, osteoproliferation) and skin manifestation (acne fulminans or psoriasis pustulosa). Frequently, it is associated with spondyloarthropathy; then, sacroiliitis and oligosegmental sclerosing or ankylosing vertebral involvement are typical. ▶Fractures, Stress ▶Spondyloarthropathies, Seronegative

Sarcoid Arthropathy There are three patterns of sarcoidosis joint involvement– Lo¨fgren’s syndrome, and early and late patterns of sarcoidosis arthropathy. Lo¨fgren’s syndrome is characterized by arthralgias, erythema nodosum and bilateral hilar lymphadenopathy. The early sarcoidal arthropathy pattern is manifested by polyarticular involvement of the distal extremities. The late manifestation of sarcoid arthropathy (6 months or more after sarcoidosis diagnosis) is a granulomatous synovitis. The diagnosis of sarcoidosis arthropathy is usually based on clinical presentation rather than imaging. ▶Sarcoidosis, Musculoskeletal System


Bibliography 1.

2.

3. 4.

Dougados M, van der Linden S, Juhlin R et al (1991) The European Spondyloarthropathy Study Group: preliminary criteria for the classification of spondyloarthropathy. Arthritis Rheum 34:1218–1227 van der Linden SM, Valkenburg HA, Cats A (1984) Evaluation of diagnostic criteria for ankylosing spondylitis: a proposal for modification of the New York criteria. Arthritis Rheum 27:361–368 Taylor WJ (2005) Understanding psoriatic arthritis. Hosp Med 66:163–167 Dihlmann W (1993) Acquired hyperostosis syndrome (so-called pustulotic arthro-osteitis): review including 73 self-observed cases. Wien klin Wochenschr 105:127–138

S ANDRA M OORE NYU Medical Center, NY [email protected]

S Synonyms Nonnecrotizing granulomatous disease

Definition and Pathology/Histopathology

SAPHO SAPHO syndrome (synonyms: acquired hyperostosis syndrome, pustulotic arthroosteitis, sternoclavicular hyperostosis) is a varied rheumatic symptom complex mainly with involvement of the anterior chest wall

▶Sarcoidosis is an inflammatory disorder of unclear etiology, characterized by the presence of noncaseating ▶granulomas in the tissues. This is usually confirmed by fiber optic bronchoscopy, biopsy of lung or lymph node, a positive Kveim-Siltzbach test, or a combination of these methods, revealing nonnecrotizing epithelioid granulomas in the absence of other granulomagenic agents.

1642



Sarcoidal Muscle Lesions

Sarcoidosis most commonly affects the pulmonary and lymphatic systems, but may involve the muscles, joints, and bones, with osteoarticular complaints commonplace. These include muscle weakness, generalized muscle pain, soft tissue swelling and/or multiarticular stiffness, and pain. Lesions detected at imaging in sarcoidosis patients maybe symptomatic or asymptomatic.

Sarcoidal muscle involvement includes myopathy/atrophy, myositis, and sarcoidal muscle nodules (3). Sarcoid-related myopathy presents clinically with elevated serum creatine kinase and aldolase, symmetric proximal weakness, and electromyographic evidence of myopathy. On MRI, there is fatty replacement of muscle. Sarcoidal myopathy cannot be differentiated from atrophy secondary to corticosteroid treatment on the basis of MR appearance, however MRI is useful to delineate the extent and severity of involvement and determine optimal muscle biopsy sites. Sarcoidal muscle nodules are uncommonly observed, and present as tender palpable masses. On MRI these show a central low signal and rim enhancement, due to fibrosis, which has been described as a “dark star” (Fig. 1).

Imaging Neither skeletal survey nor conventional bone scan reliably detects sarcoidal bone lesions, except in the small bones of the hands and feet (1). The radiographic appearance of lacy osteolysis in the bones of the hands is familiar to radiologists, but there are myriad musculoskeletal manifestations, which may be revealed in advanced imaging (specifically MRI), including subcutaneous infiltration, various muscle lesions, sarcoidal arthropathy, and large and axial bone lesions (2).

Sarcoidosis Soft Tissue Masses Sarcoidosis soft tissue lesions are uncommon, and present as palpable mass or swelling. These may be well defined or indistinctly marginated, with nonspecific signal characteristics—low on T1 weighted sequences, high on fluid sensitive sequences, and enhancing. These can be detected in the subcutaneous, inter- or intramuscular spaces. Differential diagnosis includes cellulitis (if subcutaneous and reticulated), or an inflammatory versus neoplastic mass. Biopsy is usually required, as MR imaging features are not pathognomonic.

Sarcoidal Arthropathies Three patterns of sarcoidosis joint disease may be encountered—Lo¨fgren’s syndrome, and early and late patterns of sarcoidosis arthropathy. Lo¨fgren’s syndrome is characterized by arthralgias, erythema nodosum, and bilateral hilar lymphadenopathy. The early sarcoidal arthropathy pattern (presenting during the first 6 months of sarcoidal symptoms) is manifested by polyarticular involvement of the distal extremities—ankles, knees, PIP joints, wrists, and elbows. Radiographs are usually negative, or show osteoporosis and soft tissue swelling, with rheumatoid arthritis as the principal radiographic differential diagnosis. A later manifestation of ▶sarcoid arthropathy (6 months or more after the diagnosis of sarcoidosis) usually involves 2–3 joints including the knees, ankles, proximal interphalangeal joints, and

Sarcoidosis, Musculoskeletal System. Figure 1 Fifty-one year-old woman with sarcoidosis who presented with tender nodules along both calves. T2 weighted axial image demonstrates edematous nodules within the right posterior and left anterior compartment muscles. After contrast administration, these enhanced peripherally (not shown), with “dark star” appearance characteristic of nodular muscle sarcoid lesions. Noncaseating granulomas were found at biopsy.


occasionally the wrists or shoulders. Dactylitis may occur. These patients usually have granulomatous synovitis, and the disease course is transient or relapsing. Radiographic changes consist of subchondral cystic change. The joint abnormalities detected on radiographs and MRI in sarcoidosis patients reflect nonspecific arthropathy of indeterminate origin, which includes the common arthritides. Whether sarcoidosis contributes to the imaging findings is speculative, as synovial biopsy is not routinely obtained, and the diagnosis is usually based on clinical rather than imaging parameters. When MRI is performed, joints affected by sarcoidosis arthropathy may show lesions that are occult on plain films. Tenosynovitis, tendonitis, bursitis, synovitis, and prominent subchondral cysts can be demonstrated on MRI, but are nonspecific findings. Large subchondral cysts in the absence of osteophytosis may be seen in patients younger than those usually presenting with osteoarthritis with geodes.

Bone Sarcoidosis Osseous involvement is estimated to occur in about 5% of sarcoid patients, although it is important to consider that this percentage is derived from radiographic studies and may underestimate disease occult on radiographs (4). The diagnosis of sarcoidosis has usually been established by clinical criteria prior to the detection of suspicious bone lesions. Bone lesions may be biopsied to confirm sarcoidosis if the diagnosis is in doubt, or to exclude neoplasm. The radiographic appearance of lacy osteolysis in the bones of the distal extremities is considered virtually pathognomonic of sarcoidal involvement. The alignment deformities in the hands and feet are often due to periarticular pathologic fractures rather than joint abnormalities. Large bone and spinal sarcoidal lesions, often occult on radiographs, maybe detected on MRI either incidentally, or in evaluation of musculoskeletal symptoms (5). On MRI, osseous sarcoidosis demonstrates a range of morphologies, from indistinctly marginated marrow infiltration, to discrete “cannon ball” and “starry sky” type lesions (Figs 2 and 3). The signal characteristics—low on T1 weighted images, increased signal intensity on water sensitive images, enhancing with contrast—are nonspecific. The principal differential considerations include bone metastases, lymphoma, myeloma, or rarely, disseminated nonmalignant entities such as fibrous dysplasia, disseminated fungal disease or tuberculosis, or serous atrophy. In some cases, we observed that the sarcoidosis lesions resolved on follow-up studies, with “ghosts” of the prior lesions showing central low signal correlating with fat and/or fibrosis (2). More study is needed to understand the natural history of these lesions on MRI. Our preliminary experience with

1643

Sarcoidosis, Musculoskeletal System. Figure 2 Forty-eight year-old man with sarcoidosis and hip pain. Radiographs were negative. T1 weighted coronal image through the posterior of the pelvis demonstrates multiple rounded (“cannon ball”) low signal foci. Six months later, these showed resolution on MR images.

S

Sarcoidosis, Musculoskeletal System. Figure 3 Forty-six year-old man with shoulder pain. T2 fat saturated oblique coronal image of the shoulder demonstrates multiple round and irregular intramedullary high signal foci. These were detected throughout the shoulder girdle. Subsequent whole body MRI (not shown) revealed innumerable similar foci in bilateral upper and lower extremities, spine, and pelvis. Biopsy of the right humerus demonstrated noncaseating granulomas.

1644

Sarcoma

opposed phase imaging suggests that signal intensity changes on out-of-phase images is variable (i.e., may or may not drop significantly), which maybe related to the chronicity and/or activity of these lesions.

Sarcoma ▶Neoplasms, Soft Tissue, Malignant

Nuclear Medicine Sarcoidosis lesions show variable uptake on bone scan (1). Sarcoidosis bone lesions also may present a false positive for metastases on PET scanning. Whole body MR scanning may be superior to PET imaging for the detection of diffuse sarcoidal bony lesions.

Conclusion There are numerous musculoskeletal manifestations of sarcoidosis, many of which require advanced imaging for detection. The MRI findings of musculoskeletal sarcoidosis lesions may reveal a greater granulomatous burden than had been previously appraised. When musculoskeletal manifestations are detected, MRI is useful to follow response to treatment, and can guide biopsy in patients with unproven sarcoidosis. With routine MRI technique, with the exception of sarcoidal muscle nodules, no pathognomonic imaging findings allow differentiation from other pathologies, including neoplasm. Therefore correlation with clinical and laboratory findings is essential for correct diagnosis. Radiologists should consider sarcoidosis in the differential diagnosis of musculoskeletal lesions detected at MRI in the appropriate setting (i.e., clinically proven sarcoidosis), and should be alerted that bony sarcoidosis lesions encountered on MRI can resemble metastatic lesions. In some patients in whom sarcoidosis is not proven or a second comorbidity is suspected, or to confirm sarcoidal histology, biopsy of the musculoskeletal lesion would be indicated.

Bibliography 1.

2. 3. 4. 5.

Milman N, Lund JO, Graudal N et al (2000) Diagnostic value of routine radioisotope bone scanning in a series of 63 patients with pulmonary sarcoidosis. Sarcoidosis Vasc Diffuse Lung Dis 17:67–70 Moore SL, Teirstein AS (2003) Musculoskeletal sarcoidosis: spectrum of appearances at MR imaging. Radiographics 23:1389–1399 Otake S, Ishigaki T (2001) Muscular sarcoidosis. Semin Musculoskelet Radiol 5(2):167–170 James DG, Timmis B, Barter S et al (1989) Radiology of sarcoidosis. Sarcoidosis 6:7–14 Moore SL, Teirstein A, Golimbu C (2005) MRI of sarcoidosis patients with musculoskeletal symptoms. Am J Roentgenol 185 (1):154–159

SBO Small bowel obstruction. ▶Occlusion, Bowel in Childhood

Scalloping An abnormality of the vertebral body characterized by concavity of the posterior wall. ▶Neurofibromatosis, Musculoskeletal Manifestations

Scapholunate Advanced Collapse The prognostically poor late stage of scapholunate instability with secondary degenerative osteoarthritis. ▶Rheumatoid Arthritis

SCAVF ▶Spinal Cord Arteriovenous Fistula (SCAVF)

SCAVM ▶Spinal Cord Arteriovenous Malformation (SCAVM)

Schimmelbusch’s Disease ▶Fibrocystic Disease, Breast

Scintigraphy

Schwannoma A benign neurogenic soft tissue tumor of the peripheral nerve sheath, also designated as neurilemmoma. ▶Neurofibromatosis, Musculoskeletal Manifestations ▶Tumors, Spine, Intradural, Extramedullary

Sciatica ▶Spinal Nerve Roots, Clinical Syndromes ▶Radicular Syndrome of the Spine, Conservative Therapy for

Scintigram A Scintigram is an image which assigns the data measured by scintigraphic means to spatial coordinates and, if applicable, to time. ▶Scintigraphy

Scintigraphy L ILLI G EWORSKI Charite´-Universita¨tsmedizin, Berlin, Germany [email protected]

1645

In static scintigraphy, the spatial distribution of activity in an organ is measured under the assumption that there is no significant change of activity distribution during the time interval of measurement (see Fig. 1). If appropriate, the organ under study can be measured from different views, e.g., ventral, dorsal, lateral, or oblique. In dynamic scintigraphy, attention is focused on the temporal change of activity distribution of the radiopharmaceutical used. The organ under study is imaged in a temporal sequence of frames maintaining the viewing angle. Usually, from this sequence and one or more regions of interest (ROI) within the images, time–activity curves are generated for each ROI. The subsequent analysis of these curves (e.g., maximum rise time, decay time) results in the determination of numerous physiological parameters (e.g., organ clearance, perfusion). Whole-body scintigraphy is the obtainment of scintigraphic information for the whole human body during one acquisition. As the dimensions of the human body are much larger than the axial field of view of a gamma camera, the detector head(s) and the patient must be moved relative to each other in the direction of the body’s long axis. In addition, the data acquired must be combined by appropriate means to form the final image. Gated scintigraphy is a multiple-frame imaging, controlled by external physiological signals of periodic cycles, e.g., triggered by ECG or respiratory signals, either to overcome the image blur caused by organ movement of or to collect information on organ movement. The cycle is divided into a number of phase intervals, and driven by the external trigger the scintigraphic information is sorted into the corresponding interval. Mostly, data are summed over a multitude of cycles. The final outcome is a series of images associated with the corresponding phase intervals.

Definition Diagnostic procedures in nuclear medicine are based on the external detection of the g-radiation emitted by incorporated radionuclides. The spatial distribution of these radionuclides in the (three-dimensional) object (patient) is recorded as a two-dimensional ▶projection image by means of a ▶gamma camera. This projection method is called planar ▶scintigraphy, and the image itself assigns measured data to spatial coordinates and, if applicable, to time information. It is referred to as a ▶scintigram.

S

Characteristics Types of Scintigraphy Planar scintigraphy can be performed as single-frame (static), multiple-frame (dynamic or gated), or whole-body imaging.

Scintigraphy. Figure 1

Scintigram of the thyroid gland.

1646

Scintigraphy

Imaging The most often used imaging device in nuclear medicine is the gamma camera with a detector head utilizing the Anger principle (2) (see Fig. 2). The actual detector is formed by a large-scale single crystal scintillator, which is coupled via a light guide to a number of photomultiplier tubes (PMTs). In addition to the detector, the detector head is composed of a ▶collimator, detector shield, and positional electronics. When a g-quant (photon) is absorbed in the detector crystal, the absorbed energy is emitted partly as nondirectional scintillation light. These flashes of light, whose integral is proportional to the absorbed energy, are distributed via a light guide, which is designed to optimize the spatial spread of the scintillation light, to the photocathode of a PMT. To prevent light from exiting the crystal via surfaces not facing the PMTs, these crystal boundaries are coated by reflective materials such as magnesium or aluminum oxide. The photocathode converts the scintillation light to electrons by means of the outer photo effect. These electrons are accelerated by an electrostatic field toward the dynodes (usually 10–12), thereby generating new electrons when crashing on a dynode. When finally reaching the anode of the PMT, each electron emitted by the photocathode has been multiplied by a factor of approximately 106. This avalanche of electrons generates an electrical pulse at the anode, which after additional shaping amplification and integration has to pass pulse height analysis until finally being processed and registered by the positioning electronics. Looking at just one PMT and one detected event, the amount of light and, hence, the magnitude of the measured signal are dependent on the absorbed energy and on the location of the absorption event relative the wore of the light to the position of the PMT (the shorter the distance the more the light is seen). Now, taking into account all PMTs coupled to the crystal, the weighted sum of all PMT output signals (after correcting for the absorbed energy) yields the detected position of the scintillation, whereas the unweighted sum results in an estimate of the absorbed energy, which is independent of position (1, 3). The accuracy of positioning is mostly dependent on the efficiency by which the light is collected, i.e., on the light yield of the detector material (conversion from g-energy to

Scintigraphy. Figure 2

light photons) and on the proper design of the detector. To a lesser degree it also depends on the number of PMTs per area (usually ranging from 37 to 105 per detector head). For a scintillation detector, there is a linear relationship between absorbed energy and the amount of light emitted and hence output pulse amplitude. Thus, comparing pulse amplitude with appropriate discriminator levels, the energy range of photons to be included into the detection process can be selected by pulse height analysis. The emission of g-rays from the patient is nondirectional (as is the emission of normal light). In order to generate an image, directional information must be added. This is done by inserting a device called a collimator between the object and detector crystal, which can be penetrated only by g-rays striking the collimator out of well-defined directions. The principle of mechanical collimation is absorption: a thick plate made from highly absorbing material (usually lead or tungsten) and containing thousands of well-defined bores can be penetrated only by g-rays that pass through the long bores. The most often used geometry is a parallel hole collimator, but other geometries exist, e.g., converging, diverging, and conical collimators. Other construction parameters are energy range (thickness of septa and plate), resolution (hole diameter and length), and sensitivity, which is inversely related to resolution (and energy). The detector head itself is attached to a gantry designed to facilitate detector positioning. Today, the gamma camera is interfaced to computers for data acquisition, processing, and storage. To summarize, the four physical steps of gamma camera imaging are 1. Preselection of patient volume, in which the spatial distribution of activity is to be measured, by proper detector positioning and collimation. 2. Absorption of g-quants emitted under directions preselected by collimation in the detector crystal and energy transfer by scintillation. 3. Transformation of the scintillation light to an electrical signal (PMT). 4. Determination of the absorbed energy and the position of the scintillation, energy discrimination, and final registration and storage.

Detector head of an Anger camera (principle).

Sclerosing Adenosis, Breast

The scintigraphic information is given by the sum of all events, which are registered along a ▶projection beam by the detector head. This projection beam is defined as the smallest possible volume that can be resolved by the detector head when detecting the radiation emitted by the object. Its shape is limited by the detector head’s spatial resolution in all three dimensions. Mostly, it is a long, thin, diverging cone, as a gamma camera has no spatial resolution in the direction perpendicular to the detector, i.e., the direction of collimator bores. Supposing that there is no absorption or scattering of g-radiation in the object and that sensitivity and spatial resolution are constant, the ray sum is proportional to the linear sum of activity concentrations along the projection beam. No information is contained about the depth of the source (its distance to the detector). In order to collect information on the third dimension of the object, it must be looked at from a multitude of different projection angles. In addition, the data measured have to be reconstructed by appropriate mathematical algorithms— the principle of tomography. Tomography requires either detectors that rotate around the patient or a detector ring surrounding the patient. The most prominent tomographic methods in nuclear medicine are single photon emission computed tomography (SPECT) and positron emission tomography (PET).

1647

Scleroelastotic Scar ▶Radial Scar, Breast

Sclerosing Adenosis, Breast LUIS P INA Department of Radiology, Clıńica Universitaria de Navarra, Navarra, Spain [email protected]

Synonyms Adenosis tumor; Nodular adenosis

Definition Sclerosing ▶adenosis is a benign proliferative lobulocentric lesion, consisting of stromal sclerosis and proliferative adenosis.

References 1. 2. 3.

Anger HO (1958) Scintillation camera. Rev Sci Instr 29:27–33 Anger HO (1966) Survey of radioisotope cameras. ISA Trans 5:311–334 Birks JB (1964) The Theory and Practice of Scintillation Counting. Pergamon Press, Oxford, England

Scintigraphy Scintigraphy is an imaging method to represent the spatial (and temporal) distribution of incorporated radio nuclides.

Pathology Sclerosing adenosis is composed of distorted epithelial, myoepithelial and sclerotic stromal elements arising in association with the terminal duct lobular unit (1). Sclerosing adenosis is part of the proliferative lesions without atypia, but it may be involved by atypical lobular hyperplasia, lobular carcinoma in situ, atypical ductal hyperplasia, or ductal carcinoma in situ. Perineural pseudoinvasion is seen in 2% of cases. Because of the distorted glandular pattern, sclerosing adenosis may be confused with tubular carcinoma or with radial scar (1, 2).

Scirrhous Carcinoma Clinical Presentation ▶Carcinoma, Ductal, Invasive

Sclerocystic Disease ▶Fibrocystic Disease, Breast

Sclerosing adenosis is usually an incidental finding in surgical specimens, although it may be detected on mammography (1, 2). Most cases are non-palpable, but occasionally may present clinically as a palpable mass (adenosis tumour) that can be confused with breast cancer. Sclerosing adenosis is also a cause of breast pain.

S

1648

Sclerosing Adenosis, Breast

Imaging Mammography There are not specific mammographic features to diagnose sclerosing adenosis. Round, punctate, or pleomorphic microcalcifications are usually the most frequent finding. Also well circumscribed to spiculated masses, ▶architectural distortions and asymmetric densities have been reported to be associated with sclerosing adenosis (3, 4) (Figs 1 and 2).

Ultrasound Focal acoustic shadowing without a mass and both welland ill-delimited hypoechoic masses have been reported (3, 4) (Fig. 3).

Sclerosing Adenosis, Breast. Figure 3 Ultrasonography reveals an irregular ill-delimited hypoechoic mass. Sclerosing adenosis.

Magnetic Resonance Areas of sclerosing adenosis may enhance after the injection of paramagnetic contrast, simulating breast cancer.

Nuclear Medicine Nuclear medicine techniques do not play any role in the diagnosis of sclerosing adenosis.

Diagnosis

Sclerosing Adenosis, Breast. Figure 1 Irregular spiculated mass, highly suspicious of carcinoma (mediolateral oblique view, detail). Biopsy revealed sclerosing adenosis.

The diagnosis of sclerosing adenosis is made by biopsy. Imaging techniques usually show suspicious findings, mimicking malignancy, so a biopsy is recommended. Due to the possible association with premalignant or even malignant lesions, a result of sclerosing adenosis in a core biopsy should be carefully evaluated. It may be acceptable for circumscribed masses and non-palpable indistinctly marginated masses. The result may be concordant also for vacuum-assisted biopsy of punctate, amorphous, or pleomorphic calcifications. However, the result is not acceptable for BI-RADS 5 lesions. In these cases, a surgical biopsy is recommended (5).

Bibliography 1.

2.

3. 4.

Sclerosing Adenosis, Breast. Figure 2 Architectural distortion with punctate calcifications. The lesion was palpable. Biopsy revealed sclerosing adenosis.

5.

Schnitt S, Connolly JL (2004) Pathology of benign breast disorders. In: Harris JR, Lippman ME, Morrow M (eds) et al Diseases of the Breast. Lippincott Williams & Wilkins, Philadelphia, pp 77–99 Hughes LE (2000) Sclerosing adenosis, radial scar and complex sclerosing lesions. In: Hughes LE, Mansel RE, Webster DJT (eds) Benign Disorders and Diseases of the Breast. Concepts and Clinical Management, Saunders, London, pp 137–142 Cardenõsa G (2001) Lobules. In: Cardenõsa G (ed) Breast Imaging Companion. Lippincott Williams & Wilkins, Philadelphia, pp 298–300 ¨ stun EE et al (2002) Sclerosing Gu¨nhan-Bilgen I, Memis A, U adenosis: mammographic and ultrasonographic findings with clinical and histopathological correlation. Eur J Radiol 44:232–238 Gill HK, Ioffe OB, Berg WA (2003) When Is a diagnosis of sclerosing adenosis acceptable at core biopsy? Radiology 228:50–57

Scoliosis

Sclerosing Papillary Lesion ▶Radial Scar, Breast

Sclerosing Papillary Proliferation ▶Radial Scar, Breast

Scoliosis K ARL LUDWIG , C ORINNA E.E. N IERHOFF Department of Diagnostic Radiology, Orthopedic University Hospital Heidelberg, Heidelberg, Germany [email protected]

Definition Scoliosis is defined as a fixed lateral curvature and torsion of the spine. According to the criteria of the Scoliosis Research Society, the diagnosis of structural ▶scoliosis should be reported only if the curve in the frontal plane, as viewed on an anteroposterior radiograph, is greater than 10˚.

Pathology/Histopathology Scoliosis represents a deformity of growth. The vertebral bodies grow slower on the concavity than on the convexity. This disturbance of growth of one or more vertebral bodies results in a rotation and lordosis of the spine. Eighty-five percent of scolioses are idiopathic, which means there is no known underlying pathology. Idiopathic scoliosis worsens mainly in the growth period. Generally, curvatures less than 30˚ will not progress after childhood. With greater curvatures after adolescence, the degree of scoliosis may progress by about 1˚ per year. Studies have shown that the incidence of idiopathic scoliosis is higher in relatives, which suggests an autosomal-dominant inheritance pattern with variable phenotypic expression (1).

1649

Clinical Presentation The incidence of idiopathic scoliosis with lateral curvatures of 10˚ or greater varies between 1 and 3% in the general population. The female:male ratio is 4:1. According to the onset of symptoms, idiopathic scoliosis can be classified into three groups: . Infantile: birth to age 3 years . Juvenile: 3–11 years . Adolescent An etiological classification of scoliosis differentiates between different entities: . Idiopathic . Neuropathic: cerebral palsy, syringomyelia, Charcot Marie-Tooth, polio, spinal muscle atrophy, myelomeningocele . Myopathic: muscular dystrophies . Congenital: diastematomyelia, neurofibromatosis, . Mesenchymal: Marfan syndrome, Ehlers–Danlos syndrome . Systematic diseases: achondroplasia, spondyloepiphysary dysplasia, mucopolysaccharidosis . Metabolic: rickets, juvenile osteoporosis, osteogenesis imperfecta . Radiation-associated: radiotherapy in childhood . Posttraumatic . Neoplastic . Inflammatory . Vertebral malformations (hemivertebrae, block vertebrae, butterfly and wedged vertebra, unsegmented bars) If scoliosis progresses in the thoracic region, it can create significant physical deformity. Respiratory problems can result in severe cases with curvatures approaching 90˚.

Imaging The role of radiology in scoliosis management is to confirm the diagnosis, identify any underlying pathology, and monitor the degree of curvature. Early diagnosis and treatment help prevent curve progression and deformity.

Plain Radiography Images of the erect entire spine in anteroposterior (AP) and lateral views are mandatory. In the AP projection, the iliac crest should be visible. Images are usually taken on a 36-in. cassette with a film focus distance of 3 m with the central beam in the middle of the thoracic spine. For distinguishing between reversible functional scoliosis and fixed structural scoliosis of babies and small

S

1650

Scoliosis

children (not yet able to walk), a supplementary stress picture is taken in the prone position (Bending test). The child is actively bent to the opposite side of the scoliosis while the image is taken. Is the scoliosis is fully compensated, a functional scoliosis is assumed. In structural scoliosis, muscle contractures or changes at the vertebral bodies, the disks, and the facet joints prevent compensation of the curvature.

Cobb Method A quantitative assessment of scoliosis is performed using the ▶Cobb angle. Besides some inter- and intraobserver variability, the Cobb method for scoliosis angle measurement is the most accepted standard technique worldwide. In an AP view, the two neutral vertebrae first have to be determined. Neutral vertebrae have parallel endplates with—in contrast to the other vertebrae—minimal or no change in shape and are maximally tilted to the horizontal line. Lines are drawn in extension of the uppermost and lowermost endplates of the neutral vertebra. Because these lines usually intersect outside the X-ray film, two perpendicular help-lines are constructed. At the point of intersection, the complementary angle is measured. This angle is the so-called Cobb-angle or angle of scoliosis (Fig. 1). When reporting the scoliosis angles, one should mention which measurement method was used and which neutral vertebrae were chosen. In follow-up imaging, the same vertebrae should be chosen for evaluating disease progression.

Evaluation of Rotation (Method of Nash and Moe) For evaluation of the scoliotic rotation of the vertebrae, the method of Nash and Moe (2) is widely used. There are five grades of rotation. Grade zero rotation means the pedicle outlines are symmetric and equidistant to the lateral borders of the vertebral body. Grades 1–4 illustrate the movement of the pedicle toward and over the midline (Fig. 2).

Determination of Skeletal Maturity (Risser’s Sign) A further goal of the radiographic examination is to determine skeletal maturity. Because curvature of mild scoliosis does not progress after adolescence, the skeletal age is a prognostic factor of the likelihood of curvature progression. Since the iliac crest is usually visible in the total

Scoliosis. Figure 1 Measurement of the scoliosis angle (Cobb 1948). Anteroposterior radiograph of a 22-year-old man with a single right thoracic curve. Horizontal lines are drawn parallel to the endplates of the neutral vertebrae at the end of the curve. Where perpendicular lines intersect, the angle of scoliosis is measured. In this case the angle of Cobb measures 32˚.

spine study, it can be conveniently used to determine skeletal maturity. The Risser classification (3) of the apophysis of the iliac crest enables the radiologist estimate the expected growth of the vertebral body and therefore estimate the progression tendency of the scoliosis.

Risser Classifies the Progression of Ossification of the Apophysis of the Os ileum into Six Groups 0: Apophysis cannot be seen 1: Apophysis begins to develop laterally and measures up to 25% of the iliac crest. 2: Apophysis covers up to half of the iliac crest 3: Apophysis reaches up to 75% of the iliac crest 4: Complete excursion of the iliac apophysis 5: Complete fusion of the apophysis with the iliac crest

Scoliosis

1651

Scoliosis. Figure 2 The method of Nash and Moe (2) measures the degree of rotation from grade zero (no rotation) to grade 5 (pedicle outline wanders over the midline).

Cross-Sectional Imaging Besides conventional imaging measurement, techniques focusing on MRI and CT imaging are being used more and more. They have the drawback that images are acquired with the patient in the prone position, so the influence of the physiological vertical stress of the standing body on the curvature of scoliosis cannot be evaluated. Multiplanar CT imaging especially enables visualization of complex deformities and a three-dimensional understanding of the scoliotic spine (Fig. 3).

Nuclear Medicine The role of nuclear medicine in routine evaluation of scoliosis is limited. An isotope bone scan (99mTcmethylene diphosphonate) can be helpful when evaluating a child with focal alterations of the bone, such as infection or neoplasia.

Alternative Imaging Methods To reduce radiation exposure, alternative imaging techniques have been investigated for documentation and to control the course of disease. Light optical methods with the possibility of three-dimensional surface rendering can quantify the cosmetic deformity of the thorax.

Scoliosis. Figure 3 Coronal multiplanar reconstruction of a helical computed tomography scan showing a single bended lumbar scoliosis with the concavity on the right side on the basis of a complex spinal deformity, including a hemivertebra L3.

Before surgery lateral-bending films are sometimes taken to differentiate structural fixed curves from nonstructural flexible curves. Because scoliosis is often associated with other congenital deformities, skeletal, cardiac, and urogenital deformaties must be excluded. Last not least, one must be aware of the radiation exposure caused by scoliosis follow-up. Nash et al estimated that 22 radiographic examinations are performed in the course of scoliosis management (4). Statistically, an increased risk of developing breast cancer in girls with scoliosis has been reported (5).

Diagnosis A complete evaluation should include a physical examination and imaging of the total spine. Plain radiography remains the initial imaging method of choice and in certain cases is followed by MRI or CT imaging. In the radiologic report, the form of the deformity (scoliosis, kyphosis, lordosis, or S-shaped), the side and the degree of scoliosis should be described. The rotational component of scoliosis is described by the Nash and Moe scale, and the skeletal age should be mentioned, for instance using the Risser classification.


5.

Wynne-Davies R (1968) Familial (idiopathic) scoliosis. A family survey. J Bone Joint Surg Br 50:24–30 Nash CL, Moe JH (1969) A study of the vertebral rotation. J Bone Joint Surg Am 51:223–229 Risser JC (1958) The iliac apophysis: an invaluable sign in the management of scoliosis. Clin Orthop 11:111 Nash CL, Gregg EC, Brown RH et al (1979) Risks of exposure to x-rays in patients undergoing long-term treatment for scoliosis. J Bone Joint Surg Am 61:371–374 Ron E (2003) Cancer risk from medical radiation. Health Phys 85(1):47–59

S

1652

Screening, Breast Cancer

Screening, Breast Cancer M ARGRIT R EICHEL Referenzzentrumsleiterin Mammographie-Screening Wiesbaden Michelsberg, Wiesbaden [email protected]

Synonyms Breast cancer screening

Definition Breast cancer is the most common cancer among women worldwide, with over one million new cases diagnosed in 2000. Although substantial improvements in survival have been recorded in Western countries since the late 1970s, breast cancer still causes the most deaths from cancer among women. Studies have shown that mass screening for breast cancer using mammography as the screening test (mammography screening), with or without clinical breast examination, has the potential to further decrease breast cancer mortality by detecting asymptomatic lesions which can be treated more effectively than symptomatic disease (1). In order to achieve these benefits, however, a number of conditions need to be fulfilled, which are discussed here.

Characteristics In 2002, an IARC Working Group on the Evaluation of Breast Cancer found sufficient evidence for the efficacy of mammography as the sole screening modality for breast cancer in women aged 50–69 years. The group found only limited or insufficient evidence for other age groups or screening modalities (1).

Target Population Mammography screening is generally offered to the asymptomatic 50–69-year-old female population in 2-year intervals. If the screening mammogram is normal, no further action will be taken and the woman will be invited to the next round of screening. Unlike diagnostic mammography, mammography screening should not be offered to patients presenting with signs or symptoms of breast cancer. A large proportion of women in a diagnostic breast service will present with benign abnormalities; work-up of benign lesions in a diagnostic service is

frequently necessary in order to reassure the patient and avoid unnecessary treatment. The opposite is the case in mammography screening, however. Screening clients are not informed about benign lesions in order to avoid unnecessary anxiety and unnecessary further diagnostic procedures in large numbers of women.

Screening Process Since the vast majority of women attending screening are disease free, considerable precautions must be taken to ensure that the benefits of screening are maximized while minimizing the potential harm. The physician responsible for the overall performance of the screening service and the attendant staff must constantly strive to optimize the individual components of the entire screening process, as outlined below (2, 3).

Information and Invitation A significant factor affecting the overall performance of a screening program is the proportion of women in the eligible ▶target population who attend. Organized, population-based mammography screening programs have been shown to achieve a high participation rate (over 70% attending in a given round) using invitation letters addressed to each individual woman in the eligible target population. The invitation letter should give a specific appointment (date, time, and place) and a telephone number in case the woman needs to change the appointment. The availability of population registry data can substantially simplify and improve the effectiveness of invitation procedures.

Taking the Screening Mammogram Screening mammograms (mediolateral oblique and craniocaudal views recommended) should be taken in dedicated units by specially trained staff. The physicotechnical quality must be monitored daily by unit staff and by a regional technical quality control office, which fulfills the standards of the European Guidelines (2). In some programs, screening radiographers may be trained to perform additional views on their own initiative. Such a policy can result in lower recall rates, although the overall rate of additional imaging may be higher.

Reading Screening Mammograms The classification of screening mammograms differs somewhat from diagnostic mammograms, because of

Screening, Breast Cancer

the lengthy (2 year) screening interval. In mammography screening, the reader does not have the option of recalling a woman in a shorter (e.g., 6 month) interval because that would generate inordinate anxiety. Thus, the reader must decide whether to recall a woman immediately for further assessment or in 2 years for the next round of screening. Screening mammograms should also be read in batches of several women. A batch should be of sufficient size to enable the reader to roughly appraise ▶sensitivity and ▶specificity and to judge whether repeated reading of the entire batch is necessary. Roller viewers should be used and reading should be performed in an undisturbed session with optimal viewing conditions (darkened room). Double reading with arbitration by a third very experienced reader is recommended. The first two readers should be specially trained and experienced in reading screening mammograms. The double readers should strive for optimum sensitivity. The third reader should read any mammograms that one or both double readers find suspicious, with the aim of optimizing specificity.

Assessment of Suspicious Screening Mammograms All women, the screening mammograms of whom are still deemed suspicious after arbitration, are recalled for assessment that initially consists of additional imaging and clinical breast examination. If suspicion of abnormality persists after additional imaging, biopsy should be recommended. In most cases (>99%) fine-needle aspiration, core cut, or vacuum biopsy can be used.

Preoperative Multidisciplinary Conference For each case in which suspicion of abnormality persists after imaging, the final decision on the result of screening should be reached in a multidisciplinary preoperative conference in which the assessment team (screening radiologist, pathologist, radiographer, and breast surgeon) consider the correlation between the histopathological assessment of the preoperative biopsy and the other assessment findings. A radiotherapist or radio-oncologist and breast care nurse should also be present so that the entire team can plan the operation/treatment, if breast cancer is diagnosed. The breast care nurse should also be present when a woman is informed of a diagnosis of breast cancer because the woman will have special communication needs, which the breast care nurse is specially trained to deal with.

1653

Postoperative Multidisciplinary Conference The result of each breast cancer operation and particularly the accuracy of the diagnosis and the appropriateness of the recommendations for management, made at the preoperative conference should be evaluated in a multidisciplinary postoperative conference attended by the same team present at the preoperative conference.

Quality Assurance At any given time, the vast majority of the target population is free of breast cancer and precursor lesions. Thus, even minimal harm resulting from anxiety or a small proportion of breast cancer cases detected in screening, which may not have become manifest or led to serious complications during a woman’s life can shift the delicate balance unfavorably between benefit and harm of screening. It is therefore imperative to continuously strive for optimum quality in mammography screening. Comprehensive guidelines on quality assurance of the mammography screening process have been developed in the Europe against Cancer Programme (2). Staff performing mammography screening should ensure that the minimal standards and procedures recommended in the guidelines are followed. The European guidelines also emphasize the importance of performing screening in an organized setting as opposed to ▶opportunistic screening by referring women for a mammogram in a clinical setting. ▶Organized screening programs include an administrative structure responsible for implementation, quality assurance, and evaluation. Accountability for the quality of performance and results is clearer in an organized program and managerial control of the screening process is greater, allowing more rapid response to changes in quality indicators or adverse conditions. Another key quality aspect is population-based implementation of invitation and evaluation. Identifying and inviting all eligible women in the target population is not merely important for maximizing the potential impact of screening. Population-based invitation and a high uptake rate make screening data representative for the local population. Such representative data are essential for internal and external monitoring of screening performance.

References 1.

Vainio H, Bianchini F (2002) IARC Handbooks of Cancer Prevention. Vol 7, Breast Cancer Screening. International Agency of Research on Cancer, WHO, Lyon, France

S

1654

2.

3.

Scrotal Disorders

Perry N, Broeders M, de Wolf C (eds). et al (2001) European Guidelines for Quality Assurance in Mammography Screening. 3rd edn. Office for Official Publications of the European Communities, Luxembourg Dronkers Daniel J, Hendriks Jan H C L (1999) Holland Roland, Rosenbusch Gerd Radiologische Mammadiagnostik. Thieme Verlag, Stuttgart

Scrotal Disorders A RON C OHEN 1 , H UBERT J OHN 2 1

Department of Urology, University Hospital Zurich, 8091 Zurich, Switzerland 2 Klinik Hirslanden, Zentrum fu¨r Urologie, 8008 Zurich, Switzerland [email protected], [email protected] [email protected]

Scrotal Disorders. Figure 1 Testicular torsion. Color Doppler ultrasound reveals lack of vascularization in the twisted spermatic cord.

Definitions Scrotal disorders can be subdivided into acute and painful versus chronic and asymptomatic lesions. In acute scrotal pain, the differential diagnosis must be focused on ▶testicular torsion, epididymitis, orchitis, torsion of the testicular appendage, or scrotal trauma. Inguinoscrotal hernia or distal ureterolithiasis might be further nonscrotal sources of acute pain. Painless lesions of the scrotal organs present most often as tumor masses, either of the testis (testicular cancer), the epididymis (▶spermatocele), the vascular drainage (▶varicocele), or the surrounding tissue (▶hydrocele).

Pathology Painful scrotal diseases

Scrotal Disorders. Figure 2 Testicular torsion. Intraoperative situs of an intravaginal testicular torsion without infarction.

Testicular Torsion Usually there is torsion of the testis within the tunica vaginalis, which covers the anterior surface of the testis and extends over the epididymis and the spermatic cord. Where the covering extends to the cord, the testis is suspended freely within the tunical cavity, within which it may rotate. Torsion of the spermatic cord will strangulate the main vessels and cause ischemic damage to the testis. The main peak of testicular torsion is seen in adolescence. In the neonatal period, testicular torsion presents as torsion including the tunica vaginalis, which is called an extravaginal torsion (Figs 1–3).

Epididymitis, Orchitis Epididymitis is usually a bacterial infection, although viral and fungal infections are possible as well. Most common infections in younger patients (5 years). Spinal involvement occurs late in the course of the disease and frequently goes unobserved because of the extensive involvement of the peripheral joints. The most common abnormality is erosion of the odontoid process (47%), followed by atlanto-axial dislocation and apophyseal joint

1717

involvement (24%). Atlanto-axial dislocation is present when the distance between the anterior arch of C1 and the odontoid process of C2 exceeds 2.5 mm and increases when the cervical spine moves from extension (lordosis) to flexion (kyphosis). CT can be used to demonstrate destructive bony changes and ligamentous calcifications and may also allow evaluation of synovial hypertrophy (pannus formation) in RA. MRI is best for pointing out soft-tissue changes. Periodontal pannus is the most frequent finding. Pannus is homogeneous on T1-weighted images and shows contrast enhancement and is usually hyperintense on T2-weighted images; in cases of predominantly fibrous tissue, it is hypointense. Odontoid erosions are less frequently observed by MRI. They may show a heterogeneous MRI signal, and sometimes complete destruction. Spinal cord involvement may occur as a result of severe anterior C1–C2 subluxation, atlanto-axial settling, or compression by retrodental pannus. T2 or T2-weighted images well demonstrate cord deformation and intramedullary signal increase (myelomalacia). MRI studies in cervical flexion and extension are more difficult to perform than dynamic plain X-ray films and do not provide more information. MRI may also demonstrate spinal canal narrowing, with cord compression occurring in the lower cervical region in RA. In RA, the thoracic and lumbar spine are infrequently involved. This helps to distinguish RA from ankylosing spondylitis, psoriatic arthritis, and Reiter’s syndrome. However, cervical destructive changes with atlanto-axial subluxation may occur in all these conditions.


Manelfe C (ed) (1992) Imaging of the Spine and Spinal Cord. Raven Press, New York (2002) Annals of Rheumatic Disease 61, Supplement III Resnick D (1995) Diagnosis of Bone and Joint Disorders. 3rd edn. WB Saunders, Philadelphia Brower AC (1988) Arthritis in Black and White. WB Saunders, Philadelphia

Spine, Intramedullary Tumors DANIELLE B ALERIAUX , N ESLI¨ HAN G U¨ LTAS¸ LI Clinique de Neuroradiologie, Hoˆpital Erasme, Universite´ Libre de Bruxelles, Brussels, Belgium [email protected] [email protected]

S

1718

Spine, Intramedullary Tumors

Definitions The spinal cord is a small but extremely important element of the central nervous system (▶CNS) and is especially well protected by the surrounding cerebrospinal fluid (CSF), the meningeal envelopes, and the bony spine. Although the anatomy is well known for a long time, the radiological approach has remained a sensitive subject and was difficult before the advent of magnetic resonance imaging (MRI). Major progress has been made in this field thanks to MRI (1, 2, 3). Tumors located within the spinal cord are rare: they account for 4% of all CNS tumors. The vast majority of ▶intramedullary tumors are glial tumors: the most frequently encountered neoplasms in adults are ependymoma (40%) and astrocytoma (28%). In children, however, astrocytomas are by far the most frequent tumors (90%) and ependymomas occur only in children affected by neurofibromatosis. Hemangioblastomas represent 14% of all spinal cord tumors. Less frequent tumors include intramedullary metastasis, lymphoma, epidermoid cyst, lipoma, ganglioglioma, oligodendroglioma, paraganglioma, intramedullary schwannoma, and teratoma (4, 5). Cavernomas represent 8% of intramedullary mass lesions; however, they are not true neoplasms but vascular malformations. Generally, spinal cord tumors enlarge the spinal cord. Besides a solid nodule, associated cysts may often involve large segments of the cord and should be distinguished from intratumoral cysts.

Histopathology/Anatomy Only the most commonly found intramedullary tumors are described hereafter. Astrocytomas are often large lesions with ill-defined limits; a cleavage plane is often not found at surgery. Astrocytomas are hypercellular lesions with no surrounding capsule. At histology, enlarged, irregularly shaped, hyperchromatic nuclei are present. Like in the brain, four different grades are recognized according to the World Health Organization (WHO). Grade I, considered most benign, is pilocytic astrocytoma. Grade II is fibrillary astrocytoma (low grade astrocytoma), which is the most common type in our series (75% of astrocytomas). Anaplastic astrocytoma is grade III, characterized by more hypercellularity and necrotic regions. The most malignant is grade IV, with glioblastoma multiforme showing endothelial proliferation and larger areas of necrosis and hemorrhage. Contrary to the brain, they are rare lesions: 0.2–1.5% of spinal astrocytomas (4).

Ependymomas originate from the ependymal cells surrounding the central canal of the spinal cord. Associated satellite cysts are frequently seen. Different types of ependymomas are identified: cellular (the classic and most common type), papillary, clear cell, tanycytic, and melanotic (less common type) (4). Perivascular pseudorosettes are required for the diagnosis of ependymoma. The majority of ependymomas are of low grade and are well-circumscribed tumors corresponding to a grade I or II according to WHO classification. Malignant types are rarer. Ependymomas are prone to hemorrhage, especially at the margins of the tumor; this is responsible for the so-called cap sign seen on gradient echo ▶T2-weighted images (WI). Calcifications may be found but are rare. Myxopapillary ependymomas represent a distinct entity: they are typically located at the level of the filum terminale. These tumors are lobulated with a distinct capsule. At histology, tumor cells are heterogeneous, including mucin-producing cells, papillary-type cells mixed with the typical rosettes and pseudorosettes elements. Although usually of low grade, some tumors are more aggressive, type III lesions. Hemangioblastomas are well-delineated nodular masses with extremely rich vascular components. Histologic examination shows pale stromal cells packed between blood vessels of varying size (4). Commonly, small tumors are associated with huge and extensive cystic components that may involve the entire cord. Metastasis: Longer survival of cancer patients on the one hand and better imaging techniques on the other hand today allow the detection of a higher number of intramedullary metastases. The most common primary tumors include lung and breast cancer as well as melanoma.

Clinical Presentation Spinal cord tumors grow very slowly, being mainly low-grade tumors (80%). They occur in younger adult patients: the mean age at diagnosis is 34 years for astrocytomas and 42 years for ependymomas. A slight male predominance (52–55%) is reported. The most frequent and early symptom is pain, and this may be the only one at the onset of the disease. Variations in tumor location will be responsible for variable motor and sensory deficits. Diagnosis is unfortunately made after a long period (mean 2.5 years after the onset of symptoms) as the clinical presentation is not specific. Urinary disturbances and impotence are less frequent and usually appear late in the clinical course of the disease, coincident with paralysis of the legs.


In children as well, pain is the most frequent symptom as reported in 42% of cases. Motor regression is present in 36%, gait abnormality in 27%, torticollis in 27%, and progressive kyphoscoliosis in 24% of cases. Low-grade lesions account for 89% of cases (6). Acute worsening of the symptoms may occur due to intratumoral hemorrhage. Subarachnoid hemorrhage (SAH) of spinal origin is rare and the most common source is an arteriovenous malformation. Nevertheless, cases of SAH due to bleeding from hemangioblastoma have been reported in the literature. Hydrocephalus may be observed associated with an intradural spinal tumor: this is a rare but classical finding. Finally, it is extremely rare that a spinal cord metastasis presents as the first

1719

clinical manifestation of a distant primary cancer. Clinical symptoms are nonspecific, but usually involve root pain.

Imaging Astrocytomas According to our experience, astrocytomas are found at the cervical (33%), cervicothoracic (21%), or thoracic (36%) levels, while the remaining 7% are found at the thoracolumbar and lumbar region. The mean size of the solid tumor component involves four vertebral segments. However, especially in the pediatric population, the entire spinal cord may be involved. Cystic components within

S

Spine, Intramedullary Tumors. Figure 1 Cervical fibrillary astrocytoma (Grade II). (a) Sagittal T2-WI The tumor extends from C2 to C6 and appears heterogeneously hyperintense with ill-delineated borders. (b) Sagittal T1-WI: the cervical spinal cord is moderately enlarged and slightly hypointense. (c) Sagittal Gd-enhanced T1-WI: small areas of patchy enhancement are seen at the level of C5. (d) Axial T2-WI and (e) axial Gd-enhanced T1-WI: tumor infiltration is excentrically located towards the left part of the cervical cord.

1720


the tumor are found in 27% of the cases. Associated satellite cysts and secondary hydromyelia are seen in 50% of astrocytomas. On ▶T1-WI, astrocytomas are mostly hypointense (83%), whereas on T2-WI they are predominantly hyperintense. Low-grade astrocytomas usually do not enhance, although moderate enhancement may be observed (Fig.1). Pilocytic astrocytomas, however, do enhance intensely. High-grade astrocytomas and glioblastomas tend to be more heterogeneous with necroticocytic areas and they enhance typically in a patchy mode (60%). Intratumoral hemorrhage is best detected on gradient echo T2-WI. Intratumoral cysts and necrosis are common. Associated huge syringomyelia may occur: the borders of these associated cavities do not enhance after contrast medium injection.

usually have well-defined borders. However, no or very limited contrast enhancement may be seen. As ependymomas originate from the cells bordering the ependymal canal they are located more centromedullary compared to astrocytomas. The mean tumor size of ependymomas corresponds to three vertebral bodies (min. 2 to max. 13), while astrocytomas are usually more extensive involving 2 to 19 vertebrae. Myxopapillary ependymomas of the conus medullaris and filum terminale are often large, well-circumscribed lesions appearing isointense compared to the cord on T1-WI and hyperintense on T2-WI. Strong enhancement is seen after gadolinium injection. Hemorrhage may be observed, explaining the sudden worsening of clinical symptoms with occurrence of acute leg weakness and sphincter disturbances.

Ependymomas Typically, ependymoma is more frequently found in the cervical spinal cord (50%). Associated large satellite cysts are seen in 60% of the cases. A so-called cap sign is seen in almost one out of four cases (27%) and corresponds to low-signal-intensity areas seen on T2-WI and even better on gradient echo T2-WI, capping the tumor limits on both sides (Fig. 2). These caps are hemosiderin deposits caused by chronic hemorrhage. Ependymomas enhance vividly and homogeneously in 91% of the cases, and

Hemangioblastomas Hemangioblastomas have two different but quite typical presentations: either they are small nodular lesions located in the subpial compartment and surrounded by extensive edema or they are small nodules associated to huge and extensive cystic components (Fig. 3). These tumors are either solitary (80%) or multiple, when associated with von Hippel-Lindau disease. The solid nodule is isointense to hypointense on T1-WI and

Spine, Intramedullary Tumors. Figure 2 Thoracic ependymoma (Grade II). (a) Sagittal T1-WI. (b) Sagittal T2-WI. (c) Gradient echo T2-WI and (d) sagittal Gd-enhanced T1-WI. The solid component of the tumor is located at the level of Th2-Th3. Associated cystic components are found at both tumor extremities. Hemosiderin deposits are best seen on (c). The tumor enhances homogeneously after gadolinium injection.


1721

Spine, Intramedullary Tumors. Figure 3 Cervical hemangioblastoma. (a) Sagittal T2-WI. (b) Sagittal T1-WI and (c) Sagittal Gd-enhanced T1-WI. The small tumor nodule is best detected after gadolinium injection. The borders of the associated cyst do not enhance. Significant edema is best evaluated on the T2-WI.

isointense to slightly hyperintense on T2-WI. A rich vascular network in the tumor as well as enlarged feeding arteries and dilated draining veins may best be seen on proton density images and T2-WI. After gadolinium injection, intense and homogeneous contrast uptake is seen exclusively within the nodular part. Contrast administration is especially useful in order to pick up small, multiple nodules when associated to large cystic components whose borders do not enhance.

Metastases The high sensitivity of MRI enables easy detection of intramedullary metastases; however, no specific MRI characteristics are seen. Usually, spinal cord metastases are small, nodular, well-defined lesions, hypointense on T2-WI, surrounded by mild to extensive edema. The enhancement pattern may be either ring-like or homogeneous and intense. Melanoma metastasis, on the contrary, has a more specific appearance exhibiting a spontaneously hyperintense aspect on T1-WI linked to the presence of melanin.

Cavernomas Cavernomas represent in our experience 8% of all intramedullary tumors. On MRI, intramedullary cavernomas are usually easily recognized thanks to a typical “black and white” appearance due to areas of mixed signal

intensity on both T1- and T2- or T2*-WI. Contrast enhancement is variable. As cavernomas may be multiple, we recommend cerebral MRI whenever the diagnosis of cavernoma is suspected; indeed, if multiple similar lesions are found in the brain, this should support the final diagnosis of cavernoma of the spinal cord.

Gangliogliomas Gangliogliomas are rare tumors, representing only 3.8% of all CNS tumors. They are more frequent in children. They mostly involve the upper cervical cord. The diagnosis of ganglioglioma can be strongly suspected in young patients (mean 12 years) whenever a large tumor is found presenting with a tumoral cyst, no edema, mixed signal intensity on T1-WI, and patchy enhancement. Associated bone erosion and scoliosis have been reported.

Rare Tumors As mentioned above, rare tumors such as lymphoma, epidermoid cyst, lipoma, oligodendroglioma, paraganglioma, intramedullary schwannoma, and teratoma may be found. These often have nonspecific imaging features except for lipomas, which are typically high-signalintensity lesions on T1-WI. Melanocytic schwannoma could be suspected preoperatively in our series thanks to the hyperintense appearance of the tumor on T1-WI.

S

1722

Spleen, Infectious Diseases

However, in these rare tumors, the ultimate diagnosis will still be ascertained by histopathology after biopsy, which is strongly recommended whenever an intramedullary tumor is suspected both clinically and on the basis of imaging findings.

References 1.

2. 3. 4.

5. 6.

Bale´riaux D, Parizel P, Bank WO (1992) Intraspinal and intramedullary pathology. In Manelfe C (ed) Imaging of the Spine and Spinal Cord. Raven Press New York pp 513–564 Bale´riaux D (1999) Spinal cord tumours. Eur Radiol 9:1252–1258 Bale´riaux D, Brotchi J, Salmon I (2004) Spinal cord tumours: a pictorial essay. Rivista di Neuroradiologia 17:336–348 Koeller KK, Rosenblum RS, Morrison AL (2000) Neoplasms of the spinal cord and filum terminale: Radiologic-pathologic correlation. Radiographics 20:1721–1749 Van Goethem JW, Van den Hauwe L, Ozsarlak O et al (2004) Spinal tumours. Eur J Radiol 50(2):159–176 Tortori Donati P and Rossi A (2005) Tumours of the Spine and Spinal Cord. In Pediatric Neuroradiology. Springer, Berlin Heidelberg, New York ISBN 3-540-41077-5, Chapter 40:1609–1653

Spleen, Infectious Diseases P IERO B ORASCHI 1 , F RANCESCAMARIA D ONATI 1 , A LESSIO AUCI 2 , S IMONETTA S ALEMI 2 , M ARZIO P ERRI 2 , C ARLA C APPELLI 2 1

2nd Department of Radiology, University Hospital of Pisa, Pisa, Italy 2 Department of Diagnostic and Interventional Radiology, University of Pisa, Pisa, Italy [email protected] [email protected]

Synonyms Infections of the spleen; Splenic abscess; Splenic inflammation

Definition Infectious diseases of the spleen are related to splenic phagocytic immune functions and are characterized by primary splenic abscess.

Pathology and Histopathology Splenic abscess is a rare condition that tends to occur in patients with predisposing factors such as preceding pyogenic infection, immunodeficiency, and contiguous disease in the pancreas. Several different mechanisms are

presently accepted to explain its etiology. Hematogenous spread from an infective focus elsewhere in the body (endocarditis, intraabdominal sepsis, osteomyelitis, or chest infection) is the most common cause of splenic abscess, but it may also occur as direct spread of infection in contiguous areas, such as pancreatitis, retroperitoneal and subphrenic abscesses, and diverticulitis. Finally, in some cases, splenic abscess represents a delayed infective complication of either traumatic lesions or large infarctions of the spleen. Depending on the causative organisms, pyogenic, fungal, and tubercular splenic abscess may be distinguished (1, 2).

Pyogenic Abscess The most common organisms obtained from culture of pyogenic abscess are aerobic microbes, in particular the staphylococci, streptococci, Salmonella, and Escherichia coli. Anaerobic organisms, such as Pseudomonas, are less frequently involved (1). At gross examination, most pyogenic splenic abscesses are solitary and unilocular lesions ranging from a few millimeters to several centimeters in diameter. However, in immunocompromised patients, multiple splenic abscesses usually associated with abscess in other viscera are frequently observed. At histopathologic analysis, the abscess cavity is filled with purulent material, while the edges are composed of a chronic inflammatory infiltrate and fibrous tissue. The fibrotic edge is often a centimeter or more thick. Depending on the stage, suppuration, liquefaction with presence of debris, and fibrosis are found at microscopic analysis.

Fungal Abscess Fungal abscesses are most often caused by Candida albicans, Aspergillus, and Cryptococcus neoformans and typically occur in immunocompromised patients, representing a manifestation of disseminated fungal disease. Fungal abscesses are multiple small lesions, typically only a few millimeters in diameter and often also involving the liver and, occasionally, the kidney (1–3). The typical histological pattern of splenic candidiasis is characterized by microabscesses, with fungi in the center of the lesion and a surrounding area of necrosis and polymorphonuclear infiltrate; in the healing stage there is a fibrotic evolution of the lesions.

Tuberculosis Tuberculosis of the spleen is rarely seen in isolation and is more frequently seen as part of a multifocal or

Spleen, Infectious Diseases

disseminated disease. The causative organism is Mycobacterium tuberculosis or, particularly in immunocompromised hosts, Mycobacterium avium-intracellulare. Splenic involvement usually occurs by hematogenous spread of infection in the form of microabscesses in a miliary tuberculosis pattern, which can become calcified, or, rarely, as larger abscesses or granulomas (4).

Clinical Presentation The clinical presentation is often subtle and diagnosis delayed (1). Fever, left upper abdominal pain, pleuritic chest pain, and malaise are the most common symptoms. The most important findings at physical examination are left upper quadrant tenderness and splenomegaly. Leucocytosis is invariably present in all patients.

1723

Fungal abscesses have variable appearances on US. In most cases they appear as rounded, hypoechoic lesions with a central echogenic area, creating a target or bull’seye pattern corresponding to fibrotic tissue surrounding a central inflammatory core at histopathologic analysis. The wheel-within-a-wheel appearance is observed when the central hyperechoic portion becomes necrotic and hypoechoic (2). CT shows multiple small, low-attenuation lesions. The lesions may be missed unless intravenous contrast medium is used. At MR, fungal abscess appears as multiple small lesions that are hypointense on T1-weighted images and hyperintense on T2-weighted images (Fig. 1) (3).

Tuberculosis Splenic tuberculosis can occur in the form of microabscesses or in the form of larger abscesses or granulomas.

Imaging Pyogenic Abscess On ultrasound (US), splenic pyogenic abscess shows an anechoic or hypoechoic pattern, with an irregular wall, associated splenomegaly, and variable amounts of internal echogenicity and acoustic transmission due to necrotic debris. Intraabscess gas may be observed and is indicative of infection by gas-producing agents. These features are similar to those for abscess in other abdominal organs and are not pathognomonic for splenic abscess. At color Doppler US evaluation, avascular lesions with no hypervascular rim are usually observed. Splenic pyogenic abscesses appear as low-density areas on nonenhanced computed tomography (CT) scans and with rim enhancement after intravenous contrast medium administration. A gas or fluid level within the lesion may be observed. Depending on the stage of abscess development, abscesses may be clearly demarcated from the surrounding tissue and have a ringlike, sharply marginated wall. At magnetic resonance imaging (MR), pyogenic abscesses appear as areas of decreased signal intensity on T1-weighted images and increased signal intensity on T2-weighted images. After contrast administration, the lesion shows peripheral enhancement.

Fungal Abscess Typically, splenic abscesses are observed in immunosuppressed patients and have a miliary distribution appearing on imaging as multiple small splenic lesions. They are usually associated with hepatic involvement.

S

Spleen, Infectious Diseases. Figure 1 At magnetic resonance imaging, the abscesses appear as areas of increased signal intensity on T2-weighted imaging (a) and decreased signal intensity on T1-weighted imaging (b) on a postcontrast T1-weighted image, (c) the lesions show a slight peripheral enhancement.

1724

Splenic Abnormalities

Nuclear Medicine Radionuclide scans are sensitive, though less specific, tools for diagnosing splenic abscess and were largely employed before the advent of US and CT. Particularly, 99m-Technetium sulfur-colloid liver and spleen scanning is the scintigraphic technique most commonly used and shows a focal photopenic defect or delayed uptake, while on 67-Gallium citrate scans a focal uptake in the spleen—occasionally with a perisplenic halo—is usually observed (1).

Diagnosis

Splenic Abnormalities. Figure 2 Splenic calcifications in an immunodeficient patient with a previous history of tuberculosis and pneumocystis pneumonia. Ultrasound (a, b) clearly demonstrates multiple calcifications in the spleen.

The micronodular pattern is typically observed in the military type of pulmonary tuberculosis and is usually noted as a moderate splenomegaly. On US a typical coarsened echotexture is often seen, while on CT the miliary nodules cannot be detected, and a heterogeneous, moderately enlarged spleen, sometimes with calcifications, is usually the more frequent finding (Fig. 2). However, these radiological features are usually not specific and may be observed in several different benign or malignant conditions of the spleen. The macronodular form is characterized by the presence of single or multiple larger abscesses or granulomas. Splenic tubercular abscess are often hypoechoic on US and round or ovoid lesions with low attenuation value on CT. Peripheral calcifications and internal septations may be observed. After contrast administration, the rim of the lesion enhances moderately, while no contrast uptake is observed in the low-density center. On MR a hypointense nodule with a hypointense rim on T1-weighted images and an isointense or hyperintense nodule with a less intense rim on T2-weighted images are usually observed (5).

Early diagnosis and timely treatment reduce the morbidity and mortality associated with splenic abscess. US and/ or CT should be used routinely for evaluating patients with fever and abdominal pain. Although both US and CT of the abdomen are of diagnostic value, CT is more accurate and reliable in the diagnosis of splenic abscess. There are several differential diagnoses, including splenic infarct, pseudocysts, hematomas, and tumors. The presence of a gas or fluid level within the spleen, although quite rare, is pathognomonic for pyogenic infection, while multiple small, hypovascular splenic lesions in immunocompromised patients are highly suggestive for fungal abscess (1, 3). Moreover, the CT and US appearance of splenic abscess is a valuable predictor of the patient’s response to treatment, with multiple and gas-containing abscesses indicating a poor prognosis.

Bibliography 1. 2.

3. 4.

5.

Ooi LL and Leong SS (1997) Splenic abscesses from 1987 to 1995. Am J Surg 174: 87–93 Paterson A, Frush DP, Donnelly 3rd et al (1999) A pattern-oriented approach to splenic imaging in infants and children. Radiographics 19:1465–1485 Robertson F, Leander P, and Ekberg O (2001) Radiology of the spleen. Eur Radiol 11:80–95 Batra A, Gulati MS, Sarma D et al (2000) Sonographic appearances in abdominal tuberculosis. J Clin Ultrasound 28(5): 233–245 Akhan O and Pringot J (2002) Imaging of abdominal tuberculosis. Eur Radiol 12:312–323

Splenic Abnormalities ▶Splenic Anomalies

Splenic Anomalies

Splenic Abscess ▶Spleen, Infectious Diseases

1725

Shape Abnormalities Splenic clefts, notches, and lobules may persist in adult life as variations of the normal shape and are quite common findings (1).

Abnormal Location

Splenic Anomalies P IERO B ORASCHI 1 , F RANCESCAMARIA D ONATI 1 , A LESSIO AUCI 2 , S IMONETTA S ALEMI 2 , M ARZIO P ERRI 2 , C ARLA C APPELLI 2

The spleen may be found in a variety of abnormal locations. In congenital diaphragmatic hernia and eventration of the diaphragm, the spleen may have an intrathoracic location; if the lateral peritoneal recess is particularly deep, the spleen is found posterior to the left kidney. The spleen may rarely be located in the right hypochondrium, usually in patients with situs viscerum inversus (1).

1

2nd Department of Radiology, University Hospital of Pisa, Pisa, Italy 2 Department of Diagnostic and Interventional Radiology, University of Pisa, Pisa, Italy [email protected] [email protected]

Synonyms Splenic abnormalities; Splenic malformations

Definition Splenic anomalies include congenital anomalies of shape, location, number, and size of the spleen, due to aberrant embryologic development.

Splenogonadal Fusion and Splenorenal Fusion Splenogonadal fusion is a rare anomaly. Owing to the close relationship between the developing spleen and the left gonadal-mesonephric structures, an accessory spleen may be found attached to the left ovary or within the scrotum or the left kidney. Splenogonadal fusion can be classified into two types: continuous (direct connection between spleen and gonad) and discontinuous (no anatomic connection between ectopic splenic tissue and the principal spleen). This anomaly predominates in males. It can occur as an isolated condition or can be associated with other abnormalities, such as cryptorchidism and orofacial and limb abnormalities. In splenorenal fusion, normally functioning splenic tissue is abnormally located in close proximity to the kidney, usually on the left side (1, 3).

Wandering Spleen

Pathology and Histopathology Accessory Spleen The spleen is a mesodermal derivate that first appears as a mesenchymal cell condensation inside the dorsal mesogastrium. One or more additional smaller splenic condensations sometimes develop and give origin to accessory spleens, representing by far the most common congenital abnormality of the spleen. Accessory spleens are small round masses that perfectly resemble the splenic structure. They may be located anywhere in the abdomen, but the most common sites are near the splenic hilum and the tail of the pancreas. Other possible locations are along the splenic vessels, in the gastrosplenic and splenorenal ligaments, in the mesentery, and in the omentum. Any accessory splenic tissue is capable of hypertrophy. When splenectomy is performed for hypersplenism, hypertrophy of an accessory spleen may cause recurrent disease (1, 2).

Absence, laxity, or excessive length of splenic ligaments leads to an abnormal mobility of the organ. This condition is known as “wandering spleen.” Torsion of the long vascular pedicle may occur, followed by vascular occlusion and splenic ischemia or even infarction (1, 4).

Asplenia and Polysplenia The absence of the spleen (asplenia) and the presence of multiple small spleens (polysplenia) are rare conditions usually associated with other congenital malformations, especially cardiovascular anomalies. Either polysplenia or asplenia may be seen in association with abdominal situs ambiguous; these conditions have been classically called asplenia and polysplenia syndromes (1, 5).

Splenic Agenesis Congenital splenic agenesis is quite uncommon and is associated with recurrent bacterial infections (1).

S

1726

Splenic Anomalies

Clinical Presentation Accessory spleens and splenic shape abnormalities are typically incidental findings at imaging. Splenogonadal fusion usually manifests as a mobile and painless left scrotal mass in males. Other presentations include cryptorchidism, testicular torsion, and inguinal hernia. Splenogonadal fusion is often asymptomatic in females. In cases of intrathoracic location of the spleen, patients usually have respiratory symptoms. The wandering spleen is usually symptomatic in childhood and may present with an abdominal mass and acute, chronic, or intermittent symptoms due to torsion of the pedicle. In children with asplenia and polysplenia syndromes, the clinical manifestations may be related to congenital heart disease, immunodeficiency due to splenic absence, or volvulus due to intestinal malrotation (1).

Imaging Accessory Spleen Accessory spleens may vary in number and size, usually ranging from a few millimeters to several centimeters in size (Fig. 1). Typically they appear as round or oval masses near the splenic hilum, but they may be found anywhere in the abdomen. Intrapancreatic accessory spleen, typically in the tail, can mimic a neoplastic mass. The demonstration of imaging features identical to those of normal splenic tissue allows the diagnosis (Fig. 2). At ultrasound (US), the mass shows the same echotexture as the splenic parenchyma. At computed tomography (CT), the attenuation values before and after contrast administration are identical to those of the spleen. In particular, depiction of the characteristic inhomogeneous enhancement during the arterial phase is crucial to demonstrate the nature of the mass. On magnetic resonance (MR) images, an accessory spleen has signal intensities identical to the spleen on all sequences both before and after contrast administration; in particular, the use of reticuloendothelial-targeted contrast media can confirm the splenic nature of the mass (1, 2).

Shape Abnormalities Splenic clefts, notches, and lobules are quite common findings and must be distinguished from traumatic lesions. The demonstration at US and CT of characteristic smooth edges and the absence of extravasation suggest a congenitally cleft spleen (1).

Splenic Anomalies. Figure 1 Accessory spleen. Typically this anomaly appears as a nodule near the splenic hilum. Ultrasound (a) demonstrates a nodule that shows the same echotexture as the splenic parenchyma. On magnetic resonance images, the accessory spleen is hypointense on T1-weighted scan (b) and hyperintense on T2-weighted scan (c) with respect to the liver, showing signal intensities identical to the spleen on all sequences.

Splenogonadal Fusion and Splenorenal Fusion In the case of left scrotal mass, US is usually the first examination. The splenic tissue appears as a homogeneous, well-encapsulated mass with the same echotexture as the normal spleen. At Doppler US a vascular architecture analogous to that of the spleen can be depicted. If splenogonadal fusion is suspected, a cordlike structure connecting the spleen to the mass should be sought. CT may be helpful to demonstrate associated splenic-renal

Splenic Anomalies

1727

and medially. Torsion of the pedicle may lead to ischemia or even infarction. The congested or infarcted spleen may have a normal echotexture or show diffuse increased echogenicity corresponding to hemorrhagic phenomena. The spleen’s comma-shaped configuration is usually preserved. Splenomegaly with rounded edges of the organ, when present, is strongly suggestive for torsion and has been attributed to congestion. Doppler US demonstrates no flow within the splenic parenchyma. Nonenhanced CT usually shows decreased attenuation of splenic parenchyma, and contrast medium administration may show a partial or total lack of enhancement. A highly specific sign of torsion is a whirl appearance of the splenic vessels and surrounding fat, with alternating bands of high attenuation and low attenuation. This finding is usually noted at the splenic hilum (1). MR can provide useful information about the precise location of the wandering spleen, the viability of the splenic parenchyma (assessed both with nonenhanced and enhanced scans), and the splenic vessel anatomy, thanks to MR angiography (4).

Polysplenia and Asplenia In polysplenia, numerous small splenic masses can be seen predominantly in the right upper quadrant at US, CT, MR, and scintigraphy. Documenting the absence of the spleen is more difficult than confirming its presence. Scintigraphy is the standard examination (1, 5).

Nuclear Medicine

Splenic Anomalies. Figure 2 Intrapancreatic accessory spleen. The accessory spleen appears hypointense on coronal T1-weighted image (a) and hyperintense on axial T2-weighted image (b) with respect to the pancreas. On an axial reticuloendothelial-enhanced T2-weighted image (c) a significant signal intensity decrease is consequently observed at the level of the intrapancreatic accessory spleen, spleen, and liver.

fusion or cryptorchidism and to show the typical enhancement pattern after contrast administration (1, 3).

Wandering Spleen US and especially CT well demonstrate the absence of the spleen in its expected location and the presence of a homogeneous soft-tissue mass located more inferiorly

Technetium-99m sulfur colloid scintigraphy and technetium-99m-labeled heat-damaged red blood cells offer functional images and are therefore highly specific for differentiating spleen from other tissues. Although nuclear medicine offers the most specific imaging techniques for identifying functional ectopic splenic tissue, CT and MR offer superior anatomic resolution (2). Scintigraphic studies are used to confirm the presence of normally functioning splenic tissue in cases of suspected accessory spleens, ectopic spleen, splenogonadal and splenorenal fusion, and polysplenia. Scintigraphy is also the examination of choice in documenting absence of the spleen. However, absence of uptake of radiotracer can also occur with the so-called functional asplenia, in which the splenic phagocytic function is markedly reduced despite the presence in the body of splenic tissue. Functional asplenia may occur secondary to radiation therapy and chemotherapy, secondary to tumor invasion of the spleen, in sickle cell disease, with splenic anoxia, and after bone marrow transplantation (1).

S

1728

Splenic Atrophy

Diagnosis Ectopic splenic tissue may mimic neoplasms and lymphadenopathies. Imaging can achieve a definite diagnosis, thus avoiding open biopsy. Ectopic splenic tissue shows imaging features identical to those of the spleen in all imaging modalities. The enhancement pattern, especially the inhomogeneity in the arterial phase, is very specific. MR reticuloendothelial-targeted contrast media can confirm the splenic nature of the mass. 99m-Technetium sulfur colloid scintigraphy and technetium-99m-labeled heat-damaged red blood cells represent the most specific imaging techniques to confirm the presence of functioning splenic tissue (2). In splenogonadal fusion presenting with scrotal mass, US is usually the first examination to be performed, particularly for identifying a cordlike structure connecting the mass with the normal spleen. However, surgical exploration is generally required to rule out malignancy (3). Accurate preoperative diagnosis of wandering spleen with or without torsion represents an imaging challenge and can be done with US, CT, and MR. Information concerning splenic perfusion and viability and the performance of Doppler ultrasonography and contrastenhanced CT and MR studies is important for the surgeon, especially in younger children in whom splenopexy instead of splenectomy is the treatment of choice for uncomplicated wandering spleen.

Splenic Benign Tumors ▶Neoplasms, Splenic, Benign

Splenic Congenital Abnormalities ▶Congenital Anomalies, Splenic

Splenic Inflammation ▶Spleen, Infectious Diseases

Splenic Malformations ▶Splenic Anomalies ▶Congenital Anomalies, Splenic

Splenic Malignancies Bibliography 1.

2.

3. 4.

5.

Paterson A, Frush DP, Donnelly LF et al 3rd. (1999) A patternoriented approach to splenic imaging in infants and children. Radiographics November–December 19(6):1465–1485 Miyayama S, Matsui O, Yamamoto T et al (2003) Intrapancreatic accessory spleen: evaluation by CT arteriography. Abdom Imaging November–December 28(6):862–865 Pomara G (2004) Splenogonadal fusion: a rare extratesticular scrotal mass. Radiographics March–April 24(2): 417 Deux JF, Salomon L, Barrier A et al (2004) Acute torsion of wandering spleen: MRI findings. Am J Roentgenol June 182 (6):1607–1608 Applegate KE, Goske MJ, Pierce G et al (1999) Situs revisited: imaging of the heterotaxy syndrome. Radiographics July–August 19 (4):837–852; discussion 853–854

▶Neoplasms, Splenic, Malignant

Splenic Malignant Tumors ▶Neoplasms, Splenic, Malignant

Splenic Traumatic Injury ▶Trauma, Splenic

Splenic Atrophy Splenic Traumatic Lesions Congenital splenic atrophy is quite uncommon and is associated with recurrent bacterial infections. ▶Congenital Abnormalities, Splenic

▶Trauma, Splenic

Splenomegaly

Splenic Traumatic Rupture

1729

Splenogonadal Fusion, Wandering Spleen

▶Trauma, Splenic ▶Congenital Malformations, Splenic

Spleno-Gonadal Fusion Spleno-gonadal fusion is a rare anomaly in which an accessory spleen may be found attached to the left ovary or within the scrotum or the left kidney. Spleno-gonadal fusion can be classified into two types: continuous (direct connection between the spleen and gonad) and discontinuous (no anatomic connection between ectopic splenic tissue and the principal spleen). This anomaly predominates in males and may occur as an isolated condition or associated with other abnormalities, such as cryptorchidism, orofacial, and limb abnormalities. Spleno-gonadal fusion typically manifests as a mobile and painless left scrotal mass in males. In females, spleno-gonadal fusion is often asymptomatic and is usually an incidental finding at US or CT. In spleno-gonadal fusion presenting with scrotal mass, US is usually the first examination to be performed. The homogeneity of the echotexture, the regularity of the vascular architecture and the presence of well-defined margins suggest a nonneoplastic nature of the mass. When spleno-gonadal fusion is suspected, a comparison with the US appearance of the spleen and a study directed to the visualization of a cord-like structure connecting the mass with the normal spleen should be performed. A definitive diagnosis cannot be made solely on the basis of US findings. Nuclear medicine imaging can confirm the presence of splenic areas of activity. However, surgical exploration is generally required to rule out malignancy. Nevertheless, orchiectomy can be avoided because splenic tissue can be dissected away from the tunica albuginea. ▶Congenital Abnormalities, Splenic

Spleno-Renal Fusion In spleno-renal fusion, normally functioning splenic tissue is abnormally located in close proximity of the kidney, usually on the left side. This anomaly presents as a renal mass which shows imaging features identical to those of the normal spleen. ▶Congenital Abnormalities, Splenic

Splenomegaly P IERO B ORASCHI 1 , F RANCESCAMARIA D ONATI 1 , A LESSIO AUCI 2 , S IMONETTA S ALEMI 2 , M ARZIO P ERRI 2 , C ARLA C APPELLI 2 1

2nd Department of Radiology, University Hospital of Pisa, Pisa, Italy 2 Department of Diagnostic and Interventional Radiology University of Pisa, Pisa, Italy [email protected] [email protected]

Synonym Enlargement of the spleen

Definition Splenomegaly is the enlargement of the spleen. It represents a manifestation of a splenic disorder rather than being a specific entity. In fact, it is the most common pathological finding of splenic disease. Many of the mechanisms of splenic enlargement are exaggerated forms of the spleen’s normal function. Although a wide variety of diseases are associated with splenomegaly, the most frequent causes include immune response work hypertrophy, such as in infectious mononucleosis; red blood cell destruction work hypertrophy, such as in ▶hereditary spherocytosis or thalassemia major; congestive causes, such as in splenic vein thrombosis or ▶portal hypertension; myeloproliferative causes, such as in chronic myeloid metaplasia; infiltrative causes, such as in sarcoidosis and some neoplasms; and neoplastic causes, such as in chronic lymphocytic leukemia and the lymphomas. Splenomegaly can also occur as the result of ▶extramedullary hematopoiesis. Miscellaneous causes of splenomegaly include trauma, cysts, hemangiomas, and metastasis.

Pathology and Histopathology When an enlarged spleen is referred to as hypertrophied, the underlying cause may be hypertrophy or hyperplasia

S

1730

Splenomegaly

of individual cells. In specific diseases, the splenic architecture is remodeled. For example, in Niemann– Pick disease, sphingomyelin and cholesterol accumulate within large foamy cells. With amyloidosis involving the spleen and resulting in splenomegaly, large hyaline masses are seen as lesions occupying the white pulp space. Two forms exist, including the “sago spleen,” in which amyloid deposits are limited to follicles, and the “lardaceous spleen,” in which amyloid is deposited in the walls of the splenic sinusoids. In a rare complication of typhoid fever, reactive splenic vasculitis may develop. In inflammatory splenomegaly, the demand for increased antigen clearance from the blood may lead to increased numbers of reticuloendothelial cells in the spleen and may stimulate accelerated antibody production with resultant lymphoid hyperplasia. In infiltrative splenomegaly (Gaucher’s disease, amyloidosis), engorgement of macrophages with indigestible materials can be found.

Clinical Presentation The most common history is mild abdominal pain that is vague in nature. Increased abdominal girth is less common. Early satiety from gastric displacement occurs with massive splenomegaly. Associated symptoms and signs include febrile illness (infectious process); pallor, dyspnea, bruising, and/or petechiae (hemolytic process); symptoms of liver disease (congestive process); and weight loss and constitutional symptoms (neoplastic process). Examination should include palpation with the patient in the supine and right lateral decubitus positions. Additional signs that identify possible etiologies of splenomegaly include signs of cirrhosis (asterixis, jaundice, telangiectasias, gynecomastia, caput medusae, ascites), heart murmur (endocarditis, congestive failure), jaundice and scleral icterus (spherocytosis, cirrhosis), and petechiae (any cause of thrombocytopenia). A patient with an enlarged spleen is more likely to have splenic rupture from blunt abdominal or low thoracic trauma, leading to typical symptoms. In particular, in elderly persons the combination of capsular thinning with increased spleen weight and size makes splenic injury more common. These factors account for the increased likelihood of splenectomy for trauma in this subgroup.

diffuse abnormalities (1). The size of the organ can be evaluated by measuring the length in the longitudinal scan, which allows simultaneous visualization of both the upper and lower poles and the hilum. A craniocaudal measurement of 11–13 cm is frequently used as the upper limit of normal for splenic size. Another possible method to demonstrate splenomegaly is to find an anteroposterior measurement up to two-thirds of the distance between the anterior and posterior abdominal walls. However, because of wide variations in shape, no consistent correlation has been recognized between the spleen’s length and its overall volume.

Computed Tomography Enlargement of the spleen is detectable by a variety of means, including physical examination and US; computed tomography (CT) is rarely necessary to document the presence of splenomegaly. When the spleen is enlarged, the concavity of its visceral surface is often lost as the spleen assumes a more globular shape (Fig. 1). CT can demonstrate findings that suggest the cause of the splenic enlargement (2). Neoplasm, abscess, or cystic lesions can be appreciated; associated abdominal lymph node enlargement can suggest a lymphoma. Cirrhotic patients show characteristic alterations in the size and shape of the liver and in the prominence of the venous structures in the splenic hilum. An increase in the attenuation value of the spleen (as well as the liver) can be found in patients with hemochromatosis.

Magnetic Resonance Imaging Magnetic resonance imaging (MR) is usually performed for problem solving, such as in the differential diagnosis

Imaging Ultrasonography Ultrasound (US) often represents the first imaging modality performed to evaluate the spleen. It allows the clinician to estimate the spleen’s size and morphology and, in some cases, to demonstrate splenic lesions or

Splenomegaly. Figure 1 Splenomegaly in acquired immunodeficiency syndrome. Computed tomography scan exhibits homogeneous enlargement of the spleen.

Splenomegaly

1731

when focal or diffuse disease of the spleen is present (3). For example, lymphoma dynamic sequences after injection of gadolinium-diethylenetriamine pentaacetic acid (DTPA) show large irregularly enhancing regions of highand low-signal intensity representing diffuse infiltration or focal low-signal intensity mass lesions scattered throughout the spleen. Superparagmagnetic iron oxide (SPIO) particles are selectively taken up by reticuloendothelial system (RES) cells, and not by malignant cells, improving the depiction of malignant lesions of the spleen. In splenomegaly caused by portal hypertension, arciform high-signal intensity enhancement on images obtained after gadolinium-DTPA injection is characteristic and excludes the presence of malignant disease. Gamna-Gandy bodies are spots of organized hemorrhage caused by portal hypertension and can be detected as multiple low-signal intensity nodules both on T1-weighted and T2-weighted images (Fig. 2) according to paramagnetic effects caused by hemosiderin deposits in the lesions (4). Patients with splenic enlargement secondary to hematologic disorders generally show no consistent

Splenomegaly. Figure 3 Splenomegaly in a patient with hereditary spherocytosis. Coronal fat-saturated T1-weighted (a) and T2-weighted (b) magnetic resonance images well demonstrate the spleen’s increased dimensions.

pattern of signal intensity alteration, and the spleen is often unchanged on MR images (Fig. 3). Patients affected by thalassemia have splenomegaly from extramedullary hematopoiesis and systemic iron overload from blood transfusions, with massive iron deposition in the spleen, which shortens T2.

Nuclear Medicine

Splenomegaly. Figure 2 Gamna-Gandy bodies in a patient with splenomegaly who previously underwent liver transplantation. Multiple low-signal intensity nodules are shown both on axial fat-saturated T1-weighted (a) and T2-weighted (b) magnetic resonance sequences.

A spleen scan is a good noninvasive technique for evaluating spleen size; a close correlation exists between spleen length on scan images and spleen weight after splenectomy. Erythrocytes should be labeled with chromium-51, mercury-197, rubidium-81, or 99m-technetium, and the cells altered by treatment with heat, antibody, chemicals, or metal ions so that the spleen sequesters them after infusion. A spleen scan is useful for detecting space-occupying lesions in the splenic substance and for evaluating loss of spleen function.

S

1732

Splenosis

Diagnosis Laboratory studies must always be performed to identify possible alterations that may suggest the cause of the splenomegaly. Moreover, splenomegaly produces hypersplenism, which is characterized by anemia, leukopenia, thrombocytopenia, or combinations thereof. Increased splenic platelet pooling is the primary cause of the thrombocytopenia of hypersplenism. In patients with hypersplenism, as much as 90% of the total platelet mass can be found in the spleen. In hypersplenism, the platelet count is usually 50,000–150,000/L. The etiology of the anemia observed in splenomegaly is the result of sequestration and hemodilution. Leukopenia is caused by increased destruction or sequestration of leukocytes. Sequestration may also play a role in the genesis of neutropenia. A complete laboratory study must be done in patients without evident cause of splenomegaly. Imaging studies are useful to assess the splenic enlargement and identify the cause of the splenomegaly.

be a useful tool to increase the diagnostic confidence if splenosis is suspected. 99m-Technetium sulfur colloid scintigraphy and 99m-technetium-labeled heat-damaged red blood cells represent the most specific and the most sensitive imaging techniques to confirm the presence of functioning splenic tissue. Splenosis may cause recurrence of hematologic disorders after therapeutic splenectomy. ▶Congenital Abnormalities, Splenic

Split Pleura Sign CT finding suggestive of an empyema which refers to enhancement and thickening of the visceral and parietal pleura separated by pleural fluid. ▶Pleural Effusion

Split-Liver Transplant Bibliography 1. 2. 3.

4.

Andrews MW (2000) Ultrasound of the spleen. World J Surg 24:183–187 Robertson F, Leander P, Ekberg O (2001) Radiology of the spleen. Eur Radiol 11:80–95 Semelka RC, Marcos HB (1998) Spleen. In: Heuck A and Reiser M (eds) Abdominal and Pelvis MRI. Springer, Berlin-Heidelberg, pp 51–58 Sagoh T, Itoh K, Togashi K et al (1989) Gamna-Gandy bodies of the spleen: evaluation with MR imaging. Radiology 172:685–687

Splenosis Splenosis results from the autotransplantation of splenic tissue occurring after splenic trauma or after splenectomy. The incidence of splenosis after traumatic injury of the spleen varies from 27% to 67%. Splenic implants are usually numerous and are spread throughout the peritoneal cavity. In patients with history of thoraco-abdominal trauma with splenic and diaphragmatic injury, splenosis may be intrathoracic. Splenic implants derive their blood supply from the surrounding tissue and do not have a capsule. Splenosis is usually an incidental finding at imaging, and may mimic peritoneal metastases, lymphomegalies and other neoplastic or nonneoplastic masses. Splenic implants appear as round or oval, sharp-marginated nodules resembling the features of splenic parenchyma in the different imaging modalities (US, CT, or MR). The characteristic inhomogeneity in the arterial phase is very typical. MR reticuloendothelial-targeted contrast media can

▶Transplantation, Liver

Split-liver Transplantation Liver transplantation technique consists in cadaveric liver division so that the lateral segment of the left lobe may be transplanted into a pediatric patient and the remainder of the liver may be transplanted into an adult. ▶Transplantation, Hepatic

Spondylarthrosis ▶Degenerative Conditions, Spine

Spondylitis Marginalis Spondylitis marginalis is a subdiscal osteitis of the posterior upper corner of the vertebral body with triangular sclerosis resembling an inverse Romanus lesion. ▶Spondyloarthropathies, Seronegative

Spondyloarthropathies, Seronegative



Sacroiliitis begins with hypertrophic synovitis with focal mild intima hyperplasia. Granulation tissue and pannus formation lead to cartilage destruction, reflecting the character of the sacroiliac joints as partly synovial joints. Subchondral inflammation progresses to bony erosion, typically with surrounding reactive bony sclerosis. The bone marrow is filled with inflammatory cells. Later, fibroblast proliferation anticipates fibrous scar and bone formation. Remodeling to mature bone advances toward ankylosis. Reactive sclerosis and bone formation in and following inflammation are typical.

G M L INGG , C S CHORN Department of Radiology, Sana Rheumazentrum Rheinland Pfalz, Bad Kreuznach, Germany [email protected]

Definition Seronegative ▶spondyloarthropathy comprises of the following five diseases: ankylosing spondylitis, psoriatic arthritis, reactive arthritis, enteropathic arthritis in Crohn’s disease, and ulcerating colitis, as well as undifferentiated spondyloarthropathy. SAPHO syndrome may show overlap and ▶SAPHO syndrome. These are systemic autoimmune inflammatory disorders of unknown etiology, promoted by infection and primarily affecting the discovertebral complex, synovial membranes, articular surfaces, insertion of joint capsules, and tendino-osseous junctions. Apart from joint destruction, they are characterized by new bone formation as a highly characteristic feature, with consecutive ankylosis and vertebral column stiffness. There is a hereditary predisposition in patients with HLA B 27 (Table 1).

Spondyloarthropathies, Seronegative. Table 1 European spondyloarthropathy study group (ESSG) classification of seronegative spondyloarthropathies. (≥1 clinical sign; ≥1 anamnestic clue = > sensitivity 77%; specificity 89%; + radiological sign = > sensitivity 86%; specificity 87%) Clinical signs Inflammatory back pain Asymmetrical synovitis of the lower extremity

Anamnestic clues Family history

1733

Radiological sign Sacroiliitis

Enteropathy (Crohn’s disease or ulcerosing colitis) Psoriasis Alternating bottom pain Enthesitis

Source: Dougados M, van der Linden S, Juhlin R et al (1991) The European Spondyloarthropathy Study Group: preliminary criteria for the classification of spondyloarthropathy. Arthritis Rheum 34: 1218–1227.

Clinical Presentation Ankylosing Spondylitis Ankylosing spondylitis is the prototypical form of the spondyloarthropathies. Because sacroiliitis is the main manifestation, the best diagnostic clue is long-standing low back pain. Typically in inflammatory back pain there is relief not with rest but with exercise. Patients suffer most seriously in the early morning hours, often waking up with low back pain and feeling relief when walking around. Even before bony ankylosis is present, stiffness is a common symptom. Motion restriction of the lumbar spine, neck, hip, and thorax worsens in the course of the disease. Deformity with severe kyphosis may badly handicap patients. Among the extraskeletal manifestations, iritis is most common. Patients often complain about fatigue. Ninety percent show HLA B 27, whereas only 6–9% of the normal population do.

Imaging The primary imaging tool for inflammatory back pain is plain X-ray of the lumbar spine, despite the fact that early sacroiliac arthritis may not be conspicuous at all. Whenever X-ray imaging is equivocal, further evaluation with computed tomography (CT) or magnetic resonance (MR) imaging should follow. CT is preferred in elderly patients with long-standing symptoms and in cases of known paraarticular sclerosis because bony sclerosis leads to blurring of joint contours in MR imaging, and bony destructions are best viewed by means of a bone technique like CT. MR imaging, on the other hand, is advantageous in young patients due to the lack of radiation and in early cases because of its high sensitivity to inflammatory activity. Diagnosis of lumbar, thoracic, and cervical spine as well as articular involvement is usually confirmed using X-ray imaging, even when MR imaging shows better sensitivity.

S

1734


Signs and Patterns in Ankylosing Spondylitis ▶Sacroiliacal joint arthritis is the imperative diagnostic clue of ankylosing spondylitis. A typical X-ray sign is the triad of sclerosis, destruction, and ankylosis (Fig. 1). In early stages, sclerosis may be mild, and when present it is extended, woolly, and more pronounced in the iliac bone. Erosions and bony destructions are primarily seen as indistinct and irregular contours. A pearl-string aspect of confluent erosions is typical but not very common. Most often, destruction and bony bridging go alongside. Bony bridges with blurring and disappearance of the joint space are signs of increasing ankylosis. In the end stage, the sacroiliac joint space may be completely obliterated and disappeared; however, persisting bits of joint space are not uncommon (Table 2). The differential diagnosis includes degenerative osteoarthritis, which exhibits bandlike subchondral sclerosis or, especially in obese females, multiparae triangular sclerotic areas particularly in the lower aspect of the iliac bone. Variants of the joint form are common and may simulate defects or irregularities. Destruction and pearlstring aspect occur in hyperparathyroidism. Iliacal and sacral bony sclerosis can be signs of insufficiency fractures in severe osteoporosis and should be considered as a differential in patients with low back pain, sacral sclerosis, and generalized osteoporosis.

In MR imaging, periarticular edema, marrow fat accumulation, sclerosis, and blurring of the joint contours are nonspecific findings. Erosion and destruction of the joint contours and ankylosis as well as joint effusion as signs of arthritis are detected in unenhanced images. Contrast material enhancement can be seen in the joint capsule, erosions, subchondral granulation tissue, and areas of edema as well as intra-articularly in pannus formation (Fig. 1). Spine involvement: The typical feature of spine involvement in ankylosing spondylitis is the ▶syndesmophyte, which grows marginally from the vertebral body corner in the exact position of the annulus fibrosus (Fig. 2). Sometimes, in degenerated disks, it is modified, with a more bulging contour. Often it begins consecutively in ▶Romanus lesions commonly in the thoracolumbar junction segments. Finally, the disc is completely bridged

Spondyloarthropathies, Seronegative. Table 2 classification of sacroiliitis ˚0 ˚1 ˚2 ˚3 ˚4

X-ray

Normal Suspicious Minimal arthritis Moderate arthritis Ankylosis

Spondyloarthropathies, Seronegative. Figure 1 Ankylosing spondylitis. (a) X-ray imaging shows paraarticular sclerosis and erosions of the sacroiliacal joint. (b) Computed tomography reveals erosions and iliacal sclerosis as well as bony bridging. (c) Magnetic resonance imaging in another patient allows early diagnosis because of its sensitivity to inflammatory processes. The so-called bone marrow edema is seen mainly in the iliac bone paraarticularly; here on short TR TE images, it is mildly hypointense. (d) Marked hypointensity, however, is a sign of sclerosis. Erosions are visible. Contrast material enhancement is found in the areas of bone marrow edema, in the joint space itself, and in the erosions.


1735

Spondyloarthropathies, Seronegative. Figure 2 Ankylosing spondylitis. (a) Anteroposterior view of the lumbar spine of a patient with long-standing ankylosing spondylitis shows the typical cane of bamboo with multisegmental syndesmophytic bridging of the disc spaces and the ▶Trolley-truck sign referring to the polysegmental bony ankylosis of the ▶apophyseal joints. The sacroiliacal joints are completely postarthritically ankylosed, with only mild residual sclerosis. (b) Lateral view of the same patient shows the syndesmophytes growing vertically in the location of the annulus fibrosus and the broad ossification along the apophyseal joints. (c) In another patient, the lateral view reveals multisegmental ▶spondylitis marginalis, in the anterior location the so-called Romanus lesions, and in the posterior location the characteristic triangular sclerosis of the vertebral corner. ▶Square vertebrae and the faint barrel shape of the vertebrae are other typical features.

by circular ossification of the annulus fibrosus. When multiple segments are affected, the spine looks like a ▶cane of bamboo. Osteitis of the peridiscal bone (discitis, rheumatic spondylodiscitis) is another typical aspect of inflammation in ankylosing spondylitis. It may be limited to the anterior upper corner and is then called a Romanus lesion (Fig. 2c). A Romanus lesion consists of a triangular sclerosis and an erosion of the anterior upper vertebral corner. It usually heals within months or years with syndesmophyte formation. When it is minimal and only sclerosis is present, it is referred to by the term “▶Shinig corner.” Similar lesions exist in the posterior upper corner of the vertebra and are named spondylitis marginalis. The inflammation of the central part of the disc results in an aspect resembling spondylitis, with erosions or destructions of the subchondral bone. It is called ▶Andersson lesion type A or inflammatory type. In comparison to bacterial spondylitis, the destruction remains mild, focal, and unchanged for months or even years. Perivertebral

fluid or abscesses are not an aspect of rheumatic disc disease. In the late stage, bony ankylosis is present. ▶Andersson lesion type B refers to a transdiscal insufficiency fracture. It occurs mostly in the thoracolumbar junction segments of osteoporotic multisegmentally ankylotic vertebral columns with marked kyphosis. Sometimes a minimal trauma is the cause. The prognosis for local control is poor. Square and barrel-shaped vertebrae are the result of inflammatory and osteoproliferative affection of the ventral vertebral aspect. They typically occur in the thoracolumbar vertebral column. ▶Apophyseal joint arthritis with progression to fibrous or bony ankylosis and consecutive early stiffness is very common in adolescent patients. X-ray images, however, are sometimes not very conspicuous. In late stages, bands of broad ossification over the dorsolateral aspect of the vertebral column are seen. ▶Arthritis of the costotransversal and costovertebral joints gives rise to persistent thoracolumbar pain and is the reason for respiratory

S

1736


movement restriction. Most often it is occult on X-ray images. MR imaging, however, can show extensive contrast material enhancement in these locations. Ligament ossification such as of the interspinal and iliolumbar ligaments occurs in late stages of ankylosing spondylitis, some of which are specially named: “▶Dagger sign” is the polysegmental ossification of the interspinal ligaments. If occurring along with extended bridging ossification of the intervertebral joint capsules, the term “trolley-truck sign” is used. In ankylosing spondylitis, the ▶large joints are commonly affected: hips, knees, and shoulders. Joint effusion, cartilage destruction with consecutive concentric joint space narrowing, and paraarticular demineralization are the radiologic signs of arthritis. In ankylosing spondylitis, frank bony destructions are rarely found. Inflammatory proliferation at ligament and tendinous insertions is more common. Sometimes, premature degenerative disease is the only sign of postarthritic change. The ▶small joints are quite rarely affected in ankylosing spondylitis. Extraarticular inflammatory proliferation is seen in addition to arthritic destruction. In comparison to rheumatoid arthritis, the lower extremities and the DIP joints are more often affected. Postarthritic degenerative osteoarthritis is more common than ankylosis. Bursitis compromising the underlying bone is most often seen in bursa subachillea, bursa trochanterica, and iliopsoas bursa. It appears as soft tissue swelling in the typical places. Pressure erosions of the bone and inflammatory destructions are both possible. Osteoproliferative changes are common. In ankylosing spondylitis, an inflammatory reaction of tendon and ligament insertions is a leading feature. Common locations are the iliac crest, tubera ischiadica, greater trochanter, plantar fascia (calcaneopathy), and olecranon, but it can appear anywhere. Proliferative changes with indistinct, hairy contours as well as destructions with small grooves or combinations of both are possible features. ▶Synchondritis of the symphysis or manubriosternal junction exhibits contour defects or broad indistinct defects and surrounding bony sclerosis.

Nuclear Medicine In bone scanning, inflammatory affection appears as hot spots. Sacroiliacal joints, however, normally show mild tracer accumulation. Asymmetric massive tracer accumulation may be a sign of arthritis, but specificity is very low. Radiologic work-up is mandatory for the differential diagnosis. Bone scanning can show foci of inflammatory affection all over the skeleton, and therefore can direct the following work-up to these foci.

Diagnosis Since 1966 standardized diagnostic criteria have been employed. The most recent modification of the New York criteria dates from 1984 and provides high specificity and moderate sensitivity. One or more clinical signs and radiological evidence of sacroiliitis must be present for proper diagnosis (Table 3). Early diagnosis, however, may be difficult. MR imaging is one tool to close the diagnostic gap, since sacroiliitis can be found much earlier and with more conspicuity than in X-ray imaging.

Interventional Radiological Treatment As in rheumatoid arthritis, radiosynovectomy is a promising tool for local control in limited disease of the large joints. CT-guided corticosteroid injection into the sacroiliac joints with satisfactory results is used by several authors.

Clinical Presentation Psoriatic Arthritis Psoriatic arthritis occurs in 5–10% of patients suffering from psoriasis. Spondyloarthropathy is found in 25% of patients with psoriatic arthritis and is highly associated with HLA B 27. Inflammatory back pain is the main symptom in sacroiliitis. Lumbar spine involvement is characterized by pain, morning stiffness, and motion restriction. Kyphosis or total vertebral stiffness as seen in ankylosing spondylitis is not a sign of psoriatic spondyloarthropathy.

Imaging Signs and Patterns in Psoriatic Arthritis Peripheral arthritis of the small joints is common, but not symmetrical in most patients. Irregular involvement is

Spondyloarthropathies, Seronegative. Table 3 Modified New York criteria (1984) for diagnosing ankylosing spondylitis (≥1 clinical sign and 1 radiological sign) Clinical signs Low back pain and stiffness >3 months, no pain relief with rest but relief with exercise Motion restriction of the lumbar spine sagittally and frontally Respiratory motion restriction (age-related, about 80%) or obstruction at the prevertebral part of the subclavian artery or at the origin of the innominate artery that results in retrograde flow in the ipsilateral vertebral artery. This is primarily an imaging finding that maybe accompanied by clinical symptoms in one third of the patients (1).

Pathology The flow reversal in the vertebral artery can be either permanent, the upper extremity being perfused by the ipsilateral vertebral artery (permanent vertebral steal), or induced after exercise or hyperemia of the ipsilateral arm (latent or temporary steal) (Fig. 1). It has been shown that decreased regional cerebral flow is present in patients with vertebral steal. Other types of vascular steal may be encountered: ▶carotid artery steal in the presence of innominate artery occlusive lesion; or in patients with internal mammary artery to coronary artery by pass graft, the presence of a proximal subclavian occlusion may cause ▶coronary artery steal phenomenon (Fig. 2).


Staging The local extent and distant spread of a tumour is classified by using a combination of primary tumour characteristics, nodal status and metastatic disease (TNM). The stage of a tumour will predict the probable prognosis as well as optimising treatment requirements. ▶Neoplasms, Gastroduodenal

Steal Syndrome Vertebral E LIAS N. B ROUNTZOS Department of Radiology, Athens University, Medical School, Greece [email protected]

Symptoms of this syndrome include those of upper limb ischemia, usually presenting as claudication, vertebrobasilar insufficiency (visual disturbances, dizziness, drop attack, ataxia, vertigo, syncope), and anterior circulation transient ischemic attacks. Many patients (35%–85%) with subclavian or innominate artery lesion associated vertebral steal syndrome have also concomitant carotid or vertebral artery occlusive disease, so it is not always possible to attribute their cerebral symptoms to the steal syndrome. Patients with prior internal mammary to coronary artery bypass graft and associated subclavian steal may present with angina.

Imaging Duplex or color Doppler sonography: Flow reversal or absence of flow may be depicted with accuracy using these modalities. Compression of the ipsilateral brachial artery is accompanied with clear change in the direction of the

S

1740

Steal Syndrome Vertebral

Steal Syndrome Vertebral. Figure 1 (a) arch DSA in the LAO projection in a patient complaining of dizziness during arm exercise, depicts occlusion of the left subclavian artery. (b) arch DSA in the LAO projection of the same patient in a later phase depicts reconstitution of the left subclavian artery from retrograde flow in the left vertebral artery.

Steal Syndrome Vertebral. Figure 2 (a) Selective DSA of the left subclavian artery in a patient presenting with angina 2 years after left internal mammary artery (LIMA) to coronary artery bypass graft depicts severe stenosis at the origin of the subclavian artery. (b) Selective DSA of the LIMA depicts good patency of the graft, with minimal coronary artery atherosclerotic disease. (c) Arch DSA after placement of a 9-mm balloon expandable stent depicts restoration of the flow in the subclavian artery. Ischemic cardiac symptoms of the patient were eliminated.

flow (2). In patients with the temporary type of vertebral steal syndrome, flow reversal is depicted during exercise or hyperemia of the ipsilateral upper extremity. Diagnostic accuracy is 100% (2).

MRA: The cervical vessels are ideal for MRA. Several techniques have been used to demonstrate the flow reversal in the vertebral artery. These include twodimensional time-of-flight (2-D TOF) with a selective

Steatosis, Hepatic

presaturation band applied first above and then below the volume of interest. 2-D phase-contrast MRA demonstrates reversal of flow in the superior to inferior direction. 3-D gadolinium-enhanced MRA is excellent in imaging the vertebral arteries, but cannot be used to detect flow direction. Recently it has been suggested that vertebral steal may be suspected when one observes differences in the time of peak enhancement in the vertebral arteries during a bolus-timing examination performed in the setting of a dynamic 3-D MRA (3). Conventional angiography: Aortic arch DSA with a pigtail catheter placed in the ascending aorta is the gold standard for the depiction of the flow reversal in the vertebral artery, while the proximal arterial lesion can be evaluated in detail. Interventional radiological treatment: Percutaneous angioplasty with or without stent placement has evolved as an alternative to the standard surgical management. Only symptomatic patients should be treated. Femoral artery approach is more commonly used; but when dealing with complete occlusions, a brachial artery approach may be necessary to cross the obstruction with the guidewire. Balloon expandable stents are recommended because they can be accurately placed. The reversed flow in the ipsilateral vertebral artery has been shown to provide protection against stroke complication.

References 1. 2. 3.

Fisher CM (1961) A new vascular syndrome—“the subclavian steal”. N England J Med 265:912–913 Hennerici M, Neuerburg-Heusler D (eds) (1998) Vertebral Artery System G Thieme Verlag, Stuttgart pp 79–84 Wu C, Zhang J, Ladner CJ et al (2005) Subclavian steal syndrome: diagnosis with perfusion metrics from contrast-enhanced MR angiographic bolus-timing examination- initial experience. Radiology 235:927–933

Steatosis, Hepatic M ASSIMO B AZZOCCHI , G IUSEPPE C OMO, C HIARA F RANCHINI Universita` dezli Studi di Udine, Udine, Italy [email protected]

Synonyms Fatty change of the liver; Fatty degeneration of the liver; Fatty infiltration of the liver; Fatty liver

Definition Accumulation of fat in tiny sacs within the hepatocytes.

1741

Pathology and Histopathology Hepatic steatosis is a common condition which represents a non-specific response to many metabolic disorders and toxic insults. There are many causes of fat accumulation in the liver. Non-alcoholic fatty liver disease (NAFLD) is considered the primary fatty liver disease. The most common risk factors for NAFLD are obesity, diabetes, and elevated cholesterol blood levels. The secondary fatty liver diseases include alcoholic liver disease (ALD), hepatic steatosis occurring in chronic viral hepatitis C (HCV), chronic viral hepatitis B (HBV), chronic autoimmune hepatitis (AIH) and Wilson’s disease. In all of these secondary fatty liver diseases, fat accumulation is associated with other abnormalities and is thought to result from liver cell injury. Another type of secondary fatty liver disease is unrelated to other liver diseases, but is due to metabolic disorders. Such causes include certain drugs, some gastrointestinal surgical interventions, malnutrition, parenteral nutrition and metabolic genetic diseases. Simple fatty liver is a benign condition and it occurs due to the accumulation of fat in the liver cells without flogosis or fibrosis. The fat is composed of triglycerides that accumulate in tiny sacs within the liver cells. Simple fatty liver is rapidly reversible after removing the causing factor. Only a minority of patients with simple fatty liver will develop the so-called ‘steatohepatitis’. In this entity, fatty infiltration of the liver is associated with inflammation. The inflammatory cells can destroy the hepatocytes leading to necrosis (steatonecrosis). Necrosis is followed by fibrotic phenomena, while the last, irreversible stage is cirrhosis. Hepatic steatosis may present with a diffuse, subtotal, segmental or focal distribution. In diffuse fatty infiltration, the liver may be enlarged and have smooth margins. In non-diffuse steatosis, the fat may have an anatomic distribution (segmental steatosis) or a nonanatomic distribution (focal steatosis). The focal distribution of fat is related to local differences in portal perfusion; areas with lower portal supply tend to accumulate less fat. Some areas are commonly spared by fatty infiltration, such as the dorsal portion of the IV segment just anterior to the portal vein, the gallbladder bed and the medial portion of the left lobe, along the great interlobar fissure. In these regions, the portal flow is decreased due to anomalous systemic venous drainage into portal venules from structure such as the gallbladder or the stomach. In other regions of the liver, focal sparing in fatty infiltration may reflect segmental portal vein obstruction (1).

Clinical Presentation Simple steatosis and steatohepatitis give very slight, aspecific symptoms. Actually, most patients are completely asymptomatic, but they may occasionally have vague

S

1742

Steatosis, Hepatic

right upper-quadrant abdominal pain. The symptoms and signs of liver failure include jaundice, severe weakness, loss of appetite, nausea, vomiting.

Imaging In case of steatosis, the typical finding at ultrasound (US) is the so called ‘▶bright liver’, occurring due to a diffuse increase in echogenicity of the hepatic parenchyma, characterised by the presence of numerous, fine, tightly packed echoes of high signal intensity uniformly distributed throughout the liver parenchyma (Fig. 1). The liver may be enlarged with smooth margins. When the hyperechogenicity is particularly strong, a significant attenuation of the signal may be observed. In diffuse forms the increase of echogenicity is homogeneously distributed in the hepatic parenchyma whereas in nondiffuse forms some areas are not interested by the fatty infiltration. These islands of normal liver tissue are called ▶‘skip areas’ or ‘spared areas’ and appear as relatively hypoechoic areas. Typically, the skip areas are found in the gallbladder fossa, in the IV segment anteriorly to the portal vein or, in individuals with anomalous venous drainage through the gastric veins, in the medial portion of the left lobe, along the great interlobar fissure. In focal steatosis, hyperechoic areas of various sizes with a ‘geographic’ distribution are depicted in a normally

echogenic liver. In rare cases, only the portion of the IV segment just adjacent to the portal vein, is involved by steatosis. This ▶focal fatty infiltration appears as a hyperechoic area (2). Focal fat and spared areas can mimic tumour lesions, while heterogeneous fat can sometimes be difficult to distinguish from diffuse malignant diseases. At Doppler US, the finding of normal hepatic vasculature, without any dislocation or interruption, may suggest the diagnosis of steatosis (Fig. 2a). At contrast-enhanced US, skip areas and focal fatty changes become homogeneous and isoechoic with respect to the surrounding parenchyma, having the same vascularisation as the normal liver. The increased echogenicity of the hepatic parenchyma can modify the relative echogenicity of other focal lesions. A quite common finding is the presence of haemangiomas showing a relative hypoechoic appearance over a hyperechoic liver. Furthermore, haemangiomas detected at US before the onset of steatosis may not be visibile or have a reduced detectability in a hyperechoic background. In contrast, hypoechoic lesions, such as metastases, become more clearly depicted, due to the high contrast with the surrounding parenchyma (3). Computed tomography (CT) is less sensitive, but more specific than US in diagnosing hepatic steatosis. On unenhanced CT scans, the hepatic parenchyma has an attenuation value significantly lower than the spleen and the expected normal liver (normal values 45–65 HU), sometimes showing a density close to the water (0–15 HU).

Steatosis, Hepatic. Figure 1 In case of steatosis the typical finding at ultrasound (US) is the so-called ‘bright liver’, occurring due to a diffuse increase in echogenicity of the hepatic parenchyma, characterised by the presence of numerous, fine, tightly packed echoes of high signal intensity uniformly distributed throughout the liver parenchyma. We can see the hyperechogenicity of the liver if compared to the echogenicity of renal parenchyma (on the left) and of spleen parenchyma (on the right).

Steatosis, Hepatic

1743

Steatosis, Hepatic. Figure 3 On unenhanced CT scans, the hepatic parenchyma has an attenuation value significantly lower than the spleen and the expected normal liver (normal values 45–65 HU), sometimes showing a density close to the water (0–15 HU). Peculiar finding in steatosis is the clear evidence of hepatic vessels on unenhanced scans which appear relatively hyperdense compared to the fatty liver (contrast reversal).

Steatosis, Hepatic. Figure 2 Segmental or focal forms of steatosis: the hepatic parenchyma appears inhomogeneous with areas of hyperechogenicity and normal echogenicity in the US study (a) and areas of hypodensity and areas of normal density in the CT scan (b). Distinctive features of focal fatty infiltration are non-spherical shape, absence of mass effect, normal vascular architecture, segmental or subsegmental distribution.

A reduction in density at unenhanced CT scans is also observed in amyloidosis and glycogen storage disease. Another peculiar finding in steatosis is the clear evidence of hepatic vessels on unenhanced scans which appear relatively hyperdense compared to the fatty liver (▶contrast reversal) (Fig. 3). In segmental and focal forms, the hepatic parenchyma appears inhomogeneous with areas of hypodensity and areas of normal density. Distinctive features of focal fatty infiltration are non-spherical shape, absence of mass effect, normal vascular architecture, segmental or subsegmental distribution (4) (Fig. 2b). After contrast medium administration, the difference in density between normal and fatty liver does not change. Conventional MR SE sequences have not been proven to be useful in detecting fatty infiltration of the liver on the basis of signal intensity and relaxation times. Actually,

hepatocytes with fatty degeneration are usually mixed with normal hepatocytes within the voxel volume, thus preventing substantial changes in signal intensity. Only when the concentration of fatty cells is very high, there is a brightening of fat-containing areas on T1-weighted images, whereas T2-weighted images usually do not show any change in signal intensity. The hyperintensity of the fatty liver on T1-weighted scans creates a marked contrast with hypointense lesions, improving their detection. The low sensitivity of T2-weighted scans in depicting fatty changes can be useful to exclude the presence of a focal liver lesion in case of focal steatosis. Grading fatty infiltration of the liver is not possible using conventional MR because of its insensitivity to fat. MR techniques based on the frequency difference between water and fat can be successfully used for this purpose. These techniques rely on the difference in resonance frequency between water protons and the methyl and methylene protons of fatty acid chains (chemical shift phenomenon). Two populations of protons (water and fat) exist in fatty livers. A specific method for the detection of the two populations of protons has been developed (Dixon’s method). It consists in the acquisition of two images in which water and fat protons are ‘in-phase’ and ‘out-of-phase’. In tissues where fat and water protons populations are equally represented, as in fatty liver, a significant signal loss is seen on the out-of-phase scans (Fig. 4a, b). By using this method, areas of focal fatty infiltration may be diagnosed due to the lower signal

S

1744

Stenosis

Steatosis, Hepatic. Figure 4 MR scan in-phase and out-of-phase of fatty liver: a specific method for the detection of the two entity of fatty infiltration of the liver has been developed (Dixon’s method). It consists in the acquisition of two images with two different TE in which water and fat protons are in the first ‘in-phase’ (a) and in the second ‘out-of-phase’ (b). In tissues where fat and water proton populations are equally represented, as in fatty liver, a significant signal loss is seen on the out-of-phase scans.

intensity compared to the surrounding liver in the out-ofphase images. In these images focal liver lesions in a fatty liver show a high contrast and their detection is increased (1). Fat-suppressed images may be also used, but they have a lower sensitivity for fatty infiltration compared to the out-of-phase sequences. After administration of hepato-specific MR contrast agents, fatty and normal hepatic parenchyma show similar contrast uptake, becoming isointense. This finding can rule out malignancy in case of focal fatty infiltration and skip areas.

Nuclear Medicine Tc99m-colloid scintigraphy can differentiate focal liver lesions from focal fatty sparing and focal fatty infiltration: in fatty liver, the colloid uptake by Kupffer cells may vary depending on the degree of steatosis. A ▶spared area shows a greater uptake than the surrounding liver whereas a focal fatty infiltration has a variable uptake. Malignant focal liver lesions have no uptake.

Diagnosis US is the most sensitive imaging technique in the diagnosis of steatosis and it is usually the only examination to be performed in these cases, due to the low clinical relevance of this affection. Anyway its specificity is limited, because it cannot distinguish diffuse infiltration from hepatic fibrosis. Furthermore, in case of focal steatosis or spared areas in atypical locations, US alone

may not be able to exclude the presence of a focal liver lesion. The use of Doppler US and US contrast media can be helpful by demonstrating the presence of normal vascular architecture within the focal alterations. CT is highly specific in diagnosing fatty infiltration of the liver, although it is not as sensitive as US. In those cases in which CT findings do not allow to reach a differential diagnosis, particularly in segmental or focal steatosis, MR can give a definite characterisation. In particular, the in-phase and out-of-phase sequences are very useful in assessing fatty infiltration of the liver and the use of hepato-specific MR contrast agents allow to rule out malignancy in case of focal steatosis and spared areas.

Bibliography 1. 2.

3. 4.

Lencioni R, Bartolozzi C (2003) Diffuse liver disease. In: Magnetic Resonance Imaging in Liver Disease. Thieme, New York Rubaltelli L, Savastano S, Khadivi Y et al (2002) Targetlike appearance of pseudotumors in segment IV of the liver on sonography Am J Roentgenol 178(1):75–77 Hisham T, Philip WR, Randall R et al (2002) Sonography of diffuse liver disease. J Ultrasound Med 21(9):1023–1032 Mergo PJ, Ros PR (1998) Imaging of diffuse liver disease. Radiol Clin North Am 36(2):365–375

Stenosis ▶GI Tract, Pediatric, Congenital Malformations

Stenosis, Aortic, Abdominal


1745

angina – buttock pain, and erectile dysfunction in males. In occlusions, acute bilateral ischemia may occur if no preexisting stenotic process has promoted earlier development of collateral pathways.

D IERK VORWERK Klinikum Ingolstadt GmbH, Direktor des Institutes fu¨r diagnostische und interventionelle Radiologie, Ingolstadt, Germany [email protected]

Synonyms Aortic obstruction; Aortic syndrome

Definition Aortic stenosis usually occurs in the infrarenal segment of the abdominal aorta. If stenotic, it leads to bilateral claudication of both legs including the upper thigh muscles. Bilateral absence or weakness of femoral pulses is usually found.

Imaging Many imaging modalities can be used to make the diagnosis of aortic obstruction. Color-coded duplex sonography, magnetic resonance angiography (MRA), computed tomography (CT), CT angiography (CTA), and intraarterial angiography are useful tools for detecting the location and extent of an aortic obstruction. Due to access limitations, angiography is preferably performed via a transbrachial access if the clinical findings suggest an aortic obstruction.

Nuclear Medicine Nuclear medicine plays no particular role in the diagnosis of aortic stenosis.

Pathology/Histopathology In more than 90% of cases, the cause for infrarenal aortic obstruction is due to atherosclerotic disease (2). Clinically, simple infrarenal aortic stenoses with no relation to the aortic bifurcation are infrequently found, but a stenosis of both the very distal aortic segment and the common iliac arteries is more frequent. This may be complicated by an acute or subacute thrombosis of the aortic bifurcation known as Leriche’s syndrome. A small distal aortic caliber, especially in female patients, may be a predisposing factor. Atherosclerotic stenosis above the orifice of the inferior mesenteric artery is rare (2). Embolic occlusion of the distal aorta is, on the contrary, exceptional, but may occur in patients with arrhythmias. Other rare causes of aortic obstruction are fibromuscular dysplasia, Takayasu’s disease, and retroperitoneal fibrosis. The typical age of patients with aortic obstruction is between 40 and 70 years. In aortic occlusions, 55% are located at the level of the aortic bifurcation, 8% involve the complete infrarenal segment, and 37% involve aortic segments alone (2). Collateral pathways are manifold via lumbar, epigastric, and mesenteric arteries.

Clinical Presentation Clinical symptoms in aortic obstruction are bilateral claudication predominantly with upper thigh symptoms – which may be similar to findings in spinal

Diagnosis Diagnosis is reliably achieved by angiography, CT, CTA, and MRA. Detection by duplex sonography is sometimes limited due to unfavorable anatomic conditions

Interventional Radiological Treatment More and more, aortic aneurysmal disease is becoming a field of interest for interventional radiologists and may be combined with atherosclerotic obstructive disease.

Indications for Percutaneous Treatment Indications for percutaneous versus surgical treatment largely depend on the location, extension, acuteness, and type of obstruction. Accepted indications for sole balloon angioplasty are: 1. concentric segmental stenosis 2. short-segment aortic bifurcational stenosis Balloon angioplasty may be followed by stent insertion in the case of insufficient luminal gain after adequate balloon angioplasty or the occurrence of significant dissection after percutaneous transluminal angioplasty (PTA). Balloon angioplasty is contraindicated if a complete calcified ring is present at the site of obstruction, since aortic rupture has been occasionally reported under these circumstances (3). According to Laplace’s law, the aorta is

S

1746


theoretically more easily prone to rupture than smallerdiameter vessels such as the iliac artery. In these cases, primary placement of a stent graft can be considered. Long-segment diffuse disease of both the aorta and iliac arteries may be considered as a contraindication, while surgical aortobifemoral bypass grafts might be a better choice. Bifurcational aortic stenosis may also be treated by balloon angioplasty using a simultaneous “kissing balloon” technique in order to dilate the distal aortic segment and both iliac orifices. Stent implantation has been increasingly used to achieve a stable postangioplasty widening by use of kissing stents in the distal aorta and both iliac arteries. In the case of distal aortic occlusion, few reports exist on remodeling the distal aortic segment or the aortic bifurcation by use of metallic stents (4, 5). In these difficult situations, use of advanced interventional techniques over simple balloon angioplasty is certainly of advantage.

Technical Aspects There are no major differences between the aortic and the iliac segment concerning techniques of lesion passage or traversal of occluded segments, whether they are located purely in the aortic segment or also involve the iliac segment. This is also true for balloon dilation, which does not differ significantly from angioplasty elsewhere. A large diameter of the aortic lumen, however, is a particular problem. Kissing balloon technique: Until recently, a major difficulty was the lack of suitable balloons with sufficient diameters of 16–20 mm. Thus, a double or triple balloon technique was recommended to open the aortic stenosis to a sufficiently large diameter. The aortic lumen can be widened to its original diameter with two or three kissing balloons that are inserted by a bifemoral and an additional transbrachial approach and are inflated simultaneously. A kissing technique is still recommendable for dilatation of bilateral stenosis of the aortic bifurcation and the very distal aortic segment close to the bifurcation that allows remodeling of the aortic bifurcation. Single balloon technique: Recently, large-diameter balloons, from 16 to 25 mm, have become available (Boston Scientific, Cordis), allowing the use of a single balloon technique by a unifemoral approach. With these balloons, it is strictly necessary to locate the balloon within the aortic lumen not overriding the bifurcation in order to avoid overdilatation and rupture of the proximal iliac segment. Stent insertion into the infrarenal aortic segment obeys the same rules as elsewhere. Use of a stent of appropriate size, at least 14–16 mm, is necessary to avoid

undersizing. The largest stent diameter can be achieved by use of the balloon-expandable Palmaz XXL stent (Johnson and Johnson) that can be mounted on a large balloon and inflated up to 25 mm in diameter. Single stent technique: In lesions without involvement of the aortic bifurcation, a single stent can be implanted with no specific technical requirements. The type of stent that should be used depends on the experience of the interventionalist. This is particularly true if the lesion is of considerable distance from the bifurcation. Depending on the length of the stent or the location of the lesion, overstenting of the inferior mesenteric artery may be unavoidable. However, if the lesion ends very close to the bifurcation, placement of a single stent, especially of a balloon-expandable stent, may become difficult without overdilation of an iliac orifice. Under these circumstances, use of a self-expanding stent may be advantageous, while the aortic bifurcation is protected by a cross-over catheter inserted via a contralateral approach. An alternative technique in very distal aortic lesions or bifurcational lesions is use of kissing stents. Kissing stent technique: In analogy to the kissing balloon method, stents of preferably an identical diameter and length are placed in kissing fashion within the distal aorta with its distal end in the common iliac artery. Very frequently, the stents tend to meet the opposite aortic wall. Thus – instead of being shaped in a kissing fashion – they cross each other forming a mirror-sided artificial iliac orifice. There are not many reports on this technique and some questions remain open. Especially in stenotic lesions, it is not yet known whether there are potential sequelae from using two open stents that will remain partly nonendothelialized in their aortic portion and may lead to embolic disease or a higher tendency for thrombosis. Whether covered stents are advantageous in these cases is still under discussion. Use of noncovered stents is less problematic in distal aortic occlusive disease. If a kissing stent technique is applied, it is mandatory that the proximal ends of both stents are exactly parallel to avoid one stent compromising the inflow into the other. For this reason, use of noncompressible stents such as the Palmaz stent or the new Perflex stent (both Johnson and Johnson), which are both balloon-expandable, seems of benefit.

Results As reported in the literature, aortic PTA has an excellent outcome compared to other PTA sites. A primary success rate of 95% and a cumulative patency of 98% after 1 year and 80% after 5 years have been reported from different series (6).

Stenosis, Artery, Renal

Single cases of aortic stenting in stenoses have been reported, usually associated with an excellent outcome. Long reported on two cases of successful stenting in Leriche’s syndrome (7), and Dietrich reported on six patients with chronic aortic occlusion who underwent thrombolysis and stenting by use of Palmaz stents (8). The mean patency was about 11 cases with success in all.

Complications Complications that may occur after aortic dilation are not different from those in other vascular areas; however, they might be of major clinical impact. While severe dissection, recollapse, or residual stenoses are simply treatable by additional stent implantation, aortic rupture has been reported rarely but is potentially life threatening, and therefore patients must undergo immediate surgery. To limit the extent of exsanguination, a large occlusion latex balloon (Boston Scientific) should be positioned just below the renal arteries or covering the site of rupture and left inflated until the patient is prepared for surgical repair. To avoid this complication, CT is recommended before the intervention to exclude complete or nearly complete circular calcification of the aortic wall, which is said to be a risk factor for rupture. Theoretically, a covered stent graft may be placed across the site of rupture percutaneously; however, this method has only been reported for iliac arterial rupture and may risk occlusion of major collaterals and the inferior mesenteric artery. Embolization of occlusion material may occur in less than 1% of cases (6). Subacute complications include thrombosis, which has not been reported for pure aortic dilation or stenting, but is a risk in remodeling techniques of aortic bifurcations. In these cases, the patient may be predisposed to thrombosis by adjacent aortic disease with plaques hanging over the stent orifice, thus causing inflow obstruction or by adjacent outflow problems. If a technical reason has caused stent or post-PTA thrombosis, surgery is a reasonable option. If not, thrombolysis may be tried. Thanks to their large diameter, reobstruction of aortic stents occurs rarely; however, patients may undergo repeat balloon dilatation as for iliac stents. In kissing stents, obstruction may be caused by neointimal hyperplasia that may be treated with reballooning, atherectomy, or a second stent.

References 1.

2.

The TASC Working Group Management of peripheral arterial disease (PAD) (2000) Transatlantic inter-society consensus (TASC). J Vasc Surg 31 S1–S296 Vollmar J (1996) Rekonstruktive Chirurgie der Arterien. Stuttgart, Thieme, pp 207–214

3. 4.

5.

6.

7.

8. 9.

1747

Berger T, So¨rensen R, Konrad J (1986) Aortic rupture. A complication of transluminal angioplasty. Am J Roentgenol 146:373–374 Strecker E, Hogan B, Liermann D et al (1993) Iliac and femoropopliteal vascular occlusive disease treated with flexible tantalum stents. Cardiovasc Intervent Radiol 16:158–164 Dietrich EB, Santiago O, Gustafson G (1993) Preliminary observation on the use of the Palmaz stent in the distal portion of the abdominal aorta. Am Heart J 125:490–500 Rholl K, Van Breda A (1994) Percutaneous intervention for aortoiliac disease. In Strandness E, Van Breda A (eds) Vascular Diseases. Churchill Livingstone, New York, pp 433–466 Long A, Gaux J, Raynaud A (1994) Infrarenal aortic stents. Initial clinical experience and angiographic follow-up. Cardiovasc Intervent Radiol 16:203–208 Dietrich EB (1993) Endovascular techniques for abdominal aortic occlusions. Int Angiol 12:270–280 Becker G, Katzen B, Dake M (1989) Noncoronary angioplasty. Radiology 170:921–940

Stenosis, Artery, Iliac In contrast to iliac occlusions, in iliac artery stenosis the arterial lumen is narrowed but not yet completely occluded. ▶Iliac Artery Occlusion

Stenosis, Artery, Renal P HILIPPE O TAL , F. J OFFRE CHU, Toulouse, France [email protected]

Synonym Renal artery stenosis (RAS)

Definition Renal artery stenosis (RAS) is defined by a narrowing of the renal artery trunk or proximal branches by 50% or greater, expressed as a percentage of the diameter of a normal renal artery at the same level. There has been an increase in the incidence of RAS. This is partially due to improved detection imaging techniques such as color Doppler ultrasound (CD-US), computed tomography angiography (CTA), and magnetic resonance angiography (MRA). Another reason is the greater awareness in screening patients with RAS, due to the deleterious clinical consequences in cases of spontaneous evolution

S

1748


and the interesting possibilities of proposed simple endovascular treatment modalities. However, much controversy remains, mostly concerning the benefit-risk ratio of RAS treatment.

Pathology A variety of pathologic lesions may cause RAS, but the main etiologies are atherosclerosis (75–80%) and fibromuscular dysplasia (15–20%). Atherosclerotic RAS is noted in individuals of either gender who are 50 years of age or older, and it is frequently associated with other atherosclerotic peripheral diseases. The lesion is generally proximal and concerns the first centimeter of the trunk, and is frequently related to atheromatous aortic plaques extending into the proximal part of the renal artery. Fibromuscular dysplasia generally occurs in women in the second to fifth decades of life and is mainly located in the medial and distal thirds of the renal artery trunk. Fibromuscular dysplasia represents about 90% of pediatric RAS. Uncommon causes of RAS are vasculitis (Takayasu’s or Buerger’s disease, polyarteritis nodosa), neurofibromatosis, radiation-induced RAS, retroperitoneal fibrosis, aortic dissection, renal artery aneurysm or dissection, and external compression. A special situation is represented by RAS of a transplanted kidney, which occurs in 5–10% of cases (1). An important point to consider is the spontaneous evolution of RAS. The risk of worsening of a fibrodysplastic RAS varies between 10 and 40% according to the pathologic lesion. Concerning the atherosclerotic lesion, progression of RAS to arterial occlusion and atrophy of the kidney is directly correlated to the degree of the stenosis. The viability of the kidney may be totally or partially preserved with collaterals, but its function is likely to be severely impaired and revascularization procedures more difficult to perform. Conversely, the progression of atherosclerotic RAS can be slowed by lifestyle changes and aggressive statin therapy.

Clinical Presentation Significant RAS can be asymptomatic in about 10–15% of cases. However, high-grade RAS is generally responsible for arterial hypertension and renal insufficiency. ▶Renovascular hypertension is the most common cause of secondary hypertension and depends on the renin– angiotensin system. Insufficiently controlled hypertension can lead to heart failure with frequent flush pulmonary edema and/or further deterioration of renal function. RAS can also cause ▶ischemic nephropathy in cases of severe bilateral stenosis or in a single kidney.

Because severe RAS may be present in patients with normal blood pressure and no azotemia, and because it is difficulty to assert that symptoms are related to RAS, the main issue is to identify those patients in whom RAS is responsible for their symptoms and who are amenable to revascularization. Clinical situations suggestive of RAS include: . Onset of hypertension before age 30 years or after 55 years . Severe hypertension in a child of a female patient between 20 and 50 years . Hypertension that is malignant, accelerated, or refractory to appropriate treatment . Unexplained or progressive renal dysfunction . Development of acute or subacute azotemia after therapy with angiotensin-converting enzyme (ACE) inhibitors or angiotensin receptor blockers . Recurrent pulmonary edema . Diffuse atherosclerotic vascular disease . Abdominal bruit . Unilateral unexplained small kidney.

Imaging Different modalities are available for imaging RAS. CD-US has become a leading technique for screening RAS. It allows the following: . Direct visualization of the renal arteries and detection of hemodynamically significant stenosis (Fig. 1). A complete examination of both renal artery trunks can be achieved in 80–90% of cases (2). The main criteria are a peak systolic flow velocity of 180 cm/sec or more and a renal-to-aortic ratio of peak systolic velocity >3.5. The technical limitations and difficulties in detecting accessory renal arteries explain the mixed results, with sensitivity between 50 and 95% for detecting RAS. Other changes are less contributory: turbulent flow in the stenotic area and absence of signal in cases of occlusion. . Analysis of waveforms of renal branches downstream from the lesion. This yields quantitative criteria that improve the global sensitivity of detection. . Measurement of the distal resistive index to predict clinical success after revascularization. Radermacher et al retrospectively found a cut-off point of 0.80 useful for separating patients who failed to show a clinical response to revascularization (IR ≥ 0.80) compared with patients who did show a favorable response (IR < 80) (3). However, these interesting results were not confirmed in others studies and cannot be considered conclusive.


1749

Stenosis, Artery, Renal. Figure 1 Color Doppler ultrasonography of a high-grade truncal right renal artery stenosis. (a) Presence of an aliasing artifact (arrow) at the origin of the right renal artery. (b) Doppler spectrum shows flow acceleration close to 400 cm/sec.

. Morphologic information about the kidneys: global or segmental atrophy (a long axis inferior to 7–8 cm is generally considered a contraindication for revascularization), calcifications of the renal artery or aortic wall, aortic aneurysm, or other renal-associated lesions. CD-US is noninvasive and inexpensive. It can be performed on patients with renal insufficiency and without necessitating discontinuation of their hypertensive medications. It is of great interest for monitoring patients after treatment. Limitations are the technical difficulties associated with excessive bowel gas or body habitus and the lack of generalized expertise in its use. CTA: With the advent of helical CTand particularly with the multidetector technology, it is now possible to provide angiographic images of renal arteries with excellent spatial resolution (submillimetric). The best reconstruction algorithms are maximum intensity projection (MIP) and volume rendering (Fig. 2a). Sensitivity is about 90–95% and specificity 95% (4). Besides depiction of the lesion, CTA provides important pretherapeutic information, including the exact anatomic type of the stenosis and the presence of aortic associated plaques and calcifications (5). Limitations are the use of ionizing radiation and iodinated contrast media, which is a relative contraindication in patients with renal insufficiency. MRA: MRA (Fig. 3) has made significant progress in the past few years, particularly with the use of

Stenosis, Artery, Renal. Figure 2 Computed tomography angiography diagnosis and percutaneous transluminal renal angioplasty of renal artery stenosis. (a) Curvilinear reconstruction of the right renal artery ostium shows calcification and aortic atheromatous plaque extending into the renal artery (arrows), which is responsible for a significant stenosis. (b) Angiography performed immediately after stenting.

gadolinium-enhanced three-dimensional sequences. Compared to arteriography, sensitivity and specificity are superior, at 95 and 90%, respectively. However, there is a tendency to overestimate the stenosis, and spatial resolution is inferior compared with CTA (2). In addition to three-dimensional morphologic information, it is now possible to provide hemodynamic data about the parenchymal perfusion.

S

1750


Stenosis, Artery, Renal. Figure 3 Magnetic resonance angiography of bilateral renal artery stenosis in a patient with ischemic nephropathy. The stenosis is estimated at 70% on the right side, whereas the left lesion is overestimated as a pseudo-occlusion.

The main limitations involve cost, patient issues such as claustrophobia, the presence of metallic stents, and evaluation of small branches such as accessory arteries. Intra-arterial angiography: Intra-arterial angiography currently has no role in screening RAS. Its greatest advantage is its ability to guide intervention and control endovascular correction of the lesions.

Nuclear Medicine The role of renal scintigraphy is controversial. ACE inhibitors (captopril) decrease the perfusion pressure of the glomerulus with a resultant drop in the GFR. In patients with unilateral RAS, captopril induces significant changes of the scintigraphy compared with the baseline curve. Such changes are not observed in the contralateral normal kidney. In clinical practice, the sensitivity varies between 60 and 80%, and the specificity is 59% (5). Some factors decrease the sensitivity: bilateral RAS (30%), renal insufficiency, and chronic use of ACE inhibitors.

Diagnosis The roles of RAS imaging are multiple: (i) screening of RAS, (ii) providing precise morphologic information to determine the best therapeutic strategy, (iii) establishing relationships between RAS and symptoms, (iv) guiding

endovascular treatment, and (v) monitoring the patient after treatment. Several considerations must lead to a rigorous selection of patients before imaging: (i) the high cost of imaging, (ii) the relatively low prevalence of RAS in the general population, (iii) the frequency of association of RAS, without evident relationships, with essential hypertension, renal failure, or both, (iv) difficulties in predicting a real benefit after revascularization, (v) potential risks of intervention, and (vi) the unpredictability of spontaneous worsening of RAS. Clinical features that must prompt further evaluation of RAS are listed in Table 1. Screening of RAS must be done by CD-US. Some centers use captopril scintigraphy, but the cost-effectiveness is probably inferior to that of CD-US; operator expertise is increasing, and the probability of failing to detect significant RAS is very small. MRA and CTA are not screening methods except in cases of high clinical suspicion and indeterminate diagnosis after CD-US. CTA or MRA (in cases of renal insufficiency) can also be used in cases of difficult therapeutic decisions for patients at risk (renal insufficiency, complex atherosclerotic lesions).

Endovascular Treatment Since the first description in 1978 by Gruntzig, ▶percutaneous transluminal renal angioplasty (PTRA) has been progressively considered the initial therapeutic approach to RAS whenever possible. Although truncal RAS can be treated very well with angioplasty alone (particularly fibrodysplastic lesions), the introduction of renal artery stenting has considerably improved the anatomic results, and primary stent placement is nowadays the treatment of choice, mainly for the more frequent atheromatous lesions located in the ostial zone and the first centimeter of the artery. With the use of stenting, the global results are comparable to those of surgery but with lower morbidity and mortality. PTRA is routinely performed via a femoral approach, more rarely via a brachial approach, using microguidewires (0.018 in.) and microcatheters (3–4 French). Renal arteriography is the first step in endovascular treatment of RAS. It provides useful information for determining the therapeutic strategy, for guiding the procedure, and for evaluating the immediate results. Balloon catheters and stents are generally 12–20 mm in length and 5–6 mm in diameter. Periprocedural treatment consists of intraarterial heparin, clopidogrel (plavix), and acetylsalicylic acid (aspirin). Postintervention care consists of daily monitoring of renal function and blood pressure before discharge. CD-US is the main tool for patient follow-up for evaluating the renal artery patency. It must be performed at 6 and 12 months and annually thereafter, except in cases of symptom recurrence.

Stenosis, Spinal

The more recent studies (largely using stents) show a primary success rate between 95 and 100% (6). Midterm recurrences are observed in 15% of cases and are closely related to the quality of the immediate result. Interventionrelated mortality is low, at 1%. Major complications occur in less than 5% of cases, and immediate surgery is required in less than 0.5% of patients. Clinical results are not always related to anatomic results and are totally different in atheromatous patients and in cases of fibromuscular dysplasia. In hypertension, a clinical benefit (cure or improvement) is obtained in approximately 50% of all atheromatous stenoses and in 85% of stenoses due to fibromuscular dysplasia. Results are worse in patients with renal insufficiency, many of them being older with extensive atheromatous lesions and probably nephroangiosclerosis. A clinical benefit is obtained in approximately 50% of cases. Despite the good results for several categories of patients such as those with fibromuscular dysplasia, multiple points of controversy persist, particularly concerning the indications. Due to morbidity of treatment and limited efficacy in some patient categories, a multidisciplinary discussion estimating risks and benefits is mandatory before making therapeutic decisions. Some anatomic parameters such as difficulties of arterial access, the lesion’s complexity and severity, diffusion of atherosclerotic lesions, the importance of aortorenal plaques, the presence of calcifications, and kidney size should be taken into account and are often necessary to evaluate before even discussing CTA or MRA. Age, general status, comorbidities, and expected duration of life are other important parameters. In terms of evidence-based medicine, the universally recognized indications for PTRA are significant and accessible RAS with (i) uncontrolled hypertension with medication, (ii) progressive renal insufficiency with bilateral RAS or RAS of a single kidney, or (iii) acute onset of cardiac insufficiency with recurrent pulmonary edema (6). In the following situations, the benefit of PTRA has never been clearly demonstrated, and indications must be discussed case by case, as previously described: . Arterial hypertension well controlled by medical treatment . Severe asymptomatic RAS . Renal insufficiency with unilateral significant RAS . Need for treatment with conversion enzyme inhibitors in patients with RAS . RAS associated with contralateral lesions necessitating nephrectomy. Contraindications for endovascular treatment are associated lesions of the abdominal aorta (aneurysms) and complex atheromatous or dysplastic lesions. RAS with a narrowing of less than 50% in diameter should never be treated.

1751

Bibliography 1. 2.

3.

4.

5.

6.

Hartman RP, Kawashima A, King BF (2003) Evaluation of renal causes of hypertension. Radiol Clin North Am 41:909–929 Soulez G, Oliva VL, Turpin S et al (2000) Imaging of renovascular hypertension: respective value of renal scintigraphy, renal Doppler US, and MR angiography. Radiographics 20:1355–1368 Radermacher J, Chavan A, Bleck J et al (2001) Use of Doppler ultrasonography to predict the outcome of therapy for renal artery stenosis. N Engl J Med 344:410–417 Beregi JP, El Kohen M, Deklunder G et al (1996) Helical CT angiography compared with arteriography in the detection of renal artery stenosis. Am J Roentgenol 167:495–501 Huot SJ, Hansson JH, Dey H et al (2002) Utility of captopril renal scan for detecting renal artery stenosis. Arch Intern Med 162:1981–1984 Martin LG, Rundback JH, Sacks D (2002) Quality improvement guidelines for angiography, angioplasty and stent placement in the diagnosis and treatment of renal artery stenosis in adults. J Vasc Interv Radiol 13:1069–1085

Stenosis, Spinal J AN T. W ILMINK Radiology Department University Hospital Maastricht The Netherlands [email protected]

Synonyms Spinal narrowing; Verbiest syndrome of lumbar spine

Definitions Spinal stenosis is defined as abnormal general or regional narrowing of the spinal canal. The condition may affect the entire spine as in achondroplasia, but usually sections of the spine are more severely involved than others. Spinal stenosis manifests itself most frequently in the lumbar spine, is encountered less frequently in the cervical region and is occasionally seen limited to the thoracic spinal canal. According to the reports by Mixter and Barr on lumbar disc herniation as a cause of low back pain with sciatica, the 1930s and 1940s could be described as the “decades of the disc” as far as concepts on causes of low back pain and sciatica were concerned. In the 1950s and 1960s, the realization began to grow that, besides being caused by herniation of the lumbar intervertebral disc, low back complaints irradiating to the legs could also be due to reduced dimensions of the spinal canal or, quite frequently, to a combination of these two conditions.

S

1752

Stenosis, Spinal

It was also recognized that long-tract spinal cord symptoms originating in the cervical region were caused frequently by spinal cord compression due to narrowing of the cervical spinal canal and not by a disease of the cord itself. Understanding of etiology, symptomatology, and diagnosis in patients with spinal stenosis is greatly enhanced by insight into the effects of postural changes, such as lordosis and kyphosis upon spinal structures and the contents of the spinal canal.

Etiology, Pathology The classification of causes of spinal stenosis is extensive (1). Distinction can be made between congenital-developmental and acquired stenosis. The original report by Verbiest (2) described seven patients with what was thought to be

developmental narrowness of the lumbar spinal canal, with heavily developed laminae and articular processes compressing the dural sac (Fig. 1). In fact this classic form of stenosis is quite rare, and the most important single cause of spinal narrowing is considered to be degenerative. Epstein focused attention on abnormal narrowing of the lateral recesses of the spinal canal and of the intervertebral foramina as a cause of chronic sciatica (3). The application of computed tomography to spinal imaging in the 1970s and 1980s provided new insights into the various factors causing narrowing of the spinal canal and compression of the dural sac or the emerging nerve root. Combinations of these factors in spinal stenosis are the rule rather than the exception (4). In the lumbar region the spinal canal may be somewhat narrowed on developmental basis. This need not cause symptoms by itself, but in later life degenerative hypertrophy of the facets and flaval ligaments may further reduce spinal

Stenosis, Spinal. Figure 1 Developmental stenosis of lumbar spine. Sagittal T1-(1a) and T2 weighted (1b) images demonstrate a shallow spinal canal with bony midsagittal diameter of 11mm. measured at level of L5 lamina. Note L3-4 disc protrusion further reducing available space. Axial T2 weighted cuts at level of lamina (1c) show normal interpedicular diameter and reduced AP diameter, at disc level dorsolateral encroachment by facets is seen (1d). Note that bony AP diameter is smallest at level of lamina, but effective spinal dimensions are smallest at disc level, where CSF block is seen at L3-4 and L4-5.

Stenosis, Spinal

dimensions, as may degenerative spondylolisthesis due to facet destruction, and/or degenerative disc bulging. Increase in epidural fat in obesity or steroid use can further aggravate the situation. In this way a complex of factors, each in themselves insufficient to cause complaints, can combine to compress the cauda equina and it goes without saying that a small disc herniation or even a bulging disc in such a narrowed spinal canal will cause symptoms much sooner than a similar disc lesion in a patient with a roomy canal (Fig. 2). In the cervical spine, a developmentally narrow spinal canal will be associated not only with an increased risk of acute cord lesion in trauma, but also with greater risk of chronic progressive myelomalacia as degenerative osteophyte formation causes gradual further narrowing of the spinal canal with compression and wasting of the cord (Fig. 3). In the thoracic spine, degenerative hypertrophy of the flaval ligaments can occur, with a greater risk of cord compression in a relatively narrow bony spinal canal.

Clinical Presentation In lumbar spinal stenosis the most common symptoms are tiredness and loss of power in the legs, backache, and bilateral leg pain, occasionally anesthesia, and a feeling of numbness in the sacral dermatomes. Typically, these complaints present and progress when the patient is standing or walking in lordotic posture, and are immediately relieved by sitting or lying down, by squatting or any other posture which reduces lordosis and induces kyphosis. Verbiest named this clinical picture “intermittent neurogenic claudication,” and the striking effect of postural changes serves to distinguish it from intermittent vascular claudication which is exercisedependent and not posture-dependent. When the patient suffering from neurogenic claudication is lying recumbent there are no complaints, and tests of radicular compression such as straight leg raising are negative. The clinical picture may not be so clear-cut for neurogenic claudication when only a single exiting nerve root is compressed by narrowing of the lateral recess or the intervertebral foramen, or when there is a concomitant disc herniation. The pain is then more monoradicular in nature, although posture-dependency may be present. In narrowing of the cervical or thoracic spinal canal it is not the cauda equina which is compressed but the spinal cord. The clinical presentation is that of gradually progressive long-tract neurologic dysfunction, with disturbances of walking and micturition, and muscular weakness with increased tendon reflexes and pathologic extensor plantar responses at neurologic examination.

1753

Neck pain radiating to the arms may be present, especially when the cervical foramina are narrowed by osteophyte formation and the exiting nerve roots are compressed. Clinical signs of cord compression in cervical stenosis are not as strikingly posture-dependent as those of cauda equina compression in lumbar stenosis.

Imaging Plain X-ray Films These are of limited value in assessing dimensions of the spinal canal. Verbiest defined absolute lumbar stenosis as a condition in which the bony midsagittal diameter at the upper laminar level is less than 10 mm, while relative stenosis is said to be present when the same diameter is between 10 and 12 mm (mean normal value 17 mm). This diameter cannot be measured on a plain film, however, as it is obscured by overlying bony structures, and the same applies to the dimensions of the lumbar facets and the interfacet diameter which may also be reduced in stenosis. The size of the vertebral body and the interpedicular diameter can be measured accurately, but these are usually normal in patients with spinal stenosis. Signs of degenerative change such as spondylosis, spondylarthrosis, degenerative anterolisthesis, and foraminal narrowing can be identified on plain films and provide a clue to the possible presence of degenerative narrowing of the spinal canal, but these signs are frequently also present in elderly subjects without symptoms.

Conventional Myelography After intradural injection of a radiologic contrast medium, compression of the dural sac by lumbar spinal stenosis is revealed by the presence of a posturedependent myelographic block present in lumbar retroflexion or lordosis and relieved by anteflexion or kyphosis. This block is invariably present at the level of the intervertebral disc and not at the level of the bony lamina. Posture-related changes in spinal dimensions are a normal phenomenon at the disc level, where the spinal canal is bordered by ligamentous surfaces. In lordosis the annulus fibrosus of the disc bulges backward into the spinal canal and the flaval ligaments dorsolateral to the canal also bulge inwards. In a normal spinal canal this does not produce symptoms, but in a stenotic canal the extra reduction of space induced by lordosis is sufficient to compress the dural sac and cauda equina and cause neurogenic claudication (5). In the cervical spine, cord compression in symptomatic stenosis is also seen by a myelographic block in lordosis, relieved when lordosis is reduced.

S

1754

Stenosis, Spinal

Stenosis, Spinal. Figure 2

(Continued)

Stenosis, Spinal

1755

Stenosis, Spinal. Figure 3 Cervical stenosis shown on T1-(3a) and T2 weighted (3b) images. Note obliteration of CSF spaces around spinal cord, compression of cord and a high-signal intramedullary lesion in 3b (arrow).

Computed Tomography The axial images produced by these techniques greatly improved understanding of various types of spinal stenosis. Accurate measurement of the bony dimensions of the spinal canal was now possible, although limited soft tissue resolution made employment of CT myelography necessary in some cases. As mentioned above, the soft tissues at disc level are instrumental in producing cauda equina compression in spinal stenosis, and soft tissue hypertrophy or bulging may produce symptoms in individuals with bony spinal dimensions well above those defined by Verbiest (5). Undue reliance on bony spinal measurements can therefore prove misleading, and the only soft tissue structure within the canal that can be measured on plain CT is the dural sac. A study comparing myelographic and CT features of nerve root compression in a group of 100 patients with sciatica revealed that a CSF block at myelography in spinal stenosis occurred when the CT cross-sectional area of the dural sac in the lower lumbar region decreased to below 40 mm2 (6). In the cervical region the cross-sectional area of the spinal cord can be measured by CTmyelography. In normal

individuals this value is around 90 mm2. The cervical cord is surprisingly resilient to chronic compression, and longtract signs usually do not begin to appear before cord area has been reduced by one third to around 60 mm2 (7) CT myelography also provides a detailed image of the cervical nerve root as it departs from the spinal cord and later also from the dural sac within its root sleeve. Nerve root compression by bony foraminal narrowing can best be diagnosed by this technique.

Magnetic Resonance Imaging (MRI) This is presently the best method for imaging the spine in virtually all conditions. It has been held by some that CT is better suited for the diagnosis of spinal stenosis because of the bright signal of bony structures produced by this technique. In fact, cortical bone outlining the normal and abnormal spinal canal can be easily distinguished in the MR image which has the additional advantage of demonstrating concomitant soft-tissue pathology such as flaval hypertrophy and bulging or protruded disc much more clearly than CT. An important benefit of MRI is the possibility of

Stenosis, Spinal. Figure 2 Combined stenosis. Bony AP diameter is relatively narrow; 12mm. at level of L4 lamina. T1 weighted sagittal image (2a) also shows some increase in depth of retrodural fat pad. Bulging discs at multiple levels seen on T2 weighted midsagittal (2b) and lateral-sagittal images (2c). At L2-3 this results in a CSF block with elongated and coiled nerve roots above level of block; so-called redundant roots, confirmed by MR myelogram (2d, arrow). At L4-5 there is also some encroachment by hypertrophic facets and flaval ligaments (2e), resulting in kinking of traversing L5 nerve roots, especially at right (2d, arrowhead). At L3-4 there is no disc bulging and no facet hypertrophy (2f), myelogram shows only slight displacement of nerve roots.

S

1756

Stent

producing heavily T2-weighted myelographic images which make it possible to identify features associated with symptomatic spinal stenosis such as a myelographic CSF block or the presence of redundant nerve roots. In the cervical region, compression of the spinal cord can be demonstrated directly, and the presence of myelomalacia can be diagnosed by the presence of an intramedullary lesion with increased T2 signal. As mentioned above, an ancillary role is still reserved for CT in the demonstration of osteophytes at the foraminal entrance, potentially compressing the exiting nerve root. Functional imaging studies in spinal anteflexion and retroflexion are helpful in the diagnosis of spinal stenosis. In the cervical spine such studies can be performed in MRI systems as well as CT scanners. Lumbar functional studies are not possible in most MRI systems presently in use, but the development of open upright MRI systems make loaded functional studies possible and will provide a new dimension in lumbar spinal imaging.

Stent, Carotid Artery K. A LFKE , O. J ANSEN Neuroradiology, UK SH, Kiel, Germany [email protected]

Synonyms Endovascular treatment of internal carotid artery stenosis; Stenting

Definition Treatment of a carotid artery stenosis with a stent is an endovascular intervention. It is performed to reduce the risk of stroke.

Bibliography 1.

2. 3. 4. 5. 6. 7.

Arnoldi CC, Brodski AE, Cauchoix J et al (1976) Lumbar spinal stenosis and entrapment syndromes. Definition and classification. Clin Orthop 115:4–5 Verbiest H (1954) A radicular syndrome from developmental narrowing of the spinal canal. J Bone Joint Surg Br 36B:230–237 Epstein BS, Epstein JA, Jones MD (1977) Lumbar spinal stenosis. Radiol Clin North Am 15:227–239 Paine KWE, Haung PWH (1972) Lumbar disc syndrome. J Neurosurg 37:75–82 Penning L (1992) Functional pathology of lumbar spinal stenosis. Clin Biomech 7:3–17 Wilmink JT (1989) CT morphology of intrathecal lumbosacral nerve-root compression. AJNR Am J Neuroradiol 10:233–248. Penning L, Wilmink JT, Van Woerden HH et al (1986) CT myelographic findings in degenerative disorders of the cervical spine: clinical significance. AJNR Am J Neuroradiol 7:119–127.

Stent Small tube like device mounted on a balloon catheter or fixed to the tip of a catheter that can be expanded or is self-expending and thus dilates and stabilizes a vessel. ▶Stroke, Interventional Radiology

Stent Graft A stent graft consists of a metallic wire mesh tube covered with a graft material. It is used mainly in the vascular system to treat aneurysms, vessel ruptures, or fistulas. ▶Aneurysm, Aortic and Thoracic

Characteristics Introduction Treatment of carotid artery stenosis is performed to reduce the risk of stroke. A symptomatic stenosis with a grade of over 50% according to ▶North American Symptomatic Carotid Endarterectomy Trial criteria (NASCET) (1) or over 70% according to ▶European Carotid Surgery Trial criteria (ECST) (2) should be treated. Regarding timing of treatment after brain infarction, in patients with larger infarcts an interval of 2 weeks seems to be safe to avoid reperfusion injury with the risk of secondary hemorrhage. In cases of smaller infarctions, earlier treatment is possible.

Indication Endovascular treatment with stenting and operative endarterectomy are alternative methods. The largest and most recent randomized study comparing these two methods is the Stent-Supported Percutaneous Angioplasty of the Carotid Artery versus Endarterectomy (SPACE) trial (3). It included 1,200 patients with symptomatic carotid artery stenosis (>50% NASCET, > 70% ECST). The rate of the primary endpoint, death, or ipsilateral ischemic stroke from randomization to 30 days after the procedure was 6.84% with carotid stenting and 6.34% with endarterectomy. These results did not prove equivalence of the two methods statistically, but the difference was not statistically significant (difference of only four events). Long-term results for a period of 2 years will follow.

Stent, Carotid Artery

The use of stenting as therapy for carotid artery stenosis is increasingly widespread, and guidelines are needed for everyday practice. Therapeutic decision should be based on different parameters and should differentiate between three groups of patients: The first group has a high surgical risk and therefore a clear indication for stenting. This group includes patients with restenosis after endarterectomy, radiogenic stenosis, and stenosis that is surgically not accessible, such as distal extracranial or intracranial carotid artery stenosis or a combined intra-/extracranial stenosis. Stenosis caused by nonarteriosclerotic diseases such as dissection, fibromuscular dysplasia, or Takayasu arteritis should be stented if treatment is necessary (4). Also, patients with high surgical risk due to comorbidity should be treated with local anesthesia with stenting. Patients in the second group with symptomatic highgrade stenosis can be treated with either method. In this group, the patient’s preference for one of the two methods and also the expertise of the different local therapists should be considered. The third group includes patients with asymptomatic stenosis over 50% (NASCET) who might benefit from surgical treatment. New trials including SPACE II and CAVATAS II will compare the endovascular and operative treatment of asymptomatic stenosis.

Contraindications for Stenting Contraindications for stenting are allergies to contrast media or to the necessary antiplatelet treatment with aspirin or clopidogrel. Renal insufficiency or hyperthyroidism must be considered. A floating intraluminal thrombus might prevent passage of the stenosis with endovascular tools.

Preparation for the Stenting Procedure Necessary preinterventional preparation includes clinical neurological assessment of the patient and his or her symptoms and imaging of the stenosis, the supraaortic vessel anatomy, and the brain. Magnetic resonance (MR) or computed tomography (CT) angiography gives an overview of brain-feeding arteries to evaluate the anatomy of the stenosis in relation to surrounding structures and to show, for example, tandem stenosis or occlusion of the contralateral carotid artery. Imaging of the brain with CT or MR gives information about the mechanism of stroke (e.g., embolic or hemodynamic stroke) and should exclude contraindications such as large acute infarction or bleeding. Ultrasound Doppler examination is necessary to estimate the grade of the stenosis. Antiplatelet medication is mandatory. Different schemes have been described. Patients included in the

1757

▶SPACE trial received 100 mg aspirin and 75 mg clopidogrel daily for at least 3 days before and 30 days after stenting. The aspirin was then continued.

The Stenting Procedure The procedure is performed using local anesthesia via a percutaneous transfemoral or transbrachial access with a long sheath or guiding catheter. The sheath or catheter is advanced coaxially over the catheter and guide wire into the common carotid artery. The sheath should be flushed permanently with heparinized saline. After introduction of the sheath heparin can be administered intravenously. The heparin effect can be controlled by bedside measurement of activated clotting time. To prevent bradycardia, atropine can be given subcutaneously at the beginning of the procedure and again intravenously before balloon dilatation of the carotid bulb. Angiography should include at least two projections showing the stenosis without overlay and two projections of the intracranial vasculature before stenting. For stent placement, fluoroscopy and the road-map technique should be used. The biplane angiography technique is extremely helpful. After stenting, the result should be documented angiographically (Fig. 1). After the sheath is placed in the common carotid artery, the stenosis is passed with a microguide wire. Balloons and the stent delivery system are commonly monorail systems. Predilation of the stenosis with a 2.5- or 3-mm balloon might be necessary when primary passage with the stent delivery system is not possible. The suitable stent size is then estimated from angiography, depending on the vessel’s diameter and the location of the stenosis. After stent placement, secondary balloon dilation is performed. Finally, the sheath is removed. Protection devices with filters or flow reversal have not yet shown a benefit in controlled studies. They probably reduce the risk of a larger embolism but prolong and complicate the procedure, which might cause other complications.

Possible Complications Stroke or transient neurological deficit due to periinterventional embolism is probably the most severe complication of stenting. Parts of a thrombus or plaque can be loosened during the passage of the stenosis with the wire, stent delivery system, or balloon. In particular, balloon dilatation of the stent and arterial wall might cause rupture of a plaque followed by embolism. Dissection of the arterial wall in the site of the stenosis and also in adjacent segments can lead to severe stenosis or thromboembolism. After treatment of high-grade stenosis, hyperperfusion in the territory of the internal carotid artery can

S

1758

Stenting

Step-off. Figure 1 Three steps of a stenting procedure of a carotid artery stenosis. From left to right: First, the stenosis is shown in a lateral angiographic view without subtraction. Then the stent is delivered, expanding the stenosis with its radial force. Finally, after balloon dilation of the stent, a satisfying result is achieved.

lead to edema or hemorrhage with neurological impairment, headache, seizures, and focal neurological deficits. Potential side effects of contrast media administration include allergic reaction, impairment of renal function, and hyperthyroidism. At the puncture site, a groin hematoma might occur. Restenosis caused by intimal hyperplasia is possible but normally asymptomatic. Intimal hyperplasia typically appears during the first 6 months after stenting. Follow-up with ultrasound over a period of approximately 2 years is necessary to detect severe restenosis, which can be treated with balloon angioplasty or even stenting again.

Future Prospects Stenting of carotid artery stenosis has the great advantage of avoiding the risks of operative treatment. Incisions, with the risk of cranial nerve palsy or cervical wound hematoma, and longer occlusion of the treated vessel are not necessary. Often, hospitalization of the patients is shorter. Nevertheless, larger trials are needed to show the equivalence of neurological complication rates and long-term results. Improvement of the interventional tools such as stents and protection devices that reduce the rate of thromboembolism, the growing experience of the interventionists, and the refining of the preintervention diagnostic processes (e.g., plaque imaging) will further decrease complication rates and optimize results.

Bibliography 1.

2.

3.

4.

European Carotid Surgery Trialists’ Collaborative Group (1991) MRC European carotid surgery trial: interim results for symptomatic patients with severe (70–99%) or with mild (0–29%) carotid stenosis. Lancet 337:1235–1243 Ferguson GG, Eliasziw M, Barr HW et al (1999) The North American symptomatic carotid endarterectomy trial: surgical results in 1415 patients. Stroke 30(9):1751–1758 The SPACE Collaborative Group (2006) 30-day results from the SPACE trial of stent-protected angioplasty versus carotid endarterectomy in symptomatic patients: a randomised non-inferiority trial. Lancet 368:1239–1247 Barr JD, Connors JJ III, Sacks D et al (2003) American Society of Interventional and Therapeutic Neuroradiology; American Society of Neuroradiology; Society of Interventional Radiology. Quality improvement guidelines for the performance of cervical carotid angioplasty and stent placement. J Vasc Interv Radiol 14:S321–335

Stenting ▶Stent, Carotid Artery

Step-off The normal change in contour at metaphysis between the 1–3 mm straight metaphyseal collar and the

Stomach and Duodenum in Adults Postoperative

monotonically curved metaphysis and shaft; not to be confused with buckle fracture. ▶Battered Child Syndrome

Sterile Effusion in One or More Joints

1759

Stomach and Duodenum in Adults Postoperative M ATHIEU H. R ODALLEC , M ARC Z INS Department of Radiology Fondation Hoˆpital Saint-Joseph Cedex 14, Paris, France [email protected]

▶Transient Synovitis

Definitions

Sterile Inflammation of the Hip ▶Transient Synovitis

Sternoclavicular Hyperostosis ▶Spondyloarthropathies, Seronegative

Stimulated Acoustic Emission (SAE)

Current operations on the stomach and duodenum are performed for the treatment of gastroesophageal reflux, for the complications of peptic ulcer, for the resection of benign and malignant masses, for the complications of chronic pancreatitis, and for the management of obesity (1–3). Indications for surgical management of gastroesophageal reflux and peptic ulcer disease have decreased with the advent of proton pump inhibitors (4). Whereas rates of bariatric surgery have dramatically increased in recent years because morbid obesity is increasing in epidemic proportions (5, 6). Radiologic studies are crucial in the evaluation of patients with symptoms after gastric or duodenal operations. The radiologist may be asked to demonstrate complications in the early postoperative period or to define anatomy and detect disease years later.

Gastroesophageal Reflux SAE is a nonlinear response of ultrasound-induced disintegrating gas-filled microbubbles. In particular, gasfilled microbubbles can be destroyed when excited with ultrasound under certain conditions. The microbubbles disintegrate rapidly, thereby emitting a strong nonlinear signal, which is misinterpreted as quick movement and mapped as a characteristic random color pattern in the color Doppler mode of the ultrasound device. ▶Local Drug and Gene Delivery with Microbubbles

Stochastic Risk

Indications for operative treatment of gastroesophageal reflux include clinical symptoms or esophagitis refractory to medical treatment, and Barrett esophagus (2–4). Standard operations attempt to increase the mechanical component of a competent esophagogastric junction mechanism. The Nissen ▶fundoplication creates an intraabdominal wrap of gastric fundus completely around the lower esophagus. Various other antireflux procedures attempt to restore the abdominal esophagus and ensure that it responds to intraabdominal pressure. Wraps can be noncircumferantial as in the Toupet partial fundoplication.

Peptic Ulcer Disease This is dependent upon cell transformation, with the severity of the effects being independent of the radiation dose received. However, the greater the absorbed radiation dose received by the patient, the more likely a stochastic effect is to occur. ▶Radiation Issues in Childhood

Current indications for surgery in patients with peptic ulcer disease include hemorrhage, perforation, obstruction, ulcer intractability, and the inability to exclude malignancy in a gastric ulcer (2, 3). Generally, this can be accomplished with partial gastrectomy, with removal of

S

1760


the ulcer. Combined gastric and duodenal ulcers and prepyloric ulcers are managed according to the principles of surgery for duodenal ulcer.

Duodenal Ulcers Nowadays, surgery for duodenal ulcers is rare. But physicians who treat postoperative patients must be familiar with procedures that were performed over the past several decades. Antral resection and vagotomy has been used in the treatment of duodenal ulcers to decrease the secretion of hydrochloric acid and to remove ulcerbearing gastric mucosa. Distal gastric resection with reconstruction by anastomosis of the proximal stomach to the end of the duodenum is known as a Billroth I reconstruction. When anastomosis to the duodenum is not possible, a gastrojejunostomy (Billroth II) is performed. A variable length of jejunum forms the afferent loop and carries duodenal contents toward the stomach. The efferent loop is the jejunum on the side of the gastrojejunostomy that is distal to the duodenum. The gastrojejunostomy may be anterior to the transverse colon (antecolic), or the small bowel may be brought up posteriorly through an opening made in the transverse mesocolon (retrocolic) (3). An alternate method of reestablishing gastrointestinal continuity after gastric resection is Roux-en-Y gastrojejunostomy. The jejunum is divided just distal to the ligament of Treitz: the proximal end or side of the small bowel is attached to the stomach while the distal end is fashioned into an enterostomy downstream. This procedure diverts duodenal contents away from the gastric anastomosis and prevents bile reflux.

Gastric and Duodenal Masses Resection Stomach malignancies require formal anatomic resection, with distal or total gastrectomy. Gastrointestinal continuity is restored with loop or Roux-en-Y esophagojejunostomy. Duodenal malignancies are resected with pancreatoduodenectomy, with reanastomosis of the gastrointestinal tract, bile duct, and pancreatic remnant (1, 3). Pancreatoduodenectomy may involve distal hemigastrectomy, or the stomach and pylorus may be preserved by reconstruction with end-to-side duodenojejunostomy. The bile duct is anastomosed to the jejunum, and the distal pancreas is anastomosed to the jejunum or the posterior wall of the stomach.

procedures create a small gastric pouch with a restricted outlet to cause early satiety, decreased caloric intake, and weight loss (5, 6). Currently, most surgeons prefer the gastric bypass approach or the vertical gastroplasty procedure. Bariatric surgeons use gastric silastic rings in both gastroplasty and gastric bypass operations to maintain the function and width of the stoma. Gastric Bypass (GBP). A small proximal pouch is anastomosed to the jejunum via a loop or Roux-en-Y gastrojejunostomy. The stomach is separated from the food path but it is not transected. With a Roux-en-Y type anastomosis, the jejunum is divided, the proximal end or side of the distal bowel is attached to the stomach, and the Roux-en-Y loop is anastomosed to the jejunum downstream. With a loop gastroenterostomy, the standard right-to-left anastomosis is performed. Vertical Banded Gastroplasty. A small tubular pouch and channel is created on the lesser curvature portion of the stomach.

Pathology/Histopathology Complications occur when operations on the gastroduodenal tract fail to resect, reconstruct, or redesign the tissues and function of the gastroduodenal tract. There are common complications like anastomotic leaks, abscesses (Fig. 1), strictures, and bowel obstruction that occur during any operation on the alimentary tract. There are other complications that are unique to each operative procedure.

Gastroesophageal Reflux If the fundoplication is too tight dysphagia or so-called gas bloat syndrome may develop, in which patient is unable to belch or to vomit. Long-term complications include fundoplication dehiscence, recurrent hernia, and slip wrap (4).

Peptic Ulcer Disease After operations for peptic ulcer, the most undesirable gastrointestinal sequelae are consequences of vagotomy, gastric resection, or pyloric ablation (3). Diarrhea, dumping, and bile reflux can cause clinical debility and nutritional deficiency in a small percentage of patients. Early mechanical and anatomic complications are not common but can usually be diagnosed radiologically. Cancer arising in gastric remnants is associated with longterm metabolic effects (2).

Surgery for Obesity

Duodenal Ulcers

Surgery is an effective alternative to failed medical and dietary therapy for life-threatening obesity. These stapling

Antrectomy and total vagotomy increase the risk of undesirable side effects such as dumping, diarrhea, bile reflux, and potential cancer arising in gastric remnants (1, 3).


1761

through the surgical defect in the transverse mesocolon. The herniated bowel is usually the Roux-en-Y loop itself.

Clinical Presentation Anastomotic Leaks The presentation can be acute (peritonitis) or progressive (subphrenic abscess).

Afferent Loop Syndrome

Stomach and Duodenum in Adults Postoperative. Figure 1 Duodenal stump fistula complicated with intraperitoneal abscesses in a patient with partial distal gastrectomy for gastric adenocarcinoma. On axial enhanced CT with water-contrast agent, the gastro-jejunal anastomosis is clearly depicted (a). Axial enhanced CT at the level of the duodenal stump (b) depicts extradigestive gas in contact with the staples, and intraperitoneal abscesses.

Gastric and Duodenal Masses Resection Recurrent neoplasm is a potential complication. Diarrhea, dumping, and bile reflux may follow gastric resection. Pancreatic resection may cause a deficiency of digestive enzymes. Metabolic deficiencies (iron or vitamin B12) are accentuated in patients with total gastrectomy.

Surgery for Obesity Gastric staple line dehiscence is most often caused by repeated overdistention of the gastric pouch with food. Massive gaseous distention of the excluded stomach may develop occasionally after gastric bypass procedure. It may occur because of edema at the enteroenterostomy, or it may be secondary to small bowel obstruction. Internal hernias, following Roux-en-Y GBP are much more common than in the general population and are likely to be underdiagnosed (5). The most common type of internal hernia reported after laparoscopic Roux-en-Y GBP is transmesenteric, in which the small bowel herniates

Afferent loop syndrome is caused by increased intraluminal pressure and distension due to accumulation of enteric secretions in an obstructed afferent duodenal stump in a Billroth II gastrojejunostomy (3). Afferent loop syndrome is one of the main causes of duodenal stump blowout in the early postoperative period and is also an etiology for postoperative obstructive jaundice, ascending cholangitis, and pancreatitis due to transmission of high pressures back to the biliopancreatic ductal system. Prolonged stasis and pooling of secretions facilitate bacterial overgrowth in the afferent loop. Bacteria deconjugate bile acids, which can lead to steatorrhea, malnutrition, and vitamin B-12 deficiency. Iron deficiency can occur because of bypassing of the proximal small bowel. Patients with acute afferent loop syndrome typically present with a sudden onset of abdominal pain with nausea and vomiting. If the afferent loop is not decompressed, the patient can develop peritonitis and shock if intestinal perforation or infarction ensues.

Bezoar Bezoar is reported in patients with impaired digestion and decreased gastric motility. Bezoar may cause vomiting, diarrhea, pain, and gastric ulcers.

S Imaging Before beginning the examination, specific information about the postoperative anatomy should be obtained. If an anastomosis has been fashioned, drug-induced hypotonia, which renders the gastric remnant and small bowel hypotonic, prevents rapid slipping of contrast material into the distal small bowel.

Conventional Radiographs The pattern of surgical clips and staples may help determine the anatomy with reliability. Unexpected foreign bodies, extraluminal gas collections, or signs of bowel obstruction should be sought.

1762


Contrast-Enhanced Studies Water-soluble contrast agents are used when anastomotic leaks, staple line dehiscences, perforations, fistulas, or abscesses are suspected. Barium studies provide an assessment of mucosal detail that cannot be obtained with water-soluble agents. Biphasic techniques that use both high-density barium sulfate suspension and effervescent agents and low-density contrast material with compression and palpation provide an excellent view of mucosal detail and the anastigmatic area (3). Patient cooperation and mobility are crucial for these studies. Protocols for position changes and radiographic sequences cannot be specified because each patient has unique surgical variations.

US, CT, and MRI US, CT, and MRI are useful to detect fluid collections, abscesses, leaks, or fluid-filled obstructed bowel loops. For CT scanning, water-soluble contrast agents should be used liberally. CT is better suited to detect extradigestive gas.

Nuclear Medicine Nuclear medicine emptying studies with radiolabeled solids and liquids can be used for patients with suspected postgastrectomy stasis syndromes, afferent loop dysfunction, motility problems of dumping and diarrhea, and postoperative symptoms of an unclear cause (3).

Diagnosis Anastomotic Leaks Extravasation of contrast material beyond the bowel lumen can be recognized without difficulty if the normal postoperative appearances are known.

Afferent Loop Syndrome CT demonstrates the obstructed segment as a U-shaped, fluid-filled tubular structure confined to the subhepatic area or crossing the midline between the abdominal aorta and the superior mesenteric artery. CT is also useful to predict the underlying pathology causing afferent loop syndrome (carcinomatosis, adhesion, or internal hernia).

Stomach and Duodenum in Adults Postoperative. Figure 2 Partial gastrectomy complicated with bowel obstruction after bezoar migration. Axial enhanced CT at the level of the stomach demonstrates an ovoid intraluminal mass with mottled gas pattern (a). Axial enhanced CT also demonstrates small bowel obstruction due to migration of bezoar (arrowhead) (b).

Tumor Recurrence Local tumor recurrence is usually diagnosed with endoscopy. US, CT, and MRI are useful for loco-regional staging (1, 2).

Long-Term Complications of Fundoplication Long-term complications are generally evaluated with barium studies (4). A completely dehisced fundoplication often mimics normal findings in a healthy patient who has not undergone surgical intervention. In partial fundoplication dehiscence, there is no significant tapering of the distal esophagus. In slipped fundoplication, the gastroesophageal junction lies above the level of the wrap. In recurrent hernia, the entire wrap lies above the esophagus hiatus.

Bezoar Conventional radiographs and CT demonstrate an ovoid intraluminal mass with mottled gas pattern within the stomach (2). Small bowel obstruction can follow bezoar migration (Fig. 2).

Interventional Radiological Treatment CT and fluoroscopic guidance are useful for treatment of the postsurgical complications.

Stone Disease, Urinary

Bibliography 1.

2.

3. 4.

5.

6.

Zins M, Sauvanet A (1999) Pancreás et duodenum. In: Zins M, Sauvanet A (eds) Imagerie de l’appareil digestif ope´re´. Me´decineSciences, Flammarion, Paris, pp 39–61 Vullierme MP, Sauvanet A, Zins M (1999) Cardia et estomac. In: Zins M, Sauvanet A (eds) Imagerie de l’appareil digestif ope´re´. Me´decine-Sciences, Flammarion, Paris, pp 21–37 Smith C, Deziel DJ, Kubicka RA (1994) Evaluation of the postoperative stomach and duodenum. Radiographics 14:67–86 Canon CL, Morgan DE, Einstein DM et al (2005) Surgical approach to gastroesophageal reflux disease: what the radiologists needs to know. Radiographics 25:1485–1499. Yu J, Turner MA, Cho SR et al (2004) Normal anatomy and complications after gastric bypass surgery: helical CT findings. Radiology 231:753–760 Blachard A, Federle MP, Pealer KM et al (2002) Gastrointestinal complications of laparoscopic Roux-en-Y gastric bypass surgery: clinical and imaging findings. Radiology 223:625–632

Stone Disease, Urinary G. H EINZ -P EER Department of Radiology, Medical University, Vienna, Austria [email protected]

Epidemiology Urinary stone disease is a common problem among people from industrialized nations. The incidence is 7 per 1,000 and it is responsible for 7 to 10 of every 1,000 hospital admissions in the USA (1). With increased abdominal imaging more stones are diagnosed as an incidental finding. In most cases, no clear precipitating factor is identified and the natural cumulative recurrence rate is reported to be 14% at 1 year, 35% at 5 years, and 52% at 10 years. In general the overall lifetime risk is accepted as 35–65% with a male preponderance (male: female ratio of 2:1) and the peak age of onset is 20–30 years. Caucasian or Asian origin confers a higher risk (1).

Pathophysiology/Histopathology The main types of stones encountered are calcium containing, composed mainly of pure calcium oxalate or calcium oxalate mixed with calcium phosphate. The majority of the remainder is so-called struvite or matrix stones composed of magnesium ammonium phosphate. Uric acid and cystine stones account for less than 10% of all calculi. An organic matrix of mucoprotein, consisting 1–5% of the stone by weight, is present in all calculi.

1763

In most cases, a precise etiological factor is not identified but certain predisposing factors are recognized. Incidence is higher in patients with an anatomical abnormality of the urinary tract and this may be related to urinary stasis. Stones form when the concentration of two ions in solution exceeds the saturation point. The condition at which this occurs depends on a number of patient-related factors (1).

Clinical Presentation Urinary stone disease may present in a number of ways. The classic presentation is with acute severe ipsilateral loin to groin pain, associated with nausea and vomiting. This history combined with renal angle tenderness and microscopic hematuria is highly accurate in making a clinical diagnosis of urinary tract stones with a reported sensitivity and specificity of 84 and 99%, respectively (2). Delayed presentation or diagnosis is often complicated by infection proximal to an obstructing calculus. Matrix calculi are most often infected. Renal impairment at presentation implies a complicating factor, such as underlying renal disease or septicemia. Rarely renal failure may be secondary to bilateral obstructing calculi or an obstructing stone in a single functioning kidney. Others present with vague symptoms or microscopic or gross hematuria, and in yet others the finding is incidental. Of this last group only one third will become symptomatic. There are no reliable predictive factors to identify those patients who will develop symptoms (2).

Imaging The purpose of imaging is not only to confirm the presence of urinary stone disease, but also to provide some indication of outcome.

Plain Abdominal Radiograph KUB has a low sensitivity in the diagnosis of stones, ranging from 45–60% (1). Despite this modest accuracy, the KUB is important in the management of a known radiopaque ureteric calculus especially in planning fluoroscopically guided, ▶extracorporeal shockwave lithotripsy (ESWL) or monitoring the progress of stone fragments after ESWL. It is also valuable for assessing stone status in those who are managed conservatively.

Excretory Urography Over many years, the excretory urography has been the traditional imaging modality of choice for evaluation of

S

1764


patients suspected of having ▶urolithiasis. However, small or radiolucent stones may be missed on IVU. Reported sensitivities range from 64–97% for calculus detection (3). The IVU is easy to perform and relatively save and the radiation dose from a standard three film IVU series is 1.5 mSv (1). In case of renal obstruction, additional series are required and the radiation dose may increase up to 10 mSv (4). The radiographic findings are highly accurate in diagnosing renal obstruction and allow excellent evaluation of the topography of the calyceal anatomy. These are most important contributions to management of urinary stone disease (Fig. 1).

proximal ureter or at the vesico-ureteric junction (VUJ). However, by using the color Doppler twinkle sign the sensitivity of calculus detection may be improved (Fig. 2). The role of ultrasound in chronic stone disease is also unclear. Pelvicalyceal anatomy cannot be confidently delineated on ultrasound unless the pelvicalyceal system is dilated. For follow-up of renal stones triaged to conservative therapy, ultrasound is of use. Ultrasound is also recommended in patients in whom radiation exposure is a concern, such as pregnant or pediatric patients (3).

Unenhanced Computed Tomography Ultrasound Ultrasound has an important role in the diagnosis and management of urinary tract stones but it has limitations. The sensitivity for calculus detection ranges from 37–64% and for detection of acute obstruction from 74–85% (3). Stones in the pelvicalyceal system can be reliably identified only if they are greater than 5 mm. Furthermore, calculi in the ureter are poorly visualized unless sited within the

The unequivocal advantage of UCTover all other techniques is its diagnostic accuracy. Over 99% of calculi, including those that are radiolucent on the KUB will be seen on UCT. The exceptions are pure matrix stones and stones made of indinavir (and related drugs), a human immunodeficiency virus (HIV) protease inhibitor. UCT has the highest accuracy (approximately 95% compared with around 80% for IVU) in acute ureteric colic (3). As well as

Stone Disease, Urinary. Figure 1 On KUB a 1 cm lower calyceal stone is depicted in this patient with uretero-pelvic junction (UPJ) stenosis as shown by the excretory urography. Since the passage of stones or fragments through this stenosis is unlikely, this particular patient is no candidate for ESWL. On unenhanced CT, the presence of a low or medium graded UPJ stenosis may be missed.

Stone Disease, Urinary. Figure 2(a–c) Patient presenting with a small calculus at the uretero-vesical junction (UVJ) (arrow) that is nicely depicted on unenhanced CT (a), and hardly delineated (arrow) on grey scale ultrasound (b). The color Doppler twinkle sign improves detection of urinary stones as shown in this patient with a 4 mm calculus at the UVJ (c).


demonstrating the size and site of the calculus, measurement of stone density may be useful, as stones of greater than 1,000 HU appear to respond less well to ESWL (1). A number of secondary CT signs thought to reflect the pathophysiology of acute urinary tract obstruction have been described (3). It has been suggested that these signs indicate a protection mechanism against renal pressure. The value of UCT in the management of renal, as distinct from ureteric calculi, is less clearly established. Factors other than stone size that influence the choice of ESWL versus ▶percutaneous nephrolithotomy (PCNL) are not easily assessed. Similarly, planning the route for percutaneous entry is imprecise. Multiplanar and volume reconstructions may overcome these limitations (1). For the diagnosis of ureteral stones, the calculated mean effective radiation dose reported for standard-dose CT ranges from 3.5 to 10 mSv, for low dose CT from 1.5 to 3.5 mSv, respectively.

Multislice CT Urography CT data acquisition during urographic contrast enhancement for contiguous imaging of the entire upper urinary

1765

tract is termed “multislice CT urography” (MSCTU). Multiplanar reconstructions, maximum intensity projections (MIPs), and average intensity projections can be rendered from the volume datasets to view the urogenital tract. Similar to the excretory urography, MSCTU provides valuable information on the degree of renal obstruction and allows excellent evaluation of the topography of the calyceal anatomy. The radiation dose reported for MSCTU ranges from 3.9 to 22.6 mSv (4). This great variety may be explained by different examination protocols and to a less extent to scanner- and patient-related factors.

Magnetic Resonance Urography The current role of MRU in diagnosing urinary tract stones is limited to patients in whom other investigations are contraindicated. MIP reconstructions from threedimensional datasets can resemble a conventional IVU but the spatial resolution is inferior to alternative techniques. Although the site of obstruction may be demonstrated definitively identification of a stone may be difficult (5) (Fig. 3).

Additional Radiological Investigations

Stone Disease, Urinary. Figure 3 This macroscopic image shows a urinary tract stone of mixed composition (calcium oxalate and calcium phosphate) after stone extraction.

The frequency of retrograde ureterography/scopy has been markedly decreased for evaluation of suspected ureteric stones. It is occasionally used in the patient with persistent suspicion of ureteric stone and unclear findings on UCT or IVU (e.g., differentiation of phleboliths and calculi) (1) (Fig. 4). Nuclear medicine is useful in selected cases. Its principal role is in determining relative renal function to decide whether a minimally invasive therapy is appropriate for a stone-bearing kidney or if the patient should undergo nephrectomy. There is no defined

Stone Disease, Urinary. Figure 4(a–c) Patient with right lower urinary calculus (a, arrow) and multiple calcifications in the small pelvis. Differentiation of prevesical calculi and phleboliths (arrows) may be limited on unenhanced CT as shown in these coronal (b) and axial (c) images.

S

1766

Strangulating Obstruction

threshold, but 25 mm) of the ICA, VA, or even MCA or false aneurysms located more distally, occlusion of the parent artery with coils or balloon must be considered. A balloon test occlusion can show whether the permanent occlusion will be tolerated or not. In selected patients, parent vessel occlusion should be combined with bypass surgery. A new approach is the placement of a covered stent over the aneurysm neck.

S

1772

Stroke, Interventional Radiology

Stroke, Interventional Radiology. Figure 1 A 70-year-old woman with minor left hemispheric stroke and repeated TIA of her left MCA territory due to tight stenosis of the left M1-segment since 3 months despite antiplatelet therapy. (a) shows old and acute lesions on DWI in the end-supply areas of the left MCA. (b) time-to-peak map demonstrates delayed contrast inflow into the left MCA-territory. (c) DSA of the left ICA shows (ap-view) occlusion of the left MCA-trunk. (d) direct injection of contrast into the left MCA-trunk demonstrates that the obstruction is incomplete. (e) after passing the tight stenosis with a microguide wire, angioplasty with a 2 mm-balloon recanalizes the left MCA. No further TGIA occurred.

Subcapsular Hematoma, Splenic

These stents are, however, relatively stiff and not well suited to be advanced over the carotid siphon. Arterial vasoconstrictions typically occur 1 week after SAH and jeopardize cerebral tissue. Local arterial infusions of nimodipine and balloon angioplasty can improve blood flow and prevent infarction.

Cerebral Hematoma Dural and pial AVMs are among the causes of nontraumatic cerebral hemorrhages and should be considered in all patients in whom arterial hypertension, coagulation disorder, or amyloid angiopathy is not the obvious reason for the bleeding. Patients with hemorrhage from AVM have an increased risk for a recurrent bleeding and should be treated. It is widely accepted that endovascular treatment is the first choice for dural AVMs. Interventional neuroradiologists treat dural AVMs by occlusion of the fistula with glue or coils after transarterial or transvenous approach. Pial AVMs can be treated with open surgery and resection of the AVM nidus, with radiation of the nidus, or with embolization of the nidus. Because radiation does not immediately decrease the risk of recurrent hemorrhage, it is considered only for patients with small AVMs, which cannot be reached by neurosurgeons or neuroradiologists. A complete embolization of the AVM nidus should be achieved to decrease the risk of another brain hemorrhage. A complete embolization of the AVM nidus can be achieved in 10–20% of the patients only, so that embolization has to be combined with surgery in the majority of patients. Preoperative embolization can, however, diminish the extent of the nidus and thus decrease the risk of surgery in these patients. Embolization with glue has the risk of venous obstruction and subsequent hemorrhage, gluing the microcatheter within the feeding artery, or embolization of nutritive arteries with subsequent ischemia and stroke.

Bibliography 1.

2. 3. 4.

5.

SPACE Colloborative Group, Ringleb P, Allenberg J, Bru¨ckmann H et al (2006) 30 day result from the SPACE trial of stent-protected angioplasty versus carotid endarterectomy in symptomatic patients: a randomised non-inferiority trial. Lancet 368:1239–1247 Marks M, Wojak J, Al-Ali F et al (2006) Angioplasty for symptomatic intracranial stenosis: clinical outcome. Stroke 37:1016–1020 Furlan A, Higashida R, Wechsler L et al (1999) Intra-arterial Prourokinase for acute ischemic stroke. JAMA 282:2003–2011 Lindsberg P, Mattle H (2006) Therapy of basilar artery occlusion: a systematic analysis comparing intra-arterial and intravenous thrombolysis. Stroke 37:922–928 Molyneux A, Kerr R, Yu L et al (2005) International Subarachnoid Aneurysms Trial(ISAT) of neurosurgical clipping versus endovascular coiling in 2143 patients with ruptured intracranial aneurysms: a randomised camparison of effects on survival, dependency, seizures, rebleeding, subgroups, and aneurysm occlusion. Lancet 366:809–817

1773

Stromal Tumor Gastrointestinal stromal tumor is the current designation for the major subset of gastrointestinal mesenchymal tumors (gastrointestinal smooth muscle tumors); absence of expression of the gamma-smooth muscle isoactin gene correlates with malignancy. ▶Neoplasms Small Bowel

Stubbed Toe Fracture A dorsal Salter II metaphyseal corner fracture of a distal phalanx, equivalent to an open fracture. ▶Osteomyelitis, Neonates, Infants, Childhood: Including Septic Arthritis and Other Important Soft Tissue Infections/Abscesses

Sturge–Weber Syndrome A meningofacial angiomatosis. Although there have been familial cases, as a rule there is no evidence of heredity. A meningeal vascular malformation with cortical calcification is the typical finding on imaging. ▶Neurocutaneous Syndromes

Subaxial Cervical Arthritis Subaxial cervical arthritis is a feature of rheumatoid arthritis and spondarthropathies. It comprises discitis and apophyseal joint arthritis. ▶Rheumatoid Arthritis

Subcapsular Hematoma, Splenic A collection of blood with a lenticular shape that lies just below the splenic capsule. It can be caused by blunt abdominal trauma or may originate from a spontaneous splenic rupture. ▶Trauma, Splenic

S

1774

Subchondral Bone Cysts

Subchondral Bone Cysts Subchondral bone cysts are a prominent feature of osteoarthritis, occasionally forming as sequelae of bone injury. Radiographically the lesions are radiolucent, with a surrounding sclerotic margin. Joint-space narrowing and bone sclerosis are accompanying features. These lesions are also found in hemochromatosis. ▶Hemochromatosis, Skeletal

Sublingual Gland The sublingual glands are salivary glands in the mouth. They lie anterior to the submandibular gland under the tongue, beneath the mucous membrane of the floor of the mouth. ▶Inflammation, Chronic, Acute, Salivary Glands

Subluxation Subchondral Cysts Subchondral cysts can be found with or without a sclerotic rim. Sclerotic subchondral cysts occur in degenerative osteoarthritis, gout, and SLE. In chronic gout they tend to be large and in a metaphyseal location. In degenerative osteoarthritis the predominant location is eccentric epiphyseal. SLE cysts are tiny and thought to be micronecrotic foci. Arthritic cysts tend not to have the sclerotic rim. Additional signs should be considered to establish the final diagnosis. Small osteolytic lesions have to be differentiated. ▶Connective Tissue Disorders, Musculoskeletal System

Displacement of the femoral head in the hip that still leaves some surface contact with the acetabulum. ▶Dysplasia, Hip, Developmental

Submandibular Gland The paired submandibular glands (or submaxillary glands) are salivary glands located beneath the floor of the mouth. In humans, they account for 70% of the salivary volume. ▶Inflammation, Chronic, Acute, Salivary Glands

Subclavian Steal Syndrome ▶Steal Syndrome, Vertebral

SUI ▶Stress Urinary Incontinence (SUI)

Subcutaneous Rheumatoid Nodules Sump Syndrome Subcutaneous rheumatoid nodules can be seen in seropositive rheumatoid arthritis. They are characteristic but not very common and are typically located at bony prominences. There is a typical histologic appearance. ▶Rheumatoid Arthritis

Subependymal Hemorrhage ▶Hemorrhage, Intracranial, Neonates (Neuro View)

Uncommon complication of side-to-side choledochoduodenostomy performed to improve biliary drainage in cases of retained stones or biliary dilatation. The segment of common bile duct between the anastomosis and the ampulla of Vater in cases of malfunction may act as a stagnant reservoir or stump wherein stones, debris, or infected bile accumulate. Bacterial overgrowth inside the stump results in acute cholangitis. Clinically, it presents as recurrent biliary pain or pancreatitis. Imaging findings include debris or stones in the common bile duct that may

Swallowing Disorders

appear dilated, dilated pancreatic ducts, and changes due to pancreatitis, cholangitis, or liver abscesses. ▶Cholangitis

Superior Caval Vein Occlusion

1775

SVC Obstruction Narrowing or blockage of the SVC by malignant or benign pathology resulting in characteristic signs and symptoms. ▶Varices, Oesophagus

▶Thrombosis, Caval Vein, Superior

Swallowed Foreign Bodies Superior Vena Cava Syndrome The obstruction of venous drainage due to tumour infiltration into the superior vena cava. This leads to dilation of collateral veins in the upper part of the chest and neck; oedema and plethora of the face, neck and upper part of the torso, including the breasts; suffusion and oedema of the conjunctiva; breathlessness when supine; and CNS symptoms such as headache, visual distortion and disturbed states of consciousness. Although a dramatic clinical situation, this syndrome requires urgent but not emergency care. ▶Neoplasms Pulmonary

Suppurative Fluid Collection ▶Abscess, Renal

Supramesocolic Peritoneal Cavity This extends from the diaphragm to the transverse mesocolon and is divided by peritoneal reflections into the following major spaces: right subphrenic, right anteriorposterior subhepatic, left subhepatic, and lesser sac. ▶Peritoneal Collections

Surgical Treatment of Crohn’s Disease Treatment for patients with severe complications or for those not amenable to medical treatment. ▶Crohn’s Disease

▶Foreign Bodies, Gastrointestinal

Swallowing Disorders A. W UTTGE - HANNIG , C. H ANNIG Radiologie-Strahlentherapie-Nuklearmedizin Mynich Germany [email protected]

Definitions Swallowing disorders are due to a dyscoordination, a dysfunction or an anatomic alteration in the esophagus or at the cross way of the respiratory and the alimentary tract. The swallowing disorders range from minor nutrition disabilities to the incapacity to ingest food with the consequence of aspiration of food or liquids into the airways. The swallowing disorders may even cause voice disabilities and may be associated with ronchopathy.

Pathology/Histopathology During pharyngeal phase of deglutition in only 0,7 s 24 muscle groups governated by five cranial nerves transport the bolus from the oral cavity to the entrance of the esophageal tube. The speed of bolus transport in the pharynx is about 70 cm/sec. The reduction of the transport velocity in the esophagus is due to the progressive substitution of the striated musculature by smooth musculature. Therefore the speed of the peristaltic wave in the lower esophagus is reduced to 2–4 cm/sec (1). The swallowing reflex in adults is triggered at the arch of the fauces and the corresponding dorsal pharyngeal

S

1776


wall. In neonates, the trigger occurs physiologically in the valleculae, the substitutive area for adult swallowing. The trigger of the swallowing reflex happens when the critical soil of receptor information from the different types of receptors in the oral cavity, in the valleculae, in the piriform sinuses and even in the laryngeal vestibule is reached (2). Due to surgery, radio- or chemotherapy or in the case of infections and also physiologically in the geriatric age the number of these receptors decreases causing a delayed triggering of the swallowing reflex (3). The upper esophageal sphincter opens and closes regulated by the swallowing reflex. The sphincter function is modulated not only by influences of the swallowing center in the brainstem and cortico-bulbar afferences but also by humoral transmission due to disorders of the esophageal peristalsis. The esophagus is innervated by vagal afferences and by an intrinsic autonom system, the Meissner and Auerbach plexus. The propulsion of the bolus is normally performed by a primary peristaltic wave, which is regulated by a single swallowing trigger. Multiple trigger inputs during an ongoing swallowing peristalsis interrupt the first wave; this is the so-called “intradeglutitive swallowing inhibition.” The closure of the lower esophageal sphincter is depending from the synergism of its resting tonus and the correct position in the ferrule of the diaphragmatic hiatus. The opening of the lower esophageal sphincter is coordinated by the swallowing reflex. The failure of this function is caused by an alteration of the angle of Hiss, the intrinsic sphincteric resting pressure and the length of the infradiaphragmatic esophageal segment. The pathology ranges from tumors to neurological and muscular disorders. The functional disorders are

Swallowing Disorders. Figure 1

mainly due to paraphysiologic alterations in the swallowing act and in the protection of the airways. The malignant tumors in the oral cavity, the pharynx, and the esophagus are mainly squamous cell carcinoma, distally adenocarcinoma and rarely leiomyosarcoma. Benign lesions are generally not frequent and rage in order of occurrence from leiomyoma, cysts to fibroma. The neurological induced swallowing disorders are referable to stroke, cerebral trauma or posttherapeutic, and tumoral deficiencies. Neuromuscular disorders are caused by neuro-degenerative illnesses as ALS, multiple sclerosis, Parkinson’s Disease, Guillaume Barre´ or by myogen diseases like polymyositis, myastenia gravis, or inclusion bodies myositis. Functional disorders are frequently caused by a disturbed pharyngo–esophageal interaction like in GERD (Fig. 1) (3).

Clinical Presentation The symptomatology ranges from hoarseness, globus sensation, heartburn, bolus impaction to dysphagia and odynophagia. According to the definition of Vantrappen Hellemans dysphagia is present when the deglutition of food, but only rarely saliva, causes discomfort. The globus sensation instead describes a discomfort during deglutition of saliva, but not during food ingestion. Odynophagia is painful swallowing. Neurogenic disorders are often correlated with the aging of population. In nursing homes about 40% of the elderly suffer from a not treated swallowing disorder (4), a so-called silent aspiration. Eight percent of patients

Flow sheet of the pharyngo–esophageal interactions.


1777

surviving a stroke event die in the first year from aspiration pneumonia due to an unobserved swallowing disorder (5). GERD may induce hoarseness, laryngitis posterior, and reflux induced asthma.

Generally, colloids marked with 99mTc in liquid, semisolid, or solid bolus preparations are use for this type of studies.

Imaging

Diagnosis

For the analysis of pharyngeal swallowing disorders, the imaging frequence needs to be very high. This can be achieved by the video fluoroscopy or the digital spot imaging at a high frame rate more than 20 images/sec. The high-frequence cineradiography at 50 images/sec is now reserved for scientific elaborations, due to the relatively higher radiation dose needed. The frame-by-frame evaluation is of crucial importance for the diagnostic reading process. In the esophagus the high temporal resolution is not useful. A normal fluoroscopy with spot imaging is sufficient. Double contrast imaging with BaSO4 is normally applied in every swallowing study. Different preparations with different consistencies adapted to the patients complaints should be carefully chosen. For the mucosal imaging often the application of myorelaxant agent is needed. Special contrast mediums should be used in patients with suspected or clinical appear ant tracheal aspiration or with suspected fistulas to the bronchial system. An almost iso-osmolar water-soluble iodine contrast medium should be applied, since hyperosmolar solutions lead to a pulmonary edema and/or a vaso-vagal reaction when aspirated. The modern techniques like CT or MRI have still the huge disadvantage of the obliged prone or supine position. Only in special studies like the analysis of velopharyngeal dysfunctions and its surgical therapy planning dynamic CT or dynamic MRI are used in clinical research (6, 7). Diluted gadolinium-EDTA or plain water can be used in MRI studies.

Tumoral lesions are visible directly by obstructions or mucosal alterations and indirectly in amotil regions. The extra luminal growth of tumors and the lymph node staging is the domain of the MRI and/or CT. The functional or motility disorders are best examined by video fluoroscopy or digital spot imaging or by combined methods like radiomanometry. The description of the functional diagnostic findings can be classified according to the three swallowing phases: the oral, the pharyngeal, and the esophageal phase.

Nuclear Medicine First pharyngeal studies for the analysis of neurogenic disorders were performed for the presence of aspiration or nasal penetration (8). The use of a tracer-marked bolus for the esophageal motility disorder has a long tradition. Quantifications of motility disorders like achalasia, hypomotilities in scleroderma or segmental spasms and the quantification of refluxed material in GERD are possible with a low radiation exposure. The time of observation is not limited in comparison to fluoroscopy (9). Gastric emptying with radiotracers allows a more physiological assessment of gastric motility, especially in the case of gastroparesis with secondary gastro-esophageal reflux.

Oral Pathologies Functional disorders can be divided in neurologic/neuromuscular disorders with or without a tracheal aspiration and gastroenterological disturbances. In the oral cavity, the bolus preparation and the bolus compression can be disturbed due to dismotility or an atrophy of the tongue or due to a malocclusion of the velum. Posttherapeutic neurologic or functional disorders can be seen in treated cleft.

Pharyngeal Pathologies In the pharyngeal phase the predominant finding is a functional alteration of the upper esophageal sphincter caused by GERD or an idiopathic esophageal motility disorder. The delayed and incomplete opening and the premature closure of the upper esophageal sphincter are frequent findings in GERD or neutral mass-reflux (Fig. 2). The pressure elevation in the pharynx leads to diverticula, pouches, and pharyngoceles in patients with a muscular predisposition (3). The entrance of acidity in the pharynx gives origin to web-like scar rings or inflammatory mucosal irritations in the pharynx and in the laryngeal vestibule. Indirectly the delayed opening of the upper esophageal sphincter leads to a closure of the laryngeal vestibule, thus causing a laryngeal penetration of the bolus (10). The second entity corresponds to the aspirationrelated neurogenic or neuromuscular disorders. The division in a pre-, intra-, and postdeglutitive form of aspiration (Figs 3–5), this means an aspiration before, during or after the triggering of the swallowing reflex, give useful hints for an adequate conservative rehabilitation or a surgical therapy (11).

S

1778


Swallowing Disorders. Figure 2 esophageal sphincter.

Dysfunction of the upper

Dysphagia without aspiration is generally a precursor of aspiration associated dysfunctions. The dyscoordination of the events during a swallowing act is one of the first signs: The so-called cervical achalasia represents the impossible opening or the reduced opening of the upper esophageal sphincter which is due to a loss of interruption of the resting pressure of the sphincter (Fig. 6). The reduced activity in the pharyngeal constriction is often disturbed by myogenic disease like polymyositis, dermatomyositis, muscular amyloidosis, or postradiation or autoimmunologic myofibrosis. The delayed or incomplete opening or a premature closure of the UES might be due to a reduced laryngeal anterior and cranial movement by a deficiency of the extrinsic laryngeal musculature. The lack of a consecutive propulsion in the pharyngeal constrictor muscles, corresponds to the so-called Inclusion Body Myositis. Undefined muscular dyscoordinations might be seen in Parkinson’s disease, multiple sclerosis, in the initial stage of amyotrophic lateral sclerosis, and in chorea Huntington (12).

Swallowing Disorders. Figure 3 Predeglutitive aspiration: Sequence (a–f). Sixty-year-old patient with a media infarct left hemisphere. (a, b) Early leakage of CM in the valleculae and rec. piriformes (black arrowheads). (c, d) The CM enters in the hypopharynx and the laryngeal vestibule (black arrow) before the triggering of the swallowing reflex (white arrow) causing a tracheal aspiration.


1779

Swallowing Disorders. Figure 4 Intradeglutitive aspiration. Fifty-six-year-old patient after a stroke, (a, b) pharyngeal retention with a disturbed pharyngeal contraction and a simultaneous spasm the upper esophageal sphincter (radiologically only the diagnosis of a disturbed opening of the upper esophageal sphincter can be observed). The spasm must be evaluated manometrically.

Esophageal Pathologies The functional disorders can be differentiated in primary or secondary forms.

The so-called nutcracker-esophagus does not present a typical radiological feature.

Primary Forms

Secondary Forms

The best defined primary form is the esophageal achalasia, the incomplete or absent reflex induced opening of the lower esophageal sphincter (Fig. 7). A subdivision in three types, the hypomotile, the amotile, and the hypermotile form, is useful for the therapeutic approach. The diffuse esophageal spasm is the second manometrically defined entity, which can be observed in fluoroscopy as a long segmental contraction or etage-like contractions of the esophageal body (Fig. 8). The symptomatology can be easily be mistaken for a cardiac attack.

The most frequent is the reflux-associated motility disorder in GERD or large neutral mass refluxes. The lack of a primary peristalsis is the leading symptom. The influence on the pharyngeal phase is explained above. GERD is often associated with a constant or inconstant hiatal hernia (Fig. 9), diagnosed by the classical threefolds sign in prone position. The typical GERD-related mucosal changes are in stage I of esophagitis are the focal granulomatous mucosal alterations and microulcera in the distal esophagus; in stage II superficial ulcerations

S

1780


Swallowing Disorders. Figure 5 (a–f) Postdeglutitive aspiration. (a) Already in the resting phase before the swallowing a retention in the right rec. Piriformis can be observed. (b) Beginning of the pharyngeal contraction with extrusion of a pharyngozele of the right side and a CM retention in the right hemipharynx (black arrow). (c, d) During the unilateral opening of the upper esophageal sphincter transport of the CM retention from the right hemipharynx to the left side (black arrow). (f) Due to the return of the larynx to the resting position for respiration the pharyngeal capacity diminishes. Large aspiration of the CM stored in supraglottic position (black arrow).

Swallowing Disorders. Figure 6 Cervical achalasia due to a insufficient sphincter reflex triggered opening. Major retentions in the pharynx can be observed.

Swallowing Disorders. Figure 7 Brachyesophagus.

Amotile achalasia with


1781

frequently found in the elderly population suffering from diabetes or polyneuropathy for example. The secondary achalasia is mostly caused by a submucosal growth of a carcinoma of the esophagogastric junction. The differential diagnosis can be oriented by anamnesis and manometric findings, but the final decision remains reserved to biopsy.

Bibliography

Swallowing Disorders. Figure 8 Etagenspasm in the middle and lower third of the esophagus with a hiatal hernia.

Swallowing Disorders. Figure 9 Hiatal hernia with a Schatzki’s ring and gastroesophageal reflux.

in the lower third of the esophagus; in stage III larger confluating circular ulcerations, and in stage IV peptic stenosis. The thickened mucosal folds in the esophagus and the so-called feline esophagus with the cat skin like appearance can be observed in stage I (13, 14). The diagnosis “presbyesophagus” summarizes a lot of primary and secondary esophageal motility disorders

1. Dodds WJ (1988) Physiology of swallowing. In: Ravich WJ, Donner MW, Johnes B (eds) Second Symposium on Dysphagia. Logemann, Baltimore 2. Kennedy JG, Kent RD (1988) Physiological substrates of normal deglutition. Dysphagia 3(1):24–38 3. Hannig C (1995) Radiologische Funktionsdiagnostik des Pharynx ¨ sophagus Springer-Verlag Berlin Heidelberg und des O 4. Ekberg O, Hamdy S, Woisard V et al (2002) Social and psychological burden of dysphagia: its impact on diagnosis and treatment. Dysphagia Spring 17(2):139–146 5. Logemann JA (1988) Swallowing physiology and pathophysiology. Otolaryngol Clin North Am 21(4):613–623 6. Hannig Chr., Wuttge-Hannig A, Daschner H et al (1993) Pre- and postoperative evaluation of Cleft-patients by cineradigraphic imaging.; Scientific Programme and Abstracts, ECR’93 Springer International, 1993: 273 award for outstanding scientific presentation of this session 7. Wuttge-Hannig A, Beer A, Gebhardt A et al (2001) (Alternative methods for the diagnostic of deglutition) in Deglutologia Edit. Schindler O, Ruoppolo G, Schindler Omega Edizioni, Italy, Milano. 8. Galli J, Volante M, Parrilla C et al (2005) Oropharyngoesophageal scintigraphy in the diagnostic algorithm of laryngopharyngeal reflux disease: a useful exam? Otolaryngol Head Neck Surg May 132 (5):717–721 9. Katschinski M, Schro¨ttle W, Wuttge-Hannig A et al (2002) ¨ sophagusfunktionsszintigraphie: Indikation, Durchfu¨hrung und O Auswertung Empfehlungen des Arbeitskreises Neurogastroenterologie und Motilita¨t. Z Gastroenterol 40:1–5 10. Wuttge-Hannig A, Herrman M, Hannig C (2004) Evaluation of Swallowing Reflex in Stress Drinking in Videofluoroscopy: Is it a Tool for the Measurement of pharyngeal Dysfunction in Reflux Disease? ECR, Abstract book 11. Hannig C, Wuttge-Hannig A (2007) Erkrankungen des Oesophagus In: St. Feuerbach (ed.) Gastrointestinales System, In: J. Freyschmidt (ed.) Handbuch Diagnostische Radiologi, Chapter 3: 27–137; Springer-Verlag, Berlin, Heidelberg, New York. ISBN 978-3-54041418-6 12. Wuttge-Hannig A, Hannig C (1995) Radiologische Differentialdiagnose neurologisch bedingter Schlucksto¨rungen. Der Radiologe 35:733–740 13. Hannig C, Wuttge-Hannig A (2006) Radiologische Funktionsdiagnostik von Schlucksto¨rungen bei neurologischen Krankheitsbildern und bei therapierten onkologischen Kopf-Hals-Erkrankungen In: G. Bartolome, H. Schro¨ter-Morasch (ed.) Schlucksto¨rungen, Diagnostik und Rehabilitation, Chapter 6: 99–154; Elsevier Urban & Fischer, Mu¨nchen, Jena. ISBN 978-3-437-47160-5 14. Ekberg O (2003) The symptomatic esophagus and GERD. Acta Radiol Mar 44(2):119–120 15. Hannig C, Wuttge-Hannig A (2006) Anatomie des Schluckvorgangs In: G. Bartolome, H. Schro¨ter-Morasch (ed.) Schlucksto¨rungen, Diagnostik and Rehabilitation, Chapter 1: 1–14; Elsevier Urban & Fisher, Mu¨nchen, Jena. ISBN 978-3-437-47160-5

S

1782

Swan-neck Deformity

Swan-neck Deformity A typical deformity in late-stage rheumatoid arthritis with hyperextension of the proximal interphalangeal joint and flexion of the distal interphalangeal joint. ▶Rheumatoid Arthritis

Synovial Osteochondromatosis Multiple ossified or merely cartilaginous fragments in a joint, presumably of traumatic etiology. ▶Transient Synovitis

Synovial Sarcoma Synchondritis of the Symphysis or Manubriosternal Junction Synchondritis of the symphysis or manubriosternal junction is characterized by adjacent sclerosis and bony destruction. Symphyseal involvement is typically seen in ankylosing spondylitis. Sternal manifestations are seen in ankylosing spondylitis, psoriatic arthritis, and SAPHO syndrome. ▶Spondyloarthropathies, Seronegative

More prevalent sarcoma in younger age groups, often near to joints however not within. May have a variety of appearances from small and well circumscribed to large, solid, infiltrating with areas of hemorrhage and cystic formation. Some 30% have radiographic evidence of soft tissue calcification. Tumor of high malignancy with often slow growth. ▶Neoplasms, Soft Tissues, Malignant

Syringomyelia, Posttraumatic Syndactyly Fused fingers or toes with or without synostosis. ▶Congenital Malformations of the Musculoskeletal System

Syndesmophyte An intervertebral osteophyte with special radiologic appearance as a vertebral-based ossification in the exact location and form of the annulus fibrosus. It is seen in ankylosing spondylitis and in lumbar manifestations of reactive arthritis, enteropathic arthritis, and SAPHO syndrome. ▶Spondyloarthropathies, Seronegative

Syndromic Hepatic Ductular Hypoplasia ▶Congenital Malformations, Bile Ducts

Post-traumatic intramedullary cavity, often progressive, with a length that could vary from 2 vertebral segments to the whole spinal cord, isointense to CSF in all pulse sequences. ▶Spinal Trauma

Systemic Blood Supply to the Lung There are normal anastomoses between the pulmonary and systemic (bronchial and ▶non-bronchial arteries) circulations but any abnormal process creating obstruction, compression of pulmonary artery branches or destruction of the pulmonary capillary bed may induce a compensatory development of systemic supply. ▶Hemoptysis

Systemic Mastocytosis Mastocytosis is characterized by abnormal growth and accumulation of neoplastic mast cells. Bony changes are

Systemic-Enteric Drainage, Pancreatic

frequently shown and include osteoblastic changes in conventional radiographs. MRI is more sensitive in depicting bone marrow extension of the disease, but findings are not specific. ▶Myeloproliferative Disorders

Systemic Supply to PAVM Bronchial arteries or other systemic branches of the aorta or supra-aortic trunks (intercostal, internal mammary or inferior diaphragmatic arteries) may supply PAVM before or after embolization. This systemic supply is a possible source of hemoptysis. ▶Pulmonary Arteriovenous Malformations

Systemic-Bladder Drainage, Pancreatic

1783

are anastomosed with the recipient’s common or external iliac artery and vein, respectively, whereas the exocrine pancreatic secretions are drained by an anastomosis between the donor’s duodenum and the bladder. ▶Transplantation, Pancreatic

Systemic-Enteric Drainage, Pancreatic In pancreatic transplantation with systemic-enteric drainage the arterial graft and the portal vein are anastomosed with the recipient’s common or external iliac artery and vein, respectively, whereas the exocrine pancreatic secretions are drained by an anastomosis between the donor’s duodenum and a small bowel loop. In this technique the insulin is released in the systemic circulation, while the exocrine secretions drain physiologically in a bowel loop. ▶Transplantation, Pancreatic

In pancreatic transplantation with systemic-bladder drainage technique the arterial graft and the portal vein

S

T

TACE Transarterial chemoembolization (TACE) involves the periodic injection of a chemotherapeutic agent mixed with an embolic material into selected branches of the hepatic arteries feeding a liver tumor. ▶Chemoembolization

Talipes Equinovarus (Clubfoot) Pes adductus, metatarsus, varus, external rotation of the ankle joint and parallel angle between talus and calcaneus. ▶Congenital Malformations of the Musculoskeletal System

Target Population The age-eligible population for screening, e.g., all persons offered screening according to the policy of the program. ▶Screening, Breast Cancer

Target-specific Imaging Target-specific imaging is the imaging of specific surface molecules or structures using a contrast agent binding to this particular structure. After wash-out of the unattached agent from the vascular system, the bound contrast agent can be detected. Specific accumulation of the contrast agent can be obtained by molecular binding of the agent (e.g., via antibodies) or by intracellular uptake of the agent (e.g., phagocytosis). ▶Contrast Media, Ultrasound, New Clinical Development

Targeted Delivery ▶Local Drug and Gene Delivery with Microbubbles

Targeted Imaging ▶Targeted Microbubbles ▶Direct Imaging

Targeted Microbubbles Target-Specific Gas-Filled Microbubbles ▶Targeted Microbubbles

Target-Specific Probe ▶Molecular Probes, Optical Probes

P ETER H AUFF Research Laboratories, Bayer Schering Pharma, Berlin, Germany [email protected]

Synonyms Molecular imaging; Specific microbubbles; Specific targeting; Target-specific gas-filled microbubbles; Targeted imaging; Targeted specific ultrasound contrast agents

1786

Targeted Microbubbles

Definition Targeted imaging using ultrasound relies on ultrasound contrast agents (USCAs) to localize physiological or pathophysiological molecular or cellular processes. Two mechanisms are described for in vivo targeting of USCAs: passive and active. ▶Passive targeting is a nonspecific accumulation of microbubbles at the target site after their administration and does not require a shell labeling with specific ligands. ▶Active targeting (specific targeting) requires modification of the bubble shell to allow selective binding to cellular epitopes or other receptors of interest. In general, specific USCAs consist of a stabilized microbubble as signaling moiety and shell-surface-bound ligands (such as antibodies, peptides, polysaccharides, or aptamers) as binding moiety (Fig. 1).

Mode of Action Passive Targeting Three main mechanisms (phagocytosis, interaction with cell membranes, and ▶lymph flow transport, LFT) are known for passive targeting depending on the size of microbubbles, the chemical properties of the shell, the type of the encapsulated gas, the physiologic system, and the route of administration (Fig. 2, upper part). Phagocytosis: Cells of the ▶reticuloendothelial system (RES), which clears the body from microorganisms and damaged cells, take up stabilized microbubbles after their intravenous administration leading to their accumulation in organs such as liver, spleen, lung, lymph nodes, and bone marrow. Activated neutrophil leukocytes are able to phagocytose albumin- or lipid-shell microbubbles after their cell-surface adhesion via special integrins or complement mediated opsonization.

Interaction with cell membranes: Depending on the shell composition or their surface charge, microbubbles can be adhered to the surface of cells such as activated leukocytes or the ▶endothelium. It could be shown that interactions between leukocytes and lipid-microbubbles are mediated by serum complement, which could be accelerated by inclusion of the apoptosis marker phosphatidylserine into the lipid shell. On the other hand, the interactions between leukocytes and albumin microbubbles are mediated largely by leukocyte b2 integrin. In the presence of a strong negative shell-surface charge, nonspecific adhesion and accumulation of microbubbles on the vascular endothelium has been noted (1, 2). Lymph flow transport: Passive targeting of lymph nodes was demonstrated after interstitial injection of small microbubbles (≤1 mm) consisting of a low-solubility gas or air-microbubbles stabilized with a polymeric hardshell by using gray scale or color Doppler ultrasound. Interstitial injected microbubbles enter the lymph vessels through gaps between lymphatic endothelial cells or by transcellular endo- or exocytosis and is transported by the lymph flow to the respective regional lymph node.

Active Targeting Active targeted USCAs have been made specific to cellular epitopes or other receptors by means of coupling of respective ligands with the shell of stabilized microbubbles (Fig. 2, lower part). The attachment of targeting ligands to microbubble shells can be made either before or after the bubble preparation, depending on the composition of the USCA. This can be done either by direct coupling of the ligand to shell-forming molecules or by covalent or noncovalent attachment of ligands to preformed microbubbles. Because of the size of microbubbles (mm-range), they are not able to move through the endothelium and thus do not usually leave the

Targeted Microbubbles. Figure 1 Principle of target-specific microbubbles. The conjugation of a targeting moiety with the signaling moiety results in target-specific microbubbles.

Targeted Microbubbles

1787

Targeted Microbubbles. Figure 2 Strategies for passive and active targeted microbubbles. For passive targeting, intrinsic chemical or electrostatical properties of the shell can be used to carry microbubbles to the target. Active-targeted microbubbles can be produced by different coupling strategies depending on the targeting moiety (e.g., antibodies, peptides, polysaccharides, or aptamers).

vascular system after their intravenous injection. Therefore, selective targeting with microbubbles is generally directed (limited) to specific receptors of physiological or pathological conditions on the surface of the vascular endothelium. However, the endothelium in a human adult has a total mass of about 1,000 g (like the liver mass) and mediates a wide variety of messages from local tissue to the systemic system (e.g., inflammation to immune response). Once a targeted USCAs has been administered, the microbubbles circulate in the blood stream for several minutes and accumulate in the area of interest via a ligand–receptor interaction. Free circulating microbubbles will be cleared from the blood by the RES within 10–20 min (depending on the dose). The rapid blood clearance guarantees a strong signal-to-noise ratio for specific accumulated microbubbles at the target site (Fig. 3). The exceptional ultrasound sensitivity of microbubbles allows the detection of signals from single bubbles and their quantification at the target even in high concentrations.

the so-called late phase effect with Levovist® (Schering AG, Berlin, Germany) in the liver and spleen after their intravenous injection. The late phase effect (bubble retention) allows contrast imaging in these organs long after blood-pool enhancement has disappeared and is used for the detection of neoplastic lesions. Another example is the use of Imagent (Alliance Pharmaceuticals, San Diego, California) for passive lymph node targeting after their interstitial administration (3). The nonspecific targeting of activated leukocytes could be shown in preclinical settings with different types of microbubbles, which provides a new approach for ultrasound imaging, the detection of acute inflammation including ischemiareperfusion injuries. The targeting of myocardial endothelial cells has been achieved in anesthetized dogs with negative-charged microbubbles indicating that those types of agents could provide information on myocardial perfusion and viability (1, 2).

Active Targeting

Indication Passive Targeting Few of the commercially available USCAs are already suitable for passive targeting in humans. One example is

Although targeted imaging with specific USCAs is currently at an experimental preclinical stage, researchers were able to show the potential of this ▶molecular imaging approach in various animal models by targeting several molecular structures. The potential for

T

1788

Targeted Optical Contrast Agents

Targeted Microbubbles. Figure 3 Mode of action of target-specific microbubbles (principle). The rapid blood clearance of nontargeted MBs by cells of the reticuloendothelial system within 20 min, depending on the dose used, guarantees a strong signal-to-noise ratio for targeted MBs. Moreover, it provides the opportunity of repeated investigations within one session using MBs targeted to the same or other endothelial cell receptors.

the depiction of receptors in healthy tissue (physiological conditions) could be demonstrated in mouse and dog lymph nodes after intravenous injection of L-selectine ligand-specific microbubbles (3). Targeting of microbubbles to molecules which are overexpressed in certain pathological conditions, such as inflammation (P-selectin, ICAM-1, VCAM-1), ischemia-reperfusin injury (P-selectin), thrombosis (GPIIb/IIIa), or angiogenesis (av-integrins) was achieved in several experimental animal studies (1, 4, 5). However, each of the broad variety of molecules expressed on the endothelial surface can potentially be targeted with specific microbubbles. Only recently, a new method has been reported allowing the in vivo quantification of targeted microbubbles even in high concentrations. It could be shown that this new technique, referred to as sensitive particle acoustic quantification (SPAQ), allows a quantitative molecular imaging which has been demonstrated by counting ICAM-1 and VCAM-1 in an inflammatory model under treatment conditions (5).

Bibliography 1.

Klibanov AL (2005) Ligand-carrying gas-filed microbubbles: ultrasound contrast agents for targeted molecular imaging. Bioconjug Chem 16:9–17

2.

3.

4.

5.

Lindner JR, Song J, Xu F et al (2000) Noninvasive ultrasound imaging of inflammation using microbubbles targeted to activated leukocytes. Circulation 102:2745–2750 Hauff P, Reinhardt M, Briel A et al (2004) Molecular targeting of lymph nodes with L-selectin ligand-specific US contrast agent: a feasibility study in mice and dogs. Radiology 231:667–673 Heppner P, Lindner JR (2005) Contrast ultrasound assessment of angiogenesis by perfusion and molecular imaging. Expert Rev Mol Diagn 5:447–455 Reinhardt M, Hauff P, Linker RA, et al (2005) Ultrasound derived imaging and quantification of cell adhesion molecules in experimental autoimmune encephalomyelitis (EAE) by sensitive particle acoustic quantification (SPAQ). Neuroimage 27:267–278

Targeted Optical Contrast Agents ▶Molecular Probes, Optical Probes

Targeted Probes ▶Molecular Probes, Optical Probes

Temporal Bone, Inflammatory Diseases, Acute, Chronic

Targeted Specific Ultrasound Contrast Agents ▶Targeted Microbubbles

Targeted Tumor Imaging ▶Receptor Studies, Neoplasms

TCC ▶Transitional Cell Carcinoma

Telescope Phenomenon Telescope phenomenon is a characteristic aspect of inflammatory mutilation in late stages of psoriatic arthritis, with destruction not only of the epiphysis but also of the meta- and diaphysis of the phalanges, resulting in excessive shorting of the digit. ▶Spondyloarthropathies, Seronegative

Temporal Bone, Inflammatory Diseases, Acute, Chronic G IOVANNI C ARLO E TTORRE , PAOLA M ILILLO Istituto di Radiologia, Universita` degli Studi di Catania, Catania, Italy [email protected]

Synonyms Acute cholesteatoma; Chronic otitis media

Definition Inflammatory diseases of the temporal bone indicate all those inflammatory affections that involve the structures

1789

of the external, middle, and inner ear, including the associated complications and the sequelae. The infections can be of microbial or viral origin and include a series of clinical entities for which several classifications have been proposed. The first distinction must be made between acute and chronic inflammatory, where chronic is defined as an inflammation that does not resolve itself within three months from its appearance. The inflammatory diseases of the temporal bone are generally classified according to the site of origin: external ear, middle ear and mastoid, inner ear, and petrous apex. The pattern of inflammation is strictly related to regional anatomy.

Pathology/Histopathology Acute otomastoiditis is an inflammation of bacterial origin. In acute middle ear and mastoid inflammation the mucopurulent fluid is a result of vasodilatation, increasing glandular secretion with mucous production, and polymorphonuclear reaction occurring from the neutrophil cells. Resolution frequently occurs, but, if for some reason, the condition is prolonged, such as inability of the secretions to be drained out of the eustachian tube, the number of glands and goblet cells increases. The areas formerly covered by a cuboidal or flat epithelium change into areas of a less differentiated pseudostratified columnar epithelium. Granulation tissue results from the chronicity of the inflammatory process. Localized areas of the mucosa become hyperplastic with invasion of fibroblasts, capillaries, macrophages, plasma cells, and lymphocytes. Acute otomastoiditis can cause complications such as coalescence and erosion of the bony septa with osteomyelitis, subperiosteal abscess, labyrinthitis, petrous apicitis (Gradenigo’s syndrome), thrombosis of the sigmoid sinus, intracranial empyema, meningitis, or cerebral abscess. Chronic middle otitis can be distinguished into two main groups according to the integrity of the tympanic membrane: (a) otitis with integrity of the tympanic membrane, which generally represents serous-mucous otitis; (b) otitis with open tympanum, which includes chronic suppurative otitis, chronic noncholesteatomatous otitis with ossicular erosion, and cholesteatomatous otitis. In serous-mucous chronic otitis the aerial cavities of the middle ear are filled by a transuded serous and serousmucous fluid that can change into an exudate due to bacterial superimposition. Chronic noncholesteatomatous otitis is characterized by a tympanic perforation with otorrhea and a range of irreversible modifications that can involve the mucous membrane, the temporal bone, the ossicular chain, the mastoid air cells, and the eustachian tube.

T

1790

Temporal Bone, Inflammatory Diseases, Acute, Chronic

Cholesteatoma is the evolution or the complication of chronic middle otitis. It consists of a cyst-like mass lined with stratified squamous epithelium, usually keratinizing, and filled with desquamating debris often including cholesterol crystals. It behaves as an expansive lesion whose slow growth determines erosion and discharge of bony fragments. Sequelae or variants of chronic noncholesteatomatous otitis are polypoid otitis, adhesive or fibroadhesive otitis, and tympanosclerosis. External malignant otitis is the most significant pathological process of the external ear. It is a serious infection, which in 90% of cases is caused by Pseudomonas aeruginosa, a bacterial organism that secretes lithic enzymes and toxins that favor the spreading of the inflammatory process into the bone and the adipose spaces of the cranial base, often along the vascular– nervous route. Acute labyrinthitis is an inflammation of the fluidfilled space of the inner ear and the membranous labyrinth. Abnormal ossification can occur in the chronic phase (ossificans labyrinthitis).

Clinical Presentation The inflammatory process of the middle ear is generally the consequence of acute or chronic inflammatory affections of the main airway and almost always it also involves the mastoid (otomastoiditis). Inflammatory disease of the ear is easily assessed clinically: conductive hearing loss, otalgia (earache), and secretion represent the typical clinical signs of the inflammatory process of the external and middle ear. The symptomatology of acute labyrinthitis is characterized by vertigo and sensorineural hearing loss; tinnitus can sometimes be present. In malignant external otitis, together with general symptoms of inflammation, the local symptomatology consists in strong pain during movements of the temporomandibular joint due to solicitations of the tympanic membrane. The illness usually affects immunocompromised and/or elderly subjects.

High-resolution computed tomography (HRCT) of the petrous bone is widely employed in the evaluation of bony structures and it has supplanted conventional radiography. The advent of spiral and multidetector CT (MDCT) has been a further progress in the field. It allows information to be obtained about the bony tissue and the content of the middle ear cavity. HRCT scan with multiplanar reformation (MPR) today represents the most accurate technique for the diagnosis of external and middle inflammatory diseases. Magnetic resonance imaging (MRI) has a limited role in this field of pathology, and is generally employed in the above-mentioned intracranial complications of otomastoiditis, in acute external malignant otitis, in acute labyrinthitis, and for postsurgical patients. In chronic noncholesteatomatous middle otitis with ossicular erosion, the CT features are characterized by a thickening of the mucosal layer of the middle ear with subtle decalcification and erosion of part of the ossicular chain (Fig. 1). In cholesteatomatous otitis the main CT features include: a nondependent soft tissue mass in the middle ear cavity with displacement and erosion of the auditory ossicles as well as erosion of the anterior and posterior tegmen tympany and scutum. In some advanced cases, erosion of the lateral wall of the lateral semicircular canal can be found (labyrinthine fistulas). Involvement of the intrapetrous facial nerve can occur. All the information obtained by means of HRCT is pivotal in presurgical planning (Fig. 2). In external malignant otitis, CT represents an emergency examination because of the dramatic and rapid spreading of the disease.

Imaging Acute otomastoiditis does not need evaluation by means of diagnostic imaging, with the exception of rare forms that are accompanied by intracranial complications. Chronic inflammatory diseases of the middle ear represent the main field of application of radiological imaging.

Temporal Bone, Inflammatory Diseases, Acute, Chronic. Figure 1 Chronic noncholesteatomatous middle otitis: coronal CT scan. Partial erosion of the malleus and polypoid hyperplasia of the internal wall of the tympanic membrane (arrow).

Temporomandibular Joint Disorders

1791

In ossificans labyrinthitis the absence of signal of the labyrinthine fluid on T2-W MR images is the main clue for the diagnosis.

Nuclear Medicine Bone scintigraphy does not have a role in the assessment of acute and chronic diseases of the temporal bone.

Diagnosis Temporal Bone, Inflammatory Diseases, Acute, Chronic. Figure 2 Cholesteatoma of the middle ear: the coronal CT scan shows erosion of the scutum (arrow) and of the head of malleus.

In almost all cases of inflammatory disease of the external and middle ear, the diagnosis is based on the clinical and otoscopy data. Diagnostic imaging is indicated in all those situations in which suitable information is not available for correct therapeutic planning and in local or cranial complications such as labyrinthine fistulas and/or intracranial spreading of the disease. In cholesteatomatous disease it is well known that microscopic fiberoptic otoscopy allows an accurate evaluation of the tympanic cavity only, and is not able to assess the involvement of the attic, the antrum, and the mastoid.

Interventional Radiological Treatment Interventional radiology does not have a role in the treatment of acute and chronic diseases of the temporal bone.

Bibliography 1. 2.

Temporal Bone, Inflammatory Diseases, Acute, Chronic. Figure 3 External malignant otitis: axial contrastenhanced CT scan. Inflammatory infiltration of adipose spaces (arrow) and bony erosion of the mandibular condyle.

3.

Aberg B, Westin T, Tjellstrom A (1991) Clinical characteristics of cholesteatoma. Am J Otolaryngol 12:254–258 Ettorre GC (1992) Diagnostica Per Immagini Dell’orecchio. Ed Idelson Liviana, Naples Sanjeev S, Pitkin L (2005) Middle Ear, Inflammatory Diseases. eMedicine.com Inc

T The main CT features are: erosion of the bony wall of the external ear conduct and middle ear and infiltration of neighboring soft tissues. Involvement of the base of the skull and the temporomandibular joint can also be demonstrated. The inflammatory infiltration of the adipose spaces and the subcutaneous planes of the face can be better demonstrated by means of contrastenhanced CT and/or MRI (Fig. 3). The diagnosis of acute labyrinthitis is based on MR spin-echo (SE) T1-weighted (T1-W) sequences with contrast medium that shows focal increasing of the labyrinth membrane signal.

Temporomandibular Arthropathy ▶Internal Derangement, Temporomandibular Joint

Temporomandibular Joint Disorders ▶Internal Derangement, Temporomandibular Joint

1792

Teratocarcinomas

Teratocarcinomas ▶Neoplasms, Chest, Childhood

Definition Teratoma is a tumor consisting of tissues foreign to the site of origin, originating from all three germ layers. Teratomas have solid and cystic components. Overall most teratomas are benign lesions, but malignant degeneration is always possible, increasing with age. Calcification is present in two-thirds of patients.

Teratoma Incidence A benign germ cell tumor with mature tissues derived from pluripotential cells of all three germinal layers. It is also known as dermoid cyst. Extragonadal teratomas are rare and location in the pancreas is unusual. These tumors are usually large and consist of unilocular or multilocular cysts filled with solid components, including sebaceous material, hair, teeth, calcium and skin. Microscopically the cysts are lined by ciliated or squamous epithelium, while the wall contains dermal appendages (sebaceous glands and hair follicles) and sometimes other tissues (teeth, cartilage, bone). At US examination the lesion has a polymorph appearance because of different prevalence of the three histological structures; in most cases dermoid cysts are complex masses with both liquid and solid components. They contain hyperechoic material (due to hair, teeth and fat) and hypoechoic areas due to fluid material. On CT imaging areas with low density, corresponding to fat, mixed with high density structures can be suggestive of dermoid cyst; a fat-fluid level can be frequently seen. On MR, dermoid cysts present a variable appearance, depending on their fat and fluid content. ▶Cystic Neoplasms, Pancreatic

Teratoma, Childhood J OHAN G. B LICKMAN 1 , D EWI A SHI 2 1

Department of Radiology, UMC St Radboud Nijmegen, The Netherlands 2 Department of Radiology, Faculty of Medicine, University of Indonesia, Cipto Mangunkusumo General Hospital, Salemba 4, Jakarta Pusat, Indonesia [email protected]

Synonym Germ cell tumor

1. Approximately 1 of every 40,000 live births. 2. The tumor has a 4:1 female preponderance ratio. 3. When identified at birth, the neoplasms are more aggressive in boys.

Pathology/Histopathology Ectodermal, mesodermal, and endodermal tissue.

Clinical Presentation Teratoma may be found in any part of the body. ▶Sacrococcygeal teratoma is the most frequent site (40% of all cases). The majority of cases are apparent at birth. They present as a mass at the level of the buttocks and extent below the gluteal fold. ▶Presacral teratoma: pain, constipation, and urinary frequency or signs of urinary tract obstruction. There is weakness of the lower extremities if it invades spine. Ovarian teratomata are often discovered incidentally. Symptoms may include a mass, abdominal pain, and abnormal vaginal bleeding. Bladder symptoms and gastrointestinal (GI) tract pain are less frequent. Acute symptoms may occur due to torsion. Testicular teratomata presents as a painless scrotal mass and/or a hydrocele. Mediastinal teratomata are often asymptomatic. When they become large, tracheal/bronchial compression resulting in chest pain, coughing, and wheezing may occur. Hyperglycemia or precocious puberty due to ectopic production of insulin or sex hormones is possible.

Imaging The classic imaging appearance shows the presence of all three tissues types: bone, fat, soft tissue. Conventional: soft tissue mass, scoliosis, mediastinal widening, canal/foramina widening, pedicle erosion, abnormal calcification.

Teratoma, Ovaries, Mature, Ovalar

2.

3.

4.

1793

Carty H (2001) Paediatric Imaging. In: The Encyclopaedia of Medical Imaging, Vol. 7. Published by The NICER Institute, Oslo, pp 520–521 Cohen HL, Haller JO (2004) Abnormalities of the female genital tract. In: Kuhn JP, Slovis TL, Haller JO (eds) Caffey’s Pediatric Diagnostic Imaging. 10th edn. Mosby, Philadelphia, pp 1960–1963 ▶http://www.emedicine.com/med. Teratoma, Cystic. Last update September 17, 2004

Teratoma, Ovaries, Mature, Ovalar R OSEMARIE F ORSTNER Department of Radiology, Paracelsus Private Medical University, Salzburger Landeskliniken, Austria, Salzburg [email protected] Teratoma, Childhood. Figure 1 Sagittal CT-reconstruction shows a large mass originating from the sacral area containing fat and calcifications with enhancement after contrast injection.

Synonyms Benign teratoma; Cystic teratoma; Dermoid; Dermoid cyst

Ultrasound assists in determining the cystic, or partial cystic, part of the tumor and the extent of the tumor. Computed tomography (CT) often reveals the diagnosis, as it demonstrates fat and/or calcifications, their extent, and any bone destruction. After contrast medium administration, the solid component may show enhancement (Fig. 1). Magnetic resonance imaging (MRI) is superior in delineating the extent of this tumor. Occasionally, if there is blood within the cyst, it may show a fluid level.

Nuclear Medicine Nuclear medicine is used for the evaluation of bone metastasis if the tumor shows malignant degeneration.

Definitions Ovarian teratomas comprise tumors composed of mature or immature tissues of primordial germ cell origin. They can be classified into three main categories: ▶mature teratomas, ▶immature teratomas, and monodermal teratomas (1). The mature cystic teratoma which constitutes 99% of these tumors contains mature elements of ectodermal, mesodermal, and endodermal origin. The mature cystic teratoma also known as dermoid cyst or dermoid is the most common ovarian neoplasm of women younger than 45 years. Although teratomas may occur at any age, they typically affect a younger age group than epithelial tumors, and are the most common ovarian masses in children (2). Approximately 10% of teratomas occur bilaterally. ▶Malignant degeneration of benign teratomas is rare (1–2%), and usually occurs in the sixth to seventh decade of life (2).

Diagnosis Laboratory : Elevated levels of serum alpha fetoprotein (AFP), and beta human chorionic gonadotropin (HCG). Biopsy : To differentiate benign from malignant tumor.

Bibliography 1.

Barnes PD (1998) Neuroimaging. In: Hans Blickman (ed) Pediatric Radiology: The Requisites. 2nd edn. Mosby St. Louis, pp 294–295

Pathology/Histopathology Sixty percent of cystic teratomas measure 5–10 cm in size (1). Macroscopically, benign teratomas present most often as unilocular tumors surrounded by a firm capsule of varying thickness (1). They contain fatty, sebaceous contents, and hair. The former is liquid at body temperature and semisolid at room temperature (1).

T

1794


Squamous epithelium covers the inner wall of the dermoid. Hair follicles, skin glands, muscles, and fat are located within this wall. Arising from the cyst wall is commonly one or more protuberances, the dermoid plug or Rokitansky nodule, composed of a variety of different tissues. Hair, bone, or teeth and rarely cartilage as well as phalanges may be encountered in this region (1).

Clinical Presentation The majority of benign teratomas remain clinically asymptomatic. They are often discovered incidentally during clinical examinations or imaging studies. Especially larger teratomas may cause nonspecific symptoms including pelvic pain, abdominal fullness, abdominal swelling, and dysuria. Rarely, abnormal uterine bleeding and hemolytic anemia may be associated with a dermoid (1). Teratomas may become clinically apparent due to complications which include in order of decreasing frequency ▶torsion (16%), chemical peritonitis caused by ▶rupture, and infection (3). Mature teratomas tend to grow slowly, with an average diameter of 1.8 mm reported per year (2). Rarely, giant dermoids may be encountered, which grow to a size of up to 40 cm (1). Rapid growth of a dermoid in a postmenopausal woman is a sign of malignant degeneration (2).

Teratoma, Ovaries, Mature, Ovalar. Figure 1 Dermoid cyst in a 20-year-old patient: transvaginal sonography shows a well-delineated oval adnexal mass containing multiple homogeneous fine echoes and a solid protruding nodule, the Rokitansky nodule. In this patient the latter did not contain calcifications or teeth, and may be misdiagnosed as a solid element in ovarian cancer. (Courtesy of Dr. R. Gruber, Salzburg)

Imaging Conventional radiographs are not suited for the diagnosis of teratomas. Rarely, teeth, bone, or wall calcifications may be detected on abdominal plain films. Sonography is the primary imaging modality for assessing suspected adnexal masses. The combination of transabdominal and endovaginal sonography combined with Doppler findings allows optimal characterization of the morphology and perfusion of adnexal masses. However, at ultrasound teratomas display a broad spectrum of imaging findings. They most commonly present as diffusely or partially echogenic masses with acoustic shadowing due to hair balls, calcifications, teeth, or the Rokitansky protuberance (Fig. 1). Layered lines and dots as well as fat-fluid levels include other findings. However, imaging findings are often noncharacteristic, and the dermoid plug may be misinterpreted as a solid lesion (4) (Fig. 1). In the case of unclear sonography findings, computed tomography (CT) and magnetic resonance imaging (MRI) allow reliable demonstration of intralesional fat, which is the hallmark of benign teratomas.

Teratoma, Ovaries, Mature, Ovalar. Figure 2 Dermoid cyst in CT. Noncontrast CT scan shows a 6.5-cm left adnexal mass that is located adjacent the uterine corpus. Its fatty density values are pathognomonic of a dermoid. The lesion contains areas of higher density representing floating hair and mural calcifications. The homogeneous thickening of the wall was caused by wall edema due to torsion.

In CT, attenuation values indicative of fat (−20 to −140 HU) are the pathognomonic finding (Fig. 2). Bone, teeth, and cyst wall calcifications can also be reliably detected. Contrast-enhanced CT assists in the demonstration of complications of dermoids, for example, torsion or malignant degeneration. In MRI, detection of fat is also the key for the diagnosis of teratomas. In suspected dermoids, tailored studies consisting of T1-weighted imaging (WI) and T1-WI with frequency-selective fat saturation allow confirmation of dermoids in the majority of cases.


1795

Suppression of the high-signal-intensity sebaceous contents or fat within the lesions is diagnostic (2). Gradientecho opposed-phase imaging demonstrates a fat–water interface within the lesion. Furthermore, chemical shift artifacts in the frequency-encoding direction may assist in detecting fat and in its differentiation from hemorrhage (2). Contrast-enhanced MRI aids in further characterization in atypical cases and in the assessment of complications of teratomas.

Nuclear Medicine Nuclear medicine has a very limited role in the assessment of teratomas. It may be useful only in detecting monodermal teratomas, for example, struma ovarii or carcinoids.

Diagnosis The diagnosis of teratomas is based on the demonstration of sebaceous material or fat within an adnexal mass (Figs 2, 3). The size of teratomas ranges from 0.5 to 40 cm, with the majority measuring less than 10 cm. As CT or MRT are more sensitive for fat than sonography, they are complementary in indeterminate cases or performed for the assessment of complications caused by dermoids (2). In CT, demonstration of fat within a cystic adnexal lesion with or without calcifications is the pathognomonic finding of a benign teratoma. A fat–fluid interface may be produced by floating hair. In dermoid cysts, fat is demonstrated in more than 90% of cases, teeth in 31%, and calcifications in the dermoid plug, or more often in the cyst wall, in 56% (5) (Fig. 2). In MRI, the fatty sebaceous contents display typically very high signal intensity (SI) on T1-WI (2). The SI on T2-WI can be variable. As hemorrhagic adnexal masses, endometriomas and hemorrhagic cysts in particular may display similar signal intensities. Loss of signal on T1-WI with fat saturation allows the specific diagnosis of a dermoid. Approximately 8% of benign teratomas do not show these straightforward findings in CT or MRI (3). In these patients, small foci of fat may be found in the dermoid cyst wall or in the dermoid plug (Fig. 3). When dermoids show no fat they cannot be differentiated from other cystic and solid ovarian tumors, including ovarian cancer. Malignant degeneration in mature cystic teratomas is typically due to carcinomatous or sarcomatous malignant transformation. Its prevalence is less than 1–2%, and it occurs in the sixth to seventh decade of life (2). CT and MRI findings include the presence of a dermoid with fatty

Teratoma, Ovaries, Mature, Ovalar. Figure 3 Benign teratoma with small amounts of fat. In a 15-year-old girl, transaxial T1-WI (a) and T1-WI with fat saturation (b) show a large predominantly cystic lesion in the upper pelvis. At its left periphery it contains a mural nodule that shows intermediate to high signal intensity on T1-WI. The corresponding T1 fat saturation image shows fat suppression of an ovoid area smaller than 1 cm in diameter. At surgery, a 15-cm cystic dermoid was found, the Rokitansky nodule contained hair and only small areas of fat.

contents and a heterogeneous solid component extending transmurally outside the capsule or enhancement of the Rokitansky nodule (3). In children dermoid cysts may be associated with an ipsi- or contralateral immature teratoma. The latter presents typically as a large, often solid mass that may contain tiny foci of fat or calcifications.

Bibliography 1.

2. 3.

Talerman A (2002) Germ cell tumors of the ovary. In: Kurman RJ (ed) Blaustein’s Pathology of the Female Genital Tract. Springer, Berlin, Heidelberg, NY, pp 967–1033 Outwater EK, Siegelman ES, Junt JL (2001) Ovarian teratomas: tumor types and imaging characteristics. Radiographics 21:475–490 Rha SE, Byun JY, Jung SE et al (2004) Atypical CT and MRI manifestations of mature ovarian cystic teratomas. Am J Roentgenol 183:743–750

T

1796

4.

5.

Teratomas

Quinn SF, Erickson S, Black WC (1985) Cystic ovarian teratomas: the sonographic appearance of the dermoid plug. Radiology 155:477–478 Buy JN, Ghossain MA, Moss AA et al (1989) cystic teratoma of the ovary: CT detection. Radiology 171:697–670

Teratomas ▶Neoplasms, Chest, Childhood

muscle atrophy, decreased or hyperactive reflexes, urinary incontinence, spastic gait, and orthopedic deformities such as scoliosis or foot and hip deformity. ▶Congenital Malformations, Spine and Spinal Cord

THAD ▶Transient Hepatic Attenuation Differences

Thoracic Aneurysm Tertiary Hyperparathyroidism ▶Aneurysm, Aortic and Thoracic An autonomous production of PTH by the parathyroid glands, which was originally induced by a metabolic disorder with secondary hyperparathyroidism. ▶Hyperparathyroidism

Testicular Torsion Torsion of the testis with twisting of the spermatic cord. Strangulation of the vessels will cause ischemic damage to the testis. It presents as acute scrotal pain and needs immediate surgical treatment to avoid loss of the testis or of testicular function. ▶Scrotal Disorders

Tethered Cord Often confused with the tight filum terminale, a CSD characterized by a short, rigid filum terminale. This common belief is entirely inappropriate, while the term should in fact be used to indicate a clinical condition, the tethered cord syndrome (TCS). TCS occurs as a consequence of traction on a low-lying conus medullaris with progressive neurological deterioration due to metabolic derangement, and may ensue as a complication of myelomeningocele repair or as the presentation of several forms of CSD, including spinal lipomas, the tight filum terminale, diastematomyelia, and caudal agenesis. The clinical picture of TCS includes sensorimotor dysfunction,

Thoracic Aorta The thoracic aorta carries blood away from the heart and distributes it to the head, the arms, and passes into the abdomen where it becomes the abdominal aorta. ▶Aneurysm, Aortic and Thoracic

Thoracic Aortic Aneurysm ▶Aneurysm, Aortic and Thoracic

Thoracic Outlet Syndrome Upper extremity symptoms that are caused by of the compression of the neurovascular bundle of the upper extremity as it exits the bony thorax. ▶Brachial Ischemia

Thrombocytopenia ▶Thrombopoietic System, Diseases of the

Thrombopoietic System, Diseases of the

Thrombolysis Intravenous infusion or intra-arterial local application of agents with the potential to resolve blood clots. ▶Stroke, Interventional Radiology

Thrombopoietic System, Diseases of the O TMAR S CHOBER Department of Nuclear Medicine University Hospital Mu¨nster Mu¨nster, Germany [email protected]

Synonyms Thrombocytopenia

Definition Thrombocytopenia is defined as low platelet counts (sixth decade Spinal arteries (ASA, PSA) M = F Second and third decade Spinal arteries (ASA, PSA) M = F Second and third decade Spinal arteries (ASA > PSA) Children, Young adults no AV shunt F>M fourth and fifth decade

Clinical course

Treatment

chronic myelopathy acute; hemorrhage common

1. emboliz. 2. surgery 1. emboliz. 2. surgery

acute; hemorrhage Common

1. emboliz. 2. surgery

acute or chronic; may hemorrhage acute or chronic; may hemorrhage

1. emboliz. 2. surgery Surgery

SDAVF-Spinal Dural Arteriovenous Fistula. SCAVM-Spinal Cord Arteriovenous Malformation. SCAVF-Spinal Cord Arteriovenous Fistula. CM-Cavernous Malformation.

of low signal intensity representing iron storage products completely surrounds the lesion. The findings are characteristic and often permit specific diagnosis. The surrounding parenchyma may have abnormal signal intensity as a result of gliosis, edema, or syrinx. Following acute hemorrhage, the MRI appearance may be less specific and other differential considerations may require consideration. With typical MRI features, angiographic evaluation is unnecessary. MRI findings in spinal cord ischemia are similar regardless of the etiology of the infarction (9). Cord signal abnormality and subtle enlargement in the acute stage are best seen in proton density and T2-weighted images. Signal abnormality may involve only the gray matter but often extends throughout the entire cross section of the cord. Enhancement may occur, especially involving the gray matter. In the chronic stages, cord atrophy may be present.

3. 4. 5.

6.

7. 8. 9.

Gilbertson J et al (1995) Spinal dural arteriovenous fistulas: MR and myelographic findings. AJNR 16:2049–2057 Bowen BC, Saraf-Lavi E, Pattany PM (2003) MR angiography of the spine: update. Magn Reson Imaging Clin N Am 11(4):559–584 Berenstein A, Lasjaunias P (1992) Surgical Neuroangiography Vol 5: Endovascular Treatment of Spine and Spinal Cord Lesions. Vol 5. Springer-Verlag, Berlin Heidelberg New York: pp 1–109 Gueguen B et al (1987) Vascular malformations of the spinal cord: Intrathecal perimedullary arteriovenous fistulas fed by medullary arteries. Neurology 37:969–979 Ogilvy C, Louis D, Ojemann R (1992) Intramedullary cavernous malformations of the spine. Neurosurgery 31(2):219–230 Sandson T, Friedman J (1989) Spinal cord infarction: report of 8 cases and a review of the literature. Medicine 68:282–292 Mawad, M et al (1990) Spinal cord ischemia after resection of thoracoabdominal aortic aneurysms: MR findings in 24 patients. AJNR 11:987–991

Vascular Disorders, Hepatic

Interventional Treatment Endovascular occlusion of SDAVF should be considered at the time of the diagnostic angiogram. Surgical treatment with coagulation or resection of the site of the nidus and surrounding dura also represents viable therapy for many patients.

M ASSIMO B AZZOCCHI 1 , G IUSEPPE C OMO 1 , M ARZIO P ERRI 2 1

Department of Diagnostic and Interventional Radiology, University of Udine, Udine, Italy 2 Department of Diagnostic and Interventional Radiology, University of Pisa, Pisa, Italy [email protected]

Bibliography 1. 2.

Spetzler RF et al (2002) Modified classification of spinal cord vascular lesions. J Neurosurg 96(Suppl 2):145–156 Symon L, Kuyama H, Kendall B (1984) Dural arteriovenous malformations of the spine Clinical features and surgical results in 55 cases. J Neurosurg 60:238–247

Synonyms Hepatic vascular pathologies; Primary circulatory liver disorders


Definition Vascular disorders of the liver are several different pathological conditions including ▶portal vein thrombosis, ▶Budd–Chiari syndrome, ▶veno-occlusive disease (VOD), ▶peliosis hepatis, ▶arterio-venous malformation, and ▶congestive cardiac cirrhosis.

1943

remain patent. It is usually associated with the use of chemotherapeutic agents or radiation and secondary to progressive inflammation with deposition of coagulation factors, red cells, and hemosiderin-laden macrophages within terminal hepatic venules resulting in a nonthrombotic venule occlusion (4). Histologic features consist of centrilobular sinusoidal fibrosis followed by necrosis of the zone 3 hepatocytes.

Pathology and Histopathology Portal Vein Thrombosis Etiological factors of portal vein thrombosis are slow flow secondary to cirrhosis, direct invasion by cancer, inflammatory changes in pancreatitis, sclerosing cholangitis, and abdominal infections, coagulation disorders, and surgical interventions. In chronic occlusion, the portal vein is retracted and dilatated, periportal collateral veins (▶portal cavernoma) may be observed, while the chronic occlusion of a branch of the portal vein is usually accompanied by segmental atrophy and compensatory hypertrophy of other segments (1).

Budd–Chiari syndrome The Budd–Chiari syndrome is a hepatic venous obstruction at any level from the small hepatic veins to the junction of the inferior vena cava and the right atrium. Outflow obstruction caused by hepatic VOD and cardiac disorders is excluded from this definition (2). Distinction is made between primary and secondary forms. The primary type is caused by endoluminal venous lesion (thrombosis, webs, endophlebitis) and is usually associated with underlying thrombogenic conditions. The secondary type includes obstruction originating from a lesion outside the venous system (tumor, abscess, cysts) and compressing or infiltrating the hepatic venous system. Venous compression can be complicated by thrombosis. Both in the primary and secondary forms, obstruction of the hepatic venous outflow tract results in increased hepatic sinusoidal pressure and portal hypertension. Hepatomegaly is usually seen. However, in later stage of Budd–Chiari syndrome peripheral atrophy associated with caudate and central hypertrophy is usually observed. Centrilobular necrosis is histological observed. In chronic cases the affected areas become fibrosed and shrink in size (2, 3).

Veno-Occlusive Disease Hepatic VOD is a common complication of bone marrow transplantation (BMT). It is characterized by obstruction of small and medium-size intrahepatic veins due to vasculitis. It differs from Budd–Chiari syndrome in that the major hepatic veins and inferior vena cava primarily

Peliosis Hepatis Hepatic peliosis is a rare benign pathological entity characterized by focal, multifocal, segmental, or diffuse dilatation of liver sinusoids. At gross examination multiple cyst-like, blood-filled cavities within hepatic parenchyma are observed (1). Although several theories, including outflow obstruction of the blood flow at the sinusoidal level, hepatocellular necrosis, or direct lesions of the sinusoidal barrier, have been postulated to explain its etiology, the pathogenesis of peliosis hepatis is still unclear.

Clinical Presentation Portal Vein Thrombosis In most cases, portal vein thrombosis occurs slowly and silently. Biochemical tests of the liver are normal or only mildly elevated. Ascites and encephalopathy are uncommon. Splenomegaly and esophageus varices are the most common clinical findings and gastrointestinal hemorrhage for portal hypertension is usually the first manifestation of portal vein thrombosis. Rarely, thrombosis extends from the portal vein to the mesenteric arcades, leading to bowel ischemia and infarction.

Budd–Chiari Syndrome Since the extent and rapidity of the hepatic vein occlusion and of the development of venous collateral circulation is highly variable, fulminant, acute, subacute, or chronic clinical presentation may be observed (3). The acute and fulminant forms are associated with an acute thrombosis of the main hepatic veins and the etiologic factors are related to hypercoagulability of blood. Patients rapidly develop hepatomegaly, intractable ascites, jaundice, and abdominal pain leading to liver failure and hepatic encephalopathy. The subacute and chronic types are most common and have an insidious onset; since the development of venous collateral vessels, ascites, and hepatic necrosis may be minimal and splenomegaly or esophagogastric varices may represent the only clinical findings. The degree of aminotransferase, alkaline phosphatase and

V

1944


bilirubin levels elevation is variable, depending on the rapidity of the onset. Patients with Budd–Chiari syndrome who have deteriorating liver function and complications usually need to undergo liver transplantation.

Veno-Occlusive Disease VOD occurs in over half of BMT patients. Risk factor include preexisting liver disease or liver function test abnormalities and underlying infection or antibiotic use during conditioning chemotherapy. Jaundice, weight gain, and ascites occurring within the first 3 weeks after BMT represent the typical clinical manifestations. At laboratory exams, elevated transaminases and hyperbilirubinemia (direct bilirubin >2) are observed (1, 4). In severe cases, multiorgan failure with pulmonary infiltrates, pleural effusions, congestive heart failure (CHF), and renal failure may occur.

Peliosis Hepatis Peliosis hepatis commonly occurs in patients treated with anabolic steroids, corticosteroids, or oral contraceptives. The disease is associated with chronic wasting diseases such as tuberculosis, hematological malignancies, diabetes, acquired immunodeficiency syndrome (AIDS), posttransplant immunodeficiency, and exposure to toxic agents (1, 5). Hepatic peliosis is usually asymptomatic and focal liver lesions are discovered incidentally. However, in rare cases, the lesions may be revealed by major complications such as liver failure, portal hypertension, or hemorrhage.

Imaging Portal Vein Thrombosis Ultrasound (US) features are echogenic thrombus within the lumen of the vein, demonstration of portal vein collateral circulation, enlargement of the thrombosed segment of the vein, and the cavernomatous transformation of the portal vein (Fig. 1). Color Doppler US is a useful tool for detection of portal vein thrombosis (Fig. 2). It is also useful for differential diagnosis between nonneoplastic and neoplastic thrombosis in presence of hepatocellular carcinoma. The demonstration of arterial flow within the thrombus is specific for malignancy. Nonenhanced computed tomography (CT) may show focal high density in enlarged portal system in acute thrombosis while chronic venous thrombosis appears as linear areas of calcification within the thrombus. A filling defect in the vessel lumen during the portal venous phase represents the typical finding of portal vein thrombosis on contrast-enhanced CT (Fig. 2). In a patient with hepatocellular carcinoma enhancement of the thrombus during the arterial phase is suggestive for neoplastic thrombosis. Typically, a regional parenchymal hyperdensity due to a compensatory increase in arterial blood flow to the segments poorly perfused by the portal vein is observed during the arterial phase while the locally decreased portal vein perfusion results in a hypodense appearance on the portal-venous phase. On magnetic resonance (MR), acute thrombosis manifests as absent flow signal void within a normal or dilated vein, generally associated with edge enhancement of the thrombosed

Vascular Disease, Spine. Figure 1 US scan of hilum hepatis: for chronic occlusion, the portal vein is retracted and we can see dilated periportal collateral veins (portal cavernoma).


1945

Vascular Disease, Spine. Figure 2 A filling defect in the vessel lumen during the portal venous phase represents the typical finding of portal vein thrombosis on contrast-enhanced CT. We can see axial scan and coronal MPR reconstruction. At US scan features are echogenic thrombus within the lumen of the vein with enlargement of the thrombosed segment.

vein because of blood flowing around the thrombus, or inflammatory response of the venous wall. The clot is isointense or hyperintense on T1-weighted images, and hyperintense on T2 (1). Moreover, using angiographic sequences (contrast-enhanced 3D MR portography), MR provides detailed information not only about the location and length of portal vein obstruction but also about portal collateral pathways.

Budd–Chiari Syndrome Doppler US is the modality of choice for diagnosis of Budd–Chiari syndrome, with a sensitivity of more than 85% (3). A spectrum of hepatic venous abnormalities (hepatic veins tortuosity, thrombosis, flow reversal, and monophasic flow waveform) may be observed and the association with intrahepatic or subcapsular portosystemic shunts and venous collaterals is suggestive for the diagnosis (2). US may demonstrate the occlusion of the hepatic veins due to thrombus or tumor as well as the narrowing of the inferior caval vein due to a web, appearing as an echogenic structure without acoustic shadowing (Fig. 3). Finally, US may demonstrate hepatomegaly, ascites, and parenchymal changes such as necrotic or fibrotic areas. CT scan and MR imaging allow a most extensive and accurate evaluation of the parenchymal changes. A heterogeneous enhancement of the liver parenchyma, hypertrophy of the caudate lobe, direct findings of hepatic venous occlusion and extrahepatic findings such as ascites, splenomegaly, and portosystemic collateral vessels

Vascular Disease, Spine. Figure 3 US scan demonstrates the occlusion of the hepatic veins that appear as echogenic structures.

represent the most common features of Budd–Chiari syndrome on both CTand MR contrast-enhanced imaging. Moreover, MR, by the use of angiographic sequences, allows also an extremely accurate evaluation of the vascular abnormalities and is a highly sensitive tool for the diagnosis of hepatic veins thrombosis (3).

V

1946


Veno-Occlusive Disease

Budd–Chiari

The diagnostic imaging study of choice is US with Doppler modalities. Nonspecific US findings have been described, including ascites, gallbladder wall thickening, and hepatosplenomegaly. Moreover, the absence of obstruction of hepatic veins or inferior cava at Doppler US or MR imaging enables differential diagnosis between Budd– Chiari syndrome and VOD (4).

The diagnosis of Budd–Chiari syndrome is established only upon demonstration of an obstructed hepatic venous outflow tract. Doppler US has a sensitivity of more than 85% and should be the first line of investigation, while the association of Doppler US and MR allow the diagnosis in almost all cases and an optimal delineation of the venous obstruction is usually obtained (2). In a minority of patients, mostly individuals with cirrhosis, liver biopsy may be required to firmly establish the diagnosis and to rule out other processes such as VOD and cirrhosis of other etiologies. Direct venography is no longer considered necessary. However, since it allows a precise delineation of outflow obstruction and pressure measurements, venography should be performed when percutaneous or surgical shunting is considered.

Peliosis Hepatis Imaging findings are variable depending on differing pathological patterns and on often concomitant liver steatosis. On US, numerous poorly defined foci of varying hypoechogenicity may be observed; however, these findings are nonspecific and may simulate hepatic metastasis, cirrhosis, hemangiomatosis, or adenomatosis. The CT and MR findings may differ according to the size of the lesions, the extent of communication with the sinusoids, the presence or absence of thrombi within the cavities, and the presence or absence of hemorrhage (5). However, multiple small hepatic lesions with variable enhancement represent the most common findings, while larger and even solitary lesions are uncommon. The lesions appear usually as irregular, hypodense regions on unenhanced CT scan and, occasionally, hyperdense spots for focal hemorrhage may be observed. After contrast medium administration, a slight peripheral enhancement is usually seen. A typical finding of hepatic peliosis is the lack of mass effect on hepatic vessels. On MR, the multiple hepatic foci in peliosis hepatis show a high signal on T2-weighted images and a variable signal on T1-weighted images, presumably reflecting various stages of subacute hemorrhage. A late and slow but intense enhancement is observed on contrastenhanced MR imaging.

Diagnosis Portal Vein Thrombosis Portal vein thrombosis is usually asymptomatic and diagnosis is mainly done by imaging. Doppler US is the first imaging modality used. In patients who are candidates for surgery, that is, for portosystemic shunt procedures, a more accurate imaging evaluation covering the whole portal venous system is required. Traditionally, assessment of the portal venous system has been achieved with intraarterial splenoportal and mesentericoportal angiography, but, recently, the value of MR angiography as another noninvasive procedure for the assessment of the portal venous system has been increasingly recognized.

Venous-Occlusive Disease Diagnosis can usually be made based on clinical criteria. Direct bilirubin 2 mg/dL, hepatomegaly and weight gain due to fluid accumulation occurring within 20 days of transplantation are the typical clinical findings. The diagnosis can only be made when hyperacute graftversus-host disease, sepsis, cardiac failure, and tumor infiltration have been ruled out. US can exclude tumor infiltration and biliary tract disease and confirm hepatomegaly and ascites while color Doppler US, MR imaging, and angiography can accurately determine the level of involvement enabling the differentiation between Budd– Chiari syndrome and VOD (4). Finally, transvenous liver biopsy is required in patients with thrombocytopenia posttransplantation to differentiate VOD from acute graft-versus-host disease.

Peliosis Hepatis Imaging findings are often not sufficiently specific and histology is the only certain way of making the diagnosis. However, the risk of performing a needle biopsy in peliosis hepatis is unclear and severe hemorrhage and even fatal outcome following needle biopsy have been reported.

Bibliography 1. 2.

3.

Schneider G, Grazioli L, Saini S (eds) (2002) MRI of the Liver. Berlin, Heidelberg: Springer Janssen HL, Garcia-Pagan JC, Elias E et al (2003) European group for the study of vascular disorders of the liver. Budd–Chiari syndrome: a review by an expert panel. J Hepatol 38(3):364–371 Menon KV, Shah V, Kamath PS (2004) The Budd–Chiari syndrome. N Engl J Med 350(6):578–585

Vascular Disorders, Gastrointestinal Tract

4. 5.

Mortele KJ, Van Vlierberghe H, Wiesner W et al (2002) Hepatic venoocclusive disease: MRI findings. Abdom Imaging 27(5):523–526 Steinke K, Terraciano L, Wiesner W (2003) Unusual cross-sectional imaging findings in hepatic peliosis. Eur Radiol 13:1916–1919

Vascular Disorders, the Gastrointestinal Tract A LBAN D ENYS Centre Hospitalies Universitire Vandois Lausanne, Lausanne, Switzerland [email protected]

Synonyms Angiodysplasia; Arteriovenous malformation; Vascular ectasia

Definition Vascular disorders of the gastrointestinal tract include occlusive, aneurysmal diseases, and congenital malformations of the arterial or venous mesenteric vessels. These disorders are rare and difficult to diagnose and therefore account for many malpractice claims. Symptoms can be divided into two groups: ischemic and hemorrhagic. Ischemic diseases can lead to chronic or acute intestinal ischemia. The small bowel and the right colon are the parts of the digestive system most sensitive to ischemia. Etiologies of acute mesenteric ischemia can be divided into arterial occlusive diseases, venous thrombosis, and nonocclusive arterial diseases (1). Arterial occlusions are related either to embolism in the superior mesenteric artery, mainly from a cardiac origin, or to thrombosis of the superior mesenteric artery. Venous thrombosis of the superior mesenteric veins is either primary, due to coagulation disorders, or related to various abdominal diseases favoring venous thrombosis, such as cirrhosis and intraperitoneal infectious diseases. Nonocclusive mesenteric ischemia means that there is an insufficient oxygen supply to the digestive wall despite patent mesenteric vessels. As a result of systemic hypoperfusion (cerebral and/or cardiac), mesenteric vasospasm occurs to preserve cardiac and cerebral blood flow. Therefore, this condition is encountered in severely ill patients with advanced cardiac insufficiency and is favored by the use of vasoconstrictive agents. Chronic mesenteric ischemia occurs due to occlusion or stenosis of at least two of the three digestive arterial trunks, including the superior mesenteric artery (celiac

1947

trunk, superior, and inferior mesenteric artery). The main cause is atheroma (2). Gastrointestinal bleeding can be caused by many pathologies, including cancers and inflammatory diseases, but some vascular diseases such as ▶angiodysplasia can cause recurrent bleeding in elderly patients (3). Angiodysplasias are acquired capillary vascular malformations encountered mainly in the cecum (78%), jejunum (10%), ileum (5%), and stomach (4%) in elderly patients.

Pathology/Histopathology Acute mesenteric ischemia can lead to various degree of ischemia of the bowel wall, from transient mucosal ischemia to irreversible bowel wall necrosis. Arterial etiologies first induce digestive mucosal ischemia, which extends progressively to the muscularis mucosa and to the serosa. In the early phase, the intestine is pale and spastic; in later phases, the loops are dilated, with a gray appearance and edematous infiltration of the mesocolon. In the late phase, the loop is atonic and dilated, with focal hemorrhage in the bowel wall and focal areas of necrosis. Thirty minutes of ischemia are sufficient to produce necrosis. Venous etiologies are characterized by early mesenteric edema and bowel wall hemorrhage. Thrombosis of distal venous branches is mandatory for producing ischemia. In chronic mesenteric ischemia, histopathology of the digestive wall is usually normal. Analysis of the main digestive trunks shows various degrees of occlusion of the main trunks. Occlusion of the celiac trunk is usually proximal, favored by compression from a median arcuate ligament, and is found in 15–24% of patients. Occlusion of the superior mesenteric artery is usually found in the first 5 cm after its origin. Angiodysplasia is found in the submucosa and is characterized by one or multiple dilated submucosal veins and a cluster of capillaries and small arteries.

Clinical Presentation Acute mesenteric ischemia has various clinical presentations. The most common symptom is a sudden pain localized in the periumbilical region of the right iliac fossa. This pain is usually severe and induces agitation. It occurs in older patients (mean age 70 years) with multiple cardiac and vascular comorbidities. Cardiac disease, myocardial infarction, and cerebrovascular disorders, and congestive heart failure are found in 91%, 27%, 23%, and 45% cases of mesenteric embolism, respectively. The association of sudden abdominal pain with multiple cardiovascular events in the previous

V

1948

Vascular Disorders, Gastrointestinal Tract

history is highly suggestive of acute mesenteric ischemia. If the diagnosed is delayed, peritonism as well as shock can be encountered. Patient survival is inversely correlated with diagnostic delay. Clinical symptoms of chronic mesenteric ischemia include so-called ▶intestinal angina. Patients with intestinal angina complain of periumbilical abdominal pain that starts approximately 30 min after eating and lasts for 1–2 h. The patient may recognize the relationship between eating and abdominal pain, resulting in a fear of eating and subsequent weight loss. Some patients, however, will present with anorexia and weight loss without complaints of abdominal pain. Symptoms may be less specific, such as diarrhea or vomiting, resulting in delayed diagnosis. Angiodysplasia is usually silent, but bleeding may occur in up to 10% of cases. Depending on the lesion’s localization and the volume of blood loss, symptoms may range from hematemesis to melena as well as hematochezia or only symptoms of chronic anemia.

Imaging For the diagnosis of acute mesenteric ischemia, the most critical issue is to save time and not delay treatment. Plain X-ray films are usually done at the onset of acute abdominal pain. Although useful for eliminating other causes of abdominal pain such as nephrolithiasis or bowel perforation, they are insufficient for diagnosing acute mesenteric ischemia. Ultrasound (US) and Doppler of mesenteric vessels are nonspecific and very difficult to perform in this situation and should not be done. Computed tomography (CT) and, more recently, multidetector CT of the abdomen are now recommended (4). They allow visualization both of bowel wall ischemia (wall thickening with submucosal edema) and necrosis (lack of enhancement of the bowel wall after injection of contrast media, pneumatosis; Fig. 1a) and of its vascular etiology, with an embolus visible as a filling defect in the lumen of the superior mesenteric artery (Fig. 1b) or a thrombosis of the superior mesenteric artery or vein. If CT does not provide sufficient information for treatment, angiography can be performed to visualize mesenteric vessels. Imaging of chronic mesenteric ischemia relies mainly on Doppler US. In thin patients, this technique is able to depict stenoses of the superior mesenteric artery, celiac axis, and inferior mesenteric artery with excellent sensitivity. The diagnosis of hemodynamically significant stenosis is based on velocity measurements at the origin of these trunks. A velocity of over 275 cm/sec in the superior mesenteric artery and 200 cm/sec in the celiac axis is specific. The relationship between the stenoses and the pain usually remains unresolved until correction of the

Vascular Disorders, the Gastrointestinal Tract. Figure 1 A 67-year-old woman who presented with sudden acute abdominal pain. (a) Computed tomography slice after injection of contrast media shows a dilated jejunal loop without enhancement of the bowel wall (white arrow), suggesting bowel wall ischemia. (b) Zoom view of the mesenteric vessel shows occlusion of the superior mesenteric artery by an embolus (bold arrow).

stenoses, either by intravascular methods or surgical bypass techniques. Angio-CT and magnetic resonance angiography have also been advocated, but their performance does not seem superior to Doppler US. Diagnosis of angiodysplasia is usually done by endoscopy, but during bleeding episodes, lesions can be obscured by the intestinal contents and blood clots. Mesenteric angiography is the method of choice, showing a nidus typically located on the antimesenteric border of the right colon that is associated with early filling of a large adjacent drainage vein. The feeding artery is inconsistently enlarged. Bleeding becomes visible as a pooling area of contrast persistent on late images when bleeding exceeds 0.5 mL/min. CT has recently been proposed to demonstrate these lesions. On the arterial phase after injection, the filling defect as well as the large draining vein can be visible. The performance of CT in detecting angiodysplasia is unknown.

Vascular Imaging and Vascular Interventional Radiology

Nuclear Medicine Nuclear medicine has no role in diagnosing acute or chronic mesenteric ischemia, but it has been proposed for diagnosing chronic bleeding of unknown origin. Technetium-99m-labeled red blood cell scintigraphy has proved to be efficient in localizing the bleeding site if a pooling is visible in the first 2 h after injection when bleeding exceeds 0.1 mL/min. This technique allows the bleeding site to be localized but does not give information about the bleeding etiology. There is no specific sign on technetium-99m-labeled red blood cell scintigraphy for angiodysplasia.

1949

stenoses and not occlusions. Stent placement allows similar symptom relief as well as a long-term symptom-free survival rate.

Bibliography 1. 2.

3. 4.

Oldenburg WA, Lau LL, Rodenberg TJ et al (2004) Acute mesenteric ischemia: a clinical review. Arch Intern Med 164(10):1054–1062 Cognet F, Ben Salem D, Dranssart M et al (2004) Chronic mesenteric ischemia: imaging and percutaneous treatment. Radiographics 22(4):863–879; discussion 879–880 Krevsky B (1997) Detection and treatment of angiodysplasia. Gastrointest Endosc Clin N Am 7(3):509–524 Taourel PG, Deneuville M, Pradel JA (1996) Acute mesenteric ischemia: diagnosis with contrast-enhanced CT. Radiology 199(3):632–636

Diagnosis The diagnosis of acute mesenteric ischemia should be suspected when an acute onset of abdominal pain is associated with a previous history of a cardiovascular disorder in an elderly patient. Diagnosis is usually done by CT because both bowel wall ischemia and the cause of the ischemia are visible. Nowadays, mesenteric angiography is reserved for equivocal cases and/or associated treatment. When signs of peritonism are concomitant, laparoscopy, or laparotomy can be proposed directly without preoperative imaging. The diagnosis of chronic mesenteric ischemia relies on the association of typical symptoms associated with weight loss and Doppler signs of mesenteric vessel stenoses or occlusions. Because symptoms are aspecific (abdominal pain and weight loss are encountered in many abdominal cancers) and stenoses of two mesenteric vessels can be completely asymptomatic due to collateral vessel recruitment, the final diagnosis can be made only after having treated the stenoses if symptoms disappears afterward. The diagnostic strategy of lower gastrointestinal bleeding is based on the results of endoscopy. Gastroscopy and colposcopy are typically always performed. Angiodysplasia usually presents at endoscopy as an intense red stellate flat lesion. If the bleeding site is not localized and bleeding persists, mesenteric angiography can be proposed to detect the lesion and embolize the feeding artery.

Interventional Radiological Treatment Endovascular treatment of acute mesenteric ischemia is the treatment of choice if type B aortic dissection is the etiology. In other situations, endovascular methods should be discussed on a case-to-case basis. The use of thrombolytic medication is not without risk if necrosis of the bowel wall is already present. Endovascular treatment of chronic mesenteric ischemia has now become widely accepted when the lesions are

Vascular Ectasia ▶Vascular Disorders, the Gastrointestinal Tract

Vascular Imaging and Vascular Interventional Radiology D IERK VORWERK Department of Diagnostic and Interventional Radiology, Klinikum Ingolstadt, Ingolstadt, Germany [email protected]

Imaging of vascular structures was a very early goal of radiologists before researchers were able to introduce fast imaging facilities, fluoroscopy, nontoxic contrast media, and suitable selective catheter techniques into radiology, which were all necessary for the depiction of arteries and veins. After definition of these invaluable tools, conventional angiography became one of the early revolutionizing techniques that was added to plain-film radiography, and until the development of computer tomography (CT), it was one of the most important tools for the specific diagnosis of cerebral and abdominal problems. In the meantime, diagnostic angiography—albeit refined by digital subtraction techniques—has been widely replaced by cross-sectional imaging in the diagnosis of organ-related diseases and will be more and more replaced by CT angiography, magnetic resonance (MR) angiography, and duplex ultrasound for many purely arterial or venous problems. There are, however, some special diagnostic problems that still require selective angiography for detailed

V

1950

Vascular Origin Malignant Tumors

diagnosis, such as cerebral vascular dysplasias or searching for sources of bleeding. Selective arteriography is still the guiding tool for intravascular interventional radiology, and to date there is no alternative available for this purpose. Anatomically, vascular imaging includes arteries, veins, arteriovenous connections such as hemodialysis fistulas and grafts, large visceral veins, and pulmonary veins and arteries. Classical imaging modalities for vascular imaging are nonsubtracted and subtracted angiography and phlebography preferably with direct contrast medium injection. Intravenous digital subtraction angiography (DSA) is no longer accepted as a sufficient diagnostic tool, with very few exceptions. Further imaging tools are duplex sonography with its modifications such as color-coded and power duplex, CT angiography, and MR angiography. In arteries, pathological patterns include acute disease with arterial rupture, perforation, embolic or thrombotic occlusion, and formation of pseudoaneurysms. Chronic arterial disease shows chronic occlusions, atherosclerotic stenoses, and chronic aneurysms as well as inflammatory disease. In veins, acute and subacute thrombosis, rarely stenoses, collateral pathways, and inborn variations and changes are found. In hemodialysis fistulas and grafts, stenoses and thrombotic occlusions as well as aneurysm formation are relevant findings. The choice of the primary imaging tool depends on the location of the disease. In intracranial cerebral arteries, MR angiography or CTangiography are the modalities of choice, and selective angiography is used for specialized indications or in connection with an interventional procedure. The same is true for the intracranial venous structures. Extracranial cerebral arteries are diagnosed with duplex sonography, MR or CT angiography, and selective angiography, particularly in preparation for interventions. In the thorax, imaging of the large arteries and veins is a domain for CT and MR angiography. Small vessels such as bronchial arteries require selective angiography. In the abdomen, visceral arteries and veins are usually well depicted by CT angiography, duplex sonography, and MR angiography; in arteries, however, selective arteriography is used to verify the diagnosis or to intervene. For peripheral arteries, DSA is a very reliable diagnostic tool, depicting location of lesions as well as their morphology, but it is being replaced by MR angiography. Duplex sonography as a sole diagnostic tool is also frequently used, but lacks demonstrability and is time consuming. For peripheral arterial interventions, angiography is the working modality. In peripheral veins, duplex sonography is the modality of choice and phlebography is preserved for nondiagnostic cases.

Vascular interventional radiology involves all vascular areas and vascular systems. The guiding tool is still contrast-enhanced X-ray fluoroscopy, and no other tool is yet available that gives a comparable timely resolution and precision. Treatment options include reopening of stenosed and occluded arteries and veins with different tools such as balloon angioplasty, stent implantation, and mechanical and chemical thrombolysis as well as temporary or permanent occlusion of vessels by embolization with particles, fluids, or coils or with chemoembolization. Exclusion of aneurysms is performed by stent graft implantation or coil embolization. Access for percutaneous arterial interventions is transfemoral, transbrachial, transpopliteal, or rarely by direct puncture. For venous interventions, a transfemoral, transbrachial, or transparenchymal access may be used. In hemodialysis problems, direct puncture is preferred. Interventional vascular procedures in arteries have been accepted by the medical community as a valuable treatment option in many regions and as the treatment of choice particularly in the iliac arterial region. Embolization procedures are frequently used as life-saving interventions in cases of acute bleeding.

Vascular Origin Malignant Tumors Group of tumors with lower to higher grade malignancy including hemangioendothelioma, hemangiopericytoma, angiosarcoma, Kaposi sarcoma. ▶Neoplasms, Soft Tissues, Malignant

Vasculitis Vasculitis is an inflammation of blood vessels caused by autoimmune disorders, infections, exposure to radiation or toxins and other conditions. ▶Oral Cavity, Inflammatory Diseases

VB Vacuum biopsy or vacuum assisted biopsy is a form of biopsy where several large specimens (usually 11 or 8G) can be taken from the breast under stereotactic or sonographic guidance. ▶Carcinoma, Ductal, In Situ, Breast

Verbiest Syndrome of Lumbar Spine

VCUG ▶Voiding Cystourethrography

VE ▶Virtual Endoscopy

1951

Venous Stasis Venous dysfunction causes swelling of the lower extremities with dermatitis and ulceration. In advanced stages of the disease, periostitis may occur at tibia, fibula, femur, metatarsal bones, and phalanges. The periosteal appositions have an undulated osseous contour and cortical thickening appears as the appositions are not well separated from the original cortex. ▶Hypertrophic, Osteoarthropathy

Venous Stenosis Vector A vector is a vehicle or carrier used to deliver genetic material into the body for gene therapy. ▶Local Drug and Gene Delivery with Microbubbles

VEGF Vascular epithelial growth factor. ▶Carcinoma, Ductal, In Situ, Breast

Focal narrowing of a segment of the efferent vein or of the arteriovenous anastomosis mostly owing to intimal hyperplasia; other causes of stenoses are focal fibrosis of the venous wall secondary to chronically high pressure in the venous circulation or to repeated traumatic needle punctures during hemodialysis. Finally, a central venous stenosis is nearly always the result of neointimal thickening and formation of a fibrous sheath as a result of a previously placed indwelling dialysis catheter or central line. If this stenosis becomes significant or eventually critical, the dialysis session will become insufficient and intervention is mandatory to rescue the fistula. ▶Fistula, Hemodialysis

Venous Thromboembolic Disease Veno-occlusive Disease ▶Chest, Thromboembolic Diseases Hepatic veno-occlusive disease (VOD) refers to the obstruction of small and medium-size intrahepatic veins. It is a common complication following bone marrow transplantation and it is usually associated with the use of chemotherapeutic agents or radiation. Diagnosis can usually be made based on clinical criteria. Commonly used clinical criteria for the diagnosis are direct bilirubin ≥2 mg/dL, hepatomegaly, and >2% weight gain due to fluid accumulation; the diagnosis requires 2 of 3 criteria, occurring within 20 days of transplantation. US can exclude tumor infiltration and biliary tract disease and can confirm hepatomegaly and ascites. Transvenous liver biopsy is required in patients with thrombocytopenia posttransplantation to differentiate VOD from acute graft-versus-host disease. ▶Vascular Disorders, Hepatic

Ventral Induction Anomalies ▶Congenital Malformations, Cerebrum

Verbiest Syndrome of Lumbar Spine ▶Stenosis, Spinal

V

1952

Vertebral Body Degeneration

Vertebral Body Degeneration According to Modic there are three types of vertebral body degeneration changes, which can be well assessed on MR images. Type 1 lesions (aseptic spondylodiscitis) exhibit decreased signal on T1-weighted images due to increase in bone marrow water content, and increased water signal on T2-weighted images. Type 2 changes (fatty degeneration) have increased fat signal on T1-weighted images and intermediate signal on T2-weighted images (high fat signal with turbo spin-echo T2 weighted images). Type 3 (osteosclerosis) have low signal on both T1 and T2-weighted images as well as high density on plain films and CT. All three types of lesions are located adjacent to the endplates of the vertebral bodies. Other common manifestations of vertebral body degeneration are osteophytes, which can protrude into the spinal canal and intervertebral foramina and are easily detected with plain films, CT or MRI. ▶Degenerative Conditions, Spine

Vertebral Steal Latent or Temporary Reversed flow in the vertebral artery, which is induced after exercise or hyperemia of the ipsilateral arm. ▶Steal Syndrome Vertebral

▶Contrast Media, Ultrasound, Applications in Vesicoureteral Reflux ▶Reflux, Vesicoureteral, Adults ▶Reflux, Vesicoureteral, Childhood ▶Urethra, Stenosis

Vesicoureteric or vesicoureteral reflux ▶Reflux, Vesicoureteral, Childhood

Vibration-Induced White Finger Symptoms similar to that of Raynaud’s syndrome (blanching of the fingers, coolness, paresthesias) but occurring in patients who use vibratory tools. ▶Ischemia, Brachial

Virtual Cystoscopy Technique of virtual endoscopy applied to urinary bladder. ▶Neoplasms, Bladder

Virtual Endoscopy Vertebral Steal Permanent Flow reversal in the vertebral artery that does not depend on the patient’s state of activity or other parameters. ▶Steal Syndrome, Vertebral

Vesicoureteral Reflux Pathological condition in which, during voiding, some urine goes back up into the ureters and possibly up to the kidneys. Reflux exposes the kidneys to infection. The injury to the kidney may result in renal scarring. Untreated reflux on both sides can, in the most severe instances, result in kidney failure requiring dialysis or kidney transplantation.

New method of diagnosis using computer processing of 3D image datasets (such as CT or MRI scans) to provide simulated visualization of patient specific organs similar or equivalent to those produced by standard endoscopic procedures. ▶Neoplasms, Bladder

Vitamin Deficiency D OROTHY B ULAS Division of Diagnostic Imaging and Radiology, Children’s National Medical Center, The George Washington University School of Medicine and Health Sciences Washington DC, USA [email protected]

Vitamin Deficiency

Synonyms Rickets—vitamin D Deficiency; Scurvy—vitamin C Deficiency

Definitions There are several vitamin deficiencies that can result in skeletal changes. The most common deficiencies that will be described include deficiencies in vitamin D and C. Vitamin D deficiency results in under mineralization of the boney matrix-osteomalacia. Rickets, “to bend,” is the result of decreased mineralization of the growing physis resulting in delayed skeletal development and growth. Rickets is found in infants and children prior to growth plate fusion, while osteomalacia can occur at any age. Vitamin C (ascorbic acid) deficiency results in scurvy. In the 1700s, a British naval surgeon established that oranges and lemons would cure scurvy. By the 1900s, infant scurvy was eradicated when heated formulas (boiling destroys vitamin C) was supplemented with fresh fruits or vegetables.

Pathology/Histopathology Rickets is caused by a deficiency or abnormal metabolism of vitamin D or abnormal excretion of phosphate leading to bony deformities and hypocalcemia (Table 1). The physis is a region of chondrocyte hypertrophy, proliferation and vascular invasion which then converts into primary bone spongiosa. Vascular invasion requires mineralization of the physeal cartilage. If calcium or phosphorus is deficient, the growth plate thickens, the chondrocytes become disorganized, and osteoid accumulates. The metaphysis widens, compensating for decreased bone strength. Vitamin D requires activation to regulate absorption and renal retention of calcium and phosphorus, and modulates osteoblastic function. In humans, vitamin D Vitamin Deficiency. Table 1

Causes of rickets

Deficient intake: Dietary, lack of sunlight Poor absorption: Malabsorption, antacids, anticonvulsants, renal insufficiency, hepatic insufficiency, cystic fibrosis Increased excretion: Renal tubular acidosis, renal tubular damage Medication: Anticonvulsants, fluoride, diphosphonate Chronic renal failure Hereditary: X-linked hypophosphatemic rickets, type I and II pseudovitamin D deficiency

1953

can be exogenous or endogenous. Vitamin D is synthesized endogenously in the skin following sunlight exposure. Dietary vitamin D from plants is known as ergocalciferol—vitamin D2 and that from animal is known as cholecalciferol—vitamin D3. Both vitamins D2 and D3 must be further metabolized. Activation occurs by hydroxylation at two sites of the vitamin D molecule. Hydroxylation at the 25th site occurs in the liver (25-OHD). Hydroxylation at the first site takes place in the kidneys and is regulated by 25-hydroxy D-1 alpha hydroxylase. 1,25-Dihydroxycholecalciferol is the active form of vitamin D (1,25-(OH)2D). This process is regulated by parathyroid hormone (PTH) which inhibits 25-hydroxy D-1 alpha hydroxylase. If PTH is suppressed, 25-hydroxy vitamin D can be converted to a less potent 24,25-dihydroxy vitamin D. Very low birth weight infants can develop metabolic bone disease of prematurity. The need for calcium and phosphorus is at its peak in the third trimester. If the dietary calcium is low, renal alpha1 hydroxylase is activated and 1,25 hydroxy vitamin D is made, increasing gastrointestinal uptake of calcium and inhibiting PTH release. There is a net decrease in bone mass unless adequate amounts of calcium and phosphorus are supplemented. This is particularly true if breast milk is the primary source of nutrition as it contains less than half of the needed calcium and phosphorus for growing premature infants. Scurvy is caused by a deficiency of vitamin C. Vitamin C is required for the formation of hydroxyproline, crucial for collagen formation which makes up 90% of mature boney matrix. Lack of collagen severely affects bone formation in childhood, resulting in scurvy and osteoporosis in adults. Defective collagen synthesis also leads to poor dentine formation, hemorrhage of gums, and diffuse bleeding. The pathophysiological changes in scurvy are the results of depression of normal cellular activity. Osteoblastic activity is suppressed with failure to form osteoid, yet resorption continues with resultant osteoporosis. At the physis, cartilage proliferation is decreased yet mineralization is unimpaired, thus the zone of provisional calcification appears wide and dense. Changes around the epiphyseal ossification center result in a thin ring of increased density. Vascular invasion in the zone of provisional calcification with suppressed osteoblastic activity results in decreased density in the zone of primary and secondary spongiosa. The zone of provisional calcification extends beyond the margins of the metaphysis resulting in periosteal elevation and marginal spur formation. Increased capillary fragility results in subperiosteal hemorrhage elevating the periosteum. These changes are most marked in most active endochondral bone growth. Intraarticular hemorrhage is rare as periosteal attachment to the growth plate is strong.

V

1954

Vitamin Deficiency

Clinical Presentation Rickets: With the introduction of dietary supplements, the incidence of rickets has decreased significantly. Cases still occur, particularly in breastfed infants with limited sunlight exposure who are not supplemented with vitamins. The recommended vitamin D intake is 200 IU for infants, yet breast milk contains less than 60 IU/L. Mothers with vitamin D deficiency can result in decreased transfer to the fetus. Muslim women wearing veils are at high risk of vitamin D deficiency secondary to underexposure to sunlight. Blacks are more affected than whites possibly due to decreased penetration of ultraviolet light. Premature infants are at highest risk with an incidence of 5–20% particularly when triphosphopyridine nucleotide total parental nutrition (TPN)-dependent. Rickets occurs in children whose growth plates have not yet closed. Children present with genu varum or anterior tibial bowing. Late teeth eruption, scoliosis, thickened wrist, and ankles may be noted. Ligaments are lax and muscle tone decreased. Rarely, hypocalcaemic seizures may be a presenting symptom. Vitamin D resistant rickets is found in children older than 30 months of age. These children are short and bow-egged. Ectopic calcification can be featured. Scurvy is quite rare now due to vitamin C supplementation. Patients who are elderly or alcoholic with diets lacking in fresh fruits and vegetables are most vulnerable. It rarely occurs in children less than 6 months of age. Initial symptoms include poor weight gain, anorexia, diarrhea, and tachypnea. More specific symptoms include bleeding gums, pseudoparalysis, with subperiosteal bleeding causing severe pain. Petechial hemorrhage, hematuria, and corkscrew hair can develop. Sudden death due to cardiac failure has been reported. Prognosis is excellent if diagnosed and treated appropriately.

The epiphysis may develop hazy cortical margins. The enlarged ends of the ribs can be palpable at the costochondral junction clinically, called “the rachitic rosary.” The sternum becomes prominent and a pectus carinatum may develop (Fig. 2). The skull becomes demineralized. Basilar invagination, indistinct sutures, delayed tooth eruption, and premature craniostenosis can occur. Green stick fracture of the cortex may develop with

Vitamin Deficiency. Figure 1 Rickets. AP radiograph of the wrist of a 9-month-old demonstrates cupping and fraying of the metaphysis of the distal radius and ulna. There is periosteal elevation along the metacarpals with an indistinct cortex.

Imaging Radiographs are most useful in diagnosing rickets and scurvy. CT/MRI/nuclear studies have limited roles. Dual X-ray absorptiometry (DXA) may document severity of osteomalacia. In rickets, skeletal changes are due to lack of calcification of osteoid. The first radiographic changes appear in rapidly growing distal ends of the radius and ulna. There is a loss of ossification at the zone of provisional calcification. The metaphyseal margin becomes indistinct then frayed with widening of the growth plate (Fig. 1). Weight bearing and stress on uncalcified bone result in splaying and cupping of the metaphysis. The subperiosteal layer becomes indistinct. The bone density is reduced with course trabeculation.

Vitamin Deficiency. Figure 2 Rickets. AP view of the chest demonstrates indistinct humeral metaphyseal margin with loss of ossification at the zone of provisional calcification. The costochondral junctions are widened (“rachitic rosary”).

Vitamin Deficiency

Vitamin Deficiency. Table 2

1955

Key findings in rickets

Widening and cupping of the metaphysis Fraying of the metaphysis Craniotabes Bowing of the long bones Scoliosis Triradiate pelvis—impression of femur into pelvis— protrusion acetabuli

Vitamin Deficiency. Table 3

Key findings in scurvy

Subperiosteal hemorrhages Osteopenia—Ground-glass cortex Corner sign Wimberger’s sign Radiolucent metaphyseal band

bowing of the long bones. Vertebra becomes biconcave with resultant scoliosis (Table 2). Following treatment, there is ossification of the provisional zone of calcification. The radiolucent metaphyseal bands become more prominent as the zone of provisional calcification becomes denser. Distortion of the spongiosa may occur and remain visible for years. Calvarial thickening develops with frontal and parietal bossing and premature suture closure. During healing, there is extensive periosteal new bone formation along the cornices of the diaphyses. Angulation deformities secondary to pathological fractures can result in genu valga or varum. Cortical thickening may persist though remodeling of bowing deformities eventually occurs. In scurvy, radiologic changes occur at the physis of bones with rapid growth. The knee, wrist, proximal humerus, and sternal ends of the ribs are typical sites of involvement (Table 3). In the early phase, the cortex becomes thin and the trabecular structure of the medulla atrophies and develops a ground-glass appearance. The zone of provisional calcification becomes dense and wide referred to as the ▶white line of Frankle. The epiphysis demonstrates a ground glass appearance with a sharp sclerotic rim called the ▶Wimberger’s ring. As scurvy becomes advanced, a zone of rarefaction occurs at the metaphysis below the white line of Frankle. This transverse band of radiolucency beneath the zone of calcification is called the Trummerfeld zone of lucency (Fig. 3). The zone involves the lateral aspects of the white line, resulting in triangular defects called the corner sign of park. This region has multiple microscopic fractures and can collapse with impaction of the calcified cartilage onto the shaft. The lateral aspect of the calcified cartilage can project resulting in a spur called Pelkan’s spur. Subperiosteal hemorrhages are frequent and are most commonly noted

Vitamin Deficiency. Figure 3 Scurvy. AP view of the knee demonstrates a thin cortex with ground-glass trabecula. The zone of provisional calcification is sclerotic. The epiphysis has a sharp sclerotic rim called the Wimberger’s ring. Extensive subperiosteal hemorrhages have calcified, resulting in significantly elevated periosteum.

in larger tubular bones of the femur, tibia, and humerus. Rarely, subperiosteal hematomas develop on the flat bones of the orbit causing proptosis. Subperiosteal hemorrhages do not extend to the epiphysis. Following treatment, the cortex becomes thicker. The metaphyseal transverse band of decreased density normalizes. The thick provisional zone of calcification becomes a thin transverse line. The raised periosteum layers the periphery of the hematoma with subperiosteal bone. Cortical thickening, central and epiphyseal rarefaction may persist for years. Multiple vitamin deficiencies can modify the radiologic appearance of growing bone (Fig. 4). When scurvy and rickets are both present, scurvy findings are predominant because of diminished osteoblastic activity. Scurvy could potentially be mistaken for nonaccidental trauma when corner fractures and subperiosteal elevation is noted. Response to treatment with vitamin C can establish the diagnosis.

Diagnosis Rickets: The diagnosis of vitamin D deficiency is made on clinical, radiographic, and laboratory values. In the early

V

1956

Voiding Cystourethrography

Vitamin Deficiency. Figure 4 Metabolic bone disease of prematurity. AP view of the knee (a) and the pelvis (b) of a 3-month-old very low birth weight infant on total.

stages, histological changes are developing and blood serum levels become abnormal several weeks prior to radiographic changes. Radiologic findings are first noted in the most rapidly growing physis. The rarefaction and irregular fraying of the zone of provisional calcification are virtually diagnostic but the underlying cause needs to be evaluated via biochemical and clinical assessment. Dietary history, medication history, and measurement of creatinine and liver enzymes may help pinpoint the cause. Alkaline phosphatase is usually increased and is an excellent marker of disease activity. Serum 25-OH vitamin D is low. Scurvy: Radiologic findings of scurvy are diagnostic with laboratory tests not significantly helpful. Low alkaline phosphatase levels are noted due to decreased osteoid production. Anemia may be present. Serum ascorbic acid levels

Encyclopedia of Diagnostic Imaging

Encyclopedia of Diagnostic Imaging

Encyclopedia of diagnostic imaging

Diagnostic Imaging

Primer of Diagnostic Imaging

Diagnostic Imaging: Emergency

Diagnostic Imaging: Pediatric Neuroradiology

Diagnostic Imaging: Spine

Diagnostic Imaging: Orthopaedics

Diagnostic Imaging: Brain

Diagnostic Imaging: Ultrasound

Exercises In Diagnostic Imaging

Diagnostic Imaging: Pediatrics

Anatomy for Diagnostic Imaging

Diagnostic Imaging: Orthopaedics

Diagnostic imaging: Chest

Anatomy in Diagnostic Imaging

Diagnostic Imaging: Obstetrics

Diagnostic Imaging: Gynecology

Diagnostic Imaging: Chest

Diagnostic Imaging: Ultrasound

Diagnostic Imaging: Orthopaedics

Diagnostic Imaging: Brain

Diagnostic Nuclear Medicine (Medical Radiology Diagnostic Imaging)

Veterinary Diagnostic Imaging - The Horse

Diagnostic Imaging: Head and Neck

Spinal Imaging Diagnostic Imaging of the Spine and Spinal Cord

Primer of Diagnostic Imaging, 4th Edition

Primer of Diagnostic Imaging, 5th Edition

Primer of Diagnostic Imaging, 4th Edition

Encyclopedia of Diagnostic Imaging