Mauro - Image-Guided Interventions 2e [PDF]

Zitiervorschau

IMAGE-GUIDED INTERVENTIONS SECOND EDITION

MATTHEW A. MAURO, MD, FACR The Ernest H. Wood Distinguished Professor of Radiology and Surgery Chairman, Department of Radiology University of North Carolina School of Medicine Chapel Hill, North Carolina KIERAN P.J. MURPHY, MB, FRCPC, FSIR Professor of Radiology Joint Department of Medical Imaging University Health Network Mount Sinai Hospital Women’s College Hospital Toronto, Ontario, Canada

KENNETH R. THOMSON, MD, FRANZCR

Professor of Radiology Monash University Faculty of Medicine, Nursing, and Health Sciences Director, Radiology Department Alfred Hospital Melbourne, Victoria, Australia

ANTHONY C. VENBRUX, MD Director, Cardiovascular and Interventional Radiology George Washington University School of Medicine and Health Sciences Washington, DC ROBERT A. MORGAN, MBChB, MRCP, FRCR, EBIR Consultant, Vascular and Interventional Radiologist Radiology Department St. George’s NHS Trust London, United Kingdom

1600 John F. Kennedy Blvd. Ste 1800 Philadelphia, PA 19103-2899

IMAGE-GUIDED INTERVENTIONS (A Volume in the Expert Radiology Series)

ISBN: 978-1-4557-0596-2

Copyright © 2014, 2008 by Saunders, an imprint of Elsevier Inc. Chapter 111: “Arteriography and Arterial Stimulation with Venous Sampling for Localizing Pancreatic Endocrine Tumors” by Anthony W. Kam, Bradford J. Wood, and Richard Chang is in the Public Domain. Chapter 151: “Thermal Ablation of the Adrenal Gland” by Deepak Sudheendra and Bradford J. Wood is in the Public Domain. No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system, without permission in writing from the publisher. Details on how to seek permission, further information about the Publisher’s permissions policies and our arrangements with organizations such as the Copyright Clearance Center and the Copyright Licensing Agency, can be found at our website: www.elsevier.com/ permissions. This book and the individual contributions contained in it are protected under copyright by the Publisher (other than as may be noted herein).

Notices Knowledge and best practice in this field are constantly changing. As new research and experience broaden our understanding, changes in research methods, professional practices, or medical treatment may become necessary. Practitioners and researchers must always rely on their own experience and knowledge in evaluating and using any information, methods, compounds, or experiments described herein. In using such information or methods they should be mindful of their own safety and the safety of others, including parties for whom they have a professional responsibility. With respect to any drug or pharmaceutical products identified, readers are advised to check the most current information provided (i) on procedures featured or (ii) by the manufacturer of each product to be administered, to verify the recommended dose or formula, the method and duration of administration, and contraindications. It is the responsibility of practitioners, relying on their own experience and knowledge of their patients, to make diagnoses, to determine dosages and the best treatment for each individual patient, and to take all appropriate safety precautions. To the fullest extent of the law, neither the Publisher nor the authors, contributors, or editors assume any liability for any injury and/or damage to persons or property as a matter of products liability, negligence or otherwise, or from any use or operation of any methods, products, instructions, or ideas contained in the material herein. Library of Congress Cataloging-in-Publication Data Image-guided interventions / Matthew A. Mauro … [et al.].—2nd ed. p. ; cm.—(Expert radiology series) Includes bibliographical references and index. ISBN 978-1-4557-0596-2 (hardcover : alk. paper) I. Mauro, Matthew A., 1951- II. Series: Expert radiology series. [DNLM: 1. Radiography, Interventional–methods. 2. Diagnostic Techniques, Surgical. 3. Vascular Diseases–radiography. 4. Vascular Diseases–therapy. 5. Vascular Surgical Procedures– methods. WN 200] 616.1′307572–dc23 2012046151 Content Strategist: Helene Caprari Senior Content Development Specialist: Jennifer Shreiner Publishing Services Manager: Anne Altepeter Project Manager: Cindy Thoms Design Direction: Steven Stave

Working together to grow libraries in developing countries www.elsevier.com | www.bookaid.org | www.sabre.org

Printed in China Last digit is the print number: 9

8

7

6

5

4

3

2 1

To Pat, my wife, for her unwavering love and support and to our children, Lauren and David, my most precious accomplishments Matt Mauro To Rulan, my wife, Ronan and Anya, the three most interesting people I know in the world, thank you for allowing me the time and space to work on this book and all my other crazy ideas. To everyone who has taken the time to teach me, thank you for your patience Kieran Murphy To my wife, Barbara Ken Thomson To my family; to my patients, who have taught me so much; to my mentors; and to Michael Heinl Tony Venbrux To my family and friends for their forbearance, support, and good cheer over the years Robert Morgan

Contributors Hani H. Abujudeh, MD, MBA Associate Professor of Radiology Massachusetts General Hospital Harvard Medical School Boston, Massachusetts

Morvarid Alaghmand, MD Internist Riverside Primary Care Associates District Heights, Maryland

Andreas Adam, PhD, FRCP, FRCS, FRCR Professor Department of Interventional Radiology King’s College Consultant Radiologist St. Thomas’ Hospital London, United Kingdom

Ali Albayati, MBChB Fellow Division of Cardiovascular and Interventional Radiology George Washington University Medical Center Washington, DC

Acute Lower Extremity Ischemia

Esophageal Intervention in Malignant and Benign Esophageal Disease

Allison S. Aguado, MD Fellow Department of Radiology Memorial Sloan-Kettering Cancer Center New York, New York

Embolotherapy for the Management of Liver Malignancies Other Than Hepatocellular Carcinoma

Muneeb Ahmed, MD Assistant Professor of Radiology Section of Interventional Radiology Harvard Medical School Beth Israel Deaconess Medical Center Boston, Massachusets

Image-Guided Thermal Tumor Ablation: Basic Science and Combination Therapies

Kamran Ahrar, MD Professor Departments of Diagnostic Radiology and Thoracic and Cardivascular Surgery University of Texas MD Anderson Cancer Center Houston, Texas Thermal Ablation of Renal Cell Carcinoma

Andrew Akman, MD Assistant Professor of Radiology Interventional Radiology George Washington University School of Medicine and Health Sciences Washington, DC

Alimentary Tract Vasculature Renal Vasculature

Principles of Intraprocedural Analgesics and Sedatives Management of Male Varicocele

Agaicha Alfidja, MD Faculty of Medicine University Hospital Clermont-Ferrand, France Acute Mesenteric Ischemia

Ziyad Al-Otaibi, MD Department of Anesthesiology and Critical Care Medicine George Washington University Medical Canter Washington, DC Treatment of Medical Emergencies

John F. Angle, MD Professor Department of Radiology and Medical Imaging Director Division of Angiography and Interventional Radiology University of Virgina Health System Charlottesville, Virginia Balloon Catheters Closure Devices Tracheobronchial Interventions

Gary M. Ansel, MD, FACC Director, Center for Critical Limb Care Riverside Methodist Hospital Columbus, Ohio Assistant Clinical Professor of Medicine Medical University of Ohio Toledo, Ohio

Chronic Upper Extremity Ischemia and Revascularization

Management of Male Varicocele

vii

viii

CONTRIBUTORS Julien Auriol, MD Praticien Hospitalier Service de Radiologie L’Hôpital Rangueil Universitaire de Toulouse Toulouse, France

Carlo Bartolozzi, MD Professor Diagnostic and Interventional Radiology University of Pisa Pisa, Italy

Energy-Based Ablation of Hepatocellular Cancer

Vascular Intervention in the Liver Transplant Patient

Chad Baarson, DO Department of Radiology National Capital Consortium Walter Reed National Military Medical Canter Bethesda, Maryland Bipedal Lymphangiography

Juan Carlos Baez, MD Resident Department of Radiology Brigham and Woman’s Hospital Boston, Massachusetts Subarachnoid Hemorrhage

Image-Guided Intervention for Symptomatic Tarlov Cysts Scalene Blocks and Their Role in Thoracic Outlet Syndrome

Jason R. Bauer, MD, RVT Director of Interventional Oncology Interventional and Vascular Consultants, PC Portland, Oregon Embolization Agents

Richard A. Baum, MD, MPA Director, Interventional Radiology The Herbert L. Abrams Director of Angiography and Interventional Radiology Brigham and Women’s Hospital Associate Professor Department of Radiology Harvard Medical School Boston, Massachusetts

Thoracic Duct Embolization for Postoperative Chylothorax

Curtis W. Bakal, MD, MPH Chair, Radiology Lahey Clinic Burlington, Massachusetts Professor of Radiology Tufts University School of Medicine Boston, Massachusetts

Kevin W. Bell, MBBS, FRANZCR Clinical Associate Professor of Radiology University of Melbourne Faculty of Medicine Director Department of Radiology Western Health Melbourne, Victoria, Australia

Jörn O. Balzer, MD, PhD Director Department of Radiology and Nuclear Medicine Catholic Clinic Mainz Mainz, Germany

Jacqueline A. Bello, MD Professor Departments of Radiology and Neurosurgery Albert Einstein College of Medicine Director of Neuroradiology Department of Radiology Montefiore Medical Center Bronx, New York

Diagnostic Catheters and Guidewires

Endovascular Laser Therapy

Alex M. Barnacle, MRCP, FRCR Consultant Interventional Radiologist Department of Radiology Great Ormond Street Hospital for Children London, United Kingdom

Management of High-Flow Vascular Anomalies

Bradley P. Barnett, MD, PhD Research Fellow Department of Radiology Johns Hopkins Medical Institutions Baltimore, Maryland Craniocervical Vascular Anatomy

Gamal Baroud, PhD Professor Department of Mechanical Engineering Director, Biomechanics Laboratory Canada Research Chair in Skeletal Reconstruction University of Sherbrooke Sherbrooke, Quebec, Canada New Directions in Bone Materials

Vascular Anatomy of the Thorax, Including the Heart

Management of Head and Neck Tumors

Jennifer Berkeley, MD, PhD Neurointensivist Department of Neurology Sinai Hospital of Baltimore Baltimore, Maryland Cervical Artery Dissection

Michael A. Bettmann, MD, FACR Professor of Radiology Emeritus Wake Forest University School of Medicine Winston-Salem, North Carolina Co-Chair Task Force on Imaging Decision Support American College of Radiology Reston, Virginia Contrast Agents

CONTRIBUTORS José I. Bilbao, MD, PhD Professor of Radiology Universidad de Navarra Consultant Radiologist Clínica Universitaria de Navarra Pamplona, Spain

Portal-Mesenteric Venous Thrombosis

Deniz Bilecen, MD, PhD Chief Department of Radiology Kantonsspital Laufen Laufen, Switzerland Transjugular Liver Biopsy

Tiago Bilhim, MD Interventional Radiologist Department of Radiology St. Louis Hospital Faculty of Medical Sciences New University of Lisbon Lisbon, Portugal

Treatment of High-Flow Priapism and Erectile Dysfunction

Christoph A. Binkert, MD, MBA Chairman of Radiology Institute of Radiology Kantonsspital Winterthur Winterthur, Switzerland Inferior Vena Cava Filters Caval Filtration

Haraldur Bjarnason, MD Professor of Radiology Division of Vascular and Interventional Radiology Mayo Clinic Rochester, Minnesota Acute Lower Extremity Deep Venous Thrombosis

James H. Black, III, MD, FACS Bertram M. Bernheim, MD Associate Professor of Surgery Johns Hopkins University School of Medicine Baltimore, Maryland Clinical Vascular Examination

Brian M. Block, MD, PhD President Baltimore Spine Center Towson, Maryland Instructor Department of Anesthesiology and Critical Care Medicine Johns Hopkins University School of Medicine Baltimore, Maryland Stellate Ganglion Block Facet Joint Injection Epidural Steroid Injection Scalene Blocks and Their Role in Thoracic Outlet Syndrome

Marc Bohner, PhD Material Science Engineer Dr. Robert Mathys Foundation Bettlach, Switzerland

New Directions in Bone Materials

Amman Bolia, MBChB, DMRD, FRCR Consultant Vascular Radiologist Departments of Imaging and Interventions University Hospital of Leicester NHS Trust Leicester, United Kingdom Subintimal Angioplasty

Irene Boos, VMD Clinical Research Interventional Radiology Woerth, Germany

Intraarterial Ports for Chemotherapy

Charles F. Botti, Jr., MD, FACC Cardiac and Peripheral Vascular Interventionist MidOhio Cardiology and Vascular Consultants Riverside Methodist Hospital Columbus, Ohio

Chronic Upper Extremity Ischemia and Revascularization

Nina M. Bowens, MD General Surgery Resident Department of Surgery Hospital of the University of Pennsylvania Philadelphia, Pennsylvania

Thoracic Aortic Stent-Grafting and Management of Traumatic Thoracic Aortic Lesions

Louis Boyer, MD, PhD Professor of Radiology Faculty of Medicine ISIT, UMR CNRS 6284 University D’Auvergne Head, Department of Radiology University Hospital Clermont-Ferrand, France Acute Mesenteric Ischemia

Elena Bozzi, MD Department of Diagnostic and Interventional Radiology University of Pisa Pisa, Italy Energy-Based Ablation of Hepatocellular Cancer

Peter R. Bream, Jr., MD Associate Professor Departments of Radiology and Radiological Sciences and Medicine Vanderbilt University School of Medicine Director, PICC Service Program Director, Interventional and Vascular Radiology Vanderbilt University Medical Center Nashville, Tennessee Tunneled Central Venous Catheters

ix

x

CONTRIBUTORS Rachel F. Brem, MD Professor Department of Radiology George Washington University School of Medicine and Health Sciences Director Breast Imaging and Interventional Center Vice Chair of Radiology GW Medical Faculty Associates Washington, DC

Minimally Invasive Image-Guided Breast Biopsy and Ablation

Mark F. Brodie, MD Interventional Radiologist Department of Radiology Naval Medical Center San Diego, California

Fallopian Tube Interventions

Allan L. Brook, MD Associate Professor of Clinical Radiology and Neurosurgery Departments of Radiology and Surgery Albert Einstein College of Medicine Bronx, New York Director Interventional Neuroradiology Department of Radiology Montefiore Medical Center Brooklyn, New York

Jozef M. Brozyna, BS West Virginia School of Osteopathic Medicine Lewisburg, West Virginia A Brief History of Image-Guided Therapy

Charles T. Burke, MD Associate Professor Department of Radiology University of North Carolina School of Medicine Chapel Hill, North Carolina Management of Failing Hemodialysis Access Hemodialysis Access: Catheters and Ports

James P. Burnes, MBBS, FRACR Director Body Intervention Unit Department of Diagnostic Imaging Monash Medical Centre Moorabbin Southern Health Clayton, Victoria, Australia

Endovascular Treatment of Peripheral Aneurysms

Patricia E. Burrows, MD Professor of Radiology Medical College of Wisconsin Vascular Interventional Radiology Children’s Hospital of Wisconsin Milwaukee, Wisconsin

Management of Low-Flow Vascular Malformations

Management of Head and Neck Tumors

Benjamin S. Brooke, MD, PhD Fellow Vascular Surgery Department of Surgery Dartmouth-Hitchcock Medical Center Lebanon, New Hampshire

Justin J. Campbell, MD Staff Radiologist Department of Radiology South Shore Hospital Weymouth, Massachusetts

Percutaneous Abscess Drainage Within the Abdomen and Pelvis Renal and Perirenal Fluid Collection Drainage

Clinical Vascular Examination

Mark Duncan Brooks, MBBS, FRANZCR Consultant Radiologist Department of Radiology Austin Hospital Melbourne, Victoria, Australia

Transjugular Intrahepatic Portosystemic Shunts

Daniel B. Brown, MD Director Interventional Radiology Department of Radiology Thomas Jefferson University Philadelphia, Pennsylvania

Colin P. Cantwell, MBBCh, BAO, Msc, MRCS, FRCR, FFR(RCSI) Consultant Radiologist Department of Radiology St. Vincent’s University Hospital Dublin, Ireland

Peripherally Inserted Central Catheters and Nontunneled Central Venous Catheters

Thierry Carrères, MD Cardio-vasculaire Radiologie Hopital Européen Georges Pompidou Paris, France Infrapopliteal Revascularization

Management of Biliary Calculi

Karen T. Brown, MD, FSIR Interventional Radiologist Department of Medical Imaging Memorial Sloan-Kettering Cancer Center Professor of Clinical Radiology Department of Radiology Weill Cornell Medical College New York, New York

Bland Embolization for Hepatic Malignancies

John A. Carrino, MD, MPH Associate Professor of Radiology and Orthopedic Surgery Section Chief, Musculoskeletal Radiology The Russell H. Morgan Department of Radiology and Radiological Science Johns Hopkins University School of Medicine Baltimore, Maryland

Image-Guided Percutaneous Biopsy of Musculoskeletal Lesions

CONTRIBUTORS Lucie Cassagnes, MD Department of Radiology University Hospital Clemont-Ferrand, France

Acute Mesenteric Ischemia

Pascal Chabrot, MD, PhD Faculty of Medicine ISIT, UMR CNRS 6284 University D’Auvergne Department of Radiology University Hospital Clermont-Ferrand, France Acute Mesenteric Ischemia

Suma K. Chandra, MD George Washington University Medical Center Washington, DC A Brief History of Image-Guided Therapy

Richard Chang, MD Chief, Endocrine and Venous Services Section Senior Clinician Section of Interventional Radiology National Institutes of Health Bethesda, Maryland

Arteriography and Arterial Stimulation with Venous Sampling for Localizing Pancreatic Endocrine Tumors

Rishabh Chaudhari, BA Medical Student George Washington University School of Medicine and Health Sciences Washington, DC Renal Vasculature

Lakhmir S. Chawla, MD Associate Professor Department of Medicine George Washington University Medical Center Washington, DC

Rush H. Chewning, MD Chief Resident Department of Radiology University of Washington School of Medicine Seattle, Washington Endovascular Management of Epistaxis Subarachnoid Hemorrhage

Albert K. Chun, MD Assistant Professor of Radiology and Surgery Division of Angiography and Interventional Radiology George Washington University School of Medicine and Health Sciences Washington, DC Principles of Arterial Access

Joo-Young Chun, MMBS, BSc, MSc, MRCS, FBCR Interventional Radiology Fellow Department of Radiology St. George’s Hospital London, United Kingdom Aortic Endografting

Timothy W.I. Clark, MD, MSc, FRCP(C), FSIR Associate Professor of Clinical Radiology Department of Radiology University of Pennsylvania School of Medicine Philadelphia, Pennsylvania

Acute Upper Extremity Deep Venous Thrombosis Management of Fluid Collections in Acute Pancreatitis Chemical and Thermal Ablation of Desmoid Tumors

Wendy A. Cohen, MD Professor of Radiology University of Washington School of Medicine Chief of Service Department of Radiology Harborview Medical Center Seattle, Washington

Use of Skull Views in Visualization of Cerebral Vascular Anatomy

Treatment of Medical Emergencies

Hank K. Chen, MD Section of Interventional Radiology Department of Radiology George Washington University Medical Center Washington, DC A Brief History of Image-Guided Therapy

Yung-Hsin Chen, MD Assistant Clinical Professor of Radiology Tufts University School of Medicine Boston, Massachusetts

Image-Guided Percutaneous Biopsy of Musculoskeletal Lesions

Bairbre Connolly, MB, FRCP(C) Pediatric Interventional Radiologist Department of Diagnostic Imaging Hospital for Sick Children Associate Professor Medical Imaging Department Earl Glenwood Coulson Chair University of Toronto Faculty of Medicine Toronto, Ontario, Canada

Pediatric Gastrostomy and Gastrojejunostomy

Alison Corr, MBBCh, BSc Faculty of Radiologists Royal College of Surgeons Dublin, Ireland

Percutaneous Biopsy of the Lung, Mediastinum, and Pleura

xi

xii

CONTRIBUTORS Anne M. Covey, MD, FSIR Associate Professor of Radiology Department of Diagnostic Radiology Memorial Sloan-Kettering Cancer Center New York, New York

Robert S.M. Davies, MBChB, MRCS, MMed, FRCS (England) Senior Clinical Fellow British Society of Endovascular Therapy Leicester, United Kingdom

Laura Crocetti, MD, PhD Assistant Professor of Radiology Department of Oncology, Transplants, and Advanced Technologies in Medicine University of Pisa Faculty of Medicine Pisa, Italy

Thierry de Baère, MD Chief Department of Interventional Radiology Institut Gustave Roussy Villejuif, France

Charles D. Crum, MD Neuroradiology Fellow Department of Radiology St. Joseph’s Hospital and Medical Center Phoenix, Arizona

L. Mark Dean, MD Interventional Radiologist Riverside Methodist Hospital Columbus, Ohio

Management of Malignant Biliary Tract Obstruction

Energy-Based Ablation of Hepatocellular Cancer

Subintimal Angioplasty

Portal Vein Embolization Intraarterial Ports for Chemotherapy

Selective Nerve Root Block

Cryoablation of Liver Tumors

T. Andrew Currier, BSEE, LLB, P ENG Barrister and Solicitor Perry and Currier Patent and Trademark Agents Toronto, Ontario, Canada

Sudhen B. Desai, MD Interventional Radiologist Methodist Sugar Land Hospital Houston, Texas

Percutaneous Biopsy Treatment of Effusions and Abscesses

Intellectual Property Management

Ferenc Czeyda-Pommersheim, MD Resident in Radiology Mallinckrodt Institute of Radiology Washington University Saint Louis, Missouri Uterine Fibroid Embolization

Michael D. Dake, MD Professor of Cardiothoracic Surgery Stanford University School of Medicine Director Catheterization Angiography Laboratory Stanford University Medical Center Stanford, California

Balloon Catheters Endovascular Treatment of Dissection of the Aorta and Its Branches

Michael D. Darcy, MD Professor of Radiology Washington University St. Louis Chief of Intereventional Radiology Mallinckrodt Institute of Radiology Saint Louis, Missouri

Management of Lower Gastrointestinal Bleeding Ambulatory Phlebectomy

Sean R. Dariushnia, MD Assistant Professor of Radiology Division of Interventional Radiology and Image-Guided Medicine Emory School of Medicine Atlanta, Georgia

Management of Extremity Vascular Trauma

Massimiliano di Primio, MD Division of Interventional Radiology Hopital Européen Georges Pompidou Paris, France Infrapopliteal Revascularization

Robert G. Dixon, MD Associate Professor Department of Radiology University of North Carolina School of Medicine Chapel Hill, North Carolina Subcutaneous Ports

Pablo D. Domínguez, MD Department of Radiology Clínica Universitaria de Navarra Pamplona, Spain

Portal-Mesenteric Venous Thrombosis

Robert F. Dondelinger, MD, Hon FRCR Professor and Chair Department of Medical Imaging University of Liège Faculty of Medicine Liège, Belgium

Management of Trauma to the Liver and Spleen Splenic Embolization in Nontraumatized Patients Superior Vena Cava Occlusive Disease

CONTRIBUTORS Gregory J. Dubel, MD Assistant Professor Division of Vascular and Interventional Radiology Warren Alpert Medical School Brown University Providence, Rhode Island Stents

Damian E. Dupuy, MD, FACR Professor Department of Diagnostic Imaging Warren Alpert Medical School Brown University Director of Tumor Ablation Rhode Island Hospital Providence, Rhode Island Lung Ablation

Ghassan E. El-Haddad, MD Assistant Member Interventional Radiology H. Lee Moffitt Cancer Center and Research Institute Assistant Professor of Oncologic Sciences University of South Florida Tampa, Florida Management of Renal Angiomyolipoma

Joseph P. Erinjeri, MD, PhD Interventional Radiology Service Memorial Sloan-Kettering Cancer Center New York, New York

Chemical and Thermal Ablation of Desmoid Tumors

Clifford J. Eskey, MD Assistant Professor Department of Radiology Dartmouth-Hitchcock Medical Center Lebanon, New Hampshire Vertebroplasty and Kyphoplasty

Thomas M. Fahrbach, MD Radiology Fellow Division of Interventional Radiology and Image-Guided Medicine Emory School of Medicine Atlanta, Georgia Percutaneous Cholecystostomy

Ronald M. Fairman, MD Clyde F. Barker-William Maul Measey Professor Surgery University of Pennsylvania School of Medicine Chief of Vascular Surgery and Endovascular Therapy University of Pennsylvania Health System Philadelphia, Pennsylvania

Thoracic Aortic Stent-Grafting and Management of Traumatic Thoracic Aortic Lesions

Chieh-Min Fan, MD Associate Director Division of Angiography and Interventional Radiology Brigham and Women’s Hospital Assistant Professor Department of Radiology Harvard Medical School Boston, Massachusetts

Thoracic Duct Embolization for Postoperative Chylothorax

Mark A. Farber, MD Associate Professor Departments of Radiology and Surgery Director Heart and Vascular Center University of North Carolina School of Medicine Chapel Hill, North Carolina

Aortic Stent-Grafts Thoracic Aortic Stent-Grafting and Management of Traumatic Thoracic Aortic Lesions

Laura M. Fayad, MD Associate Professor of Radiology, Orthopedic Surgery, and Oncology The Russell H. Morgan Department of Radiology and Radiological Science Johns Hopkins University School of Medicine Baltimore, Maryland

Image-Guided Percutaneous Biopsy of Musculoskeletal Lesions

Dimitrios Filippiadis, MD, PhD Consultant Attikon University Hospital Athens, Greece

Ablation and Combination Treatments of Bony Lesions Sacroiliac Joint Injections

Kathleen R. Fink, MD Assistant Professor of Neuroradiology University of Washington School of Medicine Seattle, Washington Use of Skull Views in Visualization of Cerebral Vascular Anatomy

Sebastian Flacke, MD, PhD Chief Interventional Radiology Director of Non-Invasive Cardiovascular Imaging Vice Chair, Department of Radiology Lahey Clinic Burlington, Massachusetts Professor of Radiology Tufts University Medical School Boston, Massachusetts Diagnostic Catheters and Guidewires

Karen Flood, BMedSci, BMBS, MMedSciClinEd, MRCS, FRCR Interventional Radiology Registrar Department of Vascular Interventional Radiology Leeds Teaching Hospitals NHS Trust Leeds, United Kingdom Principles of Venous Access

xiii

xiv

CONTRIBUTORS Matthew D. Forrester, MD Resident Department of Cardiothoracic Surgery Stanford University School of Medicine Stanford, California

Endovascular Treatment of Dissection of the Aorta and Its Branches

Brian Funaki, MD Professor Department of Radiology University of Chicago Pritzker School of Medicine Section Chief Division of Vascular and Interventional Radiology University of Chicago Medical Center Chicago, Illinois Thombectomy Devices

Dimitri A. Gagarin, MD Associate Professor of Medicine Department of Internal Medicine Virginia Commonwealth University Richmond, Virginia

A Brief History of Image-Guided Therapy

Philippe Gailloud MD Director Division of Interventional Neuroradiology Johns Hopkins Hospital Baltimore, Maryland

Craniocervical Vascular Anatomy Arterial Anatomy of the Spine and Spinal Cord Endovascular Management of Epistaxis

Bhaskar Ganai, BSc(MedSci) Hons, MBChB, MRCS, FRCR SpR in Radiology Department of Interventional Radiology Freeman Hospital Newcastle upon Tyne, United Kingdom Embolic Protection Devices

Debra A. Gervais, MD Associate Professor Department of Radiology Harvard Medical School Division Head Abdominal Imaging and Intervention Massachusetts General Hospital Boston, Massachusetts

Percutaneous Abscess Drainage Within the Abdomen and Pelvis Renal and Perirenal Fluid Collection Drainage

Jean-François H. Geschwind, MD Professor Departments of Radiology, Surgery, and Oncology Johns Hopkins University School of Medicine Director Interventional Radiology Center Director Vascular and Interventional Radiology Johns Hopkins Medical Institutions Baltimore, Maryland Radioembolization for Hepatocellular Carcinoma

Basavaraj V. Ghodke, MD Associate Professor Departments of Neuroradiology and Neurological Surgery University of Washington School of Medicine Director Neuro-Interventional Radiology Childrens Hospital and Research Center Seattle, Washington Use of Skull Views in Visualization of Cerebral Vascular Anatomy

Brian B. Ghoshhajra, MD, MBA Director Clinical Cardiac Imaging Department of Radiology Massachusetts General Hospital Boston, Massachusetts Noninvasive Vascular Diagnosis

Mark F. Given, MD, BCh, BAO, AFRCSI, FFRCR Consultant Radiologist Beaumont Hospital Dublin, Ireland

Anatomy of the Lower Limb Acute Arterial Occlusive Disease of the Upper Extremity Endovascular Treatment of Peripheral Aneurysms Adrenal Venous Sampling Parathyroid Venous Sampling Percutaneous Biopsy of the Lung, Mediastinum, and Pleura

Y. Pierre Gobin, MD, MS Professor of Radiology Neurosurgery Department Weill Cornell Medical College New York, New York

Acute Stroke Management Endovascular Management of Chronic Cerebral Ischemia

S. Nahum Goldberg, MD Professor Department of Radiology Harvard Medical School Beth Israel Deaconess Medical Center Boston, Massachusetts Vice Chair Research Department of Radiology Hadassah Hebrew University Medical Center Hadassah, Israel

Image-Guided Thermal Tumor Ablation: Basic Science and Combination Therapies

Theodore S. Grabow, MD Baltimore Spine Center and Maryland Pain Specialists Adjunct Assistant Professor Department of Anesthesiology and Critical Care Medicine Johns Hopkins University School of Medicine Baltimore, Maryland Facet Joint Injection Epidural Steroid Injection

CONTRIBUTORS Edward D. Greenberg, MD Interventional Neuroradiology Fellow Department of Radiology Weill Cornell Medical College New York-Presbyterian Hospital New York, New York

Acute Stroke Management Endovascular Management of Chronic Cerebral Ischemia

Bryan Grieme, MD Medical Director Interventional Radiology St. Anthony Hospital Oklahoma City, Oklahoma

Management of Male Varicocele

Gianluigi Guarnieri, MD Neuroradiology Service Cardarelli Hospital Naples, Italy

Minimally Invasive Disk Interventions

Jeffrey P. Guenette, MD Department of Diagnostic Imaging Warren Alpert Medical School Brown University Providence, Rhode Island

Stéphan Haulon, MD, PhD Professor Department of Vascular Surgery Chief of Vascular Surgery INSERM U1008, Université Lille Nord de France Hôpital Cardiologique—CHRU Lille, France

Management of Thoracoabdominal Aneurysms by Branched Endograft Technology

Klaus A. Hausegger, MD Associate Professor and Head Department of Radiology Institute of Interventional and Diagnostic Radiology Klagenfurt, Austria Endoleaks: Classification, Diagnosis, and Treatment Management of Biliary Leaks

Markus H. Heim, MD Professor of Medicine Department of Bioscience University of Basel Division of Gastroenterology and Hepatology Department of Medicine University Hospital Basel Basel, Switzerland Transjugular Liver Biopsy

Lung Ablation

Klaus D. Hagspiel, MD Professor Departments of Radiology, Medicine (Cardiology), and Pediatrics Chief Division of Noninvasive Cardiovascular Imaging Department of Radiology and Medical Imaging University of Virginia Health System Charlottesville, Virginia Balloon Catheters

Lee D. Hall, MD Interventional Radiologist Department of Radiology Naval Medical Center San Diego, California

Fallopian Tube Interventions

Danial K. Hallam, MD, MSc Associate Professor Departments of Radiology and Neurological Surgery University of Washington School of Medicine Seattle, Washington Use of Skull Views in Visualization of Cerebral Vascular Anatomy

Mohomad Hamady, MBChB, FRCR Consultant and Senior Lecturer Department of Interventional Radiology Imperial College London, United Kingdom

Fenestrated Stent-Grafting of Juxtarenal Aortic Aneurysms

Katharine Henderson, MS Genetic Counselor Yale School of Medicine Co-Director Yale HHT Center New Haven, Connecticut

Pulmonary Arteriovenous Malformations: Diagnosis and Management

Robert C. Heng, MBBS, FRANZCR Interventional Radiologist Department of Radiology and Medical Imaging Launceston General Hospital Launceston, Tasmania, Australia

Vascular Anatomy of the Thorax, Including the Heart

Joshua A. Hirsch, MD Associate Professor of Radiology Harvard Medical School Department of Radiology Massachusetts General Hospital Boston, Massachusetts Vertebroplasty and Kyphoplasty

Andrew Hugh Holden, MBChB, FRANCZR Associate Professor Department of Radiology Auckland University School of Medicine Director Body Imaging and Interventional Services Auckland City Hospital Auckland, New Zealand Acute Renal Ischemia

xv

xvi

CONTRIBUTORS Edward T. Horn, PharmD, BCPS Clinical Pharmacy Specialist Department of Pharmacy Allegheny General Hospital Pittsburgh, Pennsylvania Vasoactive Agents

Abdel Aziz A. Jaffan, MD Assistant Professor of Radiology Division of Interventional Radiology and Image-Guided Medicine Emory University School of Medicine Atlanta, Georgia Aortoiliac Revascularization

Joseph A. Hughes, MD Fellow, Cardiovascular and Interventional Radiology Department of Radiology Pennsylvania State University College of Medicine Penn State Hershey Medical Center Hershey, Pennsylvania

Peripherally Inserted Central Catheters and Nontunneled Central Venous Catheters

Elizabeth A. Ignacio, MD Associate Professor of Radiology Division of Vascular and Interventional Radiology George Washington University School of Medicine and Health Sciences Washington, DC Interventional Radiologist Department of Radiology Maui Memorial Medical Center Wailuku, Hawaii Vascular Anatomy of the Pelvis Peripartum Hemorrhage

Zubin D. Irani, MD, MBBS Instructor in Radiology Harvard Medical School Department of Medical Imaging Massachusetts General Hospital Boston, Massachusetts Bronchial Artery Embolization

Priya Jagia, MD, DNB Associate Professor Department of Cardiac Radiology All India Institute of Medical Sciences New Delhi, India Management of Vascular Arteritidies

Raagsudha Jhavar, MD Special Volunteer Urologic Oncology National Cancer Institute US National Institutes of Health Bethesda, Maryland

Principles of Thrombolytic Agents

Francis Joffre, MD Professeur Service de Radiologie L’Hôpital de Rangueil Universitaire de Toulouse Toulouse, France

Vascular Intervention in the Liver Transplant Patient

Matthew S. Johnson, MD Professor Departments of Radiology and Surgery Indiana University School of Medicine Indianapolis, Indiana Restenosis

Roberto Izzo, MD Neuroradiology Service Carderelli Hospital Naples, Italy James E. Jackson, MBBS, FRCP, FRCR Consultant Interventional Radiologist Department of Imaging Hammersmith Hospital London, United Kingdom

Management of High-Flow Vascular Anomalies Management of Visceral Aneurysms

Augustinus Ludwig Jacob, Prof MD Professor Department of Radiology University of Basel Head Division of Interventional Radiology Institute of Radiology Basel, Switzerland Transjugular Liver Biopsy

Amber Jones, CCRP Senior Research Program Coordinator Fellowship Program Coordinator Endovascular Surgical Neuroradiology Division of Interventional Neuroradiology Johns Hopkins University School of Medicine Baltimore, Maryland Vascular Anatomy of the Upper Extremity Abdominal Aorta and the Inferior Vena Cava

Verena Kahn, MD Department of Diagnostic and Interventional Radiology University of Frankfurt am Main Frankfurt, am Main, Germany Endovascular Laser Therapy

Özlem Tugçe ˘ Kalayci, MD Department of Radiology Inonu University Medical Faculty Malayta, Turkey

Congenital Vascular Anomalies: Classification and Terminology

CONTRIBUTORS xvii Sanjeeva Prasad Kalva, MD Assistant Radiologist Department of Imaging Massachusetts General Hospital Assistant Professor Department of Radiology Harvard Medical School Boston, Massachusetts Noninvasive Vascular Diagnosis

Anthony W. Kam, MD, PhD Assistant Professor Department of Radiology Johns Hopkins Medical Institutions Baltimore, Maryland

Arteriography and Arterial Stimulation with Venous Sampling for Localizing Pancreatic Endocrine Tumors

Krishna Kandarpa, MD, PhD Professor and Chair Department of Radiology University of Massachusetts School of Medicine Adjunct Professor Biomedical Engineering Worcester Polytechnic Institute Radiologist-in-Chief Department of Radiology University of Massachusetts Memorial Health Care Worcester, Massachusetts Acute Lower Extremity Ischemia

Zinvoy M. Katz, MD Clinical Instructor Division of Interventional Neuroradiology Johns Hopkins University School of Medicine Baltimore, Maryland

Frederick S. Keller, MD Cook Professor Dotter Interventional Institute Oregon Health and Sciences University Portland, Oregon Bronchial Artery Embolization

Robert K. Kerlan, Jr., MD Professor of Clinical Radiology and Surgery Department of Radiology University of California–San Francisco San Francisco, California

Biliary Complications Associated with Liver Transplantation

David O. Kessel, MBBS, MA, MRCP, FRCR, EBIR Consultant Radiologist Department of Vascular Interventional Radiology Leeds Teaching Hospitals NHS Trust Leeds, United Kingdom Principles of Venous Access

Ramy Khalil, MS Department of Radiology George Washington University School of Medicine and Health Sciences Washington, DC Renal Vasculature

Nadia J. Khati, MD Associate Professor of Radiology Abdominal Imaging Section George Washington University Medical Center Washington, DC Alimentary Tract Vasculature Renal Vasculature Vascular Anatomy of the Pelvis

Carotid Revascularization

John A. Kaufman, MD, MS Director Dotter Interventional Institute Oregon Health and Science University Portland, Oregon Invasive Vascular Diagnosis Superior Vena Cava Occlusive Disease

Linda Kelahan, MD Department of Radiology Washington Hospital Center Washington, DC Bipedal Lymphangiography

Alexis D. Kelekis, MD, PhD, EBIR Assistant Professor of Interventional Radiology Second Radiology Department Attikon University Hospital University of Athens Athens, Greece

Ablation and Combination Treatments of Bony Lesions Sacroiliac Joint Injections

Neil M. Khilnani, MD Associate Professor of Clinical Radiology Division of Vascular and Interventional Radiology Weill Cornell Medical College Division of Cardiovascular and Interventional Radiology New York-Presbyterian Hospital New York, New York

Great Saphenous Vein Ablation Magnetic Resonance–Guided Focused Ultrasound Treatment of Uterine Leiomyomas

Darren D. Kies, MD Assistant Professor of Radiology Department of Radiology and Imaging Sciences Emory School of Medicine Atlanta, Georgia

Management of Female Venous Congestion Syndrome

Hyun S. Kim, MD Associate Professor of Radiology, Obstetrics and Gynecology, Hematology and Medical Oncology, and Surgery Director Interventional Radiology and Image-Guided Medicine Department of Radiology and Imaging Sciences Emory School of Medicine Atlanta, Georgia Management of Female Venous Congestion Syndrome Percutaneous Cholecystostomy

xviii CONTRIBUTORS Jin Hyoung Kim, MD, PhD Assistant Professor Department of Radiology Asan Medical Center University of Ulsan College of Medicine Seoul, Republic of Korea

Intervention for Gastric Outlet and Duodenal Obstruction

Kyung Rae Kim, MD Assistant Professor Department of Radiology University of North Carolina School of Medicine Chapel Hill, North Carolina Hemodialysis Access: Catheters and Ports

Hiro Kiyosue, MD Associate Professor Department of Radiology Oita Medical University Yufu, Oita, Japan

Retrograde Balloon Occlusion Variceal Ablation

Sebastian Kos, MD, EBIR Chairman Institute of Radiology and Nuclear Medicine Lucerne, Switzerland Transjugular Liver Biopsy

Jim Koukounaras, MBBS, FRANCZR Interventional Radiologist Department of Radiology The Alfred Hospital Melbourne, Victoria, Australia Adrenal Venous Sampling

Andres Krauthamer, MD Resident Department of Diagnostic Radiology George Washington University Medical Center Washington, DC

Principles of Intraprocedural Analgesics and Sedatives Vascular Anatomy of the Pelvis

Venkatesh Krishnasamy, MD Interventional Radiology Fellow Division of Vascular and Interventional Radiology George Washington University School of Medicine and Health Sciences Washington, DC Vascular Anatomy of the Pelvis

William T. Kuo, MD Associate Professor Director, IR Fellowship Program Division of Vascular and Interventional Radiology Stanford University Medical Center Stanford, California

Max Kupershmidt, MBBS, FRANCZR, MMed (Radiology) Director of Ultrasound Barwon Medical Imaging Geelong, Victoria, Australia Treatment of High-Flow Priapism and Erectile Dysfunction

Vineel Kurli, MMBS Vascular and Interventional Radiologist Medical Diagnostic Imaging Group Phoenix, Arizona

Acute Upper Extremity Deep Venous Thrombosis

Jeanne M. LaBerge, MD Professor in Residence Department of Radiology University of California–San Francisco San Francisco, California

Biliary Complications Associated with Liver Transplantation

Pierre-Yves Laffy, MD Cardio-vasculaire Radiologie Hopital Européen Georges Pompidou Paris, France Infrapopliteal Revascularization

Leo P. Lawler, MD, MBBS, BAO, FRCR Assistant Professor of Radiology The Russell H. Morgan Department of Radiology and Radiological Science Johns Hopkins Medical Institutions Baltimore, Maryland Percutaneous Cholecystostomy

McKinley C. Lawson, MD, PhD Resident Diagnostic Radiology Department University of Colorado–Denver Anschutz Medical Center Aurora, Colorado Embolization Agents

Judy M. Lee, MD Assistant Professor Department of Obstetrics and Gynecology Johns Hopkins University School of Medicine Baltimore, Maryland

Management of Female Venous Congestion Syndrome

Michael J. Lee, M.Sc, FRCPI, FRCR, FFR(RCSI), FSIR, EBIR Professor of Radiology Consultant Interventional Radiologist Beaumont Hospital Professor of Radiology Royal College of Surgeons in Ireland Department of Radiology Dublin, Ireland Gastrostomy and Gastrojejunostomy

Percutaneous Interventions for Acute Pulmonary Embolism

Thomas Lemettre, MD Radiologue Clinique Claude Bernard Albi, France

Vascular Intervention in the Liver Transplant Patient

CONTRIBUTORS Riccardo Lencioni, MD, PhD Associate Professor Department of Radiology University of Pisa Faculty of Medicine Director Division of Diagnostic Imaging and Intervention Department of Hepatology and Liver Transplantation Pisa University Hospital Pisa, Italy Energy-Based Ablation of Hepatocellular Cancer

Robert J. Lewandowski, MD Associate Professor Department of Radiology Feinberg School of Medicine Chicago, Illinois

Radioembolization of Liver Metastases Percutaneous Biopsy Treatment of Effusions and Abscesses

John J. Lewin, III, PharmD, MBA, BCPS Division Director Critical Care and Surgery Department of Pharmacy Johns Hopkins Hospital Adjunct Assistant Professor Department of Anesthesiology and Critical Care Medicine Johns Hopkins University School of Medicine Baltimore, Maryland Vasoactive Agents

Yean L. Lim, MD, PhD Professioral Fellow Department of Cardiology University of Melbourne Faculty of Medicine Professor and Director Centre for Cardiovascular Therapeutics Western Health Melbourne, Victoria, Australia Director Raffles Heart Hospital Changhi, Singapore Permanent Secretary Asian Pacific Society of Interventional Cardiology Wanchai, Hong Kong

Vascular Anatomy of the Thorax, Including the Heart

Raymond W. Liu, MD Assistant Radiologist Division of Vascular Imaging and Intervention Massachusetts General Hospital Boston, Massachusetts Acute Lower Extremity Ischemia

Rafael H. Llinas, MD Associate Professor Department of Neurology Johns Hopkins University School of Medicine Clinical Vice Chair of Neurology Johns Hopkins Hospital Baltimore, Maryland

Cerebral Functional Anatomy and Rapid Neurological Examination

Curtis A. Lewis, MD, MBA, JD Assistant Professor of Radiology Division of Interventional Radiology and Image-Guided Medicine Emory School of Medicine Atlanta, Georgia

Reinhard Loose, MD, PhD Department of Diagnostic and Interventional Radiology Klinikum Nürmberg-Nord Nürmberg, Bavaria, Germany Angioplasty

Management of Extremity Vascular Trauma

Changqing Li, MB Interventional Radiologist Department of Radiology Beijing Ditan Hospital Beijing, China

Transjugular Intrahepatic Portosystemic Shunts

Eleni A. Liapi, MD Instructor Interventional Radiology Department Johns Hopkins Medical Institutions Baltimore, Maryland

Radioembolization for Hepatocellular Carcinoma

Stuart M. Lyon, MBBS, FRNZCR Adjunct Clinical Associate Professor Department of Surgery Central Clinical School Monash University Director of Interventional Radiology Alfred Hospital Melbourne, Victoria, Australia

Acute Arterial Occlusive Disease of the Upper Extremity Endovascular Treatment of Peripheral Aneurysms Percutaneous Biopsy of the Lung, Mediastinum, and Pleura

Sumaira Macdonald, MBChB, FRCR, PhD, EBIR Consultant Vascular Radiologist Department of Interventional Radiology Freeman Hospital Newcastle upon Tyne United Kingdom Embolic Protection Devices

xix

xx

CONTRIBUTORS Patrick C. Malloy, MD Chairman Department of Radiology VA New York Harbor Healthcare System New York, New York

Management of Pelvic Hemorrhage in Trauma

Mark D. Mamlouk, MD Department of Radiological Sciences University of California–Irvine Orange, California Cryoablation of Liver Tumors

Michael J. Manzano, MD Radiologist Private Practice Pittsford, New York

Alimentary Tract Vasculature

Marie Agnès Marachet, MD Praticien Hospitalier Service de Radiologie L’Hôpital de Rangueil Universitaire de Toulouse Toulouse, France

Vascular Intervention in the Liver Transplant Patient

Jean-Baptiste Martin, MD Associate Radiologist Radiology Department Geneva University Hospital Geneva, Switzerland

Ablation and Combination Treatments of Bony Lesions

Antonio Martínez-Cuesta, MD, MSc, FRCR Department of Radiology Hospital de Navarra Pamplona, Spain Portal-Mesenteric Venous Thrombosis

M. Victoria Marx, MD Professor of Clinical Radiology Radiology Keck School of Medicine University of Southern California Los Angeles, California

Radiation Safety and Protection in the Interventional Fluoroscopy Environment

Surena F. Matin, MD Associate Professor Department of Urology University of Texas MD Anderson Cancer Center Houston, Texas

Thermal Ablation of Renal Cell Carcinoma

Alan H. Matsumoto, MD, FSIR, FACR, FAHA Professor and Chair Department of Radiology University of Virginia School of Medicine Division of Interventional Radiology, Angiography, and Special Procedures Department of Radiology and Medical Imaging University of Virginia Health System Charlottesville, Virginia Balloon Catheters Closure Devices Chronic Mesenetric Ischemia Tracheobronchial Interventions

Matthew A. Mauro, MD, FACR The Ernest H. Wood Distinguished Professor of Radiology and Surgery Chairman, Department of Radiology University of North Carolina School of Medicine Chapel Hill, North Carolina

Percutaneous Management of Chronic Lower Extremity Venous Occlusive Disease

Gordon McLennan, MD, FISR Department of Diagnostic Radiology Cleveland Clinic Foundation Cleveland, Ohio Restenosis

Simon J. McPherson, MRCP, FRCR Consultant Vascular and Interventional Radiologist Department of Radiology Leeds General Infirmary Leeds, United Kingdom

Management of Upper Gastrointestinal Hemorrhage

Khairuddin Memon, MD Clinical Research Associate Department of Radiology Feinberg School of Medicine Chicago, Illinois

Radioembolization of Liver Metastases

Steven G. Meranze, MD Professor of Radiology and Surgery Vice-Chair Department of Radiology and Radiological Science Vanderbilt University School of Medicine Nashville, Tennessee

Percutaneous Nephrostomy, Cystostomy, and Nephroureteral Stenting

Todd S. Miller, MD Faculty Department of Interventional Neuroradiology Montefiore Medical Center Assistant Professor of Clinical Radiology Department of Radiology Albert Einstein College of Medicine Bronx, New York Management of Head and Neck Tumors

CONTRIBUTORS Robert J. Min, MD, MBA Chairman of Radiology Weill Cornell Medical College Radiologist-in-Chief New York-Presbyterian Hospital New York, New York Great Saphenous Vein Ablation

Sally E. Mitchell, MD, FISR, FCIRSE Professor of Radiology, Surgery, and Pediatrics Division of Interventional Radiology Johns Hopkins Hospital Baltimore, Maryland

Kieran P.J. Murphy, MB, FRCPC, FSIR Professor of Radiology Joint Department of Medical Imaging University Health Network Mount Sinai Hospital Women’s College Hospital Toronto, Ontario, Canada

Endovascular Management of Epistaxis Subarachnoid Hemorrhage Image-Guided Intervention for Symptomatic Tarlov Cysts Scalene Blocks and Their Role in Thoracic Outlet Syndrome

Stephan Moll, MD Associate Professor Department of Medicine Division of Hematology-Oncology University of North Carolina School of Medicine Chapel Hill, North Carolina

Timothy P. Murphy, MD, FSIR, FAHA, FSVB, FACR Professor Department of Diagnostic Imaging Warren Alpert Medical School Brown University Director Vascular Disease Research Center Department of Diagnostic Imaging Rhode Island Hospital Providence, Rhode Island

Robert A. Morgan, MBChB, MRCP, FRCR, EBIR Consultant, Vascular and Interventional Radiologist Radiology Department St George’s NHS Trust London, United Kingdom

Mario Muto, MD Chair of Neuroradiology Service Carderelli Hospital Naples, Italy

Congenital Vascular Anomalies: Classification and Terminology Urodynamics

Antiplatelet Agents and Anticoagulants

Aortic Endografting

Hiromu Mori, MD Professor and Chairperson Department of Radiology Oita Medical University Yufu, Oita, Japan

Retrograde Balloon Occlusion Variceal Ablation

Paul R. Morrison, MS Medical Physicist Department of Radiology Harvard Medical School Brigham and Women’s Hospital Boston, Massachusetts Cryoablation of Liver Tumors

Stefan Müller-Hülsbeck, MD, EBIR, FCIRSE, FICA Professor of Radiology Head Department of Diagnostic and Interventional Radiology/Neuroradiology Ev.-Luth. Diakonissenanstalt zu Flensburg Flensburg, Germany Atherectomy Devices

Stents Aortoiliac Revascularization

Minimally Invasive Disk Interventions Periradicular Therapy

Albert A. Nemcek, Jr., MD Professor Department of Radiology Feinberg School of Medicine Staff Interventional Radiologist Northwestern Memorial Hospital Chicago, Illinois

Percutaneous Biopsy Treatment of Effusions and Abscesses

David B. Nicholson, C-RT Radiologic Technologist Department of Radiology University of Virginia School of Medicine Charlottesville, Virginia Balloon Catheters

Ali Noor, MD Department of Radiology Mount Sinai Medical Center New York, New York Renal Vasculature Bipedal Lymphangiography

Philippe Otal, MD Professeur Service de Radiologie L’Hôpital Rangueil Universitaire de Toulouse Toulouse, France

Vascular Intervention in the Liver Transplant Patient

xxi

xxii CONTRIBUTORS Randall P. Owen, MD Department of Otolaryngology-Head and Neck Surgery Mount Sinai Medical Center New York, New York Management of Head and Neck Tumors

Auh Whan Park, MD Associate Professor Department of Radiology University of Virginia Medical Center Charlottesville, Virginia Closure Devices Tracheobronchial Interventions

David A. Pastel, MD Assistant Professor Department of Radiology Dartmouth-Hitchcock Medical Center Lebanon, New Hampshire Vertebroplasty and Kyphoplasty

Aalpen A. Patel, MD Vice Chair Clinical Operations System Radiology Geisinger Health System Danville, Pennsylvania

Management of Clotted Hemodialysis Access Grafts

Rafiuddin Patel, MBChB (Honours), MRCS, FRCR SpR in Interventional Radiology Department of Vascular Radiology Leeds General Infirmary Leeds, United Kingdom Management of Upper Gastrointestinal Hemorrhage

Daniel D. Picus, MD Professor Departments of Radiology and Surgery Washington University School of Medicine Chief Division of Diagnostic Radiology Interventional Radiology Section Mallinckrodt Institute of Radiology Barnes-Jewish Hospital Saint Louis, Missouri Management of Biliary Calculi

João M. Pisco, MD Professor and Chair Department of Radiology Centro Hospitalar Lisboa Norte EPE Hospital Pulido Valente, Lisbon, Portugal

Treatment of High-Flow Priapism and Erectile Dysfunction

Jeffrey S. Pollak, MD Professor of Radiology Co-Section Chief Division of Vascular and Interventional Radiology Department of Diagnostic Radiology Yale School of Medicine New Haven, Connecticut Pulmonary Arteriovenous Malformations: Diagnosis and Management

Rupert H. Portugaller, MD Associate Professor Department of Vascular and Interventional Radiology University Clinic of Radiology Medical University Graz Graz, Austria Management of Biliary Leaks

Monica Smith Pearl, MD Assistant Professor of Radiology Division of Interventional Neuroradiology Johns Hopkins Hospital Baltimore, Maryland Director of Pediatric Neurointervention Department of Radiology Children’s National Medical Center Washington, DC

Vascular Anatomy of the Upper Extremity Abdominal Aorta and the Inferior Vena Cava

Olivier Pellerin, MD, MSc Faculté de Médecine Université Paris Descartes Sorbonne Paris-Cité Paris, France

Infrapopliteal Revascularization

Sarah Power, MD Resident Department of Radiology Beaumont Hospital Dublin, Ireland

Gastrostomy and Gastrojejunostomy

Denis Primakov, MD Interventional Radiologist Walter Reed National Military Medical Center Assistant Professor of Radiology and Radiological Sciences Uniformed Services University of the Health Sciences Bethesda, Maryland Bipedal Lymphangiography

David S. Pryluck, MD Fellow Vascular and Interventional Radiology Department of Radiology University of Pennsylvania School of Medicine Philadelphia, Pennsylvania

Acute Upper Extremity Deep Venous Thrombosis Management of Fluid Collections in Acute Pancreatitis Chemical and Thermal Ablation of Desmoid Tumors

CONTRIBUTORS xxiii Martin G. Radvany, MD Assistant Professor of Radiology, Neurological Surgery, and Neurology Division of Interventional Neuroradiology Johns Hopkins University School of Medicine Baltimore, Maryland Carotid Revascularization

Batya R. Radzik, MSN, CRNP, BC Nurse Practitioner Neurocritical Care Department of Anesthesia and Critical Care Medicine Johns Hopkins Hospital Baltimore, Maryland Cerebral Functional Anatomy and Rapid Neurological Examination

Suman Rathbun, MD, MS, RVT Professor Department of Medicine University of Oklahoma Health Sciences Center Oklahoma City, Oklahoma

Anne Roberts, MD Professor Department of Radiology University of California–San Diego Chief, Vascular and Interventional Radiology UCSD Medical Center Department of Radiology VA Medical Center San Diego, California Surveillance of Hemodialysis Access

Alain Roche, MD Professor and Chief Department of Medical Imaging Institut Gustave Roussy Villejuif, France Director Laboratory UPRES EA 4040 Department of Interventional Radiology University of Paris South Paris, France Portal Vein Embolization

Management of Risk Factors for Peripheral Artery Disease

Anne Ravel, MD Faculty of Medicine University Hospital Clermont-Ferrand, France Acute Mesenteric Ischemia

Charles E. Ray, Jr., MD, PhD Professor and Vice-Chair of Research Department of Radiology Chief Division of Interventional Radiology University of Colorado–Denver Anschutz Medical Center Aurora, Colorado Embolization Agents

Hervé Rousseau, MD Professeur et Chef de service Service de Radiologie L’Hôpital Rangueil Universitaire de Toulouse Toulouse, France

Vascular Intervention in the Liver Transplant Patient

Stefan G. Ruehm, MD, PhD Professor Department of Radiological Sciences David Geffen School of Medicine at UCLA University of California–Los Angeles Director Cardiovascular Imaging Santa Monica-UCLA Medical Center and Orthopedic Hospital Los Angeles, California Clinical Manifestations of Lymphatic Disease

Mahmood K. Razavi, MD Director Center for Clinical Trials and Research Heart and Vascular Center St. Joseph Hospital Orange, California

Diego San Millán Ruiz, MD Neuroradiologist Department of Diagnostic and Interventional Radiology Hospital of Sion Sion, Switzerland

Aaron Reposar Medical Student George Washington University School of Medicine and Health Sciences Washington, DC

John H. Rundback, MD, FAHA, FSVM, FSIR Medical Director Interventional Institute Holy Name Medical Center Teaneck, New Jersey

Endovascular Management of Chronic Femoropopliteal Disease

Renal Vasculature

Craniocervical Vascular Anatomy

Renovascular Intervention Chemical Ablation of Liver Lesions

Wael E.A. Saad, MD, FSIR Associate Professor of Radiology Division of Angiography and Interventional Radiology University of Virginia School of Medicine Charlottesville, Virginia Balloon Catheters Management of Postcatheterization Pseudoaneurysms Management of Benign Biliary Strictures

xxiv CONTRIBUTORS Tarun Sabharwal, MBChB, FRCSI, FRCR Consultant Interventional Radiologist and Honorary Senior Lecturer Department of Radiology Guy’s and St. Thomas’ Hospital London, United Kingdom Esophageal Intervention in Malignant and Benign Esophageal Disease

Riad Salem, MD, MBA Professor Departments of Radiology, Medicine (Hematology-Oncology), and Surgery Feinberg School of Medicine Director Interventional Oncology Robert H. Lurie Comprehensive Cancer Center Northwestern Memorial Hospital Chicago, Illinois Radioembolization of Liver Metastases

Marc Sapoval, MD, PhD Head Division of Interventional Radiology Hopital Européen Georges Pompidou Paris, France Infrapopliteal Revascularization

Shawn N. Sarin, MD Assistant Professor Departments of Radiology and Surgery Section of Vascular and Interventional Radiology George Washington University Medical Center Washington, DC Balloon Catheters Angioplasty

Matthew P. Schenker, MD Associate Radiologist Division of Angiography and Interventional Radiology Brigham and Women’s Hospital Boston, Massachusetts

Thoracic Duct Embolization for Postoperative Chylothorax

Marc H. Schiffman, MD Assistant Professor of Radiology Division of Interventional Radiology New York Hospital Weill Cornell Medical College New York, New York

Magnetic Resonance–Guided Focused Ultrasound Treatment of Uterine Leiomyomas

Sanjiv Sharma, MD Professor Department of Cardiac Radiology All India Institute of Medical Sciences New Delhi, India Management of Vascular Arteritidies

Ji Hoon Shin, MD, PhD Associate Professor Department of Radiology University of Ulsan College of Medicine Research Institute of Radiology Asan Medical Center Seoul, Korea Tracheobronchial Interventions

H. Omur Sildiroglu, MD Department of Radiology and Medical Imaging University of Virginia Health System Charlottesville, Virginia Chronic Mesenteric Ischemia

Naomi N. Silva, MD Resident Department of Radiology Robert Wood Johnson University Hospital New Brunswick, New York

Vascular Anatomy of the Pelvis

Stuart G. Silverman, MD Professor Department of Radiology Harvard Medical School Director of Abdominal Imaging and Intervention Brigham and Women’s Hospital Boston, Massachusetts Cryoablation of Liver Tumors

Charan K. Singh, MMBS Assistant Professor of Radiology Department of Diagnostic Imaging and Therapeutics University of Connecticut School of Medicine Farmingham, Connecticut Acute Upper Extremity Deep Venous Thrombosis Management of Fluid Collections in Acute Pancreatitis

Tony P. Smith, MD Professor Department of Radiology Duke University School of Medicine Division Chief Peripheral and Neurological Interventional Radiology Department of Radiology Duke University Medical Center Durham, North Carolina Antibiotic Prophylaxis in Interventional Radiology

Constantinos T. Sofocleous, MD, PhD, FSIR Interventional Radiologist Department of Radiology Memorial Sloan-Kettering Cancer Center New York, New York

Embolotherapy for the Management of Liver Malignancies Other Than Hepatocellular Carcinoma

CONTRIBUTORS Luigi Solbiati, MD University of Milan Postgraduate Medical School Milan, Italy Department of Radiology and Interventional Radiology Busto Arsizio General Hospital Busto Arsizio, Italy

Stavros Spiliopoulos, MD, PhD, EBIR Board Certified Interventional Radiologist Department of Interventional Radiology Patras University Hospital Rion, Greece

Esophageal Intervention in Malignant and Benign Esophageal Disease

Energy-Based Ablation of Other Liver Lesions

Stephen B. Solomon, MD Chief Interventional Radiology Service Director Center for Image-Guided Intervention Memorial Sloan-Kettering Cancer Center New York, New York

Magnetic Resonance–Guided Focused Ultrasound Treatment of Uterine Leiomyomas

Ho-Young Song, MD, PhD Professor Department of Radiology Asan Medical Center University of Ulsan College of Medicine Seoul, Republic of Korea

Intervention for Gastric Outlet and Duodenal Obstruction

Kean H. Soon, MMBS, PhD, FRACP, FCSANZ Interventional Cardiologist Centre for Cardiovascular Therapeutics Western Hospital Melbourne, Victoria, Australia

Vascular Anatomy of the Thorax, Including the Heart

David R. Sopko, MD Associate Department of Radiology Duke University School of Medicine Durham, North Carolina

Antibiotic Prophylaxis in Interventional Radiology

Thomas A. Sos, MD Professor Department of Radiology Weill Cornell Medical College New York-Presbyterian Hospital New York, New York Renal Vein Renin Sampling

Michael C. Soulen, MD, FSIR, FCIRSE Professor of Radiology Department of Interventional Radiology University of Pennsylvania School of Medicine Philadelphia, Pennsylvania

Chemoembolization for Hepatocellular Carcinoma

James B. Spies, MD, MPH Professor and Chair Department of Radiology Georgetown University Hospital Washington, DC Uterine Fibroid Embolization

M.J. Bernadette Stallmeyer, MD, PhD Director Division of Interventional Neuroradiology Reading Hospital and Medical Center West Reading, Pennsylvania Management of Head and Neck Injuries

Joseph M. Stavas, MD Professor Department of Radiology University of North Carolina School of Medicine Chapel Hill, North Carolina Biopsy Devices

LeAnn S. Stokes, MD Assistant Professor Department of Radiology and Radiological Sciences Vanderbilt University Medical Center Nashville, Tennessee Percutaneous Nephrostomy, Cystostomy, and Nephroureteral Stenting

Ernst-Peter Strecker, MD Professor Emeritus Consultant Physician Department of Diagnostic and Interventional Radiology Trudpert Klinikum Pforzheim Pforzheim, Germany Intraarterial Ports for Chemotherapy

Michael B. Streiff, MD Associate Professor of Medicine Department of Medicine (Hematology) Johns Hopkins Medical Institutions Baltimore, Maryland Principles of Thrombolytic Agents

Deepak Sudheendra, MD Assistant Professor of Clinical Radiology and Surgery University of Pennsylvania School of Medicine Division of Interventional Radiology Hospital of the University of Pennsylvania Philadelphia, Pennsylvania Management of Renal Angiomyolipoma Thermal Ablation of the Adrenal Gland

Paul V. Suhocki, MD Associate Professor Department of Radiology Duke University Medical Center Durham, North Carolina Foreign Body Retrieval

xxv

xxvi CONTRIBUTORS Alfonso Tafur, MD, RPVI Assistant Professor of Medicine Department of Vascular Medicine Oklahoma University Health and Science Center Oklahoma City, Oklahoma

Management of Risk Factors for Peripheral Artery Disease

M. Reza Taheri, MD, PhD Assistant Professor Department of Radiology George Washington University School of Medicine and Health Sciences Washington, DC

Use of Skull Views in Visualization of Cerebral Vascular Anatomy

Jeff Dai-Chee Tam, MBBS, FRANCZR Fellow Interventional Radiology Department of Radiology Alfred Hospital Melbourne, Victoria, Australia

Anatomy of the Lower Limb Acute Arterial Occlusive Disease of the Upper Extremity Parathyroid Venous Sampling

Elizabeth R. Tang, MD Resident Department of Radiology Boston University Boston, Massachusetts

Management of Head and Neck Tumors

Emily M. Tanski, PA-C Physicians Assistant Division of Vascular and Interventional Radiology George Washington University School of Medicine Washington, DC Vascular Anatomy of the Pelvis Management of Male Varicocele

Kiang Hiong Tay, MBBS, FRCR, FAMS, FSIR Head and Senior Consultant Department of Diagnostic Radiology Singapore General Hospital Associate Professor of Radiology Duke NUS Graduate Medical School Yong Loo Lin School of Medicine National University of Singapore Chairman Cardiovascular and Interventional Radiology Subsection Singapore Radiological Society Honorary Secretary College of Radiologists Singapore Peripartum Hemorrhage

Aylin Tekes, MD Assistant Professor Department of Radiology Johns Hopkins University School of Medicine Radiologist Division of Pediatric Radiology Johns Hopkins Hospital Baltimore, Maryland

Congenital Vascular Anomalies: Classification and Terminology

Matthew M. Thompson, MD, FRCS Professor of Vascular Surgery St. George’s Vascular Institute St. George’s Hospital London, United Kingdom Aortic Endografting

Kenneth R. Thomson, MD, FRANZCR Adjunct Clinical Professor of Radiology Department of Surgery Central Clinical School Monash University Program Director Department of Radiology and Nuclear Medicine Alfred Hospital Melbourne, Victoria, Australia

Anatomy of the Lower Limb Acute Arterial Occlusive Disease of the Upper Extremity Endovascular Treatment of Peripheral Aneurysms Treatment of High-Flow Priapism and Erectile Dysfunction Adrenal Venous Sampling Parathyroid Venous Sampling Percutaneous Biopsy of the Lung, Mediastinum, and Pleura

Raymond H. Thornton, MD Vice Chair for Quality, Safety, and Performance Improvement Department of Radiology Division of Interventional Radiology Memorial Sloan-Kettering Cancer Center New York, New York Management of Malignant Biliary Tract Obstruction

Emily J. Timmreck, RN, MSN, ACNP-BC Nurse Practitioner Division of Vascular and Interventional Radiology George Washington University School of Medicine Washington, DC Vascular Anatomy of the Pelvis Management of Male Varicocele

Jessica Torrente, MD Assistant Professor of Radiology Division of Breast Imaging and Intervention George Washington University School of Medicine and Health Sciences Washington, DC

Minimally Invasive Image-Guided Breast Biopsy and Ablation

CONTRIBUTORS xxvii Gina D. Tran, MD Resident Department of Family and Community Medicine University of Nevada School of Medicine Las Vegas, Nevada A Brief History of Image-Guided Therapy

Scott Trerotola, MD Associate Chair and Chief Division of Interventional Radiology University of Pennsylvania Medical Center Philadelphia, Pennsylvania

Eric vanSonnenberg, MD Visiting Professor of Medicine David Geffen School of Medicine at UCLA Chief Academic Officer Chief of Interventional Radiology and Interventional Oncology Kern/UCLA Medical Center Bakersfield, California Professor Arizona State University Tempe, Arizona Cryoablation of Liver Tumors

Management of Clotted Hemodialysis Access Grafts

David W. Trost, MD Associate Professor of Clinical Radiology Division of Vascular and Interventional Radiology Weill Medical College of Cornell University Associate Attending Radiologist Department of Radiology New York-Presbyterian Hospital New York, New York Acute Lower Extremity Ischemia Renal Vein Renin Sampling

Kemal Tuncali, MD Instructor in Radiology Harvard Medical School Brigham and Women’s Hospital Boston, Massachusetts Cryoablation of Liver Tumors

Ulku C. Turba, MD Associate Professor of Radiology Division of Interventional Radiology, Angiography, and Special Procedures Rush University Medical Center Chicago, Illinois Balloon Catheters Chronic Mesenteric Ischemia

Mark R. Tyrrell, PhD, FRCS Consultant Vascular Surgeon Department of Vascular Surgery King’s College Hospital London, United Kingdom

Management of Thoracoabdominal Aneurysms by Branched Endograft Technology

Raghuveer Vallabhaneni, MD Assistant Professor of Surgery Division of Vascular Surgery University of North Carolina School of Medicine Chapel Hill, North Carolina Aortic Stent-Grafts

Prasanna Vasudevan, MD Chief Resident Department of Diagnostic Radiology George Washington University Medical Center Washington, DC Angioplasty

Anthony C. Venbrux, MD Director, Cardiovascular and Interventional Radiology George Washington University School of Medicine and Health Sciences Washington, DC A Brief History of Image-Guided Therapy Angioplasty Alimentary Tract Vasculature Renal Vasculature Management of Male Varicocele Management of Female Venous Congestion Syndrome Bipedal Lymphangiography

Bogdan Vierasu, MD Praticien Hospitalier Service Radiologie L’Hôpital de Rangueil Universitaire de Toulouse Toulouse, France

Vascular Intervention in the Liver Transplant Patient

Isabel Vivas, MD Professor of Radiology Universidad de Navarra Consultant Radiologist Clínica Universitaria de Navarra Pamplona, Spain

Portal-Mesenteric Venous Thrombosis

Dierk Vorwerk, MD Professor and Director Department of Radiology Klinikum Ingolstadt Ingolstadt, Germany

Renal Artery Embolization Percutaneous Management of Thrombosis in Native Hemodialysis Shunts

xxviii CONTRIBUTORS David L. Waldman, MD, PhD Professor and Chair Department of Imaging Sciences University of Rochester Strong Memorial Hospital and Highland Hospital Rochester, New York FF Thompson Hospital Canadaigua, New York

Joshua L. Weintraub, MD, FSIR Executive Vice Chairman Department of Radiology Professor of Radiology and Surgery Columbia University College of Physicians and Surgeons New York Presbyterian Hospital New York, New York

Michael J. Wallace, MD Professor Department of Diagnostic Radiology Section Chief Interventional Radiology University of Texas MD Anderson Cancer Center Houston, Texas

Robert I. White, Jr., MD Professor Emeritus and Senior Research Scientist Founder and Former Director Yale HHT Center Yale School of Medicine New Haven, Connecticut

Management of Postcatheterization Pseudoaneurysms

Renovascular Interventions Chemical Ablation of Liver Lesions

Pulmonary Arteriovenous Malformations: Diagnosis and Management

Thermal Ablation of Renal Cell Carcinoma

Eric M. Walser, MD Professor and Interim Chair Department of Radiology University of Texas Medical Branch Galveston, Texas

Endovascular Management of Chronic Femoropopliteal Disease

Antony S. Walton, MD Interventional Cardiologist Alfred Hospital Prahran, Victoria, Australia

Vascular Anatomy of the Thorax, Including the Heart

Thomas J. Ward, MD Resident in Radiology Department of Radiology Mount Sinai School of Medicine New York, New York

Chemical Ablation of Liver Lesions

Anthony F. Watkinson, BSc, MSc (Oxon), MMBS, FRCS, FRCR, EBIR Honorary Professor of Radiology Peninsula College of Medicine and Dentistry Universities of Exeter and Plymouth Consultant Radiologist Royal Devon and Exeter Hospital Exeter, United Kingdom

Mark H. Wholey, MD Director of Peripheral Vascular Interventions Chairman Pittsburgh Vascular Institute University of Pittsburgh Medical Center Shadyside Hospital Pittsburgh, Pennsylvania Carotid Revascularization

C. Jason Wilkins, BMBCh, MRCP, FRCR Consultant Department of Radiology King’s College Hospital London, United Kingdom

Management of Thoracoabdominal Aneurysms by Branched Endograft Technology

Bradford D. Winters, MD, PhD Associate Professor Departments of Anesthesiology and Critical Care Medicine, Neurology and Surgery Johns Hopkins University School of Medicine Baltimore, Maryland Vasoactive Agents

Robert Wityk, MD Associate Professor of Neurology Johns Hopkins University School of Medicine Baltimore, Maryland Cervical Artery Dissection

Transvenous Renal Biopsy

Peter N. Waybill, MD, FSIR Professor of Radiology, Medicine, and Surgery Chief, Division of Cardiovascular and Interventional Radiology Department of Radiology Pennsylvania State University College of Medicine Penn State Hershey Medical Center Hershey, Pennsylvania Peripherally Inserted Central Catheters and Nontunneled Central Venous Catheters

Edward Y. Woo, MD Associate Professor Department of Surgery University of Pennsylvania School of Medicine Vice-Chief and Program Director Division of Vascular Surgery and Endovascular Therapy Director Vascular Laboratory University of Pennsylvania Health System Philadelphia, Pennsylvania

Thoracic Aortic Stent-Grafting and Management of Traumatic Thoracic Aortic Lesions

CONTRIBUTORS xxix Bradford J. Wood, MD Director, Center for Interventional Oncology Chief, Section of Interventional Radiology Senior Investigator National Institutes of Health Bethesda, Maryland

Arteriography and Arterial Stimulation with Venous Sampling for Localizing Pancreatic Endocrine Tumors Thermal Ablation of the Adrenal Gland

Gerald M. Wyse, MBBCh, BAO, MRCPI, FFRCSI Consultant Neuroradiologist Department of Radiology Cork University Hospital Wilton, Cork, Ireland

Steven Zangan, MD Assistant Professor Department of Radiology University of Chicago Pritzker School of Medicine Chicago, Illinois Thombectomy Devices

Fabio Zeccolini, MD Neuroradiology Service Cardarelli Hospital Naples, Italy

Minimally Invasive Disk Interventions

†

Eberhard Zeitler, MD Professor Department of Diagnostic and Interventional Radiology Friedrich-Alexander University Erlangen-Nürmberg Emeritus Director Department of Diagnostic and Interventional Radiology Klinikum Nürmberg-Nord Nürmberg, Bavaria, Germany

Endovascular Management of Epistaxis Subarachnoid Hemorrhage Percutaneous Cholecystostomy Image-Guided Intervention for Symptomatic Tarlov Cysts

Albert J. Yoo, MD Assistant Professor of Radiology Harvard Medical School Department of Radiology Massachusetts General Hospital Boston, Masachusetts Vertebroplasty and Kyphoplasty

Chang Jin Yoon, MD, PhD Associate Professor College of Medicine Seoul National University Seoul, Republic of Korea

Intervention for Gastric Outlet and Duodenal Obstruction

Hyeon Yu, MD Assistant Professor Department of Radiology University of North Carolina School of Medicine Chapel Hill, North Carolina

Percutaneous Management of Chronic Lower Extremity Venous Occlusive Disease Hemodialysis Access: Catheters and Ports

Angioplasty

Dianbo Zhang, MD Assistant Professor Department of Radiology State University of New York SUNY Upstate Medical Center Syracuse, New York Restenosis

Christoph L. Zollikofer, MD Professor Department of Radiology Kantonsspital Baden Baden, Switzerland

Preoperative and Palliative Colonic Stenting

†

Deceased

Preface The second edition of Image-Guided Interventions builds on the success of the first as an international body of work that draws on the knowledge and experience of distinguished contributors from around the world. We believe the first edition fulfilled our intent of producing a practical, concise, well-illustrated textbook representative of the state of the art of interventional practice across the globe. Because modern-day textbooks are no longer needed for an exhaustive list of the current literature, only critical references were included. The second edition is a continuation of this goal. Its five principal editors were selected for their expertise and global representation. Four have returned for the second edition: Drs. Murphy, Venbrux, and Mauro from North America and Dr. Thomson from the Asia-Oceania region. We bid a fond farewell to Christoph Zollikofer (Europe), who has finally entered a well-deserved retirement, and welcome Robert Morgan as our European representative. The second edition has been reduced to one volume. Several components of the text are offered only as an online resource; these include references and supplementary material. All of the core material with accompanying highquality color illustrations are present in the traditional printed book form. The first edition received particular acclaim for its radiographic images, tables, charts, and color anatomic illustrations. These have been maintained and enhanced. We have also retained the basic organization of the text into two basic parts: vascular interventions and nonvascular Interventions. Part 1, Vascular Interventions, includes not only diagnosis and intervention of primary vascular disorders (arterial, venous, and lymphatic) but also diseases in other organ systems in which procedures are performed using the vascular system as a conduit for intervention. Part 2, Nonvascular Interventions, covers procedures performed via direct image-guided percutaneous access into the organ or site of interest. Part 1 again begins with core principles including vascular diagnosis, the instruments of intervention, patient care, and the principles

of vascular intervention. The subsequent focus of Part 1 is on primary vascular disorders, including arterial, venous, and lymphatic diseases, followed by an anatomic-based discussion of a wide variety of entities in which the vascular tree is used as a conduit for intervention. Part 2 begins with an updated description of biopsy devices, followed by a similarly anatomic-based discussion of image-guided procedures utilizing direct access, including biopsy, drainages, stenting, ablation, injections, and augmentation. All chapters have been updated, and more than a third of the authors are newly appointed and offer specific expertise in the subject matter. Image-Guided Interventions is designed to be read either primarily from cover to cover, or selectively in preparation for a specific procedure. We have maintained an organizational structure that highlights indications and contraindications for procedures at the outset. Materials required for procedures are clearly listed for easy reference, as are “Key Points” that sum up each chapter. High-quality illustrations and drawings are in abundance because they are typically worth the proverbial thousand words. We have received tremendous positive feedback from seasoned interventionalists and trainees alike for the first edition of Image-Guided Interventions. The editors and Elsevier have strived in earnest to produce a sequel worthy of its namesake. Like its predecessor, Image-Guided Interventions, Second Edition, will maintain its relevance through constant Internet updating. It is a textbook created and designed not to gather dust on a bookshelf, but to be present in offices and procedural areas alike, well worn from constant use through reading and referral. MATTHEW A. MAURO, MD, FACR KIERAN P.J. MURPHY, MB, FRCPC, FSIR KENNETH R. THOMSON, MD, FRANZCR ANTHONY C. VENBRUX, MD ROBERT A. MORGAN, MBChB, MRCP, FRCR, EBIR

xxxi

PART

1

Vascular Interventions SECTION ONE

HISTORY OF ANGIOGRAPHY AND INTERVENTION CHAPTER

1

A Brief History of Image-Guided Therapy

Anthony C. Venbrux, Jozef M. Brozyna, Suma Chandra, Hank K. Chen, Gina D. Tran, and Dmitri A. Gagarin

HISTORICAL HIGHLIGHTS OF ENDOVASCULAR THERAPY

Arterial Endovascular Therapy: Revascularization and Vessel Reconstruction

• British dentist Charles Stent develops a plastic material for taking mouth impressions (i.e., creates a “scaffold”) (1856). • Image guidance was made possible by the discovery of x-rays by Wilhelm Conrad Röntgen, November 8, 1895 (Figs. 1-1 to 1-3). • Sven Ivar Seldinger develops percutaneous vascular catheterization (1952) (Figs. 1-4 and 1-5). • Percutaneous revascularization is accomplished by Charles Dotter with coaxial dilators (1964) (Figs. 1-6 and 1-7). • Transcatheter embolotherapy is performed by Charles Dotter to control acute upper gastrointestinal bleeding (1970). • Lazar Greenfield pioneers caval interruption with a vena cava filter (1973) (Fig. 1-8). • Andreas Gruentzig performs percutaneous angioplasty (1977) (Figs. 1-9 and 1-10). • Chemoembolization was performed by Cato (1981). • Julio Palmaz develops the endovascular balloonexpandable stent (1985). • Juan C. Parodi, Julio C. Palmaz, and H. D. Barone develop stent-grafts (1990).

An English dentist, Dr. Charles Stent, developed a thermoplastic material for taking impressions of toothless mouths in 1856. Thus the “stent” may be considered a “scaffold.” For purposes of this discussion, a stent is used for reconstruction in the vascular (or nonvascular) systems. Percutaneous vascular catheterization as a viable endovascular technique was described in June 1952, when Sven Ivar Seldinger presented his idea of replacing an arteriography needle with a catheter. Translumbar aortography and direct carotid puncture were effectively relegated to historical descriptions in textbooks. The technique of percutaneous revascularization advanced rapidly in 1964, when Charles Dotter used coaxial “pencil-point” dilators to treat a superficial femoral artery stenosis. The era of image-guided revascularization of the lower extremities had begun. Although Dotter was successful in dilating femoral arterial stenoses, the use of coaxial “pencil-point” dilators (Van Andel Catheters [Cook Inc., Bloomington, Ind.]) required progressive enlargement of the percutaneous puncture site. Development of the angioplasty balloon by Andreas Gruentzig in 1977 resulted in another major step forward that allowed the percutaneous arterial entry site to be kept to a minimum (see Figs. 1-9 and 1-10). Early balloon designs were hampered by uneven balloon dilation and frequent rupture. Such events often led to pseudoaneurysm formation or dissections and resulted in vessel thrombosis. The arterial metallic stent was invented in the 1980s by Julio Palmaz at the University of Texas Health Science Center in San Antonio. Dr. Palmaz described his stent in 1985 and continued work on his device in 1986. Later developments included the use of a stent design consisting of a single stainless steel tube with parallel staggered slots in the wall. When the stainless steel tube was expanded, the slots formed diamond-shaped spaces that resisted arterial compression. This design became the first stent approved by the U.S. Food and Drug Administration (FDA) for vascular use. Stanley Baum and Moreye Nusbaum pioneered catheter embolization in the mid-1960s for the purpose of treating

ENDOVASCULAR MILESTONES November 18, 1895, was a day of historical significance. In a physics laboratory in the southern part of Germany, Conrad Wilhelm von Roentgen accidentally discovered x-rays. The mysterious rays illuminated medical science, and radiology was born. From this serendipitous discovery, image-guided minimally invasive procedures evolved to provide patients with therapeutic options in the management of vascular and nonvascular diseases. This chapter is neither comprehensive nor complete. Of the many potential technologic advances, arterial and venous endovascular therapy is the focus in this section. Later in the chapter, the history of nonvascular interventions is covered.

1

2

PART 1 VASCULAR INTERVENTIONS · SECTION 1 HISTORY OF ANGIOGRAPHY AND INTERVENTION

FIGURE 1-1. Photograph of Roentgen taken in 1906 while he was director of the Institute of Physics at the University of Munich. (From Eisenberg RL. Radiology: an illustrated history. St Louis: Mosby–Year Book; 1992, p. 38.)

FIGURE 1-2. First roentgen photograph of Mrs. Roentgen’s hand. (From Glasser O. Wilhelm Conrad Röntgen and the early history of the roentgen rays. Springfield, Ill.: Charles C Thomas; 1933.)

FIGURE 1-3. Roentgen’s first communication. Left, First page of handwritten manuscript (1895). Middle, First page of published article on new type of ray. Right, Front cover of a reprint of initial paper. (From Eisenberg RL. Radiology: an illustrated history. St Louis: Mosby–Year Book; 1992, p. 25.)

CHAPTER 1 A BRIEF HISTORY OF IMAGE-GUIDED THERAPY acute gastrointestinal bleeding. In 1970, Charles Dotter reported utilizing an autologous clot as the embolic agent to control acute upper gastrointestinal bleeding by selective embolization of the right gastroepiploic artery in a patient who was a poor surgical candidate. Robert White used this technique in 1974 to control bleeding duodenal ulcers when the hemorrhage was unresponsive to intraarterial injections of vasopressin. Chemoembolization was largely pioneered by the Japanese urologist Cato in 1981. Cato used particles about 200 μm in size to demonstrate that chemoembolization with microcapsules containing chemotherapeutic agents was superior to local intraarterial injection of antitumor agents.1 Stent-grafts, pioneered by Juan Parodi and Charles Dotter, became the major impetus for future endovascular reconstruction procedures used to exclude an aneurysm, close an arteriovenous fistula, and reconstruct the central lumen of a dissected vessel. Specifics of the history of endovascular grafts is worth mentioning. The technology has fundamentally changed the management of diseases of the abdominal and thoracic aorta. The first abdominal aortic aneurysm (AAA) was described by Andreas Vesalius in the 16th century. By the 1800s, aortic ligation had become the surgical procedure of

FIGURE 1-4. Sven Ivar Seldinger. (From Eisenberg RL. Radiology: an illustrated history. St Louis: Mosby–Year Book; 1992, p. 442.)

A

1

3

2

4

5

6

B FIGURE 1-5. Seldinger technique (1953). A, Equipment. Stiletto is removed and leader inserted through needle and catheter. B, Diagram of technique used: (1) artery punctured and needle pushed upward, (2) leader inserted, (3) needle withdrawn and artery compressed, (4) catheter threaded onto leader, (5) catheter inserted into artery, (6) leader withdrawn. (From Seldinger SI. Catheter replacement of the needle in the percutaneous arteriography: a new technique. Acta Radiol 1953;39:368–76.)

3

4

PART 1 VASCULAR INTERVENTIONS · SECTION 1 HISTORY OF ANGIOGRAPHY AND INTERVENTION

FIGURE 1-6. Dotter coaxial Teflon catheter system consisting of 12F catheter with tapered and beveled tip over inner 8F catheter with 0.044-inch guidewire. (From Waltman AC, Greenfield AJ, Athanasoulis CA. Transluminal angioplasty: general rules and basic considerations. In Athanasoulis CA, Greene RE, Pfister RC, Roberson GH, editors. Interventional radiology. Philadelphia: WB Saunders; 1982.)

FIGURE 1-7. First percutaneous transluminal angioplasty (1964). A, Control arteriogram showing segmental narrowing, with threadlike lumen of left superficial femoral artery in region of adductor hiatus. B, Study immediately after dilation with catheter having an outer diameter of 3.2 mm. C, Three weeks after transluminal dilation, lumen remains open. Clinical and plethysmographic studies indicated continuing patency more than 6 months later. (From Dotter CT, Judkins MP. Transluminal treatment of arteriosclerotic obstruction. Circulation 1964;30:654–70.)

choice when attempting to treat iliac and abdominal aortic aneurysms. The longest survivor was a patient of Keen in 1899, who survived 48 days after aortic ligation at the diaphragm for a ruptured AAA. This was deemed a great success because previous aortic ligations for iliac aneurysms performed by Astley Cooper in 1817 and J.H. James in 1829 resulted in patient death within 48 hours. As the

FIGURE 1-8. Kimray-Greenfield filter. (From Dedrick CG, Novelline RA. Transvenous interruption of the inferior vena cava. In Athanasoulis CA, Greene RE, Pfister RC, Roberson GH, editors. Interventional radiology. Philadelphia: WB Saunders; 1982.)

years passed, many others attempted aortic ligations, and most were met with similar results; the most common mechanism of failure was ligature erosion into the aorta and massive hemorrhage. In April of 1923, over 100 years after Cooper’s first attempts, Rudolph Matas performed the first successful aortic ligation on a patient with a syphilitic aortic and bilateral common iliac aneurysms, using cotton tape. The patient survived for 17 months before succumbing to tuberculosis. The subsequent evolution of aortic aneurysm repair involved many varied techniques. Physicians experimented with different variations of aortic ligation, wires to induce thrombosis, and most notably, reactive polyethylene cellophane wrapping of aneurysms. This particular technique was performed on Albert Einstein in 1949; however, he eventually died of an AAA rupture 6 years later. In 1951, the first homograft was used for AAA repair by Charles DuBost, only to be superseded by nylon and then polytetrafluoroethylene (PTFE) grafts shortly thereafter. Javid and Creech first introduced surgical endoaneurysmorrhaphy in 1961, greatly reducing the high mortality rates associated with aneurysm excision and graft repair. With the advent of minimally invasive techniques for the treatment of AAAs, the evolution of aneurysm repair changed dramatically. In the late 1970s, Juan Parodi had begun thinking about endovascular aneurysm repair (EVAR) during his vascular surgery fellowship at the Cleveland Clinic. After creating many rudimentary prototypes of varying designs, he became convinced that AAA repair was possible without laparotomy. However, he continuously ran into two problems: (1) the mechanics of delivering a minimally invasive system into the precise location needed,

CHAPTER 1 A BRIEF HISTORY OF IMAGE-GUIDED THERAPY

A

B

C FIGURE 1-9. Percutaneous transluminal coronary angioplasty (1979). A, Stenosis of coronary artery. B, Double-lumen balloon catheter is introduced with guiding catheter positioned at orifice of left or right coronary artery. At tip of dilating catheter is a short soft wire to guide catheter through vessel. Proximal to wire is a side hole connected to main lumen of dilating catheter. This lumen is used for pressure recording and injection of contrast material. Dilating catheter is advanced through coronary artery with balloon deflated. C, Balloon is inflated across stenosis to its predetermined maximal outer diameter, thereby enlarging lumen. After balloon deflation, catheter is withdrawn. (From Grüntzig AR, Senning A, Siegenthaler WE. Non-operative dilation of coronary artery stenosis. Percutaneous transluminal coronary angioplasty. N Engl J Med 1979;301:61–8. Copyright 1979 Massachusetts Medical Society. All rights reserved.)

FIGURE 1-10. Transluminal dilation of coronary artery stenosis (1978). Left, Initial angiogram in 43-year-old man with severe angina pectoris reveals severe stenosis of main left coronary artery. Middle, After passage of dilation catheter, distensible balloon segment was inflated twice to a maximum outer diameter of 3.7 mm. Right, Postprocedure angiogram showing good result without complications. (From Grüntzig AR. Transluminal dilation of coronary artery stenosis. Lancet 1978;1:263.)

and (2) the absence of a fastening device that would take the place of surgical sutures in securing the graft in place and creating a seal between the intravascular graft and vessel wall. The solution to these challenges came when Parodi met Julio Palmaz at the Transcatheter Cardiovascular Therapeutics meeting in Washington, DC, in 1988. Palmaz was attending to discuss a new stent design. Parodi was informed that Palmaz was presenting and thus was encouraged to attend. It became apparent to Parodi that the Palmaz design was precisely what he was looking for: a device with a high radial force (i.e., a balloon-expandable stent) that was deployed using endovascular techniques and would provide placement, precision, a secure anchor, and an aortic wall seal that his EVAR concept required. Palmaz was initially apprehensive about joining the project because of his many

professional commitments and perceived limitations of his device in the aorta. However, he later admitted that he envisioned his stent playing a role in aneurysm repair as early as 1986. Ultimately he decided to join efforts with Parodi. Work then began on the first EVAR in both Buenos Aires, where Parodi worked at the Instituto Cardiovascular de Buenos Aires (ICBA), and San Antonio, where Palmaz was Chief of Special Procedures at the University of Texas Health Science Center. During their efforts to perfect the device and the technique, Parodi was searching for an ideal candidate for the first EVAR. On September 7, 1990, two patients were to undergo the new procedure, to be performed by both Parodi and Palmaz at the cardiac catheterization laboratory of ICBA. During the first procedure, the proximal portion of the stent was deployed just inferior to the renal arteries, and the distal

5

6

PART 1 VASCULAR INTERVENTIONS · SECTION 1 HISTORY OF ANGIOGRAPHY AND INTERVENTION portion placed just superior to the aortic bifurcation. The postprocedure angiogram displayed a completely excluded aneurysm, concluding the first EVAR procedure. The second patient was converted to traditional surgical AAA repair, because the proximal portion of the stent was deployed too low in the aorta, and the distal end of the graft ended in one common iliac artery. It was not until later that Parodi realized that the reflection of pressure waves from the aortic bifurcation and iliac arteries would cause an endoleak and failure of a single stent device. This initial experience provided Parodi and Palmaz with clinical insight into patient benefits such as improvements in recovery time and quality of life, using an endovascular approach. The first patient quickly recovered and was discharged without complications, while the second remained intubated in the intensive care unit. Additionally, the second case provided the team with a possible mechanism of failure for their new procedure and delivery system. Over time, the procedure, delivery system, and methods of Parodi’s team evolved. By December of 1992, 24 patients had undergone EVAR by one of three techniques that had evolved since the first case, the most common being an aorto-uni-iliac approach with contralateral common iliac embolization and femoral-femoral bypass. Prior to this, an interventional radiologist from Buenos Aires, Claudio Schonholz, had also joined the international team and played a significant role in the first EVAR performed in the United States. In August of 1992, the vascular surgery department at Montefiore Medical Center in Bronx, New York, was consulted on a 76-year-old male patient with a 7.5-cm infrarenal AAA with several comorbidities that suggested he was not a candidate for open surgical repair. Frank Veith and Michael Marin quickly contacted Parodi and began discussing the logistics involved in traveling to South America, learning the procedure, and treating the aneurysm endovascularly. After several meetings, and convincing Johnson & Johnson (who was producing the Palmaz stent) executives to allow them to use a large Palmaz stent for the procedure (no FDA Investigational Device Exemption existed for it yet), they were ready to proceed. On November 23, 1992, Parodi, Schonholz, Veith, Marin, and Jacob Cynamon performed the first EVAR in the United States, using a 22-mm Dacron prosthesis sewn over a large Palmaz-like stent. The patient was discharged several days later and remained symptom free until his death 9 months later. Death was due to problems unrelated to the EVAR procedure. Although Parodi is often credited with the first EVAR, it is important to mention that a Ukrainian surgeon, Nicholas Volodos, published an article in 1986 describing the endovascular repair of a traumatic thoracic aortic aneurysm with a homemade stent-graft. Also of note, Harrison Lazarus was awarded a U.S. patent in 1988 for an endovascular stent-graft that eventually served as the basis for the Guidant (Indianapolis, Ind.) Ancure device. With advancements in technology and refinement of techniques, EVAR has become for selected patients the standard of care for treatment of AAAs. By the end of 2010, it was estimated that nearly three quarters of all abdominal aortic aneurysms in the United States were repaired endovascularly. There are an estimated 1.1 million Americans between the ages of 50 and 84 diagnosed with an AAA, and over 15,000 deaths annually due to abdominal aortic

aneurysms, so it is no surprise that this minimally invasive treatment option continues to evolve.

Venous Endovascular Therapy: Caval Interruption Treatment of venous thromboembolic disease has also rapidly evolved. Deep venous thrombosis is one of the major causes of morbidity and mortality worldwide. “Caval interruption” with a filter was initially performed via surgical cutdown by Mobbin and Uddin. In 1973, Lazar Greenfield introduced a cone-shaped surgically placed vena cava filter. Greenfield pioneered his work with the Kimray Corporation. This device was first deployed percutaneously in 1984. The diameter of the sheath for the filter was 29F (outer diameter [OD]). Later developments reduced Greenfield’s percutaneous puncture size from 29F (OD) to 16F (OD). Because all permanent vena cava filters are associated with complications, including caval thrombosis (3%-40%), recent advances have provided patients at risk for venous thromboembolic disease with “optional devices,” such as vena cava filters that may be permanent or removed, the latter when the patient’s “window of vulnerability” has clinically passed. Removable devices were historically called temporary filters and were classified as either tethered or retrievable. A more accurate term in current clinical use is an optional filter. Optional implies that the device may either be a permanent implant or be used for a short interval. A tethered inferior vena cava (IVC) filter consists of a filter attached to a central venous catheter. A retrievable IVC filter is a device deployed in the IVC that attaches to the IVC wall with hooks but has no external tether. (The FDA approved use of the first tethered temporary vena cava filter in the United States in the early 1990s.) This tethered device (Tempofilter [B. Braun, Evanston, Ill.]), placed in a young trauma patient, remained in place for 13 days and trapped a large embolus, thereby preventing a potentially life-threatening pulmonary embolic event. The FDA approved optional vena cava filters in the United States in 2003. The Recovery Filter (Bard Peripheral Vascular, Tempe, Ariz.), a permanent implant with the “option to remove” at any future date, was first deployed in the United States in July 2003. Since 2003, the Günther Tulip (Cook Medical, Bloomington, Ind.), the Opt Ease (Cordis Endovascular, Warren, N.J.), and other filters have also received FDA approval as optional vena cava filters. Thus, management of venous thromboembolic disease has further evolved with newer therapeutic options for patients who are at risk for pulmonary emboli but have a contraindication to or have had a complication of anticoagulant therapy.

SUMMARY Arterial and venous endovascular procedures continue to progress rapidly. Future developments that might take place include continued use of mechanical devices in the vascular system, combination therapies to rapidly lyse and remove residual thrombi, further design modifications in endovascular devices to reduce the risk of metal fatigue/fracture, and the application of newer technologies to reduce intimal hyperplasia (e.g., drug-eluting stents). Such improvements will, one hopes, provide solutions for some of the limitations of our current technology.

CHAPTER 1 A BRIEF HISTORY OF IMAGE-GUIDED THERAPY KEY POINTS •

Percutaneous vascular access is a technique proposed and used by Sven Ivar Seldinger.

•

Several pioneers who made contributions to the endovascular procedures used today include Charles Dotter, Andreas Gruentzig, Juan Parodi, and Julio Palmaz.

•

The word stent (named after Charles Stent) originated with the development of dental technology.

•

Advances in the use of newer vena cava filters include “optional” vena cava filters.

• Percutaneous image-guided drainage is extensively used and described by Eric Vansonnenberg, Peter R. Mueller, and Joseph T. Ferucci (1980s).

Urologic Interventions In 1954, Wickbom opacified the renal pelvis directly by injecting contrast medium through a long needle. Thus, the antegrade pyeloureterogram (antegrade nephrostogram) as we know it was first performed. The following year, Goodwin et al. used a “catheter through a needle” for drainage. In

HISTORICAL HIGHLIGHTS OF NONVASCULAR IMAGE-GUIDED THERAPY • Wickbom, Weens, and Florence directly opacified the urinary tract with a needle (1954) (Fig. 1-11). • Percutaneous urinary tract drainage is accomplished by Goodwin and colleagues (1955) (Figs. 1-12 and 1-13). • Percutaneous biopsy of opacified lymph nodes is described by Sidney Wallace (1961). • Percutaneous transhepatic cholangiography is described by Evans et al. (1962) (Fig. 1-14). • Drainage of an abdominal viscus (e.g., gallbladder, stomach) with a retrievable anchoring device is described by Constantin Cope (1986).

FIGURE 1-12. Percutaneous trocar nephrostomy (1955): method and landmarks. Optimum puncture site is usually about five fingerbreadths lateral to midline and at a level where a 13th rib would be. (From Wickbom I. Pyelography after direct puncture of the renal pelvis. Acta Radiol 1954;41:505–12.)

FIGURE 1-11. Antegrade pyelography (1954). After direct puncture of the left renal pelvis, contrast material demonstrates dilation of the pelvis and upper part of the ureter. There is complete obstruction of the ureter at the pelvic inlet (arrow).

FIGURE 1-13. Cross-sectional anatomy of left renal area (after Brodel). About 2 inches of tubing is allowed to coil in hydronephrotic pelvis. (From Wickbom I. Pyelography after direct puncture of the renal pelvis. Acta Radiol 1954;41:505–12.)

7

8

PART 1 VASCULAR INTERVENTIONS · SECTION 1 HISTORY OF ANGIOGRAPHY AND INTERVENTION

FIGURE 1-14. Operative cholangiography (1931). Left, Radiograph demonstrating biliary tract, with two sharply defined clear spaces in retroduodenal portion and at level of papilla of Vater. Right, Corresponding diagram with extracted calculi superimposed. Superior calculus is size of a hazelnut and inferior one size of a chickpea. (From Mirizzi PL, Losada CQ. Exploration of the bile ducts during an operation. Paper presented at the Third Argentine Congress of Surgery, 1931, p. 694–703.)

1965, Bartley used a guidewire technique for drainage of the urinary tract, and in 1976, Frenstrom and Johansson reported dilation of the nephrostomy tract for stone removal.1 By 1978, Stables reviewed the techniques used in the performance of 516 nephrostomies appearing in the medical literature. This series included 53 of his own patients. Thus, percutaneous techniques had received increasing attention in the published medical literature and were found to be a viable option for open surgical procedures. Smith, in 1979, coined the term endourology. At that time he had begun work at the University of Minnesota, where the scientific/academic environment was conducive to new techniques. Smith and Amplatz together developed numerous endourologic techniques, including the Amplatz retention catheter in 1986. In 1982, Castaneda-Zuniga and Amplatz published the technique for urinary stone removal. Once the percutaneous tract was dilated, they used fluoroscopic guidance to extract stones with a Randall forceps or a Dormia basket at the tip of the catheter. Coleman, in 1985, reported that the results of percutaneous removal of stones had improved since the early 1980s to a success rate of 99% in a study of 450 patients. Thus, the technique of percutaneous drainage of the urinary tract with subsequent percutaneous removal of obstructing stones made significant advancements during the 1970s and 1980s. Today, percutaneous urinary tract interventions are an integral part of image-guided therapy.

Biliary Interventions Percutaneous transhepatic cholangiography and percutaneous biliary drainage are based on needle/guidewire/catheter techniques developed simultaneously with endourology.

Nonsurgical management of biliary stones and strictures continues to be an essential part of the treatment of patients with biliary disease. In 1973, Burhenne described a technique for extraction of retained common duct stones through a T-tube tract. The feasibility of percutaneous biliary duct dilation with balloon dilation catheters was reported by Berhenne in 1975.2 Berhenne performed the technique through a mature T-tube tract. Molnar and Stockum described the dilation of biliary strictures via a transhepatic route in 1978.3 In 1980, Constantin Cope developed a simple drainage catheter to reduce the problem of nephrostomy tube dislodgement. Foley or Malecot nephrostomy drainage catheters were frequently pulled out or became dislodged. The Cope loop catheter was one of the most significant developments of nonvascular interventional procedures in the 1980s. A suture coursing through the lumen of the catheter allowed a locking mechanism, which today is the standard for most drainage catheters. Such catheters are used in the urinary tract and biliary system and for drainage of percutaneous abscesses. Percutaneous application of biliary endoprostheses (plastic or metallic) to palliate patients with malignant biliary obstruction was described by Dotter, Gianturco, and Ring, among others.

Lymph Node Applications As early as 1933, Hudack and McMaster used blue dye as a contrast agent to visualize the lymphatic system. The surgeon Servele dissected lymphatics in 1994 and inserted a needle, followed by the injection of Thorotrast. Percutaneous biopsy of opacified lymph nodes (opacified with contrast material) was described by Sidney Wallace in 1961. The diagnostic and therapeutic potential of

CHAPTER 1 A BRIEF HISTORY OF IMAGE-GUIDED THERAPY lymphangiography brought oncologic interventions to the forefront.

Other Nonvascular Interventions The ability to nonsurgically access an abdominal viscus (percutaneous gastrostomy, etc.) was improved largely through the innovative efforts of Constantin Cope. Cope developed a suture-anchoring device. The retrievable anchoring device was reported by Cope in 1986 after he successfully drained the gallbladder and stomach without leakage. The technique of percutaneous abscess drainage was applied and described extensively by E. Vansonnenberg, P.R. Mueller, and J.T. Ferucci. Drainage of a percutaneous abscess or fluid collection is an integral part of image-guided nonvascular patient management. KEY POINTS •

Nonvascular percutaneous interventions are an integral part of image-guided therapy.

•

Significant advancements have allowed nonsurgical management of multiple medical conditions that previously required operative (open) procedures.

•

Important percutaneous techniques in nonvascular interventions include those in the urinary tract, biliary system, lymphatic system, gastrointestinal tract, and drainage of abdominal fluid collections.

ACKNOWLEDGMENT The authors would like to thank Shundra Dinkins, Toni Acfalle, and Dana Murphy for their expertise in preparation of this manuscript. ▶ SUGGESTED READINGS Criado F. EVAR at 20: the unfolding of a revolutionary new technique that changed everything. J Endovasc Ther 2010;17:789–96. Dotter CT, Frische LH, Judkins MP, Mueller R. Nonsurgical treatment of iliofemoral arteriosclerotic obstruction. Radiology 1966;86:871–5. Dotter CT, Rosch J, Anderson JM, et al. Transluminal iliac artery dilation. JAMA 1974;230:117–24.

Friedman S. A history of vascular surgery. 2nd ed. Malden, MA: Blackwell Futura; 2005. Gruentzig A, Senning A, Siegenthaler WE. Nonoperative dilation of coronary-artery stenosis. N Engl J Med 1979;301:61–8. Grüntzig A, Hopf H. Perkutane recanalization chronisher arteriller verschüsse mit einen neuen dilatationskatheter: modification der Dottertechnik. Dtsch Med Wochenschr 1974;99:1502–51. Hedin M. The origin of the word stent. Acta Radiol 1997;38:937–9. Hurst JW. The first coronary angioplasty as described by Andreas Gruentzig. Am J Cardiol 1986;57:185–6. Kent KC, Zwolak RM, Egorova NN, et al. Analysis of risk factors for abdominal aortic aneurysm in a cohort of more than 3 million individuals. J Vasc Surg 2010;52:539–48. Lubarsky M, Ray CE, Funaki B. Embolization agents–which one should be used when? Part 1: large-vessel embolization. Semin Intervent Radiol 2009;26:352–7. Margulis AR. Interventional diagnostic radiology; a new subspecialty [editorial]. AJR Am J Roentgenol 1967;99:761–2. Palmaz JC, Parodi JC, Barone HD. Transluminal bypass of experimental abdominal aortic aneurysm. J Radiol 1990;11:177–202. Palmaz JC, Sibbitt RR, Reuter SR, et al. Expandable intraluminal graft: a preliminary study. Work in progress. Radiology 1985;156:72–7. Palmaz JC, Sibbitt RR, Tio FO, et al. Expandable intraluminal graft: a feasibility study. Surgery 1986;98:199–205. Parodi JC, Ferreira M. Why endovascular abdominal aortic aneurysm repair? Semin Intervent Cardiol 2000;5:3–6. Pearce W, Rowe VL. Abdominal aortic aneurysm. MedScape. Accessed September 4, 2011. Ring ME. How a dentist’s name became a synonym for a lifesaving device: the story of Dr. Charles Stent. J Hist Dent 2001;49:77–80. Rosch J, Keller FS, Kaufman JA. The birth, early years, and future of interventional radiology. J Vasc Interv Radiol 2003;14:841–53. Seldinger SI. Catheter replacement of needle in percutaneous arteriography: a new technique. Acta Radiol 1953;39:368–76. Veith FJ, Marin MJ, Cynamon J, et al. 1992: Parodi, Montefiore, and the first abdominal aortic aneurysm stent graft in the United States. Ann Vasc Surg 2005;19(5):749–51. Volodos NL, Shekhanin VE, Karpovich IP, et al. A self-fixing synthetic blood vessel endoprosthesis [in Russian]. Vestn Khir IM I I Grek 1986;137(11):123–5. White RI, Giargian FA, Bell William. Bleeding duodenal ulcer control: selective arterial embolization with autologous blood clot. JAMA 1974;299:546–8.

The complete reference list is available onine at www.expertconsult.com.

9

CHAPTER 1 A BRIEF HISTORY OF IMAGE-GUIDED THERAPY 9.e1 REFERENCES 1. Geddes LA, Geddes LE. The catheter introducers. Chicago: Mobius Press; 1993. p. 83–4, 106. 2. Berhenne HJ. Dilatation of biliary tract strictures: a new roentgenologic technique. Radiol Clin 1975;44:153–9.

3. Molnar W, Stockum AE. Transhepatic dilatation of choledochoenterostomy strictures. Radiology 1978;129:59–64.

SECTION TWO

VASCULAR DIAGNOSIS CHAPTER

2

Noninvasive Vascular Diagnosis Brian B. Ghoshhajra and Sanjeeva Prasad Kalva

Most vascular diseases can be diagnosed with a combination of clinical history and examination, relevant laboratory tests, and appropriate noninvasive tests. Such tests include the ankle-brachial index, segmental pressure and pulse volume recording measurements, Doppler assessment of blood flow, ultrasound evaluation of vessels, computed tomographic angiography (CTA), and magnetic resonance angiography (MRA). Computed tomography (CT) and magnetic resonance imaging (MRI) are now the first-line tests or replacements for diagnostic catheter angiography in many cases. The imaging test should be chosen based on the clinical problem. Other factors that may affect the choice of a particular test include cost, availability, urgency, pretest probability, and underlying contraindications. As with any radiologic tests, radiation exposure should be kept as low as reasonably achievable. In this chapter we briefly discuss the principles of ultrasonography, CTA, and MRA and their clinical applications. Ultrasonography is highly sensitive and specific for diagnosis of venous thrombosis in the extremities and superficial vessels. It remains the test of choice for evaluation of venous insufficiency, and is often useful in assessing hemodynamic significance of an arterial stenosis (sometimes in conjunction with pulse-volume recordings for peripheral arterial disease). Ultrasound is widely available, relatively inexpensive, and rapid, but is highly operator dependent and subject to technical challenges such as acoustic window limitations, especially in the obese patient. In peripheral vessels, ultrasound performs well and can obviate the need for further or invasive testing. In the abdominal vasculature, ultrasound can be reliably used when image quality allows. CTA and MRA are comparable for evaluation of thoracoabdominal vessels and peripheral arteries. CTA is preferred in the acute setting, whereas MRA enjoys unique advantages when assessing inflammatory diseases affecting the large vessels. Both modalities are routinely useful in the neurovascular system, particularly in the setting of ischemic stroke workup. MRA is preferable for evaluation of the vessel lumen in patients with extensive mural calcification. It is also

10

operator dependent, often requiring expert physician oversight, and subject to numerous artifacts. Acquisition times can be long, particularly when imaging small vessels or using electrocardiogram (ECG) gating and respiratory gating or breath-hold techniques (especially in the cardiothoracic anatomy, where this is frequently necessary). However, MRA is less often limited by body habitus and generally allows unrestricted imaging planes. Recent advances in noncontrast MRA allow the technique to benefit those patients with contraindications to intravenous contrast. CTA has also progressed dramatically in recent years as helical CT has moved to multidetector CT, and now widearea-detector CT and dual-source CT. Recent advances now allow rapid and reliable angiographic imaging in nearly every circulatory bed. CT is less operator dependent than MRA but does require expert care. Although most acquisitions are axial, postprocessing of volumetric datasets allows near-unrestricted imaging planes. CT with cardiac gating is now widely available, allowing routine noninvasive ECGsynchronized imaging of the heart and great vessels. Coronary CTA is useful in assessing coronary artery anomalies, bypass grafts, and coronary artery disease. Although the capabilities of modern CTA are impressive, downsides include the need for rapid iodinated intravenous contrast administration and ionizing radiation. Numerous advances have decreased the radiation dose associated with CTA, particularly in the thorax. Limitations of the technique include poor assessment of small and peripherally calcified vessels, in which CTA can overestimate stenosis. Noninvasive imaging now offers a large and diverse tool kit to the cardiovascular imager. Although ultrasound remains the first-line test in many vascular beds, MRI and CTA now offer reliable high-quality imaging of the vasculature in many settings. Choice of the appropriate test varies with the available technology, expertise, target vessel, and clinical question at hand. The full discussion of noninvasive vascular diagnosis, as well as 10 illustrations, are available at www. expertconsult.com.

SECTION TWO

VASCULAR DIAGNOSIS CHAPTER

2

Noninvasive Vascular Diagnosis Brian B. Ghoshhajra and Sanjeeva Prasad Kalva

Most vascular diseases can be diagnosed with a combination of clinical history and examination, relevant laboratory tests, and appropriate noninvasive tests. Such tests include the ankle-brachial index, segmental limb pressure and pulse volume recording measurements, Doppler assessment of blood flow, ultrasound evaluation of vessels, computed tomographic angiography (CTA), and magnetic resonance angiography (MRA). Computed tomography (CT) and magnetic resonance imaging (MRI) are now the first-line tests or replacements for diagnostic catheter angiography in many cases. The imaging test should be chosen based on the clinical problem. Other factors that may affect the choice of a particular test include cost, availability, urgency, pretest probability, and underlying contraindications. As with any radiologic tests, radiation exposure should be kept as low as reasonably achievable. In this chapter we briefly discuss the principles of ultrasonography, CTA, and MRA and their clinical applications.

beam) depending on whether the motion is away or toward the transducer. Such change in the frequency is referred to as Doppler shift, which is governed by the following equation:

ULTRASONOGRAPHY

When the angle of interrogation between the incident beam and direction of blood flow approaches 90 degrees, the cos θ approaches 0 and no Doppler shift is detected. In general, the angle of interrogation is kept between 30 and 60 degrees with the vessel lumen to receive reliable velocity measurements. The Doppler shift is measured either with continuous or pulsed-wave Doppler. In continuous-wave Doppler, a transducer is placed over the area of interest and the reflected echoes are detected by another transducer. The ultrasound beam is emitted and received continuously. The magnitude of Doppler shift in the path of the beam is detected without spatial localization. Pulsed-wave Doppler imaging uses a single transducer that acts as both transmitter and receiver. Multiple short pulses of ultrasound beam are emitted, and the transducer turns to receiving mode. Generally, pulsed-wave Doppler is combined with grayscale ultrasound imaging (duplex ultrasonography), and thus the location of the ultrasound interrogation with the blood vessels can be identified and changed as required. With pulsed Doppler imaging, the maximum Doppler shift that can be measured depends on the pulse repetition frequency and the frequency of the transducer. When set conditions do not allow detection of the Doppler shift frequency, aliasing occurs, which is seen as wraparound Doppler spectrum on spectral wave form display or as complex colors on color flow imaging. Aliasing can be reduced by increasing the pulse repetition frequency or by using a low-frequency transducer. The Doppler shift can be displayed in three different ways: color flow imaging, spectral waveforms, and audible sound. The mean velocity is color coded depending on

Real-time grayscale ultrasonography is based on the principle of reflection of sound waves. The ultrasound transducer emits short pulses of high-frequency (1-10 MHz) ultrasound waves into the body and receives reflected echoes from the tissues. The computer analyzes the data and displays a grayscale image based on the intensity and timing of the reflected echoes. The reflection of sound waves depends on the differences in acoustic impedance of various tissues that form the interface. Large, smooth interfaces produce a high degree of reflection, which also depends on the orientation of the interface with respect to the angle of the incident ultrasound beam. In addition, small reflections, referred to as scatter, arise from multiple interfaces; these contribute to the diagnostic detail in the ultrasound images of solid organs. The scatter is not affected by the angle of the incident beam. Rayleigh scatter refers to diffuse scatter that occurs when the ultrasound beam interacts with tissues that are smaller than the wavelength of the ultrasound beam. This occurs commonly in the blood vessels, where red blood cells measure less than the wavelength of the ultrasound beam. Such scatter depends on many factors, including the number of scatterers (the red cells), size of the scatterers, and the ultrasound frequency. This scatter is responsible for Doppler evaluation of blood flow. The Doppler effect refers to change in the perceived frequency when the source and/or the detector are moving. The red cells (the reflectors or the source) moving in the blood vessels change the frequency of reflected echoes (compared with the incident frequency of the ultrasound

fD = fR − fl = (2fl ⋅ V ⋅ cos θ)/C where fD = Doppler shift or Doppler frequency; fR = frequency of the reflected beam; fl is frequency of incident beam, V = velocity of the blood flow, θ = the angle between the direction of flow and the axis of the ultrasound beam, and C = velocity of the sound in tissues. The computer detects the frequency of the reflected beam, and we know the frequency of incident beam and the velocity of sound in the tissues (1540 m/s). We can rewrite the equation to learn the velocity of the blood flow: V = (fR − fl )C/(2fl ⋅ cos θ)

e1

e2

PART 1 VASCULAR INTERVENTIONS · SECTION 2 VASCULAR DIAGNOSIS whether the flow is toward or away from the transducer. The color flow maps are displayed over the grayscale image to provide better identification of the vessels. However, color flow imaging is less useful for quantification of blood flow and understanding the phasic nature of the blood flow. Spectral waveform display allows depiction of the direction of blood flow, phasic nature of blood flow, change in velocity over time (systolic, diastolic), and spectrum of velocities within the volume of interrogation. Audible sound is less useful in quantification and understanding the nature of the flow. Power Doppler flow imaging differs from color Doppler imaging in that the intensity (or amplitude) of the Doppler shift is displayed in color, rather than the Doppler shift per se. It does not display the direction of blood flow. It is useful to detect “any flow” in the region of interest, and it is more sensitive in detecting blood flow and less dependent on the angle of interrogation.

CLINICAL APPLICATIONS Lower Extremities Veins Doppler evaluation of the venous system in lower extremities is usually performed to detect deep or superficial venous thrombosis (compression ultrasonography) or valvular insufficiency.1 Normal patent veins are anechoic and directly compressible with the transducer (Fig. e2-1, A). The Doppler spectrum demonstrates flow toward the heart, with respiratory variations in the flow pattern (the phasicity). Normal phasic flow in the distal veins suggests patent central veins. In addition, augmentation of flow can be demonstrated with distal compression (Fig. e2-1, B). An augmentation response usually suggests patent veins peripherally from the site of Doppler interrogation to the site of compression. Acute deep venous thrombosis (DVT) is predisposed by hypercoagulable states (congenital or acquired), endothelial injury, and stasis of blood flow (Virchow triad). The common predisposing clinical conditions for DVT are malignancy, trauma, perioperative states, and postpartum, congenital, or acquired hypercoagulable states. Ultrasonography is highly sensitive in the diagnosis of acute DVT.2 Acute thrombosis results in an enlarged vein with intraluminal low-level echoes. The vein is noncompressible, and the Doppler spectrum shows no demonstrable flow. Subacute thrombus is seen as moderate high-level echoes in the lumen, with minimal or no enlargement of the vein. Chronic thrombosis is difficult to identify because the veins become small and may sometimes demonstrate channels of flow (partial recanalization of thrombus). Rarely, calcification of the vein may be seen as a focal highly echogenic area with distal acoustic shadowing. Venous insufficiency predisposes to varicose veins, leg edema, venous stasis dermatitis, and venous ulcers. It can occur in the superficial or deep veins. Chronic venous insufficiency affecting the deep veins is usually predisposed by prior DVT. Venous insufficiency affecting the superficial veins may be related to prior thrombosis, but most often no cause is found. The elderly are predisposed to its development, and pregnancy and genetic factors also play a role. To plan appropriate therapy, it is necessary to document the

FIGURE e2-1. Normal veins. A, Ultrasound image shows common femoral vein (arrows) is compressible. Compressibility is a sign of a patent vein. B, Doppler evaluation shows normal phasicity (vertical arrow) and demonstrates augmentation (horizontal arrow) on distal compression.

site and degree of venous insufficiency and assess the anatomy of the superficial veins, perforator veins, and their relative contribution to varicosities. Venous insufficiency should ideally be assessed with the patient standing. A competent vein demonstrates flow toward the heart and transient reflux (1/2 systemic pressure), an occlusion balloon catheter is useful for temporary occlusion of the PAVM while monitoring pressures with a small catheter introduced from the contralateral femoral vein for 30 minutes. In our experience, temporary occlusion of the PAVM for 30 minutes has not led to significant elevations in pulmonary artery pressure, and we then proceed with the occlusion. Double-lumen occlusion balloon catheters (Boston Scientific, Natick, Mass.) are sometimes used for occluding high-flow large-lumen PAVMs (>10 mm diameter). This is particularly true if there is not a good “anchor” vessel present. Oversized MReye coils (Cook) can be placed through the occlusion balloon catheter to form a “scaffold” proximal to the sac. Once several high-radial-force coils are placed, the occlusion balloon is removed, and we complete cross-sectional occlusion of the PAVM by “nesting” soft platinum coils within the scaffold, using the standard 7/5F LuMax system. For patients with short high-flow arteries (i.e., < 2.5 cm long), embolization is performed through the occlusion balloon catheter (Boston Scientific) with a double-marker Renegade 0.021-inch microcatheter (Boston Scientific) and deployment-suitable pushable, and perhaps detachable, 0.018-inch coils. The resultant coil mass should be at least twice the diameter of the draining pulmonary veins and should fill the aneurysmal sac. Once the sac is filled with oversized microcoils, the occlusion balloon catheter is withdrawn and the occlusion completed with the 7/5F LuMax system and standard 0.035- or 0.038-inch fibered coils.12

TECHNIQUE Anatomy and Approaches Most PAVMs consist of a variable-length single segmental artery, but 10% are complex, with more than one segmental artery connecting the aneurysmal sac to the draining veins.9,16,17 Despite the magnificent images provided by highresolution multidetector CT scanners, as well as similar images from magnetic resonance angiography (MRA), we still depend on multiprojection diagnostic pulmonary angiography before embolotherapy. It is particularly important to be confident of the morphology of the artery as it enters the sac. The goal of therapy is to occlude the feeding artery as close to the sac as possible, thereby limiting reperfusion from adjacent branches in the lungs of children3 and

limiting reperfusion from the bronchial circulation, which is known to occur with proximal occlusions.18,19 Perhaps the most important aspect of imaging occurs just before deployment of coils. We perform a small series of hand injections of contrast material into the feeding artery in multiple projections, settling on the projection that provides us with the best view of the morphology of the feeding artery as it enters the aneurysmal sac. This image is preserved on the monitor next to our real-time monitor to guide us during deployment of coils. In many instances, the artery narrows before entering the sac. This is very favorable anatomy, and pushable fibered coils 2 mm larger than the diameter of the artery are simply placed and nested into a tight coil mass to produce cross-sectional occlusion. If the artery widens before entering the sac, a small adjacent artery that would be occluded by the coil mass is selectively entered, and a long Nester coil or stainless steel MReye coil (Cook) is deployed so that at least 2 cm of the first coil is anchored in this “anchor” branch. By using the anchor approach, we have not had a paradoxical embolization of a pushable fibered coil since 1994.9,17 Because of self-limited pleurisy, which occurs in approximately 15% of patients after occlusion of single or multiple PAVM in one lung, we tend to perform occlusion of one lung at a time. We prefer to have the patient return for treatment of the other side within 6 weeks. This also allows us to detect recanalization on the earlier occluded side. This is easier on the patient as well as the interventional radiologist. Figures 85-3 and 85-4 show various views of the anchor and scaffold techniques.

Technical Aspects Morphology of the artery as it enters the sac is a critical determinant of occlusion method (to anchor or not to anchor). Patients with a single PAVM receive heparin (50 units/ kg) during the occlusion. If the procedure is for multiple PAVMs, patients are fully heparinized (100 units/kg). All guidewires should be withdrawn under water to void a vacuum effect. Inadvertent air in the catheter can result in angina as a result of air paradoxically embolizing and entering the coronary circulation as the next coil or guidewire is introduced. The anchor technique describer earlier provides a safe method for deploying pushable fibered coils. If there is no suitable anchor vessel adjacent to the sac, and one is uncomfortable with the control of pushable oversized coils placed coaxially, the first coil can be a detachable coil followed by packing pushable coils to create cross-sectional occlusion. Use of a 7F guide catheter proximal in the feeding artery allows control of the inner 5F catheter so that pushable fibered coils can be packed into a dense “nest,” thus achieving cross-sectional occlusion. For high-flow arteries 10 mm in diameter or larger, the scaffold technique, consisting of building an endoskeleton close to the sac with oversized stainless steel or MReye coils is used. Cross-sectional occlusion is achieved by finishing the occlusion with Nester coils. We also use Amplatzer II plugs for larger feeding arteries (Fig. 85-5).

CHAPTER 85 PULMONARY ARTERIOVENOUS MALFORMATIONS: DIAGNOSIS AND MANAGEMENT 609

A

D

B

C

E

F

FIGURE 85-3. “Anchor” technique in asymptomatic patient with pulmonary arteriovenous malformation (PAVM). A-B, A 7F guide catheter with a 5F endhole catheter has been advanced into segmental artery to PAVM. Shallow left anterior oblique view best demonstrates entry of artery into aneurysmal sac. Because artery bifurcates immediately before entering sac, one branch is selected to anchor the first 2 cm of a 14-cm, 6-mm-diameter fibered coil; rest of coil is deposited in the other branch and proximal artery (C-D). Key to control of fibered coil placement is having a guide catheter held steady in feeding artery, which allows precise deployment of coils through the 5F end-hole catheter. E-F, Final placement of two additional 6-mm coils is demonstrated; they are tightly packed (nested). This prevents recanalization and again is facilitated by having a guide catheter proximal in the feeding segmental artery.

Because occlusion of PAVM is usually elective and there are a great number of anatomic variants, complete diagnostic angiography is required, as well as considering referring a patient with difficult anatomy to an experienced center.

CONTROVERSIES Although there is no worldwide consensus yet about which PAVMs should be occluded or exactly how to occlude them, consensus in North America and Europe is beginning to be achieved. For example, there is one report of a stroke event via a paradoxical embolus through a 2.8-mm-diameter artery,20 but for the most part in Great Britain, Canada, Denmark, Netherlands, and France, as well as in North America, the 3-mm-diameter artery appears to be the “threshold size” that warrants occlusion.21-28 With regard to occlusion devices, results equal to those obtained with pushable fibered coils had been achieved with detachable silicone balloons (Boston Scientific), but

since they were discontinued in 2002, we are occluding most PAVMs with standard pushable fibered coils, facilitated by use of a coaxial guide catheter system.9,29 Some authors have favored routinely occluding the aneurysmal sac with detachable coils.30,31 We have rarely needed detachable coils for occluding PAVMs, and their expense is a downside. In addition, occluding the sac of a PAVM removes our primary anatomic outcome of the procedure (i.e., thrombosis and disappearance of the sac).32 In 2006, the nitinol plug developed by Amplatz and associates was reported for occluding the feeding artery to the PAVM.9,33,34 Over the last 6 years, a number of reports of success using the Amplatzer I plug with coils or the Amplatzer II plug alone have been reported.14,15 We believe that to adequately treat patients with PAVMs, the interventionalist must be familiar with a variety of approaches, and ideally patients should be referred to experienced centers with a commitment to management and follow-up of patients with HHT/PAVM.35

610 PART 1 VASCULAR INTERVENTIONS · SECTION 10 THORACIC VASCULAR INTERVENTION

A

E

B

F

C

G

D

H

FIGURE 85-4. “Scaffold” and “anchor” techniques for large pulmonary arteriovenous malformation (PAVM) in 78-year-old woman with significant liver AVM as well as mild pulmonary artery hypertension (44/17 mmHg), and 6-year follow-up. A-B, Arterial and venous phase of a selective left pulmonary angiogram demonstrate a large PAVM with aneurysmal sac and no immediate evidence of a good anchor artery next to sac. C, A good anchor artery is demonstrated by a selective hand injection through guide catheter. It was decided to use anchor technique with oversized 20-mm stainless steel coils to form a scaffold. Main artery to PAVM measured 17 mm in diameter. D, Two 20-mm-diameter high-radial-force stainless steel coils have been sequentially anchored and allowed to reform in feeding artery. E, Additional 12- and 15-mm-diameter coils have been placed in scaffold. F, Final cross-sectional occlusion has been obtained by packing long fibered platinum coils within scaffold. G, Final selective angiogram demonstrates cross-sectional occlusion. H, Six-year follow-up angiogram (2008) demonstrated that left lung AVM remains occluded. We now use MReye coils to form a scaffold; they are compatible with magnetic resonance imaging and have the same radial force as stainless steel coils. Currently, one could consider a 20-mm Amplatzer II Plug and possibly achieve the same long-term result.

OUTCOMES Outcomes from multiple single-center series have been reported, and recently we reported our prospective outcomes of 155 patients observed closely for a mean of 6.5 years.9 All series have demonstrated reduction in shunt measured by radionuclides or 100% oxygen studies. We have maintained and demonstrated that anatomic as well as physiologic outcomes are necessary. In our series, similar to the Dutch series,8 there were a few recurrences due to reperfusion of incompletely occluded arteries or accessory arteries. Reperfusion can also occur through collateral perfusion from adjacent normal arteries, particularly in children and adolescents, as well as hypertrophy of bronchial arteries, the latter more common in the small subset of patients with diffuse PAVMs.4,34,35 All recurrences can be relatively easily reoccluded at the 1-year follow-up visit. Most importantly in the Dutch and our series, enlargement of unoccluded PAVM was demonstrated by thinsection noncontrast CT assessment at 1 year and every 5 years. When PAVMs were multiple or if there was any doubt about reperfusion, often an outpatient pulmonary angiogram was performed and, at the same time, reocclusion of any reperfused PAVM.

Since a small risk of infection remains from unoccluded small or tiny PAVMs, prophylactic antibiotics continue to be recommended prior to procedures that may cause a bacteremia, such as dental work. Stroke or TIA events were reduced and often found to be due to other causes than paradoxical embolus through a PAVM (e.g., atrial fibrillation or other cardiac events).9 Similar results were obtained by Mager et al. from the HHT center in The Netherlands.8

COMPLICATIONS Those complications occurring during the procedure are due in most part to inadvertent air traversing the PAVM during flushing or passing wires, coils, or other devices. In addition, bland thrombus occurring on catheters can pass through a PAVM and cause serious downstream events. Angina and/or a TIA are now rare events, occurring in no more than 1% of patients treated by experienced teams. In most instances they can be prevented by withdrawing all wires under water and by heparinization during the procedure. The most feared complication is paradoxical embolization of a device, and these occurrences have been well

CHAPTER 85 PULMONARY ARTERIOVENOUS MALFORMATIONS: DIAGNOSIS AND MANAGEMENT 611

A

D

B

E

C

F

FIGURE 85-5. Amplatzer II Vascular Plug in medium-sized pulmonary arteriovenous malformation (PAVM) without good anchor vessel in 57-year-old with long-standing history of dyspnea and fatigue. A, Right lower lobe PAVM is identified. B, We did not identify a suitable anchor artery, and a 10-mm Amplatzer II device was temporarily placed (C). C-D, Amplatzer plug in place before detachment; control angiogram demonstrates excellent occlusion. E-F, Four months later, patient returned for treatment of other PAVMs. Magnification view shows Amplatzer occluder in excellent position and confirms PAVM occlusion.

described.17,36 Now that we have adopted the anchor technique, we have not encountered any paradoxical embolization of a device. Groups treating PAVM should be familiar with “snare devices” for retrieval of misplaced coils. Although not strictly considered a complication, selflimited pleurisy occurs in 10% to 20% of patients, usually beginning 48 to 96 hours after the procedure. It is thought that this occurs by thrombosis of the aneurysmal sac when adjacent to visceral pleura. Irritation of the visceral pleura is associated with pain. This is typically self-limited and can be treated with ibuprofen for 2 to 3 days. Patients should be warned about this and encouraged to follow up with the treatment center. In a very small percentage of patients, there may be late pleurisy occurring 2 to 4 months after a successful procedure, which we attribute to delayed thrombosis of the

sac. This may be associated with high fever, and in addition to ibuprofen we often give a short course of oral antibiotics.

POSTPROCEDURAL AND FOLLOW-UP CARE Most patients are seen in our interventional clinic the day before the procedure, and we obtain a history and perform a thorough physical examination. Embolotherapy is performed as a same-day procedure, with discharge 4 to 5 hours after it is finished. All patients call our office or are contacted for 2 days after the procedure to check their progress. Most important is the follow-up noncontrast CT of the chest 6 months to 1 year after treatment. If the patient

612 PART 1 VASCULAR INTERVENTIONS · SECTION 10 THORACIC VASCULAR INTERVENTION cannot return for an outpatient visit at this time, the chest CT is sent for review; after reviewing the CT, we dictate a letter to the referring doctor outlining our findings and follow-up recommendations. In patients with multiple PAVMs in both lungs, it is our impression that follow-up pulmonary angiography at 1 year is more helpful in excluding recurrences than non–contrast enhanced CT. It is recommended that all patients be seen every 5 years after their occlusion for follow-up clinical evaluation for growth of unoccluded PAVMs and for other organ manifestations of HHT. KEY POINTS •

Pulmonary arteriovenous malformation (PAVM) is a marker for a relatively common genetic disorder, hereditary hemorrhagic telangiectasia (HHT).

•

Treatment of PAVM is performed by embolotherapy techniques using guiding catheters and pushable fibered coils, and in some instances with detachable coils, fully retrievable before final deployment, and Amplatzer II plugs.

•

Permanent ablation of PAVMs is possible, with a high success and low complication rate in experienced HHT centers with an interventional radiologist performing these embolizations once or twice per week.

•

Most patients with HHT and PAVM have small PAVMs remaining after closure of large ones, and follow-up is important to detect recurrences and monitor growth of small PAVMs to the size when treatment is needed.

▶ SUGGESTED READINGS Gossage J, Kanj G. State of the art: pulmonary arteriovenous malformations. Am J Respir Crit Care Med 1988;158:643–77. Guttmacher AE, Marchuk DA, White RI Jr. Hereditary hemorrhagic telangiectasia. N Engl J Med 1995;333:918–24. Shovlin CL, Guttmacher AE, Buscarini E, et al. Diagnostic criteria for hereditary hemorrhagic telangiectasia. Am J Med Genet 2000;91: 66–7.

The complete reference list is available online at www.expertconsult.com.

CHAPTER 85 PULMONARY ARTERIOVENOUS MALFORMATIONS: DIAGNOSIS AND MANAGEMENT612.e1 REFERENCES 1. Guttmacher AE, Marchuk DA, White RI Jr. Hereditary hemorrhagic telangiectasia. N Engl J Med 1995;333:918–24. 2. Gershon AS, Faughnan ME, Chon KS, et al. Transcatheter embolotherapy of maternal pulmonary arteriovenous malformations during pregnancy. Chest 2001;119:470–7. 3. Faughnan M, Thabet A, Mei-Zahav M, et al. Pulmonary arteriovenous malformations in children: outcomes of transcatheter embolotherapy. J Pediatr 2004;145:826–31. 4. Pierucci P, Murphy J, Henderson KJ, et al. New definition and natural history of patients with diffuse pulmonary arteriovenous malformations: twenty-seven-year experience. Chest 2008;133(3):653–61. 5. White RI Jr, Pollak JS, Wirth JA. Pulmonary arteriovenous malformations: diagnosis and transcatheter embolotherapy. J Vasc Interv Radiol 1996;7(6):787–804. 6. Kjeldsen AD, Oxhoj H, Andersen PE, et al. Pulmonary arteriovenous malformations: screening procedures and pulmonary angiography in patients with hereditary hemorrhagic telangiectasia. Chest 1999;116: 432–9. 7. Nanthakumar K, Hyland RH, Graham AT, et al. Contrast echocardiography for detection of pulmonary arteriovenous malformations. Am Heart J 2001;141:243–6. 8. Mager JJ, Overtoom TT, Blauw H, et al. Embolotherapy of pulmonary arteriovenous malformations: long-term results in 112 patients. J Vasc Interv Radiol 2004;15:451–6. 9. Pollak JS, Saluja S, Thabet A, et al. Clinical and anatomic outcomes after embolotherapy of pulmonary arteriovenous malformations. J Vasc Interv Radiol 2006;17:35–45. 10. Li W, Niu B, Henderson K, Northrup V, et al. Reproducibility of oxygen saturation monitoring during six-minute walk test and exercise stress test in patients with pulmonary arteriovenous malformations associated with hereditary hemorrhagic telangiectasia. Pediatr Cardiol 2011;32(5):590–4. 11. Harrison RE, Flanagan JA, Sankelo M, et al. Molecular and functional analysis identifies ALK-1 as the predominant cause of pulmonary hypertension related to hereditary hemorrhagic telangiectasia. J Med Genet 2003;40:865–71. 12. Tal MG, Saluja S, Henderson KJ, White RI Jr. Vein of Galen technique for occluding the aneurysmal sac of pulmonary arteriovenous malformations. J Vasc Interv Radiol 2002;13:1261–4. 13. White RI Jr, Pollak JS. Controlled delivery of pushable fibered coils for large vessel embolotherapy. In Golzarian J, Sun S, Sharafuddin MJ, editors. Vascular embolotherapy: a comprehensive approach. Berlin: Springer; 2006, p. 35–42. 14. Trerotola SO, Pyeritz RE. Does use of coils in addition to Amplatzer vascular plugs prevent recanalization? AJR Am J Roentgenol 2010; 195(3):766–71. 15. Tapping CR, Ettles DF, Robinson GJ. Long-term follow-up of treatment of pulmonary arteriovenous malformations with Amplatzer Vascular Plug and Amplatzer Vascular Plug II devices. J Vasc Interv Radiol 2011;22(12):1740–6. 16. White RI Jr, Mitchell SE, Barth KH, et al. Angioarchitecture of pulmonary arteriovenous malformations: an important consideration before embolotherapy. AJR Am J Roentgenol 1983;140:681–6. 17. White RI Jr, Pollak JS, Wirth JA. Pulmonary arteriovenous malformations: diagnosis and transcatheter embolotherapy. J Vasc Interv Radiol 1996;7:787–804. 18. Sagara K, Miyazono N, Inoue H, et al. Recanalization after coil embolotherapy of pulmonary arteriovenous malformations: study of

long-term outcome and mechanism of recanalization. AJR Am J Roentgenol 1998;170:727–30. 19. White RI Jr. Recanalization after embolotherapy of pulmonary arteriovenous malformations: Significance? Outcome? AJR Am J Roentgenol 1998;171:1704. 20. Todo K, Moriwaki H, Higashi M, et al. A small pulmonary arteriovenous malformation as a cause of recurrent brain embolism. Am J Neuroradiol 2004;25:428–30. 21. Dutton JA, Jackson JE, Hughes JM, et al. Pulmonary arteriovenous malformations: results of treatment with coil embolization in 53 patients. AJR Am J Roentgenol 1995;165:1119–25. 22. Milic A, Chan RP, Cohen JH, et al. Reperfusion of pulmonary arteriovenous malformations after embolotherapy. J Vasc Interv Radiol 2005;16:1675–83. 23. Andersen PE, Kjeldsen AD. Clinical and radiological long-term follow-up after embolization of pulmonary arteriovenous malformations. Cardiovasc Intervent Radiol 2006;29:70–4. 24. Mager JJ, Overtoom TT, Blauw H, et al. Embolotherapy of pulmonary arteriovenous malformations: long-term results in 112 patients. J Vasc Interv Radiol 2004;15:451–6. 25. Hewes RC, Auster M, White RI Jr. Cerebral embolism: first manifestation of pulmonary arteriovenous malformation in patients with hereditary hemorrhagic telangiectasia. Cardiovasc Intervent Radiol 1985; 8:151–5. 26. Pelage J-P, Lacombe P, White RI Jr, Pollak JS. Pulmonary arteriovenous malformations. In Golzarian J, Sun S, Sharafuddin MJ, editors. Vascular embolotherapy: a comprehensive approach. Berlin: Springer; 2006. p 279–96. 27. White RI Jr, Nyhan AL, Terry P, et al. Pulmonary arteriovenous malformations: techniques and long-term outcome of embolotherapy. Radiology 1988;169:633–69. 28. White RI Jr, Pollak JS. Pulmonary arteriovenous malformations: diagnosis with three-dimensional helical CT—a breakthrough without contrast media. Radiology 1994;191:613–4. 29. Saluja S, Sitko I, Lee DW, et al. Embolotherapy of pulmonary arteriovenous malformations with detachable balloons: long-term durability and efficacy. J Vasc Interv Radiol 1999;10:883–9. 30. Dinkel HP, Triller J. Pulmonary arteriovenous malformations: embolotherapy with superselective coaxial catheter placement and filling of venous sac with Guglielmi detachable coils. Radiology 2002;223: 709–14. 31. Takahashi K, Tanimura K, Honda M, et al. Venous sac of pulmonary arteriovenous malformation: preliminary experience using interlocking detachable coils. Cardiovasc Intervent Radiol 1999;22:210–3. 32. White RI Jr, Pollak JS, Picus D. Letter to editor. Are Guglielmi detachable coils necessary for treating pulmonary arteriovenous malformations? Radiology 2003;226:599–600. 33. Cil B, Canyigit M, Ozkan OS, et al. Bilateral multiple pulmonary arteriovenous malformations: endovascular treatment with the Amplatzer vascular plug. J Vasc Interv Radiol 2006;17:141–5. 34. White RI Jr. Letter to editor. Re: Bilateral multiple pulmonary arteriovenous malformations: endovascular treatment with the Amplatzer vascular plug. J Vasc Interv Radiol 2006;17:913–5. 35. Faughnan ME, Lui YW, Wirth JA, et al. Diffuse pulmonary arteriovenous malformations: characteristics and prognosis. Chest 2000; 117(1):31–8. 36. Haitjema R, ten Berg JM, Overtoom TT, et al. Unusual complications after embolization of a pulmonary arteriovenous malformation. Chest 1996;109:1401–4.

CHAPTER

86

Percutaneous Interventions for Acute Pulmonary Embolism William T. Kuo

Venous thromboembolism (VTE) is a global health problem that encompasses acute deep venous thrombosis (DVT) and acute pulmonary embolism (PE). Although the true incidence of PE is unknown, it is recognized as a significant cause of morbidity and mortality in hospitalized patients.1 In the United States alone, it is estimated there are between 500,000 and 600,000 cases per year,2 and approximately 300,000 people die every year from acute PE.3 Indeed, acute PE is believed to be the third most common cause of death among hospitalized patients.4 Common risk factors for acute PE are related to underlying genetic conditions (e.g., factor V Leiden mutation, prothrombin 20210A mutation, and multiple other thrombophilias), acquired conditions (e.g., cancer, immobilization, trauma, surgery, prior DVT), and acquired hypercoagulable states (e.g., oral contraceptive use, nephrotic syndrome, antiphospholipid syndrome, disseminated intravascular coagulation, hyperestrogenic states, obesity). There are three basic categories of acute PE: (1) simple PE with no associated heart strain and no hypotension, (2) submassive PE with associated right heart strain, (3) massive PE with associated right heart strain and hemodynamic shock. Patients with simple acute PE require treatment with therapeutic anticoagulation alone. Patients with submassive PE may require treatment escalation beyond anticoagulation; and patients with massive PE certainly require treatment escalation beyond anticoagulation alone.5 The immediate mortality rate related to simple PE is less than 8% when the condition is recognized and treated with anticoagulation.1,6,7 However, patients with submassive PE have a higher cumulative mortality rate, reaching approximately 20% over a 90-day period.8 Patients with massive PE have the highest mortality rate, which can exceed 58%, including a high risk of sudden death.9 The pathophysiology of PE consists of direct physical obstruction of the pulmonary arteries, hypoxemic vasoconstriction, and release of potent pulmonary arterial vasoconstrictors that further increase pulmonary vascular resistance and right ventricular (RV) afterload. Acute RV pressure overload may result in RV hypokinesis and dilation, tricuspid regurgitation, and ultimately RV failure. RV pressure overload may also result in increased wall stress and ischemia by increasing myocardial oxygen demand while simultaneously limiting its supply. Ultimately, cardiac failure due to acute PE results from a combination of the increased wall stress and cardiac ischemia that compromise RV function and impair left ventricular (LV) output, resulting in lifethreatening hemodynamic shock.9 Depending on underlying cardiopulmonary reserve, patients with acute massive PE may deteriorate over the course of several hours to days and develop systemic arterial hypotension, cardiogenic shock, and cardiac arrest. Owing to the risk of sudden death, these critically ill patients with massive PE should be

quickly identified as candidates for rapid endovascular treatment as a lifesaving procedure.5

INDICATIONS Because of the high mortality associated with acute massive PE, successful management requires prompt risk stratification and decisive early intervention (Table 86-1). Confirmation of hemodynamic shock attributed to central obstructing embolus should be present to justify treatment escalation beyond anticoagulation. The American College of Chest Physicians has recommended that percutaneous catheterdirected therapy (CDT) be considered in acute massive PE patients who are unable to receive systemic thrombolytic therapy because of bleeding risk.10 In addition, global metaanalytic data has demonstrated that percutaneous CDT can be considered as a first-line treatment option in lieu of intravenous (IV) tissue plasminogen activator (tPA) for patients with massive PE.11 Pulmonary angiography was once considered the gold standard for diagnosing PE, but it has largely been replaced by the wide availability of cross-sectional imaging. Historically, many types of imaging studies have been used in diagnosing acute pulmonary embolism, including ventilation/perfusion (V/Q) scanning, magnetic resonance angiography (MRA), and computed tomographic angiography (CTA). CTA is the preferred modality and has proven to be advantageous thanks to its wide availability, superior speed, characterization of nonvascular structures, and detection of venous thrombosis. CTA has the greatest sensitivity and specificity for detecting emboli in the main, lobar, or segmental pulmonary arteries. Systematic reviews and randomized trials suggest that outpatients with suspected pulmonary embolism and negative CTA studies have excellent outcomes without therapy.12 If a patient has either acute or chronic renal insufficiency and contrast administration is undesirable, echocardiography may be used to evaluate for right heart dysfunction as an indication for underlying acute PE. The echocardiogram can be performed at bedside, and the study may reveal findings that strongly support hemodynamically significant pulmonary embolism,13 offering the potential to guide treatment escalation to thrombolytic and/or endovascular therapy. Large emboli moving from the heart to the lungs are occasionally confirmed with this technique. In addition, intravascular ultrasonography has also been used at the bedside to visualize central pulmonary emboli.14 Although the diagnosis of submassive PE follows a similar workup to evaluating massive PE, these patients do not present with systemic arterial hypotension, and particular attention must be paid to detecting the presence of right heart strain, which clinches the diagnosis of submassive PE. Identifying right heart strain allows risk stratification for 613

614 PART 1 VASCULAR INTERVENTIONS · SECTION 10 THORACIC VASCULAR INTERVENTION TABLE 86-1. Risk Stratification and Indications for Aggressive Intervention to Treat Massive Pulmonary Embolism At least one of the following criteria must be present: 1. Arterial hypotension ( 40 mmHg) 2. Cardiogenic shock with peripheral hypoperfusion and hypoxia 3. Circulatory collapse with need for cardiopulmonary resuscitation From Uflacker R. Interventional therapy for pulmonary embolism. J Vasc Interv Radiol 2001;12:147–64.

possible treatment escalation beyond anticoagulation in normotensive PE patients.5 Echocardiography is the best imaging study to detect RV dysfunction in the setting of acute PE. Characteristic echocardiographic findings in patients with submassive PE include RV hypokinesis and dilatation, interventricular septal flattening and paradoxical motion toward the LV, abnormal transmitral Doppler flow profile, tricuspid regurgitation, pulmonary hypertension as identified by a peak tricuspid regurgitant jet velocity over 2.6 m/s, and loss of inspiratory collapse of the inferior vena cava (IVC).15 An RV-to-LV end-diastolic diameter ratio of 0.9 or greater, assessed in the left parasternal long-axis or subcostal view, is an independent predictor of hospital mortality.16 Detection of RV enlargement by chest CTA is especially convenient for diagnosis of submassive PE, because it uses data acquired during the initial diagnostic scan. Submassive PE can be diagnosed when RV enlargement on chest CT, defined by an RV-to-LV diameter ratio greater than 0.9, is observed17; RV enlargement on chest CTA also predicts increased 30-day mortality in patients with acute PE.17,18 Even if shock and death do not ensue, survivors of acute submassive PE remain at risk for developing chronic PE and thromboembolic pulmonary hypertension.19 Additionally, monitoring cardiac troponin T (cTnT) identifies the high-risk group of normotensive patients with submassive PE.20 A persistent increased cTnT level (>0.01 ng/mL) in PE patients with right heart strain predicts a significant risk of a complicated clinical course and fatal outcome, and these patients therefore require more aggressive treatment.20 Identifying submassive PE for treatment escalation is important because these normotensive PE patients demonstrate increased short-term mortality and high risk of adverse outcomes when the degree of heart strain results in elevations in levels of cardiac troponins and brain-type natriuretic peptide.21,22 The optimal protocol for treatment of acute submassive PE is still in evolution, but a proposed algorithm for managing submassive PE has been published23 describing treatment escalation beyond anticoagulation (Fig. 86-1). Based on the history, angiographic findings, and outcomes of patients undergoing CDT for acute PE, three groups of patients have been identified24: • Type I patients with fresh clots that have recently embolized should respond well to mechanical thrombectomy with increased peripheral flow and oxygenation. In general, 2 to 3 weeks is the upper age limit of thrombus when considering the option of CDT. • Type II patients with older and more organized clots respond less effectively to mechanical thrombectomy

Submassive pulmonary embolism ↑Cardiac biomarkers with: • RV dysfunction on echocardiography OR Yes

• RV enlargement on CT

Contraindications to fibrinolysis? Yes Contraindications to catheter-assisted embolectomy Yes Contraindications to surgical embolectomy? Yes Anticoagulation ! IVC filter insertion

No Anticoagulation alone

No

Consider fibrinolysis No

No

Consider catheterassisted embolectomy Consider surgical embolectomy

FIGURE 86-1. Algorithm for management of submassive pulmonary embolism patients. According to the algorithm, catheter-directed therapy should be considered in particular when there are contraindications to systemic thrombolysis. CT, Computed tomography; IVC, inferior vena cava; RV, right ventricle. (From Piazza G, Goldhaber SZ. Management of submassive pulmonary embolism. Circulation 2010;122:1124–9.)

alone. Although more chronic clots are likely to remain, there is a chance for pulmonary flow improvement if pharmacologic thrombolysis of overlying acute thrombus can be achieved. • Type III patients with organized chronic PE do not respond well to the effects of CDT.

CONTRAINDICATIONS Because of the risk of major hemorrhage, aggressive anticoagulation, systemic thrombolysis, and local catheterdirected thrombolysis may be contraindicated in patients with recent major general or intracranial surgery. In such cases, mechanical methods of percutaneous catheterdirected thrombectomy, fragmentation, and/or aspiration should be considered without pharmacologic agents. Since pulmonary hypertension is a relative contraindication to pulmonary angiography, the degree of pulmonary hypertension and underlying cardiopulmonary reserve are important considerations prior to performing pulmonary angiography. These factors must be used to determine a safe rate and volume of contrast injection into the pulmonary circulation. For instance, when systolic pulmonary artery pressure (PAP) exceeds 55 mmHg, or right ventricular end diastolic pressure (RVEDP) is greater than 20 mmHg, the mortality associated with pulmonary angiography using large-volume power injection has been reported to be as high as 3%.2,3 Therefore, such patients with massive PE should only receive a limited rate of power injection, controlled by hand. In the submassive PE patient, a selective contrast injection into the main left or right PA should not exceed a 20-mL volume at a rate of 10 mL/s. Lower injection parameters by hand may be considered depending on

CHAPTER 86 PERCUTANEOUS INTERVENTIONS FOR ACUTE PULMONARY EMBOLISM 615 Severity of embolism Right

Left 1

1

Flow 1

0123

3

0123

7

9

6

1

1

2

16

0123 5

2

1

Flow

1

1 1

2

0123

3

4 0123

0123 1

Involvement

1

1

1

1

1 Involvement

0-9

1 0-7

Reduction of flow 0 - 9

Reduction of flow 0 - 9

Total score

Total score

0 - 18

0 - 16

Overall total 0 - 34

FIGURE 86-2. Schematic of pulmonary arterial segments used to grade severity of pulmonary embolism based on angiographic findings before and after treatment with thrombolytic and/or catheter-directed therapy. Segmental clot involvement and reduction of flow are scored separately. (From Miller GAH, Sutton GC, Kerr IH, et al. Comparison of streptokinase and heparin in treatment of isolated acute massive pulmonary embolism. BMJ 1971;2:681–4.)

degree of heart failure, as determined by the operator’s judgment, to achieve adequate vessel opacification without endangering the patient. Pulmonary angiography can be used to calculate the degree of pulmonary obstruction before and after treatment, using the calculated Miller Index (Fig. 86-2). Finally, in PE patients with contraindications to anticoagulation and/or thrombolytic agents, placement of an IVC filter should be considered for prophylaxis of recurrent PE.

EQUIPMENT Modern CDT for massive PE has been defined according to the following criteria: use of low-profile catheters and devices (≤10F), catheter-directed mechanical fragmentation and/or aspiration of emboli, and intraclot thrombolytic injection if a local drug is infused.11 Therefore, a variety of devices can be successfully used to treat PE, so long as they meet criteria for modern CDT (Table 86-2). Depending on anticipated bleeding risk, CDT may be performed with either no or low-dose local tPA injection. The initial goal of all these techniques is rapid debulking of central thrombus to relieve life-threatening heart strain, immediately improve pulmonary perfusion, and improve oxygenation (Fig. 86-3).5 In contrast to CDT for massive PE, which requires mechanical methods, the protocol for submassive PE consists of gentle image-guided infusion catheter placement and overnight pharmacologic thrombolysis without aggressive mechanical intervention. This treatment may be offered to patients who present primarily with submassive PE (Fig. 86-4), or as adjunctive treatment in patients with prior

massive PE who have been “downstaged” via central clot debulking to submassive PE. For at least one of the access sites, it is preferable that the sheath be sized at least 2F larger than the infusion catheter to allow subsequent central pulmonary artery (PA) pressure measurements through the sheath. An 8F to 10F vascular sheath (e.g., Flexor sheath [Cook Medical, Bloomington, Ind.]) is recommended, and the sheath should be long enough to reach from the access point to the main pulmonary trunk. If a second sheath is placed for dual catheter infusions, the operator can decide on the specific diameter and length, based on vessel tortuosity, sufficient to provide stability for the second infusion catheter.

TECHNIQUE Anatomy and Approaches Solid catheter-based skills along with knowledge of the cardiopulmonary anatomy are important for safe intravascular placement and manipulation of catheters and thrombectomy devices. Catheterization of the PA can be performed using a variety of methods, depending on the operator’s preference and experience. An example via the transfemoral approach is using a 5F or 6F pigtail catheter in conjunction with a hydrophilic Glidewire (Terumo Medical Corp., Somerset, N.J.) and torque control device. An example via the right transjugular approach is using a C2 catheter (Cook Medical) in conjunction with a Glidewire and torque control device. Regardless of technique used, the electrocardiogram should be continuously monitored by trained nursing staff. Catheterization should be meticulously performed to

616 PART 1 VASCULAR INTERVENTIONS · SECTION 10 THORACIC VASCULAR INTERVENTION TABLE 86-2. Catheter-Directed Therapy for Massive Pulmonary Embolism in 594 Patients

Author, Year (Reference)

Country

Patients, n

Sexn

Local Intraclot Lytic During CDT-n

Local Intraclot Lytic, Extended Infusion-n

Age: Mean, Range

Techniquen PF-10

8

1

0

0

8/10 (80)

PF-20

0

0

1

0

16/20 (80)

ATD-9

0

5

0

0

9/9 (100)

PF-20

0

16

0

0

14/20 (70)

PF & AT-25

25

21

0

0

25/25 (100)

PF-10, (ATD-na)

0

0

0

0

9/10 (90)

PF-1, MC-2

2

2

0

0

3/3 (100)

MC-5

5

5

1

0

5/5 (100)

ATD-5

1

1

0

1

3/5 (60)

PF-16, (BA-na)

16

16

3

0

14/16 (88)

PF & BA-5

5

5

0

2

5/5 (100)

Minor Cxs

Major Cxs

Clinical Success (%)

Prospective Studies SchmitzRode et al. 199848

Germany

10

M-6, F-4

54

SchmitzRode et al. 200049

Germany

20

M-10, F-10

59

MullerHulsbeck et al. 200150

Germany

9

M-4, F-5

55

Prokubovsky et al. 200351

Russia

20

na

51

Tajima et al. 200452

Japan

25

M-8, F-17

61

Barbosa et al. 200853

Brazil

10

M-7, F-3

57

(36-70)

(48-60)

(27-85)

(32-75)

(35-77) (39-75)

Retrospective Studies Brady et al. 199154

England

3

M-0, F-3

36

Rafique et al. 199255

South Africa

5

M-1, F-4

35

Uflacker et al. 199624

U.S.A.

5

M-4, F-1

45

Fava et al. 199732

Chile

16

M-8, F-8

49

Stock et al. 199756

Switzerland

5

M-3, F-2

50

Basche et al. 199757

Germany

na

na (21-73)

PF & BA-2, BA-13

na

na

0

0

12/15 (80)

Hiramatsu et al. 199958

Japan

8

M-4, F-4

58

AT & WD-8

0

8

0

0

7/8 (88)

Wong et al. 199959

England

4

M-2, F-2

33

PF-1,PF & G-1, G-2

0

4

0

1

3/4 (75)

Murphy et al. 199960

Ireland

4

M-2, F-2

60

MC & WD-4

4

4

0

0

4/4 (100)

Voigtlander 199961

Germany

5

M-4, F-1

57

RT-5

0

0

4

0

3/5 (60)

Fava et al. 200062

Chile

11

M-3, F-8

61

Hy-11

0

4

0

0

10/11 (91)

Egge et al. 200263

Norway

3

M-2, F-1

49

PF-3

3

3

0

0

3/3 (100)

De Gregorio et al. 200233

Spain

59

M-25, F-34

56

59

57

8

0

56/59 (95)

Zeni et al. 200336

U.S.A.

16

M-9, F-8

52

0

10

2

1

14/16 (88)

15

(18-71) (21-47)

(25-64)

(20-68) (21-80)

(42-87)

(18-46) (46-66)

(25-72) (37-79) (40-54) (22-85)

(30-86)

PF-52, PF & BA-4, PF & DB -3 RT-16

CHAPTER 86 PERCUTANEOUS INTERVENTIONS FOR ACUTE PULMONARY EMBOLISM 617 TABLE 86-2. Catheter-Directed Therapy for Massive Pulmonary Embolism in 594 Patients—cont’d

Author, Year (Reference)

Country

Reekers et al. 200364

Netherlands

Tajima et al. 200465

Patients, n

Sexn

Age: Mean, Range

Techniquen

Local Intraclot Lytic During CDT-n

Hy-6, Oa-1

7

0

0

0

6/7 (86)

AT-15

9

0

0

0

15/15 (100)

Hy-4, Oa-3

3

3

1

1

6/7/ (86)

RT-6

4

0

2

0

5/6 (83)

na

na

0

1

7/8 (88)

6

0

0

0

15/15 (100)

164

164

0

0

138/164 (84)

RT-6

2

0

5

2

4/6 (67)

PF-3, PF & AT-2

3

0

0

0

2/5 (40)

PF-10, PF & AT-5, PF+SR-4

19

na

0

0

18/19 (95)

RT-20

na

0

8

8

17/20 (85)

RT-13

na

0

6

8

8/13 (62)

ATD-17, SR-9

21

0

1

0

26/26 (100)

PF-5, PF & SR-13

2

0

0

0

16/18 (90)

PF & AT-6, PF & AT & BA-2, RT & AT-2, AT & IC-2

8

na

1

0

10/12 (83)

(7.9%)* [5.011.3%]

(2.4%)* [1.94.3%]

7

M-2, F-6

46

Japan

15

M-4, F-11

60

Fava et al. 200566

Chile

7

M-3, F-4

56

Siablis et al. 200537

Greece

6

M-4, F-2

59

Yoshida et al. 200667

Japan

8

M-4, F-4

61

Li J-J et al. 200668

China

15

M-11, F-4

56

Pieri and Agresti. 200769

Italy

na

68

Chauhan et al. 200770

U.S.A.

6

M-2 F-4

64

Krajina 200771

Czech Rep

5

M-1, F-4

67

Yang 200772

China

19

M-13, F-6

62

Margheri 200873

Italy

20

M-12, F-8

66

Vecchio et al 200838

Italy

13

na

68

Chen et al 200874

China

26

M-15, F-11

53

Eid-Lidt et al. 200835

Mexico

18

M-6, F-12

51

Kuo et al. 200831

U.S.A.

12

M-7, F-5

56

TOTAL = 35

164

594

53

(28-76)

(27-79) (30-79) (42-76) (47-75)

(19-73)

(35-78)

(49-78)

(52-80) (22-87)

(32-85) (54-80) (36-71) (47-55)

(21-80)

PF & AT-8

PF & ATD-13, PF & Hy-1, PF & Oa-1 PF-164

(18-87)

356/535 67%

Local Intraclot Lytic, Extended Infusion-n

329/552 60%

Minor Cxs

Major Cxs

Clinical Success (%)

(86.5%)* [82.290.2%]

*

Pooled estimates from random effects model. [%], 95% confidence intervals; AT, aspiration thrombectomy; ATD, Amplatz thrombectomy device (Microvena, White Bear Lake, Minn.); B, Dormia basket (Cook Europe, Bjaeverskov, Denmark); BA, balloon fragmentation; Cxs, complications; G, Gensini (Cordis Corp., Miami, Fla.); Hy, Hydrolyzer (Cordis Corp., Miami, Fla.); IC, infusion catheter; MC, multipurpose catheter; na, data not available; Oa, Oasis (Boston Scientific, Galway, Ireland); PF, pigtail fragmentation; RT, rheolytic AngioJet thrombectomy (Possis Medical, Minneapolis, Minn.); SR, Straub Rotarex (Straub Medical, Wangs, Switzerland); WD, wire disruption. From Kuo WT, Gould MK, Louie JD, et al. Catheter-directed therapy for the treatment of massive pulmonary embolism: Systematic review and metaanalysis of modern techniques. J Vasc Interv Radiol 2009;20:1431–40.

minimize wire contact with cardiac structures to reduce the risk of ventricular tachycardia and vascular perforation while passing through the right heart and into the pulmonary trunk. Once the pulmonary outflow has been selected, depending on operator preference, the hydrophilic wire can

be exchanged for a non-hydrophilic wire such as a 0.035inch Rosen wire (Boston Scientific, Natick, Mass.) to provide stability for subsequent vascular sheath placement. Selective catheterization of the main right and left PAs is routinely performed, but selective and subselective

618 PART 1 VASCULAR INTERVENTIONS · SECTION 10 THORACIC VASCULAR INTERVENTION

A

B

E

C

D

F

FIGURE 86-3. This patient with massive pulmonary embolism presented with syncope. A, Right pulmonary angiogram showed a large central embolism, with thrombus occluding most of the artery and sparing just a portion of the right upper lobe. B, Left pulmonary angiogram showed a large filling defect in main left pulmonary artery (PA). Catheter-directed therapy was initiated first to debulk central emboli. An infusion catheter was then wedged into main thrombus in the right PA, and urokinase was started at 100,000 IU/h for 36 hours, with alternating catheterization of the left PA for 12 hours. C-D, Posttreatment angiography showed significant improvement of bilateral pulmonary circulation. Patient experienced dramatic clinical improvement along with reduction of PA pressures. E-F, Pre- and post-treatment lung perfusion scans showed marked improvement in right lung perfusion, with some improvement in left lower lobe.

catheterization of pulmonary segments is often necessary for further treatment. The femoral approach is preferred in most patients who are candidates for catheter-directed pulmonary thrombolysis or thrombectomy, but the right internal jugular approach is also feasible and may be preferred in the presence of IVC or iliofemoral thrombus. Most thrombectomy devices are easily passed from the right internal jugular vein into the right ventricle and PA. Anticoagulation Initial IV administration of heparin is the therapy of choice to treat all forms of acute pulmonary thromboembolism. Heparin binds to and accelerates the activity of antithrombin III, prevents additional thrombus formation, and permits endogenous fibrinolytic mechanisms to thrombolyse the clot that has already formed. During CDT for massive PE, full heparin anticoagulation may be continued during the clot debulking procedure. However, once the massive PE has been converted to submassive PE and overnight catheter-directed thrombolytic infusion is initiated, full heparin anticoagulation should be discontinued to minimize the risk of bleeding complications. During concomitant low-dose tPA infusion, a subtherapeutic heparin dose is desirable (partial thromboplastin time [PTT] < 60 seconds) to minimize risk of peri-sheath clot formation. To achieve this, the heparin infusion rate is typically between

300 and 500 units/h through a peripheral IV site. Once the patient has completed a course of catheter-directed thrombolytic therapy, full therapeutic anticoagulation can be resumed with full-dose IV heparin or low-molecular-weight heparin (LMWH). Full heparin anticoagulation should be maintained for 7 to 10 days as a bridge to subsequent oral anticoagulation. Systemic Thrombolysis Current approved medical therapy for acute massive PE consists of systemic thrombolysis with 100 mg of alteplase (Activase [Genentech, South San Francisco, Calif.]) infused IV over 2 hours,3 and the most widely accepted indication for thrombolytic therapy in these patients is cardiogenic shock from acute PE. However, contraindications prevent many patients from receiving systemic thrombolysis. Even when patients with acute PE are prescreened for absolute contraindications, the rate of major hemorrhage from systemic thrombolytic administration is still about 20%, including a 3% to 5% risk of hemorrhagic stroke.8,25 Furthermore, there may be insufficient time in the acute setting to infuse a full dose of IV thrombolytic. In some instances, in appropriate candidates, it may be desirable to initiate IV tPA while simultaneously activating the interventional team to perform CDT. For example, in selected patients who are in extremis from PE and deemed candidates for any thrombolytic treatment, some clinicians

CHAPTER 86 PERCUTANEOUS INTERVENTIONS FOR ACUTE PULMONARY EMBOLISM 619

A

B

C

D

E

F

G

H

I

K

J

L

may wish to initiate urgent “medical” treatment in the form of IV thrombolytic as a bridge to escalation “surgical” treatment with CDT. When used in this fashion, IV tPA could also be less risky. For instance, the amount of IV thrombolytic could be reduced by at least 50% (from the standard 100 mg tPA dose infused over 2 hours) if catheter

FIGURE 86-4. This 45-year-old woman with lymphoma presented with chest pain and dyspnea but remained hemodynamically stable; the diagnosis was acute submassive pulmonary embolism. A, Left pulmonary angiography showed occlusion of the posterior basal segment of the left lower lung. B, Lateral view of arteriogram showed occlusion of the posterior basal segment. Note patency of superior, anteromedial, and lateral basal segments, with a wedge of the posterior basal segment missing. C, An infusion catheter was carefully advanced into the posterior basal segment artery, and contrast medium injection showed extensive thrombosis. A catheter was wedged, and urokinase infusion was started. D, Follow-up angiography 12 hours into treatment showed restored patency of peripheral branches of the posterior basal segment, with persistent proximal occlusion of the posterior basal segment artery. E-F, Anterior and lateral views of a 24-hour follow-up angiogram showed persistent proximal occlusion and increased volume of a pleural effusion, with further collapse of the superior segment and left upper lung. G-H, Balloon angioplasty and occlusion balloon thrombectomy was performed in the persistently occluded segment of the posterior basal segment resulting in recanalization. I-J, Follow-up left pulmonary angiogram 1 month after treatment showed a patent posterior basal segment. K, Pretreatment lung perfusion scan showed reduced perfusion of the left lower lobe. L, Posttreatment scan showed persistent reduced perfusion of the left lung base due to an underlying pleural effusion.

intervention is initiated promptly, allowing discontinuation of IV tPA within 30 to 60 minutes.26 Catheter-Directed Thrombolysis Under the definition of modern CDT,11 two basic protocols of catheter-directed thrombolysis have emerged for

620 PART 1 VASCULAR INTERVENTIONS · SECTION 10 THORACIC VASCULAR INTERVENTION Right upper lobe

Left upper lobe

Left lower lobe Right intermediate artery

Pulmonary artery trunk

FIGURE 86-5. Schematic drawing of a pulmonary arterial flow model showing vortex formation immediately proximal to the level of obstruction. Note prominent vortex near the occlusion causing most circulating fluid to flow toward the nonoccluded left pulmonary artery. There is minimal fluid contact with the occluding embolus (the balloon). (From Schmitz-Rode T, Kilbinger M, Günther RW. Simulated flow pattern in massive pulmonary embolism: Significance for selective intrapulmonary thrombolysis. Cardiovasc Intervent Radiol 1998;21:199–204.)

treatment of massive and submassive PE, respectively. For massive PE, a catheter-directed bolus of thrombolytic drug is used in conjunction with mechanical clot fragmentation and/or aspiration to achieve central clot debulking. Depending on anticipated bleeding risk, CDT may be performed with either no or low-dose local tPA injection. The goal of these techniques is rapid central clot removal to relieve lifethreatening heart strain and immediately improve pulmonary perfusion. Catheter intervention is important not only for creating an immediate flow channel through the obstruction, but also for exposing a greater surface area of thrombus to the effects of locally infused thrombolytic drug. If thrombolysis is performed without intraclot drug injection, and if the thrombolytic is instead infused proximal to the target embolus (as performed in older studies), there is little added benefit compared to systemic IV infusion.27 SchmitzRode et al. demonstrated with in vitro and in vivo flow studies28 that an obstructing embolus causes proximal vortex formation that prevents a drug infused upstream from making rapid contact with the downstream embolus, and the eddy currents instead cause washout of thrombolytic into the unobstructed pulmonary arteries (Fig. 86-5). These flow studies emphasize the importance of direct intrathrombus injection as an adjunct to embolus fragmentation to achieve rapid and effective catheter-directed thrombolysis.28 Several devices meeting criteria for modern CDT have been used effectively, but the most common technique is rotating pigtail fragmentation, which has been used either alone or in combination with other methods in 70% of patients worldwide receiving CDT.11 Although pigtail clot fragmentation appears to effectively debulk proximal emboli, in some instances it has resulted in distal embolization with PAP elevation, requiring adjunctive aspiration

thrombectomy to complete treatment.29 Aspiration can be performed with virtually any end-hole catheter, such as an 8F JR4 catheter (Cook Medical). Additional clot fragmentation may also be achieved with insertion and inflation of an angioplasty balloon sized below the target arterial diameter (Fig. 86-6). Thus it is important to have adjunctive methods available to use in conjunction with pigtail rotation. The main advantage of the rotating pigtail is its wide availability and low cost relative to the mechanically-driven thrombectomy devices. For treatment of submassive PE, or once massive PE has been downstaged to submassive PE, further mechanical debulking is usually unnecessary, and the protocol consists of careful image-guided infusion catheter placement into thrombosed segments for overnight pharmacologic thrombolysis.5 Once the infusion catheters have been properly positioned and connected to IV pumps, catheter-directed thrombolysis should be initiated using alteplase at a rate of 0.5 mg/h through the drug lumen of each catheter if bilateral catheters are used. If only one catheter has been placed for unilateral treatment, the rate may be increased to 1 mg/h through the single infusion catheter. The recommended total tPA infusion rate should be 1 mg/h. To achieve the infusion dose, the reconstituted drug can be diluted in normal saline solution to yield a concentration of 0.1 mg tPA/mL of solution, and the pump can be set accordingly to deliver the prescribed dose. An alternative to catheterdirected tPA is urokinase infusion. The regimen consists of a bolus infusion of 200,000 to 500,000 IU followed by an infusion of 100,000 IU/h of urokinase for 12 to 36 hours.30 Fibrinogen levels can be monitored, particularly in those patients at greater risk of bleeding or if the infusion will be continued beyond 24 hours. When fibrinogen levels drop below 150 to 200 mg/dL, the infusion should be reduced, discontinued, or alternatively continued with transfusions of fresh frozen plasma if further thrombolysis is desired.5 Catheter-Directed Embolectomy, Fragmentation, and Thrombolysis Contrary to invasive open surgical thrombectomy, which has been associated with high perioperative morbidity and mortality, percutaneous CDT represents a safe, less invasive option for treating patients with acute massive PE.11 As mentioned, the rationale for using percutaneous devices in the pulmonary circulation is rapid central clot debulking to relieve life-threatening heart strain and immediately improve pulmonary perfusion, which can be immediately verified by follow-up angiography and hemodynamic assessments. Percutaneous embolectomy, clot fragmentation, and mechanical thrombolysis also serve to expose a greater surface area of thrombus to the effects of locally infused thrombolytic drug. However, depending on anticipated bleeding risk, CDT may be performed with either no or low-dose local tPA injection, depending on operator preference and risk assessment. Regardless of the decision to use thrombolytic drug, CDT can be used effectively when full-dose thrombolytic therapy is contraindicated or fails to resolve hemodynamic shock.3,5,11 An ideal percutaneous thrombectomy device for treating PE should have these characteristics: • Low profile (≤10F) and long enough to reach the PA from a peripheral venous access site

CHAPTER 86 PERCUTANEOUS INTERVENTIONS FOR ACUTE PULMONARY EMBOLISM 621

A

B

C

D

E

F

G

H

I

J

K

M

L

N

• Adequate flexibility for facile insertion and navigation in the pulmonary circulation • Capable of achieving embolectomy, clot fragmentation, and/or mechanical thrombolysis in a large vessel • Allows concomitant intraclot thrombolytic drug injection if desired

FIGURE 86-6. This 78-year-old woman with chronic obstructive pulmonary disease and history of myocardial infarction 1 month earlier presented to the emergency department with symptoms of deep vein thrombosis, chest pain, shortness of breath, and congestive heart failure. Chest computed tomographic angiography (CTA) was positive for acute pulmonary embolism. Ventilation/perfusion scan showed massive perfusion defects. Patient was transferred to our hospital, and a pulmonary arteriogram was requested for possible catheter-directed treatment. A, Chest CT scan showed large filling defects in both main pulmonary arteries. After hemodynamic stabilization, a pulmonary arteriogram was performed. B, Right pulmonary angiogram showed severe obstruction at the bifurcation of the right main pulmonary artery (PA). There was reduced perfusion of the right upper lobe and marked hypoperfusion of the right lower lobe. C, Left pulmonary angiogram showed a large embolism involving the left lower lobe with marked reduction in peripheral perfusion. Pulmonary artery pressure (PAP) was 39 mmHg. D, Follow-up right pulmonary angiogram after 15 hours of catheter-directed thrombolytic infusion showed marked improvement of the obstruction, with some residual clots involving the right upper and middle lobes. There was interval improvement in right lower lobe perfusion. E, Follow-up left pulmonary angiogram showed progressive occlusion of the left lower lobe likely from distal migration of a proximal thrombus. PAP was 46 mmHg. F, Selective catheterization of the left lower lobe branch through the clot was performed and showed total occlusion of the lower lobe artery. G, A 10-mm balloon catheter was advanced into the lower lobe branches over a wire and used to fragment clot. H, An occlusion balloon was trawled back several times to further fragment the clots. I, An angiogram after mechanical balloon thrombectomy showed improved lower lobe segment patency. A catheter was placed within the partially occluded artery, and thrombolytic infusion was continued. J, Follow-up left pulmonary angiogram after 15 hours of local thrombolytic infusion showed significant improvement in left lung perfusion. K, Final right pulmonary angiogram showed significant improvement in right pulmonary perfusion. L, Final left pulmonary angiogram showed near complete patency of the left PA and peripheral branches. Mean PAP after treatment was 33 mmHg. The patient improved clinically, was weaned off the ventilator, and successfully extubated. M, Pretreatment perfusion scan showed multiple wedge-shaped areas of segmental hypoperfusion. N, Perfusion scan 3 days after catheterdirected thrombolytic therapy showed remarkable improvement in bilateral pulmonary perfusion. After discharge from the intensive care unit, the only complication was acute renal failure that eventually resolved after 5 days, with return of creatinine to normal levels. The patient was discharged home within 1 week after receiving catheterdirected therapy.

• Safe in the central circulation, with low risk of vascular perforation, no widespread distal embolization, and no hemolytic side effects Since no current device possesses all these characteristics, a combination of devices and methods can be used to achieve effective CDT.

622 PART 1 VASCULAR INTERVENTIONS · SECTION 10 THORACIC VASCULAR INTERVENTION

FIGURE 86-7. Photo diagram of the rotating pigtail method most commonly used to treat acute massive pulmonary embolism. (From SchmitzRode T, Janssens U, Duda SH, et al. Massive pulmonary embolism: Percutaneous emergency treatment by pigtail rotation catheter. J Am Coll Cardiol 2000;36:375–80.)

Percutaneous Devices and Methods Rotating Pigtail Clot Fragmentation The most common technique currently used is rotating pigtail fragmentation (Fig. 86-7), which has been used either alone or in combination with other methods in 70% of patients worldwide receiving CDT.11 Although a robust pigtail has been manufactured in Europe specifically for treating PE (Cook Europe, Bjaeverskov, Denmark), virtually any type of pigtail catheter can be used. The pigtail catheter and its distal side holes can also be used to inject local thrombolytic drug directly into the thrombus. Although pigtail clot fragmentation appears to effectively debulk central emboli, in some instances it has resulted in distal embolization with PAP elevation, requiring adjunctive aspiration thrombectomy to complete treatment.29 Balloon Angioplasty for Clot Fragmentation Additional clot fragmentation may also be achieved with insertion and inflation of an angioplasty balloon sized below the target arterial diameter. Specifically, mechanical clot fragmentation has been described using angioplasty balloons between 6 and 16 mm in diameter. When used in conjunction with local pharmacologic thrombolysis for massive PE, these methods have been very successful, with an 87.5% recovery rate, as measured by PAPs, blood O2 values, and clinical outcomes.32,33 Simple Aspiration Thrombectomy Aspiration can be performed with virtually any end-hole catheter, such as an 8F JR4 catheter (Cook Medical) or 10F Pronto catheter (Vascular Solutions, Minneapolis, Minn.). This method works best on central occlusive thrombus and should be used in conjunction with the methods already mentioned. Electronic Aspiration Thrombectomy The Aspirex (Straub Medical, Wangs, Switzerland), has shown promising results for acute PE thrombectomy.34,35 This device works on the principle of a rotating Archimedes screw that resides within a low-profile catheter lumen (Fig. 86-8). The metallic spiral is connected to an electric motor drive and control unit. Electronic activation of the spiral

FIGURE 86-8. Close-up photo of the Aspirex device tip. (Courtesy Straub Medical AG, Wangs, Switzerland.)

coil produces aspiration from the open catheter tip, transporting material down the catheter shaft and into a collecting system. High-Rpm Mechanical Thrombectomy The Helix Clot Buster (ev3/Covidien, Plymouth, Minn.), formerly known as the Amplatz thrombectomy device (ATD), has been used to treat acute PE. The device is a 75or 120-cm-long, 7F reinforced polyurethane catheter with a distal metal tip containing an impeller connected to a drive shaft. The catheter is connected to an air source turbine that generates up to 140,000 rpm at pressures between 30 and 35 psi during operation. Although little data are available on the new version of this device for treatment of PE, data from off-label use of the older 8F version have been published, with use in conjunction with a 10F guide catheter.24 The possibility of hemolytic complications exists, but so far the degree has not been shown to be clinically significant.24 Despite promising results, production of the Helix device is currently on hold by the manufacturer, with possible plans for a product re-release. Rheolytic Thrombectomy The mechanism of rheolytic thrombectomy is a highpressure saline jet in conjunction with aspiration. This creates a Venturi effect that causes fragmentation and aspiration of the clot. The AngioJet (Possis Medical, Minneapolis, Minn.) is a double-lumen system with diameters ranging from 4F to 6F. Although results were promising in early small series of patients with massive PE,36-38 recent metaanalytic data revealed higher procedure-related complications associated with AngioJet rheolytic thrombectomy (ART), including bradyarrhythmia, heart block, hemoglobinuria, renal insufficiency, major hemoptysis, and procedure-related death. Several deaths related to the AngioJet have been recorded in the U.S. Food and Drug Administration’s (FDA’s) MAUDE (Manufacturer and User

CHAPTER 86 PERCUTANEOUS INTERVENTIONS FOR ACUTE PULMONARY EMBOLISM 623 Facility Device Experience) database.39 As a result, the FDA has issued a black-box warning on the device label.40 For all these reasons, the AngioJet device should probably be avoided as the initial mechanical option in CDT protocols for acute massive PE.26,41 Pulmonary Artery Stent Placement Case reports have described the use of metallic stents in critically ill patients with persistent central obstruction from presumed chronic emboli resistant to catheterdirected thrombolysis. The stents have been placed alongside organized PA emboli in which the patients presented with cor pulmonale, profound arterial hypoxemia, and hypotension that did not respond to other therapies.42,43 The paucity of evidence should make use of stents only considered when there is life-threatening PE refractory to CDT (described earlier). Ultrasound-Assisted Catheter-Directed Thrombolysis The EKOS infusion catheter (EKOS Corp., Bothell, Wash.) uses microsonic energy designed to help loosen and separate fibrin to enhance clot permeability while increasing availability of more plasminogen activation receptor sites for tPA. The microsonic energy is also intended to drive the thrombolytic agent deep into the blood clot to accelerate thrombolysis.44 If successful, it has the potential to shorten the duration of infusion and lower the total dose of thrombolytic drug in submassive PE patients.44

CONTROVERSIES In 1988, Verstraete et al. published a study27 comparing the recanalization effects of intrapulmonary versus IV infusion of rtPA and showed that transcatheter intrapulmonary delivery did not offer a significant benefit over the IV route. The major flaw in the study was that intrapulmonary drug delivery was performed proximal to the target clot, without intraclot thrombolytic injection and without mechanical intervention. As noted earlier, subsequent in vitro and in vivo flow studies28 confirmed that an obstructing embolus causes proximal vortex formation that prevents a drug infused upstream (even via catheter) from making rapid contact with the downstream embolus, and the eddy currents instead cause washout of thrombolytic into the unobstructed pulmonary arteries (see Fig. 86-5). This emphasizes the importance of direct intrathrombus injection as an adjunct to embolus fragmentation to achieve rapid and effective catheter-directed thrombolysis.28 Despite the complications associated with AngioJet rheolytic thrombectomy and the FDA black box warning, some interventionalists continue to use this device to treat acute PE.26 A meta-analysis of data on CDT11 revealed that the highest complication rates occurred in patients treated with ART, including a 40% rate of minor complications and 28% rate of major complications that included procedure-related death.11 Cumulative data indicate that most modern CDT (89%) has been performed worldwide with a high degree of safety and efficacy without using ART.

OUTCOMES In a systematic review and meta-analysis of 594 patients with acute massive PE treated with modern CDT, clinical

success was achieved in 86.5% (Fig. 86-9), where success was defined as stabilization of hemodynamics, resolution of hypoxia, and survival to hospital discharge.11 In the same study, 96% of patients received CDT as the first adjunct to heparin, with no prior systemic tPA infusion, and 33% of cases were initiated with mechanical treatment alone without local thrombolytic infusion.11 The data were derived across 18 countries from 35 studies—six prospective, 29 retrospective—and pooled results were similar in prospective versus retrospective studies, with no statistically significant difference. The pooled frequency of success was higher in studies in which at least 80% of participants received local thrombolytic therapy during the procedure (91.2% vs. 82.8%). The pooled frequency of success was also higher in studies in which at least 80% of participants received extended local thrombolytic therapy for treatment of residual submassive PE (89.2% vs. 84.2%).11 Modern CDT continues to be used worldwide. In 2011, another large-scale study of 111 PE patients confirmed that modern catheter-directed thrombolysis with mechanical fragmentation achieves rapid normalization of the pulmonary pressure and is a safe and effective method for treating acute massive PE.30 Patients in extremis from massive PE require emergent treatment escalation beyond anticoagulation, and if IV tPA is contraindicated or there is insufficient time for full-dose tPA, CDT may be the only viable treatment option. Indeed, at experienced centers, the use of modern CDT has proven to be a lifesaving treatment in patients dying from acute massive PE.11,30

Submassive Pulmonary Embolism There is growing evidence that aggressive treatment of submassive PE is beneficial. The Management Strategies and Prognosis of Pulmonary Embolism Trial-3 (MAPPET-3) randomized 256 patients with submassive PE to receive 100 mg of IV tPA over a 2-hour period, followed by unfractionated heparin infusion, versus placebo plus heparin anticoagulation.45 Compared with heparin anticoagulation alone, thrombolysis resulted in a significant reduction in the primary study endpoint of in-hospital death or clinical deterioration that required escalation of therapy (defined as catecholamine infusion, rescue thrombolysis, mechanical ventilation, cardiopulmonary resuscitation, or emergency surgical embolectomy).45 The difference was largely attributable to a higher frequency of open-label thrombolysis (breaking randomized trial protocol to offer medically necessary thrombolysis) due to clinical deterioration as determined by the treating clinician.45 In a prospective study of 200 patients with submassive PE,19 echocardiography was performed at the time of diagnosis and after 6 months to determine the frequency of pulmonary hypertension between two groups—one group treated with heparin and another group treated with IV tPA and heparin. The median decrease in PA systolic pressure was only 2 mmHg in patients treated with heparin alone, compared with 22 mmHg in those treated with tPA plus heparin.19 At 6 months, the PA systolic pressure increased in 27% of patients who had received heparin alone, and nearly half of those patients were moderately symptomatic.19 These data suggest that thrombolytic therapy may reduce the likelihood of developing chronic thromboembolic pulmonary hypertension.19

624 PART 1 VASCULAR INTERVENTIONS · SECTION 10 THORACIC VASCULAR INTERVENTION Schmitz-Rode et al. 1998 (48)

0.80 (0.44, 0.97)

Schmitz-Rode et al. 2000 (49)

0.80 (0.56, 0.94)

Muller-Hulsbeck et al. 2001 (50)

1.00 (0.66, 1.00)

Prokubovsky et al. 2003 (51)

0.70 (0.46, 0.88)

Tajima et al. 2004 (52)

1.00 (0.86, 1.00)

Barbosa et al. 2008 (53)

0.90 (0.55, 1.00)

Brady et al. 1991 (54)

1.00 (0.29, 1.00)

Rafique et al. 1992 (55)

1.00 (0.48, 1.00)

Uflacker et al. 1996 (24)

0.60 (0.15, 0.95)

Fava et al. 1997 (32)

0.88 (0.62, 0.98)

Stock et al. 1997 (56)

1.00 (0.48, 1.00)

Basche et al. 1997 (57)

0.80 (0.52, 0.96)

Hiramatsu et al. 1999 (58)

0.88 (0.47, 1.00)

Wong et al. 1999 (59)

0.75 (0.19, 0.99)

Murphy et al. 1999 (60)

1.00 (0.40, 1.00)

Voigtlander 1999 (61)

0.60 (0.15, 0.95)

Fava et al. 2000 (62)

0.91 (0.59, 1.00)

Egge et al. 2002 (63)

1.00 (0.29, 1.00)

De Gregorio et al. 2002 (33)

0.95 (0.86, 0.99)

Zeni et al. 2003 (36)

0.88 (0.62, 0.98)

Reekers et al. 2003 (64)

0.86 (0.42, 1.00)

Tajima et al. 2004 (65)

1.00 (0.78, 1.00)

Fava et al. 2005 (66)

0.86 (0.42, 1.00)

Siablis et al. 2005 (37)

0.83 (0.36, 1.00)

Yoshida et al. 2006 (67)

0.88 (0.47, 1.00)

Li J-J et al. 2006 (68)

1.00 (0.78, 1.00)

Pieri and Agresti 2007 (69)

0.84 (0.78, 0.89)

Chauhan et al. 2007 (70)

0.67 (0.22, 0.96)

Krajina et al 2007 (71)

0.40 (0.05, 0.85)

Yang et al. 2007 (72)

0.95 (0.74, 1.00)

Margheri et al. 2008 (73)

0.85 (0.62, 0.97)

Vecchio et al 2008 (38)

0.62 (0.32, 0.86)

Chen et al 2008 (74)

1.00 (0.87, 1.00)

Eid-Lidt et al. 2008 (35)

0.89 (0.65, 0.99)

Kuo et al. 2008 (31)

0.83 (0.52, 0.98)

Combined 0.0

0.865 (0.82, 0.90) 0.2

0.4

0.6

0.8

1.0

Proportion (95% confidence interval)

FIGURE 86-9. Forest plot shows clinical success rates from CDT and confidence intervals (CIs) from reported studies encompassing 594 patients with acute massive pulmonary embolism. Percentage clinical success is denoted along x-axis. Extended lines represent 95% CIs. Squares are proportional to study weight. Width of diamond corresponds to 95% CI for pooled clinical success rate of 86.5%. (From Kuo WT, Gould MK, Louie JD, et al. Catheter-directed therapy for the treatment of massive pulmonary embolism: Systematic review and meta-analysis of modern techniques. J Vasc Interv Radiol 2009;20:1431–40.)

CHAPTER 86 PERCUTANEOUS INTERVENTIONS FOR ACUTE PULMONARY EMBOLISM 625 However, for patients who are not good candidates for systemic tPA, the next logical step to consider is catheterdirected intervention. That is why the incorporation of a CDT protocol with targeted drug delivery and a lower overall thrombolytic dose could further improve outcomes while reducing hemorrhagic risk in the submassive PE group.5 Indeed, when low-dose (≤30 mg) local tPA was administered to acute massive PE patients—a group at higher risk for bleeding than submassive PE patients12— there were no major hemorrhagic complications.10,11 Furthermore, when patients with massive PE are downstaged to submassive PE via initial catheter-directed treatment of central clot, the overall clinical success was higher when these patients also received extended local thrombolytic therapy.11 Such results strongly suggest that use of endovascular treatment in the form of local thrombolytic infusion appears to be a promising option for reducing both acute and chronic complications from PE, while avoiding the bleeding risks associated with full-dose systemic thrombolysis.

COMPLICATIONS From the global meta-analysis on CDT,11 the pooled risk of major complications was only 2.4%. Among 594 patients, major procedural complications occurred in 25 patients, including 11 groin hematomas (requiring transfusion), five noncerebral hemorrhages (sites unspecified, requiring transfusion), two cases of massive hemoptysis requiring transfusion, one renal failure requiring hemodialysis, one cardiac tamponade treated with surgical repair, one death from bradyarrhythmia and apnea, one death from widespread distal embolization, one death associated with cerebrovascular hemorrhage, and two procedure-related deaths (mechanism unspecified). The highest complication rates occurred in the 68 patients who underwent CDT with the rheolytic AngioJet device, including 27 minor complications (40%) and 19 major complications (28%).11 The complications of bradyarrhythmia, heart block, hemoglobinuria, temporary renal insufficiency, one minor hemoptysis, one major hemoptysis, five major hemorrhages, and five procedure-related deaths were all associated with the AngioJet device. Interestingly, 76% of the recorded major complications (19/25) in the study were directly attributed to ART, despite the fact that it was used in only a small percentage (11%) of the 594 patients evaluated.11 Conversely, the data indicated that most modern CDT (89%) was performed worldwide with a high degree of safety and efficacy without using ART. Overall, compared to the high rate of major hemorrhagic complications from systemic tPA of 20%,8,25 the rate of major complications from modern CDT has proven to be only 2.4%.11 Since most of these complications were attributed to ART, elimination of AngioJet from the CDT protocol could further improve the overall safety of modern CDT.

If VTE is associated with a major irreversible risk factor such as cancer, these patients have at least a 15% risk of recurrence during the first year after stopping anticoagulation.46 Consequently, patients with active cancer and a first episode of VTE usually receive treatment indefinitely.10 Conversely, if VTE is provoked by a major reversible risk factor such as recent surgery, the risk of recurrence is about 3% in the first year if anticoagulation is discontinued after 3 months.46 Between these two extremes are patients who have suffered acute VTE associated with a minor reversible risk factor (e.g., estrogen therapy or soft-tissue leg injury) and those who have had an unprovoked or idiopathic VTE.47 For patients with a minor reversible risk factor, the risk of recurrence is about 5% in the first year after stopping anticoagulant therapy.47 This is considered low enough to justify stopping anticoagulant therapy at the end of 3 months.10 However, an unprovoked proximal DVT or PE has a higher risk of recurrence (about 10% in the first year after stopping therapy).46 Continuing anticoagulant therapy beyond 3 months confers a greater than 90% risk reduction for preventing recurrence among these patients; however, if anticoagulants are subsequently stopped after 6 or 12 months of treatment, the risk of recurrence appears to be the same as if anticoagulants had been stopped after 3 months.10 This high risk of recurrence is indirect evidence that patients with a first unprovoked proximal DVT or acute PE should receive anticoagulant therapy indefinitely. However, after all the major risks and benefits of long-term anticoagulation therapy have been explained, patient preference should also influence the decision.46 Because of all these complexities, consultation with a hematologist or physician experienced in treating VTE is essential in determining the optimal anticoagulation regimen. If there are contraindications to immediate anticoagulation for acute VTE, IVC filtration may be indicated. For those patients who have suffered near death from massive PE, IVC filter placement should be considered in addition to anticoagulation. For those diagnosed with submassive PE, it is unclear whether additional filter placement is necessary in addition to anticoagulation. If caval filtration is desired, a retrievable filter is often preferable, since it has the option to be removed. When lifelong filtration is not intended, such patients should be followed closely and scheduled for timely filter removal (e.g., once they are stabilized on therapeutic anticoagulation). Once filtration is no longer indicated, successful IVC filter removal can spare patients the potential risks associated with long-term filter implantation. KEY POINTS •

Rapid risk stratification by identifying acute massive and acute submassive pulmonary embolism (PE) patients is essential in determining appropriate treatment escalation beyond anticoagulation.

•

For patients with less severe or submassive PE (defined by right heart strain), endovascular treatment in the form of local thrombolytic infusion appears to be a promising option for reducing both acute and chronic complications from PE while avoiding the bleeding risks associated with full-dose systemic thrombolysis.

•

For patients in extremis from massive PE (defined by hemodynamic shock), emergent treatment escalation is necessary

POSTPROCEDURAL ANTICOAGULATION AND FOLLOW-UP CARE Therapeutic anticoagulation is the primary medical treatment for all patients diagnosed with acute VTE, and it should be prescribed for at least 3 to 6 months or indefinitely, depending on careful assessment of several factors.10

626 PART 1 VASCULAR INTERVENTIONS · SECTION 10 THORACIC VASCULAR INTERVENTION in the form of catheter-directed therapy (CDT) for rapid debulking of central obstructive emboli, especially if systemic tissue plasminogen activator (tPA) is contraindicated. •

CDT may be the only viable option for rapid restoration of pulmonary arterial flow if intravenous systemic tPA is contraindicated or has already failed to resolve hemodynamic shock.

•

Rheolytic thrombectomy with the AngioJet device has been associated with a high rate of major complications, including procedure-related death, so to improve the overall safety of modern CDT, it should probably be avoided in future protocols.

•

The major complication rate from CDT is only 2.4%, versus the 20% rate of major hemorrhage associated with systemic tPA infusion.

•

At experienced centers, the use of modern CDT has proven to be a lifesaving treatment for acute massive PE, with an overall clinical success rate of 86.5%.

•

Following intervention, therapeutic anticoagulation should be continued as the ongoing medical treatment, and duration of therapy must depend on careful assessment of several risk factors.

▶ SUGGESTED READINGS Konstantinides S, Geibel A, Heusel G, et al. Heparin plus alteplase compared with heparin alone in patients with submassive pulmonary embolism. N Engl J Med 2002;347:1143–50. Kuo WT, Gould MK, Louie JD, et al. Catheter-directed therapy for the treatment of massive pulmonary embolism: systematic review and meta-analysis of modern techniques. J Vasc Interv Radiol 2009;20: 1431–40. Kuo WT. Endovascular therapy for acute pulmonary embolism. J Vasc Interv Radiol 2012;23:167–79. Piazza G, Goldhaber SZ. Management of submassive pulmonary embolism. Circulation 2010;122:1124–9.

The complete reference list is available online at www.expertconsult.com.

CHAPTER 86 PERCUTANEOUS INTERVENTIONS FOR ACUTE PULMONARY EMBOLISM626.e1 REFERENCES 1. Olin JW. Pulmonary embolism. Rev Cardiovasc Med 2002;3:S68–S75. 2. Heit JA, Cohen AT, Anderson FA. Estimated annual number of incident and recurrent, non-fatal and fatal venous thromboembolism (VTE) events in the U.S. Blood 2005;106:267a. 3. Tapson VF. Acute Pulmonary Embolism. N Engl J 2008;358:1037–52. 4. Pulido T, Aranda A, Zevallos MA, et al. Pulmonary embolism as a cause of death in patients with heart disease: an autopsy study. Chest 2006;129:1282–7. 5. Kuo WT. Endovascular therapy for acute pulmonary embolism. J Vasc Interv Radiol 2012;23:167–79. 6. Dalen JE, Alper JS. Natural history of pulmonary embolism. Prog Cardiovasc Dis 1975;17:259–70. 7. Barritt DW, Jordan SE. Anticoagulant drugs in the treatment of pulmonary embolism: a controlled trial. Lancet 1960;1:1309–12. 8. Goldhaber SZ, Visani L, De Rosa M. Acute pulmonary embolism: clinical outcomes in the International Cooperative Pulmonary Embolism Registry (ICOPER). Lancet 1999;353:1386–9. 9. Wood KE. Major pulmonary embolism: review of a pathophysiologic approach to the golden hour of hemodynamically significant pulmonary embolism. Chest 2002;121:877–905. 10. Kearon C, Akl EA, Agnelli G, et al. Antithrombotic therapy for VTE disease: American College of Chest Physicians evidencebased clinical practice guidelines (9th Edition). Chest 2012;141: e466–e470. 11. Kuo WT, Gould MK, Louie JD, et al. Catheter-directed therapy for the treatment of massive pulmonary embolism: systematic review and meta-analysis of modern techniques. J Vasc Interv Radiol 2009;20: 1431–40. 12. Moores LK, Jackson WL, Shorr AF, Jackson JL. Meta-analysis: outcomes in patients with suspected pulmonary embolism managed with computed tomographic pulmonary angiography. Ann Intern Med 2004;141:866–74. 13. Goldhaber SZ, Haire WD, Feldstein ML, et al. Alteplase versus heparin in acute pulmonary embolism: randomised trial assessing rightventricular function and pulmonary perfusion. Lancet 1993;341: 507–11. 14. Tapson VF, Davidson CJ, Kisslo KB, Stack RS. Rapid visualization of massive pulmonary emboli utilizing intravascular ultrasound. Chest 1994;105:888–90. 15. Goldhaber SZ. Echocardiography in the management of pulmonary embolism. Ann Intern Med 2002;136:691–700. 16. Fremont B, Pacouret G, Jacobi D, et al. Prognostic value of echocardiographic right/left ventricular end-diastolic diameter ratio in patients with acute pulmonary embolism: results from a monocenter registry of 1,416 patients. Chest 2008;133. 17. Schoepf UJ, Kucher N, Kipfmueller F, et al. Right ventricular enlargement on chest computed tomography: a predictor of early death in acute pulmonary embolism. Circulation 2004;110:3276–80. 18. van der Meer RW, Pattynama PM, van Strijen MJ, et al. Right ventricular dysfunction and pulmonary obstruction index at helical CT: prediction of clinical outcome during 3-month follow-up in patients with acute pulmonary embolism. Radiology 2005;235:798–803. 19. Kline JA, Steuerwald MT, Marchick MR, et al. Prospective evaluation of right ventricular function and functional status 6 months after acute submassive pulmonary embolism: frequency of persistent or subsequent elevation in estimated pulmonary artery pressure. Chest 2009;136:1202–10. 20. Pruszczyk P, Bochowicz A, Torbicki A, et al. Cardiac troponin T monitoring identifies high-risk group of normotensive patients with acute pulmonary embolism. Chest 2003;123:1947–52. 21. Klok FA, Mos IC, Huisman MV, et al. Brain-type natriuretic peptide levels in the prediction of adverse outcome in patients with pulmonary

embolism: a systematic review and metaanalysis. Am J Respir Crit Care Med 2008;178:425–30. 22. Jimenez D, Uresandi F, Otero R, et al. Troponin-based risk stratification of patients with acute nonmassive pulmonary embolism: systematic review and metaanalysis. Chest 2009;136:974–82. 23. Piazza G, Goldhaber SZ. Management of submassive pulmonary embolism. Circulation 2010;122:1124–9. 24. Uflacker R, Strange C, Vujic I. Massive pulmonary embolism: preliminary results of treatment with the Amplatz thrombectomy device. J Vasc Interv Radiol 1996;7:519–28. 25. Fiumara K, Kucher N, Fanikos J, Goldhaber SZ. Predictors of major hemorrhage following fibrinolysis for acute pulmonary embolism. Am J Cardiol 2006;97:127–9. 26. Kuo WT, Hofmann LV. Optimizing endovascular therapy for acute PE: primum non nocere. J Vasc Interv Radiol 2010;21:1776–7. 27. Verstraete M, Miller GAH, Bounameaux H, et al. Intravenous and intrapulmonary recombinant tissue-type plasminogen activator in the treatment of acute massive pulmonary embolism. Circulation 1988; 77:353–60. 28. Schmitz-Rode T, Kilbinger M, Günther RW. Simulated flow pattern in massive pulmonary embolism: significance for selective intrapulmonary thrombolysis. Cardiovasc Intervent Radiol 1998;21:199–204. 29. Nakazawa K, Tajima H, Murata S, et al. Catheter fragmentation of acute massive pulmonary thromboembolism: distal embolisation and pulmonary arterial pressure elevation. B J Radiol 2008;81:848–54. 30. De Gregorio MA, Laborda A, De Blas I, et al. Endovascular treatment of a haemodynamically unstable massive pulmonary embolism using fibrinolysis and fragmentation. Experience with 111 patients in a single centre. Why don’t we follow ACCP recommendations? Arch Bronconeumol 2011;47:17–24. 31. Kuo WT, Van den Bosch MA, Hofmann LV, et al. Catheter-directed embolectomy, fragmentation, and thrombolysis for the treatment of massive pulmonary embolism after failure of systemic thrombolysis. Chest 2008;134:250–4. 32. Fava M, Loyola S, Huete I. Mechanical fragmentation and pharmacologic thrombolysis in massive pulmonary embolism. J Vasc Interv Radiol 1997;8:261–6. 33. De Gregorio MA, Gimeno MJ, Mainar A, et al. Mechanical and enzymatic thrombolysis for massive pulmonary embolism. J Vasc Interv Radiol 2002;13:163–9. 34. Kucher N, Windecker S, Banz Y, et al. Percutaneous catheter thrombectomy device for acute pulmonary embolism: in vitro and in vivo testing. Radiology 2005;236:852–8. 35. Eid-Lidt G, Gaspar J, Sandoval J, et al. Combined clot fragmentation and aspiration in patients with acute pulmonary embolism. Chest 2008;134:54–60. 36. Zeni PT, Blank BG, Peeler DW. Use of rheolytic thrombectomy in treatment of acute massive pulmonary embolism. J Vasc Interv Radiol 2003;14:1511–5. 37. Siablis D, Karnabatidis D, Katsanos K, et al. AngioJet rheolytic thrombectomy versus local intrapulmonary thrombolysis in massive pulmonary embolism: a retrospective data analysis. J Endovasc Ther 2005; 12:206–14. 38. Vecchio S, Vittori G, Chechi T, et al. Percutaneous rheolytic thrombectomy with AngioJet for pulmonary embolism: methods and results in the experience of a high-volume center. It J Practice Cardiol 2008;9:355–63. 39. http://www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfMAUDE/ search.CFM. (Accessed December 2011). 40. AngioJet Xpeedior [product insert]. Minneapolis, MN: Possis Medical; 2008. 41. Kuo WTSD, Hofmann LV, Goldhaber SZ. Catheter-directed intervention for acute pulmonary embolism: a shining saber. Chest 2007;133: 317–8.

626.e2PART 1 VASCULAR INTERVENTIONS · SECTION 10 THORACIC VASCULAR INTERVENTION 42. Haskal ZJ, Soulen MC, Huetti EA, et al. Life-threatening pulmonary emboli and cor pulmonale: treatment with percutaneous pulmonary artery stent placement. Radiology 1994;191:473–5. 43. Koizumi J, Kusano S, Akima T, et al. Emergent Z stent placement for treatment of cor pulmonale due to pulmonary emboli after failed lytic treatment: technical considerations. Cardiovasc Interv Radiol 1998; 21:254–5. 44. Chamsuddin A, Nazzal L, Kang B, et al. Catheter-directed thrombolysis with the EndoWave System in the treatment of acute massive pulmonary embolism: a retrospective multicenter case series. J Vasc Interv Radiol 2008;19:372–6. 45. Konstantinides S, Geibel A, Heusel G, et al. Heparin plus alteplase compared with heparin alone in patients with submassive pulmonary embolism. N Engl J Med 2002;347:1143–50. 46. Kearon C. Stopping anticoagulant therapy after an unprovoked venous thromboembolism. CMAJ 2008;179:401–2. 47. Baglin T, Luddington R, Brown K, et al. Incidence of recurrent venous thromboembolism in relation to clinical and thrombophilic risk factors: prospective cohort study. Lancet 2003;362:523–6. 48. Schmitz-Rode T, Janssens U, Schild HH, et al. Fragmentation of massive pulmonary embolism using a pigtail rotation catheter. Chest 1998;114:1427–36. 49. Schmitz-Rode T, Janssens U, Duda SH, et al. Massive pulmonary embolism: percutaneous emergency treatment by pigtail rotation catheter. J Am Coll Cardiol 2000;36:375–80. 50. Muller-Hulsbeck S, Brossmann J, Jahnke T, et al. Mechanical thrombectomy of major and massive pulmonary embolism with use of the Amplatz thrombectomy device. Invest Radiol 2001;36:317–22. 51. Prokubovsky VI, Kapranov SA, Bobrov BY. Endovascular rotary fragmentation in the treatment of massive pulmonary thromboembolism. Angiol Sosud Khir 2003;9:31–9. 52. Tajima H, Murata S, Kumazaki T, et al. Hybrid treatment of acute massive pulmonary thromboembolism: mechanical fragmentation with a modified rotating pigtail catheter, local fibrinolytic therapy, and clot aspiration followed by systemic fibrinolytic therapy. Am J Roentgenol 2004;183:589–95. 53. Barbosa MA, Oliveira DC, Barbosa AT, et al. Treatment of massive pulmonary embolism by percutaneous fragmentation of the thrombus. Arq Bras Cardiol 2008;88:279–84. 54. Brady AJB, Crake T. Percutaneous catheter fragmentation and distal dispersion of proximal pulmonary embolus. Lancet 1991;338:1186–9. 55. Rafique M, Middlemost S, Skoularigis J, Sareli P. Simultaneous mechanical clot fragmentation and pharmacologic thrombolysis in acute massive pulmonary embolism. Am J Cardiol 1992;69:427–30. 56. Stock KW, Jacob AL, Schnabel KJ, et al. Massive pulmonary embolism: treatment with thrombus fragmentation and local fibrinolysis with recombinant human-tissue plasminogen activator. Cardiovasc Intervent Radiol 1997;20:364–8. 57. Basche S, Oltmanns G. Thrombus fragmentation in massive pulmonary embolism. Die Medizinische Welt 1997;48:325–7 58. Hiramatsu S, Ogihara A, Kitano Y, et al. Clinical outcome of catheter fragmentation and aspiration therapy in patients with acute pulmonary embolism. J Cardiol 1999;34:71–8

59. Wong PS, Singh SP, Watson RD, Lip GY. Management of pulmonary thrombo-embolism using catheter manipulation: a report of four cases and review of the literature. Postgrad Med J 1999;75:737–41. 60. Murphy JM, Mulvihill N, Mulcahy D, et al. Percutaneous catheter and guidewire fragmentation with local administration of recombinant tissue plasminogen activator as a treatment for massive pulmonary embolism. Eur Radiol 1999;9:959–64. 61. Voigtlander T, Rupprecht HJ, Nowak B, et al. Clinical application of a new rheolytic thrombectomy catheter system for massive pulmonary embolism. Catheter Cardiovasc Interv 1999;47:91–6. 62. Fava M, Loyola S, Huete I. Massive pulmonary embolism: treatment with the hydrolyser thrombectomy catheter. J Vasc Interv Radiol 2000;11:1159–64. 63. Egge J, Berentsen S, Storesund B, et al. Treatment of massive pulmonary embolism with local thrombolysis. Tidsskr Nor Laegeforen 2002;122:2263–6. 64. Reekers JA, Baarslag HJ, Koolen MGJ, et al. Mechanical thrombectomy for early treatment of massive pulmonary embolism. Cardiovasc Intervent Radiol 2003;26:246–50. 65. Tajima H, Murata S, Kumazaki T, et al. Hybrid treatment of acute massive pulmonary thromboembolism: mechanical fragmentation with a modified rotating pigtail catheter, local fibrinolytic therapy, and clot aspiration followed by systemic fibrinolytic therapy. Radiat Med 2004;22:168–72. 66. Fava M, Loyola S, Bertoni H, Dougnac A. Massive pulmonary embolism: percutaneous mechanical thrombectomy during cardiopulmonary resuscitation. J Vasc Interv Radiol 2005;16:119–23. 67. Yoshida M, Inoue I, Kawagoe T, et al. Novel percutaneous catheter thrombectomy in acute massive pulmonary embolism: rotational bidirectional thrombectomy (ROBOT). Catheter Cardiovasc Interv 2006;68:112–7. 68. Li J, Zhai R, Dai D, et al. Interventional mechanical thrombectomy procedure in treating acute massive pulmonary infarction. J Intervent Radiol 2006;15:336–8 69. Pieri S, Agresti P. Hybrid treatment with angiographic catheter in massive pulmonary embolism: mechanical fragmentation and fibrinolysis. Radiol Med 2007;112:837–49. 70. Chauhan MS, Kawamura A. Percutaneous rheolytic thrombectomy for large pulmonary embolism: a promising treatment option. Catheter Cardiovasc Interv 2007;70:121–8. 71. Krajina A, Lojik M, Chovanec V, et al. Percutaneous mechanical fragmentation of emboli in the pulmonary artery. Ces Radiol 2007; 61:162–6. 72. Yang Z, Shi H, Li L, Liu S. System thrombolysis combined with percutaneous catheter fragmentation and thrombectomy in acute massive pulmonary embolism. Chin J Radiol 2007;41:1241–4. 73. Margheri M, Vittori G, Vecchio S, et al. Early and long-term clinical results of AngioJet rheolytic thrombectomy in patients with acute pulmonary embolism. Am J Cardiol 2008;101:252–8. 74. Chen L, Gu J, Lou W, et al. Interventional mechanical thrombectomy for acute pulmonary embolism. J Intervent Radiol 2008.

SECTION ELEVEN

NEUROVASCULAR AND HEAD AND NECK INTERVENTION CHAPTER

87

Craniocervical Vascular Anatomy

Diego San Millán Ruiz, Bradley P. Barnett, and Philippe Gailloud

CRANIOCERVICAL ARTERIES Cerebral blood flow is provided by the paired internal carotid arteries (ICAs) and vertebral arteries (VAs). The intradural branches of the ICA supply the anterior cerebral circulation (i.e., cerebral hemispheres, including basal ganglia) and the orbit; these branches include the ophthalmic artery (OA), posterior communicating artery (PComA), anterior choroidal artery (AChoA), and anterior and middle cerebral arteries (ACA, MCA). The vertebrobasilar system, formed by fusion of the VAs into the basilar artery (BA), supplies the posterior cerebral circulation, including the brainstem, cerebellum, and posterior aspect of the cerebral hemispheres via the posterior cerebral arteries (PCAs). Connections between the anterior and posterior circulations occur through the circle of Willis (CW) and its branches or through occasionally persistent embryologic carotid-vertebral and carotid-basilar connections (i.e., persistent trigeminal, hypoglossal, and proatlantal arteries). Cortical (or leptomeningeal) anastomoses also exist over the surface of the cerebral convexity and connect the ACA, MCA, and PCA. These anastomoses may provide important collateral pathways in patients with steno-occlusive disorders such as atheromatous disease or moyamoya syndrome. Finally, so-called watershed areas exist between the terminal vascular territories of the principal cerebral arteries. Watershed areas represent zones at risk for ischemic injury in patients with hemodynamically significant stenosis of the ICA, low cardiac output, or severe prolonged hypotension. The anatomy of the intracranial arterial system is discussed in detail in the following sections, which have been organized into the anterior and posterior circulations, the CW, and major branches of the anterior and posterior circulations (Table 87-1).

Anterior Circulation Common Carotid Artery The left and right common carotid arteries (CCAs) have a different origin pattern. The right CCA generally arises from the brachiocephalic trunk (also known as the innominate artery). When the brachiocephalic trunk is absent—for instance, in the presence of an aberrant right subclavian artery (SubcA), or arteria lusoria—the right CCA comes directly from the aortic arch. Rarely, the right and left CCAs originate from a common trunk called a bicarotid trunk, a pattern usually observed in association with an aberrant right SubcA. The left CCA generally originates directly

from the aortic arch, in between the brachiocephalic trunk and left SubcA. The left CCA is therefore usually longer than the right. A common origin for the left CCA and the brachiocephalic trunk is observed in approximately 13% of cases.1 This variation is commonly referred to as a “bovine arch,” a misnomer because the typical bovine arch consists of a single aortic trunk that provides all the supraaortic branches.2 In about 9% of cases, the left CCA arises from the brachiocephalic trunk itself.1 Rarely, the left CCA may share a common origin with the left SubcA and form a left brachiocephalic artery.3 Although the CCA does not normally have branches, variations may include variant origins of the superior thyroid, ascending pharyngeal, and vertebral arteries (Fig. 87-1). In rare instances, bronchial arteries can arise from the proximal portion of the CCA.4 Carotid Bifurcation The carotid bifurcation is most commonly located at the level of the superior border of the thyroid cartilage between the C3 and C5 vertebral levels, but extreme positions ranging from C1 to T4 may be encountered. Intrathoracic bifurcations are rare but represent a potential pitfall during angiography, when the ICA or external carotid artery (ECA) may be unintentionally catheterized because of their low origin. Intrathoracic carotid bifurcations may be more frequent in patients with Klippel-Feil anomaly.5 Both carotid bifurcations are generally found at the same level (28%) or within one vertebral level (65%).6 Immediately distal to the bifurcation, the ICA is lateral and posterior to the ECA. The ICA then bends medially to reach the outer opening of the carotid canal at the skull base, whereas the ECA ascends laterally toward the parotid gland, where it divides into its terminal branches. The ICA may infrequently arise medial (4%) or posteromedial (8%) to the ECA.7 The carotid bulb is a normal widening of the carotid bifurcation that extends from the distal CCA to the proximal ICA. It may also involve the proximal ECA. External Carotid Artery The ECA divides into numerous branches along its course between the carotid bifurcation and its termination in the maxillary and superficial temporal arteries in the parotid gland region. The classic ECA description includes eight main branches (described later) that supply the superior aspect of the thyroid gland, the viscerocranium (soft and bony structures of the face), the scalp, and a large part of the neurocranium (skull and meninges) (Fig. 87-2). Because of the existence of an extensive collateral network, selective 627

628 PART 1 VASCULAR INTERVENTIONS · SECTION 11 NEUROVASCULAR AND HEAD AND NECK INTERVENTION TABLE 87-1. Abbreviations Acronym

Expanded

ACA

Anterior cerebral artery

AChoA

Anterior choroidal artery

AComA

Anterior communicating artery

AG

Arachnoid granulation

AICA

Anterior inferior cerebellar artery

BA

Basilar artery

BVR

Basal vein of Rosenthal

CCA

Common carotid artery

CS

Cavernous sinus

CW

Circle of Willis

ECA

External carotid artery

ICA

Internal carotid artery

ICV

Internal cerebral vein

IJV

Internal jugular vein

IVVP

Internal vertebral venous plexus

MCA

Middle cerebral artery

OA

Ophthalmic artery

PCA

Posterior cerebral artery

PComA

Posterior communicating artery

PICA

Posterior inferior cerebellar artery

PTA

Persistent trigeminal artery

SubcA

Subclavian artery

SCA

Superior cerebellar artery

SMCV

Superficial middle cerebral vein

SSS

Superior sagittal sinus

VA

Vertebral artery

VAVP

Vertebral artery venous plexus

VG

Vein of Galen

WS

Vertebral venous system

embolization with occlusion of branches of the ECA or the ECA itself is generally well tolerated. Connections between the ECA and ICA or VA are often referred to as “dangerous anastomoses.” These anastomoses may play an important role as sources of collateral supply in patients with stenoocclusive disorders, but they also represent important pitfalls during embolization procedures within the ECA territory. These dangerous collateral pathways include (1) anastomoses between external carotid branches (mainly from the facial, maxillary, and superficial temporal arteries) and branches of the distal ICA, including the OA, inferiorlateral trunk, and meningohypophyseal trunk, and (2) anastomoses between external carotid branches (mainly the occipital artery) and muscular branches of the VA. The eight branches of the ECA are as follows: 1. The superior thyroid artery principally supplies the superior pole of the ipsilateral thyroid lobe. The superior and inferior thyroid arteries are connected through a bilateral anastomotic network that can play an important role in collateral cerebral supply.8 The superior

FIGURE 87-1. Digital subtraction angiogram, right innominate injection, anteroposterior view showing a right vertebral artery (VA) (arrow) originating from right common carotid artery (CCA). This variant is generally associated with an aberrant right subclavian artery (arteria lusoria). Proximal origin of VA makes it easy to overlook during selective angiography of CCA. (©2012, Philippe Gailloud, MD.)

thyroid artery also vascularizes the larynx (superior and inferior laryngeal arteries), as well as the infrahyoid and sternocleidomastoid muscles. 2. The lingual artery arises between the superior thyroid and facial arteries, although in about 20% of cases it forms a common trunk with the facial artery (linguofacial trunk). The lingual artery gives off branches for the hyoid bone, the muscles forming the floor of the mouth, the sublingual gland (sublingual branch), and the muscles of the tongue (deep lingual artery). 3. The facial artery courses over the body of the mandible toward the angle of the mouth and then ascends toward the medial epicanthus (angular artery), where it anastomoses with branches of the OA. The facial artery has numerous branches including the ascending palatine artery (for the tonsils, soft palate, and oropharynx), artery of the submandibular gland, submental artery, inferior and superior labial arteries (for the lips), lateral nasal branch, and angular artery (for the eyelids). Through its connection with the OA, the facial artery may provide collateral flow to the ICA in cases of severe stenosis or occlusion of the ICA. This pathway can be demonstrated by sonography or angiography as reversal of flow within the OA. 4. The ascending pharyngeal artery supplies the pharynx, palate, and tympanic cavity (inferior tympanic artery). Its terminal branches are meningeal arteries for the posterior fossa (clivus and posterior surface of petrous bone) that enter the cranium through the foramen lacerum, foramen jugulare, or hypoglossal canal. These branches may also participate in collateral supply of the distal ICA.

CHAPTER 87 CRANIOCERVICAL VASCULAR ANATOMY 629

FIGURE 87-3. Digital subtraction angiogram, internal carotid artery (ICA) injection, lateral view showing occlusion of ICA distal to origin of ophthalmic artery (OA). Middle meningeal artery originates from OA via its lacrimal branch. (©2012, Philippe Gailloud, MD.)

FIGURE 87-2. Digital subtraction angiogram, selective external carotid artery (ECA) injection, lateral view. Main branches of ECA are (1) superior thyroid artery, (2) lingual artery, (3) facial artery, (4) occipital artery, (5) ascending pharyngeal artery, and (6) posterior auricular artery. ECA terminates in (7) superficial temporal artery and (8) maxillary artery, which provides (9) middle meningeal artery. (©2012, Philippe Gailloud, MD.)

5. The posterior auricular artery ascends posteriorly between the auricular cartilage and the mastoid process. It may arise as a common trunk with the occipital artery. It gives rise to several muscular branches for surrounding muscles (digastric, stylohyoid, and sternocleidomastoid muscles), the parotid gland, and the auricle. It also provides a small stylomastoid branch that enters through the stylomastoid foramen into the facial canal and ends in the tympanic cavity. 6. The occipital artery courses over the mastoid process toward the suboccipital region and divides into terminal branches at the level of the external occipital protuberance. It provides a sternocleidomastoid branch, an auricular artery (with a meningeal branch entering the cranium through the mastoid foramen), and a descending branch to the suboccipital region. The latter branch is connected to the vertebral and deep cervical arteries and constitutes the suboccipital carrefour or “knot of Bosniak,” an important source of collateral supply in cases of steno-occlusive disease involving the VA or ICA.9 At the same time, this anastomotic network represents a significant pitfall during embolization procedures involving the occipital artery (dangerous anastomosis), and one has to be particularly vigilant for transient opacification (sometimes visible on one frame

only) of a portion of the VA during selective occipital angiography. The terminal branches of the occipital artery supply the scalp of the occipital region. 7. The maxillary artery originates behind the neck of the mandible and courses through the pterygomaxillary fissure into the pterygopalatine fossa. The maxillary artery has numerous branches: • The deep auricular artery, which often arises with the anterior tympanic artery, courses through the parotid gland and pierces the cartilaginous or bony wall of the external acoustic meatus. It supplies the outer surface of the tympanic membrane and the temporomandibular joint. • The anterior tympanic artery enters the tympanic cavity through the petrotympanic fissure and supplies the tympanic membrane. • The middle meningeal artery is the largest branch of the maxillary artery. It takes a short vertical extracranial course before it enters the middle cranial fossa through the foramen spinosum. It then takes a characteristic horizontal course anteriorly and laterally toward the pterion. A posterior branch arises in the middle cranial fossa and courses along the inner surface of the temporal and parietal bones. An anterior branch reaches the pterion, where it divides into lateral and medial branches. The lateral branch courses along the inner surface of the frontal bone toward the vertex. The medial branch courses under the lesser sphenoid wing and anastomoses with a recurrent branch of the lacrimal artery (branch of the OA), usually through the meningolacrimal foramen. In some cases, a branch of the middle meningeal artery or the entire artery may arise from the OA (Fig. 87-3). Conversely, the OA may originate partially or completely from the middle meningeal artery and enter the orbit through the superior orbital fissure (see later discussion). The clinical importance of the latter variant lies in the potential risk for blindness during endovascular procedures within the ECA territory.

630 PART 1 VASCULAR INTERVENTIONS · SECTION 11 NEUROVASCULAR AND HEAD AND NECK INTERVENTION • The accessory meningeal artery may arise from the middle meningeal artery or from the maxillary artery directly. Despite its name, its main supply is to the pterygoid and tensor veli palatini muscles, with a meningeal contribution representing only about 10% of its territory.10 The meningeal branch, which is inconstant, enters the middle cranial fossa through the foramen ovale and supplies the dura adjacent to the sella turcica and trigeminal ganglion. • The inferior alveolar artery enters the mandibular foramen and courses with the inferior alveolar nerve through the mandibular canal. One of its terminal branches, the mental branch, crosses the mental canal and anastomoses with the submental and inferior labial arteries (branches of the facial artery). • The anterior and posterior deep temporal arteries, the pterygoid branches, and the masseteric and buccal arteries supply the muscles of mastication and the buccinator muscle. • The posterior superior alveolar artery originates (like all the following branches) from the pterygopalatine segment of the maxillary artery, either directly or from a common trunk with the infraorbital artery. It supplies the mucosa of the maxillary sinus and the teeth. • The infraorbital artery enters the orbit through the inferior orbital fissure, runs forward along the roof of the maxillary sinus within the infraorbital canal (along with the infraorbital nerve), and exits the skull through the infraorbital foramen. It supplies the mucosa of the maxillary sinus, the cheek, lower eyelid, upper lip, lacrimal sac, upper incisors and canine teeth, and the side of the nose. It anastomoses with the facial artery and posterior superior alveolar artery. • The descending palatine artery courses down the pterygopalatine canal and gives off a large anterior branch, the greater palatine artery, that crosses the greater palatine foramen, runs forward along the alveolar border of the hard palate, crosses through the incisive canal, and then anastomoses with the sphenopalatine artery, thus supplying the mucosa of the palate, the palatine glands, and the gums. Lesser palatine branches from the descending palatine artery descend into the lesser palatine canals and supply the soft palate and palatine tonsils. • The sphenopalatine artery supplies the nasal mucosa after passing through the sphenopalatine foramen. The sphenopalatine artery is the targeted vessel in superselective embolization for epistaxis.11 In patients with inflammation of the nasal cavity (e.g., those with a common cold), a prominent mucosal blush with early venous drainage toward the superior ophthalmic vein and cavernous sinus may be observed and represents a diagnostic pitfall. • The artery of the pterygoid canal, or the vidian artery, extends backward along the pterygoid canal with the vidian nerve into the cartilage of the foramen lacerum. It supplies the upper part of the pharynx and auditory tube and the tympanic cavity through a small tympanic branch. • The pharyngeal artery leaves the pterygopalatine fossa through the pharyngeal canal and supplies the upper part of the nasopharynx and auditory tube.

• The artery of the foramen rotundum anastomoses with branches of the inferior lateral trunk through the foramen rotundum. 8. The superficial temporal artery arises within the parotid gland, crosses the zygomatic process, where its pulsations may be felt anterior to the auricle, and then divides into terminal anterior (frontal) and posterior (parietal) branches supplying the muscles, integument, and pericranium of these regions. The transverse artery of the face arises from the superficial temporal artery in the parotid gland and crosses forward between the zygomatic arch and parotid duct to the soft tissues of the cheek. Internal Carotid Artery The ICA arises from the carotid bifurcation, most frequently around the C3-C5 vertebral level. The ICA is divided into cervical, petrous, cavernous, and supraclinoid (or cisternal) segments. The course of its proximal cervical segment is closely related to that of the internal jugular vein (IJV) and vagus nerve; it follows a more or less straight trajectory from its origin to the outer opening of the carotid canal. The sharp transition between the mobile distal portion of the cervical ICA and its fixed petrous segment at the skull base represents a zone at risk for traumatic injuries such as arterial dissection12 (Fig. 87-4). Marked but

FIGURE 87-4. Digital subtraction angiogram, common carotid artery injection, unsubtracted lateral view showing arterial dissection with pseudoaneurysm formation at junction between mobile cervical and fixed petrous segments of internal carotid artery (arrow). (©2012, Philippe Gailloud, MD.)

CHAPTER 87 CRANIOCERVICAL VASCULAR ANATOMY 631

FIGURE 87-5. Digital subtraction angiogram, internal carotid artery (ICA) injection, oblique view showing smooth congenital sinuosity of ICA. (©2012, Philippe Gailloud, MD.)

smooth sinuosity of the cervical ICA, which may even lead to a 180-degree turn, can be seen in individuals of any age, including young children, and constitutes an anatomic variant. This variant, classically termed coiling, more frequently involves the distal portion of the cervical ICA13 (Fig. 87-5). Coiling does not generally produce significant lumen narrowing but seems to represent a risk factor for arterial dissection.14 In contrast, the irregular tortuosity that predominates in the proximal portion of the cervical ICA develops with advancing age (generally after 50 years), arteriosclerosis, and high blood pressure. It is often termed kinking, especially when associated with arterial hypertension, in which case it acquires a classic “lead-pipe” appearance (Fig. 87-6). The cervical portion of the ICA does not normally give off angiographically detectable branches. Its average caliber is about 6 mm, and in the absence of unusual secondary branches, it remains constant throughout its course.15 Unusual branches of the cervical ICA include the ascending pharyngeal artery (8%) and less frequently the occipital artery (one third MCA

TABLE 92-1. Exclusion Criteria for Intravenous Administration of Tissue Plasminogen Activator (tPA) • 3 hours from stroke onset • Minor or resolving deficits, NIHSS score < 4 • Sustained hypertension (systolic BP > 185 mmHg/diastolic BP > 110 mmHg) • Seizure at onset • International normalized ratio > 1.7 • Intravenous heparin with partial thromboplastin time > 40 s or low-molecular-weight heparin in past 48 hours • Platelet count < 100,000/mm3 • Glucose > 400 mg/dL or < 50 mg/dL • Stroke, myocardial infarction, or head trauma in past 3 months • Major gastrointestinal or genitourinary hemorrhage in past 3 weeks • Major surgery in past 2 weeks • Noncompressible arterial puncture in past 7 days • Past intracerebral hemorrhage, subarachnoid hemorrhage, or intraventricular hemorrhage • CT scan with intracerebral hemorrhage or hypodensity in greater than one third of middle cerebral artery territory

TABLE 92-2. Indications for Endovascular Acute Stroke Therapy • Significant neurologic deficit (NIHSS score > 8) or large territory at risk • Angiographically demonstrable occlusion of an endovascularly accessible vessel* • No hemorrhage or major infarction on noninvasive imaging • 3 hours). A DWI-PWI mismatch may be found in up to 70% of patients imaged within 6 hours from stroke onset,11 and it is hypothesized that these patients may derive the most clinical benefit from endovascular intervention. Some relative contraindications to IA therapy include rapidly improving neurologic examination, NIHSS over 30, and a history of anaphylaxis to thrombolytic drugs or iodinated contrast material.

EQUIPMENT Digital biplane angiography with road-mapping capability is optimal for the safe and efficient performance of endovascular acute stroke intervention. General endotracheal anesthesia is required for most procedures, although monitored conscious sedation is at times adequate for cooperative patients. Our experience is that patients with left hemisphere strokes are paradoxically more cooperative than right hemisphere stroke patients. The patient should have an arterial line placed for blood pressure management, which may become crucial after recanalization to prevent reperfusion hemorrhage. In the time before vessel recanalization, it is critical to maintain adequate perfusion pressure by elevating the patient’s blood pressure to significantly above the patient’s baseline values, particularly during administration of general anesthesia. A recent study comparing outcomes of AIS patients12 showed that patients placed under general anesthesia during endovascular treatment for anterior circulation had higher rates of poor neurologic outcome and mortality, a difference that could potentially be due to lack of adequate blood pressure support during general anesthesia. Common femoral artery access requires a minimum of a 5F sheath for IA thrombolysis and up to a 9F sheath for the Merci Retrieval System. A

5F diagnostic catheter is used for cerebral angiography to demonstrate and characterize the arterial occlusion. The catheter and sheath are maintained under continuous heparinized saline flush (5000 units heparin per liter normal saline), and nonionic iodinated contrast (Omnipaque [GE Healthcare, Piscataway, N.J.]) is used for angiography. Once an intervention is deemed appropriate, minimal systemic anticoagulation is initiated with 2000 to 3000 units of IV heparin. A guide catheter is required for device support during thrombolysis and embolectomy. A microcatheter and microwire facilitate navigation past the occlusion, and the microcatheter provides a conduit for IA thrombolytic infusion. More specific details about the basic equipment required for IA treatment are summarized in Table 92-3. The most commonly used drugs for IA thrombolysis include plasminogen activators such as urokinase, alteplase, reteplase, and prourokinase. Other newer thrombolytics that are not plasminogen dependent include V10153 (Vernalis, Winnersh, U.K.), a recombinant variant of human plasminogen, Microplasmin (ThromboGenics, Heverlee, Belgium), a truncated form of plasmin, Alfimeprase (Nuvelo, San Carlos, Calif.), a truncated form of fibrolase that degrades fibrin directly, and Ancrod, a purified fraction of the venom of the Malayan pit viper, which acts by directly inactivating fibrinogen. Given the prothrombotic activity of fibrinolytics, there is some rationale for adjunctive anticoagulant therapy when administering fibrinolytics in the setting of AIS. Investigations are ongoing involving combinations of fibrinolytics, anticoagulants, and platelet disaggregants for IV, IA, and combined IV/IA AIS therapy, although there is currently no clear evidence to recommend one combination over another. Periprocedural anticoagulation with IV heparin may allow for augmentation of thrombolysis and prevention of arterial reocclusion at the expense of potentially higher rates of ICH. The use of glycoprotein IIb/IIIa antagonists in ischemic stroke remains investigational; only limited data are available, mainly involving IV administration of a glycoprotein IIb/IIIa antagonist in conjunction with IV thrombolysis. A randomized double blinded placebo-controlled study in which 54 ischemic

688 PART 1 VASCULAR INTERVENTIONS · SECTION 11 NEUROVASCULAR AND HEAD AND NECK INTERVENTION stroke patients were randomly selected to receive IV abciximab did not report a single ICH complication in the group receiving abciximab.13 The CLEAR trial evaluated the combination of low-dose IV rtPA and eptifibatide, a glycoprotein IIb/IIIa antagonist, in patients with NIHSS scores of less than 5 presenting within 3 hours of stroke onset. The study showed a nonsignificant trend toward increased efficacy with the standard-dose rtPA treatment arm, and there was a non–statistically significant increased incidence on ICH in the eptifibatide group.14 A small study on the IA administration of glycoprotein IIb/IIIa antagonists as adjunctive therapy in patients with large-vessel occlusion and AIS also suggests that the added risk of ICH is low.15 However, the AbESTT II trial, a phase III multicenter randomized double-blind placebo-controlled study evaluating the safety and efficacy of IV abciximab in AIS treated within 6 hours after stroke onset or within 3 hours of awakening with stroke symptoms, was stopped early because of high rates of symptomatic ICH in the abciximab-treated patients (5.5% vs. 0.5%, P = .002).16 Mechanical embolectomy involves the use of devices for mechanical clot extraction or fragmentation. The only two devices currently approved by the FDA for revascularization are the Merci Retriever (Concentric Medical Inc., Mountain View Calif.) and the Penumbra Reperfusion System (Penumbra Inc., San Leandro, Calif.). A myriad of other devices designed for this same purpose have recently become available in this regard, but their usage is off-label and/or investigational. Among these newer devices, the Solitaire and Trevo devices seem to be the most promising. The mechanisms employed by these devices include thrombectomy, thromboaspiration, mechanical disruption, angioplasty, stenting, laser techniques, ultrasound-based methods, flow augmentation, and transvenous retrograde reperfusion. A summary of the devices currently available for IA treatment of AIS is listed in Table 92-4.

TECHNIQUE Anatomy and Approaches Specific techniques for endovascular stroke therapy differ depending on whether IA chemical thrombolysis or mechanical embolectomy will be performed in isolation or in combination and on which mechanical device will be deployed. All acute stroke management requires similar equipment (see Table 92-3) and necessitates sufficient guide catheter support from the common femoral artery through the abdominal and thoracic aorta into the ICA for ICA, MCA, or ACA occlusion (Fig. 92-2) and subclavian or vertebral artery access for vertebral artery, basilar artery, and posterior cerebral artery thrombosis (Fig. 92-3). The technical aspects of IA thrombolysis and mechanical embolectomy with the Merci Retriever, the Penumbra Reperfusion System, and the Solitaire FR Revascularization Device are discussed next.

Technical Aspects Intraarterial Thrombolytic Infusion Fibrinolytic infusion is best used in conjunction with mechanical embolectomy, especially in the setting of high

TABLE 92-4. Current Mechanical Embolectomy Devices Approved or Under Investigation Device

Company

Merci Retriever*

Concentric Medical, Mountain View, Calif.

Penumbra System*

Penumbra Inc., San Leandro, Calif.

Solitaire

ev3/Covidien, Irvine, Calif.

Trevo

Concentric Medical

Phenox Clot Retriever and Phenox Clot Retriever CAGE

Phenox, Bochum, Germany

Attracter-18 device

Target Therapeutics, Freemont, Calif.

Alligator Retriever

Chestnut Medical Technologies Inc., Menlo Park, Calif.

EnSnare

InterV, Gainesville, Fla.

AngioJet and NeuroJet

Possis Medical Inc., Minneapolis, Minn.

Oasis Thrombectomy Catheter System

Boston Scientific, Natick, Mass.

Amplatz Thrombectomy Device

Microvena, White Bear Lake, Minn.

Hydrolyzer

Cordis Endovascular, Warren, N.J.

F.A.S.T. Funnel Catheter

Genesis Medical Interventional, Redwood City, Calif.

Microsnare

Microvena

HyperGlide balloon

ev3/Covidien

Gateway angioplasty balloon

Boston Scientific

Enterprise stent

Cordis Corp., Miami Lakes, Fla.

LEO stent

Balt Extrusion, Montmorency, France

Solitaire/Solo stent

ev3/Covidien

Wingspan stent

Boston Scientific

EPAR Laser

Endovasix Inc., San Francisco, Calif.

LaTIS laser device

LaTIS, Minneapolis, Minn.

Locket retrieval

Lazarus Effect, Cambell, Calif.

OmniWave Endovascular System

OmniSonics Medical Technologies, Wilmington, Mass.

EKOS MicroLysUS infusion catheter

EKOS Corporation, Bothel, Wash.

NeuroFlo device

Co-Axia, Maple Grove, Minn.

ReviveFlow System

ReviveFlow, Quincy, Mass.

*510K clearance for recanalization of cerebral arteries.

clot burden (Fig. 92-4). When mechanical embolectomy is not considered, a smaller (6F) guide catheter can be used via a 6F groin sheath. A diagnostic angiogram is performed to identify the branch occlusion(s). The diagnostic catheter can then be exchanged out for the 6F guide catheter. Alternatively, the guide catheter can be navigated directly into the ICA or the vertebral artery and used to perform the initial angiogram in patients with straightforward anatomy. Next, using roadmap technique, the microcatheter is advanced over a microwire into the area of arterial occlusion and then slightly distal to it. A microcatheter

CHAPTER 92 ACUTE STROKE MANAGEMENT 689

A

B

FIGURE 92-2. Left common carotid artery angiogram, intracranial anteroposterior view. A, Acute mid-M1 segment left middle cerebral artery (MCA) occlusion in a 65-year-old man with atrial fibrillation. B, Complete recanalization was achieved with combination intraarterial tissue plasminogen activator (tPA) infusion and mechanical embolectomy with the Merci L5 Retriever.

FIGURE 92-3. Left vertebral artery angiogram, intracranial lateral view, shows an acute mid-basilar artery occlusion.

FIGURE 92-4. Same patient as in Figure 92-3. Left vertebral artery angiogram, intracranial lateral view. After mechanical embolectomy, basilar artery is completely recanalized.

angiogram can then delineate the anatomy of the occlusion and the distal vasculature. The microcatheter is then withdrawn into the area of arterial occlusion, and thrombolytic medication is slowly infused. Sequential passes of the microwire through the thrombus increase the surface area for thrombolysis and are typically performed during the infusion. Cerebral angiography through the guide catheter is performed serially (typically every 15 minutes) during the

infusion to assess the progress of recanalization. The procedure is completed when recanalization has been achieved or the maximum dose of medication has been administered. IA thrombolytic infusion is particularly helpful for recanalization of more distal branch occlusions that are inaccessible to mechanical devices, including the treatment of distal branch occlusion after mechanical thrombus extraction of a large proximal occlusion.

690 PART 1 VASCULAR INTERVENTIONS · SECTION 11 NEUROVASCULAR AND HEAD AND NECK INTERVENTION The Merci Retriever After acquisition and interpretation of the diagnostic cerebral angiogram, the femoral sheath used for diagnostic catheterization must be exchanged under fluoroscopic guidance for an 8F or 9F sheath to allow insertion of an 8F or 9F Concentric BGC. Alternatively, the sheath may be exchanged directly for the BGC. The diagnostic catheter is exchanged out for a BGC using an exchange-length hydrophilic 0.035- or 0.038-inch guidewire, or the BGC can be directly navigated into the ICA, subclavian, or vertebral artery over a 0.035- or 0.038-inch wire in patients with straightforward anatomy. If an ICA kink is encountered, the BGC should be kept in the common carotid artery, even though anterograde flow control may be less efficient. In the setting of an ICA stenosis severe enough to inhibit embolus removal, carotid angioplasty and stenting may be performed before embolectomy. For posterior circulation stroke, owing to the small size of the vertebral artery, the BGC should be positioned in the subclavian artery just below the origin of the vertebral artery. At the distal end of the BGC is a 10-mm-diameter soft silicone balloon that is inflated with a 50% contrast-in-saline medium by a 3-mL syringe (0.8 mL maximal inflation volume) at the time of clot retrieval (Fig. 92-5). Once the BGC is in position, the Merci microcatheter (14X for X-series retrievers and 18L for L and V series retrievers) is advanced over a microguidewire under digital roadmap technique into the target artery. The microwire is navigated past the occlusion into a patent large branch (e.g., posterior cerebral artery for basilar artery occlusion), and the microcatheter is coaxially advanced over the microwire. Careful passage of the microwire into the true lumen of the occluded artery is crucial to avoid vessel dissection or perforation. Contrast medium injection through the microcatheter will demonstrate the patency of the branches distal

to the occlusion. The microwire is then exchanged for the Merci Retriever device. The X-series retriever (X5 and X6 devices) is composed of a memory-shaped nitinol wire with five helical loops of tapering diameter at the distal tip (Fig. 92-6). The loops of the L-series retriever (L5 and L6 devices) are arranged at a 90-degree angle to the base wire and are cylindrical. L-series devices additionally have bound suture threads that encircle the helical loops and function to facilitate thrombus capture (Fig. 92-7). The V-series is a hybrid of the X and L series. The Merci device is advanced through the microcatheter and unsheathed past the microcatheter tip. The retriever and microcatheter are withdrawn as a unit into the thrombus. The device will slightly deform once it is in contact with clot (Fig. 92-8). At this point, anterograde blood flow is arrested by inflation of the BGC balloon, and the thrombusretriever-microcatheter unit is withdrawn into the BGC lumen and removed from the patient under vigorous hand suction with a 60-mL syringe placed at the proximal hub of the BGC (Fig. 92-9). Injection of contrast medium during clot retrieval may demonstrate a thrombus halo, indicating successful thrombus capture (Fig. 92-10). The BGC balloon

FIGURE 92-6. Photograph of the Merci X-series Retriever with tapered helical loops.

FIGURE 92-5. Unsubtracted anteroposterior view of an inflated Concentric balloon guide catheter used to initiate flow arrest during embolus extraction.

FIGURE 92-7. Photograph of the Merci L-series Retriever with cylindrical helical loops and bound suture thread for enhanced clot capture.

CHAPTER 92 ACUTE STROKE MANAGEMENT 691

FIGURE 92-8. Unsubtracted anteroposterior view of Merci L5 Retriever in left middle cerebral artery of patient in Figure 92-2. Note deformation of retriever helix as it captures embolus.

FIGURE 92-10. Left vertebral artery angiogram, intracranial lateral view. Contrast agent injection during flow arrest demonstrates a halo around captured thrombus.

FIGURE 92-9. Photograph of Merci X-series Retriever with ensnared red clot. (Courtesy Concentric Medical, Mountain View, Calif.)

is deflated, and cerebral angiography is then performed to confirm vessel recanalization. Multiple passes with the Merci device may be required to achieve complete vessel recanalization. No more than six passes are recommended by the manufacturer. In our own experience, more than three passes is futile. The Penumbra Reperfusion System After performing the initial diagnostic angiogram, the diagnostic catheter is exchanged out for a 6F guide catheter using an exchange-length hydrophilic 0.035- or 0.038-inch guidewire. Alternatively, the 6F guide can be directly navigated into the ICA or vertebral artery over a 0.035- or 0.038-inch wire in patients with straightforward anatomy. Next, using roadmap technique, a Penumbra Reperfusion catheter of appropriate size for use in the target vessel is advanced over a microwire into the area of the vessel just proximal to the occlusion. The 0.54 Reperfusion catheter can be used for vessels larger than 4 mm in caliber (ICA, M1, vertebral, basilar), the 0.41 Reperfusion catheter for

vessels measuring 3 to 4 mm (M1, vertebral, basilar), the 0.32 for vessels measuring 2 to 3 mm (M2, P1), and the 0.26 for vessels smaller than 2 mm, such as M3 branches. The microwire is then exchanged out for an appropriately sized Separator, which is a soft wire with a 6-mm teardrop-shaped enlargement proximal to its tip designed to physically break up the clot and keep the Reperfusion catheter tip from clogging. The RHV of the Reperfusion catheter is then connected to the aspiration pump, which is then turned on to suction, and the baseline aspiration rate is observed. The Separator is then positioned so that the radiopaque marker on it is 4 mm distal to the radioopaque marker on the Reperfusion catheter. The Reperfusion catheter is then advanced until aspiration flow slows or stops signifying that the reperfusion catheter is at the site of the thrombus. The Separator is then gently moved in and out of the Reperfusion catheter to clear it of debris and break up the thrombus. Guide catheter angiograms are performed intermittently to assess progress. Several passes with the device may be required to clear the vessel of thrombus. Solitaire FR Revascularization Device After acquisition and interpretation of the diagnostic cerebral angiogram, the femoral sheath used for diagnostic catheterization is exchanged out under fluoroscopic guidance for an 8F Cello Balloon Guide Catheter (ev3/Covidien, Irvine, Calif.), which is placed into the proximal ICA. With the balloon of the guide catheter deflated, a 0.014-inch guidewire (Silverspeed; ev3/Covidien) is advanced over a

692 PART 1 VASCULAR INTERVENTIONS · SECTION 11 NEUROVASCULAR AND HEAD AND NECK INTERVENTION Rebar 18 microcatheter (ev3/Covidien) via the guide catheter into the occluded intracranial vessel and navigated distal to the thrombus. The Rebar 18 microcatheter is then advanced over the microwire through the thrombus, and the microwire is removed. The Solitaire FR device is then advanced via the microcatheter and deployed with the distal portion of the stent placed a few millimeters distal to the thrombus. This typically results in immediate flow restoration so long as the stent is longer than the length of the thrombus. A repeat angiogram can be performed at this point to evaluate the potentially reconstituted flow through the previously occluded segment. The stent is kept deployed for a few minutes before its retrieval. The proximal third of the stent is recaptured into the microcatheter, and the balloon on the guide catheter is inflated for flow arrest in the ICA. Then, under continuous proximal aspiration of the guide catheter with a syringe, the stent and microcatheter are gently withdrawn proximally through the guide catheter. Alternatively, in cases with underlying stenosis, the Solitaire Stent can be deployed at the site of occlusion in to help maintain patency of the artery. Several recent reports on the use of the Solitaire Revascularization Device have shown promising revascularization rates of about 90%.17-20

CONTROVERSIES One major controversy in acute stroke management is whether the clinical benefit of reperfusion warrants the risks and cost of the procedures. Currently, no answer to this question exists. However, several studies show relative safety and efficacy and of endovascular stroke therapies in terms of successful vessel recanalization,11,21,22 and successful recanalization has been shown to be strongly associated with improved patient outcomes.23 Most interventional acute stroke trials use a time window rather than a tissue window for inclusion and therefore undoubtedly incorporate both patients with salvageable brain tissue and those with completed infarction. This heterogeneity of treated stroke populations makes proof of clinical benefit difficult. As already mentioned, successful clinical outcomes require appropriate patient selection. The concept of patient selection based on the degree of tissue at risk for infarction (tissue window) is being investigated in MR Rescue, a randomized trial of Merci Retriever embolectomy versus standard medical therapy based on the presence or absence of a 20% or more DWI-PWI mismatch. It is hoped that this study will help resolve this controversy. Another debate is the role of bridging stroke therapy, the concept of administering reduced-dose IV thrombolytic before IA therapy. The IMS I study24 investigated a bridging tPA protocol in 80 patients with NIHSS scores of 10 or higher. Intravenous tPA (0.6 mg/kg over 30 minutes) was administered within 3 hours of stroke onset and was followed by IA tPA to a total dose of 22 mg over a 2-hour infusion or until complete thrombolysis was achieved. Although this trial was a safety and feasibility study, clinical outcomes were modest, showing a trend toward improved clinical outcome at 30 days (odds ratio [OR]for global test, 1.35; 95% confidence interval [CI] 0.78-2.37) and numerically lower 3-month mortality (16% vs. 21%) compared with historical IV tPA-treated patients from the NINDS tPA stroke trial. The IMS II study differed from the IMS I trial mainly by incorporating the use of the MicroLysUS infusion

catheter (EKOS, Bothell, Wash.) to deliver rtPA into the clot and to use ultrasound to augment thrombolysis.25 The IMS II patients had a lower mortality and better outcomes than NINDS rt-PA–treated subjects, with 3-month mortality in IMS II subjects 16%, compared with a 24% mortality with placebo and 21% mortality in rt-PA–treated subjects in the NINDS rt-PA Stroke Trial. The IMS III study is a 3-hour time window stroke protocol randomizing standard IV tPA (0.9 mg/kg) versus a bridging protocol of IV tPA (0.6 mg/ kg) plus IA therapy that uses techniques with proven safety in phase I studies (IMS III). IA treatment may include IA tPA alone (maximum dose 22 mg), ultrasound-enhanced IA tPA with the EKOS microcatheter (demonstrated safety and feasibility in the IMS II Trial), or mechanical embolectomy with the Merci Retriever. Individual technique selection is at the discretion of the operator. These and other bridging strategies, including glycoprotein IIb/IIIa antagonists and newer fibrinolytic agents, may offer thrombolytic synergy and added safety and are currently being investigated.

OUTCOMES Outcome after endovascular stroke management is measured as both technical success and clinical benefit. Overall, endovascular techniques result in higher rates of recanalization than IV thrombolytic therapy, and these higher rates of recanalization are strongly associated with improved functional outcomes and reduced mortality in AIS patients.7 A meta-analysis of the most significant trials of IA thrombolysis, including PROACT I and II8,26 and the Middle Cerebral Artery Embolism Local Fibrinolytic Intervention Trial (MELT),27 included 204 patients treated with IAT and 130 controls. The analysis showed a lower rate of death or dependency at long-term follow-up with IAT compared with controls.28 A second, more recent meta-analysis on this topic confirmed these results, concluding that IA fibrinolysis significantly increases recanalization rates and results in improved clinical outcomes in AIS patients.29 Regarding mechanical embolectomy, in the MERCI trial, arterial recanalization was attained with the X-series Merci Retriever, with and without adjuvant IA thrombolytic therapy, in 57% and 48%, respectively. In the Multi-MERCI trial, a single-arm study of the Merci L5 retriever, the device was deployed in 111 patients, including 30 patients initially treated with IV tPA, and 43 patients were administered adjuvant IA tPA.22 With other demographics being similar to those of the MERCI trial, revascularization was achieved in 54.1%, increasing to 69.4% when an adjuvant IA thrombolytic agent was used. Clinical outcome also benefited from higher revascularization rates, with lower morbidity (34.3% vs. 27.7% independence at 90 days) and lower mortality (30.6% vs. 43.5%) in Multi-MERCI compared with the MERCI trial. The Penumbra stroke trial is a prospective multicenter single-arm study that included 125 patients with NIHSS scores of 8 or higher presenting with angiographic occlusion of an intracranial vessel within 8 hours of symptom onset. All 125 patients were treated with the Penumbra system with an overall 81.6% recanalization rate, although 28% were found to have intracranial hemorrhage on 24-hour followup CT, and all-cause mortality was 32.8% at 90 days, with 25% of the patients achieving a modified Rankin Scale score of 2 or below.30 More recent trials

CHAPTER 92 ACUTE STROKE MANAGEMENT 693 involving the Penumbra system have shown even higher recanalization rates and better clinical outcomes.31 Despite the excellent data from the above studies, there is still an overall lack of controlled clinical outcomes analyses in interventional stroke management. Because clinical outcome is intimately linked with complete vessel recanalization, current device modifications, novel technologies, and newer thrombolytic agents have to be developed to attain improved technical outcomes likely to yield higher rates of clinical success.

COMPLICATIONS The major complication of endovascular acute stroke management is symptomatic ICH, which is associated with a dismal mortality rate.32 ICH may be a consequence of reperfusion, where a rapid increase in hydrostatic pressure after revascularization leads to rupture of previously underperfused vessels or hyperperfusion injury. The latter is more common after chronic rather than acute vessel recanalization and is discussed in detail in Chapter 93, Endovascular Management of Chronic Cerebral Ischemia. Reperfusion hemorrhage occurs within infarcted brain tissue, where endothelial integrity is lost, causing small vessels to become friable and easily rupture. Reperfusion hemorrhage typically occurs shortly after recanalization and is significantly affected by hypertension and the volume and location of infarcted tissue. The basal ganglia and posterior cerebral artery territories appear to be at highest risk for reperfusion ICH. Established infarcts greater than one third of the MCA

A

territory are more likely to bleed once vessel patency is restored. Other risk factors for reperfusion ICH include delayed recanalization, higher pretreatment NIHSS score, inherent or iatrogenic coagulopathy, pretreatment hyperglycemia, and the presence of MRI gradient echo–positive microhemorrhages.32-34 When similar definitions are used to describe symptomatic ICH in the major interventional stroke trials, the rates of symptomatic ICH are 5% in the MERCI trial,11 6.3% in the IMS trial,24 9% in the MultiMERCI trial,22 and 10% in the PROACT II trial.8 These rates are of the same order of magnitude as the 6.4% rate of ICH in the NINDS IV tPA trial.2 Other complications of endovascular stroke management for the most part have less calamitous prognoses compared with reperfusion ICH. Vessel dissection or perforation may arise from sheath insertion or exchange and from diagnostic catheter, guide catheter, or guidewire manipulation anywhere in the approach to the occluded vessel, including the common femoral artery, aorta, innominate artery, subclavian artery, vertebral artery, and the common carotid or ICA. Consequences of vessel dissection or perforation range from asymptomatic and spontaneously resolving events to stenosis and thrombotic occlusion and pseudoaneurysm formation. Arterial trauma may also involve the occluded intracerebral artery secondary to microguidewire or mechanical embolectomy device manipulation and may result in vessel stenosis, thrombosis, or rupture. Depending on the location and degree of dissection, management options include conservative observation, balloon angioplasty, stent placement, or surgical repair (Fig. 92-11). Failed

B

FIGURE 92-11. Same patient as in Figure 92-1. A Neuroform intracranial stent was placed across dissection, tacking flap against arterial wall and thus preventing possible thromboembolic consequences. A, Unsubtracted anteroposterior view demonstrating proximal and distal markers of Neuroform stent. B, Right internal carotid angiogram, intracranial anteroposterior view, shows complete luminal patency after stent deployment.

694 PART 1 VASCULAR INTERVENTIONS · SECTION 11 NEUROVASCULAR AND HEAD AND NECK INTERVENTION common femoral artery hemostasis at the end of the procedure may result in groin hematoma, retroperitoneal hemorrhage, and pseudoaneurysm formation. Occasionally, surgical repair of groin complications is required to restore vessel integrity. Device fractures and thromboembolic events to uninvolved arterial territories may also occur during endovascular revascularization. Device fractures require removal with foreign-body snare devices. Low-level anticoagulation with IV heparin during the procedure diminishes the risk of catheter- and guidewire-related thromboembolic events. Adverse contrast reaction includes allergy, with potentially fatal consequences if prompt management is not instituted, and renal failure, especially in patients with baseline renal insufficiency. Unfortunately, premedication with N-acetylcysteine35 and prehydration with sodium bicarbonate drip,36 which have both been shown to diminish the risk of contrast medium-induced nephropathy, is impractical for emergent stroke intervention. Although unproved, intraprocedural and postoperative administration of these agents may help prevent contrast nephropathy even in the absence of preadministration, and should be instituted in patients with renal insufficiency.

POSTPROCEDURAL AND FOLLOW-UP CARE At the conclusion of an endovascular acute stroke intervention, common femoral artery hemostasis is obtained. Adequate hemostasis is necessary to avoid a groin complication. The nature of these interventions—namely, large groin sheaths and use of thrombolytics and platelet inhibitors— raises the risk of groin hematoma and retroperitoneal hemorrhage. Avoiding the latter involves careful arterial access, specifically avoiding arterial puncture above the inguinal ligament. After an endovascular stroke procedure, hemostasis is best obtained with a closure device (e.g., Angio-Seal Vascular Closure Device [St. Jude Medical Inc., St. Paul, Minn.]). If a closure device is contraindicated (e.g., in a septic patient), the sheath should be kept in place until the thrombolytic coagulopathy is reversed as the infused drugs are metabolized and eliminated. Manual compression is then sufficient for hemostasis. As soon as revascularization is achieved, elevated blood pressure must be rapidly controlled to avoid reperfusion hemorrhage and hyperperfusion syndrome. Blood pressure can be efficiently managed with IV labetalol or a nicardipine drip. Systolic blood pressure targets of below 140 mmHg is the goal, and lower levels (e.g., 100-120 mmHg) may be required for patients with angiographic evidence of hyperperfusion (e.g., basal ganglia blush) after recanalization. Nicardipine drip is started at 5 mg/h IV infusion and can be titrated to a narrow blood pressure goal at 2.5 mg/h every 5 minutes to a maximum dose up to 15 mg/h. Vasodilating antihypertensives, such as nitroprusside, are best avoided because of their intracranial vasodilatory effects that theoretically can increase cerebral hyperperfusion. After the procedure, the patient should be examined as quickly as possible to establish clinical improvement, no change, or detriment. If the patient’s neurologic examination is worse than before the procedure, or if there is any clinical indication of ICH, such as headache, vomiting, decreased level of consciousness, anisocoria, or Cushing

reaction (sudden hypertension and bradycardia associated with brainstem compression from cerebral herniation), an immediate head CT scan should be obtained. If ICH is encountered, procedure-related coagulopathy must be rapidly reversed, blood pressure controlled, and if the hemorrhage is of sufficient size, neurosurgical consultation for intracranial pressure monitoring and hematoma evacuation requested. Reversal of coagulopathy may include any combination of protamine (to reverse intraprocedural heparin), fresh frozen plasma, cryoprecipitate, and platelets, although coagulopathic correction is typically incomplete. Postoperative observation and management in an intensive care unit, optimally one with specialty in neurologic patients, is obligatory. Within the first 24 hours, patients should have nursing neurologic assessments every hour and immediate head CT scans obtained for any clinical deterioration. Vital signs and blood glucose monitoring are essential for maximizing clinical outcome after stroke intervention. Other poststroke comorbidities must be prevented or managed, including deep vein thrombosis, pulmonary embolus, myocardial infarction, pneumonia, urosepsis, malnutrition, and pressure ulcers. Immediate focus on these issues is the key to maintaining a successful clinical outcome. The remaining hospitalization is directed at continuing to prevent or manage poststroke complications, obtaining any necessary physical, occupational, or speech therapeutic services, and determining the etiology of the patient’s stroke. The extent of the poststroke assessment depends on the supposed etiology of the stroke and patient age, where young patients with cryptogenic stroke and few risk factors generally have the most extensive evaluations. Secondary stroke prevention involves prescribing appropriate antiplatelet agents or warfarin and risk factor modification, including management of chronic hypertension, diabetes mellitus, hyperlipidemia, and counseling for smoking cessation. Key to preventing a second stroke is treatment of symptomatic extracranial carotid artery, vertebral artery, and potentially intracerebral arterial stenosis, the endovascular management of which is discussed in Chapter 93. KEY POINTS •

The aim of endovascular stroke treatment is rapid restoration of brain perfusion to minimize cerebral tissue loss during an acute ischemic stroke.

•

Endovascular stroke interventions include intraarterial thrombolytic infusion and mechanical embolectomy.

•

These procedures are indicated for patients with contraindications for systemic tissue plasminogen activator (tPA), or who have failed to recanalize after intravenous tPA, who have accessible arterial occlusions and salvageable brain tissue.

•

Devices like the Merci Retriever, the Penumbra Reperfusion System, and stentrievers can restore vessel patency without thrombolytic drugs, extending the time window of stroke treatment.

•

Intracerebral hemorrhage, the major adverse consequence of endovascular stroke treatment, may be minimized by careful patient selection, judicious use of thrombolytic drugs, and strict postprocedural blood pressure control.

CHAPTER 92 ACUTE STROKE MANAGEMENT 695 ▶ SUGGESTED READINGS Katz JM, Gobin YP, Segal AZ, Riina HA. Mechanical embolectomy. Neurosurg Clin N Am 2005;16(3):463–74, v. Nogueira RG, Schwamm LH, Hirsch JA. Endovascular approaches to acute stroke, part 1: Drugs, devices, and data. AJNR Am J Neuroradiol 2009;30(4):649–61.

Nogueira RG, Yoo AJ, Buonanno FS, Hirsch JA. Endovascular approaches to acute stroke, part 2: A comprehensive review of studies and trials. AJNR Am J Neuroradiol 2009;30(5):859–75.

The complete reference list is available online at www.expertconsult.com.

CHAPTER 92 ACUTE STROKE MANAGEMENT695.e1 REFERENCES 1. Roger VL, Go AS, Lloyd-Jones DM, et al. Heart disease and stroke statistics–2011 update: A report from the American Heart Association. Circulation 2011;123(4):e18–e209. 2. Tissue plasminogen activator for acute ischemic stroke. The National Institute of Neurological Disorders and Stroke rt-PA Stroke Study Group. N Engl J Med 1995;333(24):1581–7. 3. Hacke W, Kaste M, Bluhmki E, et al. Thrombolysis with alteplase 3 to 4.5 hours after acute ischemic stroke. N Engl J Med 2008;359(13): 1317–29. 4. Wardlaw JM, Murray V, Berge E, Del Zoppo GJ. Thrombolysis for acute ischaemic stroke. Cochrane Database Syst Rev 2009;(4):CD000213. 5. Katzan IL, Furlan AJ, Lloyd LE, et al. Use of tissue-type plasminogen activator for acute ischemic stroke: The Cleveland area experience. JAMA 2000;283(9):1151–8. 6. Furlan AJ. CVA: Reducing the risk of a confused vascular analysis. the Feinberg lecture. Stroke 2000;31(6):1451–6. 7. Rha JH, Saver JL. The impact of recanalization on ischemic stroke outcome: A meta-analysis. Stroke 2007;38(3):967–73. 8. Furlan A, Higashida R, Wechsler L, et al. Intra-arterial prourokinase for acute ischemic stroke. the PROACT II study: A randomized controlled trial. Prolyse in acute cerebral thromboembolism. JAMA 1999;282(21):2003–11. 9. Albers GW, Amarenco P, Easton JD, et al. Antithrombotic and thrombolytic therapy for ischemic stroke: American College of Chest Physicians evidence-based clinical practice guidelines (8th edition). Chest 2008;133(6 Suppl):630S–69S. 10. Schonewille WJ, Wijman CA, Michel P, et al. Treatment and outcomes of acute basilar artery occlusion in the basilar artery international cooperation study (BASICS): A prospective registry study. Lancet Neurol 2009;8(8):724–30. 11. Smith WS, Sung G, Starkman S, et al. Safety and efficacy of mechanical embolectomy in acute ischemic stroke: Results of the MERCI trial. Stroke 2005;36(7):1432–8. 12. Abou-Chebl A, Lin R, Hussain MS, et al. Conscious sedation versus general anesthesia during endovascular therapy for acute anterior circulation stroke: Preliminary results from a retrospective, multicenter study. Stroke 2010;41(6):1175–9. 13. Abciximab in acute ischemic stroke: A randomized, double-blind, placebo-controlled, dose-escalation study. the abciximab in ischemic stroke investigators. Stroke 2000;31(3):601–9. 14. Pancioli AM, Broderick J, Brott T, et al. The combined approach to lysis utilizing eptifibatide and rt-PA in acute ischemic stroke: The CLEAR stroke trial. Stroke 2008;39(12):3268–76. 15. Deshmukh VR, Fiorella DJ, Albuquerque FC, et al. Intra-arterial thrombolysis for acute ischemic stroke: Preliminary experience with platelet glycoprotein IIb/IIIa inhibitors as adjunctive therapy. Neurosurgery 2005;56(1):46–54; discussion 54–5. 16. Adams HP Jr, Effron MB, Torner J, et al. Emergency administration of abciximab for treatment of patients with acute ischemic stroke: Results of an international phase III trial: Abciximab in emergency treatment of stroke trial (AbESTT-II). Stroke 2008;39(1):87–99. 17. Cohen JE, Gomori JM, Leker RR, et al. Preliminary experience with the use of self-expanding stent as a thrombectomy device in ischemic stroke. Neurol Res 2011;33(4):439–43. 18. Stampfl S, Hartmann M, Ringleb PA, et al. Stent placement for flow restoration in acute ischemic stroke: A single-center experience with the solitaire stent system. AJNR Am J Neuroradiol 2011.

19. Roth C, Papanagiotou P, Behnke S, et al. Stent-assisted mechanical recanalization for treatment of acute intracerebral artery occlusions. Stroke 2010;41(11):2559–67. 20. Castano C, Dorado L, Guerrero C, et al. Mechanical thrombectomy with the solitaire AB device in large artery occlusions of the anterior circulation: A pilot study. Stroke 2010;41(8):1836–40. 21. Gobin YP, Starkman S, Duckwiler GR, et al. MERCI 1: A phase 1 study of mechanical embolus removal in cerebral ischemia. Stroke 2004; 35(12):2848–54. 22. Smith WS, Sung G, Saver J, et al. Mechanical thrombectomy for acute ischemic stroke: Final results of the multi MERCI trial. Stroke 2008;39(4):1205–12. 23. Rha JH, Saver JL. The impact of recanalization on ischemic stroke outcome: A meta-analysis. Stroke 2007;38(3):967–73. 24. IMS Study Investigators. Combined intravenous and intra-arterial recanalization for acute ischemic stroke: The interventional management of stroke study. Stroke 2004;35(4):904–11. 25. IMS II Trial Investigators. The Interventional Management of Stroke (IMS) II study. Stroke 2007;38(7):2127–35. 26. del Zoppo GJ, Higashida RT, Furlan AJ, et al. PROACT: A phase II randomized trial of recombinant pro-urokinase by direct arterial delivery in acute middle cerebral artery stroke. PROACT investigators. Prolyse in acute cerebral thromboembolism. Stroke 1998;29(1): 4–11. 27. Ogawa A, Mori E, Minematsu K, et al. Randomized trial of intraarterial infusion of urokinase within 6 hours of middle cerebral artery stroke: The middle cerebral artery embolism local fibrinolytic intervention trial (MELT) Japan. Stroke 2007;38(10):2633–9. 28. Saver JL. Intra-arterial fibrinolysis for acute ischemic stroke: The message of MELT. Stroke 2007;38(10):2627–8. 29. Lee M, Hong KS, Saver JL. Efficacy of intra-arterial fibrinolysis for acute ischemic stroke: Meta-analysis of randomized controlled trials. Stroke 2010;41(5):932–7. 30. Penumbra Pivotal Stroke Trial Investigators. The Penumbra Pivotal Stroke Trial: Safety and effectiveness of a new generation of mechanical devices for clot removal in intracranial large vessel occlusive disease. Stroke 2009;40(8):2761–8. 31. Kang DH, Hwang YH, Kim YS, et al. Direct thrombus retrieval using the reperfusion catheter of the penumbra system: Forced-suction thrombectomy in acute ischemic stroke. AJNR Am J Neuroradiol 2011;32(2):283–7. 32. Kase CS, Furlan AJ, Wechsler LR, et al. Cerebral hemorrhage after intra-arterial thrombolysis for ischemic stroke: The PROACT II trial. Neurology 2001;57(9):1603–10. 33. Kidwell CS, Saver JL, Carneado J, et al. Predictors of hemorrhagic transformation in patients receiving intra-arterial thrombolysis. Stroke 2002;33(3):717–24. 34. Kidwell CS, Saver JL, Villablanca JP, et al. Magnetic resonance imaging detection of microbleeds before thrombolysis: An emerging application. Stroke 2002;33(1):95–8. 35. Tepel M, van der Giet M, Schwarzfeld C, et al. Prevention of radiographic-contrast-agent-induced reductions in renal function by acetylcysteine. N Engl J Med 2000;343(3):180–4. 36. Merten GJ, Burgess WP, Gray LV, et al. Prevention of contrast-induced nephropathy with sodium bicarbonate: A randomized controlled trial. JAMA 2004;291(19):2328–34.

CHAPTER

93

Endovascular Management of Chronic Cerebral Ischemia Pierre Y. Gobin and Edward D. Greenberg

Chronic cerebral ischemia may be the result of extracranial carotid or vertebral stenosis and/or intracranial arterial stenoses. This chapter is going to focus on intracranial arterial stenoses as well as extracranial vertebral artery stenosis. For an in-depth discussion of extracranial carotid stenosis, please refer to Chapter 91, Carotid Revascularization. Intracranial atherosclerosis is estimated to be the underlying cause of 10% of all ischemic strokes1,2 and carries a first-year ipsilateral stroke rate of at least 11%.3 Twenty percent of all ischemic strokes involve the posterior circulation, and the most common area of vascular stenosis or occlusion in the setting of a posterior circulation stroke is the extracranial vertebral artery.4 Given these strong associations with stroke, intracranial atherosclerotic disease and extracranial vertebral artery atherosclerotic disease have become prime targets for endovascular treatment.

INDICATIONS The goal of intracerebral artery and extracranial vertebral endovascular revascularization in the setting of atherosclerotic stenosis is ipsilateral stroke prevention. Given the benign nature of asymptomatic intracranial atherosclerotic disease, asymptomatic patients with intracranial atherosclerotic narrowing should not be offered endovascular treatment.5 Regarding treatment of symptomatic intracranial atherosclerosis, it is generally believed that endovascular treatment should only be considered in patients who remain symptomatic despite maximal medical therapy. The rationale leading up to this conclusion, including the preliminary results of the SAMMPRIS study (Stenting and Aggressive Medical Management for Preventing Recurrent Stroke in Intracranial Stenosis), are discussed in the following paragraphs. The utility of best medical therapy alone in the treatment of intracranial atherosclerotic disease was evaluated in the Warfarin-Aspirin Symptomatic Intra-cranial Disease Trial (WASID).3 In this study, patients with a history of stroke or transient ischemic attack (TIA) caused by an angiographically confirmed 50% to 99% intracranial artery stenosis were randomly assigned to receive aspirin (1300 mg; n = 280) or warfarin (target international normalized ratio 2.0 to 3.0; n = 289). The study was prematurely terminated because of higher death, major hemorrhage, and myocardial infarction rates in the warfarin group, without any primary endpoint benefit for either treatment (stroke/brain hemorrhage/vascular death other than stroke, 22.1% for aspirin vs. 21.8% for warfarin). Perhaps most importantly, this study illustrated the high morbidity and lack of effective medical therapy for moderate to severe intracranial atherosclerotic stenosis at that time. The overall 1- and 2-year rates of ischemic stroke in the territory of the stenosed artery were 11% and 14%, respectively. Factors associated with increased stroke risk included stenosis greater than 696

70%, recent symptoms, and female gender.6 The risk of a recent index event (TIA or stroke) and the linear increase in stroke risk associated with the degree of stenosis is analogous to the risk stratification observed in cervical carotid artery disease, which is estimated at 22% for severe symptomatic stenosis. Endovascular stenting is another treatment option that has been studied for the prevention of recurrent stroke or TIA in patients with intracranial atherosclerotic disease. However, evaluation of the efficacy of intracranial angioplasty and stenting is limited by factors such as rapidly evolving technology, the wide variation in endovascular techniques, and the overall lack of controlled data in the literature. Outcome after endovascular treatment for chronic cerebral ischemia is generally measured by technical success, incidence of restenosis, and early and long-term ipsilateral stroke rates. Technical success for intracerebral arterial and extracranial vertebral artery disease, defined as residual stenosis post procedure of less than 50%, is on the order of 95%.7-10 However, restenosis rates are high, with the Stenting of Symptomatic Atherosclerotic Lesions in the Vertebral or Intracranial Arteries (SSYLVIA) study8 reporting a 43% restenosis rate over a mean follow-up period of 16 months for extracranial vertebral artery stenting and a 32.4% restenosis rate for intracranial stenting. In estimating the overall risk of stroke and death associated with intracranial angioplasty with or without stenting, a Cochrane systematic review of 79 publications calculated the pooled periprocedural rate of stroke or death at 9.5%.11 The SSYLVIA study,8 a nonrandomized prospective protocol evaluating the feasibility of the Neurolink System (Guidant Corp., Indianapolis, Ind.) for vertebral and intracranial artery stenosis, included patients with 50% or greater stenosis by angiography with a recent history of stroke (60.7%) or TIA (39.3%). Failure of medical therapy before endovascular intervention was not a prerequisite for this study. Of the 61 patients included, 43 underwent treatment for symptomatic intracranial stenosis (15 internal carotid, 5 middle cerebral, 1 posterior cerebral, 17 basilar, 5 vertebral), and 18 underwent treatment for extracranial vertebral artery stenosis (6 ostial lesions and 12 lesions proximal to the posterior inferior cerebellar artery). Successful stenting with residual stenosis of less than 50% was attained in 95% (58/61 patients). However, as noted, restenosis rates were high: 12/37 or 32.4% for the intracranial arteries and 6/14 or 42.9% for the extracranial vertebrals. Among the 18 patients with restenosis, 7 (39%) were symptomatic. Strokes occurred in 6.6% of patients within 30 days and in 7.3% between 30 days and 1 year, and there were no deaths. In the prospective multicenter Wingspan trial,7 45 patients with symptomatic medically refractory intracranial stenosis of more than 50% were treated with angioplasty and Wingspan stent placement (Fig. 93-1). Technical success, defined as residual postprocedure stenosis of less

CHAPTER 93 ENDOVASCULAR MANAGEMENT OF CHRONIC CEREBRAL ISCHEMIA 697

A

B

FIGURE 93-1. Intracranial angioplasty and stenting. A, Right common carotid artery angiogram, intracranial anteroposterior view, demonstrates a critical right middle cerebral artery (MCA) stenosis in this 57-year-old woman with multiple right MCA distribution infarcts and hypoperfusion on computed tomographic perfusion imaging. B, Complete recanalization is achieved after angioplasty with a 2.5-mm Gateway balloon and placement of a 3 × 20 mm Wingspan stent (Boston Scientific, Natick, Mass.).

than 50%, was achieved in 100% of cases. The incidence of restenosis was only 7.5%, much lower than in the SSYLVIA study, and the ipsilateral 30-day and 6-month combined stroke and death rates were 4.4% and 7.1%, respectively. The only prospective randomized study on the topic of intracranial angioplasty and stenting is the SAMMPRIS trial, which compared aggressive medical management alone to intracranial stenting plus aggressive medical management for the prevention of stroke in patients with symptomatic severe intracranial stenosis (70%-99%) of a major artery (middle cerebral, carotid, vertebral, or basilar arteries). In the context of the study, aggressive medical management consisted of aspirin 325 mg/day for the entire follow-up, clopidogrel 75 mg/day for 90 days after enrollment, intensive management of vascular risk factors (systolic blood pressure < 140 mmHg, [=70% symptomatic intracranial stenosis after Wingspan stenting. Stroke 2011; 42(7):1971–5. 19. Fiorella DJ, Turk AS, Levy EI, et al. US Wingspan registry: 12-month follow-up results. Stroke 2011;42(7):1976–81. 20. Al-Ali F, Cree T, Hall S, et al. Predictors of unfavorable outcome in intracranial angioplasty and stenting in a single-center comparison: Results from the Borgess Medical Center-intracranial revascularization registry. AJNR Am J Neuroradiol 2011;32:1221–6. 21. Higashida RT, Meyers PM, Connors JJ 3rd, et al. Intracranial angioplasty and stenting for cerebral atherosclerosis: A position statement of the American Society of Interventional and Therapeutic Neuroradiology, Society of Interventional Radiology, and the American Society of Neuroradiology. J Vasc Interv Radiol 2009;20(7 Suppl):S312–6. 22. Abou-Chebl A, Yadav JS, Reginelli JP, et al. Intracranial hemorrhage and hyperperfusion syndrome following carotid artery stenting: Risk factors, prevention, and treatment. J Am Coll Cardiol 2004;43(9): 1596–601. 23. Ascher E, Markevich N, Schutzer RW, et al. Cerebral hyperperfusion syndrome after carotid endarterectomy: Predictive factors and hemodynamic changes. J Vasc Surg 2003;37(4):769–77. 24. Nielsen TG, Sillesen H, Schroeder TV. Seizures following carotid endarterectomy in patients with severely compromised cerebral circulation. Eur J Vasc Endovasc Surg 1995;9(1):53–7. 25. Coutts SB, Hill MD, Hu WY. Hyperperfusion syndrome: Toward a stricter definition. Neurosurgery 2003;53(5):1053–58; discussion 1058–60. 26. Zhang R, Zhou G, Xu G, Liu X. Posterior circulation hyperperfusion syndrome after bilateral vertebral artery intracranial stenting. Ann Vasc Surg 2009;23(5):686.e1–686.e5. 27. Rezende MT, Spelle L, Mounayer C, et al. Hyperperfusion syndrome after stenting for intracranial vertebral stenosis. Stroke 2006;37(1): e12–4.

CHAPTER

94

Management of Head and Neck Tumors

Todd S. Miller, Elizabeth R. Tang, Randall P. Owen, Jacqueline A. Bello, and Allan L. Brook

The types of masses encountered in the head and neck are diverse, as are their appropriate treatments. Some benign head and neck lesions are treated primarily by surgical means without endovascular intervention. For example, when possible, hemangiomas are commonly excised by laser, and lymphangiomas are sclerosed by direct puncture or surgically resected.1 Certain highly vascular benign tumors like carotid body paragangliomas and juvenile nasal angiofibromas are frequently treated with preoperative embolization. Malignancy is treated by surgery whenever possible.2,3 Often, however, these patients present with the tumor at a high stage. For stage IV squamous cell carcinoma, 5-year survival rates are lower than 25% following surgery and/or radiotherapy.4 Management of these serious late-stage malignancies is complex, involving substantial teamwork among several disciplines. The interventional radiologist’s role on this team is most necessary for advanced tumors or carotid blowout syndrome (CBS) secondary to such tumors and/or complications of treatment. Potential interventions include balloon test occlusion (BTO), preoperative embolization to aid surgery or posttreatment embolization for hemorrhage due to tumor necrosis, emergent treatment of CBS, intraarterial chemotherapy, and image-guided radiofrequency ablation (RFA).5

INDICATIONS FOR ENDOVASCULAR THERAPY OF BENIGN HEAD/NECK LESIONS The two most common endovascularly treated benign lesions of the head and neck are paragangliomas and juvenile nasal angiofibromas (JNAs). Preoperative embolization of other potentially highly vascular head/neck tumors, including angiomyomas (i.e., angioleiomyoma, vascular leiomyoma),6,7 neurofibromas,8 schwannomas,9,10 hemangioendothelioma,11 and meningioma1 has been reported. Congenital vascular malformations that have failed to regress or those whose mass effect impinges on vital structures have also been successfully treated with endovascular therapy.12

Paragangliomas Paragangliomas are slow-growing tumors that originate from the neural crest of the autonomic nervous system and can be hereditary in 7% to 9% of cases.13 Locations in the head and neck include the carotid body (35%), glomus vagale (11%), tympanic and/or jugular regions (almost 50%; jugulotympanic if the tumor is larger and melds together

over the two regions), and other locations (6%) including the orbit, nasal cavity, and the rare laryngeal case, 90% of which are supraglottic.14 The malignant potential of paragangliomas is generally low at around 4%,15 but it can be up to 10% to 18% for the vagal, carotid, and laryngeal types.3 Recommended therapy for paragangliomas is complete surgical excision. Given the high vascularity of these tumors, the role of endovascular management by preoperative angiogram and embolization can be significant. If a paraganglioma is suspected, a diagnostic superselective angiogram should precede any biopsy attempt, lest the attempt lead to arterial rupture. The angiogram should include bilateral exploration,16 since multicentric paragangliomas occur in up to 80% of familial cases and 10% to 20% of nonfamilial,17 and bilateral carotid body tumors occur in 5% of sporadic and 33% to 38% of familial lesions.13 In addition, diagnostic angiography may demonstrate occult lesions not previously demonstrated on computed tomography (CT) or magnetic resonance imaging (MRI).16 Diagnostic evaluation demonstrates the blood supply and flow dynamics of the tumor. The venous supply can be mapped, allowing the surgeon to preserve the venous drainage, limiting blood loss. The role of embolization in paraganglioma management depends on the extent of the tumor and the anatomy of its feeding branches.3,16,17 In jugular or jugulotympanic paragangliomas with more complicated vasculature involvement, preoperative embolization has become a definite part of surgical management and can significantly reduce intraoperative blood loss.17,18 In most tympanic and laryngeal cases, preoperative embolization is unnecessary.17,19 Transarterial or direct puncture embolization as an alternative to radiotherapy could be considered as the primary therapy for paragangliomas that are unresectable or where resection would be debilitating.1,20-22,63 As for the management of carotid body and vagal paragangliomas, most recommend preoperative embolization, but the role of preoperative embolization is still a matter of debate, especially it seems for carotid body tumors.14,17,23 Carotid body tumors are potentially the most challenging to resect. This tumor originates in the bifurcation of the common carotid artery (CCA), where it commonly involves the internal carotid artery (ICA) and receives feeding vessels from external carotid branches and muscular branches of the ICA.24 When small, hemostasis can be controlled during resection through careful preoperative and intraoperative consideration of the ascending pharyngeal artery (Fig. 94-1).14,13,11 Finally, if resection of the paraganglioma may require sacrifice of the carotid artery, particularly in the case of 703

704 PART 1 VASCULAR INTERVENTIONS · SECTION 11 NEUROVASCULAR AND HEAD AND NECK INTERVENTION

A

D

B

C

E

F

FIGURE 94-1. Carotid body tumor. A, Axial contrast-enhanced CT demonstrating large enhancing mass deviating airway. Vessels of carotid sheath are splayed. B, Axial Fat Sat T2-weighted fast spin-echo (FSE) magnetic resonance imaging (MRI) demonstrating same findings with flow-related signal voids seen within central tumor. C, Sagittal Fat Sat FSE T2-weighted MRI demonstrating flow-related signal voids within mass centered on carotid sheath. D, Lateral angiogram demonstrating splayed external and internal carotid arteries. E, Capillary phase from same angiogram in anteroposterior view demonstrating diffuse capillary blush. F, Lateral angiogram taken with a coaxial microcatheter and guide demonstrating multicompartmental blood supply. This technique was used to deliver polyvinyl alcohol particles 150 to 250 μm in size to each feeding vessel. Immediate surgical resection followed.

large jugulotympanic tumors and larger carotid body tumors, another necessary preoperative endovascular intervention is BTO of the ICA.24

Juvenile Nasal Angiofibromas JNAs are benign, submucosal, unencapsulated, vascular tumors that are most commonly diagnosed in adolescent males between 14 and 25 years old, but can present at any age. The tumor is locally destructive and tends to erode

bone as it spreads.24 Patients usually present with unilateral nasal obstruction or epistaxis. The majority originate in the sphenopalatine foramen and progress to the pterygopalatine fossa, infratemporal fossa, sphenoid sinus, and/or other locations within the head/neck.25,26 Though benign, the risk of bleeding and/or further intracranial extension can be life threatening.27 Surgical resection is considered one of the best treatment options, in which preoperative embolization plays a significant role to prevent excessive intraoperative blood loss.28

CHAPTER 94 MANAGEMENT OF HEAD AND NECK TUMORS 705

B

A

D

E

C

F

FIGURE 94-2. Juvenile nasopharyngeal angiofibroma. A, Axial contrast-enhanced CT demonstrating an enhancing mass expanding pterygopalatine fossa and extending through foramina. Foramina are expanded and bone is eroded. B, Axial contrast-enhanced magnetic resonance imaging (MRI) demonstrating similar findings. C, Coronal contrast-enhanced MRI with fat suppression demonstrating tumor extent relative to skull base. D, Anteroposterior (AP) angiogram demonstrating internal carotid arterial supply and tumor blush. E, AP angiogram with a 4F H1 catheter injecting left internal maxillary artery, causing intense tumor blush. F, Lateral angiogram demonstrating coaxial microcatheter and guide used to inject ascending pharyngeal artery. This technique was used to deliver 150- to 250-μm polyvinyl alcohol particles to bilateral external carotid artery branches feeding tumor. Immediate surgical resection followed.

Determining the extent of intracranial involvement is crucial to treatment planning for these lesions, as is determining the vascular supply, which may include ipsi- and contralateral, internal and external carotid artery (ECA) systems.29 Pretherapeutic planning often involves endoscopic biopsy, CT scan, MRI, and diagnostic angiography (Fig. 94-2). In cases where diagnostic angiography shows risk of surgical injury to or potential involvement of the ICA, BTO of the ICA should be used to determine

adequacy of collateral blood flow. Because profuse intraoperative bleeding is a hallmark of JNA surgery, preoperative embolization of feeding vessels from the external carotid system is consistently recommended in the ear, nose, and throat literature, although some authors rely on intraoperative ligation of feeding vessels.25,29,30,31 Decreased blood loss, surgical time and morbidity have been demonstrated with either approach. This consideration is crucial because the relatively lower intravascular blood volume of children

706 PART 1 VASCULAR INTERVENTIONS · SECTION 11 NEUROVASCULAR AND HEAD AND NECK INTERVENTION Parotid Hemangioma Diagnosis unsure MRI

Focal

Treatment Propranolol & Laser

Segmental

No Treatment

Reassess periodically

V3?, Exclude airway involvement

Exclude PHACES

Inadequate response Treatment Sclerotherapy Surgery

FIGURE 94-3. Decision tree for management of parotid hemangiomas. (From Weiss I, O TM, Lipari BA, Meyer L, Berenstein A, Waner M. Current treatment of parotid hemangiomas. Laryngoscope 2011;121: 1642–50, Fig. 5.)

Propranolol & Laser

No Treatment

Reassess periodically

Sclerotherapy Surgery

limits their tolerance for blood loss. Direct embolization with Onyx has been performed as part of endoscopic resection as well.32

ethanol sclerotherapy should be considered the primary treatment modality, followed by a combination of embolization and surgery if repeated sclerotherapy fails.39

Congenital Vascular Malformations

INDICATIONS FOR ENDOVASCULAR THERAPY OF MALIGNANT HEAD/NECK CANCER

Congenital vascular malformations are primarily treated with percutaneous therapy, where direct puncture of these lesions allows instillation of sclerosing agents. In some cases, however, endovascular techniques can assist in therapy (Fig. 94-3). Congenital vascular malformations are categorized as either low- or high-flow lesions. The low-flow group includes telangiectasias, lymphangiomas, and hemangiomas. Although sclerotherapy is the dominant therapy for this group,12,33 when hemangiomas present as superficial fungating hemorrhagic lesions large enough to sequester platelets or obstruct vision or vital structures, polyvinyl alcohol (PVA) embolization may be used conservatively.3,34 To prevent disfigurement or facial nerve injury when hemangiomas present in the face, surgical excision is usually performed but requires complete resection to minimize the high rate of recurrence. Preoperative embolization assists resection by shrinking the tumor and also helps preserve cosmetic appearance following surgery.1,35,36 The high-flow group, in addition to JNAs, includes arteriovenous fistulas (AVFs)/aneurysms and arteriovenous malformations (AVMs). Kohout et al.’s study (1998), the largest analysis of patients with head/neck AVMs, showed the most success with those lesions treated by a combination of super-selective embolization and aggressive surgical resection.37,38 Although embolization and surgery lead to a much higher curative rate for head/neck AVMs, the complication rate has been reported as almost double that of ethanol sclerotherapy.39 Thus, Jeong emphasized that

Surgical resection maintains an important role in the management of head/neck carcinomas; indeed, complete resection of head/neck malignancy with negative histologic margins is the most crucial factor in prognosis.2 Even in cases of residual or recurrent advanced head/neck carcinoma involving the carotid artery, where prognosis continues to be dim despite resection, surgery can reduce the incidence of life-threatening sequelae like fistula formation, erosion of the pharyngeal wall, or acute tumoral hemorrhage.40,41

Carcinoma Involving Carotid Artery Carotid Occlusion and Tumor Embolization Followed by Resection Most head/neck malignancies are relatively hypovascular and do not require preoperative embolization for surgical resection,42 but involvement of the carotid artery by head and neck carcinoma can hinder safe tumor resection. In these cases, preoperative BTO followed by preoperative carotid artery occlusion can greatly reduce neurologic morbidity and assist in planning for resection of tumor involving the carotid artery. Candidates for surgery with preoperative embolization include those whose carotid arteries are invaded by the less radiosensitive carcinomas (e.g., adenocarcinoma, adenoid cystic carcinoma, melanoma) or any residual or recurrent

CHAPTER 94 MANAGEMENT OF HEAD AND NECK TUMORS 707

FIGURE 94-5. Balloon test occlusion. Lateral view from angiogram demonstrates occlusive balloon in distal internal carotid artery (oblique arrow). Horizontal arrow indicates stagnant contrast material in carotid bulb. Vertical arrow indicates guiding catheter in distal common carotid artery.

FIGURE 94-4. Axial contrast agent-enhanced computed tomography scan of tumor invading carotid artery. Arrow indicates necrotic tumor surrounding internal carotid artery (right ICA with > 200-degree circumferential involvement). There is high likelihood of carotid invasion by tumor.

head/neck carcinomas including squamous cell.43 Usually such a patient will present with a fixed neck mass. Clinical presentation, however, has a highly inaccurate (32%-63%) rate of predicting carotid invasion. Circumferential carotid artery involvement of at least 270 degrees should be confirmed by MRI, CT, or ultrasound, which predicts arterial wall invasion with high sensitivity and specificity. In one series, arteriography only identified invasion in 1 of 12 patients2,44,45 (Fig. 94-4). Historically, carotid artery ligation can cause up to 45% neurologic morbidity rate and 31% to 41% reported mortality.2 Stroke can occur either immediately from decreased blood flow or later from a postprocedural thrombus or embolic event.43 Preoperative BTO with/without singlephoton emission computed tomography (SPECT) imaging or induced hypotension can help identify those who would be less neurologically vulnerable to decreased blood flow. Patients whose preocclusion diagnostic evaluation indicates inadequate collateral flow may benefit from a bypass procedure during surgical ligation, though this approach adds morbidity and mortality. For those patients with demonstrated sufficient collateral flow allowing for carotid resection without reconstruction, preoperative coil embolization of the ICA can minimize surgical blood loss as well as minimize postprocedural thromboembolic complications. In particular, coil embolization of the preophthalmic petrous carotid artery with postembolization heparinization has been shown to help prevent distal embolization after surgical resection.43 After allowing clot formation to

FIGURE 94-6. Damage to carotid artery by tumor and therapy. Anteroposterior angiogram demonstrates severe irregularity of internal and external carotid arteries.

stabilize, surgical ligation can follow 2 to 6 weeks later42 (Fig. 94-5). Carotid Blowout Syndrome Rupture of any branch in the extracranial carotid artery system, known as carotid blowout, can follow extravasation, pseudoaneurysm, or AVF from any number of causes and can lead to acute transoral or transcervical hemorrhage. Carotid blowout syndrome is a broader term that encompasses carotid blowout as well as the threat of it in cases where the carotid is injured or exposed.46 Most commonly, the cause leading to carotid blowout is therapy for squamous cell carcinoma (Figs. 94-6 and 94-7). Radiation

708 PART 1 VASCULAR INTERVENTIONS · SECTION 11 NEUROVASCULAR AND HEAD AND NECK INTERVENTION direct-puncture ethanol or cyanoacrylate as a last resort48 (Table 94-1).

Radplat/Intraarterial Chemotherapy

FIGURE 94-7. Carotid blowout syndrome. Lateral angiogram demonstrates contrast material leaking into soft tissue surrounding diseased carotid artery.

weakens the arterial wall and has been associated with a sevenfold increase in the risk of CBS. Other etiologies include other tumors, neck trauma, and functional endoscopic surgery for refractory epistaxis.47 Historically, treatment of CBS with emergent surgical ligation alone led to a 60% neurologic complication rate and 40% overall mortality.48,49 With advances in interventional endovascular therapies that include embolization, occlusion, and stenting, morbidity and mortality has been reduced substantially. Attempts should therefore be made to treat CBS endovascularly. A combination of anatomy, severity, and availability of interventional services determines the appropriate endovascular management of CBS. A helpful algorithm has been proposed by Cohen and Rad47 (Fig. 94-8). If CBS severity falls under the categories of threatened by pseudoaneurysm or neoplasm, or impending or acute, angiography and CT with contrast should be used to confirm the anatomy and the potential of sentinel hemorrhage.50 If ICA or CCA involvement is demonstrated by angiography/CT, BTO is indicated to determine extent of collateral flow. If collateral flow is inadequate, stenting is indicated to achieve hemostasis until permanent occlusion with surgical extra-anatomic bypass (i.e., subclavian to carotid) can be performed. Covered stent technologies have advanced to a point where their use in such patients may also be beneficial.51 ICA/CCA involvement with adequate collateral flow as well as ECA involvement should be managed by distal and proximal occlusion using coils or detachable balloons (currently unavailable in the United States). If the tumor itself is the source of hemorrhage, as indicated by tumor blush on angiography/CT, the entire tumor bed should be embolized. This can be accomplished with PVA particles via superselective transarterial microcatheterization of all arterial supply to the tumor or by

Surgery and radiotherapy are the primary treatments for head/neck cancer therapy. Yet surgical resection of areas of the larynx, pharynx, and/or tongue can severely diminish quality of life through limits on speech and swallowing.4 While radiotherapy alone may better preserve organ function, it does not always provide better rates of local control and survival.52 Chemotherapy is a promising treatment; studies of head/neck squamous cell carcinomas have shown that they may be highly responsive in a dose-related fashion to the drug cisplatin. Systemic delivery of cisplatin, however, can result in side effects like severe nephrotoxicity. Selective intraarterial chemotherapy may decrease the risk and severity of these systemic effects while permitting high-dose cisplatin delivery directly to the tumor bed by taking advantage of the angiographer’s ability to selectively catheterize distal external carotid branches. Intraarterial chemotherapy is often administered in conjunction with radiotherapy, as seen in the most well-known of the intraarterial head/neck regimens, RADPLAT, reported by Robbins et al.53-55 RADPLAT (radiotherapy and concomitant intraarterial cisplatin) also combines intraarterial chemotherapy with an intravenous systemic infusion of the competitive cisplatin antagonist, sodium thiosulfate, to further protect the rest of the body from cisplatin’s systemic toxic effects.56 In this manner, RADPLAT allows for delivery of high doses of cisplatin to head/neck tumors while limiting systemic exposure. Other applications of intraarterial chemotherapy under investigation include its preoperative use to improve the quality of head/neck tumor resection (neo-RADPLAT), its role in palliative therapy,57 and its use as a potential prognostic indicator that can guide further management of disease. Patients who completely respond to intraarterial chemoradiation may have a significantly higher overall survival rate than those who have a lesser response.57,58 Several authors have subsequently reported success with selective intraarterial chemotherapy using direct superficial temporal artery access.59-62

CONTRAINDICATIONS For benign lesions, in the rare event the tumor blood supply extends intradurally (e.g., via anterior inferior cerebellar artery [AICA] or posterior inferior cerebellar artery [PICA]), embolization is contraindicated owing to risks to the intracranial circulation.63 Specifically, ICA embolization is contraindicated when collateral blood supply is insufficient (Table 94-2), as when a patient fails a BTO or when the contralateral carotid system has previously undergone embolization or ligation. Finally, for congenital vascular malformations other than the cases indicated earlier, percutaneous sclerotherapy may be considered the primary treatment modality rather than embolization and resection. For malignancy, if involvement of the carotid artery is by a previously untreated case of squamous cell carcinoma, surgical resection with embolization would not be the therapy of choice. Carotid invasion is usually best treated with radiotherapy. Also, if collateral circulation is

CHAPTER 94 MANAGEMENT OF HEAD AND NECK TUMORS 709 Carotid Blowout Syndrome

Stabilization

Interventional radiology available?

No

No

Yes

Transfer possible?

Yes

Transfer Clinically

Threatened CBS

Impending CB

Angiography Negative

CT

Angiography

Positive

Positive Internal or common carotid system involvement

N e g a t i v e Surgical ligation

Acute CBS

Test balloon occlusion

External carotid system involvement

Tumor blush Only

Passed

Failed

Appropriate wound management

Stenting

Occlusion

Embolization

FIGURE 94-8. Management algorithm for carotid blowout syndrome (CBS). Morrissey and colleagues proposed a decision tree for the management CBS. (From Cohen J, Rad I. Contemporary management of carotid blowout. Curr Opin Otolaryngol Head Neck Surg 2004;12:110–5, after Morrissey DD, Andersen PE, Nesbit GM, et al. Endovascular management of hemorrhage in patients with head and neck cancer. Arch Otolaryngol Head Neck Surg 1997;123:15–9.)

TABLE 94-1. Balloon Test Occlusion Decision Matrix*: Indications for Bypass from Temporary Balloon Occlusion: Institutional Protocol Clinical

EEG

SPECT

Hypotension

Intervention

Pass

Pass

Pass

Pass

PO

Pass

Fail

Fail

Fail

PO + low-flow bypass (STA-MCA)

Fail

Fail

Fail

Fail

PO + high-flow bypass (venous/radial artery bypass)

Pass

Pass

Fail

Fail

PO + low-flow bypass (STA-MCA)

*The suggested testing matrix was devised from multiple components of balloon test occlusion and results of surgical management of the carotid artery during tumor resection. EEG, Electroencephalogram; PO, permanent occlusion; SPECT, single-photon emission computed tomography; STA-MCA, superficial temporal artery to middle cerebral artery. From Parkinson RJ, Bendok BR, O’Shaughnessy BA, et al. Temporary and permanent occlusion of cervical and cerebral arteries. Neurosurg Clin N Am 2005;16:249–56.

710 PART 1 VASCULAR INTERVENTIONS · SECTION 11 NEUROVASCULAR AND HEAD AND NECK INTERVENTION TABLE 94-2. Collateral Flow Patterns Extracranial to intracranial anastomoses External carotid artery (ECA) to internal carotid artery (ICA) Maxillary artery Middle meningeal artery (MMA), recurrent meningeal to ethmoidal branches of ophthalmic artery (OA) MMA to inferolateral trunk (ILT) (foramen spinosum) Artery of foramen rotundum to ILT Accessory meningeal artery to ILT (foramen ovale) Anterior, middle deep temporal arteries to OA Via vidian artery to petrous ICA (anterior tympanic artery) Facial artery Angular branch to orbital branches of OA Superficial temporal artery (STA) Transosseous perforators to anterior falx artery Ascending pharyngeal artery Superior pharyngeal artery to cavernous ICA (foramen lacerum) Rami to cavernous ICA branches Inferior tympanic artery to petrous ICA Posterior auricular, OA to petrous ICA Stylomastoid artery Iatrogenic STA-middle cerebral artery (MCA) bypass graft ECA to vertebral artery (VA) Occipital artery Transosseous perforators C1 anastomotic branch (first cervical space) C2 anastomotic branch (second cervical space) Ascending pharyngeal Hypoglossal (odontoid arterial arch) Musculospinal branch Subclavian artery (SCA) to VA Ascending cervical branch C3 anastomotic branch C4 anastomotic branch Extracranial anastomoses Subclavian steal (VA to basilar artery with flow reversal in contralateral VA to distal SCA) Carotid steal (common carotid artery [CCA] occlusion with a patent ECA and ICA and flow reversal in ECA) Direct VA to VA (via segmental branches) ECA to ECA (via contralateral branches) Iatrogenic (SCA to CCA anastomosis, bypass graft) From Osborn AG. Diagnostic cerebral angiography I. Philadelphia: Lippincott Williams & Wilkins; 1999.

inadequate as investigated by diagnostic angiography16 and BTO with or without SPECT, preoperative embolization is contraindicated. Instead, carotid artery resection with reconstruction, though rarely performed in these cases, can be considered.43 If carotid artery reconstruction is impossible, palliative chemotherapy or hospice care may be sought.

Carotid Blowout Syndrome An increasing number of CBS cases are recurrent CBS. In contralateral recurrent cases, if the carotid originally affected by CBS has already been occluded or ligated, contralateral occlusion or ligation is an unavailable option.48 Endovascular occlusion is also contraindicated in any patient where the contralateral carotid artery is already occluded by atherosclerosis, and in patients who have proven CBS bleeding of the ICA/CCA systems but fail BTO. In cases where occlusion is contraindicated, a selfexpandable stent should be placed.47,51 If using a stent-graft (i.e., covered stent), placement in an uninfected, unexposed

area is strongly recommended. Favorable sites are often unavailable in impending or acute CBS. A near certainty of subsequent infection and a high risk of occlusion or re-hemorrhage persists when stenting in unfavorable locations51,64,65 (see Fig. 94-8).

Radplat/Intraarterial Chemotherapy When treating patients with hemostatic disorders, other therapies may be more appropriate because there is higher risk with selective catheterization in these populations. In patients with atherosclerotic femoral anatomy, use of the superficial temporal approach is preferred.58-62 Tumors partially supplied by the ICA system cannot be treated by intraarterial chemotherapy; this would allow access of high-dose highly neurotoxic cisplatin into the intracranial circulation. Such tumors include those of the paranasal sinuses and skull base.66 In recurrent head and neck disease, which may be affected by cisplatin resistance, direct arterial chemotherapy has been reported to be much less effective, though perhaps slightly better than intravenous chemotherapy.53,55 Finally, in untreated disease where RADPLAT is unlikely to be curative, surgery should be considered as the primary therapy to avoid the increasing difficulty of operations in locations where chemoradiation has been previously applied.4 No treatment algorithm has yet been set, but such factors as tumor volume and nodal status, as well as age and T-stage classification, are potential indicators of the efficacy of RADPLAT. In particular, high tumor volume and central necrosis could hinder successful selective catheterization and thereby limit the efficacy of RADPLAT.

EQUIPMENT Standard angiographic equipment is required for all procedures. After a groin sheath is placed, a 4F diagnostic catheter appropriate for cerebrovascular access is used to obtain diagnostic images. Once vessels of interest are demonstrated, activated clotting time (ACT) is checked. Heparin is administered. An exchange Bentson wire or Glidewire is used to safely place a 5F to 8F guiding catheter with an appropriately angled tip. Microcatheters (0.014) and microwire (0.010) are used to obtain selective and superselective images.

Balloon Test Occlusion For BTO, the HyperForm nondetachable balloon (Micro Therapeutics, Irvine, Calif.) is used after diagnostic angiography demonstrates collateral circulation.16 Inflation is carefully performed through a thoroughly purged system. Temporary vascular occlusion is performed as distal as is safe to a point just central to the petrous carotid to reduce the noted increased incidence of dissection when working in this carotid segment. Full electroencephalographic (EEG) monitoring is supplied by the neurologist performing the evaluation while circulation is occluded.

Embolization Embolization procedures may employ particles, acrylate, or coils and are usually performed via a 0.014 microcatheter.

CHAPTER 94 MANAGEMENT OF HEAD AND NECK TUMORS 711 Guglielmi detachable coils (GDCs) may be delivered through this system. PVA particles may also be delivered via this system. Safe and effective particle diameters of the 150 to 250 μm variety offer excellent arteriole occlusion with decreased risk of passing through the capillary bed. Extreme care is used to segregate the embolization equipment from those catheters that may be used for diagnostic image injections after embolization. Onyx or n-butyl cyanoacrylate (NBCA) may also be injected through this system with appropriate catheters.67,68 Embolization performed by using detachable balloons for managing head/ neck cancer–related CBS has been described in depth since it was first reported.69 However, both permanent balloon occlusion and the requisite temporary BTO may add an additional 15% to 20% delayed and 0.4% immediate occurrence of permanent stroke, respectively.43,46,70

Stenting Stents may be balloon-mounted or self-expandable, uncovered or covered. Stent delivery systems continue to evolve. They frequently come packaged with their own delivery catheters and microwires. Close attention is necessary to ensure the access sheath and guiding wires are compatible with the stent delivery system. Appropriate antiplatelet medications are also necessary.51

FIGURE 94-9. Collateral flow from ophthalmic artery to internal carotid artery. Lateral angiogram demonstrates common carotid artery injection and blunt occlusion of internal carotid artery (vertical arrow). Enlarged internal maxillary artery (star) is feeding sphenopalatine segment and enlarged ophthalmic artery (oblique arrow). Supraclinoid internal carotid artery was reconstituted (horizontal arrow).

Selective Catheterization for Intraarterial Chemotherapy The success of selective arterial catheterization for targeted delivery of cisplatin has been increased with the use of CT scan selective arteriography. Although not widely available, this technique allows for the determination of which branch arteries are feeding tumor masses. A CT machine is attached to the standard fluoroscopy gantry, or cone beam techniques may eventually be used for this purpose. After selective arterial catheterization is performed, injections are made while dynamic CT images are acquired. This method allows the angiographer to elucidate tumor blood supply prior to chemotherapy infusion. Direct superficial temporal artery access requires specifically designed hook catheters.59-62

TECHNIQUE Anatomy and Approaches Avoidance of postprocedural complications can be helped by paying close attention to collateral ECA-ICA connections during embolization procedures (Figs. 94-9 to 94-13). Emphasis must be placed on the observation that flow dynamics will be altered by embolization, and collateral ECA to ICA connections may be revealed as the embolization proceeds. Specific attention should be paid to direction of ophthalmic artery flow and potential internal maxillary and occipital artery anastomosis to cerebral circulation. Paragangliomas Typically the ascending pharyngeal artery and its branches are the rostral-caudal blood supply of paragangliomas. Additional feeding arteries include inferior tympanic for tympanic, neuromeningeal for jugular, and musculospinal

FIGURE 94-10. Collateral flow from vertebral artery to external carotid artery (ECA). Lateral angiogram illustrates vertebral artery injection (horizontal arrow). Vertical arrow points to posterior circulation providing entire cerebral perfusion via posterior communicating arteries. Oblique arrow demonstrates lack of filling of central ECA, proving ECA branches communicate with vertebral artery.

to vagal and carotid body.16 This can be supplemented by additional feeding vessels such as the internal maxillary artery or superior and inferior laryngeal arteries in their given territories. Contributions from the occipital and posterior auricular arteries, vertebral arteries, or meningeal

712 PART 1 VASCULAR INTERVENTIONS · SECTION 11 NEUROVASCULAR AND HEAD AND NECK INTERVENTION

FIGURE 94-13. Angiographic image demonstrating ophthalmic artery supplied by distal internal maxillary artery branch.

FIGURE 94-11. Collateral flow from external carotid artery to vertebral artery. Lateral view from angiogram demonstrates injection of common carotid artery. Multiple oblique arrows outline vertebral artery shadow, which is reconstituted via muscular branches originating from external carotid artery.

tumor supplies only its own respective region of the tumor, as opposed to monocompartment supply, where each artery supplies the entire tumor.3,13 Angiography should therefore thoroughly investigate the bilateral ascending pharyngeal arteries, as well as the ipsilateral vertebral, internal carotid, distal external carotid, posterior auricular, and occipital arteries.63 Juvenile Nasal Angiofibroma JNAs are supplied by branches of the internal maxillary (sphenopalatine and descending palantine branches) and ascending pharyngeal artery. As tumor spreads, recruitment of other arteries include the ascending pharyngeal/ ascending palantine.66 Ipsilateral/contralateral ICA/ECA involvement has commonly been demonstrated.29 It is recommended to perform angiography with contralateral injection of the CCA and/or ICA and then ipsilateral injection of the ICA and/or CCA to demonstrate the tumor’s extension. In the ECA, attention should be given to anastomoses of the internal maxillary and ascending pharyngeal arteries with the intracranial, intraorbital, and internal carotid arteries.

FIGURE 94-12. Lateral view from internal carotid artery injection demonstrating absence of ophthalmic artery filling prior to embolization procedure. This should raise suspicion for aberrant ophthalmic artery supply, which should be confirmed prior to embolic material is injected.

branches of the ICA are also possible and depend on tumor location and extension. Paragangliomas that are solely tympanic are often supplied mainly by the inferior tympanic artery, which eases intraoperative hemostatic control and decreases the need for preoperative embolization. Most (83%) paragangliomas, however, have a “multicompartment” vascular supply; that is, each feeding artery for the

Carotid Blowout Syndrome CBS can occur at virtually any portion of the carotid system, but nearly half of cases have been reported at the ICA (43.6%). Other sites include the ECA (23.4%), CCA (11.7%), buccal artery (3.2%), inferior (3.2%) and superior thyroid artery (2.1%), internal mammary artery (2.1%), innominate artery (1.1%), facial artery (1.1%), and aorta (1.1%), as well as rare case reports of the internal maxillary artery.47 When considering possible CBS sites, it is also necessary to investigate the tumor itself (5.3% of reported cases), as well as the internal jugular vein, which although not technically a part of the carotid artery system is still also a potential source of rupture (Figs. 94-14 to 94-16). Radplat/Intraarterial Chemotherapy The physical anatomy of head/neck tumors does not necessarily reflect the vascular anatomy, as shown by Terayama

CHAPTER 94 MANAGEMENT OF HEAD AND NECK TUMORS 713 et al.,71 who studied the vascular supply of selectively catheterized laryngeal and hypopharyngeal tumors as demonstrated on their CT-visualized enhancement during arteriography. Nearly all ipsilaterally located laryngeal tumors were supplied by the ipsilateral superior thyroid and/or

laryngeal arteries, but less than half of ipsilaterally located hypopharyngeal cancers were supplied by ipsilateral vasculature. Almost one quarter of midline-crossing laryngeal tumors, on the other hand, were supplied ipsilaterally. Only 12 of 25 midline-crossing laryngeal tumors were actually supplied by the bilateral superior thyroid arteries. Inclusion of other arterial branches, including the inferior thyroid and lingual arteries, was necessary to completely catalog the arterial supply of the tumors in question (Fig. 94-17). To avoid inadequate distribution of cisplatin, complete investigation of vascular supply is an essential part of treatment planning.

Technical Aspects

FIGURE 94-14. Blowout. Bleeding was noted after incision and drainage of a supraclavicular abscess. Axial contrast-enhanced computed tomography demonstrates extravasation of contrast material into superficial wound. Pressure dressing was applied immediately, and patient was stabilized.

FIGURE 94-15. Pseudoaneurysm. Anteroposterior view from left subclavian angiogram with calibrated catheter demonstrates large subclavian pseudoaneurysm.

A

B

General Embolization Procedures in Head and Neck Standard preoperative precautions are observed. This includes verification of normal renal function and coagulation profile. Some therapeutic agents would produce too much pain in an awake patient, and hemorrhage complicating some procedures often compromises the patient’s airway and hemodynamics. For the preceding reasons, these procedures are frequently undertaken with the aid of general anesthesia to increase safety and shorten the procedure. If general anesthesia is planned, appropriate cardiopulmonary evaluation is advised. Frequently general anesthesia is initiated at the beginning of the diagnostic procedure, maintained through the embolization phase of

FIGURE 94-16. Stented pseudoaneurysm. Coronal image was reformatted from a computed tomographic angiogram after stenting of subclavian artery.

FIGURE 94-17. Computed tomographic angiogram (A) and angiographic image (B) showing superficial temporal artery access and contrast injection for delineation of tumor blood supply. (From Mitsudo K, Shigetomi T, Fujimoto Y, et al: Organ preservation with daily concurrent chemoradiotherapy using superselective intra-arterial infusion via a superficial temporal artery for T3 and T4 head and neck cancer. Int J Radiat Oncol Biol Phys 2011;79:1428– 35, Fig. 4.)

714 PART 1 VASCULAR INTERVENTIONS · SECTION 11 NEUROVASCULAR AND HEAD AND NECK INTERVENTION the angiogram, and carried into the operative suite. This approach is most useful for the younger patient population affected by JNA. Endovascular intervention is tailored to the specific disease entity. However, the majority of cases begin in the same fashion. Standard endovascular access techniques are used to obtain safe diagnostic images of the desired circulation. Therapy is usually accomplished through a microcatheter. This may be either a flow-directed-variety catheter over a microwire or a hybrid catheter. These are stabilized by placement of a larger-diameter guide catheter. The diagnostic catheter is exchanged for the guide catheter. To reduce embolic complications, heparin is administered prior to microcatheter deployment. The ACT is monitored with the goal of attaining a safe working range of 1.5 to twice normal. The microcatheter is then used to obtain diagnostic images and deliver various therapeutic agents. This includes Gelfoam, PVA, NBCA, Onyx, alcohol, or coils. Of particular importance, embolic materials should be placed in the ECA in a series of small volumes during the systolic phase; the ECA lacks diastolic forward flow, which would increase the likelihood of material reflux during this phase63 (Fig. 94-18). Contrast injections demonstrating stagnant flow within the target vessels is considered a technical success. Care should be taken regarding the ophthalmic artery,11 since inappropriate occlusion frequently leads to blindness. If a temporalis muscle flap is to be used for reconstruction, it is necessary to communicate this fact to the radiologist so care is taken not to compromise the blood supply to the temporalis muscle.29 Heparinization is usually recommended during catheterization to prevent clot formation. Corticosteroids are used only with very prolonged procedures or when a significant amount of inflammation is expected, depending on the embolization target and extent of vascular territory involved.63

FIGURE 94-18. Polyvinyl alcohol (PVA) embolization of hemorrhagic malignancy. Lateral view from angiogram demonstrates PVA embolization with coaxial guiding catheter and microcatheter. Contrast agent and PVA mixture were injected until a stagnant column of contrast material was demonstrated. Care is taken to avoid extension of contrast column to parent artery, which would signal reflux of PVA particles.

Balloon Test Occlusion BTO is an exception to the usual approach of general anesthesia. This procedure requires an awake patient because they are frequently monitored via EEG, and their neurologic status is continually monitored throughout the standard 30-minute occlusion session. A standard diagnostic exam is performed to assess collateral circulation by bilateral carotid angiogram. If inadequate collateral vessels are demonstrated, the study is terminated at this point to avoid possible seizure-induced ischemia.16 If adequate collateral vessels are demonstrated, a balloon is deployed in the proximal ICA via a guide catheter. These patients are heparinized to reduce embolic complications from the stagnant blood column past the balloon. Adjunct methods may be used to increase the sensitivity of the test.42,72 It is important to note the potential risks associated with prolonged large-diameter balloon inflation in a diseased vessel. Stenting Standard diagnostic angiography is performed following contrast CT scan, computed tomographic angiography (CTA), or MRI/MRA with contrast. Once the vessel abnormality is localized, an appropriate stent is chosen predicated on lesion length and parent vessel diameter. Additional considerations include location within the vascular tree and status of the surrounding soft tissue, which may influence the choice of a covered or bare metal stent. This topic is discussed later. The most significant potential complication is arterial dissection encountered while maneuvering the sheathed stent system across the lesion prior to final deployment. This risk is highest with the larger covered stent systems in narrowed and irregular parent vessels. Patients with atherosclerosis or who have been treated with prior radiation therapy have narrower and more irregular vessels, posing higher dissection risk during stent placement (Figs. 94-19 and 94-20). Radplat/Intraarterial Chemotherapy The common intraarterial chemoradiation regimen known as RADPLAT consists of four weekly cycles of intraarterial chemotherapy, with radiotherapy during the first week. To administer each cycle of chemotherapy, the carotid anatomy of a well-hydrated patient is assessed by transfemoral digital subtraction angiography. Superselective catheterization of the involved feeding arteries is performed, followed by CT scan. CT concurrent with angiography can help identify feeding arteries. Upon successful catheterization, an intraarterial infusion of cisplatin is rapidly administered (150 mg/m2 over 3-5 minutes), which is immediately followed by catheter removal to reduce the likelihood of clot formation. CT during the infusion is useful in confirming drug distribution. During the cisplatin infusion, a separate slower intravenous administration of sodium thiosulfate is begun (9 g/m2 over 15-20 minutes, then 12 g/m2 over 6 hours). With malignancies that cross the midline, bilateral infusions may be used. Superficial temporal artery access techniques have also been used. Local anesthesia is used for the arterial cutdown. A hooked catheter is placed in the target vessel using standard techniques.

CHAPTER 94 MANAGEMENT OF HEAD AND NECK TUMORS 715

FIGURE 94-19. Lateral angiogram demonstrating an open design nitinol stent used to reduce risk of bleeding from a diseased vessel.

hypervascular tumors. In several studies where embolization therapy involved some component of direct percutaneous acrylate injection, at least some degree of devascularization was observed in over 90% of patients, without any permanent major or minor complications.21,72,73 Practitioners therefore report increasing use of this technique. In addition to its high success and safety rate, advantages include a greater chance of total devascularization and an “intratumoral casting” of the tumor’s vascular bed.48,74 This “casting” by the colored acrylate glue not only likely aids in achieving complete occlusion but also allows for better distinction of the tumor and its vasculature from surrounding normal tissue during surgery. Despite this high rate of success, however, drawbacks remain. Clinically, the most significant disadvantage is control of glue migration. In two cases, the acrylate glue has been observed to migrate to the middle cerebral artery,75 once with no permanent effect but once with subsequent mortality. In the former case, it was attributed to inexperience that led to overinjection of the tumor bed; in the latter, it was related to poor visualization of the vasculature due to a tumor that was already opaque from the procedure. In a third case, glue migration was the cause of blindness when it occluded an ophthalmic artery that was part of the supply to another JNA that had eroded into the orbit. Therefore, great care should be taken during angiographic evaluation of the vasculature, as well as during the subsequent embolization. In cases where the ophthalmic artery is part of the feeding supply, direct glue injection is contraindicated. Other drawbacks are regulatory, technical, and surgical. It has been suggested that intratumoral injection of glue may lead to adherent tumor, making it more difficult to surgically resect; perhaps it should be reserved for larger tumors where endovascular embolization of every feeding artery is difficult to achieve, or where endovascular embolization fails.11 As we gather more experience with direct percutaneous acrylate injection for hypervascular tumors, its role in embolization therapy will be better understood.30,67,68

Carcinoma Involving Carotid Artery

FIGURE 94-20. Anteroposterior angiogram showing stent stabilizing severely diseased common and internal carotid arteries.

CONTROVERSIES Paraganglioma Cyanoacrylate Embolization Direct percutaneous injection of acrylate glue may begin to replace or at least complement intraarterial embolization of

Carotid Occlusion and Tumor Embolization Followed by Resection Although poor collateral circulation often contraindicates resection of the carotid when invaded by carcinoma, an alternative endovascular/surgical method devised by Nussbaum et al. warrants further investigation.76 This method involved placement of an endovascular stent in the carotid artery of a patient whose carotid would not tolerate ligation or reconstruction. One month later after the stent formed a complete neoendothelial barrier within the diseased carotid, Nussbaum et al. were able to successfully perform a longitudinal resection and dissection of the tumor surrounding the diseased portion of the carotid. If this method shows continued success, it could be added to the arsenal of endovascular management for head/neck malignancy.43 Carotid Blowout Syndrome Generally, permanent embolic materials have had reported success in controlling CBS hemorrhage; absorbable gelatin particles may also have a similar efficacy, with the added

716 PART 1 VASCULAR INTERVENTIONS · SECTION 11 NEUROVASCULAR AND HEAD AND NECK INTERVENTION advantage of allowing eventual reestablishment of local circulation to the normal tissue. Kakizawa et al.70 hypothesize that this absorbability could result in reduced focal necrosis and decreased pain. Furthermore, absorption of the particles could allow for repeat endovascular therapy, especially in the setting of contralateral recurrent carotid disease. A randomized prospective trial comparing permanent embolic materials with the absorbable gelatin particles would be necessary to validate these findings. Another controversy regarding management of CBS concerns the use of stents. While many advocate the use of therapeutic tumor embolization rather than stenting as the preferred course for CBS treatment, Chaloupka et al. now promote the use of uncovered stents, especially overlapping and in conjunction with coil embolization, as the preferred therapy. The group, possibly one of the most experienced groups in endovascular CBS management, argues that stenting maintains rates of hemostatic control comparable to the rates achieved by embolization and may even reduce the rate of stroke. Also, while the use of uncovered stents alone had a rate of treatment failure similar to the rate for therapeutic embolization alone in one of their series, Chaloupka’s group surmises that combining the two techniques may improve on the rates of recurrence, since in the same series, none of the patients under the combined regimen experienced treatment failure. Additional potential advantages of stenting include the immediate ability to treat CBS patients who have inadequate collateral circulation, reinforcement of the damaged arterial wall, and preservation of bilateral carotid artery patency. Maintaining patency could be useful in treating future carotid disease that occurs contralateral to the original site of CBS.51 Others argue that the difficulties in managing complications associated with stenting outweigh the advantages. Such complications include infection, rebleeding, and thrombosis. Indeed, reports in the literature concerning stent-grafts (i.e., covered stents) have demonstrated infection-related complications that lead to thrombosis, hemorrhage, stent extrusion, and in one case a brain abscess.64,65 In a study of stent-graft therapy for CBS by Chang et al., stent-grafts became completely occluded secondary to infection in 3 out of 8 patients.64 It is worthwhile to note, however, that the reported incidence of infection associated with covered stents is greater than with uncovered stents. Uncovered stents require smaller-caliber delivery systems and are not as hindered by size limitations as covered stents. Uncovered stents cannot provide an immediate virtual bypass of the affected site as covered stents do. It may be worthwhile to reprioritize the role of stenting in the management of CBS, especially when considering uncovered stents in combination with coil embolization and, perhaps later, stent-grafts if advances in material sciences produce less thrombogenic graft material.

OUTCOMES Paragangliomas Eighty percent of preoperatively embolized paragangliomas experience a reduction in tumor size.13 Often the size reduction is as high as 25%,16 which reduces surgical blood loss,

intraoperative manipulation of surrounding structures, and the need for intra- or postoperative transfusion. In their surgical management of large paragangliomas (>3 cm) with preoperative embolization, Persky et al.16 report a mean blood loss of the equivalent of only 1 transfusion unit.

Juvenile Nasal Angiofibromas Before the common use of endovascular techniques, average blood loss for JNA amounted to approximately 2000 mL per case; that figure has now been reduced to less than 1000 mL.24 With preoperative embolization, intraoperative time has been significantly reduced, as has tumoral size.63 This in turn has led to a higher rate of complete surgical resection and thereby a lower rate of recurrence and repeated recurrence.

Carcinoma Involving Carotid Artery Carotid Occlusion and Tumor Embolization Followed by Resection In terms of overall survival, resection of carcinoma involving the carotid artery is rarely curative and carries a risk of perioperative mortality. With reconstruction or preoperative embolization, the risk has been reported as less than 5%.2,40,43 The disease-free survival rate at 1 year following resection has been reported as 60%. The prognosis is worse for advanced recurrent or residual disease. Even with carotid reconstruction, preoperative embolization, or radiotherapy, resection of carcinoma involving the carotid has a 25% to 40% survival at 1 year, with a reported 6.3month median survival rate. Given the nature of the disease and its prognosis in this patient population, many patients with carotid involvement may not elect for aggressive surgical therapy. For those patients electing aggressive therapy, resection may provide greater-than-average survival even with residual or recurrent disease.43 Of those who do undergo resection, endovascular techniques have led to lower rates of morbidity. By using BTO to identify patients who require carotid reconstruction, the 30% risk of stroke from carotid ligation without reconstruction2,40 has been reduced to 10% to 20% with reconstruction.42 As for patients who pass BTO, the stroke rate with ligation is favorable compared to that seen in ligation patients who failed BTO.77 The use of preoperative permanent occlusion has lowered that rate to 10%. Preoperative endovascular occlusion decreases blood loss by precluding the need for intraoperative heparinization of a shunt and minimizes operative time by avoiding reconstruction and the need for well-vascularized coverage of a graft. In addition, by identifying patients who will tolerate ligation without carotid reconstruction, the option of radiotherapy can remain for these patients without worrying about damage to the integrity of a reconstructive graft (Figs. 94-21 and 94-22). Carotid Blowout Syndrome Using a variety of methods, endovascular management of CBS has greatly reduced morbidity and mortality. Minor and major neurologic morbidity has been reported to be 10% to 17% and 0% to 7%, respectively; mortality has also been reduced significantly (2%-14%).48 With endovascular management, however, recurrence of CBS is more

CHAPTER 94 MANAGEMENT OF HEAD AND NECK TUMORS 717

Radplat/Intraarterial Chemotherapy

FIGURE 94-21. Endovascular carotid occlusion. Lateral view demonstrates coils, with a proximal glue cast outlining internal carotid artery.

FIGURE 94-22. Coils. Scout view from head computed tomogram shows coils along course of internal carotid artery.

Early studies of intravenous cisplatin delivery for head and neck malignancies has shown an overall response rate of 40% for previously untreated disease and an average of 28% for recurrent disease.53 On the other hand, the vast majority of nonrandomized studies pertaining to intraarterial chemotherapy with radiotherapy showed promising rates of complete and partial response—up to 90% and 81%, respectively.4 In particular, local and regional control has appeared excellent. In a study of 213 patients,78 only 6% and 3% developed local and regional recurrence, respectively, while 18% were observed to develop distant metastasis. The relatively high rate of metastasis, possibly unmasked by the local therapy, had led to the observation that perhaps a systemic treatment component would be also be necessary. A phase III prospective randomized clinical trial79-81 comparing intraarterial and intravenous chemoradiation for head/neck cancer was recently completed, with results that appear at odds with previous studies. Preliminary reports of the phase III trial suggest that in 240 patients with stage IV inoperable head and neck cancer, no differences in tumor control, metastasis, or overall survival were observed between intraarterial and intravenous delivery arms. A point of debate has arisen concerning whether the intraarterial treatment was effectively administered based on the anatomy of the treated tumors. Based on preliminary results, the Dutch investigators recommend intravenous chemoradiation as standard therapy. It should be noted that despite comparable rates of control and survival, systemic complications were observed at a lower rate with intraarterial administration. This appears to confirm the results of earlier nonrandomized studies. The low toxicity profile would not preclude RADPLAT’s use as a preoperative adjunct to surgery or as a potential prognostic indicator in the management of head/ neck tumors. As a preoperative adjunct to surgery (i.e., neoRADPLAT), RADPLAT has been shown to reduce tumor burden. This has led to a reduction in the extent of subsequent resection, with improved postoperative organ function. It has allowed for surgical removal of otherwise unresectable tumors. Kovacs has reported overall survival rate estimates to be significantly higher for patients who demonstrate complete tumor response to RADPLAT (85%) than those who have poorer responses (33%-64%).82 As research continues on intraarterial chemotherapy, another potentially efficacious local therapy being studied is percutaneous image-guided radiofrequency tumor ablation. Given the advantages of local minimally invasive therapy, we should continue to consider all options for our patients.

COMPLICATIONS General Embolization in Head and Neck

likely and has been reported from 0% to as high as 33%,70 with 65% resulting from ipsilateral or contralateral progressive disease and 35% from ipsilateral treatment failure. Recurrent CBS can be managed successfully. Chaloupka et al.72 reported only one of 20 events ending in mortality, and no neurologic or ophthalmologic complications were noted.

Many postembolization complications are associated with inappropriate occlusion of the vasculature, which can occur either by inadvertent reflux or through ECA-vertebrobasilar or ECA-ICA anastomoses. This has led to stroke, cerebral ischemia, blindness, and cranial nerve (CN) palsies corresponding to affected vasculature (e.g., CN IX-XII, supplied by ascending pharyngeal artery; CN VII, supplied by middle

718 PART 1 VASCULAR INTERVENTIONS · SECTION 11 NEUROVASCULAR AND HEAD AND NECK INTERVENTION meningeal and stylomastoid arteries13). Catheterization also increases the risk of stroke and blindness. In general, however, endovascular embolization treatment of head/ neck lesions using modern techniques has less than 1% rate of severe neurologic deficit.24 Other potential major complications of embolization include necrosis of the mucosa, tongue, and/or skin, ulceration, peripheral neuropathies, and acute muscle ischemia. Muscle is highly sensitive to ischemia, so the latter can result in intense muscular pain. Potential complications more specific to the management of paragangliomas include carotid sinus syndrome and excessive catecholamine secretion, which could be managed with alpha blockade premedication. Some minor transient complications may include postembolization fever or ear pain due to tumor ischemia.40 Vasospasm during catheterization can be managed locally by percutaneous nitroglycerin without fear of systemic hypotension.63

Carcinoma Involving Carotid Artery Carotid Occlusion and Tumor Embolization Followed by Resection Stroke continues to be a concern for patients undergoing this procedure, despite lowered rates with preoperative endovascular occlusion. Other complications may occur during surgical resection. When confronting massive neck malignancy, despite the use of preoperative permanent occlusion, massive hemorrhage and hypotension is still possible during surgical resection.40 Also, one report of balloon migration during surgical manipulation of the petrous carotid artery resulted in postoperative stroke. In coil-embolized carotid artery systems, dislodging of the clot formation is a theoretical risk of intraoperative manipulation. Detachable balloons are no longer available in the United States. Carotid Blowout Syndrome The neurologic morbidity from CBS has been greatly reduced. Reported complications include occasional cases of transient ischemia,46 Horner syndrome,47 temporary hemiplegia, or permanent hemiparesis.70 Occlusion of the facial artery may cause transient facial pain controllable by analgesics.70 Covered stent placement has been associated with infection, occlusion, and stroke.83

Radplat/Intraarterial Chemotherapy Intraarterial chemotherapy with radiotherapy can result in neurologic sequelae due to catheterization-related injury, chemotoxicity, or general cardiopulmonary effects such as pneumonia. Higher complication rates have been reported at centers with lower case volumes.4,78,84-88

POSTPROCEDURE AND FOLLOW-UP CARE General Embolization in Head and Neck If significant inflammation is expected, corticosteroid medication may be administered during and following endovascular therapy. Usually, however, postembolization

inflammation is expected to decrease after 1 to 2 days, and follow-up surgery is scheduled accordingly. With paragangliomas, surgical resection is recommended within the period after inflammation has been allowed to decrease (1-2 days postembolization) and before feeding vessels begin to recanalize (no later than 2 weeks postembolization). For JNAs, timely resection within 12 to 48 hours is recommended, before the onset of any postembolization revascularization inflammation and edema; this is particularly important to the preservation of an already tenuous airway in the rare case where angiofibroma involves the larynx (laryngeal angiofibroma).13,63

Carcinoma Involving Carotid Artery Carotid Occlusion and Tumor Embolization Followed by Resection After preoperative occlusion patients awaiting tumor resection are monitored in the intensive care unit for 3 days of heparinization and fluid dynamic monitoring. Surgical resection of the tumor can be performed 2 to 6 weeks later.42,43 This equilibration period for the newly embolized arteries helps patients tolerate sudden changes in blood pressure during anesthesia and acute blood loss during surgery. Time interval between carotid occlusion and surgery reduces embolic complications. This may be secondary to clot maturity at the ends of the occluded carotid. Waiting allows coils to stabilize and fix to the vascular wall. For previously untreated carcinomas that have undergone resection, postoperative radiation therapy is indicated. Carotid Blowout Syndrome If CBS has been treated by endovascular occlusion, patients should be hemodynamically and neurologically monitored in the intensive care unit, with heparinization for 48 hours and daily aspirin therapy thereafter.48 In patients with stent treatment, antiplatelet therapy can be administered cautiously. If the patient is at high risk for re-hemorrhage (i.e., in the impending or acute categories of CBS), heparinization should be avoided during and after stent placement.46 On the other hand, if the patient is in the threatened category or receiving a more thrombogenic stent-graft instead of an uncovered stent, the patient should be immediately placed on clopidogrel with aspirin and maintained following stent-graft placement. Patients receiving stent-grafts should also receive prophylactic antibiotics.51,64,65 Given the debate over the role of stents in the management of CBS, some follow the emergent stent placement with permanent ligation and vascular bypass surgery. This plan is more strongly advocated if stent placement occurs in a contaminated area. This is also a consideration in patients with stent-grafts, which have a higher occlusion rate.

Radplat/Intraarterial Chemotherapy After each cycle of chemotherapy, patients should receive intravenous hydration using 1 liter of normal saline with 20 mEq of potassium chloride and 2 g of magnesium sulfate over 6 hours. Patients should also receive 4 mg dexamethasone every 6 hours, either intravenously or orally, until the next morning. Given potential side effects, some patients may require a percutaneous endoscopic gastrostomy (PEG) tube if swallowing problems persist following the

CHAPTER 94 MANAGEMENT OF HEAD AND NECK TUMORS 719 completion of RADPLAT, or if hematologic disturbances are observed during therapy. Partial responders with larger nodal malignancies may require selective salvage neck dissection. Complete responders do not need follow-up neck dissection, since their nodes have consistently been found to be negative at neck dissection.4,53,55

•

Selective arterial catheterization for chemotherapy may be appropriate in some cases.

•

Management of complications may require further embolization. Use of covered stents to maintain carotid artery patency instead of endovascular occlusion may help control local complications. This approach may both maintain adequate cerebral perfusion and allow future intervention.

KEY POINTS •

Appropriate therapy for benign tumors of the head and neck may include surgery, endovascular therapy, superficial therapy, or direct puncture. Frequently a combined team approach of experienced physicians offers the best results with the least morbidity.

•

Attention to collateral flow from the external carotid artery to the internal carotid artery and the vertebral artery system is crucial to avoid complications from embolic therapy.

•

In cases where a paraganglioma is the suspected diagnosis, a preintervention diagnostic bilateral carotid angiogram reduces the likelihood of unforeseen hemorrhagic complications and helps detect bilateral tumors.

•

Juvenile nasal angiofibromas are best approached with a preoperative angiogram and possible embolization.

•

Late-stage head and neck carcinoma is a complex clinical problem that may require surgery, chemotherapy, radiation therapy, and endovascular intervention.

•

Contrast-enhanced computed tomography or magnetic resonance imaging best demonstrates the extent of carotid artery involvement by tumor. It assists in surgical and endovascular treatment planning.

•

Diagnostic angiography with balloon test occlusion is the next step when planning treatment for patients with a carotid artery surrounded by malignancy.

▶ SUGGESTED READINGS Ackerstaff AH, Rasch CR, Balm AJ, et al. Five-year quality of life results of the randomized clinical phase III (RADPLAT) trial, comparing concomitant intra-arterial versus intravenous chemoradiotherapy in locally advanced head and neck cancer. Head Neck 2011 Aug 4. doi: 10.1002/hed.21851. [Epub ahead of print] Chaudhary N, Gemmete JJ, Thompson BG, Pandey AS. Intracranial endovascular balloon test occlusion—indications, methods, and predictive value. Neurosurg Clin N Am 2009;20(3):369–75. Cherekaev VA, Golbin DA, Kapitanov DN, et al. Advanced craniofacial juvenile nasopharyngeal angiofibroma. Description of surgical series, case report, and review of literature. Acta Neurochir (Wien) 2011; 153(3):499–508. Epub 2011 Jan 28. Mendenhall WM, Amdur RJ, Vaysberg M, et al. Head and neck paragangliomas. Head Neck 2011;33(10):1530–4. doi: 10.1002/hed.21524. Epub 2010 Nov 4. Sekhar LN, Biswas A, Hallam D, et al. Neuroendovascular management of tumors and vascular malformations of the head and neck. Neurosurg Clin N Am 2009;20(4):453–85. Zussman B, Gonzalez LF, Dumont A, et al. Endovascular management of carotid blowout: institutional experience and literature review. World Neurosurg 2011;78:109–14.

The complete reference list is available online at www.expertconsult.com.

CHAPTER 94 MANAGEMENT OF HEAD AND NECK TUMORS719.e1 REFERENCES 1. Weiss I, O TM, Lipari BA, et al. Current treatment of parotid hemangiomas. Laryngoscope 2011;121(8):1642–50. 2. Weissler M. Head and neck cancer: emerging perspectives. Otolaryngol Head Neck Surg 2004;130(3):395. 3. Broomfield S, Bruce I, Birzgalis A, et al. The expanding role of interventional radiology in head and neck surgery. JR Soc Med 2009;102(6):228–34. 4. Alkureishi LW, de Bree R, Ross GL. RADPLAT: an alternative to surgery? Oncologist 2006;11(5):469–80. 5. Brook AL, Gold MM, Miller TS, et al. CT-guided radiofrequency ablation in the palliative treatment of recurrent advanced head and neck malignancies. J Vasc Interv Radiol 2008;19(5):725–35. 6. Nall AV, Stringer SP, Baughman RA. Vascular leiomyoma of the superior turbinate: first reported case. Head Neck 1997;19(1):63–7. 7. Marden FA, Calilao GC, Guzman G, et al. Glossal angiomyoma: imaging findings and endovascular treatment. Head Neck 2004;26(12): 1084–8. 8. Tsutsumi M, Kazekawa K, Tanaka A, et al. Rapid expansion of benign scalp neurofibroma caused by massive intratumoral hemorrhage–case report. Neurol Med Chir (Tokyo) 2002;42(8):338–40. 9. Allcutt DA, Hoffman HJ, Isla A, et al. Acoustic schwannomas in children. Neurosurgery 1991;29(1):14–8. 10. Abramowitz J, Dion JE, Jensen ME, et al. Angiographic diagnosis and management of head and neck schwannomas. AJNR Am J Neuroradiol 1991;12(5):977–84. 11. Liu DG, Ma XC, Li BM, et al. Clinical study of preoperative angiography and embolization of hypervascular neoplasms in the oral and maxillofacial region. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 2006;101(1):102–9. 12. Lewin JS, Merkle EM, Duerk JL, et al. Low-flow vascular malformations in the head and neck: safety and feasibility of MR imaging-guided percutaneous sclerotherapy–preliminary experience with 14 procedures in three patients. Radiology 1999;211(2):566–70. 13. Rao AB, Koeller KK, Adair CF. From the archives of the AFIP. Paragangliomas of the head and neck: radiologic-pathologic correlation. Armed Forces Institute of Pathology. Radiographics 1999;19(6): 1605–32. 14. Van den Berg R. Imaging and management of head and neck paragangliomas. Eur Radiol 2005;15(7):1310–8. 15. Bradshaw JW, Jansen JC. Management of vagal paraganglioma: is operative resection really the best option? Surgery 2005;137(2): 225–8. 16. Persky MS, Setton A, Niimi Y, et al. Combined endovascular and surgical treatment of head and neck paragangliomas–a team approach. Head Neck 2002;24(5):423–31. 17. Boedeker CC, Ridder GJ, Schipper J. Paragangliomas of the head and neck: diagnosis and treatment. Fam Cancer 2005;4(1):55–9. 18. Sanna M, Shin SH, De Donato G, et al. Management of complex tympanojugular paragangliomas including endovascular intervention. Laryngoscope 2011;121(7):1372–82. 19. Myssiorek D, Rinaldo A, Barnes L, et al. Laryngeal paraganglioma: an updated critical review. Acta Otolaryngol 2004;124(9):995–9. 20. van den Berg R. Imaging and management of head and neck paragangliomas. Eur Radiol 2005;15(7):1310–8. 21. Abud DG, Mounayer C, Benndorf G, et al. Intratumoral injection of cyanoacrylate glue in head and neck paragangliomas. AJNR Am J Neuroradiol 2004;25(9):1457–62. 22. Chaloupka JC, Mangla S, Huddle DC, et al. Evolving experience with direct puncture therapeutic embolization for adjunctive and palliative management of head and neck hypervascular neoplasms. Laryngoscope 1999;109(11):1864–72.

23. Zeitler DM, Glick J, Har-El G. Preoperative embolization in carotid body tumor surgery: is it required? Ann Otol Rhinol Laryngol 2010;119(5):279–83. 24. Valavanis A, Christoforidis G. Applications of interventional neuroradiology in the head and neck. Semin Roentgenol 2000;35(1): 72–83. 25. Scholtz AW, Appenroth E, Kammen-Jolly K, et al. Juvenile nasopharyngeal angiofibroma: management and therapy. Laryngoscope 2001; 111(4 Pt 1):681–7. 26. Nicolai P, Schreiber A, Bolzoni Villaret A. Juvenile angiofibroma: evolution of management. Int J Pediatr 2012;2012:412545. 27. Kalani MYS, Kalani MA, Kalb S, et al. Craniofacial approaches to large juvenile angiofibromas. J Neurosurg Pediatr 2011;8(1):71–8. 28. Alvarez FL, Suárez V, Suárez C, et al. Multimodality approach for advanced-stage juvenile nasopharyngeal angiofibromas. Head Neck 2012. 29. Fagan JJ, Snyderman CH, Carrau RL, et al. Nasopharyngeal angiofibromas: selecting a surgical approach. Head Neck 1997;19(5):391–9. 30. Elhammady MS, Johnson JN, Peterson EC, et al. Preoperative embolization of juvenile nasopharyngeal angiofibromas: transarterial versus direct tumoral puncture. World Neurosurg 2011;76(3-4):328–34; discussion 263–5. 31. Moulin G, Chagnaud C, Gras R, et al. Juvenile nasopharyngeal angiofibroma: comparison of blood loss during removal in embolized group versus nonembolized group. Cardiovasc Intervent Radiol 1995;18(3): 158–61. 32. Herman BR, Bublik M, Ruiz J, et al. Endoscopic embolization with onyx prior to resection of JNA: A new approach. Int J Pediatr Otorhinolaryngol 2011;75(1):53–6. 33. Glade RS, Richter GT, James CA, et al. Diagnosis and management of pediatric cervicofacial venous malformations: retrospective review from a vascular anomalies center. Laryngoscope 2010;120(2):229–35. 34. Song JK, Niimi Y, Berenstein A. Endovascular treatment of hemangiomas. Neuroimaging Clin N Am 2007;17:165–73. 35. Yuan KH, Li Q, Yu WL, et al. Successful combination therapy for severe infantile hemangiomas: case report and literature search. Photomed Laser Surg 2009;27(6):973–7. 36. Jianhong L, Xianliang H, Xuewu J. Transcatheter arterial embolization in the treatment of extensive maxillofacial hemangioma in children. World J Surg 2005;29(12):1550–6. 37. Kang GCW, Song C. Forty-one cervicofacial vascular anomalies and their surgical treatment–retrospection and review. Ann Acad Med Singap 2008;37(3):165–79. 38. Kohout MP, Hansen M, Pribaz JJ, et al. Arteriovenous malformations of the head and neck: natural history and management. Plast Reconstr Surg 1998;102(3):643–54. 39. Jeong HS, Baek CH, Son YI, et al. Treatment for extracranial arteriovenous malformations of the head and neck. Acta Otolaryngol 2006;126(3):295–300. 40. Thorisson HM, Johnson M. Endovascular Management of Extracranial Carotid and Vertebral Disease. Neurosurg Clin N Am 2009;20(4): 487–506. 41. He XB, Li JJ, Chen YH, et al. Treatment of recurrent head and neck carcinoma involving the carotid artery: carotid reconstruction with ePTFE graft. Eur Arch Otorhinolaryngol 2011;268(12):1817–22. 42. Johnson MH, Chiang VL, Ross DA. Interventional neuroradiology adjuncts and alternatives in patients with head and neck vascular lesions. Neurosurg Clin N Am 2005;16(3):547–60, vii. 43. Freeman SB. Advanced cervical metastasis involving the carotid artery. Curr Opin Otolaryngol Head Neck Surg 2005;13(2):107–11. 44. Yousem DM, Gad K, Tufano RP. Resectability issues with head and neck cancer. AJNR Am J Neuroradiol 2006;27(10):2024–36.

719.e2PART 1 VASCULAR INTERVENTIONS · SECTION 11 NEUROVASCULAR AND HEAD AND NECK INTERVENTION 45. Yousem DM, Hatabu H, Hurst RW, et al. Carotid artery invasion by head and neck masses: prediction with MR imaging. Radiology 1995;195(3):715–20. 46. Lesley WS, Chaloupka JC, Weigele JB, et al. Preliminary experience with endovascular reconstruction for the management of carotid blowout syndrome. AJNR Am J Neuroradiol 2003;24(5):975–81. 47. Cohen J, Rad I. Contemporary management of carotid blowout. Curr Opin Otolaryngol Head Neck Surg 2004;12(2):110–5. 48. Chaloupka JC, Roth TC, Putman CM, et al. Recurrent carotid blowout syndrome: diagnostic and therapeutic challenges in a newly recognized subgroup of patients. AJNR Am J Neuroradiol 1999;20(6): 1069–77. 49. Miller T, Burns J, Farinhas J, et al. Covered stents safely utilized to prevent catastrophic hemorrhage in patients with advanced head and neck malignancy. J Neurointerv Surg 2012;4:426–34. 50. Goodman DN, Hoh BL, Rabinov JD, et al. CT angiography before embolization for hemorrhage in head and neck cancer. AJNR Am J Neuroradiol 2003;24(1):140–2. 51. Miller T, Burns J, Farinhas J, et al. Covered stents safely utilized to prevent catastrophic hemorrhage in patients with advanced head and neck malignancy. J Neurointerv Surg 2012;4:426–34. 52. Fuwa N, Kodaira T, Furutani K, et al. Intra-arterial chemoradiotherapy for locally advanced oral cavity cancer: analysis of therapeutic results in 134 cases. Br J Cancer 2008;98(6):1039–45. 53. Robbins KT, Storniolo AM, Kerber C, et al. Phase I study of highly selective supradose cisplatin infusions for advanced head and neck cancer. J Clin Oncol 1994;12(10):2113–20. 54. Rosenthal DI, Yom SS, Liu L, et al. A Phase I study of SPI-077 (Stealth® liposomal cisplatin) concurrent with radiation therapy for locally advanced head and neck cancer. Invest New Drugs 2002;20(3): 343–9. 55. Balm AJM, Rasch CRN, Schornagel JH, et al. High-dose superselective intra-arterial cisplatin and concomitant radiation (RADPLAT) for advanced head and neck cancer. Head Neck 2004;26(6):485–93. 56. Zuur CL, Simis YJ, Lansdaal PE, et al. Ototoxicity in a randomized phase iii trial of intra-arterial compared with intravenous cisplatin chemoradiation in patients with locally advanced head and neck cancer. J Clin Oncol 2007;25(24):3759–65. 57. Rohde S, Kovács AF, Turowski B, et al. Intra-arterial high-dose chemotherapy with cisplatin as part of a palliative treatment concept in oral cancer. AJNR Am J Neuroradiol 2005;26(7):1804–9. 58. Kovács AF. Response to intraarterial induction chemotherapy: a prognostic parameter in oral and oropharyngeal cancer. Head Neck 2006;28(8):678–88. 59. Fuwa N, Kodaira T, Furutani K, et al. A new method of selective intraarterial infusion therapy via the superficial temporal artery for head and neck cancer. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 2008;105(6):783–9. 60. Mitsudo K, Shigetomi T, Fujimoto Y, et al. Organ preservation with daily concurrent chemoradiotherapy using superselective intra-arterial infusion via a superficial temporal artery for T3 and T4 head and neck cancer. Int J Radiat Oncol Biol Phys 2011;79(5):1428–35. 61. Shojaku H, Takakura H, Tachino H. Response to intra-arterial cisplatin and concurrent radiotherapy in a patient with primary mucosal malignant melanoma of the nasal cavity. Head Neck 2011 Dec 16. doi: 10.1002/hed.21976. [Epub ahead of print] 62. Kobayashi W, Teh BG, Sakaki H, et al. Superselective intra-arterial chemoradiotherapy with docetaxel-nedaplatin for advanced oral cancer. Oral Oncol 2010;46(12):860–3. 63. Klurfan P, Lee SK. Vascular Embolotherapy: A Comprehensive Approach. In: Golzarian M, Sharafuddin S, Sun J, editors. New York: Springer; 2006:235–46.

64. Chang FC, Lirng JF, Luo CB, et al. Carotid blowout syndrome in patients with head-and-neck cancers: reconstructive management by self-expandable stent-grafts. AJNR Am J Neuroradiol 2007;28(1): 181–8. 65. Warren FM, Cohen JI, Nesbit GM, et al. Management of carotid “blowout” with endovascular stent grafts. Laryngoscope 2002;112(3): 428–33. 66. Gandhi D, Gemmete JJ, Ansari SA, et al. Interventional neuroradiology of the head and neck. Am J Neuroradiol 2008;29(10):1806–15. 67. Wiegand S, Kureck I, Chapot R, et al. Early side effects after embolization of a carotid body tumor using Onyx. J Vasc Surg 2010;52(3): 742–5. 68. Liebman KM, Severson 3rd MA. Techniques and Devices in Neuroendovascular Procedures. Neurosurg Clin N Am 2009;20(3):315–40. 69. Debrun G, Lacour P, Caron JP, et al. Detachable balloon and calibratedleak balloon techniques in the treatment of cerebral vascular lesions. J Neurosurg 1978;49(5):635–49. 70. Kakizawa H, Toyota N, Naito A, et al. Endovascular therapy for management of oral hemorrhage in malignant head and neck tumors. Cardiovasc Intervent Radiol 2005;28(6):722–9. 71. Yoshizaki T, Wakisaka N, Murono S, et al. Intra-arterial chemotherapy less intensive than RADPLAT with concurrent radiotherapy for resectable advanced head and neck squamous cell carcinoma: a prospective study. Ann Otol Rhinol Laryngol 2007;116(10):754–61. 72. Chaloupka JC, Putman CM, Citardi MJ, et al. Endovascular therapy for the carotid blowout syndrome in head and neck surgical patients: diagnostic and managerial considerations. AJNR Am J Neuroradiol 1996;17(5):843–52. 73. Casasco A, Herbreteau D, Houdart E, et al. Devascularization of craniofacial tumors by percutaneous tumor puncture. AJNR Am J Neuroradiol 1994;15(7):1233–9. 74. Gemmete JJ, Ansari SA, McHugh J, et al. Embolization of vascular tumors of the head and neck. Neuroimaging Clin N Am 2009;19(2):181– 98, Table of Contents. 75. Casasco A, Houdart E, Biondi A, et al. Major complications of percutaneous embolization of skull-base tumors. AJNR Am J Neuroradiol 1999;20(1):179–81. 76. Nussbaum ES, Levine SC, Hamlar D, et al. Carotid stenting and “extarterectomy” in the management of head and neck cancer involving the internal carotid artery: technical case report. Neurosurgery 2000; 47(4):981–4. 77. Wright JG, Nicholson R, Schuller DE, et al. Resection of the internal carotid artery and replacement with greater saphenous vein: a safe procedure for en bloc cancer resections with carotid involvement. J Vasc Surg 1996;23(5):775–80; discussion 781–2. 78. Robbins KT, Kumar P, Wong FS, et al. Targeted chemoradiation for advanced head and neck cancer: analysis of 213 patients. Head Neck 2000;22(7):687–93. 79. Rasch CRN, Hauptmann M, Balm AJM. Intra-arterial chemotherapy for head and neck cancer: is there a verdict? Cancer 2011;117(4):874; author reply; 874–5. 80. Ackerstaff AH, Rasch CRN, Balm AJM, et al. Five-year quality of life results of the randomized clinical phase III (RADPLAT) trial, comparing concomitant intra-arterial versus intravenous chemoradiotherapy in locally advanced head and neck cancer. Head Neck 2012;34: 974–80. 81. Ackerstaff AH, Balm AJM, Rasch CRN, et al. First-year quality of life assessment of an intra-arterial (RADPLAT) versus intravenous chemoradiation phase III trial. Head Neck 2009;31(1):77–84. 82. Kovacs AF. Response to intraarterial induction chemotherapy: a prognostic parameter in oral and oropharyngeal cancer. Head and Neck 2006;28:678–88.

CHAPTER 94 MANAGEMENT OF HEAD AND NECK TUMORS719.e3 83. Chang FC, Lirng JF, Luo CB, et al. Carotid blowout syndrome in patients with head-and-neck cancers: reconstructive management by self-expandable stent-grafts. AJNR Am J Neuroradiol 2007;28(1): 181–8. 84. Gemmete JJ. Complications associated with selective high-dose intraarterial cisplatin and concomitant radiation therapy for advanced head and neck cancer. J Vasc Interv Radiol 2003;14(6):743–8. 85. Foote RL, Kasperbauer JL, Okuno SH, et al. A pilot study of high-dose intraarterial cisplatin chemotherapy with concomitant accelerated radiotherapy for patients with previously untreated T4 and selected patients with T3N0-N3M0 squamous cell carcinoma of the upper aerodigestive tract. Cancer 2005;103(3):559–68.

86. Newman LA, Robbins KT, Logemann JA, et al. Swallowing and speech ability after treatment for head and neck cancer with targeted intraarterial versus intravenous chemoradiation. Head Neck 2002; 24(1):68–77. 87. Ackerstaff AH, Tan IB, Rasch CRN, et al. Quality-of-life assessment after supradose selective intra-arterial cisplatin and concomitant radiation (RADPLAT) for inoperable stage IV head and neck squamous cell carcinoma. Arch Otolaryngol Head Neck Surg 2002;128(10): 1185–90. 88. Taki S, Homma A, Suzuki F, et al. Combined modality therapy for laryngeal cancer with superselective intra-arterial cisplatin infusion and concomitant radiotherapy. Int J Clin Oncol 2011.

CHAPTER

95

Endovascular Management of Epistaxis

Rush H. Chewning, Gerald M. Wyse, Philippe Gailloud, and Kieran P.J. Murphy CLINICAL RELEVANCE Nosebleeds so severe they require medical attention are rare. Those severe enough to require transarterial embolization are even rarer. Epistaxis of this type usually occurs in hypertensive elderly individuals with vasculopathy. Serious complications, particularly stroke, can develop quickly during endovascular procedures. It is therefore critical that the physician performing the procedure have both meticulous technique and an understanding of the potentially dangerous anastomoses around the skull base that may allow embolic material to pass from the extracranial circulation into the intracranial circulation. Fortunately, stroke can easily be avoided through knowledge of the vasculature of the head and neck, a well-developed system of preparation for head and neck embolization procedures, and an understanding of microcatheter technique.

TREATMENT OPTIONS Local conservative treatments are first line in managing epistaxis. A common treatment is insertion of pledgets soaked in an anesthetic vasoconstrictor solution, along with pressure on both sides of the nose.1 If this or another similar local treatment fails to stop the bleeding, the nose is typically packed, from posterior to anterior, with gauze or a preformed nasal tampon (Merocel or Doyle sponge). To pack the nose, a catheter is inserted through the nostril and nasopharynx and out the mouth. A gauze pack is then attached to the end of the catheter and positioned in the posterior nasopharynx to seal the region and compress the site of bleeding.1 Complications of nasal packing procedures include septal hematomas and abscesses from traumatic packing, sinusitis, and pressure necrosis secondary to excessively tight packing. Though rare, major complications such as septicemia, arrhythmia, hypoxia, and death have also been associated with packing.2,3 Use of an epistaxis balloon is an alternative method of tamponade. For anterior bleeding, balloons are inserted along the floor of the nasal cavity and inflated slowly with sterile water or air until the bleeding stops. For posterior epistaxis, a double-balloon device is passed into the nasopharynx and seated in the posterior nasal cavity to tamponade the bleeding source. Next, the anterior balloon is inflated in the anterior nasal cavity to prevent retrograde travel of the posterior balloon and subsequent airway obstruction. If a specialized epistaxis balloon is unavailable, a Foley catheter may be substituted using the same general principles. Balloon tamponade is associated with a risk of saline aspiration,4 pressure necrosis, perforation, infection, and other potential complications of packing. Persistent bleeding requires endoscopic inspection by an ear, nose, and throat (ENT) specialist to identify the 720

bleeding site. Cautery is first-line management if an anterior source of bleeding has been visualized.5 However, cautery will not take effect through fluid, so the bleeding point itself cannot be cauterized until hemostasis is achieved. Cautery may cause rhinorrhea and crusting of the mucosa; overzealous use of the technique could lead to ulceration and perforation.6 For intractable epistaxis, surgical ligation or endoscopic cauterization of the internal maxillary, external carotid, or sphenopalatine artery may be considered. Potential complications of such procedures include excessive blood loss, periorbital infection, and paresthesia. Rare major complications include oculomotor paralysis, oroantral fistulas, and septal perforation.7 Angiographic embolization in the case of intractable epistaxis has become an increasingly common management approach. The first successful percutaneous embolization for epistaxis was reported by Sokoloff et al. in 1974.8 The technique was applied to patients with hereditary hemorrhagic telangiectasia (HHT) by Strother and Newton in 1976, and in 1980, Merland et al. were able to report a 97% success rate in a first series of 54 patients with severe epistaxis.9,10 The technique has since been widely accepted and its efficacy confirmed.11 Embolization may be considered a safe and effective therapy for epistaxis.12 In addition, it is particularly helpful in cases where ENT surgery has failed to stop bleeding, usually after an endoscopic approach. Although endovascular treatment of epistaxis is safe, serious complications are still possible. Tseng et al. reported 2 cases of stroke in 114 patients treated,11 Elden et al. noted 1 stroke in 108 patients,12 and Remonda et al. reported 2 strokes in 47 patients treated endovascularly for severe epistaxis.13 With the use of calibrated particles, complications such as soft-tissue necrosis and facial paralysis have also been noted.14 Complications related to coils and Gelfoam include temporofacial pain and headache.15 The nasal mucosa receives its blood supply principally from the maxillary artery via the sphenopalatine artery and its branches. Patients with intractable epistaxis commonly show diffuse bilateral mucosal anomalies on catheter angiography (Fig. 95-1). It is therefore not surprising that endovascular management of epistaxis has classically involved bilateral sphenopalatine and sometimes bilateral facial artery angiography, followed by targeted embolization of offending vessel branches with various embolic materials such as calibrated microparticles and Gelfoam pledgets.15,16 Although the collateral network connecting the sphenopalatine artery and its branches to the facial artery and the anterior and posterior ethmoidal arteries is well recognized,17 the possibility of mucosal blood supply from branches of the infraorbital artery has not been previously emphasized. In rare cases, the infraorbital artery should be considered a possible source of bleeding and a potential candidate for superselective embolization in patients with intractable epistaxis18 (Fig. 95-2).

CHAPTER 95 ENDOVASCULAR MANAGEMENT OF EPISTAXIS 721

FIGURE 95-1. Digital subtraction angiography. Lateral projection of left common carotid artery demonstrates diffuse hypervascularity of mucosa.

FIGURE 95-2. Digital subtraction angiography. Lateral projection of right external carotid artery demonstrates dysplastic-looking arterial network (arrow) in location corresponding to source of bleeding, derived from right infraorbital artery (arrowheads).

In addition to diffuse mucosal changes, a focal arterial anomaly can often be detected (Fig. 95-3). The combination of a bleeding source suspected clinically and a focal anomaly seen on angiography allows precise targeting of embolization to a single vessel, despite the presence of widespread mucosal hypervascularity. This strategy, which relies on close collaboration between the otorhinolaryngologist and the interventional neuroradiologist, minimizes the potential for periprocedural complications and recurrent hemorrhage. The use of n-butyl cyanoacrylate (NBCA) glue as an agent in the treatment of epistaxis has been documented. Merland et al. used Gelfoam occasionally supplemented by NBCA glue at the end of the procedure.10 Luo et al.

FIGURE 95-3. Digital subtraction angiography. Lateral projection of right common carotid artery demonstrates diffuse hypervascularity of mucosa and dysplastic arteriolar network in right anterior nasal cavity fed by right infraorbital artery (arrow).

FIGURE 95-4. Digital subtraction angiography. Lateral projection of right infraorbital artery in late stage of n-butyl cyanoacrylate injection shows distribution of glue reaching suspected source of bleeding (arrow) and even refluxing into a sphenopalatine artery (arrowheads). Note presence of surgical clips, consistent with history of previous maxillary and ethmoidal ligation.

reported using NBCA glue to treat pseudoaneurysms associated with hypervascular tumors of the head and neck.19 NBCA glue as the primary embolic agent for treatment of severe epistaxis was first documented by Gailloud et al.18 (Figs. 95-4 and 95-5). A targeted embolization approach to epistaxis with NBCA glue reduces the duration of the procedure in two ways. First, a selective technique in which only one abnormal vessel is embolized avoids the multiple catheter manipulations necessary for a classic approach involving bilateral

722 PART 1 VASCULAR INTERVENTIONS · SECTION 11 NEUROVASCULAR AND HEAD AND NECK INTERVENTION

FIGURE 95-5. Digital subtraction angiography. Lateral projection of right external carotid artery shows no opacification of dysplastic network after embolization.

maxillary and facial artery embolization. Second, the use of NBCA glue as the embolic agent is technically faster than embolization with polyvinyl alcohol (PVA) microparticles or Gelfoam. In a population of elderly patients generally manifesting associated cervicocranial atheromatous changes, reducing the complexity and duration of an intervention should help reduce the risk for potential periprocedural complications. A few technical tricks can be used to optimize embolization of the targeted vessel. Unpacking the nasal cavities in the angiography suite before embolization restores flow in the catheterized vessel and thereby improves both visualization of small focal arterial abnormalities and distal distribution of the NBCA glue. Wedging the microcatheter tip into the targeted vessel and flooding the arterial bed with 5% dextrose before NBCA injection also improves distal progression of the NBCA glue. When a wedged microcatheter tip position cannot be obtained, distal progression of the glue can be optimized by performing a continuous 5% dextrose injection through the guiding catheter simultaneously with microcatheter injection of NBCA.

INDICATIONS An estimated 60% of the population experiences epistaxis at least once in their lifetime, and about 6% of these patients require treatment.20 Although the cause of epistaxis is not determined in most cases, possibilities include postsurgical complications, facial trauma, neoplasms, coagulopathy, vascular malformations, arterial hypertension, and HHT.16 Most cases of epistaxis are managed medically, but such methods are not always effective, especially in patients with a posterior source of bleeding.15 When conservative management fails, the epistaxis is termed intractable. Patients with intractable epistaxis almost always show diffuse bilateral mucosal anomalies on catheter angiography, with no clearly identifiable source of bleeding. This appearance is often due to nasal packing. Endovascular management of epistaxis has therefore classically involved

bilateral distal internal maxillary artery embolization with calibrated microparticles or Gelfoam pledgets. However, a focal arterial anomaly may be detectable on angiography in addition to diffuse mucosal hypervascularity. A focal angiographic anomaly, when identified, may be selectively embolized with NBCA glue. Angiographic identification of the bleeding source followed by superselective single-vessel embolization with NBCA glue is a successful therapeutic approach to treating intractable epistaxis that reduces the number of endovascular manipulations and duration of the procedure. Between 1% and 6% of adults experience a large posterior nosebleed. Epistaxis can be due to numerous causes, but usual causes include dry air, nasal oxygen cannulas, hypertension, low platelet count, aspirin use, warfarin (Coumadin) use, and atherosclerosis. Unusual causes of epistaxis include a giant internal carotid artery aneurysm eroding into the sphenoid bone, aneurysmal bone cyst, pseudoaneurysm, HHT, trauma, juvenile angiofibroma, ethmoidal arteriovenous fistula, and cancer of the nose and throat.

CONTRAINDICATIONS It is critical that a physician performing transarterial embolization have an understanding of the potential dangerous anastomoses around the skull base. These anastomoses may allow embolic material to pass from the extracranial circulation into the intracranial circulation during the embolization and result in a stroke. Relevant anastomoses involve the inferolateral trunk, ethmoidal collaterals, meningohypophyseal arteries, and occipital-vertebral arteries. In addition to being aware of these potentially dangerous anastomoses, use of larger embolic particles can minimize stroke risk. For this reason, it is advisable to never use embolic material smaller than 200 μm during routine embolization for epistaxis.

EQUIPMENT When selecting a microcatheter, it is important to keep in mind the goal of atraumatic access. The “smaller, softer, and safer principle” applies to all procedures and equipment. Many microcatheters are available. They vary in stiffness, tractability, and inner and outer luminal diameter. Ideally, a catheter that will push and track yet remain stable should be used. Catheter braiding increases axial rigidity and improves stability. Many catheters have a proximal braided shaft with a softer, more flexible distal end. An appropriately sized microcatheter is selected for each procedure by considering catheter flexibility and stability, the guiding catheter, and devices to be inserted into the microcatheter. The journey traveled by the microcatheter, including the length, morphology, and tortuosity of the anatomy, must also be considered. The technique of steaming a microcatheter is very useful because a shape that conforms to a vessel will give the catheter stability during the procedure. Microcatheters are also available in preshaped form, making steaming unnecessary. A gentle 30- or 45-degree curve is usually sufficient for the procedure. Similarly, a 30- or 45-degree shape on a wire is all that is needed. Before beginning a procedure, ensure that all equipment sizes are compatible with one another and that everything

CHAPTER 95 ENDOVASCULAR MANAGEMENT OF EPISTAXIS 723 fits (sizes of wires, coil, etc.). Perform a dry table test if necessary, especially if the equipment is unfamiliar. Choose optimal obliquity. Always remove the slack from the system. Begin slowly and proceed gently. If something is not advancing, understand why. Do not simply push harder!

EMBOLIZATION Multiple embolic agents are available for use in cases of epistaxis, including PVA, Gelfoam, trisacryl gelatin microspheres, and NBCA glue.

Polyvinyl Alcohol Because it is both biocompatible and efficient as a permanent embolic agent, PVA has traditionally been the goldstandard embolic particle. However, since it is hydrophobic and irregular in shape, particles tend to cluster and create aggregates of unpredictable size.

Gelfoam Particles of Gelfoam, made from purified gelatin, may be combined with a mixture of saline and contrast material and then injected into a vessel for embolization. The spongelike material draws in blood via capillary action, and clotting ensues.

Trisacryl Gelatin Microspheres (Embospheres) These small spheres are composed of a plastic called trisacryl gelatin. They are hydrophilic, biocompatible, nonresorbable, and uniformly spherical. These calibrated microspheres are easy to deliver through a microcatheter, and they quickly and reliably reduce blood flow. Their size correlates well with the size of the occluded vessel. They have better sizing and penetration characteristics than PVA, but deaths have been reported from progressive irreversible hypoxemia—especially with smaller microspheres (40-120 μm).

Trufill Liquid Embolic System This liquid glue system is a combination of NBCA, ethiodized oil, and tantalum powder. It is inserted under fluoroscopic guidance via superselective catheter delivery to obstruct or reduce blood flow. The mixture polymerizes into a solid material on contact with blood or tissue. Higher concentrations of ethiodized oil increase polymerization time, allowing better distal penetration. High concentrations of NBCA result in a faster polymerization rate, which allows proximal embolization. Tantalum powder is added to increase radiopacity and lower viscosity.

TECHNIQUE Anatomy and Approach The key to a successful transarterial embolization procedure is proper technique. It is important to keep in mind that the purpose of the procedure is not to devascularize the nose but to allow hemostasis by decreasing arterial inflow pressure to facilitate clotting and epithelial repair.

The nose should be packed during the procedure and unpacked at the end of it by an ENT specialist. Of note, procedures should be performed under general anesthesia to protect the patient’s airway. Size 6F guiding catheters should be placed in the external carotid artery, then microcatheters and microwires should be used to access the internal maxillary artery distal to the infraorbital branch. It may be necessary to embolize this vessel bilaterally because it is common for ENT physicians to be unable to identify the source of the bleeding. Occasionally the nasal branches of the facial arteries have to be embolized, particularly in patients with very anterior bleeding, because these branches may supply the anterior aspect of the nose. In patients who have previously been treated by coil embolization, branches may have collateralized more posteriorly and should thus be investigated. Again, it is imperative that potentially dangerous anastomosis sites be identified by high-quality digital subtraction angiography and careful review. It is also essential to place all embolic materials on a separate table to prevent accidental use of a syringe containing embolic particles for a diagnostic injection or transfer of embolic particles onto a diagnostic catheter or wire, which may cause stroke. Microcatheter injection should be performed with 3-mL syringes. Syringes should be clearly labeled in standard fashion as either saline, contrast material, or particles. Avoid using particles smaller than 200 μm. The best results are achieved with particles in the range of 250 to 350 μm. Although it may seem counterintuitive in cases of bleeding, administration of heparin (5000-unit bolus, with 1000 units/h thereafter) during the embolization is critical. Prevention of stroke is still the primary concern. Meticulous attention to pressurized catheter flush bags filled with 4 units heparin/mL of normal saline (4000 units of heparin in 1000 mL normal saline) is critical. This ensures that the dead space around the microcatheter is flushed and clot does not accumulate there. Heparin activity should be reversed with protamine at the end of the procedure. It is crucial to ensure that all guiding catheters are clean and properly flushed. Just one piece of PVA is sufficient to cause a stroke during an injection. Remove all catheters and replace them with new catheters when in doubt; the expense is far less costly than stroke rehabilitation. A cautious approach to embolization is advised because it is never possible to reverse the effects of embolization, especially after devascularization and skin necrosis. Again, staying with midrange particle sizes will help avoid such complications.

Technical Aspects Though useful, biplane angiography is not required for embolization, but roadmap capability is essential. Two skilled physician operators are needed rather than a scrub nurse and one physician. Size 5F or 6F Cordis Envoy MPC (or comparable) guide catheters should be used. These catheters have 0.057- and 0.067-inch lumens, allowing use of microcatheters and roadmap injections. Use of 0.038-inchlumen 5F catheters as guiding devices can cause poor vessel visualization and lead to multiple problems. Guiding injections are not possible with this size, and flush will not flow in the dead space, thereby increasing the risk of clotting. Elderly patients with left-sided nosebleeds often need a 6F

724 PART 1 VASCULAR INTERVENTIONS · SECTION 11 NEUROVASCULAR AND HEAD AND NECK INTERVENTION Simmons-2 guide catheter or even a shuttle sheath because of a hypertensive arch and challenging left common carotid access. A few such catheters should be kept on hand for this indication.

COMPLICATIONS Complications of endovascular therapy for epistaxis, though rare, do occur. They include temporofacial pain, headache, soft-tissue necrosis, facial paralysis, and stroke.

POSTPROCEDURAL AND FOLLOW-UP CARE After the procedure, the nose should be unpacked by an ENT specialist, and heparin activity should be reversed to lower the risk for continued bleeding.

anomaly, particularly with respect to lateralization. In certain cases, postembolization endoscopic evaluation may even demonstrate embolic material in the treated vessel. When possible, this therapeutic strategy may avoid long and possibly unwarranted endovascular manipulations that put the patient at higher risk for periprocedural complications. The use of NBCA glue, though not classic in this indication, appears to be ideally suited for rapid targeting of a focal source of arterial bleeding. The tips and techniques in this chapter are routinely used in a variety of applications, including gunshot wounds to the head and neck, exsanguinating tumors, and preoperative embolization of glomus tumors. The target vessel is the primary difference between the applications. KEY POINTS •

Intractable epistaxis requiring transarterial embolization is rare.

•

Endovascular embolization is a minimally invasive, effective therapy for epistaxis.

•

Proper technique and knowledge of anatomy are essential to minimize risks.

SUMMARY Transarterial embolization of epistaxis is a safe, costeffective procedure when performed with adequate technique. During the procedure, an understanding of potential dangerous anastomoses is crucial to avoiding passage of embolic material from the extracranial circulation into the intracranial circulation, which could result in a stroke. Rapid targeted epistaxis embolization can be performed when clinical suspicion of a bleeding source correlates with a focal arterial anomaly on diagnostic angiography, despite the presence of diffuse bilateral mucosal hypervascularity. Transarterial embolization often follows an unsuccessful attempt at controlling the epistaxis via an endoscopic approach. Communication between the ENT specialist and the interventional neuroradiologist is essential for correlation of the bleeding source with an angiographic

▶ SUGGESTED READINGS Elden L, Montanera W, Terbrugge K, et al. Angiographic embolization for the treatment of epistaxis: A review of 108 cases. Otolaryngol Head Neck Surg 1994;111:44–50. Tseng E, Narducci C, Willing S, Sillers M. Angiographic embolization for epistaxis: a review of 114 cases. Laryngoscope 1998;108:615–9.

The complete reference list is available online at www.expertconsult.com.

CHAPTER 95 ENDOVASCULAR MANAGEMENT OF EPISTAXIS724.e1 REFERENCES 1. Kucik C, Clenny T. Management of epistaxis. Am Fam Physician 2005;71:305–11. 2. Papel ID, Scott JC, Fairbanks DNF. Complications of nasal surgery and epistaxis management. In: Eisele DW, editor. Complications in head and neck surgery. St Louis: Mosby; 1993. p. 447–56. 3. Loftus BC, Blitzer A, Cozine K. Epistaxis, medical history, and the nasopulmonary reflex: what is clinically relevant? Otolaryngol Head Neck Surg 1994;110:363–8. 4. McFerran DJ, Edmonds SE. The use of balloon catheters in the treatment of epistaxis. J Laryngol Otol 1993;107:197–200. 5. Manthey DE, Harrison BP. Otolaryngologic procedures. In: Roberts JR, Hedges JR, editors. Clinical Procedures in emergency medicine. Philadelphia: Saunders; 1998. 6. Toner JG, Walby AP. Comparison of electro- and chemical cautery in the treatment of anterior epistaxis. J Laryngol Otol 1990;104:617–8. 7. Schaitkin B, Strauss M, Houck JR. Epistaxis: medical versus surgical therapy: a comparison of efficacy, complications, and economic considerations. Laryngoscope 1987;97:1392–6. 8. Sokoloff J, Waskom T, McDonald D, et al. Therapeutic percutaneous embolization in intractable epistaxis. Radiology 1974;111:285–7. 9. Strother CM, Newton TH. Percutaneous embolization to control epistaxis in Rendu-Osler-Weber disease. Arch Otolaryngol 1976;102: 58–60. 10. Merland JJ, Melki JP, Chiras J, et al. Place of embolization in the treatment of severe epistaxis. Laryngoscope 1980;90:694–704.

11. Tseng E, Narducci C, Willing S, Sillers M. Angiographic embolization for epistaxis: a review of 114 cases. Laryngoscope 1998;108:615–9. 12. Elden L, Montanera W, Terbrugge K, et al. Angiographic embolization for the treatment of epistaxis: a review of 108 cases. Otolaryngol Head Neck Surg 1994;111:44–50. 13. Remonda L, Schroth G, Caversaccio M, et al. Endovascular treatment of acute and subacute hemorrhage in the head and neck. Arch Otolaryngol Head Neck Surg 2000;126:1255–61. 14. Wehrli M, Liieverherr U, Valavanis A. Superselective embolization for intractable epistaxis: experience with 19 patients. Clin Otolaryngol 1988;13:415–20. 15. Oguni T, Korogi Y, Yasunaga T, et al. Superselective embolization for intractable idiopathic epistaxis. Br J Radiol 2000;73:1148–53. 16. Vitek J. Idiopathic intractable epistaxis: endovascular therapy. Radiology 1991;181:113–6. 17. Babin E, Moreau S, Goullet de Ruby M, et al. Anatomic variations of the arteries of the nasal fossa. Otolaryngol Head Neck Surg 2003; 128:236–9. 18. Mahavedia A, Murphy K, Obray R, Gailloud P. Embolization for intractable epistaxis. Tech Vasc Interv Radiol 2005;8:134–8. 19. Luo C, Teng M, Lirng J, et al. Endovascular embolization of intractable epistaxis. Zhonghua Yi Xue Za Zhi 2000;63:205–12. 20. Small M, Murray J, Moran AG. A study of patient with epistaxis requiring admission to hospital. Health Bull 1982;40:24–9.

CHAPTER

96

Subarachnoid Hemorrhage

Juan Carlos Baez, Rush H. Chewning, Gerald Wyse, and Kieran P.J. Murphy

When a patient has a severe or “thunderclap” headache of sudden onset, subarachnoid hemorrhage (SAH) must be considered and investigated. The most common symptom described by patients with SAH is headache, often the worst they have ever experienced and associated with a sense of doom. Less specific symptoms include stiff neck, change in level of consciousness, vomiting, and focal neurologic deficits, including cranial nerve palsies. Cranial nerve palsies can arise as a result of bleeding, compression from an aneurysm, or ischemia secondary to acute vasoconstriction immediately after aneurysmal rupture. The stereotypic third nerve palsy is manifested as an inferiorly rotated abducted eye with accompanying ptosis and pupil dilation. Patients often describe visual deterioration in the form of diplopia. Onset is sudden, and pain is a common complaint. Third nerve palsy often serves as an indicator of a posterior communicating artery aneurysm, although superior cerebellar and basilar artery aneurysms can less commonly cause the same symptoms.1 Ruptured intracranial aneurysms account for approximately 85% of cases of nontraumatic SAH. Although some have suggested familial or genetic causes for the development of intracranial aneurysms, other factors such as smoking, heavy alcohol use, and hypertension play a larger role. Female sex, African American heritage, and hereditary disorders such as autosomal polycystic kidney disease constitute some of the genetic factors also linked to development of these vascular lesions.2 Other causes of SAH include cerebral arteriovenous malformation (AVM) and perimesencephalic hemorrhage. AVMs are complex bundles of arteriovenous connections without intervening capillaries. The abnormally high flow commonly results in secondary aneurysms. AVMs are less common than aneurysms as a cause of SAH and will not be discussed further. Perimesencephalic hemorrhage is manifested as blood in the cerebral cisterns, with a predominantly prepontine distribution, and accounts for 10% of all SAH cases. This benign condition is thought to be secondary to rupture of a prepontine vein and poses no threat to the patient.3 Multiple grading systems are used for subarachnoid hemorrhage. Two of the more commonly used include the Fisher scale and the Hunt and Hess classification. The Fisher scale scores SAH on computed tomography (CT) appearance and quantification of subarachnoid blood. Patients are in group 1 if no blood is detected. Group 2 consists of those with diffuse deposition of subarachnoid blood, no clots, and no layers of blood greater than 1 mm. Group 3 patients have localized clots or vertical layers of blood 1 mm or greater in thickness, whereas patients in group 4 have intracerebral or intraventricular clots or diffuse subarachnoid blood.4 The Fisher scale is a radiologic classification and should not be used to predict clinical outcomes.

The Hunt and Hess classification scale grades SAH on a scale from I to V based on clinical manifestations of the patient. A grade of I is given to patients with asymptomatic findings or mild headache. Grade II is assigned for cranial nerve palsies, moderate to severe headache, and nuchal rigidity. Grade III includes mild focal neurologic deficits, lethargy, or confusion. Grade IV consists of stupor, moderate to severe hemiparesis, and early decerebrate rigidity. Grade V is deep coma, decerebrate rigidity, and a moribund appearance.4 If the patient has a serious systemic disease such as hypertension, diabetes mellitus, or chronic obstructive pulmonary disease or if severe vasospasm occurs during arteriography, the Hunt and Hess grade is increased by one level.5 The Hunt and Hess system fails to adequately describe the patient’s stage of consciousness on admission, which ultimately proves to be the most reliable indicator of recovery. Consequently, interphysician variability exists when using this classification system. In addition to the Hunt and Hess system, use of the Glasgow Coma Scale, which attempts to objectively grade consciousness according to eye opening, verbal ability, and motor function, helps stratify patients and minimize variability among physicians.

INDICATIONS Outcomes in cases of SAH are poor, with mortality rates ranging from 32% to 67%.6 Early diagnosis is crucial in minimizing the consequences associated with this condition, because up to 30% of fatalities in individuals with previous SAH result from rebleeding. Among survivors of SAH, 30% will have moderate to severe disability after the event. Unfortunately, between a fourth and half of cases of SAH are missed by the first evaluating physician. Initial evaluation of SAH typically involves the use of noncontrastenhanced CT in the emergency department. CT scans are fast, readily available, and relatively inexpensive. CT has been demonstrated to have 93% to 95% sensitivity for SAH when conducted within 24 hours of initial bleed.7 A timely study is critical because 50% of SAH episodes are not visible on CT 1 week after the bleeding. When clinical suspicion for SAH is high, negative CT findings cannot rule it out. The literature reports a 7% falsenegative rate of head CT for the diagnosis of SAH in patients later found to have ruptured aneurysms. Because mortality with untreated SAH is exceedingly high, patients with negative CT findings but clinical suspicion for SAH should undergo additional diagnostic testing. A lumbar puncture (LP) is recommended as the next step in the diagnosis of SAH. Recent work published in the BMJ supports modern multislice head CT alone as adequate to exclude SAH if it is performed within 6 hours of ictus.8 This is being adopted 725

726 PART 1 VASCULAR INTERVENTIONS · SECTION 11 NEUROVASCULAR AND HEAD AND NECK INTERVENTION in some emergency departments but not all, and will no doubt be controversial because of cases where LP found CT-negative SAH. The typical LP involves sequentially drawing four aliquots of cerebrospinal fluid (CSF) from the L4-L5 interspace under local anesthesia of the patient’s back. Inspection of CSF for xanthochromia (yellowish tint of CSF, indicating hemolyzed blood) provides the best evidence of SAH. A perfectly performed LP in a patient with no SAH yields clear CSF with no red blood cells (RBCs) after violating the L2-L3 interspace. Common parlance refers to this as a “champagne tap,” and this result combined with the absence of xanthochromia reliably rules out SAH.9 Approximately 10% to 25% of the time, however, LP yields varying amounts of blood. The physician must then decide whether the blood represents SAH or perforation of the epidural venous plexuses and cauda equina vessels during the procedure. The “three-tube test” (named before four tubes became standard) is based on the principle that if the amount of blood present in the aliquots decreases between the first and fourth tubes, the blood is attributed to a traumatic tap. No consensus exists regarding what size decrement of blood indicates a traumatic LP. This results in uncertainty for physicians who must base therapeutic decisions on these results. Although multiple criteria have been used to differentiate between traumatic taps and true SAH, these criteria have been devised on the basis of traditional medical practice and not evidence-based medicine. A study by Heasley et al., however, showed that a 25% reduction in RBCs between the first and fourth tubes—as one would expect to see in a traumatic tap—does not rule out SAH.9 Other authors have used different cutoffs for what constitutes a traumatic tap, but these cutoffs have also yielded unacceptably high false-positive and false-negative rates. Heasley’s group considers determination of traumatic taps via the three-tube test unreliable and recommend further invasive testing to determine the presence or absence of SAH with certainty if clinical suspicion remains high. Additional CSF testing for D-dimer, clot formation, and opening pressure have all been studied in an attempt to diagnose SAH with greater accuracy, but the results have been inconclusive; these tests are rarely part of the LP procedure. In cases where a patient is complaining of the worst headache ever experienced, formal evaluation of the cerebral blood vessels is necessary subsequent to any LP that yields blood.

CONTRAINDICATIONS There are no absolute contraindications to digital subtraction angiography (DSA). Relative contraindications include allergic reactions to contrast material, hypotension, severe hypertension, coagulopathy, renal insufficiency, and congestive heart failure. The patient must also be cooperative and able to remain still during the procedure.

EQUIPMENT Safe and effective DSA for detection of intracranial aneurysms is ideally performed in an angiography suite with a biplane angiography system staffed by experienced

neurointerventionalists. Three-dimensional (3D) DSA requires additional workstations and appropriate software for image reconstruction.

TECHNIQUE Cerebral DSA is considered the gold standard for finding the aneurysmal source of SAH. This invasive imaging modality relies on manipulation of catheters within the carotid and vertebral arteries to inject radiographic contrast material into the intracranial vessels. DSA provides the best spatial and temporal resolution of aneurysmal sources of SAH, and 3D-DSA further improves aneurysm detection and characterization. After consent, the patient is brought to the angiography suite and properly draped. Arterial access is gained via a femoral approach, and arterial sheaths and diagnostic catheters are placed within the femoral artery. The sheath is continuously flushed with heparinized saline throughout the procedure. The diagnostic catheter is then advanced to the aortic arch. Imaging of the aortic arch and bilateral vertebral and carotid arteries is performed. Special attention is paid to common locations for aneurysm formation. The three most common sites of aneurysm formation are the anterior communicating artery, posterior communicating artery, and middle cerebral artery. The basilar, posterior inferior cerebellar, and vertebral arteries must also be visualized to clear the angiogram. On DSA, an aneurysm appears as a saccular or irregular dilation of a blood vessel. The interventionalist can use 3D-DSA to better visualize and characterize aneurysms. Special attention should be paid to the neck, shape, size, location, and orientation of the aneurysm because these parameters will help determine therapy. Both ruptured and unruptured aneurysms should receive immediate treatment by either surgical clipping or endovascular coiling. Once an aneurysm has ruptured, there is a 15% to 20% chance of it rebleeding, with mortality in this scenario exceeding 70%. Studies have shown improvements in mortality and morbidity if treatment of the culprit aneurysm is not delayed.10 Consequently, quick and accurate diagnosis proves vital for the patient’s prognosis. The high mortality associated with untreated SAH makes diagnosis of it imperative.11 Any patient with the worst headache ever experienced and a clinical picture consistent with SAH should undergo CT scanning to evaluate for acute bleeding. A negative CT should prompt further invasive testing with LP and possibly DSA. Although improvements in technique and expertise have reduced the risks associated with DSA, a meta-analysis showed that there was a 1.8% chance of complications and less than a 0.2% probability of sustaining permanent neurologic deficits subsequent to DSA.12 Nevertheless, the benefits of diagnosing an aneurysm causing SAH with DSA far outweigh the risks.

CONTROVERSIES The increasing popularity of computed tomographic angiography (CTA) and magnetic resonance angiography (MRA) has made prominent the question of whether these noninvasive imaging modalities can replace DSA. DSA is an invasive test that adds time, cost, discomfort, and a small

CHAPTER 96 SUBARACHNOID HEMORRHAGE 727 but finite risk to the patient’s hospital experience. If noninvasive imaging could match the high sensitivity of DSA for the diagnosis of SAH and localization of an aneurysm, one could minimize the use of DSA. 3D-CTA is increasingly being used for detection and characterization of intracranial aneurysms. In cases of suspected SAH, studies can be conducted immediately after initial noncontrast-enhanced CT scan, because the patient is already on the scanning table. These studies are quick and noninvasive, although contrast agent must be administered through a peripheral intravenous catheter. The sensitivity of 3D-CTA was shown to be 93% in a metaanalysis by Chappell et al.13 Sensitivity, particularly for small aneurysms, would have to improve for 3D-CTA to fully supplant DSA as the gold standard. Nonetheless, a positive scan proves incredibly useful. Many surgeons now depend solely on 3D-CTA for preoperative evaluation and planning of surgical clipping of aneurysms. As technologic advances in both scanning equipment and software image processing improve, the sensitivity of CT for diagnosing SAH and detecting aneurysms could approach 100%, thereby obviating the need for more invasive tests. Advances in magnetic resonance imaging (MRI) may allow its use in diagnosing SAH. MRI has been demonstrated to be superior to CT in diagnosing chronic as well as subacute SAH when imaging is performed 5 days after the initial bleed.14 Although MRI has traditionally been considered less useful than CT for evaluating acute SAH, recent reports indicate that the sensitivity of MRI with fluid-attenuated inversion recovery (FLAIR), proton density–weighted images, or gradient-echo parameters might in fact be equal to if not greater than CT in diagnosing acute SAH.15,16 Concern about availability of scanners and length of time necessary for scans might limit the utility of MRI in this application. However, acute SAH protocol MRIs can be conducted in as little as 8 minutes without the need for iodinated contrast material or exposure of the patient to radiation.16 Nevertheless, a hemodynamically unstable sick patient should never undergo MRI. Similar to 3D-CTA, MRI evaluation of SAH adds the benefit of pinpointing the location of an aneurysm, when one is present, without the need for further scans. CTA and MRA currently have problems detecting small aneurysms, with some studies stating sensitivity as low as 40% for aneurysms smaller than 3 mm. Sensitivity for aneurysms larger than this, however, is significantly higher. Because most aneurysms causing third nerve palsy exceed 5 mm in diameter, CTA and MRA are well positioned to diagnose the culprit aneurysms.17 Once an aneurysm grows large enough to compress the third nerve, there is a 30% to 60% probability of it causing SAH within the following month. Consequently, presence of a third nerve palsy acts as a herald of potentially catastrophic SAH and should raise multiple red flags in the mind of the physician. Furthermore, a noninvasive imaging modality such as CTA might prove sufficient for diagnosis in this clinical scenario. Nevertheless, until the sensitivity of CTA and MRA improve for aneurysms smaller than 3 mm, DSA will continue being the gold standard in diagnosing intracerebral aneurysms. Whole-head 320-slice CT allows exact aneurysm assessment, and clearly the technology is marching towards a day where catheter angiography will be obsolete (Fig. 96-1).

FIGURE 96-1. This 320-slice computed tomographic angiogram demonstrates computed tomography digital subtraction angiographic image of anterior communicating aneurysm filling from right A1-A2 junction, pointing superiorly, with a narrow neck.

OUTCOMES As noted, DSA is the most definitive method of diagnosing an aneurysm subsequent to SAH, but 10% to 20% of DSA studies performed subsequent to proven SAH will appear normal. In these cases, if the pattern of bleeding appears to be perimesencephalic, the bleeding is attributed to a nonaneurysmal cause, and no further diagnostic evaluation is warranted. These patients go home after observation for 7 to 10 days in the hospital, with no complications and no modifications necessary in their daily lives. If the pattern of blood on CT suggests an aneurysm, but results of DSA appear normal, repeat angiography within 1 to 6 weeks is warranted because the risk of rebleeding is high. Oblique views that fail to visualize the aneurysm, vasospasm, thrombosis of the aneurysm, or saccular obliteration by hematoma have all been posited as reasons why aneurysmal SAH can appear normal on DSA.18 If an aneurysm is diagnosed, one must decide whether therapy is to be pursued. Elderly and frail patients in whom the risk of sedation and an interventional procedure exceeds the risk of not treating the aneurysm should undergo observation. Younger and otherwise healthier patients, however, require treatment because the risk of primary or secondary bleeding from an aneurysm accumulates with each passing year. Fifteen years ago, these patients would have undergone neurosurgical clipping of the aneurysm. This technique has been shown to effectively reduce morbidity and mortality when compared with untreated aneurysms.19 Since the introduction of detachable coils, however, patients can now decide whether they would prefer endovascular therapy. The morphology and location of an aneurysm, as well as patient status and preference, help determine which therapeutic modality should be used. Coiling selection criteria include anterior circulation aneurysms in patients with Hunt and Hess grade III to V SAH, posterior circulation aneurysms, medical contraindications to surgery, advanced

728 PART 1 VASCULAR INTERVENTIONS · SECTION 11 NEUROVASCULAR AND HEAD AND NECK INTERVENTION age, and multiple aneurysms. Aneurysms with broad necks (>4 mm) or a fundus-to-neck ratio of less than 2 increase the risk for coil migration and protrusion of the coil into the parent vessel, thereby making clipping clearly favorable in these situations. In terms of location, surgical clipping of anterior circulation aneurysms, particularly anterior communicating artery aneurysms, yields the best results.20 For all other aneurysms whose morphology and location do not clearly favor one treatment modality over another, controversy still exists regarding optimal therapy. Two large multicenter studies have attempted to determine whether coiling or clipping provides a better outcome in patients who are equally good candidates for both therapies. The International Subarachnoid Aneurysm Trial (ISAT), a randomized prospective study, showed that coiling yielded significantly better outcomes at 1 year as quantified by Rankin scores, a functional scoring system.21 Whereas ISAT included patients with ruptured aneurysms, the International Study of Unruptured Intracranial Aneurysms (ISUIA) demonstrated that coiling also yields better outcomes in patients whose aneurysms have not ruptured.22 Both these studies have drawn criticism from the medical community, but their results, as well as the results of countless smaller studies, suggest that coiling results in reduced morbidity and mortality and decreased use of hospital resources in the short period during which they have been studied. Recent developments in flow-diverting stents are having a remarkable impact on giant aneurysm outcomes. These devices do have side effects, however, including delayed aneurysm rupture and stroke.23

COMPLICATIONS Serious complications from DSA in the hands of experienced interventionalists are rare. Nevertheless, the patient must be informed of the risk of temporary or permanent neurologic deficits. This risk is increased in elderly patients

with known histories of extensive atherosclerosis. Additional non-neurologic risks include femoral artery injury, hematoma, allergic reaction to the contrast agent, and nephrotoxicity. Like all invasive procedures, DSA also carries a small but finite risk of death.

POSTPROCEDURAL AND FOLLOW-UP CARE Patients with aneurysms diagnosed on DSA require treatment unless therapy is contraindicated. If the aneurysm has already led to SAH, the chance of rebleeding is exceedingly high. These patients should undergo either clipping or coiling to reduce SAH morbidity and mortality. KEY POINTS •

Negative head computed tomography (CT) does not rule out subarachnoid hemorrhage (SAH).

•

After a negative head CT, a patient with the worst headache ever experienced should undergo lumbar puncture.

•

Digital subtraction angiography remains the gold standard for the diagnosis of aneurysms causing SAH.

▶ SUGGESTED READINGS Brisman JL, Song JK, Newell DW. Cerebral aneurysms. N Engl J Med 2006;355:928–39. Shah KH, Edlow JA. Distinguishing traumatic lumbar puncture from true subarachnoid hemorrhage. J Emerg Med 2002;23:67–74. van Gijn J, Kerr RS, Rinkel GJ. Subarachnoid haemorrhage. Lancet 2007;369:306–18.

The complete reference list is available online at www.expertconsult.com.

CHAPTER 96 SUBARACHNOID HEMORRHAGE728.e1 REFERENCES 1. Biousse V, Newman NJ. Third nerve palsies. Semin Neurol 2000;20: 55–74. 2. Longstreth Jr WT, Koepsell TD, Yerby MS, van Belle G. Risk factors for subarachnoid hemorrhage. Stroke 1985;16:377–85. 3. van Gijn J, Rinkel GJE. Subarachnoid haemorrhage: diagnosis, causes, and management. Brain 2001;124:249–78. 4. Fisher CM, Kistler JP, Davis JM. Relation of cerebral vasospasm to subarachnoid hemorrhage visualized by computerized tomographic scanning. Neurosurgery 1980;6:1–9. 5. Hunt WE, Hess RM. Surgical risk as related to time of intervention in the repair of intracranial aneurysms. J Neurosurg 1968;28:14–20. 6. Hop JW, Rinkel GJE, Algra A, van Gijn J. Case-fatality rates and functional outcome after subarachnoid hemorrhage: a systematic review. Stroke 1997;28:660–4. 7. van der Wee N, Rinkel GJ, Hasan D, van Gijn J. Detection of subarachnoid hemorrhage on early CT: is lumbar puncture still needed after a negative scan? J Neurol Neurosurg Psychiatry 1995;58:357–9. 8. Perry JJ, Stiell IG, et al. Sensitivity of computed tomography performed within six hours of onset of headach for diagnosis of subarachnoid haemorrhage: prospective cohort study. BMJ 2011;343(1):d4277. PMCID21768192,. 9. Heasley DC, Mohamed MA, Yousem DM. Clearing of red blood cells in lumbar puncture does not rule out ruptured aneurysm in patients with suspected subarachnoid hemorrhage but negative head CT findings. AJNR Am J Neuroradiol 2005;26:820–4. 10. Ross N, Hutchinson PJ, Seeley H, Kirkpatrick PJ. Timing of surgery for supratentorial aneurysmal subarachnoid haemorrhage: report of a prospective study. J Neurol Neurosurg Psychiatry 2002;72:480–4. 11. Schievink WI. Intracranial aneurysms. N Engl J Med 1997;336: 28–40. 12. Cloft HJ, Joseph GJ, Dion JE. Risk of cerebral angiography in patients with subarachnoid hemorrhage, cerebral aneurysm, and arteriovenous malformation: a meta-analysis. Stroke 1999;30:317–20.

13. Chappell ET, Moure FT, Good MC. Comparison of computed tomographic angiography with digital subtraction angiography in the diagnosis of cerebral aneurysms: a meta-analysis. Neurosurgery 2003; 52:624–31. 14. Weismann M, Bruckmann H. Magnetic resonance imaging of subarachnoid hemorrhage. Fortschr Rontgenstr 2004;176:500–5. 15. Matsumura K, Matsuda M, Handa J, Todo G. Magnetic resonance imaging with aneurysmal subarachnoid hemorrhage: comparison with computed tomography scan. Surg Neurol 1990;34:71–8. 16. Fiebach JB, Schellinger PD, Geletneky K, et al. MRI in acute subarachnoid haemorrhage; findings with a standardised stroke protocol. Neuroradiology 2004;46:44–8. 17. Jacobson DM, Trobe JD. The emerging role of magnetic resonance angiography in the management of patients with third cranial nerve palsy. Am J Ophthalmol 1999;128:94–6. 18. van Gijn J, Rinkel GJ. Subarachnoid haemorrhage: diagnosis, causes and management. Brain 2001;124:249–78. 19. Krisht AF, Gomez J, Partington S. Outcome of surgical clipping of unruptured aneurysms as it compares with a 10-year nonclipping survival period. Neurosurgery 2006;58:207–16; discussion 207–16. 20. Shanno GB, Armonda RA, Benitez RP, Rosenwasser RH. Assessment of acutely unsuccessful attempts at detachable coiling in intracranial aneurysms. Neurosurgery 2001;48:1066–72. 21. Molyneux A, Kerr R, Stratton I, et al. International Subarachnoid Aneurysm Trial (ISAT) of neurosurgical clipping versus endovascular coiling in 2143 patients with ruptured intracranial aneurysms: a randomised trial. Lancet 2002;360:1267–74. 22. International Study of Unruptured Intracranial Aneurysms Investigators. N Engl J Med 1998;339:1725–33. Erratum in N Engl J Med 1999;340:744. 23. D’Urso P, Karadell H, Kallmes D, et al. Coiling for paraclinoid aneurysms: time to make way for flow diverters? AJNR Am J Neuroradiol 2012;33:1470–4.

CHAPTER

97

Management of Head and Neck Injuries M.J. Bernadette Stallmeyer

Traumatic head and neck vascular injuries such as dissections, transections, pseudoaneurysms, arteriovenous fistulas, and large artery occlusions are relatively uncommon but can result in potentially devastating stroke, severe blood loss, and even exsanguination and death. Rapid, early diagnostic imaging followed by effective management of neurovascular injuries can help improve patient outcome by focusing attention on prompt treatment of these lesions.1,2 Advances in multidetector computed tomography (CT) technology (Fig. 97-1) have greatly facilitated rapid noninvasive screening of trauma patients with suspected neurovascular injury. Although the use of computed tomographic angiography (CTA) for definitive diagnosis remains somewhat controversial, it can provide sufficient information to “fast-track” a patient with significant vascular injury to definitive surgical or endovascular intervention.3 Alternatively, CTA may help determine whether a patient with minor injury is better served by delaying angiographic evaluation or treatment in favor of managing concomitant visceral or orthopedic injuries. Angiographic and endovascular resources can thus be directed where they are most appropriate and most needed.

INDICATIONS • Obvious signs of trauma, including facial fracture; penetrating wounds; epistaxis; hemorrhage from the mouth, orbit, or ears; periorbital ecchymosis, or Battle sign; expanding cervical hematoma or a pulsatile mass, cervical bruising or abrasions; or cervical bruit in a young patient1,2 • Fractures in proximity to the internal carotid or vertebral arteries, such as basilar skull fractures, cervical spine fracture-dislocations, or fractures extending into the foramen transversarium1,2 • History of significant blunt force injury to the head and neck, especially with a mechanism of injury involving cervical hyperextension-rotation, hyperflexion, or direct blow to the neck • New focal neurologic deficit or stroke of unexplained cause on head CT in a patient with history of trauma; focal neurologic deficit in the setting of questionable or minimal trauma, especially if the patient was normal at admission1 • Pain in the neck, face, scalp, or head in the setting of questionable or minor trauma

CONTRAINDICATIONS Relative contraindications include: • Presence of dangerous collaterals or routes of blood flow to the orbit and brain; if these are present and cannot be

protected, endovascular therapy may pose a significant risk for stroke. • Reopening of an acute traumatic vascular occlusion should be done with extreme caution because of the risk of thromboembolic stroke, and should be performed only when essential for cerebral perfusion. In many instances, particularly in patients with an intact circle of Willis and good intracranial collaterals, intentional occlusion of severely injured vessels or coil embolization to secure a traumatic vascular occlusion should be considered among the treatment options. • Some patients may better be served by medical or surgical therapy. • Triage issues may dictate the timing or extent of treatment.

EQUIPMENT • Angiography tray • Rotating hemostatic valves • Continuous flush system using heparinized saline (20004000 units/L) for sheaths, catheters, and microcatheters) (Fig. e97-1) • 6F (or larger) access sheath • 5F to 6F guide catheter • 6F (or larger) introducer guide sheath, 90 to 100 cm long, for stent delivery • End-hole microcatheter appropriate for use with chosen embolic agents, and 0.014-inch microwire; exchange length microwire may be needed for some stent delivery systems or tortuous anatomy • Particulate embolic agents for epistaxis and facial hemorrhage • Liquid embolic agents (n-butyl cyanoacrylate [NBCA], Onyx) have been used for treatment of life-threatening facial hemorrhage. • Fibered 0.018-inch push coils and coil pusher for distal arterial injuries • Fibered 0.035-inch push coils and Benson 0.035-inch wire for proximal arterial injuries • Detachable 0.020-inch, 0.018-inch, 0.014-inch, 0.012-inch coils for occlusion of pseudoaneurysm sacs • Stent(s) suitable for delivery into the extracranial internal carotid or vertebral arteries for treatment of cervical arterial dissection or pseudoaneurysm • Covered stent(s) suitable for delivery into the extracranial internal carotid artery for treatment of cervical arterial lacerations and transections, including arteriovenous and cavernous carotid fistulae (CCF) 729

CHAPTER 97 MANAGEMENT OF HEAD AND NECK INJURIES729.e1

FIGURE e97-1. Sample continuous flush system for neurointerventional procedures. On the left, a one-way stopcock is interposed between guide catheter and a rotating hemostatic valve (RHV) to prevent back-bleeding during microcatheter insertion. Three-way stopcock is connected to sidearm of RHV for attachment of a continuous flush line and for injection of contrast material. Microcatheter is also attached to an RHV to permit continuous flushing during insertion of microwire.

730 PART 1 VASCULAR INTERVENTIONS · SECTION 11 NEUROVASCULAR AND HEAD AND NECK INTERVENTION

FIGURE 97-1. Computed tomographic angiography coronal multiplanar reconstruction image demonstrates left internal carotid artery (ICA) dissection and pseudoaneurysm (arrow), and normal right ICA.

TECHNIQUE: EPISTAXIS, FACIAL FRACTURES Anatomy and Approaches The cervical, facial, and cranial regions are supplied by a rich vascular tree, including the external carotid artery (ECA), internal carotid artery (ICA), vertebral artery (VA), thyrocervical trunk, and costocervical trunk. Branching patterns can be variable, and numerous potential pathways of collateral flow exist between the extracranial and intracranial arteries.4 Thus, careful vigilance for the presence of these dangerous collaterals is essential in preventing inadvertent embolization of the retina, brain, or cranial nerves. If time and the patient’s condition permit, CTA to evaluate the extent of bony injury and identify potential sites of arterial injury often facilitates delivery of definitive treatment, and is well worth the additional “cost” in examination time, contrast administration, and patient radiation exposure.3 In epistaxis and facial smash injuries, the most common culprits amenable to embolization are branches of the facial and internal maxillary arteries. The facial artery arises as a proximal branch of the ECA (either directly or as a common trunk with the lingual artery), ascends superficial to the mandible, then courses medially in the superficial soft tissues of the face, giving off branches to the lips, mandible, cheeks, and nasal cavity. The internal maxillary artery (IMA) arises as one of the terminal branches of the ECA at the level of the ramus of the mandible. It passes forward almost horizontally, medial to the mandibular ramus, giving off branches to the soft tissues (transverse facial artery), meninges (middle and accessory meningeal arteries) and, as it subsequently courses either deep or superficial to the lateral pterygoid muscle, to the muscles of mastication (masseter, temporalis, mylohyoid, and pterygoids). The

most distal part of the IMA passes through the pterygomaxillary fissure and into the pterygopalatine fossa; this distal segment gives rise to numerous branches that can be injured in facial fractures, particularly LeFort fractures, including the anterior and posterior superior alveolar arteries, descending palatine artery, infraorbital artery, sphenopalatine artery, and branches to the nasal cavity (Fig. 97-2). Facial and nasal cavity bleeding can also occur as a consequence of injury to intracranial branches of the ICA. Persistent nasal bleeding may arise from the posterior and anterior ethmoidal arteries, which arise from the ophthalmic artery (OphA) distal to the takeoff of the central retinal artery. The posterior ethmoidal artery courses medially, passing between the medial rectus and superior oblique muscles, and enters the nasal cavity via the posterior ethmoidal canal to supply posterior nasal air cells and the upper nasal septum. The anterior ethmoidal artery also courses medially, passing through the anterior ethmoidal canal, and supplies the anterior and middle ethmoidal air cells, frontal sinus, and anterosuperior aspect of the lateral nasal wall. These ethmoidal branches are generally not amenable to safe direct embolization, and a failure to recognize them prior to ECA branch embolization can result in unilateral vision loss or stroke (Fig. 97-3).

Technical Aspects In patients with epistaxis or facial injury, it is generally prudent to secure the airway by intubating the patient. The common femoral artery is entered using Seldinger technique, and a 6F or larger sheath inserted. A 5F or 6F guide catheter is then passed over a suitable 0.035- or 0.038-inch guidewire into the aortic arch. The large internal diameter of many available guide catheters is adequate to obtain a good roadmap by injecting around a coaxially placed microcatheter. There are two general approaches to angiographic protocol: (1) diagnostic angiography of all vessels that are potentially a source of bleeding, followed by embolization of relevant arteries, or (2) angiography and embolization of one vessel, then moving on to another. The second approach is recommended only when the source of bleeding is clear on the basis of antecedent clinical or imaging findings. In either case, careful evaluation of the injured vessel and its runoff for possible communication with cerebral collaterals must be performed prior to intervention. The common carotid artery (CCA) is selected, with care to keep the catheter and wire below any potential site of injury. Digital subtraction angiography (DSA) over the neck, skull base, and head is then performed to evaluate for ICA dissection, pseudoaneurysm, or extravasation arising from the cervical ICA or ECA. Treatment of cervical ICA dissections, pseudoaneurysms, and lacerations is addressed in a separate section later. Large ECA branch lacerations causing profuse hemorrhage are best treated with particulate embolization followed by coil occlusion. Although the ECA main trunk and proximal portions of ECA branches can be occluded using 0.035-inch fibered push coils (delivered via a 5F catheter) or by 0.018-inch fibered push coils (delivered via microcatheter) alone, the offending ECA branch may continue to fill via retrograde collaterals from the opposite side. Selective roadmap imaging of the ECA, followed by passage of a

CHAPTER 97 MANAGEMENT OF HEAD AND NECK INJURIES 731

A

B

FIGURE 97-2. Lateral (A) and anteroposterior (B) angiograms of external carotid artery demonstrate main branches, including (1) lingual artery, (2) facial artery, (3) ascending pharyngeal artery, (4) internal maxillary artery, (5) sphenopalatine artery, (6) middle meningeal artery, (7) superficial temporal artery, and (8) occipital artery.

A

B

C

FIGURE 97-3. Left external carotid artery (ECA) anteroposterior (A) and lateral (B) angiograms, and left internal carotid artery (ICA) lateral (C) angiogram reveal anastomosis between distal internal maxillary artery and ophthalmic artery (OphA) (black arrow) via posterior ethmoidal artery collaterals. Note retrograde filling of OphA; identical vessel morphology is seen on ICA injection (open arrows).

microwire and microcatheter across (if possible) or closely proximal to the site of extravasation, are performed. Subsequent DSA with microcatheter injection is then done to evaluate for dangerous collaterals. If none are present, small branches that could serve as potential channels for collateral supply are then occluded with polyvinyl alcohol (PVA) particles larger than 250 μm, followed by occlusion of the lacerated branch with 0.018-inch fibered push coils, using either the microwire or a coil pusher for delivery. In many if not most cases of epistaxis or bleeding due to facial fractures, however, no obvious source of bleeding is

evident on the initial CCA angiogram. Often the source of hemorrhage is seen only on selective or superselective angiography. Selective ICA angiography should be performed to rule out laceration of this vessel as a direct source of bleeding into the sphenoid sinus, evaluate ethmoidal branches of the OphA as either direct or collateral sources of hemorrhage, and look for other potential dangerous collaterals. Selective ECA angiography should then be performed to evaluate for arterial truncations (and thus potential lacerations), extravasation, pseudoaneurysms, or regions of unusual vascular blush (Fig. 97-4). The ECA

732 PART 1 VASCULAR INTERVENTIONS · SECTION 11 NEUROVASCULAR AND HEAD AND NECK INTERVENTION (right IMA, left IMA, right facial, left facial) should be embolized. Internal Maxillary Artery Technique After obtaining a roadmap image of the ECA, the microcatheter and microwire are passed coaxially through the guide catheter into the IMA. Superselective anteroposterior (AP) and lateral angiography, with hand injection of the microcatheter from a position in the mid-IMA, can be performed prior to embolization to check for dangerous collaterals. If any are seen, great care must be taken to prevent reflux, or embolization of this branch may have to be abandoned. If it appears safe to proceed, the microcatheter is advanced toward the terminal sphenopalatine branch. Facial Artery Technique After obtaining roadmap imaging of the ECA, the microcatheter and microwire are passed coaxially into the facial artery, usually just beyond the point where it crosses the mandible. Hand-injection DSA is performed to evaluate for lacerations that may require coil occlusion and for potential dangerous collaterals to the orbit. If embolization is being performed primarily for nasal injury or epistaxis, the catheter is moved sufficiently distal to prevent inadvertent reflux into the main ECA trunk, and even more importantly the ICA. FIGURE 97-4. Lateral external carotid artery angiogram demonstrating a small pseudoaneurysm (arrowhead) arising from a branch of internal maxillary artery.

angiogram and any selective branch angiograms should be carefully evaluated for the presence of any branches projecting over the orbit and any collateral branches filling the ICA or OphA prior to performing embolization (see Fig. 97-3). Therapeutic small-artery occlusion in the setting of epistaxis or trauma is most commonly performed using various preparations of PVA particles, typically 250 to 355 μm in size, suspended in a 1 : 1 or 2 : 1 mixture of contrast material to saline. This choice of particulate size effectively controls bleeding but allows small mucosal arteries to remain intact.5 Particles larger than 250 μm also help avoid inadvertent embolization of perineural vessels and some potentially dangerous collaterals.4 The concentration of PVA in the suspension should be sufficiently dilute to avoid clumping or clogging of the microcatheter. Great care should be taken to prevent proximal reflux of particulate material into potential anastomotic branches and the ICA. It is also important to avoid a wedged catheter position, because the resultant increased injection pressure can sometimes open up previously unseen dangerous collateral branches. If obvious sources of bleeding are seen on the diagnostic angiograms, these vessels are embolized first. If the site of bleeding is occult but clearly unilateral, the ipsilateral IMA and facial arteries may be embolized empirically. When the source of difficult-to-control bleeding is not evident, branches of the IMA are often to blame. In all cases, however, a maximum of three out of the four arteries providing most of the supply to the midline midface

Internal Maxillary or Facial Artery Technique A blank roadmap image is obtained. The microcatheter is disconnected from the continuous flush system, and hand injection of particles is performed via a 1-mL syringe, with 0.1- to 0.3-mL aliquots at a time injected under constant direct fluoroscopic visualization. Some operators prefer to inject particulate embolic using a “closed” system: a 3-way stopcock is interposed between the microcatheter hub and 1-mL syringe, with a reservoir syringe containing particulate suspension attached to the 3-way sidearm. Other operators simply draw the embolic suspension from a small medicine cup. Alternating injection of particulate suspension with injection of saline helps prevent clogging of the microcatheter, as does frequent inspection of the microcatheter hub for layering of particles. Embolization should continue until distal branches become “pruned” (Fig. 97-5); nasal blush will decrease concomitantly. Control angiograms via both the microcatheter and guide catheter should be performed intermittently during embolization to assess for additional suspicious vessels and any dangerous collaterals that may open up as small branches are occluded.

Controversies ECA embolization for control of hemorrhage is safe and effective so long as appropriate care is taken to ensure delivery of embolic agent into target vessels and avoid delivery into intracranial branches and dangerous collaterals.

Outcomes Technical success is defined as occlusion of the targeted vessels, and in most reported series is greater than 90%.5 Clinical success is defined in most cases as cessation or control of hemorrhage; this is achieved in the majority of cases.

CHAPTER 97 MANAGEMENT OF HEAD AND NECK INJURIES 733

A

B

FIGURE 97-5. Lateral external carotid artery (ECA) angiogram before (A) and after (B) polyvinyl alcohol embolization, demonstrating desired pruning (arrowheads) of distal ECA branches.

Complications Major potential complications include vision loss or cranial nerve palsy (110 pg/mL increase in serum gastrin over basal levels after 0.4 μg/kg is given intravenously).1 The role of preoperative imaging in patients with ZES is to identify the location and number of primary tumors and assess for metastases. The initial imaging study of choice is somatostatin receptor scintigraphy (SRS).7 Gastrinomas express somatostatin receptors to which the somatostatin analog octreotide binds. SRS is the most sensitive noninvasive imaging modality for detecting primary gastrinomas or liver metastases. It detects approximately 58% of primary tumors and 92% of liver metastases,8 but sensitivity depends on tumor size. SRS detects 30% of tumors 1.1 cm or less, 64% of tumors between 1.1 and 2 cm, and 96% of tumors larger than 2 cm.9 SRS misses mainly

CHAPTER 111 ARTERIOGRAPHY AND ASVS FOR LOCALIZING PANCREATIC ENDOCRINE TUMORS 817 duodenal gastrinomas, which tend to be small (≤1 cm).9 Because SRS does not provide information on tumor size and gives only an approximate location, CT or MRI, also performed routinely, are useful for detecting larger primary tumors and distant metastases. Single-photon emission computed tomography (SPECT)-CT hybrid imaging and software fusion of SRS with CT or MRI may help localize areas of increased activity on SRS. If results of both SRS and CT or MRI are negative, an invasive localization study such as endoscopic ultrasonography (EUS) or arteriography with ASVS using secretin should be considered. The choice depends on local expertise. EUS has a sensitivity of about 85% in detecting pancreatic gastrinomas and about 43% in detecting duodenal gastrinomas.4,10 Limitations of EUS include low sensitivity for tumors in the duodenal wall, tumors in the pancreatic tail, and small ( 1.5 mg/ dL) is also a contraindication or should be carefully treated.

EQUIPMENT Balloon Catheter A 6F to 7F Simmons-shaped balloon catheter has generally been used for occlusion of gastrorenal shunts (Fig. 113-4). The balloon catheter should have a wide lumen through which a microcatheter can be inserted and should also have a large-diameter balloon (>15 mm) to occlude a large shunt completely. A 5F or 6F C-shaped catheter with a balloon has been used for occlusion of gastrocaval shunts. For the technique of selective BRTO, a 9F/5F coaxial balloon catheter that can easily be advanced proximally to the varices (Fig. 113-5) is used.10 A microcatheter can be inserted through the 5F coaxial balloon catheter. The 9F guiding balloon is positioned at the outlet of the gastrorenal shunt, and the 5F balloon catheter is introduced coaxially into the proximal portion of the shunt over a microcatheter advanced into the varices. The sclerosing agent is then infused via the microcatheter under balloon occlusion of the shunt at the outlet and proximal portion. Microballoon catheters are recently available, which is useful for selective BRTO in cases with a small and tortuous access route (Fig. 113-6). These catheters come in 3.3F to 3.5F sizes with a 6- to 10-mm balloon (see Fig. 113-5), and they can be introduced through a 6F guiding catheter or 4F sheath over a microguidewire of 0.014 inch.

Sclerosing Agent For BRTO, 5% ethanolamine oleate iopamidol (EOI), which consists of a mixture of 10% ethanolamine oleate (Oldamin) and the same dose of a contrast agent (iopamidol 300), is commonly used. Because some types of contrast material cannot be mixed with ethanolamine oleate, we used iopamidol 300. Ethanolamine oleate induces hemolysis in blood vessels, and free hemoglobin is released, which may result in renal tubular disturbances and acute renal failure. To prevent renal insufficiency, intravenous administration of 4000 units of haptoglobin, which combines with free hemoglobin, is performed during the procedure. Polidocanol in the form of foam has been recently used as a sclerosing agent alternative to ethanolamine olerate.11 Foam sclerotherapy using polidocanol may reduce the drug amount and maximize the effect by increasing the contact surface area with the venous wall. However, one should be aware of the 829

830 PART 1 VASCULAR INTERVENTIONS · SECTION 14 INTERVENTION IN PORTAL HYPERTENSION

FIGURE 113-1. Gastroesophageal varices. Coronal multiplanar reconstructed computed tomography image shows cardiac gastric varices (GV) draining through esophageal varices (white arrows) into azygos vein (black arrows).

Sclerosant (purple color) in varices

Balloon catheter occluding gastro-renal shunt and injecting sclerosant

FIGURE 113-3. Schematic drawing of standard balloon-occluded retrograde transvenous obliteration technique. Balloon catheter is introduced into outlet of gastrorenal shunt from a femoral venous approach, then sclerosant is slowly injected via balloon catheter, with balloon occlusion of shunt. Sclerosant fills varices.

FIGURE 113-2. Isolated gastric varices. Coronal multiplanar reconstructed computed tomography image shows gastric varices (GV) draining through left inferior phrenic vein (IPV) (gastrorenal shunt [black arrows]) into left renal vein (LRV). Note that varices also drain through subdiaphragmatic branch of left inferior phrenic vein into pericardiophrenic vein (white arrows).

FIGURE 113-4. Photograph of standard Simmons-shaped balloon catheter for balloon-occluded retrograde transvenous obliteration procedure.

CHAPTER 113 RETROGRADE BALLOON OCCLUSION VARICEAL ABLATION 831 potential risks of air embolism. Some authors used 50% glucose solution, which is infused before injection of EOI to occluded collateral veins or to replace the blood. This technique can reduce the amount of EOI.12 Sodium tetradecyl sulfate (STS) is the most commonly used sclerosant within the United States. It is used at the 3% concentration and is often mixed with Ethiodol (for opacity) and air to create a foam solution. A 3 :2 : 1 ratio (air/ STS/Ethiodol) has been used successfully by the University of Virginia group. Other ratios have also been used successfully. The added Ethiodol reduces the sclerosing effect to some degree, so there is a balance between opacity and effectiveness.

TECHNIQUE FIGURE 113-5. Photograph of a 9F/5F coaxial double-balloon catheter system (left) and microballoons with 3.3F (middle) and (right) for the selective balloon-occluded retrograde transvenous obliteration technique.

Anatomy and Approach Isolated gastric varices are usually located at the fundus or at the cardia and fundus and develop independently as part

A

B

C

D

FIGURE 113-6. Gastric varices with gastrocaval shunt treated by selective balloon-occluded retrograde transvenous obliteration (BRTO) with microballoon catheter. A-B, Postcontrast computed tomography shows gastric varices (white arrows) draining via a small gastrocaval shunt (arrowheads) into inferior vena cava. Stenosis (arrow) at proximal portion of gastrocaval shunt is noted. C, Retrograde venography using a microcatheter shows significant stenosis (arrows) at proximal portion of gastrocaval shunt. Multiple collateral drainages are also noted. D, A 7F guiding catheter is positioned proximally to stenotic portion of gastrocaval shunt, and a 3.3F microballoon catheter (arrowhead) is advanced distally beyond collateral drainage into variceal draining vein. Fluoroscopic image during BRTO shows gastric varices sufficiently opacified with sclerosant. Arrows indicate a guidewire introduced into peripheral branch of inferior phrenic vein to stabilize guiding catheter.

832 PART 1 VASCULAR INTERVENTIONS · SECTION 14 INTERVENTION IN PORTAL HYPERTENSION of a large portosystemic shunt that runs through the stomach wall and drains into the left renal vein or IVC.2 The portosystemic shunt consists of afferent gastric veins and the left inferior phrenic vein, which drains into the left renal vein (gastrorenal shunt; 80%-85%) or directly into the IVC (gastrocaval shunt; 10%-15%) or into both the renal vein and the IVC. The left inferior phrenic vein often communicates with other peridiaphragmatic veins, including the left pericardiophrenic vein, intercostal vein, and azygos vein. These peridiaphragmatic veins, along with the main portosystemic shunt, act as collateral veins draining the gastric varices and would be an obstacle to sufficient filling of sclerosant in the gastric varices during BRTO. Knowledge of potential communications between the gastric varices and peridiaphragmatic veins is very important for successful treatment of gastric varices by BRTO.4,5,13 The left inferior phrenic vein often ends in the left renal vein together with the left adrenal vein or passes in front of the esophageal hiatus in the diaphragm and ends in the IVC

or left hepatic vein. The proximal portion of the left inferior phrenic vein runs inferior to the diaphragm, and its peripheral branches run superior to the diaphragm. The left inferior phrenic vein potentially communicates with other peridiaphragmatic veins, including the left pericardiophrenic vein, intercostal vein, and anastomotic veins to the right inferior phrenic vein or azygos venous system14; consequently, fundic varices can also drain through these potential communications (Fig. 113-7; also see Fig. 113-2). In our experience with portography, balloon-occluded venography, or both in 85 patients with isolated gastric varices, the main drainage routes from the gastric varices were gastrorenal shunts in 52 cases (61.2%), gastrocaval shunts in 3 cases (3.5%), both gastrorenal and gastrocaval shunts in 26 cases (30.6%), and other peridiaphragmatic veins in 4 cases (4.7%). In a portographic study of 20 patients by Chikamori et al., isolated gastric varices drained into a gastrorenal shunt in 85%, a gastrocaval shunt in 10%, and the pericardiophrenic vein in 5%.2

Peridiaphragmic veins

Azygos and Hemiazygos venous system

Pericardiophrenic vein

Gastrocaval (GC) shunt

Inferior vena cava

Esophageal plexus

Gastrorenal (GR) shunt Gastric veins

Subcostal and Intercostal veins Gastric varices protruding into gastric lumen Paragastric varices

Suprarenal vein Left kidney

Left inferior phrenic vein

Left renal vein

FIGURE 113-7. Schematic drawing of isolated gastric varices draining through various peridiaphragmatic drainage routes. Isolated gastric varices can drain through gastrorenal shunt into left renal vein or through gastrocaval shunt into inferior vena cava or pericardiophrenic vein (or both), intercostal vein, and azygos/hemiazygos vein.

CHAPTER 113 RETROGRADE BALLOON OCCLUSION VARICEAL ABLATION 833 Gastrorenal Shunt A gastrorenal shunt ends in the left renal vein together with or separately from the left adrenal vein. It often has small retroperitoneal veins communicating with the azygos venous system, and the left adrenal vein and is sometimes associated with other large collaterals through the gastrocaval shunt or left pericardiophrenic vein (see Fig. 113-2). Gastrocaval Shunt A gastrocaval shunt ends in the IVC or left hepatic vein (Fig. 113-8). The left inferior phrenic vein runs above the diaphragm peripherally, penetrates the diaphragm near the cardiac apex, and then travels on the inferior surface of

the diaphragm. The peripheral portion of the left inferior phrenic vein has many collateral veins, so the majority of gastrocaval shunts are always associated with additional collateral drainage, including the left pericardial vein, veins of the thoracic wall, and an anastomotic vein to the right inferior phrenic vein (see Fig. 113-7). BRTO for this type of gastric variceal drainage is often more difficult than that for varices draining only into the gastrorenal shunt. Pericardiophrenic Vein The diaphragmatic branches of the left pericardiophrenic vein anastomose with the supradiaphragmatic portion of the inferior phrenic vein, which ascends parallel to the left cardiac border and drains into the left brachiocephalic vein, the superior intercostal vein, or the internal mammary vein. The vein can function as a collateral pathway in patients with occlusion of the IVC or superior vena cava (SVC) and portal hypertension. Pericardiophrenic venous drainage from varices is usually associated with a gastrocaval or gastrorenal shunt (Fig. 113-9), but it can exist alone. Performance of BRTO through the left pericardiophrenic vein has a potential risk for serious arrhythmia secondary to stimulation by catheter manipulation or sclerosant. Veins of the Thoracic Wall Peripheral branches of the left inferior phrenic vein anastomose with the left lower intercostal veins that drain into the hemiazygos vein posteriorly or into the internal mammary vein anteriorly. These thoracic wall veins can also function as drainage veins supplementing the main drainage routes from the gastric varices (Fig. 113-10).

FIGURE 113-8. Coronal maximum intensity projection shows gastric varices (GV) draining through gastrocaval shunt (arrows) into inferior vena cava.

A

B

Azygos and Hemiazygos Veins The azygos vein begins as a continuation of the right ascending lumbar vein, enters the thorax through the aortic hiatus in the diaphragm, ascends along the right side of the vertebral column, and enters the posterior aspect of the SVC. The hemiazygos vein begins in the left ascending lumbar or renal vein, ascends on the left side of the vertebral column,

C

FIGURE 113-9. Gastric varices draining through a gastrorenal shunt and pericardiophrenic vein. A, Balloon-occluded venography of gastrorenal shunt shows diaphragmatic branch of left inferior phrenic vein continuing to left pericardiophrenic vein (arrows). B, Diaphragmatic branch was occluded with coils (arrow). C, Fluoroscopic image during balloon-occluded retrograde transvenous obliteration after embolization of collateral vein shows sufficient opacification of sclerosant in varices.

834 PART 1 VASCULAR INTERVENTIONS · SECTION 14 INTERVENTION IN PORTAL HYPERTENSION

FIGURE 113-10. Balloon-occluded venography of a gastrorenal shunt shows contrast medium draining through diaphragmatic branch of left inferior phrenic vein into inferior vena cava (gastrocaval shunt [arrowheads]) and into internal mammary vein (arrows).

FIGURE 113-11. Balloon-occluded venography of a gastrorenal shunt shows small veins draining to paravertebral plexus and azygos vein (arrows). Note anastomosis with pericardiophrenic vein (arrowheads).

and then joins the azygos vein at the T8-T9 level. The azygos and hemiazygos veins receive the intercostal, pericardial, esophageal, and mediastinal veins. Although the azygos/ hemiazygos vein is the most important drainage route from esophageal and gastroesophageal varices that drain directly into the azygos system (see Fig. 113-1), isolated gastric varices less frequently and indirectly drain into the azygos venous system.2 There are small venous communications between the paravertebral venous plexus and the left inferior phrenic vein. These small anastomotic veins are often observed on balloon-occluded venography of the gastrorenal shunt (Fig. 113-11). Another important issue regarding the anatomy of gastric varices for BRTO is the existence of a direct shunting vein (paragastric varices).5,9,13 Gastric varices are located in the submucosal layer and protrude into the gastric lumen. Paragastric varices, which represent variceal shunting veins that connect the afferent gastric veins directly to the draining veins, run around the gastric wall and are sometimes associated with gastric varices. When gastric varices with a large paragastric component are treated by BRTO, sclerosant injected via a balloon catheter may fill the paragastric varices more than the gastric varices.

is slowly and intermittently injected through the balloon catheter until the gastric varices are completely filled with EOI. After 30 to 50 minutes, EOI is aspirated via the balloon catheter as much as possible. The balloon is then deflated and the balloon catheter withdrawn. The degree of collateral venous drainage is one of the most important factors for a successful procedure. Gastric varices without collateral drainage or with minimum-flow collaterals can easily be treated with the standard BRTO technique because sclerosant infused via the balloon catheter can fill and stagnate in the varices (Fig. 113-12). Some modification in technique is required for the treatment of gastric varices with moderate- or high-flow collateral drainage. We routinely use a 9F/5F coaxial balloon catheter and a microcatheter for selective infusion of sclerosant as mentioned earlier.10 When the 5F coaxial balloon catheter can be advanced beyond the collateral veins close to the varices, the infused sclerosant can stagnate in the varices under proximal balloon occlusion (Fig. 113-13). For the treatment of cases with moderate- or high-flow collateral drainage, the selective BRTO technique should be attempted first because it requires no further equipment such as coils or an additional balloon (Fig. 113-14). However, in some cases the balloon catheter cannot be advanced beyond the collaterals because of significant tortuosity of the shunt. If proximal balloon occlusion fails, occlusion of these collateral veins is required. Techniques for occlusion of collateral drainage veins, including coil embolization, infusion of liquid occlusive agents such as ethanol or 50% glucose, and balloon occlusion (double-balloon technique), are applied according to the size and ability to catheterize the collateral veins (Figs. 113-15 and 113-16; also see Fig. 113-9). When the collateral drainage vein is large enough that it can be

Technical Aspects The standard technique of BRTO was introduced by Kanagawa et al.3 (see Fig. 113-3). According to their description, a balloon catheter is inserted into the outlet of the gastrorenal shunt from a femoral venous approach. After occlusion of the shunt, balloon-occluded venography is performed with the injection of approximately 8 mL of contrast material (iopamidol 370) via the balloon catheter. Then 5% EOI

CHAPTER 113 RETROGRADE BALLOON OCCLUSION VARICEAL ABLATION 835 Gastric varices 5F balloon catheter

2-3F microcatheter

9F balloon catheter

FIGURE 113-12. Gastric varices without collateral venous drainage. Fluoroscopic image during a standard balloon-occluded retrograde transvenous obliteration procedure shows sufficient opacification of sclerosant in varices.

retrogradely catheterized with a balloon catheter, the double-balloon technique is performed.5 Collateral veins with a diameter of 2 to 7 mm can be embolized with coils. Small collateral veins less than 2 mm or those that cannot be catheterized are treated with liquid occlusive agents. Although ethanol can occlude these veins instantly, it is more frequently associated with side effects (e.g., serious arrhythmia) than 50% glucose, so we use 50% glucose for such cases. Treatment strategies related to the size of the collateral drainage vein are summarized in Table 113-1. Types of afferent veins are another important factor for successful BRTO. Insufficient filling of sclerosant is found in some cases with varices supplied by multiple gastric veins of the left gastric vein, posterior gastric vein, and the short gastric vein, although the systemic drainage from gastric varices is completely blocked by balloon occlusion. Insufficient filling of gastric varices is caused by alternative flow through the gastric varices from a gastric vein to another gastric vein due to pressure gradient under balloon occlusion of systemic drainage. Sclerosant injected retrogradely flows into the gastric vein with lower pressure without filling into the higher pressure part of the gastric varices in such cases (Fig. 113-17). This could result in incomplete thrombosis of the gastric varices after treatment.15 The addition of afferent venous occlusion with transportal approaches are thought to be useful techniques for ensuring sufficient distribution of sclerosant into complex gastric varices. However, these techniques are associated with periprocedural risks such as peritoneal hemorrhage. Temporary occlusion of the splenic artery during BRTO is a safer and useful technique that can reduce the venous pressure

FIGURE 113-13. Schematic drawing of proximal balloon occlusion technique using a coaxial double-balloon catheter system. A 9F balloon catheter is positioned at outlet of gastrorenal shunt, a 5F flexible balloon catheter is coaxially advanced beyond collateral drainage veins, and then a microcatheter is introduced through the 5F catheter into varices. Sclerosant (purple color) is slowly injected via microcatheter into varices under balloon occlusion of shunt at both proximal and distal portions.

of the posterior gastric vein and short gastric vein, so altering the pressure gradient in the varices and promoting filling of the entire varices with sclerosant (Fig 113-18).16

CONTROVERSIES In the BRTO technique, there are no significant controversies except for some details. The duration of balloon occlusion varies among physicians.3-5,9,15 We usually perform BRTO with 30- to 60-minute balloon occlusion times, but several authors have recommended overnight balloon occlusion. In such long-duration balloon occlusion techniques, most physicians perform balloon-occluded venography or contrast-enhanced computed tomography (CT) to evaluate the degree of thrombosis of the varices just before deflation of the balloon. Sclerosant is again injected through the balloon catheter when residual varices are demonstrated by balloon-occluded venography or contrastenhanced CT. Therefore, the long-duration balloon occlusion technique is more effective for thrombosis of varices than the short-duration balloon occlusion technique. Pulmonary embolism, which can be caused by migration of thrombi formed in the gastric varices and drainage vein, is extremely rare but one of the most serious complications of BRTO. Some physicians believe the risk of

836 PART 1 VASCULAR INTERVENTIONS · SECTION 14 INTERVENTION IN PORTAL HYPERTENSION

A

B

C

D

E

F

FIGURE 113-14. Gastric varices treated by coaxial double-balloon catheter system. A, Computed tomography (CT) after enhancement with contrast material shows large gastric varices (arrows) at fundus. B, Gastroendoscopy shows large tumorous gastric varices. C, Balloon-occluded venography of gastrorenal shunt shows multiple collateral drainage veins. Gastric varices are not opacified. D, Fluoroscopic image during balloon-occluded retrograde transvenous obliteration (BRTO) with a coaxial double-balloon catheter system. A 9F balloon catheter (black arrow) is positioned at outlet of gastrorenal shunt. A 5F balloon catheter (arrowhead) is coaxially advanced beyond collateral drainage, and a microcatheter (white arrow) is introduced into varices. Gastric varices are well opacified with sclerosant under double-balloon occlusion of shunt. E, CT during a BRTO procedure shows sclerosant sufficiently filling varices. F, Follow-up CT 1 week after BRTO shows complete thrombosis of varices. G, Gastroendoscopy 1 month after BRTO shows marked regression of gastric varices.

pulmonary embolism after BRTO can be reduced by longduration balloon occlusion techniques. However, from our data and those reported by other investigators, the results of short-duration balloon occlusion are satisfactory and similar to those of long-duration balloon occlusion techniques in both effectiveness and occurrence of pulmonary embolism.3,9 Furthermore, the long-duration balloon occlusion technique requires patients to be bedridden for a long period, with the potential risk of malpositioning or thrombogenesis of the balloon catheter. For these reasons we do not recommend the long-duration balloon occlusion technique initially, except for the special case of a very large draining shunt. For the treatment of collateral drainage, some physicians have infused ethanol via the balloon catheter before infusion of ethanolamine oleate for devascularization of the collaterals. Although infusion of ethanol via the balloon catheter is a simple and effective technique for devascularization of collaterals, it carries a potential risk of minor and major complications such as chest pain, intoxication, and arrhythmia.

Even though the BRTO technique has become the firstline treatment of isolated gastric varices in Japan, it has not yet spread worldwide. Some endoscopists or interventional radiologists recommend endoscopic treatment or a transjugular intrahepatic portosystemic stent shunt (TIPSS) for the treatment of gastric varices.16-23 TIPSS or endoscopic treatment may be effective for the management of esophageal or gastroesophageal varices that are not associated with a large portosystemic shunt, but these techniques are less effective for isolated varices with a large portosystemic shunt and low portal pressure. Although prospective randomized trials including a large number of cases have not been performed to clarify the efficacy and safety of these three techniques, the reported results of BRTO, including safety and long-term efficacy, are better than those of endoscopic treatment and TIPSS.3-5,9,18-24,25

OUTCOMES Reported technical success rates for BRTO procedures have ranged from 90% to 100%, and regression or disappearance

CHAPTER 113 RETROGRADE BALLOON OCCLUSION VARICEAL ABLATION 837 Gastric varices

EV

Coil emboliization

Microcatheter SGV

LGV

Balloon catheter in gastrorenal shunt

PGV GRS

SV

FIGURE 113-15. Schematic drawing of a technique for coil embolization of a collateral vein. A balloon catheter is positioned at outlet of gastrorenal shunt. Collateral drainage veins are occluded with coils, and then a microcatheter is introduced through balloon catheter into gastric varices. Sclerosant (purple color) is slowly injected via microcatheter into varices under balloon occlusion of shunt. Gastric varices

Balloon catheter in gastrocaval shunt Microcatheter

Balloon catheter in gastrorenal shunt

FIGURE 113-17. Schematic drawing of hemodynamic changes occurring in gastric varices supplied by multiple gastric veins after balloon occlusion of drainage shunt. When gastrorenal shunt (GRS) is occluded by a balloon, left gastric vein (LGV) and (para)esophageal varices (EV) can work as a draining vein (arrows) because they communicate with systemic vein of azygos venous system, and pressure in LGV would be lower than in other gastric veins. PGV, Posterior gastric vein; SGV, short gastric vein; SV, splenic vein.

TABLE 113-1. Use of BRTO Technique According to Types of Collateral Venous Drainage Type of Collateral Drainage

Recommended Technique

Collaterals with no or minimum flow

Standard or selective BRTO technique

Collaterals with moderate or high flow Small collaterals ( 5 mm)

Infusion of liquid embolic material (ethanol, 50% glucose) Coil embolization Double-balloon technique

BRTO, Balloon-occluded retrograde transvenous obliteration.

FIGURE 113-16. Schematic drawing of double-balloon catheter technique in a patient with two large portal systemic venous shunts. A balloon catheter is positioned at outlet of gastrorenal shunt, and another balloon catheter is retrogradely introduced into gastrocaval shunt. A microcatheter is then advanced through either catheter into gastric varices, and sclerosant is injected via microcatheter into varices.

of gastric varices on endoscopy was achieved in 80% to 100% of patients after BRTO. Some studies with long-term follow-up have shown recurrent gastric varices in 0% to 10% of cases. From our data, including 69 cases monitored over a period of 6 months, recurrence/regrowth of gastric varices was observed in 5 cases (7.2 %) on follow-up endoscopy. In all 5 recurrences, partial thrombosis of the gastric varices was noted on follow-up CT 1 or 2 weeks after BRTO. By contrast, in all 58 cases in which complete thrombosis of gastric varices was observed on CT 1 to 2 weeks after BRTO, no recurrences developed. Therefore, CT performed 1 to 2 weeks after BRTO can predict the long-term efficacy of BRTO. Ninoi et al. investigated prognostic factors in 78 patients after BRTO and demonstrated that the presence

838 PART 1 VASCULAR INTERVENTIONS · SECTION 14 INTERVENTION IN PORTAL HYPERTENSION

A

B

C

D

FIGURE 113-18. Gastric varices treated by balloon-occluded retrograde transvenous obliteration (BRTO) combined with temporary balloon occlusion of splenic artery. A, Maximum-intensity projection image of enhanced computed tomography shows large gastric varices (GV) supplied by multiple gastric veins of posterior gastric vein (PGV), left gastric vein (LGV), and short gastric vein (SGV). LGV continued to paraesophageal varices (paraEV). Selective balloonoccluded venography at early (B) and delayed (C) phases with injection of contrast materials via a microcatheter positioned in gastric varices. Contrast material does not stagnate in varices, which drain into LGV (arrows) and paraEV. D, Fluoroscopic image during BRTO with temporary occlusion of splenic artery. Sclerosant sufficiently fills and stagnates in gastric varices. Black arrowhead indicates balloon occluding splenic artery. White arrowhead indicates tip of microcatheter introduced into gastric varices. Arrows indicate double coaxial balloon catheter system placed in gastrorenal shunt.

of hepatocellular carcinoma (relative risk [RR] 24.342) and the Child-Pugh classification (RR 5.780) were statistically associated with decreased survival after BRTO.8 In two studies of BRTO for bleeding gastric varices, rebleeding after BRTO occurred in 0% and 9%.18,25 Several authors have reported improvement in hepatic function, including serum albumin levels, indocyanine green test, and symptoms of portosystemic encephalopathy after BRTO.6,26-28 These effects on hepatic function and portosystemic encephalopathy are due to an increase in hepatofugal portal flow as a result of obliteration of a large shunt. Because the majority of patients have hepatic cirrhosis secondary to hepatis B or C and BRTO does not have a direct effect on liver cirrhosis, these improvements in hepatic function and portosystemic

encephalopathy would be transient, and long-term efficacy would depend on the extent of residual hepatic function.

COMPLICATIONS Hemoglobinuria, abdominal/back pain, and low-grade fever are often observed during and for a few days after BRTO and can be treated conservatively with medication. To prevent renal dysfunction secondary to hemoglobinuria from the hemolytic effect of ethanolamine oleate, intravenous administration of haptoglobin is usually performed during the procedure. We observed one case of acute renal failure (0.5%) in 200 BRTO procedures. Therefore, the dosage of ethanolamine oleate should be decreased

CHAPTER 113 RETROGRADE BALLOON OCCLUSION VARICEAL ABLATION 839 as much as possible. We have not encountered any other serious complications, which are extremely rare; however, cardiogenic shock, atrial fibrillation, and pulmonary embolism have been reported.29 Increased portal venous pressure after BRTO secondary to obliteration of a large portosystemic shunt occurs frequently and would lead to development of ascites and aggravation of esophageal varices. The risk for aggravation of esophageal varices after BRTO was reported to be 27% at 1 year and 58% at 3 years.8 From our data, aggravation of esophageal varices after BRTO occurred in 45% at 1 year. Therefore, careful follow-up by endoscopy is required. Transient ascites was observed in 15%. Although all such cases could be controlled by medication, it may worsen the condition of patients.

regrow and drain via a large portosystemic shunt, they can be treated by BRTO again. Other treatment options, including TIPSS and endoscopic treatment, should be performed when an accessible shunt has been already thrombosed by the previous BRTO procedure. Esophageal varices that are exacerbated after BRTO should be treated endoscopically. KEY POINTS •

Balloon-occluded retrograde transvenous obliteration (BRTO) has become an effective minimally invasive treatment of fundal varices associated with a large portosystemic venous shunt such as a gastrorenal shunt.

•

Gastric varices are obliterated by injection of sclerosant via the shunt retrogradely into the varices under balloon occlusion of the shunt.

•

Complete obliteration of varices with long-term efficacy can be achieved when sclerosant fills and stagnates in the varices.

•

Knowledge of the peridiaphragmatic veins associated with varices is important because the presence of such veins draining gastric varices is well correlated with difficulty performing and reduced efficacy of BRTO.

POSTPROCEDURAL AND FOLLOW-UP CARE Because of the potential risk for acute renal failure secondary to the hemolytic effect of ethanolamine oleate and hepatic injury secondary to migration of the sclerosant into the portal venous system, we routinely check laboratory data, including serum creatinine, blood urea nitrogen, transaminase, and red blood cell and platelet counts on the first, third, and seventh days after BRTO. CT should be performed within 2 weeks after BRTO for evaluation of the degree of thrombosis of the varices and the development of ascites and for screening for other potential complications. As mentioned earlier, CT performed 1 to 2 weeks after BRTO is useful for predicting the long-term effect.9 We usually perform gastroendoscopy 1 week, 1 month, and 3 months after BRTO to assess changes in gastric varices, esophageal varices, and portal hypertensive gastropathy. Insufficient regression or regrowth of gastric varices on follow-up gastroendoscopy at 3 months would require repeat treatment. When varices remain or

▶ SUGGESTED READINGS Kiyosue H, Mori H, Matsumoto S, et al. Transcatheter obliteration of gastric varices. Part 1. Anatomic classification. Radiographics 2003;23: 911–20. Kiyosue H, Mori H, Matsumoto S, et al. Transcatheter obliteration of gastric varices: Part 2. Strategy and techniques based on hemodynamic features. Radiographics 2003;23:921–37.

The complete reference list is available online at www.expertconsult.com.

CHAPTER 113 RETROGRADE BALLOON OCCLUSION VARICEAL ABLATION839.e1 REFERENCES 1. Sarin SK, Kumar A. Gastric varices: profile, classification, and management. Am J Gastroenterol 1989;84:1244–9. 2. Chikamori F, Kuniyoshi N, Shibuya S, Takase Y. Correlation between endoscopic and angiographic findings in patients with esophageal and isolated gastric varices. Dig Surg 2001;18:176–81. 3. Kanagawa H, Mima S, Kouyama H, et al. Treatment of gastric fundal varices by balloon-occluded retrograde transvenous obliteration. J Gastroenterol Hepatol 1996;11:51–8. 4. Hirota S, Matsumoto S, Tomita M, et al. Retrograde transvenous obliteration of gastric varices. Radiology 1999;211:349–56. 5. Kiyosue H, Mori H, Matsumoto S, et al. Transcatheter obliteration of gastric varices: Part 2. Strategy and techniques based on hemodynamic features. Radiographics 2003;23:921–37. 6. Kato T, Uematsu T, Nishigaki Y, et al. Therapeutic effect of balloonoccluded retrograde transvenous obliteration on portal-systemic encephalopathy in patients with liver cirrhosis. Intern Med 2001;40: 688–91. 7. Takahashi K, Yamada T, Hyodoh H, et al. Selective balloon-occluded retrograde sclerosis of gastric varices using a coaxial microcatheter system. AJR Am J Roentgenol 2001;117:1091–3. 8. Ninoi T, Nishida N, Kaminou T, et al. Balloon-occluded retrograde transvenous obliteration of gastric varices with gastrorenal shunt: long-term follow-up in 78 patients. AJR Am J Roentgenol 2005;184: 1340–6. 9. Kiyosue H, Matsumoto S, Onishi R, et al. Balloon-occluded retrograde transvenous obliteration (B-RTO) for gastric varices: therapeutic results and problems. Nippon Igaku Hoshasen Gakkai Zasshi 1999;59: 12–9. 10. Tanoue S, Kiyosue H, Matsumoto S, et al. Development of a new coaxial balloon catheter system for balloon-occluded retrograde transvenous obliteration (B-RTO). Cardiovasc Intervent Radiol 2006;29: 991–6. 11. Koizumi J, Hashimoto T, Myojin K, et al. C-arm CT-guided foam sclerotherapy for the treatment of gastric varices. J Vasc Interv Radiol 2010;21:1583–7. 12. Yamagami T, Kato T, Hirota T, et al. Infusion of 50% glucose solution before injection of ethanolamine oleate during balloon-occluded retrograde transvenous obliteration. Australas Radiol 2007;51:334–8. 13. Kiyosue H, Mori H, Matsumoto S, et al. Transcatheter obliteration of gastric varices. Part 1. Anatomic classification. Radiographics 2003;23: 911–20. 14. Ibukuro K, Tsukiyama T, Mori K, Inoue Y. Precaval draining vein from paraesophageal varices: radiologic-anatomic correlation. AJR Am J Roentgenol 1999;172:651–4. 15. Takaji R, Kiyosue H, Matsumoto S, et al. Partial thrombosis of gastric varices after balloon-occluded retrograde transvenous obliteration: CT findings and endoscopic correlation. AJR Am J Roentgenol 2011;196:686–91 16. Kiyosue H, Tanoue S, Kondo Y, et al. Balloon-occluded retrograde transvenous obliteration of complex gastric varices assisted by

temporary balloon occlusion of the splenic artery. J Vasc Interv Radiol 2011;22:1045–8. 17. Shiba M, Higuchi K, Nakamura K, et al. Efficacy and safety of balloonoccluded endoscopic injection sclerotherapy as a prophylactic treatment for high-risk gastric fundal varices: a prospective, randomized, comparative clinical trial. Gastrointest Endosc 2002;56:522–8. 18. Kitamoto M, Imamura M, Kamada K, et al. Balloon-occluded retrograde transvenous obliteration of gastric fundal varices with hemorrhage. AJR Am J Roentgenol 2002;178:1167–74. 19. Arakaki Y, Murakami K, Takahashi K, et al. Clinical evaluation of combined endoscopic variceal ligation and sclerotherapy of gastric varices in liver cirrhosis. Endoscopy 2003;35:940–5. 20. Tripathi D, Therapondos G, Jackson E, et al. The role of the trans-jugular intrahepatic portosystemic stent shunt (TIPSS) in the management of bleeding gastric varices: clinical and haemodynamic correlations. Gut 2002;51:270–4. 21. Chau TN, Patch D, Chan YW, et al. “Salvage” transjugular intrahepatic portosystemic shunts: gastric fundal compared with esophageal variceal bleeding. Gastroenterology 1998;114:981–7. 22. Mahadeva S, Bellamy MC, Kessel D, et al. Cost-effectiveness of N-butyl-2-cyanoacrylate (Histoacryl) glue injections versus transjugular intrahepatic portosystemic shunt in the management of acute gastric variceal bleeding. Am J Gastroenterol 2003;98:2688–93. 23. Pomier-Layrargues G, Villeneuve J-P, Deschênes M, et al. Transjugular intrahepatic portosystemic shunt (TIPS) versus endoscopic variceal ligation in the prevention of variceal rebleeding in patients with cirrhosis: a randomised trial. Gut 2001;48:390–6. 24. Ninoi T, Nakamura K, Kaminou T, et al. TIPS versus transcatheter sclerotherapy for gastric varices. AJR Am J Roentgenol 2004;183: 369–76. 25. Choi YH, Yoon CJ, Park JH, et al. Balloon-occluded retrograde transvenous obliteration for gastric variceal bleeding: its feasibility compared with transjugular intrahepatic portosystemic shunt. Korean J Radiol 2003;4:109–16. 26. Akahane T, Iwasaki T, Kobayashi N, et al. Changes in liver function parameters after occlusion of gastrorenal shunts with balloon-occluded retrograde transvenous obliteration. Am J Gastroenterol 1997;92: 1026–30. 27. Miyamoto Y, Oho K, Kumamoto M, et al. Balloon-occluded retrograde transvenous obliteration improves liver function in patients with cirrhosis and portal hypertension. J Gastroenterol Hepatol 2003;18: 934–42. 28. Fukuda T, Hirota S, Sugimura K. Long-term results of balloonoccluded retrograde transvenous obliteration for the treatment of gastric varices and hepatic encephalopathy. J Vasc Interv Radiol 2001;12:327–36. 29. Shimoda R, Horiuchi K, Hagiwara S, et al. Short-term complications of retrograde transvenous obliteration of gastric varices in patients with portal hypertension: effects of obliteration of major portosystemic shunts. Abdom Imaging 2005;30:306–13.

SECTION FIFTEEN

HEMODIALYSIS ACCESS MANAGEMENT CHAPTER

114

Surveillance of Hemodialysis Access Anne Roberts

Large numbers of patients depend on hemodialysis for their survival. In the United States, approximately 370,000 individuals were receiving hemodialysis for end-stage renal disease (ESRD) through 2009.1 The increased incidence of diabetes in the U.S. population has led to an increased number of diabetic patients being treated for end-stage renal disease with hemodialysis. Growth in the incident population is particularly marked in the number of patients aged 45 to 64; growth in the population aged 75 and older has grown 12% since 2000.1 Because older patients are less likely to be candidates for renal transplantation, the number of patients on hemodialysis will increase as the population continues to age. Even patients who are transplant candidates will have a significant period of time on dialysis, because median wait times in the United States are more than 2 years and significantly increasing.1 In 2009 there were about 83,000 patients listed for transplants.1 For patients with ESRD either awaiting transplantation or who are not transplant candidates, their hemodialysis access is their “lifeline.” Dialysis access may take the form of dialysis catheters, dialysis grafts, or arteriovenous fistulas (AVFs). The AVF is the preferred access because it provides the greatest chance for long-term function. In the United States, the National Kidney Foundation Disease Outcomes Quality Initiative (NKF KDOQI) guidelines affect all dialysis centers.2 These guidelines have set a goal of 65% of all patients on hemodialysis having a functional AVF (“Fistula First Initiative”) and fewer than 10% of patients having a catheter for permanent dialysis. The rationale behind the encouragement of dialysis fistulas is the fistula’s excellent patency and low rate of complications.2 Synthetic grafts are an alternative to native fistulas, but they do not have the longevity that can be achieved with an endogenous fistula. Primary patency rates at 3 years are reported to be 40% to 60%.3 Secondary patency rates range from 50% to 90%.4 Vascular access function and patency are essential for patients on dialysis. If the access is not functioning properly, the patient will have inadequate dialysis, leading to increased morbidity and mortality.2 Numerous studies have shown a correlation between delivered dose of hemodialysis and patient mortality and morbidity.5 Vascular access–related complications account for 15% to 20% of hospitalizations for patients on hemodialysis.2 Because of the adverse outcomes for dialysis patients, preventing dialysis access dysfunction has become a focus of the KDOQI guidelines. Dysfunction and failure of AVFs or dialysis grafts are primarily due to development of stenoses in the veins associated with the fistula or graft. These stenoses lead to inadequate dialysis and then, potentially, thrombosis and failure 840

of the access. However, if the stenoses can be detected and treated, such problems can be reduced. Detection of stenoses can often be achieved through a systematic monitoring and surveillance program. It is critical to extend the functional life of each fistula for as long as possible because the sites available for dialysis fistulas are limited, and many patients are dependent on dialysis for survival.

INDICATIONS All patients who have hemodialysis fistulas or grafts should have periodic surveillance of the dialysis access flows. The KDOQI guidelines recommend an organized monitoring/ surveillance approach with regular assessment of clinical parameters of the dialysis access and hemodialysis adequacy. Data from the clinical assessment and hemodialysis measurements should be collected and tracked as part of a quality assurance program. In the KDOQI guidelines, monitoring refers to examination and evaluation of the vascular access by means of physical examination to detect physical signs that suggest the presence of dysfunction.2 Surveillance refers to periodic evaluation of the vascular access by using tests that may involve special instrumentation, and for which an abnormal test result suggests the presence of dysfunction.2 Diagnostic testing refers to specialized testing prompted by some abnormality or other medical indication and undertaken to diagnose the cause of the vascular access dysfunction. Physical examination should be performed on all fistulas and grafts at least monthly to detect dysfunction. Physical examination is useful, reliable, easily performed and inexpensive.6 A number of techniques may be used for surveillance of grafts. Those preferred include intra-access flow measurements, outlined in Table 114-1 and directly measured or derived from static venous dialysis pressures. In fistulas, recirculation may be measured using a non–ureabased dilutional method. Abnormal values include a flow rate less than 600 mL/ min in grafts and less than 400 to 500 mL/min in fistulas, a venous static pressure ratio greater than 0.5 in grafts or in fistulas, or an arterial segment static pressure ratio greater than 0.75 in grafts. One should not necessarily respond to a single abnormal value, but if a trend is noted, the patient should be referred for diagnosis and treatment.

CONTRAINDICATIONS There are no contraindications to dialysis access surveillance. Angiographic diagnosis and treatment of the dialysis access may be contraindicated in some patients and would

CHAPTER 114 SURVEILLANCE OF HEMODIALYSIS ACCESS 841 TABLE 114-1. Methods for Measuring Flow in Dialysis Access • Duplex Doppler ultrasonography (quantitative color velocity imaging) • Magnetic resonance angiography • Variable flow Doppler ultrasonography • Ultrasound dilution • Crit-Line III (optodilution by ultrafiltration) • Crit-Line II (direct transcutaneous) • Glucose pump infusion technique • Urea dilution • Differential conductivity • In-line dialysance Based on data from NKF KDOQI Guidelines. Available at http:// www.kidney.org/professionals/kdoqi/guidelines.cfm.

include those associated with contrast media, primarily severe allergic reactions. Other contraindications would be systemic infections, particularly in the case of a thrombosed access where the thrombus may be infected. Early failure is usually due to a technical error in creation of the fistula/graft. Rarely, no underlying anatomic lesion can be identified. These patients may have had excessive postdialysis compression, hypotension, hypovolemia, compression of the graft due to sleeping position, or a hypercoagulable state, any of which may lead to thrombosis.7

EQUIPMENT Surveillance can be performed in a number of ways (see Table 114-1). Flow decreases by less than 20% until the stenosis is 40% to 50%, and then the decrease in flow is rapid as the degree of stenosis increases to 80%.2 Decreasing flows can be measured by direct flow measurements, most commonly using ultrasound dilution. Graft function can also be evaluated using venous pressure measurements that will slowly increase with the development of stenoses. Recirculation is a very late manifestation of stenosis and a poor predictor of imminent thrombosis, so it is no longer used as a measurement of graft dysfunction in the KDOQI guidelines.2 The KDOQI Venous Access Working Group strongly encourages use of physical examination for monitoring dialysis fistulas and grafts.2 There is a concern that the basic skills of inspection, palpation, and auscultation have been largely abandoned in favor of technology, and the Working Group believes these skills should be taught to all individuals who perform hemodialysis procedures. Physical findings of persistent swelling of the arm, presence of collateral veins, prolonged bleeding after needle withdrawal, or altered characteristics of pulse or thrill in the access suggest the possibility of access dysfunction. Flow through the access can be evaluated by assessing the thrill (vibration) at various points along the access.8 When there is a stenosis present, a “water-hammer” pulse is felt below the stenosis, and above the stenosis the pulse goes away abruptly.6 If accessory veins have developed, they may frequently be visible, and the pulse in the main outflow vein decreases above the level of the side branch.6 Auscultation of the graft, particularly at the region of the venous anastomosis, can be very informative. When a stenosis is present, there is often a high-pitched, harsh, or discontinuous bruit at the affected

site. Because the velocity of flow increases in an area of stenosis, there will be a localized bruit or localized increase in the pitch of the bruit.6 Marked arm swelling usually indicates the presence of a central venous stenosis. Multiple dilated veins and collaterals are usually apparent in the patient’s upper arm and chest, and swelling of the breast may also occur with central venous stenosis.6 If an abnormality is found on physical examination, about 90% of patients will be found to have an angiographically significant abnormality.6,9,10 Duplex Doppler ultrasonography can be used to measure the diameter of the access. The method is operator dependent and subject to error caused by variation in crosssectional area and the angle of insonation.2 Duplex Doppler ultrasonography flow measurements do not have good sensitivity and specificity. Duplex ultrasound is much more accurate in performing anatomic assessment and direct evidence for the presence, location, and severity of access stenosis. It is particularly useful to determine reasons for maturation failure of fistulas. Because many fistulas cannot be studied using other surveillance techniques, routine duplex Doppler surveillance of primary fistulas should be considered to identify and refer for correction flow-limiting stenoses that may compromise the long-term patency and use of the fistula. Magnetic resonance angiography can also be used to evaluate the dialysis access noninvasively and allows anatomic assessment. Its high cost limits its use except in research settings. A number of indirect methods use some type of dilution technique. The major techniques are ultrasound dilution, timed ultrafiltration method, transcutaneous access flow rate, glucose infusion, differential conductivity, and ionic dialysance.2 Ultrasound dilution is probably the most common method of measuring access flow. It determines flow by measuring recirculation induced by reversing the blood lines during dialysis. The method detects dilution of blood in the arterial line after injecting saline into the venous line. Glucose pump infusion technology is another way to measure flow. In this method, a glucose solution is infused directly into the access, and the glucose level in a downstream blood sample is compared with a baseline sample.11 This measures flow with good accuracy when compared with ultrasound dilution measurement.11 Access flow is useful for identifying inflow stenosis.10 However, variations in access flow during dialysis can result from changes in cardiac output, mean arterial pressure, and blood volume.2 Access flow can increase by up to 11% or decrease by up to 30% from initial values by the end of dialysis, potentially impairing the ability of flow to predict impending vascular access failure.2 A single abnormal value is thus less valuable than demonstrating a trend over time.10 When access flow is measured repeatedly, trends of decreasing flow add predictive power for detection of access stenosis or thrombosis.2 Access pressures have also been used to evaluate dialysis access function. The pressure required to infuse blood back into the access is recorded as a venous pressure. When a stenosis develops, the pressure increases and the flow decreases. One of the components of the venous pressure is the “static pressure”; when static pressure increases to greater than 50% of the mean arterial pressure, the graft flow commonly has decreased into the thrombosis-prone

842 PART 1 VASCULAR INTERVENTIONS · SECTION 15 HEMODIALYSIS ACCESS MANAGEMENT range of 600 to 800 mL/min, and the presence of a stenosis is likely.2 Static venous pressure is a better tool for outflow lesions.10 A mean venous access pressure ratio can be calculated by simultaneous measurement of the static venous access pressure divided by the mean arterial pressure measurement.12 Although the venous pressure measurement may work relatively well for a graft, in a fistula there may be collateral veins decompressing the fistula, making venous pressures less valuable as a surveillance tool.2,10 There are situations in which the venous pressure is not accurate, including arterial stenoses or stenosis between the area used for the arterial needle and the area used for the venous needle. In both instances, the venous pressure may be normal or even decrease despite increasing stenosis and decreased flows.2 Recirculation, the return of dialyzed blood to the dialyzer, had been used as a measure of dialysis access function, but as noted earlier is no longer recommended by the KDOQI guidelines. It can be a sign of low access blood flow and a marker for stenosis,2 but it is not recommended as a surveillance test in grafts. This technique may be used in fistulas, although it tends to be a late finding and should have a minor role.2

TECHNIQUE Anatomy and Approaches In grafts, stenoses tend to develop in the venous outflow, usually within a few centimeters of the graft-vein anastomosis; they are due to neointimal hyperplasia. The cause of intimal hyperplasia is not clear but may be a response to turbulent blood flow, shear stress, endothelial dysfunction, vibratory effects from the placement of the graft,13 compliance mismatch between the graft material and the vein,7 or angulation and stretching of the vein.14 Neointimal hyperplasia includes myointimal proliferation, matrix deposition, infiltration with macrophages, and neovascularization.15 Venous stenoses increase the amount of recirculation in the graft, increase pressure throughout the graft, and increase the likelihood of thrombosis. In fistulas, stenoses tend to occur just distal to the arterial anastomosis or in the puncture zone of the vein. There may be ischemic effects16 as well as injury to the vein resulting from recurrent cannulation and subsequent fibrosis.2 Stenoses in the fistula have variable effects depending on the anatomy. A stenosis just distal to the arterial anastomosis will cause the fistula to appear flat because blood flow is not going through the outflow vein. A stenosis farther into the outflow vein may increase the pressure in the more proximal vein, or if there are collateral veins, these veins may decompress the fistula and there will be the development of large collateral veins.6 If side branches are occluded, a venous stenosis will give rise to a situation very similar to a graft, with increased recirculation, increased pressure, and increased risk of thrombosis. Stenoses can be treated with either angioplasty or surgical revision but are likely to recur, particularly in dialysis grafts, because the conditions that caused the original stenoses are unchanged. These lesions are the most common cause for dialysis graft thrombosis, and the long-term durability of any procedure to improve venous outflow is

constrained by development of such stenoses. Fistula stenoses seem to have a better outcome, although it may take several procedures to get the vein to a suitable caliber.

Technical Aspects Monitoring (physical examination) and surveillance of the fistula or graft should occur at least monthly. Flow assessment should be performed during the first 1.5 hours of treatment to eliminate error caused by decrease in cardiac output or blood pressure related to the ultrafiltration/ hypotension. Venographic evaluation of a failing dialysis access can be easily performed either before or after dialysis. If performed after dialysis, the dialysis needles can be left in place and a venogram performed. If a stenosis is discovered at the time of the venogram, angioplasty can be performed. Angioplasty should improve the flow through the graft by at least 20%.

CONTROVERSIES There continues to be controversy regarding the benefit of vascular access surveillance.15 The KDOQI guidelines recommend diagnostic fistulography if the intragraft blood flow is less than 600 mL/min in a graft, or when the fistula flow has decreased by over 25% (≥33% to avoid potential hemodynamic variation).17 Fistula referral is recommended when flow is less than 400 to 500 mL/min. It seems there is a relatively good correlation between intragraft blood flow of less than 600 mL/min and development of at least one significant (>50%) stenosis.18 It is not clear that early detection and treatment of stenoses leads to an improvement in graft patency or longevity.19 However, one study that evaluated whether prophylactic percutaneous transluminal angioplasty (PTA) of stenosis improved survival in native virgin radiocephalic forearm fistulas suggested that improved fistula survival is the result of increased access flow.2 PTA was associated with a significant decrease in access-related morbidity, halving the risk for hospitalization, central venous catheterization, and thrombectomy.2 Many of the studies of surveillance have been nonrandomized, and these studies have shown significant improvement. However, randomized trials show thrombosis rates that are not significantly different between groups.17 Flow and venous pressure measurements should be combined with clinical monitoring and demonstration of a trend of decrease flow.17

OUTCOMES Stenotic lesions are detected more often by using flow rates of less than 650 mL/min or a 25% to 33% decrease in flow. Early correction of venous stenoses is believed to reduce thrombosis rates, and some studies show a 50% to 90% decrease in thrombosis rate can occur when monitoring is instituted.12,20,21 Studies also indicate that access viability can be prolonged with surveillance programs.21,22 Surveillance tends not to decrease the number of angioplasties, but shifts toward a greater proportion of elective PTA vs thrombolysis/thrombectomy with PTA.12 However, some studies indicate that the rates of graft loss, times to graft loss, and overall thrombosis rates did not differ between

CHAPTER 114 SURVEILLANCE OF HEMODIALYSIS ACCESS 843 patients undergoing elective PTA or those in the control group.2 A limitation of studies trying to evaluate the interventions in dialysis grafts is the small sample size. Graft survival studies require a sample size of approximately 700 patients to detect an increase in graft survival of 1 year or a 33% difference in survival by 3 years.

KEY POINTS •

Dialysis access is a “lifeline” for many patients with renal failure.

•

Vascular access continues to be a leading cause for hospitalization and morbidity in patients with chronic kidney disease.

•

Failure of dialysis grafts and, to a lesser degree, dialysis fistulas is not uncommon and may be prevented by a program of regular surveillance.

•

There are a number of surveillance methods that can be used; physical examination can be very useful in evaluating dialysis access for problems.

•

When a dialysis access is discovered to be failing, venography and percutaneous treatment can prolong its survival and decrease the incidence of thrombosis.

COMPLICATIONS The primary complication of dialysis surveillance is more invasive angiographic study of dialysis grafts and fistulas resulting from a false-positive finding on surveillance. This is why a single measurement should not be used to initiate more invasive studies or treatment. The development of a surveillance abnormality should be correlated with other findings on physical examination and adequacy of hemodialysis. Any abnormality should be confirmed before further referral for duplex Doppler ultrasonography (for stenosis characterization) or angiography.2

POSTPROCEDURAL AND FOLLOW-UP CARE After discovery of an abnormality in a fistula or graft and subsequent treatment, the patient should be placed back into a surveillance program.

▶ SUGGESTED READINGS National Kidney Foundation Disease Outcomes Quality Initiative (NKF KDOQI) Guidelines. Available at http://www.kidney.org/professionals/ kdoqi/guidelines.cfm. U.S. Renal Data System. USRDS 2009 Annual Data Report: Atlas of End-Stage Renal Disease in the United States. Available at http:// www.usrds.org.

The complete reference list is available online at www.expertconsult.com.

CHAPTER 114 SURVEILLANCE OF HEMODIALYSIS ACCESS843.e1 REFERENCES 1. U.S. Renal Data System. USRDS 2011 Annual Data Report: Atlas of Chronic Kidney Disease and End-Stage Renal Disease in the United States, National Institutes of Health, National Institute of Diabetes and Digestive and Kidney Diseases, Bethesda, MD, 2011. 2. KDOQI Clinical Practice Guidelines and Clinical Practice Recommendations for 2006 Updates: Hemodialysis Adequacy, Peritoneal Dialysis Adequacy and Vascular Access. Am J Kidney Dis 2006;48(Suppl 1):S1–S322. 3. Rosas SE, Feldman HI. Synthetic vascular hemodialysis access versus native arteriovenous fistula: a cost-utility analysis. Ann Surg, 2012; 255(1):181–6. 4. Tordoir JH, Bode AS, Peppelenbosch N, et al. Surgical or endovascular repair of thrombosed dialysis vascular access: is there any evidence? J Vasc Surg, 2009;50(4):953–6. 5. Hemodialysis Workgroup 2006: Hemodialysis Adequacy 2006. Am J Kidney Dis 2006;48(Suppl 1):S8–11. 6. Beathard GA. Physical examination of the dialysis vascular access. Semin Dial 1998;11(4):231–6. 7. Windus DW. Permanent vascular access: a nephrologist’s view. Am J Kidney Dis 1993;21(5):457–71. 8. Gelbfish GA. Surgical versus percutaneous care of arteriovenous access. Semin Vasc Surg 2007;20(3):167–74. 9. Safa AA, Valji K, Roberts AC, et al. Detection and treatment of dysfunctional hemodialysis access grafts: effect of a surveillance program on graft patency and the incidence of thrombosis. Radiology 1996;199(3):653–7. 10. Kumbar L, Karim J, Besarab A. Surveillance and monitoring of dialysis access. Int J Nephrol 2012;2012:649–735. 11. Ram SJ, Magnasco A, Jones SA, et al. In vivo validation of glucose pump test for measurement of hemodialysis access flow. Am J Kidney Dis 2003;42(4):752–60.

12. Zasuwa G, Frinak S, Besarab A, et al. Automated intravascular access pressure surveillance reduces thrombosis rates. Semin Dial 2010; 23(5):527–35. 13. Fillinger MF, Reinitz ER, Schwartz RA, et al. Graft geometry and venous intimal-medial hyperplasia in arteriovenous loop grafts. J Vasc Surg 1990;11(4):556–66. 14. Malchesky PS, Koshino I, Pennza P, et al. Analysis of the segmental venous stenosis in blood access. Trans Am Soc Artif Intern Organs 1975;21:310–9. 15. Paulson WD, Moist L, Lok CE. Vascular access surveillance: an ongoing controversy. Kidney Int 2012;81(2):132–42. 16. Badero OJ, Salifu MO, Wasse H, et al. Frequency of swing-segment stenosis in referred dialysis patients with angiographically documented lesions. Am J Kidney Dis 2008;51(1):93–8. 17. Paulson WD, Work J. Controversial vascular access surveillance mandate. Semin Dial 2010;23(1):92–4. 18. Amin MZ, Vesely TM, Pilgram T. Correlation of intragraft blood flow with characteristics of stenoses found during diagnostic fistulography. J Vasc Interv Radiol 2004;15(6):589–93. 19. Moist LM, Churchill DN, House AA, et al. Regular monitoring of access flow compared with monitoring of venous pressure fails to improve graft survival. J Am Soc Nephrol 2003;14(10):2645–53. 20. Besarab A, Lubkowski T, Frinak S, et al. Detecting vascular access dysfunction. ASAIO J 1997;43(5):M539–43. 21. McCarley P, Wingard RL, Shyr Y, et al. Vascular access blood flow monitoring reduces access morbidity and costs. Kidney Int 2001; 60(3):1164–72. 22. Cayco AV, Abu-Alfa AK, Mahnensmith RL, et al. Reduction in arteriovenous graft impairment: results of a vascular access surveillance protocol. Am J Kidney Dis 1998;32(2):302–8.

CHAPTER

115

Management of Failing Hemodialysis Access Charles T. Burke

The natural history of a vascular access graft or fistula is progressive development of neointimal hyperplastic stenoses that reduces blood flow and compromises performance of the vascular access. If left untreated, these lesions will eventually lead to thrombosis of the vascular access. Early detection and treatment of hemodynamically significant stenosis is a primary tenet of a vascular access management program. Angioplasty remains the most important technique for treating neointimal hyperplastic stenoses associated with hemodialysis grafts and fistulas.

INDICATIONS The criteria that define failing vascular access and the indications for percutaneous or surgical repair are described in the National Kidney Foundation (NKF) Clinical Practice Guidelines for Vascular Access.1 These criteria and indications have been incorporated into the Society of Interventional Radiology (SIR) quality improvement guidelines, the American College of Radiology (ACR)-SIR practice guidelines, and the SIR document on standards of reporting.1-3 Primary indications for percutaneous intervention of either polytetrafluoroethylene (PTFE) grafts or arteriovenous fistulas (AVFs) include (1) documentation of a clinical or hemodynamic abnormality and (2) a stenosis causing over 50% reduction in luminal diameter. Intervention is not indicated unless the clinical and angiographic findings are both abnormal and correlative. Clinical and hemodynamic abnormalities indicative of significant stenosis include elevated venous pressure, decreased intra-access blood flow, prolonged bleeding after needle removal, and arm swelling. Several clinical studies have reported that periodic assessment of intra-access blood flow may be the most reliable method for early detection of a developing stenosis.4-6 Intra-access blood flow is measured during hemodialysis, and if intra-access blood flow is less than 600 mL/min, a fistulogram is recommended to evaluate the entire vascular access circuit.1 Over time, there may be degeneration of graft material, resulting in overlying pseudoaneurysms. This is often due to fragmentation of the graft material from repeated cannulation and increased intragraft pressure from venous outflow stenosis.7 Treatment may be required when the pseudoaneurysm interferes with graft access or there are signs of impending rupture. Indicators for possible graft rupture include skin breakdown over the pseudoaneurysm or rapid pseudoaneurysm expansion. In severe cases or in the setting of graft infection, pseudoaneurysms are treated surgically. In patients with limited surgical options and in whom the anatomy is suitable, endovascular management may be considered. Early evaluation and intervention may be useful for salvage of nonmaturing fistulas.8 Recent reports 844

demonstrate that serial dilation of small-diameter veins (balloon maturation) may expedite the maturation process and provide functionality to fistulas that may otherwise have failed.9-11 Although this indication for percutaneous intervention is not yet fully defined, AVFs that have failed to mature within 3 months of creation should undergo fistulography to determine whether angioplasty or surgical revision would be beneficial. The primary reasons for a fistula that fails to mature include vascular stenosis (outflow vein or arteriovenous anastomosis), competitive outflow veins, or a deep outflow vein that is nonpalpable. Management of vascular stenoses continues to be balloon angioplasty. Small competitive outflow veins can be managed by surgical ligation or endovascular embolization. Deep nonpalpable outflow veins require surgical revision. Use of stents or stent-grafts for management of failing hemodialysis grafts and fistulas remains controversial. Although angioplasty remains the primary technique for treating neointimal hyperplastic stenosis, stent placement may be indicated in specific situations. The Kidney Disease Outcomes Quality Initiative (KDOQI) guidelines allow for stent use to treat central venous stenoses when there is (1) acute elastic recoil of the vein and over 50% residual stenosis after angioplasty and (2) stenosis that recurs within 3 months after previously successful angioplasty.1 When treating an aggressive recurring stenosis in a large central vein, a vascular stent may provide better results than angioplasty alone. Another acceptable use of stents is for acute repair of angioplasty-induced venous rupture. Although minor venous injuries can be effectively managed by prolonged inflation of an angioplasty balloon, more substantial injuries with continuing perivascular hemorrhage may require a stent or stent-graft to provide a more durable repair.12,13

CONTRAINDICATIONS The only absolute contraindication to percutaneous intervention is an infected vascular access. Manipulation of an infected graft or fistula may cause intravascular dissemination of microorganisms and thereby elicit acute sepsis or long-term infectious complications such as endocarditis or osteomyelitis. Superficial cellulitis may resolve with appropriate antibiotic treatment, but more extensive infection of a graft or fistula often requires surgical management. Another important contraindication to percutaneous intervention is allergy to a contrast agent. Patients with a known or suspected allergy should be pretreated according to standardized protocols before fistulography or any interventional procedure requiring use of intravascular contrast material. Alternatively, a noniodinated contrast agent (e.g., CO2) may be used.14 Relative contraindications include intervention on a newly placed graft or fistula; dilation of a new (95% of failures) and due to recoil of elastic fibrotic tissue at anastomosis. (From Saad WEA. Percutaneous management of postoperative anastomotic biliary strictures. Tech Vasc Interv Radiol 2008;11:143–53.)

CBD

CBD

CDS

A

B

FIGURE 135-7. Cutting balloon dilation of a hepaticocholedochostomy anastomosis in an adult right hepatic lobe living related transplant recipient. A, Fluoroscopic image of critical stenosis at bile duct–to–bile duct anastomosis (arrowheads). CBD, Common bile duct; CDS, cystic duct stump. B, Fluoroscopic image after dilation. Patient underwent 7-mm cutting balloon dilation followed by 7-mm conventional balloon dilation. Anastomotic stricture has resolved (arrowheads). CBD, Common bile duct of recipient. (From Saad WEA. Percutaneous management of postoperative anastomotic biliary strictures. Tech Vasc Interv Radiol 2008;11:143–53.)

an intent-to-treat analysis, but only evaluated patency for technically successful cases and cases with “adequate” follow-up.4,13 Amalgamation of lesion types, etiology, and locations also adds to the confusion. Overall, the 1- to 3-year intent-to-treat patency ranges from 38% to 73%, based on cholangiography and/or clinical follow-up.2,4,13,18,22-25 The 5- to 6-year patency of technically successful procedures by clinical/laboratory follow-up is 52% to

66%.4,13 The only study comparing the patency of ductto-duct anastomoses with biliary-enteric (hepaticojejunostomies) was by Saad et al., who showed there was no statistical difference (P = 0.1) between the 1-year primary unassisted patency (by cholangiography) of either method: 43% and 48%, respectively.18 This study also showed that liver transplants with patent hepatic arteries had far better 1-year patency results compared to liver transplants with

1000 PART 2 NONVASCULAR INTERVENTIONS · SECTION 19 THE BILIARY TRACT

A

B

FIGURE 135-8. Cutting balloon dilation of a choledochocholedochostomy anastomosis in an adult cadaveric liver transplant recipient. A, Fluoroscopic image of significant stenosis at bile duct–to–bile duct anastomosis (arrows). B, Fluoroscopic image after dilation. Patient underwent 8-mm cutting balloon dilation followed by 10- and 12-mm conventional balloon dilation. Anastomotic stricture has resolved (arrows). (From Saad WEA. Percutaneous management of postoperative anastomotic biliary strictures. Tech Vasc Interv Radiol 2008;11:143–53.)

stenosed or thrombosed hepatic arteries: 55% versus 0%, respectively.18

endoscopic means. Transhepatically, the WallFlex can be deployed through a 9F sheath. Both the VIABIL and WallFlex come in fully covered and partly covered stent-grafts. Essentially, retrievable stent-graft placement belongs to the school of prolonged dilation or splintering (see earlier) in which the stent dwells across the benign central/ anastomotic biliary stricture, stretching it for months and remodeling the biliary stricture during this dwell period.2 The issue is how long to keep the covered stent in place to allow remodeling and prevent restenosis. Theoretically, if the stent-graft is removed before a certain dwell time (time x) the stricture will restenose as if it was a dilation, and the only time “bought” would be the dwell time of the stent (Fig. 135-10).2 However, beyond a certain dwell time (time x) where enough remodeling has occurred, the actual patency of the benign biliary stricture will be improved (stay open longer). This particular dwell time (time x), if this hypothesis is correct, is still to be determined and is the subject of current investigation, primarily in Europe.2

Retrievable Stents for Central Benign Biliary Strictures

COMPLICATIONS

The experience of placing and removing retrievable covered stents (removable stent-grafts) from a percutaneous transhepatic approach is limited compared to endoscopically placed stent-grafts.2,19 Nevertheless, in the United States, the two most common commercially available stent-grafts used for subsequent transhepatic removals are the most commonly placed and removed stents from an endoscopic approach. These are the VIABIL (W.L. Gore) and the WallFlex (Boston Scientific) stent-grafts. The VIABIL is has expanded polytetrafluoroethylene (e-PTFE) covering a nitinol frame, and the WallFlex has a platinum-cored nitinol (Platinol) braid design covered on the inside with silicone (Permalume). The WallFlex has a retrievable loop (“bucket handle” [author’s term]) only on the bowel side (caudad), not on the transhepatic side (cephalad) (Fig. 135-9). In other words, it is currently designed primarily to be removed by

The overall risk of major and minor complications is 4% to 12%.2-4,13,18 Major complications can be classified into traumatic complications or postprocedural infectious complications (cholangitis with or without sepsis). Hemobilia not requiring blood transfusion is considered a minor complication and has been described in up to 10% of cases.3 Major traumatic complications occur in 0% to 5.6% of cases (see Table 135-1),2-4,13,18 and these include anastomotic rupture/ biliary leak (0%-1.9%), hemobilia requiring transfusion (0%1.9%), and pseudoaneurysm formation (0%-0.5%).3,4,18 Postdilation cholangitis with or without sepsis is actually rarely mentioned.2-4,13,18 This author does not believe transhepatic balloon dilation of benign biliary strictures adds to the risk of cholangitis or sepsis after a routine biliary drain cholangiogram and biliary drain exchange, which has a cholangitis rate and a sepsis rate of 2.1% and 0.4%, respectively.21

FIGURE 135-9. Photograph of partly deployed WallFlex stent (Boston Scientific, Natick, Mass.). “Bucket handle” (retrievable loop) can be seen at tip of delivery platform from percutaneous transhepatic approach. (Image courtesy Boston Scientific; © 2012 Boston Scientific Corporation or its affiliates. All rights reserved.)

CHAPTER 135 MANAGEMENT OF BENIGN BILIARY STRICTURES 1001 !tx

"tx

Stent removal

100

80

Patency

Patency

80 50 40 20 0

A

Stent removal

100

50 40 20

3

6

12

24 t months

0

B

3

6

12

24 t months

FIGURE 135-10. Graphs depicting hypothetical response of biliary anastomotic patency after stent removal, depending on dwell time of covered stents in biliary tract. A, Graph depicting hypothetical response of biliary anastomotic patency after covered stents have dwelled in biliary tract for less than time x (tx). Hypothetically, a particular dwell time for covered stent is required to achieve desired effect of improved patency after stent removal. Premature stent removal (tx) may lead to improved patency. This particular dwell time, if the theory is correct, is to be determined and is the subject of current investigation. (From Saad WEA. Percutaneous management of postoperative anastomotic biliary strictures. Tech Vasc Interv Radiol 2008;11:143–53.)

Covered stents (stent-grafts) have two main complications: migration (up to 25% of cases) and potential occlusion.

POSTPROCEDURAL AND FOLLOW-UP CARE After each over-the-wire biliary dilation session, the balloon is removed and replaced with an internal-external PTBD.24,13,18 Removal of the catheter depends on the “school of management” (see earlier). If prolonged splintering of the stricture is the preferred approach, the drain is removed 4 to 12 months after the dilation. If fast-track balloon dilation is preferred (and if there is residual stenosis < 20%-30%), the transhepatic biliary drain is externalized (pigtail placed above the level of the target stricture) and capped for up to 1 to 2 weeks. If the patient tolerates capping and the cholangiogram continues to show adequate biliary patency, the external biliary drain is removed. KEY POINTS •

Benign biliary strictures are difficult to manage, probably because of the cicatricial fibrotic nature of the lesions.

•

There are numerous variables in etiology, pathology, locations, and management techniques for benign biliary strictures.

•

Two schools/approaches exist: (1) chronic percutaneous biliary drainage ± balloon dilations and (2) sequential balloon dilation.

•

Aggressive balloon dilation (e.g., cutting balloons) have improved technical results.

•

The 1- to 3-year patency of transhepatic balloon dilation of benign biliary strictures ranges 33% to 67%.

•

Removable stent-grafts may be a better management option, but well-designed studies are needed to verify this assertion.

▶ SUGGESTED READINGS Cantwell CP, Pena CS, Gervais DA, et al. Thirty years’ experience with balloon dilation of benign postoperative biliary strictures: Long-term outcomes. Radiology 2008;249:1050–7. Saad WE, Saad NE, Davies MG, et al. Transhepatic balloon dilation of anastomotic biliary strictures in liver transplant recipients: The significance of a patent hepatic artery. J Vasc Interv Radiol 2005;16: 1221–8. Saad WE. Percutaneous management of postoperative anastomotic biliary strictures. Tech Vasc Interv Radiol 2008;11:143–53.

The complete reference list is available online at www.expertconsult.com.

CHAPTER 135 MANAGEMENT OF BENIGN BILIARY STRICTURES1001.e1 REFERENCES 1. Lubienski A, Deux M, Lubienski K, et al. Interventions for benign biliary strictures. Radiology 2005;45:1012–9. 2. Saad WE. Percutaneous management of postoperative anastomotic biliary strictures. Tech Vasc Interv Radiol 2008;11:143–53. 3. Saad WE, Davies MG, Saad NE, et al. Transhepatic dilation of anastomotic biliary strictures in liver transplant recipients with use of a combined cutting and conventional balloon protocol: Technical safety and efficacy. J Vasc Interv Radiol 2006;17:837–43. 4. Cantwell CP, Pena CS, Gervais DA, et al. Thirty years’ experience with balloon dilation of benign postoperative biliary strictures: Long-term outcomes. Radiology 2008;249:1050–7. 5. Connor S, Garden OJ. Bile duct injury in the era of laparoscopic cholecystectomy. Br J Surg 2006;93:158–68. 6. Lillemoe KD, Pitt HA, Cameron JL. Current management of benign bile duct strictures. Adv Surg 1992;25:119–74. 7. Rossi M, Salvatori FM, Giglio I, et al. Interventional radiology techniques in the treatment of complications due to videolaparoscopic cholecystectomy. J Radiol Med (Torino) 2002;103:384–95. 8. Rossi M, Salvatori FM, Ingianna D, et al. Non-vascular interventional radiology in the treatment of post-liver transplant complications: The clinico-radiological correlations and technical considerations. J Radiol Med (Torino) 1995;90:291–7. 9. Zoepf T, Maldonado-Lopez EJ, Hilgard P, et al. Balloon dilation vs. balloon dilation plus bile duct endoprosthesis for treatment of anastomotic biliary strictures after liver transplantation. Liver Transpl 2006;12:88–94. 10. Cameron AM, Busuttil RW. Ischemic cholangiopathy after liver transplantation. Hepatobiliary Pancreat Dis Int 2005;4:495–501. 11. Park JS, Kim MH, Lee SK, et al. Efficacy of endoscopic and percutaneous treatments for biliary complications after cadaveric and living donor liver transplantation. Gastrointest Endosc 2003;57:78–85. 12. Mondragon RS, Karani JB, Heaton ND, et al. The use of percutaneous transluminal angioplasty in hepatic artery stenosis after transplantation. Transplantation 1994;57:228–31. 13. Zajko AB, Sheng R, Zetti GM, et al. Transhepatic balloon dilation of biliary strictures in liver transplant patients: A 10-year experience. J Vasc Interv Radiol 1995;6:79–83.

14. Pitt HA, Kaufman SL, Coleman J, et al. Benign postoperative biliary strictures: Operate or dilate? Ann Surg 1989;210:417–25. 15. Rothlin MA, Lopfe M, Schlumpf R, Largiader F. Long-term outcome of hepaticojejunostomy for benign lesions of the bile ducts. Am J Surg 1998;175:22–6. 16. Lilemoe KD, Martin SA, Cameron JL, et al. Major bile duct injuries during laparoscopic cholecystectomy: Follow-up after combined surgical and radiologic management. Ann Surg 1997;225:459–71. 17. Pitt HA, Miyamoto T, Parapatis SK, et al. Factors influencing outcome in patients with postoperative biliary strictures. Am J Surg 1982; 144:14–21. 18. Saad WE, Saad NE, Davies MG, et al. Transhepatic balloon dilation of anastomotic biliary strictures in liver transplant recipients: The significance of a patent hepatic artery. J Vasc Interv Radiol 2005;16:1221–8. 19. Petersen BD, Timmermans HA, Uchida BT, et al. Treatment of refractory benign biliary stenoses in liver transplant patients by placement and retrieval of a temporary stent-graft: Work in progress J Vasc Interv Radiol 2000;11:919–29. 20. Saad WEA, Wallace MJ, Wojak JC, et al. Quality improvement guidelines for percutaneous transhepatic cholangiography, biliary drainage, and percutaneous cholecystostomy. For the Standards of Practice Committee of the Society of Interventional Radiology. J Vasc Interv Radiol 2010;21:789–95. 21. Ginat D, Saad WE, Davies MG, et al. Incidence of cholangitis and sepsis associated with percutaneous biliary drain cholangiography and exchange: A comparison between liver transplant and native liver patients. AJR Am J Roentgenol 2011;196: W73–77. 22. Mueller PR, vanSonnenberg E, Ferrucci JT, et al. Biliary stricture dilation: Multicenter review of clinical management in 73 patients. Radiology 1986;160:17–22. 23. Williams HJ, Bender CE, May GR. Benign postoperative biliary strictures: Dilation with fluoroscopic guidance. Radiology 1987;163: 629–34. 24. McDonald V, Matalon TAS, Patel SK, et al. Biliary strictures in hepatic transplantation. J Vasc Interv Radiol 1991;2:533–8. 25. Ward EM, Kiely MJ, Maus TP, et al. Hilar biliary strictures after liver transplantation: Cholangiography and percutaneous treatment. Radiology 1990;177:259–63.

CHAPTER

136

Management of Biliary Leaks Rupert H. Portugaller and Klaus A. Hausegger

Biliary leakage is defined as diversion of bile via a defect in the ductal wall leading to the formation of bilomas, fistulas, or free spillage into the peritoneal cavity. It is a common complication after liver and biliary surgery. With emerging laparoscopic techniques and transplantation, the rate of bile leakage is reported to range from 0.8% to 12%.1-4 Massive hepatic trauma with subsequent biliary leakage has also been reported.5 Leaks can originate from biliary-enteric anastomoses, choledochocholedochostomy, bile or cystic duct stumps, cystohepatic ducts, blind-ending aberrant bile ducts (ducts of Luschka),6 intraoperative bile duct injury, or ablation of the surface of the liver, or they may be due to loose or dislocated T-tube drainage.7 Biliary anastomotic or T-tube exit site leaks attract early attention, whereas symptoms from bile duct necrosis occur relatively late3 (Fig. 136-1). Small leaks can be managed conservatively with antibiotics and maintenance of perioperatively placed drains,8 but massive biliary extravasation can result in bilomas, fistulas, or biliary peritonitis, with subsequent abscess formation and sepsis. Patients usually have localized or diffuse abdominal pain and tenderness and fever and chills. Jaundice is indicative of concomitant biliary strictures. Leukocytosis and elevated C-reactive protein levels are signs of active inflammation. Older patients, however, may lack typical symptoms. Because biliary leakage can provoke a lifethreatening condition, early diagnosis and therapy are crucial. Intravenous administration of broad-spectrum antibiotics is mandatory. When bile or blood cultures are available, specific antibiotic therapy is initiated. Bilomas should be diverted percutaneously under ultrasound or computed tomography (CT) guidance or with perioperatively placed drains. Irrigation and low-pressure suction prevent stasis of fluid in a closed space and thus decrease the likelihood of infection.9,10

INDICATIONS • Major biliary leakage unlikely to resolve spontaneously (ongoing abdominal drainage of a significant amount of bilious fluid) • Surgical therapy not necessary (definitive healing of the leakage expected after endoluminal decompression alone) • Surgical therapy impossible because of extensive tissue adhesions (previous operations) • Poor patient condition, with high anesthesiology risk • Endoscopic management not possible because of • Enteric Roux-en-Y anastomosis • Recent duodenotomy (air insufflation can cause rupture of the suture) 1002

• Inability to probe the duodenal papilla or the common/ hepatic bile duct • Inability to negotiate an additional biliary stricture retrogradely In patients with massive biliary peritonitis, surgical revision and peritoneal lavage are usually necessary. Bile duct transection or disruption requires surgical therapy as well. In contrast to the blind-ending ducts of Luschka, the cystohepatic ducts drain liver parenchyma. Thus, injuries to the cystohepatic ducts must be treated by either surgical biliary reconstruction or diversion to the jejunum.6 If surgical treatment is not necessary or impossible, endoluminal biliary drainage is indicated for persistent biliary leakage. The principle of healing a bile leak is to create a low-pressure system along the biliary tract by means of internal or external drainage. Endoscopic therapy with nasobiliary drainage, sphincterotomy, or placement of an endoprosthesis has been reported to result in a high closure rate of up to 82%.11-13 Because percutaneous transhepatic access is more invasive, endoscopic retrograde cholangiopancreatography (ERCP) is recommended as the first option. However, when the duodenum or the common bile duct cannot be cannulated, percutaneous transhepatic biliary drainage (PTBD) is the only endoluminal treatment option. PTBD serves to decompress the biliary system and redirect bile flow from the defect into the bile ducts. This may lead to sufficient healing of the leakage that additional surgery is not necessary (see Fig. 136-1). In critically ill patients, PTBD may allow the patient’s condition to improve before subsequent definitive surgery (Fig. 136-2).

CONTRAINDICATIONS • Coagulopathy that cannot be corrected. Because the risk of uncontrollable hemorrhage from the liver is high in patients with coagulopathy, every effort should be made to correct or improve coagulation status before the procedure.14 • History of allergies to contrast material. During hepatic puncture for PTBD, contrast material is likely to be injected into blood vessels before a bile duct is opacified. If a patient is known to be allergic to contrast agents, cortisone and antihistamine should be used prophylactically.

EQUIPMENT A modern C-arm unit with pulsed fluoroscopy is desirable because fluoroscopy times can be extensive when probing nondilated bile ducts. Apart from the anesthesiology

CHAPTER 136 MANAGEMENT OF BILIARY LEAKS 1003

A

B

D

C

E

F

FIGURE 136-1. This 29-year-old woman underwent surgical resection of retropancreatic teratoma. Intraoperative injury to portal vessels resulted in massive hemorrhage with subsequent multiorgan failure. Pancreas and spleen had to be resected, and a venous bypass was performed to bridge iatrogenic occlusion of celiac trunk. Eight months later, patient had jaundice. A, Endoscopic retrograde cholangiopancreatography (ERCP) demonstrated a complex central biliary lesion with leakage (arrowheads) that was probably due to bile duct ischemia. B, Percutaneous transhepatic puncture of right (arrowheads) and left (arrow) biliary tract was performed. C, Because occlusion of left hepatic duct could not be negotiated with a guidewire, balloon-targeted needle passage was successfully performed. D, Left hepatic duct was predilated with a balloon. E, Finally, biliary system was drained internally by three percutaneously implanted plastic stents. F, One year later, because ERCP demonstrated no further biliary leakage and patent bile ducts, plastic stents were removed endoscopically.

A

B

FIGURE 136-2. This 64-year-old woman with pancreatic carcinoma underwent duodenopancreatectomy (Whipple technique). Persistent bilious effluence from abdominal drain and signs of bilious peritonitis occurred. A, Abdominal computed tomography study revealed free fluid at porta hepatis (arrowheads). Because postoperative course was complicated by cardiac insufficiency, no repeated laparotomy was performed. B, Percutaneous transhepatic cholangiography on postoperative day 8 showed leakage at bilioenteric anastomosis (arrowheads), with free spillage of contrast material into abdominal cavity. Right and left hepatic ducts were narrowed at site of anastomosis, probably because of edema.

1004 PART 2 NONVASCULAR INTERVENTIONS · SECTION 19 THE BILIARY TRACT equipment, in our unit the following material is at the interventionist’s disposal: • 5- and 10-mL syringes; sterile swabs • Nonionic water-soluble contrast agent (300-370 mg iodine/mL) • Local anesthetic: 5 to 10 mL of 2% lidocaine; injection needles • Single-puncture technique: Neff Percutaneous Access Set (William Cook Europe, Bjaeverskov, Denmark): • Small-gauge stainless steel access needle with an inner stylet • 0.018-inch guidewire • Coaxial introducer system consisting of three parts: an outer sheath, an inner tapered introducer, and an innermost metal stiffening cannula • 0.035-inch hydrophilic Radiofocus guidewire M (Terumo Medical Corp., Tokyo, Japan) • 5F Angled Taper Glidecath (Terumo) • 0.035-inch Amplatz Extra Stiff guidewire (William Cook Europe) • 8.3F ring, Lunderquist biliary drainage catheter • 3-0 cutaneous suture • 10F and 12F Munich drainage catheters (Peter Pflugbeil GmbH, Zorneding, Germany): perforated bile duct drainage with a special flange (perforation distance, 7.5, 10, or 15 cm), connecting tube, sealing cap, and adapter • Bandage material

TECHNIQUE Anatomy and Approach Biliary leakage is suspected when persistent bilious-looking effluent from the abdominal drain is noted. Injection of contrast material through the drain may result in opacification of a bile duct.15 Before PTBD, contrast-enhanced CT or magnetic resonance imaging (MRI) yields important anatomic information. Bilomas or fistulas, as well as abscesses, are especially well visualized with magnetic resonance cholangiopancreatography (MRCP),16 and the site of the bile leak may be tentatively established. Though more investigator dependent, ultrasonography is an alternative preinterventional imaging method.8 Generally, PTBD for biliary leakage is performed in similar fashion as for biliary obstruction. However, in 88% to 100% of cases, the intrahepatic bile ducts are not dilated or only minimally dilated, and hence more difficult to cannulate.5,8,15,17 The percutaneous access site is chosen according to the estimated site of leakage, because the biliary drainage catheter should finally be able to cross the injured portion of the bile duct. A single-puncture technique is acceptable for nondilated bile ducts.16 This technique requires that the initial puncture be appropriate for definitive catheter placement (Fig. 136-3). After hepatic puncture, the access needle is slowly withdrawn while contrast material is gently injected. When a bile duct is opacified, the stylet of the needle is removed and a 0.018-inch guidewire is advanced into the biliary system. After removal of the cannula, the locked coaxial introducer system is pushed over the guidewire into the bile duct. The metal stiffener is unlocked from the catheter, and only the catheter with its dilator is advanced. After

the 0.018-inch guidewire, tapered dilator, and metal stiffener are pulled out, a 0.035-inch hydrophilic steerable guidewire is advanced into the bowel lumen. Once the intestine has been entered, the hydrophilic guidewire is exchanged for a 0.035-inch extrastiff guidewire (Amplatz type). After dilation of the hepatic puncture tract, an 8.3F self-retaining loop catheter (Lunderquist biliary drainage catheter) is introduced for external/internal biliary drainage. Catheter side holes must be located on both sides of the bile leak to keep intraluminal resistance as low as possible.

Technical Aspects Sometimes the initial puncture site reveals a bile duct too central for definitive drainage because of the presence of major vessels. In such cases, cholangiography can be performed over the puncture needle. Then, with the doublepuncture technique, a proper bile duct is targeted and punctured again with the access needle under fluoroscopic guidance. If biliary ascites is present, leakage of acidic fluid along the percutaneous puncture route often causes skin irritation. Additional drainage of peritoneal fluid helps minimize puncture tract extravasation.

CONTROVERSIES There is controversy about when to remove the biliary drainage catheter. Whereas some authors remove the catheter when cholangiography shows that the leak has healed without residual stenosis,15,17 others keep the tubes in place for a certain period after it has been proved that the leak has sealed.10 Because no randomized studies have compared early and late catheter removal, no definitive advice can be given for catheter dwell time. Ernst et al. encountered residual stenoses in 4 of 16 patients after PTBD. They suggested that use of a large drainage catheter (12F) and drainage for at least 7 to 8 weeks might prevent secondary biliary stenosis after closure of the leak.15

OUTCOMES Technical success of PTBD is defined as placement of the tube to provide continuous drainage of bile. The Society of Interventional Radiology has established a success threshold of 70% for percutaneous cannulation of nondilated bile ducts.14 In a series of 130 patients with nondilated bile ducts, a technical success rate of 90% has been reported.16 In case of biliary leakage, procedural success means either closure of a biliary leak with sole percutaneous biliary drainage or improvement of the patient’s condition such that definitive surgical repair is possible. Table 136-1 presents the outcomes reported by several authors.* For biliary leakage, PTBD has been reported to be effective in 50% to 100% of patients without additional surgery. The time from PTBD to closure of the leak ranged from 7 to 150 days (mean, 10-80 days), and the time to catheter removal ranged from 7 to 274 days. Reasons for extended catheter dwell time after closure of the leak were *References 5, 7, 8, 10, 12, 15, 17-20.

CHAPTER 136 MANAGEMENT OF BILIARY LEAKS 1005

A

B

C

D

FIGURE 136-3. This 88-year-old-woman with contained perforation of an inflamed gallbladder underwent open cholecystectomy. Operation was complicated by extensive inflammatory tissue alterations. A, Postoperatively, percutaneous single puncture of a right intrahepatic bile duct revealed biliary extravasation from cystic duct stump (arrowheads) and stricture in common hepatic/common bile duct (arrow). After probing biliary system with 0.018-inch guidewire (B), an 8.3F catheter was placed for internal/external drainage (C). D, Two weeks later, inflammatory markers had decreased, and only cystic duct stump without further bile extravasation was seen. A 12F Munich drainage catheter was placed and kept on internal drainage.

TABLE 136-1. Outcomes of Percutaneous Transhepatic Biliary Drainage for Biliary Leakage

Ernst et al.15

n

Technical Success

Leak Healing

Procedural Success

Closure Time (days)

Catheter Dwell Time (days)

78 (30-150)

n.s.

2 (12%)

38 (11-76)

Complications

Follow-up (months)

16

16 (100%)

13 (81%)

13 (81%)

5

12

12 (100%)

6 (50%)

6 (50%)

27 (9-61)

140 (61-274)

3 (25%)

n.s.

4

4 (100%)

2 (50%)

3 (75%)

14-21

n.s.

0 (0%)

n.s.

de Castro et al.8

11

11 (100%)

9 (90%)

9 (90%)

n.s.

n.s.

0 (0%)

n.s.

Vaccaro et al.

3

3 (100%)

3 (100%)

3 (100%)

25-75

46-70

0 (0%)

36

Tsukamoto et al.12

3

3 (100%)

2 (100%)

2 (100%)

14

14

0 (0%)

72

Chen et al.

2

2 (100%)

2 (100%)

2 (100%)

14

n.s.

0 (0%)

n.s.

Civelli et al.20

8

8 (100%)

5 (62%)

5 (62%)

80

n.s.

0 (0%)

18

Righi et al.

22

22 (100%)

21 (91%)

21 (91%)

10 (7-41)

15 (7-74)

0 (0%)

36

Stampfl et al.17

30

30 (100%)

22 (73%)

22 (73%)

55 (15-116)

55 (15-116)

2 (7%)

n.s.

Kaufman et al. 7

Zhang et al.

10

18

19

n.s., Not specified.

1006 PART 2 NONVASCULAR INTERVENTIONS · SECTION 19 THE BILIARY TRACT concomitant strictures or calculi that were also treated percutaneously. If PTBD fails to seal a biliary leak, bile duct necrosis, complete duct disruption, or extensive laceration is frequently responsible. Typical signs of procedural failures include persistent septic states and continuous outflow of bile-stained fluid from abdominal drains. Gwon et al. used retrievable covered stents (Song retrievable stent [Taewoong Medical, Kimpo, Korea]) in complex biliary leakages.21 Eleven patients with major bile leaks (>500 mL drained bile/24 h) or leaks with concomitant biliary strictures untreatable by endoscopic means or refractory bile leaks despite more than 10 days of external/ internal drainage were included in their study. In all patients, the covered stents could be introduced via the percutaneous transhepatic puncture tracts. Additional pigtail catheters were placed coaxially in the stents and remained in place to prevent stent migration and keep the percutaneous puncture tracts patent until stent retrieval. Via the PTBD tracts, all covered stents were removed successfully after a mean of 31 days (range, 14-64 days). In all patients, sealing of the bile leaks without biliary stenosis was proven by cholangiography. The PTBD catheters were extracted after a mean indwelling period of 41 days (range, 20-80 days). No recurrent biliary leakages were noted after a mean follow-up of 366 days (range, 215-730 days). Endoscopic stent placement has been reported to be effective in sealing biliary leaks in 71% of cases. Cannulation of the bile duct could not be achieved in 4%.11 With endoscopically placed retrievable covered stents, 100% resolutions were achieved after a mean stent dwell time of 156 days (range, 67-493 days) in 11 patients with complex bile leaks.22 After liver transplantation, the outcome of patients with anastomotic bile leaks or leaks from bile duct necrosis is relatively poor. Even after surgical revision, a death rate of 32% has been reported in a series of 34 patients. All deaths were associated with uncontrollable sepsis.3

COMPLICATIONS Because bile ducts are often nondilated in patients with biliary leakage, the complication rate of PTBD for this indication may exceed that of PTBD for biliary obstructive disease. The frequency of complications is shown in Table 136-1 and ranged from 0% to 25%.5,15 Most authors did not differentiate minor from major complications. There is one report of PTBD in 130 patients with nondilated bile ducts performed for different indications, not just for biliary leakage. Major procedure-related complications occurred in 4%.16 Hemobilia secondary to arterial injury has been reported, as well as arterial bleeding after catheter removal. If bleeding does not stop spontaneously, selective arterial embolization is the treatment of choice. In another study, subcapsular hepatic hematoma developed in a patient with necrotizing pancreatitis after PTBD and was treated by laparotomy. However, the patient died of recurrent bleeding.15 Lethal gastrointestinal hemorrhage from duodenal ulcer has also been reported.5 Residual biliary stenosis occurred in up to 31% and was managed successfully by balloon dilation or stent insertion.15 Metabolic acidosis

secondary to continuous bile loss from external drainage was treated successfully by fluid replacement and administration of bicarbonate.5 Intermittent fever with chills after catheter exchange is a minor complication that can be prevented by prophylactic single-dose administration of antibiotics.

POSTPROCEDURAL AND FOLLOW-UP CARE Immediately after PTBD, patients are kept in a recovery room under cardiovascular monitoring. Depending on their general condition, they return to a normal ward or to an intensive care unit. Analgesia with opioids and nonsteroidal antiinflammatory drugs, as well as antibiotic therapy, continues according to the clinical course. Generally, the duration of antibiotic therapy should exceed the decline in clinical symptoms and inflammation by a few days. Patient monitoring plus replacement of bile losses with intravenous fluid and electrolytes is necessary during external biliary drainage.5 The percutaneous biliary drainage catheter can be irrigated during the first postinterventional days, but irrigation must be performed gently to prevent any intraductal overpressure that may keep the leak patent. Frequent catheter cholangiograms are performed during the patient’s hospitalization. When a significant reduction in biliary extravasation is noted, drainage is internalized. Before discharge, the 8.3F catheter is exchanged for a larger softer catheter. For instance, a 10F or 12F Munich catheter with various sidehole configurations can be introduced. The outer end of this catheter has the form of a flat hat with holes for placing sutures to prevent catheter dislocation. Using a special connection tube, additional external drainage or irrigation is possible. Regular sterile cleaning and bandaging of the catheter access site must be performed by an instructed relative, a nurse, or the family doctor to prevent access-related infections. Patients should not take baths or go swimming. They can take showers, with the tubes protected by plastic covering that is impervious to water. The time needed to heal a biliary leak extends beyond that required for drainage of an associated biloma.9 Prolonged biliary drainage for several weeks to months may be necessary. Catheters are usually exchanged every 2 months on an outpatient basis.5 If abdominal pain and signs of infection recur, the patient should immediately go to the interventional unit for testing of tube patency and eventual catheter exchange. If no further biliary leak is demonstrated on repeated cholangiograms and the patient is doing well, the drainage catheter is exchanged for a catheter of smaller diameter that has no side holes and therefore no draining function. This tube is left in place to preserve biliary access and let the hepatic puncture tract gradually contract. After a test period of about 4 weeks, the catheter can be removed, provided the bile ducts are normal and the patient is free of complaints. Associated biliary strictures can be treated percutaneously23 (Fig. 136-4). However, complete sealing of the biliary defect is advocated before cholangioplasty is performed at the site of the former leak. Otherwise, repeated rupture may occur when the duct wall is stressed.

CHAPTER 136 MANAGEMENT OF BILIARY LEAKS 1007

A

B

D

C

E

FIGURE 136-4. Carcinoma of duodenal papilla in a 55-year-old man. After duodenopancreatectomy (Whipple technique) and cholecystectomy, patient was evaluated for jaundice and persistent biliary effluence from abdominal drain. A, Under computed tomography guidance, a subhepatic fluid collection was drained percutaneously. B, Percutaneous transhepatic cholangiography revealed extravasation of bile from cystic duct stump (arrow) as well as stricture at biliodigestive anastomosis (arrowhead). C, Three weeks after percutaneous transhepatic biliary drainage (PTBD), no further bile extravasation was noted. Repeat dilation of biliodigestive anastomosis with a 10/40-mm balloon catheter (D) finally resulted in bile flow into jejunum (E) (arrow). Percutaneous access was discontinued 5 months after initiation of PTBD.

KEY POINTS •

Elderly patients can lack clinical symptoms of biliary leakage.

•

Whereas small bile leaks may be managed conservatively, ongoing biliary leakage has to be treated by endoscopic or percutaneous biliary decompression or surgical repair.

•

Because percutaneous biliary drainage (PTBD) is more invasive, it should be performed only when endoscopic relief is not possible.

•

In biliary leakage, the bile ducts are frequently not dilated.

•

In complex lesions, a biliary drainage catheter should cross the injured part of the bile duct with its side holes above and below the leak.

•

Prolonged dwell time of the drainage catheter is recommended after bile leak closure.

•

Retrievable covered stents can be used to treat complex biliary leaks.

▶ SUGGESTED READINGS Vecchio R, MacFadyen BV, Ricardo AE. Bile duct injury: management options during and after gallbladder surgery. Semin Laparosc Surg 1998;5:135–44.

The complete reference list is available online at www.expertconsult.com.

CHAPTER 136 MANAGEMENT OF BILIARY LEAKS1007.e1 REFERENCES 1. Wills VL, Jorgensen JO, Hunt DR. Role of relaparoscopy in the management of minor bile leakage after laparoscopic cholecystectomy. Br J Surg 2000;87:176–80. 2. Rossi P, Servili S, Contine A, et al. Bile leak from the hepatic bed after laparoscopic cholecystectomy. Chir Ital 2002;54:507–9. 3. Sheng R, Sammon JK, Zajko AB, Campbell WL. Bile leak after hepatic transplantation: cholangiographic features, prevalence, and clinical outcome. Radiology 1994;192:413–6. 4. Reed DN Jr, Vitale GC, Wrightson WR, et al. Decreasing mortality of bile leaks after elective hepatic surgery. Am J Surg 2003;185:316–8. 5. Kaufman SL, Kadir S, Mitchell SE, et al. Percutaneous transhepatic biliary drainage for bile leaks and fistulas. AJR Am J Roentgenol 1985;144:1055–8. 6. Sharif K, de Ville de Goyet J. Bile duct of Luschka leading to bile leak after cholecystectomy—revisiting the biliary anatomy. J Pediatr Surg 2003;38:E21–E23. 7. Zhang JM, Yu SA, Shen W, Zheng ZD. Pathogenesis and treatment of postoperative bile leakage: report of 38 cases. Hepatobiliary Pancreat Dis Int 2005;4:441–4. 8. De Castro SM, Kuhlmann KF, Busch OR, et al. Incidence and management of biliary leakage after hepaticojejunostomy. J Gastrointest Surg 2005;9:1163–71. 9. VanSonnenberg E, Ferrucci JT Jr, Mueller PR, et al. Percutaneous drainage of abscesses and fluid collections: technique, results, and applications. Radiology 1982;142:1–10. 10. Vaccaro JP, Dorfman GS, Lambiase RE. Treatment of biliary leaks and fistulae by simultaneous percutaneous drainage and diversion. Cardiovasc Intervent Radiol 1991;14:109–12. 11. Saab S, Martin P, Soliman GY, et al. Endoscopic management of biliary leaks after T-tube removal in liver transplant recipients: nasobiliary drainage versus biliary stenting. Liver Transpl 2000;6:627–32. 12. Tsukamoto T, Hirohashi K, Osugi H, et al. Percutaneous management of bile duct injuries after cholecystectomy. Hepatogastroenterology 2002;49:113–5.

13. McLindon JP, England RE, Martin DF. Causes, clinical features and non-operative management of bile leaks. Eur Radiol 1998;8:1602–7. 14. Society of Interventional Radiology Standards of Practice Committee. Quality guidelines for percutaneous transhepatic cholangiography, biliary drainage, and percutaneous cholecystostomy. J Vasc Interv Radiol 2010;21:789–95. Available at www.sirweb.org. 15. Ernst O, Sergent G, Mizrahi D, et al. Biliary leaks: treatment by means of percutaneous transhepatic biliary drainage. Radiology 1999;211: 345–8. 16. Funaki B, Zaleski GX, Straus CA, et al. Percutaneous biliary drainage in patients with nondilated intrahepatic bile ducts. AJR Am J Roentgenol 1999;173:1541–4. 17. Stampfl U, Hackert TH, Radeleff B, et al. Percutaneous management of postoperative bile leaks after upper gastrointestinal surgery. Cardiovasc Intervent Radiol 2011;34:808–15. 18. Chen XP, Peng SY, Peng CH, et al. A ten-year study on non-surgical treatment of postoperative bile leakage. World J Gastroenterol 2002; 8:937–42. 19. Righi D, Franchello A, Ricchiuti A, et al. Safety and efficacy of the percutaneous treatment of bile leaks in hepaticojejunostomy or splitliver transplantation without dilatation of the biliary tree. Liver Transpl 2008;14:611–5. 20. Civelli EM, Meroni R, Cozzi G, et al. The role of interventional radiology in biliary complications after orthotopic liver transplantation: a single-center experience. Eur Radiol 2004;14:579–82. 21. Gwon DI, Ko GY, Sung KB, et al. Percutaneous transhepatic treatment of postoperative bile leaks: prospective evaluation of retrievable covered stents. J Vasc Interv Radiol 2011;22:75–83. 22. Baron TH. Covered self-expandable metal stents for benign biliary tract disease. Curr Opin Gastroenterol 2011;27:262–7. 23. Citron SJ, Martin LG. Benign biliary strictures: treatment with percutaneous cholangioplasty. Radiology 1991;178:339–41.

CHAPTER

137

Percutaneous Cholecystostomy

Thomas M. Fahrbach, Gerald M. Wyse, Leo P. Lawler, and Hyun S. Kim CLINICAL RELEVANCE Acute cholecystitis (AC) is a prevalent condition that carries significant risks of morbidity and mortality.1 AC may present with a spectrum of disease stages ranging from a mild self-limited illness to a fulminant potentially lifethreatening illness.2 Laparoscopic or open surgical removal of the inflamed gallbladder remains the gold-standard therapy when the patient is a surgical candidate.3,4 However, because surgery requires general anesthesia, incisions, and postoperative recuperation, percutaneous cholecystostomy (PC) has served to largely replace surgery in patients who have significant comorbidities or anatomic variations that may preclude or delay surgical therapy.5-8 This minimally invasive nonsurgical approach has in fact been suggested as definitive therapy in critically ill and elderly patients,9 as well as to provide a bridge to surgery.10,11 The overall goal of PC is gallbladder decompression and drainage for prevention of gallbladder perforation and sepsis.12-16 The technique of PC also offers a minimally invasive means of access to the biliary system for other interventions such as stent placement and stone retrieval.17-20

INDICATIONS As just noted, there are primarily two clinical reasons patients are referred for cholecystostomy: (1) to decompress the gallbladder for management of cholecystitis21 or (2) to provide a portal of access to the biliary tract for therapeutic purposes.22 A multidisciplinary approach to determine whether to proceed with PC is ideal and should be guided by clinical symptomatology, laboratory data, and supporting imaging evidence.22 The clinical manifestations of cholecystitis include fever, elevated white blood cell count, right upper quadrant pain, and Murphy sign. Patients may suffer from either calculous or acalculous cholecystitis (Figs. 137-1 to 137-4).23,24 Those with calculous cholecystitis have gallstones and possibly sludge causing mechanical obstruction of the cystic duct. The exact pathogenesis of acalculous cholecystitis is unclear but tends to occur in the critically ill, debilitated, or intensive care patients. Associations with diabetes, malignant disease, vasculitis, and congestive heart failure also exist,25 as well as being secondary to mechanical obstruction of the cystic duct from biliary stent placement in the common bile duct. Empirical cholecystostomy may be indicated in patients with fever of unknown origin or other clinical evidence of sepsis where there is no apparent source other than classic imaging features of cholecystitis. Sonographic, computed tomography (CT), or magnetic resonance imaging (MRI) findings of uncomplicated cholecystitis include gallbladder wall thickening, pericholecystic fluid/stranding, gallbladder distension, sonographic Murphy sign (right upper quadrant pain with probe 1008

pressure), or a gallstone impacted in the neck of the gallbladder or cystic duct.26,27 Findings of complicated cholecystitis such as pericholecystic abscess, intraluminal membranes, and/or gas suggest a gangrenous or perforated gallbladder.28 Hepatobiliary nuclear scintigraphy with technetium (Tc)-99m iminodiacetic acid derivatives provides highly specific diagnostic confirmation, with nonvisualization of the gallbladder at 4 hours suggesting cystic duct obstruction and cholecystitis.29 Cholecystostomy may also be employed as a means to access the biliary system when transhepatic or endoscopic routes are not feasible or contraindicated. This portal to the biliary system provides a pathway for a variety of therapeutic interventions, including internal/external biliary drain placement, metallic stent placement, gallstone extraction, dissolution, lithotripsy, and even gallbladder ablation.

CONTRAINDICATIONS Frequently, PC is performed in gravely ill and high-risk patients, and there are few absolute contraindications. Interposed bowel (e.g., as in Chilaiditi syndrome) may preclude safe access to the gallbladder. Coagulopathy is a relative contraindication, and a severe bleeding diathesis may not allow transhepatic access. Other relative contraindications include gallbladder tumor that may be seeded by percutaneous access, or a gallbladder greatly distended by calculi that prevent drainage tube formation and locking (Fig. 137-5). Finally, it may be difficult or impossible to place a PC tube in a perforated decompressed gallbladder.

EQUIPMENT Percutaneous access to the gallbladder can be achieved via direct image guidance with ultrasound, fluoroscopy, or CT. The preferred method of ultrasound guidance is normally performed with a midrange frequency 2- to 8-Hz curvilinear array sector probe. The procedure may be performed at the bedside when necessary,16,30 but catheter insertion is best performed in a fluoroscopy suite. Axial noncontrast CT may be used, and CT fluoroscopy tools are rarely necessary but may be helpful when a significantly diseased or calcular gallbladder limits sonographic visualization of the lumen.31 Standard sterile technique should always be used. Local anesthesia and moderate sedation with subcutaneous lidocaine 2% and intravenous (IV) midazolam and fentanyl is sufficient for the vast majority of patients, although critical care support is needed in critically ill patients. IV antibiotics with gram-negative coverage is administered. The gallbladder is usually accessed with an 18G percutaneous entry needle or trocar needle, with length dependent on the distance measured to the gallbladder. A 21G micropuncture or 22G Chiba needle with a 0.018-inch wire may be employed with a wire guide exchange set that allows

CHAPTER 137 PERCUTANEOUS CHOLECYSTOSTOMY 1009

FIGURE 137-1. Ultrasound of acalculous cholecystitis showing a sludgefilled gallbladder with gallbladder wall thickening and pericholecystic fluid (arrow).

FIGURE 137-4. Axial T2-weighted magnetic resonance imaging shows a distended gallbladder with mixed signal within it (arrow), consistent with blood products and hemorrhagic cholecystitis.

FIGURE 137-2. Axial contrast-enhanced computed tomography shows a distended gallbladder and stranding of pericholecystic fat (arrow).

FIGURE 137-5. Contrast study through cholecystostomy catheter (short black arrow) shows a gallbladder packed with stones (long black arrow). Note absence of cystic duct filling.

FIGURE 137-3. Axial noncontrast computed tomography shows multiple gallstones, increased attenuation in gallbladder mucosa, and edema of gallbladder wall (arrow).

transition to a 0.035-inch system; 8F and 10F tissue dilators may be required. The biliary system is opacified with iodinated contrast such as Hypaque (diatrizoate sodium and diatrizoate meglumine [Nycomed Inc., Princeton, N.J.]). A 0.035-inch Rosen (Cook Medical, Bloomington, Ind.) or short Amplatz Super Stiff wire (Boston Scientific, Natick, Mass.) have the necessary stiffness to support drainage tube placement and may also be partly looped within the gallbladder. A self-retaining all-purpose drain with distal side holes (e.g., Flexima [Boston Scientific]) is used for gravity drainage and connected to a bag. A variety of

1010 PART 2 NONVASCULAR INTERVENTIONS · SECTION 19 THE BILIARY TRACT

FIGURE 137-6. Axial contrast-enhanced computed tomography showing potential intercostal access routes (white lines) for transhepatic cholecystostomy.

FIGURE 137-7. Coronal contrast-enhanced computed tomography shows an intercostal transhepatic route to gallbladder (white line). Note close relationship of colon (arrowhead) and small bowel (arrow).

cholangioplasty balloons, snares, and lithotripsy may be employed when gallstone extraction is contemplated. The gallbladder and biliary system may be directly visualized with a 15F choledochoscope, which will require an 18F tract.

TECHNIQUE Anatomy and Approach The gallbladder may be approached via one of two basic methods: transhepatic or transperitoneal. In general, the transhepatic route involves an arc from the right midaxillary line to the midclavicular line (Fig. 137-6), and hypothetically traverses the bare area of the gallbladder (superior third portion of the gallbladder).31 A track is chosen below the diaphragm, which may or may not be intercostal. Intercostal access is immediately above the rib to avoid the neurovascular bundle (Fig. 137-7). It is important to exclude any ascending colon or hepatic flexure interposed between the liver and abdominal wall along the anticipated track. The transhepatic approach generally traverses segments 5 or 6 and enters the gallbladder through the gallbladder fossa (see Fig. 137-7). The transperitoneal approach is generally anterior or anterolateral, and similarly, interposed bowel must be excluded (Fig. 137-8).21

Technical Aspects Fully informed consent is obtained. Prothrombin time or international normalized ratio (INR), partial thromboplastin time (PTT), and platelet count are checked, and any coagulopathy is corrected. Available imaging is reviewed

FIGURE 137-8. Axial noncontrast computed tomography shows sludge within gallbladder, a thickened gallbladder wall, and pericholecystic fluid (long arrow) in a patient with acalculous cholecystitis. Note close relationship of transverse colon (short arrows) along potential anterior or posterior approaches.

to assess a safe percutaneous window. With the patient in a supine position and right arm abducted, the abdomen is sonographically evaluated for determination of access site and the transhepatic/transperitoneal route to the gallbladder. Depth is measured for needle length selection. Under

CHAPTER 137 PERCUTANEOUS CHOLECYSTOSTOMY 1011 fluoroscopy, the level of the diaphragm and the planned level of access is marked, and a scout image is taken to avoid a transpulmonary or transpleural approach. Lidocaine 2% local anesthetic is applied, and a small nick is made at the needle puncture site. If necessary, the subcutaneous fat and muscle may be bluntly dissected with a curved forceps. There are two methods for catheter placement when performing PC: the Seldinger technique or the trocar technique.16 The Seldinger technique is better suited for difficult access or a small gallbladder and uses an 18G to 22G needle, which is advanced into the gallbladder under image guidance with visualization of the needle tip at all times. Upon aspiration of bile, contrast can be injected for cholecystography/cholangiography and to fluoroscopically confirm intraluminal placement within the gallbladder. A guidewire is then advanced through the needle and coiled within the gallbladder, taking care not to overdistend or stretch the inflamed gallbladder. An 8F to 10F self-retaining/ locking all-purpose drain is then advanced into the gallbladder and fed off its stiffener to form it. Prior soft-tissue tract dilation may be required. The trocar technique is better suited to those with a large distended gallbladder that can be easily visualized with a clear percutaneous window. With the trocar technique, the needle tip is placed within the all-purpose drain, and the entire system is advanced into the gallbladder. This technique also has the advantage of a single step and access to the gallbladder with no exchanges over a wire. With either technique, the placed drain is formed, locked, and injected with a small amount of contrast (3-5 mL) to confirm position (Fig. 137-9). It is sutured to the skin and attached to a gravity bag. A sample of bile is sent for culture and Gram stain.21

FIGURE 137-9. Contrast injection through cholecystostomy catheter shows irregular filling of gallbladder, characteristic of mucosal inflammation. Note absent cystic duct filling (arrow).

EQUIPMENT • Imaging: • 2- to 8-Hz ultrasound curvilinear array probe • Fluoroscopic systems: full angiographic suite, mobile C-arm • CT, with or without CT fluoroscopy • Medications and contrast: • Gram-negative antibiotic coverage • Lidocaine 2% • Moderate sedation: midazolam, fentanyl • Contrast (e.g., Hypaque [diatrizoate sodium and diatrizoate meglumine]) • Access: • 18G percutaneous entry needle or trocar needle with 0.035-inch wire Or • 21G micropuncture or 22G Chiba with 0.018-inch wire and a transitional wire guide exchange set • 8F to 10F tissue tract dilators • Wires: • 0.035-inch, 145-cm Rosen (Cook Medical, Bloomington, Ind.) Or • Short 0.035-inch, 75-cm Amplatz Super Stiff wire (Boston Scientific) • Drain: • Self-retaining all-purpose drain (e.g., Flexima [Boston Scientific]) • Drainage bag • 2-0 Prolene suture

CONTROVERSIES The transhepatic approach for cholecystostomy drain placement is generally preferred. Advantages include a theoretical decreased risk of intraperitoneal bile leakage, more rapid tract maturation, and greater catheter stability as the decompressed gallbladder collapses toward the tube, thus preventing bile leak. If the bare area of the gallbladder was successfully targeted, any bile leak that does occur should be extraperitoneal. The transhepatic access is also more stable, because the less mobile point of access is better suited for exchanges or interventional procedures through the tract, especially in the setting of ascites. Decreased risks of bleeding and liver contamination with infected bile have been postulated with the transperitoneal approach, but concerns remain for an increased risk of bile spillage and subsequent peritonitis due to longer tract maturation time. The transperitoneal approach has also been associated with an increased risk of colon perforation, portal vessel injury, and displacement of the catheter after decompression of the gallbladder as the decompressing gallbladder recedes away from the catheter tip.32 However, satisfactory results have been reported for the transperitoneal approach, and it may be considered favorable in patients with liver disease or coagulopathy. Its advocates note that the inflamed gallbladder is thickened and distended, so it is thus relatively immobile for ease of puncture. Simple bile aspiration without drain placement for decompression of the gallbladder may suffice, and there are limited studies in support of this.33 However, if aspiration

1012 PART 2 NONVASCULAR INTERVENTIONS · SECTION 19 THE BILIARY TRACT fails to initially resolve the acute inflammation, the patient bears the risk, pain, and cost of a repeat procedure. The lack of definitive tube placement also prevents subsequent tube cholangiograms and biliary access.

OUTCOMES The technique of cholecystostomy tube placement is relatively straightforward, and as such it has enjoyed good technical success in 95% to 100% of patients.14,15,34-36 Technical failures occur with thick bilious aspirate not amenable for drainage, decompressed gallbladders, porcelain gallbladders, and pronounced gallbladder wall thickening. The results of clinical success are more varied with efficacy from 60 to over 90%.* This is in large measure due to the wide variation in patient clinical status and patient selection. Frequently a PC is requested in patients too sick for a definitive surgery, and frequently—despite successful cholecystostomy placement—they succumb to the other comorbidities, which are difficult to discriminate from the failure of biliary decompression.1 Outcomes after PC for AC are better when the disease is primary and not precipitated by concurrent illness.38 The best clinical results are in those with good clinical and imaging evidence of cholecystitis, and the response is rapid within 72 hours of placement.15,16

FIGURE 137-10. Contrast study through cholecystostomy catheter (long white arrow) shows a bile leak (short white arrow) tracting superiorly, with formation of a biloma within mediastinum (black arrow). Arrowhead points to inferior vena cava filter.

COMPLICATIONS Major complications of PC are rare, and similar to most minimally invasive procedures, PC is considered low risk. Intraprocedural major complications include sepsis, peritonitis, abscess, hemorrhage, or transgression of adjacent structures, all of which are reported to occur in less than 3% of cases.22 A recent critical review of peer-reviewed articles reports an intraprocedural death rate of 1.7%.22 The most common complication of PC is biliary leak, which may occur during placement or removal of catheters (Fig. 137-10).34,39 The acutely inflamed gallbladder wall is friable and may perforate with wire or catheter manipulation. Upon catheter removal, major and minor bile leaks have been reported in 3% of patients.39 Leakage may be selflimited and managed conservatively. Development of a biloma may require placement of an additional subhepatic drain. Large-volume contrast injections into the gallbladder may precipitate rigors and sepsis. Without diligent evaluation of preprocedural imaging, transgression of surrounding structures may occur, with breaching of the pleural space or bowel, resulting in pneumothorax, bowel perforation, and fistula formation. Transpleural tubes must be removed and replaced because they can result in bile tracking into the pleural space, resulting in pleural reaction and effusion. Transpulmonary tubes will likely result in pneumothorax. If injury to bowel is questioned during the procedure, it may be confirmed by placing a sheath over the wire and injecting while pulling back. One may withdraw the wire or catheter immediately and replace or allow the tract to mature, creating a controlled fistula that may resolve with time if surgery is not an option. Alternately, some will primarily repair such *References 12-14, 16, 21, 23, 34, 35, 37.

injuries by surgery if clinically indicated. Hemorrhage may result from hepatic arterial, venous, or portal venous injury via transhepatic placement and may necessitate vascular intervention and embolization if hemodynamically significant. Most hemobilia is self-limited and responds well to tamponade by catheter upsizing.21,40

POSTPROCEDURE AND FOLLOW-UP CARE Watchful monitoring of postprocedure vital signs and symptoms should occur initially at short intervals for the first 2 hours, and then every subsequent 2 to 4 hours. Antibiotic coverage should be adjusted based on the results of Gram stain and culture. The cholecystostomy catheter should be attached to external bag drainage to gravity and should be flushed with 5 to 10 mL normal saline twice daily. The tube should remain to external drainage and not be capped. Early cholecystograms are not required unless they will effect a management change. The tract is left to mature for 3 to 6 weeks; early removal after as little 14 days can be safe, but tract maturation may be prolonged in immunocompromised or critically ill patients. Determination of tube removal depends on the initial intention-to-treat and clinical goals.1 For acalculous cholecystitis, PC may be the definitive treatment, and the tube may be removed once the episode has passed. A tube cholangiogram is performed in anticipation of tube removal or subsequent cholecystectomy11 to exclude cystic duct obstruction or extrahepatic biliary tract stones (Fig. 137-11).19,20 Prior to tube removal, a clinical trial may be initiated by capping the drain for 48 hours to exclude

CHAPTER 137 PERCUTANEOUS CHOLECYSTOSTOMY 1013 recurrent cholecystitis. For calculous cholecystitis, the tube usually remains in place until cholecystectomy. Nonsurgical options of percutaneous stone removal require tract dilatation up to 18F or larger for stone extraction or lithotripsy using fluoroscopic or choledochoscope guidance (Fig. 137-12).19,20

KEY POINTS

FIGURE 137-11. Contrast study through cholecystostomy catheter (long arrow) shows a large stone in gallbladder fundus (arrowhead), with a patent cystic duct filling a stone-free common bile duct (short arrow).

•

Acute cholecystitis (AC) is a prevalent condition that may present with a spectrum of disease stages ranging from a mild self-limited illness to a fulminant potentially lifethreatening illness.

•

Percutaneous cholecystostomy is a minimally invasive nonsurgical treatment proven to be a safe and effective first-line therapy for AC.

•

Cholecystostomy offers an alternate interventional access to the biliary system for other biliary procedures.

▶ SUGGESTED READINGS Ginat D, Saad WE. Cholecystostomy and transcholecystic biliary access. Tech Vasc Interv Radiol 2008;11(1):2–13. Review. PubMed PMID: 18725137. Joseph T, Unver K, Hwang GL, et al. Percutaneous cholecystostomy for acute cholecystitis: ten-year experience. J Vasc Interv Radiol 2012;23(1):83–8.e1. Epub 2011 Nov 30. PubMed PMID: 22133709. Pomerantz BJ. Biliary tract interventions.Tech Vasc Interv Radiol 2009;12(2):162–70. Review. PubMed PMID: 19853234. Saad WE, Wallace MJ, Wojak JC, et al. Quality improvement guidelines for percutaneous transhepatic cholangiography, biliary drainage, and percutaneous cholecystostomy. J Vasc Interv Radiol 2010;21(6):789– 95. Epub 2010 Mar 21. PubMed PMID: 20307987.

The complete reference list is available online at www.expertconsult.com. FIGURE 137-12. Photograph of a 15F cholangioscope being passed into biliary system for stone retrieval. Safety wire maintains biliary system access.

CHAPTER 137 PERCUTANEOUS CHOLECYSTOSTOMY1013.e1 REFERENCES 1. Yusoff IF, Barkun JS, Barkun AN. Diagnosis and management of cholecystitis and cholangitis. Gastroenterol Clin North Am 2003; 32(4):1145–68. 2. Hirota M, Takada T, Kawarada Y, et al. Diagnostic criteria and severity assessment of acute cholecystitis: Tokyo Guidelines. J Hepatobiliary Pancreat Surg 2007;14(1):78–82. Epub 2007 Jan 30. PubMed PMID: 17252300; PubMed Central PMCID: PMC2784516. 3. Ghahreman A, McCall JL, Windsor JA. Cholecystostomy: a review of recent experience. Aust N Z J Surg 1999;69(12):837–40. 4. Soper NJ, Brunt LM, Kerbl K. Laparoscopic general surgery. N Engl J Med 1994;330(6):409–19. 5. Li JC, Lee DW, Lai CW, et al. Percutaneous cholecystostomy for the treatment of acute cholecystitis in the critically ill and elderly. Hong Kong Med J 2004;10(6):389–93. 6. Macri A, Scuderi G, Saladino E, et al. Acute gallstone cholecystitis in the elderly: treatment with emergency ultrasonographic percutaneous cholecystostomy and interval laparoscopic cholecystectomy. Surg Endosc 2006;20(1):88–91. 7. Hamy A, Visset J, Likholatnikov D, et al. Percutaneous cholecystostomy for acute cholecystitis in critically ill patients. Surgery 1997; 121(4):398–401. 8. Akinci D, Akhan O, Ozmen M, et al. Outcomes of percutaneous cholecystostomy in patients with high surgical risk. Tani Girisim Radyol 2004;10(4):323–7. 9. Griniatsos J, Petrou A, Pappas P, et al. Percutaneous cholecystostomy without interval cholecystectomy as definitive treatment of acute cholecystitis in elderly and critically ill patients. South Med J 2008; 101(6):586–90. PubMed PMID: 18475218. 10. Macri A, Scuderi G, Saladino E, et al. Acute gallstone cholecystitis in the elderly: treatment with emergency ultrasonographic percutaneous cholecystostomy and interval laparoscopic cholecystectomy. Surg Endosc 2006;20(1):88–91. 11. Akyurek N, Salman B, Yuksel O, et al. Management of acute calculous cholecystitis in high-risk patients: percutaneous cholecystotomy followed by early laparoscopic cholecystectomy. Surg Laparosc Endosc Percutan Tech 2005;15(6):315–20. 12. Boland GW, Lee MJ, Leung J, Mueller PR. Percutaneous cholecystostomy in critically ill patients: early response and final outcome in 82 patients. AJR Am J Roentgenol 1994;163(2):339–42. 13. Hultman CS, Herbst CA, McCall JM, Mauro MA. The efficacy of percutaneous cholecystostomy in critically ill patients. Am Surg 1996;62(4):263–9. 14. Lo LD, Vogelzang RL, Braun MA, Nemcek AA Jr. Percutaneous cholecystostomy for the diagnosis and treatment of acute calculous and acalculous cholecystitis. J Vasc Interv Radiol 1995;6(4):629–34. 15. Sosna J, Copel L, Kane RA, Kruskal JB. Ultrasound-guided percutaneous cholecystostomy: update on technique and clinical applications. Surg Technol Int 2003;11:135–9. 16. Browning PD, McGahan JP, Gerscovich EO. Percutaneous cholecystostomy for suspected acute cholecystitis in the hospitalized patient. J Vasc Interv Radiol 1993;4(4):531–7; discussion 7-8. 17. Rozenblit GN, Eisenberger E, Rundback JH, et al. Percutaneous cholecystoduodenostomy: a case report. J Vasc Interv Radiol 2000;11(5): 629–33. 18. Miyayama S, Matsui O, Akakura Y, et al. Percutaneous cholecystocholedochostomy for cholecystitis and cystic duct obstruction in gallbladder carcinoma. J Vasc Interv Radiol 2003;14(2 Pt 1):261–3. 19. Picus D, Hicks ME, Darcy MD, et al. Percutaneous cholecystolithotomy: analysis of results and complications in 58 consecutive patients. Radiology 1992;183(3):779–84.

20. Picus D, Marx MV, Hicks ME, et al. Percutaneous cholecystolithotomy: preliminary experience and technical considerations. Radiology 1989;173(2):487–91. 21. Akhan O, Akinci D, Ozmen MN. Percutaneous cholecystostomy. Eur J Radiol 2002;43(3):229–36. 22. Saad WE, Wallace MJ, Wojak JC, et al. Quality improvement guidelines for percutaneous transhepatic cholangiography, biliary drainage, and percutaneous cholecystostomy. J Vasc Interv Radiol 2010;21(6):789– 95. Epub 2010 Mar 21. PubMed PMID: 20307987. 23. van Overhagen H, Meyers H, Tilanus HW, et al. Percutaneous cholecystectomy for patients with acute cholecystitis and an increased surgical risk. Cardiovasc Intervent Radiol 1996;19(2):72–6. 24. Welschbillig-Meunier K, Pessaux P, Lebigot J, et al. Percutaneous cholecystostomy for high-risk patients with acute cholecystitis. Surg Endosc 2005;19(9):1256–9. 25. Pomerantz BJ. Biliary tract interventions. Tech Vasc Interv Radiol 2009;12(2):162–70. Review. PubMed PMID: 19853234. 26. Ralls PW, Colletti PM, Lapin SA, et al. Real-time sonography in suspected acute cholecystitis. Prospective evaluation of primary and secondary signs. Radiology 1985;155(3):767–71. PubMed PMID: 3890007. 27. Fidler J, Paulson EK, Layfield L. CT evaluation of acute cholecystitis: findings and usefulness in diagnosis. AJR Am J Roentgenol 1996; 166(5):1085–8. PubMed PMID: 8615248. 28. Bortoff GA, Chen MY, Ott DJ, et al. Gallbladder stones: imaging and intervention. Radiographics 2000;20(3):751–66. 29. Mauro MA, McCartney WH, Melmed JR. Hepatobiliary scanning with 99mTc-PIPIDA in acute cholecystitis. Radiology 1982;142(1):193–7. PubMed PMID: 7053532. 30. Habib FA, McKenney MG. Surgeon-performed ultrasound in the ICU setting. Surg Clin North Am 2004;84(4):1151–79, vii. 31. Ginat D, Saad WE. Cholecystostomy and transcholecystic biliary access. Tech Vasc Interv Radiol 2008;11(1):2–13. Review. PubMed PMID: 18725137. 32. Malone DE, Burhenne HJ. Advantages and disadvantages of the newer “interventional” procedures for the treatment of cholecystolithiasis. Hepatogastroenterology 1989;36(5):317–26. Review. PubMed PMID: 2695445. 33. Ito K, Fujita N, Noda Y, et al. Percutaneous cholecystostomy versus gallbladder aspiration for acute cholecystitis: a prospective randomized controlled trial. AJR Am J Roentgenol 2004;183(1):193–6. 34. vanSonnenberg E, D’Agostino HB, Goodacre BW, et al. Percutaneous gallbladder puncture and cholecystostomy: results, complications, and caveats for safety. Radiology 1992;183(1):167–70. 35. Vogelzang RL, Nemcek AA, Jr. Percutaneous cholecystostomy: diagnostic and therapeutic efficacy. Radiology 1988;168(1):29–34. 36. Boggi U, Di Candio G, Campatelli A, et al. Percutaneous cholecystostomy for acute cholecystitis in critically ill patients. Hepatogastroenterology 1999;46(25):121–5. PubMed PMID: 10228775. 37. Boland GW, Lee MJ, Dawson SL, Mueller PR. Percutaneous cholecystostomy for acute acalculous cholecystitis in a critically ill patient. AJR Am J Roentgenol 1993;160(4):871–4. 38. Joseph T, Unver K, Hwang GL, et al. Percutaneous cholecystostomy for acute cholecystitis: ten-year experience. J Vasc Interv Radiol 2012;23(1):83–8.e1. Epub 2011 Nov 30. PubMed PMID: 22133709. 39. Wise JN, Gervais DA, Akman A, et al. Percutaneous cholecystostomy catheter removal and incidence of clinically significant bile leaks: a clinical approach to catheter management. AJR Am J Roentgenol 2005;184(5):1647–51. 40. Stempel LR, Vogelzang RL. Hemorrhagic cholecystitis with hemobilia: treatment with percutaneous cholecystostomy and transcatheter urokinase. J Vasc Interv Radiol 1993;4(3):377–80.

CHAPTER

138

Management of Biliary Calculi Daniel B. Brown and Daniel D. Picus CLINICAL RELEVANCE Symptomatic gallstones affect 260,000 Americans every year.1 The majority of patients are treated definitively with surgery, and with the advent of laparoscopic cholecystectomy, surgeons can treat patients less invasively and allow for a quicker recovery. Even in the setting of laparoscopic cholecystectomy, significant medical comorbidities will make some patients unsuitable candidates for surgery. This group can greatly benefit from percutaneous drainage and staged stone extraction. Patients with symptomatic gallstones often simultaneously have calculi in the intrahepatic or extrahepatic bile ducts. Endoscopic retrograde cholangiopancreatography (ERCP) is the principle technique to clear these stones in the majority of patients.2 However, in patients with challenging anatomy (duodenal diverticulum, impacted stones > 15 mm), endoscopic techniques are more prone to failure. Finally, patients who have undergone previous biliary reconstruction as part of a Billroth II or Roux-en-Y anastomosis may be inaccessible via endoscopy, and intrahepatic calculi in these patients may require percutaneous removal.

INDICATIONS Indications for percutaneous biliary stone management are relatively straightforward. The principal indication for gallstone extraction is treatment of symptomatic calculi in patients who are not candidates for surgery. The majority of patients with retained biliary duct stones following surgery will be referred following failed ERCP. If hepatolithiasis is diagnosed shortly after surgery and the patient still has a surgical T-tube in place, some surgeons will primarily refer these patients for interventional radiologic management. Peripheral to the hilum of the liver, intraductal calculi are beyond the reach of virtually all endoscopists, and percutaneous management plays a larger role.

CONTRAINDICATIONS The primary absolute contraindication is the presence of an uncorrectable coagulopathy; bleeding complications are a significant source of morbidity when using large-bore percutaneous transhepatic tracts. Relative contraindications include a nondistensible gallbladder and the presence of active infection. Gallstone extraction can be challenging if not impossible in a chronically diseased, contracted gallbladder. Lack of distensibility can limit needle access and tract dilation. In patients with active infection related to stone disease, appropriate drainage catheter(s) must be placed, with a “cooling off ” period to limit septicemia during or following stone extraction. One author has reported successful treatment of a patient with unrelenting fevers secondary to an infected stone burden.3 Only clearance of the 1014

stones allowed this patient’s fever to clear. In the majority of patients, drainage and appropriate antibiotics for 24 to 48 hours will allow recovery to perform stone extraction.

EQUIPMENT Much of the equipment used for percutaneous stone extraction is familiar to trained interventional radiologists. Procedures are staged so initially patients will undergo drainage of the gallbladder with either a locking loop retention catheter or placement of an external or (preferably) internalexternal biliary drainage catheter. Details of these initial drainage procedures and required equipment are described elsewhere in this textbook. Visualization of the gallbladder and biliary ducts is performed with injections of ionic contrast material unless the patient has a significant contrast allergy, in which case the patient is premedicated with corticosteroids, and nonionic contrast material is used. To perform stone removal, we use a 15F flexible choledochoscope (model CHF-4B [Olympus, New Hyde Park, N.Y.]) (Fig. 138-1, A). The scope contains a working channel that allows use of standard instruments measuring up to 5F in diameter, including graspers and baskets. The existing tube tract is dilated to 18F with a series of Teflon dilators (Cook Medical, Bloomington, Ind.) (Fig. 138-1, B). The outer Teflon sheath allows easy passage of the choledochoscope into the biliary tree, providing a large-bore access to flush out stone debris during the procedure. The scope is connected to continuous normal saline irrigation during use. A principal advantage of the choledochoscope is that it allows the operator to characterize filling defects identified at cholangiography as stones, mucus, air bubbles, clots, or soft-tissue masses (Fig. 138-1, C). For stones that are too large to dislodge and fragment with baskets or other tools, we perform intracorporeal electrohydraulic lithotripsy (EHL). The principle of intracorporeal EHL is similar to extracorporeal EHL. A spark from the electrode generates a shockwave in the fluid medium of the gallbladder or bile duct, which then fragments the stone. We use a foot pedal–controlled EHL shockwave generator (EL-115 [American ACMI, Stamford, Conn.]) with 70- to 80-watt power output. The EHL electrode (Northgate, Chicago, Ill.) is 3F and passes through the working channel to allow direct visualization during activation. Others have described use of an 11F choledochoscope (URF type P2 [Olympus]). Potential advantages of the smaller device are that it requires a smaller percutaneous access (14F) to use and may pass more easily beyond strictures within the intrahepatic biliary ducts. The working channel of this scope is 3F. Authors using it described use of a coumarin green pulsed dye laser (MDL-1 LaserTripter [Candela Laser, Wayland, Mass.]) to fracture resistant stones. The laser operates at a wavelength of 504 nm, and power output is between 40 and 100 mJ. A downside of this

CHAPTER 138 MANAGEMENT OF BILIARY CALCULI 1015

A

B

C

FIGURE 138-1. Equipment used for percutaneous choledochoscopy. A, 15F choledochoscope. B, Telescoping Teflon dilators used to create a tract large enough to place endoscope. C, Narrowing of duct around guidewire is identified. Biopsy of this unsuspected mass was positive for cholangiocarcinoma.

device is a smaller working channel to pass baskets and other tools. Additionally, the laser-based equipment is more costly than EHL. Use of ultrasound lithotripters has been described, but the access required for application is prohibitively large (24F) for most operators and has a small field of view.4

TECHNIQUE Percutaneous Cholecystolithotomy This procedure has three components: (1) percutaneous cholecystostomy, (2) tract dilation and stone removal, and (3) tract evaluation and tube removal. Patients are sedated and given intravenous antibiotics (3.375 g of piperacillin/tazobactam or 3 g ampicillin/ sulbactam) prior to the procedure. Some operators may prefer administration of atropine (0.6-1 mg) to limit the potential of a vagal response secondary to gallbladder dilation. Initial access may be either transhepatic or transperitoneal, although the latter approach is more commonly associated with acute pain during catheter placement. The gallbladder may be targeted with any combination of fluoroscopic, computed tomographic (CT), or ultrasonic guidance. Optimally the route of entry will be at the end of the

gallbladder fundus to simplify reaching calculi with the EHL probe and to ease clearance of the entire lumen of calculi with baskets. Via the Seldinger technique, a 10F to 14F retention loop catheter is placed (Fig. 138-2). In patients with chronic cholecystitis, tract dilation and stone extraction is performed the following day. Patients who have drains placed for acute cholecystitis should recover from their presenting symptoms prior to stone removal. A safety wire is left in the gallbladder lumen outside the 18F working access. In cases where tract dilation is difficult, removable T-tac fasteners are available.5 Stones too large to remove via the sheath and resistant to basket fragmentation can be broken up with the EHL probe. Important factors to consider when using EHL are that the probe has to be 1 to 2 mm beyond the tip of the scope to avoid damaging the choledochoscope (Fig. 138-3). The probe has to be in direct contact with the stone and not the gallbladder wall. Activation of the EHL probe while in contact with the biliary endothelium can lead to bleeding or bile leakage that will make visualization difficult for the remainder of the case. When the operator believes all stones have been removed, final choledochoscopy is performed to ensure no residual stones remain. Contrast cholangiography is performed to evaluate the cystic and biliary ducts. Calculi are identified in cystic and biliary ducts 28% and 24% of the time, respectively.6 Stones in these structures are

1016 PART 2 NONVASCULAR INTERVENTIONS · SECTION 19 THE BILIARY TRACT

A

B

C

D

FIGURE 138-2. Percutaneous access for cholecystolithotomy. A, Preliminary T2-weighted magnetic resonance imaging demonstrates gallstones (arrow) in a nonsurgical candidate. B, A 14F Cope loop catheter is placed and left to gravity drainage overnight. C, Next morning, tract was dilated to 18F, and a basket was advanced. Note stone ensnared in basket (long arrow) and safety wire outside access sheath (short arrow). D, Final cholangiography demonstrates clearance of stones. Tube was removed 4 weeks later.

A

B

C

FIGURE 138-3. Electrohydraulic lithotripsy (EHL) probe. A, Prior to being advanced through working channel of choledochoscope. B, Directly impacting calculus and avoiding contact with biliary endothelium. C, Fluoroscopic image demonstrating EHL probe appropriately positioned beyond end of choledochoscope (arrow).

CHAPTER 138 MANAGEMENT OF BILIARY CALCULI 1017

A

B

FIGURE 138-4. Percutaneous hepatolithotomy via a T-tube tract. A, Cholangiography via T-tube demonstrates calculus at sphincter of Oddi and at entry to right main hepatic duct. B, Cholangiography after electrohydraulic lithotripsy demonstrates clearance of common duct calculus.

removed via the gallbladder access if possible or by ERCP if necessary. Calculi in the cystic duct are often impossible to ensnare in a basket or treat with EHL. We have found that leaving an internal-external biliary drain through the gallbladder and cystic duct into the common bile duct and duodenum will usually passively fragment the stones and allow them to pass into the duodenum or out the cholecystostomy catheter. At the end of the procedure, a retention loop catheter is again left in the gallbladder (14F-16F). A large-bore catheter is left in place to tamponade the tract used for endoscopy and allow drainage of any residual “gravel” from calculus fragmentation. The catheter is left to gravity drainage.

Percutaneous Hepatolithotomy Planning Route of Access The route of access for these patients is variable and depends upon the timing of presentation. Patients with retained calculi following cholecystectomy may present shortly after surgery with a surgical T-tube in place. These tubes usually form a mature tract quickly and can be used for definitive access approximately 2 weeks after surgery. For patients who present in a delayed fashion following cholecystectomy or those with stones related to primary abnormalities of the biliary ducts or bilioenteric anastomosis, access is usually similar to that used for percutaneous biliary drainage. Approximately half of patients with biliary stones will have multiple calculi, and a careful search for additional stones must be performed during percutaneous cholangiography. If all intrahepatic calculi are located in one lobe of the liver, many operators prefer to access the opposite side to limit any acute angles that may be needed to enter the stone-bearing ducts. Preprocedural noninvasive imaging can be useful to determine stone burden and distribution when planning access. Magnetic resonance cholangiopancreatography has been shown superior to standard CT.7 Compared with ERCP, it has a 96% sensitivity, 100% specificity, 100% positive predictive value, and 96% negative predictive value.

Accessing the Ducts If a T-tube tract is to be used, we perform an initial cholangiogram through the existing catheter. Preprocedure sedation and antibiotic prophylaxis is performed similarly to cholecystolithotomy. The T-tube is removed over a guidewire. Depending on the treatment plan (see later), either a vascular sheath can be placed, or telescoping Teflon dilators can be placed to allow scope placement (Fig. 138-4). If de novo transhepatic access is used, the left, right, or both ducts are accessed in a fashion described in the chapters covering biliary obstruction. Similar to the staged procedure described for cholecystolithotomy, we place an internal-external catheter at the initial visit. If the patient is scheduled electively and is not acutely ill, tract dilation and lithotomy can be performed the next day (Fig. 138-5). If the patient presents acutely, a “cooling-off ” period may be required prior to stone removal. If the patient presents with septicemia and the stones are impacted in the common bile duct, it is wise to leave an external drain in place rather than risk septic shock by manipulating through the calculi into the bowel. Stone Extraction There are two principal and complementary methods to treat/remove ductal calculi: 1. Balloon dilation of the ampulla followed by use of balloons to dislodge and push calculi into the bowel. This technique works best for smaller, non-impacted stones. 2. Basket/EHL fragmentation similar to that used for cholecystolithotomy. This method is important to treat larger calculi. Balloon Manipulation The main advantage of this technique is that the procedure can be performed via a smaller access than is needed for choledochoscope use. Whether via a T-tube tract or percutaneous access, the procedure is performed similarly. For the purposes of this discussion, we will focus on access via a T-tube tract, since the variables are slightly more involved. After removing the T-tube over a guidewire, a sheath is placed in the hepatic ducts. A safety wire is then coiled in

1018 PART 2 NONVASCULAR INTERVENTIONS · SECTION 19 THE BILIARY TRACT

A

B

C

FIGURE 138-5. Percutaneous hepatolithotomy via left duct access in a patient drained for cholangitis. A, Cholangiography demonstrates calculus in common bile duct. B, After placing a safety wire in duodenum, a basket (arrow) is advanced through working port of choledochoscope. C, Follow-up cholangiography reveals clearance of stone.

the small bowel. Cholangiography confirms the location of the stone(s). A 10- or 12-mm balloon is then inflated across the ampulla of Vater for 45 to 90 seconds over one or two inflations to facilitate stone passage. A balloon approximating the size of the largest calculus is then used to fragment the stone. Authors advocating this approach have limited maximal balloon size to 12 mm.8,9 The balloon is then deflated and carefully withdrawn above the stone, reinflated, and used to push the debris into the small bowel. If the calculi are peripheral to the insertion site of the T-tube, standard catheter and guidewire techniques are used to negotiate past the stones into the respective duct(s), and the stones are pulled into the common bile duct with a Fogarty balloon. The calculi are then fragmented and pushed into the small bowel as described earlier. This technique is generally successful for smaller stones. As calculi exceed 15 mm in diameter, adjunctive techniques become necessary, including basket fragmentation and EHL. After cholangiography and/or endoscopy confirm that all calculi have been treated, an internal-external biliary drain is placed across the ampulla and into the small bowel. The catheter is left to gravity drainage overnight and capped in the morning if the patient is afebrile and otherwise asymptomatic. The decision whether to routinely perform intrahepatic choledochoscopy is mildly controversial. Preferences are likely related to expertise in choledochoscope manipulation. Although it requires a larger access, EHL via the 15F scope can successfully treat calculi as large as 4 cm in the central bile ducts.10 A principal advantage of the choledochoscope is that it allows characterization of filling defects in the bile ducts as residual calculi, mucus, blood clots, or soft tissue. Diagnosis and biopsy of unsuspected ductal tumors has been described during cholangioscopy. These tumors may have been diagnosed much later without use of the choledochoscope. Biliary endoscopy by a skilled operator likely cuts down procedure time in complex cases with large or impacted calculi. Operators may elect to use the smaller (11F) scope. This device may pass more easily

past impacted intrahepatic calculi in smaller ducts that are not dislodged by balloon techniques.

OUTCOMES Percutaneous Cholecystolithotomy Complete stone removal can be achieved in over 90% of patients.11,12 A common cause of failure to remove all calculi is missing small fragments discovered at follow-up ultrasound imaging. This factor reinforces the need for careful and thorough inspection of the entire gallbladder lumen with the choledochoscope prior to the end of the extraction procedure, along with critical review of postprocedure cholangiograms. One group found that switching from a rigid to a flexible scope helped them find more residual fragments because it facilitated inspection of the entire gallbladder lumen.11 At a mean follow-up of 33 months, Courtois found that eight out of 65 patients (12%) had recurrent symptoms.6 During follow-up, 30 patients had follow-up ultrasounds, and recurrent stones were found in 12 (40%). Interestingly, only half of these patients had recurrent symptoms. Overall recurrence rates for gallstones in patients undergoing percutaneous cholecystolithotomy are uncertain. Estimates range anywhere from 40% to 70% at 3 to 4 years following the procedure.11,13-15 Even though not all patients with recurrent stones get recurrent cholecystitis, these longterm outcomes dictate that removal of the diseased gallbladder when feasible remains the best option.

Percutaneous Hepatolithotomy Success can be achieved in over 90% of cases.4,8-10,16 Most patients are completely cleared of calculi in one treatment session. An increase in the number of sessions is likely directly related to stone burden. Approximately half of patients will have more than one stone. The majority of calculi are located in the extrahepatic bile duct. On average,

CHAPTER 138 MANAGEMENT OF BILIARY CALCULI 1019 10% to 30% of calculi will be located in the intrahepatic biliary system. In contrast to the frequency of recurrence with percutaneous cholecystolithotomy, successful outcomes with intraductal biliary calculi are typically far more durable, with one report describing no recurrences in 100 patients after a mean of 55 months.9 Patients presenting with ductal stones related to cholecystitis will be most likely to mirror these outcomes, whereas those with diseases centered in the intrahepatic ducts will be more likely to have recurrence. One study focusing on treatment in the intrahepatic ducts found a relatively lower success rate (70.3%) than in other studies. This same group found a significant rate of recurrent sepsis in their patients (26%-42% depending on whether or not the patient had associated biliary strictures).17

COMPLICATIONS Percutaneous Cholecystolithotomy Major complications occur in approximately 6% of patients.4 Most reported complications are related to bile leakage, ranging from simple loculated bilomas to severe bile peritonitis. Thirty-day mortality is approximately 3%.12 This value is in keeping with the very ill patient population undergoing this procedure. As a reference point, in one study, over 20% of treated patients expired of unrelated causes within 18 months of the procedure.11 The most important step to avoid complications related to bile leakage is to leave the tube in place long enough to allow a mature tract to form and perform an over-the-wire “tractogram” prior to removing the catheter. Doing so has diminished this specific complication in our practice. Given the prohibitive morbidity and mortality that would be associated with this patient group with surgical intervention, the outcomes with percutaneous cholecystolithotomy are very reasonable.

operators believe balloon cholangioplasty of the sphincter of Oddi is less traumatic over the long term than formal sphincterotomy performed at ERCP.8,9

POSTPROCEDURE AND FOLLOW-UP CARE Percutaneous Cholecystolithotomy In approximately 2 weeks, the patient is brought back for an outpatient catheter injection. If no residual stones are identified and the gallbladder drains appropriately into the central bile ducts, the catheter is capped to simulate internal drainage. Once the patient tolerates internal drainage for 2 weeks, the catheter is injected over a guidewire to ensure a mature cholecystocutaneous tract is present. Once the tract is mature, the catheter may be removed. Patients are typically seen in the interventional radiology clinic in follow-up on an as-needed basis. Given that many patients with recurrent calculi are asymptomatic, we do not schedule routine follow-up imaging to evaluate for stone recurrence, and rely on patient findings.

Percutaneous Hepatolithotomy Patients are similarly allowed to develop a mature tract, and the catheter is removed upon confirmation of the clearance of all calculi. Given the lack of recurrence for the majority of patients, we do not routinely follow patients with stones secondary to cholecystitis. Patients with more complex scenarios are managed conjointly with the referring surgeon or gastroenterologist. KEY POINTS •

Percutaneous biliary stone management involves two groups of patients: • Patients with symptomatic gallstones but not eligible for cholecystectomy • Patients with calculi in the extra- or intrahepatic biliary ductal system who are not treatable by endoscopic means. This group will often have recently undergone cholecystectomy.

•

With experience and a variety of equipment, operators can remove all calculi in over 90% of patients.

•

Patients with symptomatic gallstones are much more likely than patients with hepatolithiasis to have recurrence, because the diseased gallbladder remains in place.

Percutaneous Hepatolithotomy Similar to cholecystolithotomy, complications with hepatolithotomy have decreased with time and operator experience. In an early report, Stokes et al. had a 17% major complication rate, with 4% mortality within 30 days.18 Another operator directed a 20F metal introducer into the biliary tree to perform choledochoscopy.19 This group had an 8% 30-day mortality and 12% incidence of major hemobilia. Bower et al. reported only one major complication in 68 patients.10 Harris reported no bleeding complications with use of a smaller scope. However, this patient group had a 24% rate of symptomatic bacteremia following endoscopy.3 Their group involved a number of patients with intrahepatic stones secondary to vascular compromise in transplanted livers, and they recommend aggressive antibiotic coverage in this setting based upon their results. Groups who have focused on use of balloons and pushing stone fragments into the small bowel have not avoided hemobilia.8,9,16 Pushing stones into the small bowel is associated with a small but definite risk of pancreatitis. Some operators think that use of octreotide can limit the risk of pancreatitis.9 It should be remembered that although ERCP is a useful adjunct to manage hepatolithiasis, it is not without adverse events. Complications occur in 8% to 12% of patients, and mortality is 0.5% to 1%.20,21 Additionally, some

▶ SUGGESTED READINGS Akiyama H, Okazaki T, Takashima I, et al. Percutaneous treatments for biliary diseases. Radiology 1990;176:25–30. Burhenne HJ. Percutaneous extraction of retained biliary tract stones: 661 patients. AJR Am J Roentgenol 1980;134:888–98. Fache JS. Interventional radiology of the biliary tract. Transcholecystic intervention. Radiol Clin North Am 1990;28:1157–69. Malone DE. Interventional radiologic alternatives to cholecystectomy. Radiol Clin North Am 1990;28:1145–56. Picus D. Intracorporeal biliary lithotripsy. Radiol Clin North Am 1990;28:1241–9.

The complete reference list is available online at www.expertconsult.com.

CHAPTER 138 MANAGEMENT OF BILIARY CALCULI1019.e1 REFERENCES 1. Russo MW, Wei JT, Thiny MT, et al. Digestive and liver diseases statistics, 2004. Gastroenterology 2004;126:1448–53. 2. Livingston EH, Rege RV. Technical complications are rising as common duct exploration is becoming rare. J Am Coll Surg 2005;201:426–33. 3. Harris V, Sherman S, Trerotola S, et al. Complex biliary stones: treatment with a small choledochoscope and laser lithotripsy. Radiology 1996;199:71–7. 4. Burton KE, Picus D, Hicks ME, et al. Fragmentation of biliary calculi in 71 patients by use of intracorporeal electrohydraulic lithotripsy. J Vasc Interv Radiol 1993;4:251–6. 5. Cope C, Burke DR, Meranze SG. Percutaneous extraction of gallstones in 20 patients. Radiology 1990;176:19–24. 6. Courtois CS, Picus DD, Hicks ME, et al. Percutaneous gallstone removal: long-term follow-up. J Vasc Interv Radiol 1996;7:229–34. 7. Soto JA, Alvarez O, Munera F, et al. Diagnosing bile duct stones: comparison of unenhanced helical CT, oral contrast-enhanced CT cholangiography, and MR cholangiography. AJR Am J Roentgenol 2000; 175:1127–34. 8. Garcia-Garcia L, Lanciego C. Percutaneous treatment of biliary stones: sphincteroplasty and occlusion balloon for the clearance of bile duct calculi. AJR Am J Roentgenol 2004;182:663–70. 9. Garcia-Vila JH, Redondo-Ibanez M, Diaz-Ramon C. Balloon sphincteroplasty and transpapillary elimination of bile duct stones: 10 years’ experience. AJR Am J Roentgenol 2004;182:1451–8. 10. Bower BL, Picus D, Hicks ME, et al. Choledochoscopic stone removal through a T-tube tract: experience in 75 consecutive patients. J Vasc Interv Radiol 1990;1:107–12. 11. McDermott V, Arger P, Cope C. Gallstone recurrence and gallbladder function following percutaneous cholecystolithotomy. J Vasc Interv Radiol 1994;5:473–8.

12. Picus D, Hicks ME, Darcy MD, et al. Percutaneous cholecystolithotomy: analysis of results and complications in 58 consecutive patients. Radiology 1992;183:779–84. 13. Jan YY, Chen MF. Percutaneous trans-hepatic cholangioscopic lithotomy for hepatolithiasis: long-term results. Gastrointest Endosc 1995; 42:1–5. 14. O’Donnell LD, Heaton KW. Recurrence and re-recurrence of gall stones after medical dissolution: a long-term follow up. Gut 1988; 29:655–8. 15. Zou YP, Du JD, Li WM, et al. Gallstone recurrence after successful percutaneous cholecystolithotomy: a 10-year follow-up of 439 cases. Hepatobiliary Pancreat Dis Int 2007;6:199–203. 16. Gil S, de la Iglesia P, Verdu JF, et al. Effectiveness and safety of balloon dilation of the papilla and the use of an occlusion balloon for clearance of bile duct calculi. AJR Am J Roentgenol 2000;174:1455–60. 17. Cheung M-T, Kwok PC-H. Liver resection for intrahepatic stones. Arch Surg 2005;140:993–7. 18. Stokes K, Falchuk K, Clouse M. Biliary duct stones: update on 54 cases after percutaneous transhepatic removal. Radiology 1989;170: 999–1001. 19. Bonnel D, Liguory C, Cornud F, Lefebvre J. Common bile duct and intrahepatic stones: results of transhepatic electrohydraulic lithotripsy in 50 patients. Radiology 1991;180:345–8. 20. Bergman JJ, Rauws EA, Fockens P, et al. Randomised trial of endoscopic balloon dilation versus endoscopic sphincterotomy for removal of bile duct stones. Lancet 1997;349:1124–9. 21. Freeman ML, Nelson DB, Sherman S, et al. Complications of endoscopic biliary sphincterotomy. N Engl J Med 1996;335:909–18.

CHAPTER

139

Biliary Complications Associated with Liver Transplantation Robert K. Kerlan, Jr. and Jeanne M. LaBerge

Orthotopic liver transplantation (OLT) is a critical therapeutic alternative for managing patients with progressive liver failure and selected patients with hepatocellular carcinoma. Refinements in surgical technique, improved immunosuppressive agents, and the evolution of minimally invasive image-guided procedures have allowed 5-year survival rates to approach 85%.1-3 The most common source of a replacement organ is a deceased donor. Usually the whole liver is used as the allograft, but the donor organ can be split, with the right lobe or right lobe and medial segment of the left lobe donated to an adult and the left lateral segment given to a child.4-6 Unfortunately, there remains a critical shortage of donor organs when compared with the number of patients on the liver transplant waiting list. This shortage of organs stimulated the development of using right and left hepatic lobes from living donors with matching ABO blood types for urgent transplantation. Despite the use of living donors, the availability of organs falls well short of the demand, and up to 50% of potential recipients die while awaiting transplantation.7,8 As with any complex surgical procedure, complications after orthotopic liver transplantation are not infrequent and occur in both the acute postoperative period and months to years later. The most common source of significant complications in a liver transplant patient is the biliary tract.9-12 To understand the types of complications that occur, thorough comprehension of the surgical anatomy is essential.

can be monitored and the biliary tree can be investigated easily by T-tube cholangiography. Moreover, the presence of the T-tube may make the anastomosis technically easier to create. However, because of complications associated with T-tube placement, as well as biliary leakage after tube removal, many centers have abandoned routine T-tube placement during creation of a CDCD anastomosis.15-17 A recent meta-analysis suggests that biliary complications may not be increased by T-tube insertion, however, and suggests the incidence of biliary strictures may be reduced.18 In some clinical situations, a CDCD anastomosis should not be created. Such circumstances arise when the native recipient extrahepatic bile duct is unsuitable as a conduit, as commonly encountered in patients with primary sclerosing cholangitis and all pediatric patients with biliary atresia. Moreover, if there is a substantial donor/recipient duct size mismatch, a CDCD anastomosis may not be the most attractive option. When a CDCD anastomosis is not the optimal surgical option, a CDJ or hepaticojejunostomy is constructed.19,20 These anastomoses are created by attaching the end or obliquely cut side (to create a larger channel) of the common bile duct or common hepatic duct to the side of a loop of jejunum pulled up to the region of the porta hepatis in a Roux-en-Y configuration. This maneuver requires surgical closure of the free end of the jejunal loop adjacent to the biliary anastomosis, as well as distal jejunojejunostomy to maintain continuity of the alimentary tract.

SURGICAL ANATOMY OF THE BILIARY ANASTOMOSIS

Split Liver

The type of biliary anastomosis created depends on several factors,13,14 including whether a whole cadaveric liver or a split liver has been used, the condition of the recipient’s bile duct, and whether it is an adult or pediatric patient receiving the allograft.

Whole Liver For whole-liver transplantation, two types of biliary anastomosis are constructed: (1) choledochocholedochostomy (CDCD) and (2) biliodigestive choledochojejunostomy (CDJ) or hepaticojejunostomy (HJ). CDCD is an anastomosis of the donor common bile duct to the recipient common bile duct. CDCD is the preferred biliary anastomosis, because sphincteric function is preserved, bowel surgery in a patient who will be immunosuppressed is avoided, and retrograde cholangiography with appropriate endoscopic interventions can be performed postoperatively. The CDCD anastomosis may be performed over a T-tube. If a T-tube is placed, postoperative bile production 1020

The biliary anastomosis in split-liver transplantation can be more technically challenging than in whole-liver transplantation. This is particularly true with living donor transplantation, because not causing harm to the donor is of paramount importance. When harvesting a lobe from a living donor, the biliary duct is severed at a safe distance from the common hepatic duct to avoid stricture formation. This maneuver often results in only a short segment of donor right (or left) hepatic duct being available to anastomose to the recipient duct, making creation of a tensionfree anastomosis difficult. Moreover, variations in donor biliary anatomy may necessitate the creation of more than one biliary anastomosis. Numerous variations in biliary anatomy exist, with so-called normal anatomy being present in only 57% of patients.20 “Normal” anatomy is considered to be a common hepatic duct bifurcating into left and right hepatic ducts, and the right duct subsequently bifurcating into anterior and posterior segmental ducts. A trifurcation of the common hepatic duct into left, right anterior segmental, and right posterior segmental ducts is observed in 12%. Other common variations include origin of the right

CHAPTER 139 BILIARY COMPLICATIONS ASSOCIATED WITH LIVER TRANSPLANTATION 1021 anterior or posterior segmental ducts from either the common hepatic duct or the left hepatic duct. Rarely, the right posterior or anterior segmental duct joins the cystic duct prior to its entry into the extrahepatic bile duct. When variant biliary anatomy is present in the right lobe of the donor organ and it is desirable to create anastomoses to the recipient bile duct, unconventional anastomoses may be required, including the donor segmental right hepatic duct anastomosed to the recipient’s left hepatic duct or cystic duct.21,22 As an alternative, separate biliary-enteric anastomoses to a Roux-en-Y loop can be created. In some cases, the segmental duct is ligated, and that segment of the liver subsequently atrophies. When living donor transplantation is being considered, presurgical evaluation of the potential donor is performed so that the biliary and vascular anatomy can be delineated prior to transplantation. Accurate assessment of the biliary anatomy is critical in living donor transplantation because a biliary injury potentially has great significance. Complications related to the biliary tree are encountered in 4.4% of donor hepatectomies.23 The optimal method for preoperative anatomic evaluation of potential living hepatic donors remains to be defined. The ideal technique would clearly delineate the hepatic arteries, portal veins, hepatic veins, bile ducts, and hepatic parenchyma without requiring contrast media or ionizing radiation. Although magnetic resonance imaging (MRI) and MR angiography (MRA) potentially fulfill some of these requirements, delineation of small hepatic arteries and diminutive biliary radicles can be problematic.24 Threedimensional reconstruction renderings will usually provide adequate information for surgical planning. A more detailed biliary evaluation would be advantageous, but the risks associated with retrograde cholangiography are considered excessive for routine evaluation of a healthy living donor.

BILIARY COMPLICATIONS The surgical anastomosis between the donor biliary tract and the recipient may be technically difficult to create. Moreover, the biliary system is sensitive to ischemia and may be damaged either by prolonged preservation before transplantation or by problems with either hepatic arterial and possibly portal venous flow.25 Despite refinements in surgical technique, biliary complications continue to be a common source of morbidity in the transplant patient. In a systematic review of 14,359 transplants, biliary complications were observed in 12% of deceased donor recipients and 19% of live donor grafts.11 Two major types of complications related to the biliary tract are observed: bile leak and biliary obstruction.

Bile Leak Bile leaks usually occur early in the postoperative period and originate from three sources: the T-tube entry site (in patients who have had T-tubes placed), the biliary anastomosis, or the cut surface of the liver (in patients with split or reduced-size allografts). In a patient with a CDCD anastomosis performed over a T-tube, seepage through the T-tube choledochostomy is the most common cause of a bile leak.26 This type of leak is

usually identified on routine postoperative T-tube cholangiography. Initial treatment consists of opening the T-tube to gravity drainage, which is curative in up to 60% of patients.27 If this is unsuccessful, endoscopic sphincterotomy may be useful. Bile leaks may also develop when the T-tube is removed 3 to 6 months after the transplant operation. Immunosuppression often prevents formation of a connective tissue tract to allow bile to drain freely into the peritoneal cavity. Retrograde stent placement or percutaneous transhepatic drainage coupled with percutaneous drainage of the biloma controls the leak in most patients. However, up to a third of these patients require operative intervention with surgical revision for closure.28 Because of the aforementioned problems related to T-tubes, some centers have elected to perform the CDCD anastomosis without T-tube insertion. In a study by Randall et al.,29 a group of 59 patients who had T-tubes placed were compared with 51 patients who underwent CDCD anastomoses without T-tubes. This comparison revealed no difference in biliary complication rates or survival. Bile leaks that occur at the anastomotic site (Fig. 139-1) may be difficult to manage conservatively. The principles of nonoperative management of bile leaks include (1) exclusion of hepatic artery thrombosis or stenosis, (2) drainage of any associated bilomas or abscesses, (3) diversion of bile flow, (4) eradication of any coexistent infection, and (5) optimization of the nutritional status of the patient. Patency of the hepatic artery should be confirmed in all patients with a bilious postoperative collection. This is usually accomplished with duplex ultrasound though MRA, or CTA may be used when the ultrasound examination is inconclusive. When hepatic artery thrombosis (HAT) is detected, retransplantation may be required. Kaplan et al.30 reported percutaneous drainage in 15 patients with hepatic artery occlusion; 11 required retransplantation, and 4 were managed for more than 30 months with indwelling drainage catheters. Drainage of infected or uninfected bilomas is critical to percutaneous management. Healing requires the area of dehiscence to be invested with healthy connective tissue to provide blood flow as well as structural support. Evacuation of infected fluid promotes ingrowth of healthy tissue. Management of bilomas in a transplant patient is identical to that in a nontransplant patient, and consists of imageguided percutaneous drainage for accessible collections in the majority of patients. The most common area in which a biloma develops is adjacent to the extrahepatic biliary system in the subhepatic space. However, the bile may track and accumulate some distance from the site of leakage. The right subphrenic space is not an uncommon site, particularly if the size of the allograft is poorly matched to the configuration of the right subphrenic space. The right subphrenic space offers unique challenges for percutaneous drainage because of its location. Frequently a choice must be made between a transhepatic approach, which can potentially lead to damage to the allograft, and a transpleural approach, which may potentially contaminate the pleural space. Overall, percutaneous drainage of postoperative bile collections in a liver transplant patient carries high success and low complication rates.31 After drainage of any associated bilomas, bile flow should be diverted. In so doing, the amount of bile effluent through

1022 PART 2 NONVASCULAR INTERVENTIONS · SECTION 19 THE BILIARY TRACT

A

B

D

E

C

F

FIGURE 139-1. This 60-year-old man with cryptogenic cirrhosis underwent orthotopic liver transplantation. Two weeks after surgery, his alkaline phosphatase level was elevated, and he underwent endoscopic retrograde cholangiopancreatography (ERCP) for biliary tract evaluation. A, Endoscopic injection of contrast material into common bile duct demonstrates a duct-to-duct anastomotic leak. Note duct-to-duct anastomosis (black arrow) and collection of contrast material in subhepatic space (white arrow). B, Patient was treated by placement of an endoscopic stent (arrow) across anastomosis to facilitate drainage. C, Computed tomography (CT) performed after ERCP demonstrated extravasation of ERCP contrast material into subhepatic space (arrow). D, Also noted on CT were large subphrenic (arrow) and subhepatic fluid collections. E, Percutaneous drainage of bilomas was accomplished with CT guidance and placement of two 10F pigtail drains. One drain was inserted into subhepatic space (arrow). F, Another drain was placed in subphrenic space (arrow).

the defect can be diminished, thereby improving the chance of spontaneous closure. Diversion of bile flow can be accomplished by either placement of an endoscopically guided retrograde stent (if a CDCD anastomosis has been created) or by percutaneous transhepatic biliary drainage (PTBD). Retrograde endoscopic drainage via retrograde stent placement has the major advantage of not requiring a transhepatic puncture and an external drainage tube. PTBD has two advantages over endoscopically guided internal stent placement: the ability to provide both internal and external drainage and the ease with which cholangiography can be repeated. The ability to provide external drainage allows more efficient evacuation of bile. Both character and amount of bile output can be easily monitored. However, if internal drainage is deemed appropriate, it can be accomplished by merely capping the internal-external drainage tube, provided that the catheter has been positioned across the area of leakage into the bowel. The tube is not usually capped until a cholangiogram no longer demonstrates the presence of a leak. Moreover, monitoring the healing process by tube cholangiography avoids the necessity of sedation and an endoscopic procedure. Unfortunately,

performance of transhepatic biliary drainage requires the risk and discomfort of a transhepatic puncture and may be technically challenging in patients with nondilated intrahepatic ducts. For these reasons, the endoscopic retrograde approach is favored in many centers for patients with CDCD anastomoses to achieve biliary drainage. Patients with CDJ anastomoses require PTBD because it is usually impossible to reach the biliary anastomosis with an endoscope. Whatever method of biliary diversion is selected, the final two principles in the management of anastomotic leaks have to be followed: eradication of infection and maintenance of nutrition. The former can be accomplished by adequate drainage and appropriate antibiotic therapy. The latter may require enteral or parenteral hyperalimentation to maintain a positive nitrogen balance. With these principles, it may be possible to treat anastomotic leaks nonoperatively. Osorio et al.15 reported successful healing of 14 of 17 CDCD anastomotic leaks; 3 of 17 CDCD anastomoses that were entirely dehiscent required operative revision. In this same series, only 3 of 9 bile leaks originating from CDJ anastomoses healed with conservative therapy.

CHAPTER 139 BILIARY COMPLICATIONS ASSOCIATED WITH LIVER TRANSPLANTATION 1023 This resistance to healing with conservative therapy may in part be a reflection of the immunosuppressive corticosteroids these patients are taking, which inhibits the healing response. In our institution, a short trial of either PTBD or retrograde stent placement is often attempted, but surgical revision is performed if the perforation does not seal rapidly. That said, many patients have comorbid conditions that make reoperation an unattractive therapeutic option. Patients with split-liver transplants (either deceased or living donor) deserve special consideration.32-34 The technical aspects of creating a biliary anastomosis are more challenging in this group. Cadaveric split livers generally contain an ample amount of extrahepatic bile duct when the right lobe is harvested. To prevent a bile leak, care must be taken to ligate the left hepatic duct origin at the point where it is divided from the confluence of the hepatic ducts. The left lobe from the same donor may be used for pediatric transplantation, with anastomosis of the left hepatic duct to a Roux-en-Y loop. The technical demands of creating a biliary anastomosis to the right hepatic duct in patients receiving an allograft from a living donor are even greater because of the short length of right hepatic duct available for creation of the anastomosis. These anastomoses may be created under tension, which can lead to focal necrosis and development of a bile leak. Because of these anatomic restrictions, biliary complications are more frequent after living donor right lobe transplantation, with bile leakage being observed in 4.7% to 18.2% of patients.35-37 Because of the presence of the biliary anastomosis cephalically within the porta hepatis, these patients are extremely difficult to surgically revise. When anastomotic leaks occur, an extensive trial of nonoperative therapy before surgical revision is usually attempted. When transhepatic cholangiography and biliary drainage are performed on patients who are recipients of a right lobe from a split-liver transplant, detailed knowledge of the postoperative anatomy is required. Because of the variable segmental ductal anatomy of the right hepatic ducts, it is sometimes necessary to ligate a substantial-sized duct such as the right anterior or right posterior segmental duct. When this is performed in a sterile environment, the ducts dilate and the parenchyma atrophies. However, if these ducts are punctured and a tube is placed, it is not usually possible to reconnect to a biliary system that has been anastomosed to the recipient’s native duct (in the case of a ductto-duct anastomosis) or to the jejunum (in the case of a hepaticojejunal anastomosis). The punctured ligated duct may leak if the needle is removed. Alternatively, if a tube is placed, it is likely that the external drainage tube must remain indefinitely, probably for the remainder of the patient’s life. Therefore, the operative report must be carefully scrutinized and all previous imaging studies reviewed before attempting a diagnostic or therapeutic intervention on a biliary system that has been anastomosed to the recipient. MR cholangiography may be very useful before percutaneous interventions to delineate the appearance of both the ligated and anastomosed systems when a donor system has been intentionally occluded. Split-liver allografts may also leak from the cut edge of the liver (Fig. 139-2). Because surgical division of the right and left lobe occurs along the plane of the middle hepatic vein, small ductal radicles are divided, ligated, and cauterized in the course of the dissection. Occlusion of these

radicles is sometimes incomplete, so bile leakage may occur from this cut surface. In a series of 72 patients reported by Wojcicki et al.,38 bile leakage from the hepatic parenchyma was the most common biliary complication, observed in 11% of patients. Leaks from the cut edge of the liver often resolve spontaneously with simple drainage of the leak. If the leak continues, drainage of the biliary tract from either an antegrade or retrograde approach may be required. In rare circumstances, it may be necessary to occlude the offending radicle. This can be accomplished with fluoroscopically guided catheter techniques and 2,3-isobutyl cyanoacrylate as the occlusive agent.

Biliary Obstruction Biliary obstruction represents the other major complication related to the biliary tract after orthotopic liver transplantation. This complication is detected to a varying degree in 10% to 20% of patients.11,39 The most common problem is a stricture at the biliary anastomosis (Fig. 139-3). Clinical manifestations of biliary obstruction may be nonspecific. Liver transplant recipients have many potential causes of fever, abnormal liver function tests, and leukocytosis. Biliary obstruction is often just one of the potential explanations. Noninvasive evaluation for biliary obstruction with ultrasound, CT, and MRI may be difficult because patients may not have ductal dilation.40 Therefore, direct cholangiography may be necessary to exclude or confirm the diagnosis. In a patient with a CDCD anastomosis, endoscopic retrograde cholangiography is usually performed. In a patient with a CDJ anastomosis, percutaneous transhepatic cholangiography is the appropriate examination.41 When stenosis of a CDCD anastomosis is identified in the early postoperative period, it may merely represent edema or an inflammatory reaction (or both) at the anastomotic site. In this circumstance, a short period (2-6 weeks) of endoscopic stenting may be curative.42 If endoscopic cannulation cannot be achieved, a transhepatic internalexternal biliary drainage tube should be placed. However, if the stenosis persists or recurs, surgical revision to a CDJ is often the most efficient and durable solution. As an alternative to surgical reconstruction, balloon dilation via the endoscopic retrograde approach may be attempted, followed by a variable period of retrograde stenting. With this strategy, success rates ranging from 64% to 90% have been reported.43-45 When retrograde access to the biliary tree cannot be achieved, percutaneous transhepatic dilation may be attempted (Fig. 139-4). This is accomplished technically by using a standard transhepatic approach to gain access to the intrahepatic biliary system. A sheath large enough to allow passage of the appropriate-sized balloon is inserted. Using a small curved catheter and a probing guidewire, the stenotic anastomosis is crossed. The guidewire is advanced into the duodenum for patients with a CDCD anastomosis or into the Roux-en-Y loop for a patient with a CDJ. A balloon size is selected that exceeds the normal ductal caliber by approximately 10%. This balloon is inflated across the anastomosis for a variable period. Some investigators recommend prolonged balloon inflation (for periods of up to 20 minutes) and repeating the procedure two to three times within 2 weeks to achieve a durable result. Using this technique with balloons ranging from 7 to 14 mm in

1024 PART 2 NONVASCULAR INTERVENTIONS · SECTION 19 THE BILIARY TRACT

A

D

B

E

C

F

FIGURE 139-2. This 20-year-old woman with primary sclerosing cholangitis underwent liver transplantation with a split liver (she received right lobe and medial segment of left). Postoperatively, abdominal pain and fever developed. A, Initial percutaneous transhepatic cholangiography demonstrates mild biliary ductal dilation. B, After placement of a drainage catheter into bile duct, injection of contrast material reveals leak at cut edge of liver, with extravasated contrast material filling a small biloma cavity (arrow). Note presence of an operative drain in region of leak. C, A percutaneous internal-external biliary drain was inserted into biliary tree. D, Surgical drain was then exchanged for a percutaneous pigtail drain (arrow). E, After several weeks of percutaneous drainage, biloma drain was removed. A small residual leak was noted at cut edge of liver (arrow). F, Three months after transplantation, leak was healed. Injection of biliary drain did not demonstrate a leak, and biliary drainage catheter was removed.

diameter, Zajko et al.46 reported a 73% success rate at 2 years and a 66% success rate at 6 years in a series of 56 patients with anastomotic strictures. However, use of conventional dilation balloons for treatment of anastomotic strictures has not been uniformly successful. For this reason, some practitioners47 have advocated using a cutting balloon (Boston Scientific, Natick, Mass.). This device has four surgical blades embedded in the dilating balloon. During balloon inflation, the blades incise the dense fibrous tissue of the stricture in the hope of preventing elastic recoil and achieving a more durable result of the dilation. Frequently, a cutting balloon smaller than the anticipated necessary diameter is selected to achieve complete dilation, and a conventional balloon is then used to enlarge the lumen to the desired size. Using this technique, Saad et al.47 treated 22 patients with posttransplant anastomotic strictures during 49 cutting balloon sessions, usually combined with conventional balloon dilation. A technical

success rate of 93% was achieved without incurring a major complication. Long-term durability of this strategy in comparison to conventional balloon dilation has not been assessed. Before the advent of the cutting balloon, when conventional dilation of an anastomotic stricture failed in a patient who was considered to be a poor surgical risk, metallic stent placement was occasionally performed.48 Unfortunately, the rate of recurrent stenosis and formation of obstructing debris in the biliary tree was extremely high, and most centers have abandoned this therapeutic alternative. Moreover, it should be noted that stent placement can potentially preclude or complicate definitive surgical repair. To circumvent this problem, the concept of retrievable stent-graft placement was put forth by Petersen et al.49 In this series, Gianturco-Rosch Z-stents were covered with expanded polytetrafluoroethylene (ePTFE) and inserted for a period of 2 to 9 months. All patients had the stents successfully

CHAPTER 139 BILIARY COMPLICATIONS ASSOCIATED WITH LIVER TRANSPLANTATION 1025

A

B

C

FIGURE 139-3. Biliary strictures after liver transplantation. A, Duct-to-duct anastomotic stricture (arrow). Such strictures are often treated by operative repair with a Roux anastomosis. Nonoperative candidates can be treated by balloon dilation. B, In right lobe living donor or split-liver transplants, a high stricture can develop in hilum at operative site (arrow). C, Stricture can develop at enteric anastomosis (arrow) in patients who receive a Roux-en-Y. These strictures are often treated by balloon dilation because reoperation is difficult.

A

B

C

FIGURE 139-4. Stricture at hepaticojejunal anastomosis developed in this 38-year-old man who underwent orthotopic liver transplantation for sclerosing cholangitis. A, Tube cholangiogram reveals high-grade narrowing at surgical anastomosis. B, A 10 mm × 4 cm dilation balloon is inflated across anastomosis. C, Repeat cholangiogram after dilation shows good cosmetic result.

retrieved. Five of six patients with anastomotic strictures (2 CDCD, 4 CDJ) had successful outcomes during the 6- to 20-month follow-up period. Unfortunately, since no commercial devices have been created for this technique, there have been no further reports in over a decade. Special consideration must be given to patients with biliary obstruction who are recipients of split-liver transplants. Left duct anastomoses in pediatric patients receiving a left lobe transplant tend to not pose a technical problem, because the length of the left hepatic duct is longer and thus allows a relatively straightforward anastomosis, usually to a Roux-en-Y loop. Right lobe transplantation is not problematic when a cadaveric organ is used, because the extrahepatic bile duct is also harvested. However, when a living donor is used, the procedure is considerably more difficult because of the short length of

the right hepatic duct that is available. Not only is the duct small, but the arterial supply may be compromised during the donation right hepatectomy.50 It is not surprising that biliary obstructions complicate 8.3% to 31.7% of right lobe transplants.51 Multiple types of biliary anastomoses can be created with a donated right lobe. A direct anastomosis to the recipient’s common hepatic duct or right hepatic duct is usually performed if a single right donor duct drains both the anterior and posterior segments. If the donor biliary anatomy has the posterior and anterior segments draining into variant locations, such as the left hepatic duct, separate anastomoses may be required. If a Roux-en-Y loop is constructed, it is generally straightforward to anastomose each segmental duct into the loop. However, if a duct-to-duct anastomosis is desired, the ducts may be anastomosed to

1026 PART 2 NONVASCULAR INTERVENTIONS · SECTION 19 THE BILIARY TRACT

A

B

FIGURE 139-5. Stricture of proximal common hepatic duct following right lobe liver transplant. A, Tube cholangiogram reveals diffuse narrowing of hepaticojejunostomy. B, A 14F catheter with a coaxial 8F catheter exiting a proximal side hole bridges stenotic duct.

the recipient’s right and left hepatic ducts or even to the recipient’s right or common hepatic duct and cystic duct stump. Strictures that occur in living donor recipients are managed in a similar fashion to patients with cadaveric liver transplants. If the anastomosis is created to the recipient’s bile duct, retrograde cholangiography and therapy is generally performed. When the anastomosis is created to a Rouxen-Y loop, the percutaneous transhepatic route is required. Conventional balloon dilatation and cutting balloons followed by a variable period of transhepatic internal-external stenting is usually performed. An innovative method of transhepatic management of biliary strictures following living donor transplant has been recommended by Gwon et al.52 These investigators use a 14F biliary drainage tube through which an 8F catheter is coaxially inserted, exiting a side hole proximally and traversing the stricture (Fig. 139-5). By using this configuration, a 22F channel is created across the strictured area. With this technique, primary patency rates of 96%, 92%, and 91% were reported at 1, 2, and 3 years, respectively, in a series of 79 patients.46 An additional novel technique in managing biliary strictures was reported by Itoi et al.53 These authors inserted a samarium-cobalt magnet from a transhepatic approach with an additional magnet inserted from the retrograde endoscopic approach. The compressive force created through the magnetic attraction eradicated the stricture and created a new anastomosis that remained patent following removal of the magnets. In some circumstances, segmental ductal systems may be intentionally divided and ligated, which leads to gradual segmental atrophy. If postoperative obstruction develops in such patients, retrograde cholangiography may be performed in the usual manner, and the ligated system will not be opacified. However, if a percutaneous cholangiogram and drainage are necessary, it is crucial to be aware of the ligated system before embarking on the procedure. If a tube is inadvertently placed into a ligated system, the patient may

require an external drainage tube for the remainder of life or a major surgical procedure to rectify the situation. It is prudent to obtain an MR or CT cholangiogram before percutaneous interventions on these patients so that intentionally ligated systems can be avoided during percutaneous interventions. Biliary strictures unrelated to the surgical anastomosis may also develop.54 Nonanastomotic strictures related to hepatic artery thrombosis, prolonged cold preservation times, acute or chronic rejection, and cytomegalovirus infection may be encountered. In addition, recurrent intrahepatic biliary strictures develop in 5% to 20% of patients undergoing transplantation for primary sclerosing cholangitis.55 Because there are no surgical options other than retransplantation for this group of patients, balloon dilation has formed the mainstay of clinical management when symptoms related to such narrowing develop. In the previously cited Zajko study,46 a 94% success rate at 2 years and 84% at 5 years was reported with the use of balloon dilation in a group of 16 patients with non-anastomotic strictures. As with anastomotic strictures, metallic stents have also been used rarely in the management of non-anastomotic strictures that fail to respond to balloon dilation. Before placement of a metallic stent, a trial of cutting balloon cholangioplasty should be considered. Although the immediate result of metallic stent placement is often gratifying, there is a tendency for these stents to serve as a nidus for stone and debris formation. The long-term durability of stents in these patients has not been thoroughly evaluated, and such stents should be considered only when other therapeutic alternatives, including prolonged tube drainage, have been exhausted. In a patient who has had the biliary anastomosis performed with placement of a T-tube, the tube may become occluded with debris or the limb displaced, with subsequent biliary obstruction. It is often possible to rectify T-tube problems via interventional radiologic techniques. However, because transplant patients form mature tracts at a much

CHAPTER 139 BILIARY COMPLICATIONS ASSOCIATED WITH LIVER TRANSPLANTATION 1027 slower rate than patients who are not immunosuppressed, simple replacement is sometimes not possible. If access is lost during attempted revision of a T-tube, it may be necessary to perform an additional transhepatic or retrograde procedure to ensure adequate biliary drainage. Mucocele of the donor cystic duct remnant is an unusual but well-recognized cause of biliary obstruction.56 Although percutaneous aspiration of a cystic duct mucocele may provide temporary relief of obstruction,57 surgical revision is usually required for definitive therapy. Dysfunction of the sphincter of Oddi in a patient with a CDCD anastomosis can also inhibit bile flow.58 Even though the cause of this disorder is not clearly defined, its occurrence is well recognized. Endoscopically guided sphincterotomy is curative. Obstruction of the bile duct by stones, sludge, or necrotic debris may also be encountered. Although stones may develop in any patient with biliary obstruction, the presence of necrotic debris usually indicates a prolonged cold preservation injury, rejection, or hepatic arterial insufficiency. Stones and debris can generally be treated with percutaneous transhepatic or endoscopic retrograde techniques, including basket retrieval and balloon sweeping.59 However, unless the underlying cause is addressed, recurrent obstruction from newly formed material is likely to occur.

KEY POINTS •

The most common complications after orthotopic liver transplantation are related to the biliary tract. These complications can be divided into two major groups: leak and obstruction.

•

Bile leaks develop in the early postoperative period and can often be managed by draining the associated biloma in conjunction with controlling the drainage of bile from the liver with either a retrograde stent or a percutaneous transhepatic biliary drainage tube.

•

Biliary obstruction usually develops more than 1 month after transplant surgery and can be managed in many cases by balloon dilatation and stenting.

•

Special challenges are posed in patients receiving split organs, with right lobe transplants being most common in the adult population.

▶ SUGGESTED READINGS Akamatsu N, Sugawara Y, Hashimoto D. Biliary reconstruction, its complications and management of biliary complications after adult transplantation: a systematic review of the incidence, risk factors and outcome. Transpl Int 2011;24:379–92. Gwon DI, Ko GY, Sung KB, et al. A novel double stent system for palliative treatment of malignant extrahepatic biliary obstructions: a pilot study. AJR Am J Roentgenol 2011;197(5):W942–7. Perrakis A, Fortsch T, Schellerer V, et al. Biliary tract complications after orthotopic liver transplantation: still the “Achilles heel”? Transpl Proc 2010;10:4154–7. Piardi T, Greget M, Audet M, et al. Biliary strictures after liver transplantation: is percutaneous treatment indicated? Ann Transplant 2011;16: 5–13.

The complete reference list is available online at www.expertconsult.com.

CHAPTER 139 BILIARY COMPLICATIONS ASSOCIATED WITH LIVER TRANSPLANTATION1027.e1 REFERENCES 1. Kotlyar DS, Campbell MS, Reddy KR. Recurrence of diseases following orthotopic liver transplantation. Am J Gastroenterol 2006;101:1370–8. 2. Said A, Lucey MR. Liver transplantation: an update. Curr Opin Gastroenterol 2006;22:272–8. 3. Jacob M, Lewsey JD, Sharpin C, et al. Systematic review and validation of prognostic models in liver transplantation. Liver Transpl 2005; 11:814–25. 4. Wilms C, Walter J, Kaptein M, et al. Long-term outcome of split liver transplantation using right extended grafts in adulthood: a matched pair analysis. Ann Surg 2006;244:865–73. 5. Yersiz H, Cameron AM, Carmody I, et al. Split liver transplantation. Transplant Proc 2006;38:602–3. 6. Kim WR, Therneau TM, Benson JT, et al. Deaths on the liver transplant waiting list: an analysis of competing risks. Hepatology 2006; 43:345–51. 7. Wiesner RH. Patient selection in an era of donor liver shortage: current US policy. Nat Clin Pract Gastroenterol Hepatol 2005;2:24–30. 8. Florman S, Miller CM. Live donor liver transplantation. Liver Transpl 2006;12:499–510. 9. Hampe T, Dogan A, Encke J, et al. Biliary complications after liver transplantation. Clin Transplant 2006;20(Suppl 17):93–6. 10. Starzl TE, Ishikawa M, Putnam CW, et al. Progress in and deterrents to orthotopic liver transplantation with special reference to survival, resistance to hyperacute rejection, and biliary duct reconstruction. Transplant Proc 1974;6:129–39. 11. Akamatsu N, Sugawara Y, Hashimoto D. Biliary reconstruction, its complications and management of biliary complications after adult transplantation: a systematic review of the incidence, risk factors and outcome. Transpl Int 2011;24:379–92. 12. Perrakis A, Fortsch T, Schellerer V, et al. Biliary tract complications after orthotopic liver transplantation: still the “Achilles heel”? Transpl Proc 2010;10:4154–7. 13. Schweizer RT, Sutphin B, Bartus SA. Biliary reconstruction methods for liver transplantation. Transplant Proc 1981;13:842–4. 14. Alsharabi A, Zieniewicz K, Patkowski W, et al. Assessment of early biliary complications after orthotopic liver transplantation and their relationship to the technique of biliary reconstruction. Transplant Proc 2006;38:244–6. 15. Osorio RW, Freise CE, Stock PG, et al. Nonoperative management of biliary leaks after orthotopic liver transplantation. Transplantation 1993;55:1074–7. 16. Angel-Mercado M, Chan C, Orozco H, et al. Bile duct injuries related to misplacement of “T tubes.” Ann Hepatol 2006;5:44–8. 17. Amador A, Charco R, Marti J, et al. Cost/efficacy clinical trial about the use of T-tube in cadaveric donor liver transplant: preliminary results. Transplant Proc 2005;37:1129–30. 18. Huang WD, Jiang JK, Lu YQ. Value of T-tube in biliary tract reconstruction during orthotopic liver transplantation. J Zhejiang Univ Sci B 2011;12:357–64. 19. Valera-Sanchez Z, Flores-Cortes M, Romero-Vargas ME, et al. Biliodigestive anastomosis in liver transplantation: review of 13 years. Transplant Proc 2006;38:2471–2. 20. Smadia C, Blumgart LH. The biliary tract and the anatomy of biliary exposure. In: Blumgart LH, editor. Surgery of the Liver and Biliary Tract, vol 1, 2nd ed. Edinburgh: Churchill Livingstone; 1994. p. 11–24. 21. Soejima Y, Taketomi A, Yoshizumi T, et al. Biliary strictures in living donor liver transplantation: incidence, management, and technical evolution. Liver Transpl 2006;12:979–86. 22. Asonuma K, Okajima H, Ueno M, et al. Feasibility of using the cystic duct for biliary reconstruction in right-lobe living donor liver transplantation. Liver Transpl 2005;11:1431–4.

23. Yaprak O, Dayangac M, Demirbas BT, et al. Analysis of right lobe living-living donor complications: a single center experience. Exp Clin Transplant 2011;9:56–9. 24. Hsu HW, Tsang LL, Yap A, et al. Magnetic resonance cholangiography in living donor transplantation. Transplantation 2011;92:94–9. 25. Farid WR, de Jonge J, Slicker JC, et al. The importance of portal venous blood flow in ischemic-type biliary injuries after transplantation. Am J Transplant 2011;11:857–62. 26. Zajko AB, Campbell WL, Bron KM, et al. Diagnostic and interventional radiology in liver transplantation. Gastroenterol Clin North Am 1988;17:105–43. 27. Sheng R, Sammon JK, Zajko AB, et al. Bile leak after hepatic transplantation: cholangiographic features, prevalence, and clinical outcome. Radiology 1994;192:413–6. 28. Ward EM, Wiesner RH, Hughes RW, et al. Persistent bile leak after liver transplantation: biloma drainage and endoscopic retrograde cholangiopancreatographic sphincterotomy. Radiology 1991;179: 719–20. 29. Randall HB, Wach ME, Somberg KA, et al. The use of T tube after orthotopic liver transplantation. Transplantation 1996;61:258–61. 30. Kaplan SB, Zajko AB, Koneru B. Hepatic bilomas due to hepatic artery thrombosis in liver transplant recipients: percutaneous drainage and clinical outcome. Radiology 1990;174:1031–5. 31. Chang YF, Chen YS, Huang TL, et al. Interventional radiologic procedures in liver transplantation. Transpl Int 2001;14:223–9. 32. Wojcicki M, Silva MA, Jethwa P, et al. Biliary complications following adult right lobe ex vivo split liver transplantation. Liver Transpl 2006;12:839–44. 33. Yazumi S, Chiba T. Biliary complications after a right-lobe living donor liver transplantation. J Gastroenterol 2005;40:861–5. 34. Liu CL, Lo CM, Chan SC, Fan ST. Safety of duct-to-duct biliary reconstruction in right-lobe live-donor liver transplantation without biliary drainage. Transplantation 2004;77:726–32. 35. Gondolesi GE, Varotti G, Florman AA, et al. Biliary complications in 96 consecutive right lobe living donor transplant recipients. Transplantation 2004;77:1842–8. 36. Hwang S, Lee S-G, Sung K-B, et al. Long-term incidence, risk factors, and management of biliary complications after adult living donor transplantation. Liver Transpl 2006;12:831–8. 37. Franco J. Biliary complications in liver transplant recipients. Curr Gastroenterol Rep 2005;7:160–4. 38. Wojcicki M, Silva MA, Jethwa P, et al. Biliary complications following adult right lobe ex vivo split liver transplantation. Liver Transpl 2006;12:839–44. 39. Qian YB, Liu CL, Lo CM, Fan ST. Risk factors for biliary complications after liver transplantation. Arch Surg 2004;139: 1101–5. 40. Zemel G, Zajko AB, Skolnick ML, et al. The role of sonography and hepatic cholangiography in the diagnosis of biliary complications after liver transplantation. AJR Am J Roentgenol 1988;151:943–6. 41. Piardi T, Greget M, Audet M, et al. Biliary strictures after liver transplantation: is percutaneous treatment indicated? Ann Transplant 2011;16:5–13. 42. Ryu CH, Lee SK. Biliary strictures after transplantation. Gut Liver 2011;5:133–42. 43. Qin YS, Li ZS, Sun ZX, et al. Endoscopic management of biliary complications after orthotopic liver transplantation. Hepatobiliary Pancreat Dis Int 2006;5:39–42. 44. Morelli J, Mulcahy HE, Willner IR, et al. Long-term outcomes for patients with post–liver transplant anastomotic biliary strictures treated by endoscopic stent placement. Gastrointest Endosc 2003; 58:374–9. 45. Chahin NJ, De Carlis L, Slim AO, et al. Long-term efficacy of endoscopic stenting in patients with stricture of the biliary anastomosis

1027.e2PART 2 NONVASCULAR INTERVENTIONS · SECTION 19 THE BILIARY TRACT after orthotopic liver transplantation. Transplant Proc 2001;33: 2738–40. 46. Zajko AB, Sheng R, Zetti GM, et al. Transhepatic balloon dilation of biliary strictures in liver transplant patients: a 10-year experience. J Vasc Interv Radiol 1995;6:79–83. 47. Saad WE, Davies MG, Saad NE, et al. Transhepatic dilation of anastomotic biliary strictures in liver transplant recipients with use of a combined cutting and conventional balloon protocol: technical safety and efficacy. J Vasc Interv Radiol 2006;17:837–43. 48. Culp WC, McCowan TC, Lieberman RP, et al. Treatment of biliary strictures with metallic stents in liver transplant recipients. J Vasc Interv Radiol 1996;7:457–8. 49. Petersen B, Timmermans HA, Uchida BT, et al. Treatment of refractory benign biliary stenoses in liver transplant patients by placement and retrieval of a temporary stent-graft: work in progress. J Vasc Interv Radiol 2000;11:919–29. 50. Wang SF, Huang ZY, Chen XP. Biliary complications after living donor transplantation. Liver Transpl 2011;17:1127–36. 51. Olthoff KM, Merion RM, Ghobrial RM, et al. Outcomes of 385 adultto-adult living donor liver transplant recipients. Ann Surg 2005; 242:314–25.

52. Gwon DI, Sung KB, Ko GY, et al. Dual catheter placement technique for treatment of biliary anastomotic strictures after liver transplantation. Liver Transpl 2011;17:159–66. 53. Itoi T, Yamanouchi E, Ikeuchi N, et al. Magnetic compression duct-to duct anastomosis for biliary obstruction in a patient with living donor liver transplantation. Gut Liver 2010;4(suppl 1):S96–8. 54. Cameron AM, Busuttil RW. Ischemic cholangiopathy after liver transplantation. Hepatobiliary Pancreat Dis Int 2005;4:495–501. 55. Graziadei IW. Recurrence of primary sclerosing cholangitis after liver transplantation. Liver Transpl 2002;8:575–81. 56. Zajko AB, Bennett MJ, Campbell WL, Koneru B. Mucocele of the cystic duct remnant in eight liver transplant recipients: findings at cholangiography, CT, and US. Radiology 1990;177:691–3. 57. Abcarian PW, Emond JC, Ring EJ. Cystic duct remnant mucocele in a liver transplant recipient. J Vasc Interv Radiol 1994;5:127–30. 58. Douzdjian V, Abecassis MM, Johlin FC. Sphincter of Oddi dysfunction following liver transplantation. Screening by bedside manometry and definitive manometric evaluation. Dig Dis Sci 1994;39:253–6. 59. Canete JJ, Aidlen JT, Uknis ME, Cicalese L. Images of interest. Hepatobiliary and pancreatic: biliary cast syndrome. J Gastroenterol Hepatol 2005;20:791.

SECTION TWENTY

GENERAL CONSIDERATIONS IN ENERGY-BASED ABLATION CHAPTER

140

Image-Guided Thermal Tumor Ablation: Basic Science and Combination Therapies Muneeb Ahmed and S. Nahum Goldberg

UNDERSTANDING TUMOR ABLATION: AN OVERVIEW Image-guided tumor ablation is a minimally invasive strategy to treat focal tumors by inducing irreversible cellular injury through application of thermal and, more recently, nonthermal energy or chemical injection. This approach has become a widely accepted technique, incorporated into the treatment of a range of clinical circumstances, including treatment of focal tumors in the liver, lung, kidney, bone, and adrenal glands.1-4 Benefits of minimally invasive therapies compared to surgical resection include lesser mortality and morbidity, lower cost, and the ability to perform procedures on outpatients who are not good candidates for surgery.5 However, additional work is required to improve outcomes and overcome limitations in ablative efficacy, including persistent growth of residual tumor at the ablation margin, the inability to effectively treat larger tumors, and variability in complete treatment based upon tumor location.1 Given the multiplicity of treatment types and potential complexity of paradigms in oncology and the wider application of thermal ablation techniques, a thorough understanding of the basic principles and recent advances in thermal ablation is a necessary prerequisite for their effective clinical use. These key concepts related to tumor ablation can be broadly divided into (1) those that relate to performing a clinical ablation, such as understanding the goals of therapy and mechanisms of tissue heating or tumor destruction, and (2) understanding the proper role of tumor ablation and the strategies being pursued to improve overall ablation outcome. These latter concepts include a systematic approach to technologic development, understanding and using the biophysiologic environment to maximize ablation outcome, combining tumor ablation with adjuvant therapies to synergistically increase tumor destruction, and improving tumor visualization and targeting through image navigation and fusion technology. Given that the greatest experimental and clinical experience exists for radiofrequency (RF)-based ablation, this will be the representative model used to discuss many of these concepts. However, many of these principles also apply when using alternative ablative modalities.

Goals of Minimally Invasive Therapy The overall goal of minimally invasive focal tumor ablation encompasses several specific aims. First and foremost, a successful ablation completely treats all malignant cells 1028

within the target. This includes both achieving homogenous ablation within the tumor (so no focal areas of viable tumor remain) and adequately treating a rim of apparently “normal” tissue around the tumor, which often contains microscopic invasion of malignant cells at the tumor periphery. Thus, based upon examinations of tumor progression in patients undergoing surgical resection and the demonstration of viable malignant cells beyond visible tumor boundaries, in most cases (or except when otherwise indicated) tumor ablation therapies attempt to include at least a 0.5- to 1-cm “ablative margin” of seemingly normal tissue for liver and lung, though less may be needed for some tumors in the kidney.6,7 A secondary goal is to ensure accuracy of therapy to minimize the damage to nontarget normal tissue surrounding the ablative zone. As such, one significant advantage of percutaneous ablative therapies over conventional standard surgical resection is the potential to remove or destroy only a minimal amount of normal tissue. This is critical in clinical situations where the functional state of the background organ parenchyma is as important in determining patient outcome as the primary tumor. Examples of clinical situations where this is relevant include focal hepatic tumors in patients with underlying cirrhosis and limited hepatic reserve, patients with von Hippel-Lindau syndrome who have limited renal function and require treatment of multiple renal tumors, and patients with primary lung tumors with extensive underlying emphysema and limited lung function.8-10 Many of these patients are not surgical candidates owing to limited native organ functional reserve, placing them at a higher risk for postoperative complications or organ failure. Other clinical circumstances in which high specificity and accuracy of targeting have proven useful include providing symptomatic relief for patients with symptomatic osseous metastases or hormonally active neuroendocrine tumors, and in using percutaneous therapies to improve focal interstitial drug delivery.2,11,12 Finally, maximizing ablation efficiency is an additional consideration. For example, appropriate and complete tumor destruction only occurs when the entire target tumor is exposed to appropriate temperatures, and therefore determined by the pattern of tissue heating in the target tumor. For larger tumors (usually defined as > 3-5 cm in diameter), a single ablation treatment may not be sufficient to entirely encompass the target volume.13 In these cases, multiple overlapping ablations or simultaneous use of multiple applicators may be required to successfully treat the entire tumor and achieve an ablative margin, though accurate targeting and probe placement can often be technically

CHAPTER 140 IMAGE-GUIDED THERMAL TUMOR ABLATION: BASIC SCIENCE AND COMBINATION THERAPIES 1029 Background tissue

60 50 43

A

Tumor

Applicator

B

FIGURE 140-1. Schematic (A) and pictorial (B) representation of focal thermal ablation therapy. Electrode applicators are positioned either with image guidance or direct visualization within target tumor, and thermal energy is applied via electrode. This creates a central zone of high temperatures in tissue immediately around electrode (can exceed 100°C), surrounded by more peripheral zones of sublethal tissue heating ( 100°C) leads to tissue dehydration and water vaporization, which cause dramatic increases in circuit impedance, and ultimately limits further heating.28 When these effects begin to inhibit current flow from a generator, alternative methods (described later) to decrease circuit impedance, such as expanding the electrode surface area, pulsing the input power, and injections of saline, can be used to augment RF current flow.29,30 Microwave Microwaves are a second thermal energy source used for percutaneous image-guided tumor ablation. In contrast to RFA, where the inserted electrode functions as the active source, in microwave ablation (MWA) the inserted applicators (usually 14-17 gauge) function as antennae for externally applied energy at 1000 to 2450 MHz.31 The microwave energy applied to tissues results in rotation of polar molecules, which is opposed by frictional forces. As a result, there is conversion of rotational energy into heat.32 There are several potential advantages of MWA for tissue ablation.33,34 Microwaves readily penetrate through biological materials, including those with low electrical conductivity (e.g., lung, bone) and dehydrated or charred tissue. Consequently, microwave power can be continually applied to produce very high temperatures (>150°C), improving ablation efficacy by increasing thermal conduction into surrounding tissue.35 Microwaves also heat tissue more efficiently than RF energy in tissue, do not require ground pads, and multiple antennas can be operated simultaneously.36 The superposition of microwaves may even be exploited to augment performance.136-137

On the other hand, microwave energy is inherently more difficult to distribute than RF energy. Microwaves must be carried in waveguides (e.g., coaxial cable) that are typically more cumbersome than the small wires used to feed energy to RF electrodes, and prone to heating when carrying large amounts of power. It is well known that higher microwave powers increase ablation zone size, but excessive power in the antenna shaft can lead to unintended injuries to other tissues such as the skin.33,37 Recent investigations have shown that adding a cooling jacket around the antenna can reduce cable heating and eliminate skin burns while effectively increasing the amount of power that can safely be delivered to the tumor.38 In clinical practice, MWA has been utilized most in Japan and China, where several systems have been described.39,40 Most of these systems operate at 2.45 GHz and use monopole, dipole, or slotted coaxial antennas to deliver up to 60 W. Recently, 915 MHz and water-cooled systems have been described that appear to deliver up to 80 W and create larger ablations than previous generation systems.41 Recently a number of U.S. Food and Drug Administration (FDA)-approved MWA devices have become available for commercial use. Complete clinical characterization of each of these devices is currently ongoing. Laser Laser ablation or laser photocoagulation is another method for inducing thermally mediated coagulation necrosis that has been employed for percutaneous tumor ablation. For this procedure, flexible thin optic fibers are inserted into the target through percutaneously placed needles, using imaging guidance. The laser provides sufficient energy to allow for significant heat deposition surrounding the fiber tip, inducing protein denaturation and cellular death. As with RF systems, thermal profiles have been demonstrated to correlate well with the extent of coagulation necrosis observed histopathologically,42,43 as well as with ultrasound43,44 and T1-weighted magnetic resonance (MR) images.45,46 Most devices use a standard laser source (Nd:YAG, erbium, holmium, etc.) that produces precise wavelengths. Additional device modifications have been developed to increase the size of the heated volume, including: (1) type of laser fiber (flexible/glass dome), (2) modifications to the tip (i.e., flexible diffuser tip or scattering dome), (3) length of applicator and diameter of the optic fiber, and (4) the number of laser applicators used (i.e., single vs. multiple applicators).47 Similar to RF technologies, additional device modifications such as pulsing algorithms and internal cooling of the applicator have also been reported.48 Ultrasound High-intensity focused ultrasound (HIFU) is a transcutaneous technique that has recently been studied for minimally invasive treatment of localized malignancy.49,50 HIFU uses a parabolic transducer to focus the ultrasound energy at a distance, creating a focused beam of energy with very high peak intensity. This focusing of energy has been likened to using a magnifying glass to focus sunlight.51 The focused energy is transmitted transcutaneously into the targeted tissue without requiring percutaneous insertion of an electrode or transducer. The ultrasound energy absorbed by tissue is converted to heat, which ablates tissue via

CHAPTER 140 IMAGE-GUIDED THERMAL TUMOR ABLATION: BASIC SCIENCE AND COMBINATION THERAPIES 1031

Iceball

Argon input

Vacuum Vacuum Expansion chamber

A

B

FIGURE 140-2. Cryoablation. A, Schematic illustration of tip of a cryoprobe with surrounding iceball formation. B, Axial contrast-enhanced computed tomography image demonstrates multiple cryoprobes placed and iceball formation during cryoablation of tumor in right lobe of liver. (From Ahmed M, Brace CL, Lee FT Jr, Goldberg SN. Principles of and advances in percutaneous ablation. Radiology 2011;258:351–69.)

coagulation necrosis. Since insertion of a percutaneous applicator is not required, HIFU can be considered the least invasive of the “minimally invasive” therapies. However, limitations in time and coverage, especially given respiratory motion, have limited most clinical investigations to benign conditions in static organs such as the breast,52 uterus,53 and prostate.54 Cryoablation Cryoablation induces focal tissue injury using alternating sessions of freezing and tissue thawing55-59 (Fig. 140-2). Minimally invasive cryoablation is performed using a closed specially constructed cryoprobe that is placed on or inside a tumor. In the past, cryoablation had to be performed in the setting of an open procedure owing to the large caliber of available probes; however, technologic developments have led to smaller probes that can be placed percutaneously under MR or computed tomography (CT) guidance. Liquid forms of inert gases are cycled through the hollow probes, resulting in tissue cooling. The two main types of systems use gas or liquid nitrogen, or argon gas. Many of the systems that use argon gas base their freezing on the Joule-Thompson effect.60-62 More recent work has involved probes that use “super-critical” nitrogen to cool faster and deeper into tissue.63 Cryoablation induces direct cellular injury through multiple mechanisms including tissue cooling.64 At low cooling rates, freezing primarily propagates extracellularly, which draws water from the cell and results in osmotic dehydration.65 The intracellular high solute concentration that develops leads to damaged enzymatic systems, protein damage, and injury to the cell membrane.66,67 At higher cooling rates, water is trapped within the cell, intracellular ice formation occurs, and organelle and membrane injury results.65,68 A second hypothesized mechanism of cryoablative tissue destruction centers on vascular injury consisting of mechanical injury to the vessel wall, direct injury to cells lining the vessel, and post-thaw injury from reperfusion.64 Mechanical injury to the vessel wall from intravascular ice formation results in increased vessel porosity, reduced plasma oncotic pressure, and perivascular edema.69 Endothelial injury further results from the exposure of underlying connective tissue to the vessel lumen and subsequent thrombus formation.70 Damage to endothelial cells occurs in a fashion similar to the direct cellular injury described

earlier. Reperfusion injury results from the release of vasoactive factors after the tissue thaws, which leads to vasodilation and increased blood flow to treated areas. Subsequent high oxygen delivery71 and neutrophil migration72 into the damaged area results in increased free radical formation, and through the peroxidation of membrane lipids, further endothelial damage. The degree to which each mechanism of cellular and tissue injury described determines overall tissue destruction is governed by characteristics of the administration algorithm, including freezing and thawing rate, the minimum (i.e., coldest) temperature reached, and the duration that temperature was maintained.64 However, for current commonly used systems, argon provides a heat sink of about 9 kJ and can generate temperatures as low as −140°C inside the iceball, which expands by thermal conduction.73 Finally, the fact that the lethal isotherm (−20°C to −40°C) rests several millimeters inside the iceball boundary must be considered when using imaging to assess cryoablation treatment efficacy.74 Irreversible Electroporation Percutaneous irreversible electroporation (IRE) is new ablation technology and is most notable because it is presumed to be inherently nonthermal (i.e., no heat is produced to cause cell death). Rather, cells are eradicated using several microsecond- to millisecond-long pulses of electric current. The pulses generate electrical fields up to 3 kV/cm that cause irreversible damage to the cell membrane, thereby inducing apoptosis.75,76 Since IRE is predominantly nonthermal, heat sinks, such as large vessels, should have a much smaller influence on the ablation zone than with thermal treatments. IRE also appears to limit damage to more collagenous tissues and nerves, which if verified in larger trials will make it an attractive option for thermally sensitive areas (e.g., in the prostate, near large blood vessels, or the renal cortex).77 IRE electrodes consist of insulated 19-gauge (1.1 mm diameter) or larger needles with an exposed active portion of 1 to 4 cm. For most applications, multiple electrodes are required and spaced 1 to 3 cm apart to provide sufficient electrical field strengths for irreversible cell damage. A single-needle bipolar electrode is also available for more localized treatments. Initial studies required very high-voltage pulses, but recent reports have shown that

1032 PART 2 NONVASCULAR INTERVENTIONS · SECTION 20 GENERAL CONSIDERATIONS IN ENERGY-BASED ABLATION lower-voltage pulses can be used when repeated several hundred times.78 Current IRE devices do have notable drawbacks, including generation of potentially dangerous electrical harmonics that can stimulate muscle contraction or cardiac arrhythmias. These techniques require general anesthesia and paralytic induction and have treatment times on the order of seconds that prevent treatment monitoring or adjustment. There is also a requirement for accurate placement of several needles to achieve moderate-sized ablations (≈3-4 cm), and a lack of coagulation around needle insertion sites, which theoretically could elevate bleeding complication risks. Ongoing research aims to minimize these complications.

ADVANCES IN RADIOFREQUENCY ABLATION: A MODEL FOR TECHNOLOGIC EVOLUTION The full discussion of technologic advances that have been and continue to be made in RFA is available at www. expertconsult.com.

Practical Application of Technologic Developments: Choosing an Ablation Device Several different commercially available RFA devices are commonly used, and vary based on their power application algorithms and electrode designs. Selecting a Radiofrequency Ablation System: Is One System Better Than Another? Zones of adequate thermal ablation can achieved with most commercially available RFA devices when properly used as described in the instructions for use. Though the specific size of the ablation zone and the time required to achieve adequate ablation will likely vary between different manufacturers and devices, a recent study by Lin et al. compared four separate and commonly used devices in over 100 patients with primary and secondary hepatic tumors, and found little difference in ablation time and local tumor progression.101 Thus, it is important for operators to be completely familiar with their device of choice, including electrode shapes and deployment techniques, power-input algorithms, and common trouble-shooting issues, so as to permit thorough treatment planning and maximize optimal clinical outcomes. Ultimately, good operator technique and

A

B

careful patient selection contributes at least as much to treatment success as does the specific electrode choice. Understanding the Principles or Algorithm of the Selected Device Several commonly used RF devices use different power application algorithms based upon current flow, impedance, or time to deliver RF energy. Current electrode designs are paired to specific devices and companies, and RF power algorithms are also commonly tailored to specific electrode designs (such that operators can not interchange electrodes from one device to another, and the greatest efficiencies in use are likely found by following company-recommended application protocols). Several specific electrode designs are available, most commonly divided into those that are needle-like versus multi-tined expandable. Specifically, practitioners should be familiar with the shape of the coagulation zone each system generates. For example, needlelike electrodes induce a more oval ablation zone parallel to the axis of the electrode, whereas expandable electrodes generate ablation zones that are oval shaped perpendicular to the shaft of the electrode. Examples of several commercially available systems (limited to the most common FDA-approved devices) are illustrated in Figure 140-3. The Cool-Tip RF electrode system (Valleylab Inc., Boulder, Colo.) uses a pulsed application applied over a 12-minute time period that inputs high amounts of current, alternating frequently with “offperiods” triggered by rises in impedance, to reduce tissue overheating around the electrode. The Cool-Tip RF system uses 17-gauge needle-like electrodes as single, cluster, or multiple single electrodes that are internally cooled with icewater (to a temperature of < 10°C). Electrode switching technology (described earlier) is currently only available with the Cool-Tip RF system. In contrast, the Boston Scientific RF 3000 ablation system (Boston Scientific, Natick, Mass.) employs two rounds of slowly increasing power to achieve tissue heating, which occurs until tissue impedance starts to rise (thereby limiting further current input) to threshold levels, colloquially termed roll-off. The RF 3000 system uses umbrella-shaped expandable electrodes with multiple (10-12) tines extending out from the needle shaft. Finally, the RITA system (AngioDynamics, Queensbury, N.Y.), another commonly used device, administers RF energy to set temperature endpoints (usually 105°C as measured by sensors within the electrode tips) combined with incremental extension of an expandable electrode system.

C

D

FIGURE 140-3. Commonly used and commercially available electrode designs. A, A single internally cooled electrode with a 3-cm active tip (Cool-Tip system [Valleylab, Boulder, Colo.]). B, A cluster internally cooled electrode system with three 2.5-cm active tips (Cluster electrode system [Valleylab]). C-D, Two variations of an expandable electrode system: C, Starburst (RITA Medical Systems, Mountain View, Calif.) and, D, LeVeen (Boston Scientific, Natick, Mass.). (From Ahmed M, Brace CL, Lee FT Jr, Goldberg SN. Principles of and advances in percutaneous ablation. Radiology 2011;258:351–69.)

CHAPTER 140 IMAGE-GUIDED THERMAL TUMOR ABLATION: BASIC SCIENCE AND COMBINATION THERAPIES1032.e1 Most investigations to improve thermal ablation outcomes have focused on device development, much of which has been for RF-based ablation systems (see earlier). Technologic efforts to increase ablation size have focused on modifying energy deposition algorithms and electrode designs to increase both the amount of tissue exposed to the active electrode and the overall amount of energy that can be safely deposited into the target tissue.

Refinement of Energy Application Algorithms The algorithm by which energy is applied during thermal ablation is dependent on the energy source, device, and type of electrode being used. Initial power algorithms for RF-based systems were based upon a continuous and constant high-energy input, but tissue overheating and vaporization ultimately interferes with continued energy input, owing to high impedance to current flow from gas formation. Therefore, several strategies to maximize energy deposition have been developed and, in some cases, incorporated into commercially available devices. Applying high levels of energy in a pulsed manner, separated by periods of lower energy, is one such strategy that has been used with RF-based systems to increase the mean intensity of energy deposition.79 If a proper balance between high- and low-energy deposition is achieved, preferential tissue cooling occurs adjacent to the electrode during periods of minimal energy deposition, without significantly decreasing heating deeper in the tissue. Thus, even greater energy can be applied during periods of high-energy deposition, thereby enabling deeper heat penetration and greater tissue coagulation.80 Synergy between a combination of both internal cooling of the electrode and pulsing has resulted in even greater coagulation necrosis and tumor destruction than either method alone.80 Pulsed-energy techniques have also been successfully used for microwave and laser-based systems. Another strategy is to slowly increase (“ramp up”) the RF energy application in a continuous manner until the impedance to RF current flow increases prohibitively.61,81 This approach is often paired with multi-tined expandable electrodes, which have a greater contact surface area with tissue, and in which the goal is to achieve smaller ablation zones around multiple small electrode tines. This algorithm is also often combined with a staged expansion of the electrode system such that each small ablation occurs in a slightly different location within the tumor (with the overall goal being ablating the entire target region). Electrode switching is an additional technique incorporated into pulsing algorithms to further increase RF tissue heating.82 In this technique, multiple independently placed RF electrodes are connected to a single RF generator, and RF current is applied to a single electrode until an impedance spike is detected, at which point current application is applied to the next electrode, and so on. Several studies have demonstrated significant increases in ablation zone size and a reduction in application time using this technique.82,83 Brace et al. also show larger and more circular ablation zones with the switching application, with more rapid heating being 74% faster than sequential heating (12 vs. 46 minutes).82 Continued device development has also led to increases in the overall maximum amount of power that can be

delivered.84,85 For RF-based devices, the maximum amount of RF current that can be delivered is dependent on both generator output and electrode surface area. Higher surface areas reduce the current density, and therefore adequate tissue heating around the electrode cannot be achieved. Initial systems had maximum power outputs of less than 200 W, but subsequent investigation suggests that higher current output and larger ablation zones can be achieved if higher-powered generators are coupled with larger surface area electrodes. For example, Solazzo et al. used a 500-KHz (1000 W) high-powered generator in an in vivo porcine model and achieved larger coagulation zones with a 4-cm tip cluster electrode (5.2 ± 0.8 cm) compared to a 2.5-cm cluster electrode (3.9 ± 0.3 cm).85 Similar gains in ablation size have been seen in higher-powered versions of microwave-based systems.86

Electrode Modification Development of ablation applicators (electrodes for RF, antennae for microwave, and diffuser tips for laser-based systems) has contributed significantly to the ability to reliably achieve larger ablation zones. Several strategies to increase the amount of energy deposition and the overall ablation size have been balanced with the need for smallercaliber electrodes to permit continued use of these devices in a percutaneous and minimally invasive manner. These include use of multiple electrodes simultaneously (either adjacent to each other or as part of an expandable device through a single introducer needle), use of cooling systems to minimize tissue and electrode overheating, and use of bipolar systems for RFA to increase tissue heating in the target zone. Originally, simply lengthening the electrode tip increased coagulation in an asymmetric and preferentially longitudinal geometry. Use of a single electrode inserted multiple times to perform overlapping ablations requires significantly greater time and effort, making it impractical for routine use in a clinical setting. Therefore, using multiple electrodes simultaneously in a preset configuration represents a significant step forward in increasing overall ablation size. Initial work with multiple electrodes demonstrated that placement of several monopolar electrodes in a clustered arrangement (no more than 1.5 cm apart) with simultaneous RF application could increase coagulation volume by over 800% compared to a single electrode.87 Subsequently, working to overcome the technical challenges of multiprobe application, multi-tined expandable RF electrodes were developed.88 These systems involve deploying a varying number of multiple thin, curved tines in the shape of an umbrella (or more complex geometries) from a central cannula.89,90 This surmounts earlier difficulties by allowing easy placement of multiple probes to create large reproducible volumes of necrosis. Leveen et al., using a 12-hook array, were able to produce lesions measuring up to 3.5 cm in diameter in in vivo porcine liver by administering increasing amounts of RF energy from a 50 W RF generator for 10 minutes.91 More recently, Applebaum et al. were able to achieve over 5 cm of coagulation in in vivo porcine liver using currently commercially available expandable electrodes with optimized stepped-extension and power input algorithms.92

1032.e2PART 2 NONVASCULAR INTERVENTIONS · SECTION 20 GENERAL CONSIDERATIONS IN ENERGY-BASED ABLATION Although the majority of conventional RF systems use monopolar electrodes (where the current runs to remotely placed grounding pads placed on the patient’s thighs to complete the circuit), several studies have reported results using bipolar arrays to increase the volume of coagulation created by RF application. In these systems, applied RF current runs from an active electrode to a second grounding electrode in place of a grounding pad. Heat is generated around both electrodes, creating elliptical lesions. McGahan et al. used this method in ex vivo liver to induce necrosis of up to 4 cm in the long-axis diameter but could only achieve 1.4 cm of necrosis in the short diameter.93 Though this increases the overall size of coagulation volume, the shape of necrosis is unsuitable for actual tumors, making the gains in coagulation less clinically significant. Desinger et al. have described another bipolar array that contains both the active and return electrodes on the same 2-mm-diameter probe.94 Lee et al. used two multi-tined electrodes as active and return to increase coagulation during bipolar RFA.95 Finally, several studies have used a multipolar (more than two) array of electrodes (multi-tined and single internally cooled) during RFA to achieve even greater volumes of coagulation.96,97 One of the limitations for RF-based systems has been overheating surrounding the active electrode, leading to tissue charring, rising impedance, and RF circuit interruption, ultimately limiting overall RF energy deposition. One successful strategy to address this has been the use of internal electrode cooling, where electrodes contain two hollow lumens that permit continuous internal cooling of the tip with a chilled perfusate, and the removal of warmed effluent to a collection unit outside the body.29 This reduces heating

directly around the electrode, tissue charring, and rising impedance; allows greater RF energy deposition; and shifts the peak tissue temperature farther into the tumor, contributing to a broader depth of tissue heating from thermal conduction. In initial studies using cooling of 18-gauge single or clustered electrode needles using chilled saline perfusate, significant increases in RF energy deposition and ablation zone size were observed compared to conventional monopolar uncooled electrodes in ex vivo liver, with findings subsequently confirmed in in vivo large-animal models and clinical studies.29,80 Similar results were observed when chilled saline was infused in combination with expandable electrode systems (referred to as an internally cooled wet electrode), though infusion of fluid around the electrode is more difficult to control and makes the reproducibility of results more variable.98 Most recently, several investigators have used alternative cooling agents (e.g., argon or nitrogen gas) to achieve even greater cooling, and therefore larger zones of ablation, around the RF electrode tip.99 As for RF-based systems, cooling of antennae shafts for microwave applicators have also been developed and reduce shaft heating (and associated complications such as skin burns) while allowing increased power deposition through smallercaliber antennae. Although many of these technologic advances have been developed independently of each other, they may often be used concurrently to achieve even larger ablation zones. Furthermore, while much of this work has been based upon RF technology, specific techniques can be used for other thermal ablative therapies. For example, multiple electrodes and applicator cooling have been effectively applied in microwave-based systems as well.100

CHAPTER 140 IMAGE-GUIDED THERMAL TUMOR ABLATION: BASIC SCIENCE AND COMBINATION THERAPIES 1033 The RITA system uses an expandable electrode system paired with slow saline drip infusion.

MODIFICATION OF THE BIOPHYSIOLOGIC ENVIRONMENT Although larger coagulation zones have been created by modifying electrode design in ablative procedures, limitations due to the tumor physiology itself have to be considered. Recent investigations have centered on altering underlying tumor physiology as a means to improving thermal ablation. Current studies have focused on the effects of tissue characteristics in the setting of temperaturebased therapies such as tissue perfusion, thermal conductivity, and system-specific characteristics such as electrical conductivity for RFA.

Tissue Perfusion The leading factor constraining thermal ablation coagulation size in tumors continues to be tissue blood flow, which has two effects: a large-vessel heat sink effect and a microvascular perfusion-mediated tissue cooling effect. Firstly, larger-diameter blood vessels with higher flow act as heat sinks, drawing away heat from the ablative area. Lu et al. show this in an in vivo porcine model by examining the effect of hepatic vessel diameter on RFA outcome. Using CT and histopathologic analysis, more complete thermal heating and a reduced heat sink effect is seen when the heating zone was less than 3 mm in diameter.102 In contrast, vessels over 3 mm in diameter had higher patency rates, less endothelial injury, and greater viability of surrounding hepatocytes after RFA. Studies confirm the effect of hepatic blood flow on RF-induced coagulation, in which increased coagulation volumes were obtained with decreased hepatic blood flow either by embolotherapy, angiographic balloon occlusion, coil embolization, or the Pringle maneuver (total portal inflow occlusion).103,104 The second effect of tissue vasculature is due to perfusionmediated tissue cooling (capillary vascular flow), which also functions as a heat sink. By drawing heat from the treatment zone, this effect reduces the volume of tissue that receives the required minimal thermal dose for coagulation. Targeted microvascular perfusion by using pharmacologic alteration of blood flow can also improve overall RFA efficacy. Goldberg et al. modulated hepatic blood flow using intraarterial vasopressin and high-dose halothane in conjunction with RFA in in vivo porcine liver.105 Horkan et al. modulated the use of arsenic trioxide, showing reduced blood flow and increased tumor destruction.106 More recently, another antiangiogenic drug, sorafenib, has been combined with RFA to increase overall tumor coagulation in small-animal models.107 In this regard, newer nonthermal energy sources (e.g., IRE) have been shown to be less affected by these effects and will likely provide an alternative in clinical situations where perfusion-mediated cooling impacts ablation outcome.108

Thermal Conductivity Initial clinical studies using RFA for hepatocellular carcinoma in the setting of underlying cirrhosis noted an “oven”

effect (i.e., increased heating efficacy for tumors surrounded by cirrhotic liver or fat, such as exophytic renal cell carcinomas) or altered thermal transmission at the junction of tumor tissue and surrounding tissue.13 Subsequent experimental studies in ex vivo agar phantoms and bovine liver have confirmed the effects of varying tumor and surrounding tissue thermal conductivity on effective heat transmission during RFA, and further demonstrated the role of “optimal” thermal conductivity characteristics on ablation outcome.16,109 For example, very poor tumor thermal conductivity limits heat transmission centrifugally away from the electrode, with marked heating in the central portion of the tumor and limited, potentially incomplete, heating in peripheral portions of the tumor. In contrast, increased thermal conductivity (e.g., in cystic lesions) results in fast heat transmission (i.e., heat dissipation), with potentially incomplete and heterogeneous tumor heating. Furthermore, in agar phantom and computer modeling studies, Liu et al. demonstrated that differences in thermal conductivity between the tumor and surrounding background tissue (specifically, decreased thermal conductivity from increased fat content of surrounding tissue) results in increased temperatures at the tumor margin. However, heating was limited in the surrounding medium, making a 1-cm “ablative margin” more difficult to achieve.16,109 Understanding the role of thermal conductivity and tissue and tumorspecific characteristics on tissue heating may be useful when trying to predict ablation outcome in varying clinical settings (e.g., in exophytic renal cell carcinomas surrounded by perirenal fat, lung tumors surrounded by aerated normal parenchyma, or osseous metastases surrounded by cortical bone).16

Electrical Conductivity The power deposition in RF-induced tissue heating is strongly dependent on local electrical conductivity; its effect can be divided into two main categories. First, altering the electrical activity environment immediately around the RF electrode with ionic agents can increase electrical conductivity prior to or during RFA. High local saline concentration increases the area of the active surface electrode, allowing greater energy deposition, thereby increasing the extent of coagulation necrosis.110 Saline may be beneficial when ablating cavitary lesions that might not have a sufficient current path. However, it should be noted that saline infusion is not always a predictable process, since fluid can migrate to unintended locations and cause complications if not used properly.111 Second, different electrical conductivities between the tumor and surrounding background organ can affect tissue heating at the tumor margin. Several studies show increases in tissue heating at the tumor/organ interface when the surrounding medium is characterized by reduced lower electrical conductivity.109 In certain clinical settings, such as treating focal tumors in either lung or bone, marked differences in electrical conductivity may result in variable heating at the tumor/organ interface and, indeed, limit heating in the surrounding organ and make obtaining a 1-cm ablative margin difficult. Lastly, electrical conductivity must be taken into account when using techniques such as hydrodissection to protect adjacent organs. Nonionic fluids can be used to protect tissues adjacent to the ablation zone (e.g., diaphragm, bowel) from thermal

1034 PART 2 NONVASCULAR INTERVENTIONS · SECTION 20 GENERAL CONSIDERATIONS IN ENERGY-BASED ABLATION injury. For this application, fluids with low ion content, such as 5% dextrose in water (D5W) should be used, since they have been proven to electrically force RF current away from the protected organ, decrease the size and incidence of burns on the diaphragm and bowel, and reduce pain scores in patients treated with D5W when compared to ionic solutions such as saline.82,83 Ionic solutions like 0.9% saline should not be used for hydrodissection since, as noted, they actually increase RF current flow.112

Tissue Fluid Content Microwave-based systems use electromagnetic energy to forcibly align molecules with an intrinsic dipole moment (e.g., water) to the externally applied magnetic field, which generates kinetic energy and results in local tissue heating. Therefore, higher tissue water content influences the rates and maximum amounts of achievable tissue heating in these systems.35 Incorporating tissue internal water content into computer modeling of tumor ablation to more accurately predict tissue heating has also been proposed.113 In addition, Brace et al. have demonstrated that greater dehydration (and resultant contraction of the ablation zone) occurs with MWA compared to RFA systems.114 Optimization of microwave heating by modulating the tissue fluid content is also being investigated.115

RATIONALE FOR COMBINING THERMAL ABLATION WITH OTHER THERAPIES Substantial efforts have been made in modifying ablation systems and the biological environment to improve the clinical utility of percutaneous ablation, but limitations in clinical efficacy persist. For example, with further longterm follow-up of patients undergoing ablation therapy, there has been an increased incidence of detection of progressive local tumor growth for all tumor types and sizes, despite initial indications of adequate therapy, suggesting there are residual foci of viable untreated disease in a substantial but unknown number of cases.13 The ability to achieve complete and uniform eradication of all malignant cells remains a key barrier to clinical success, so strategies that can increase the completeness of RF tumor destruction, even for small lesions, are needed. Several studies have demonstrated that tumor death can be enhanced when combining RF thermal therapy with adjuvant chemo- or radiosensitizers. The goal of this combined approach is to increase tumor destruction occurring within the sizable peripheral zone of sublethal temperatures (i.e., largely reversible cell damage induced by mildly elevating tissue temperatures to 41°C-45°C) surrounding the heat-induced coagulation.116 Modeling studies demonstrate that were the threshold for cell death to be decreased by as few as 5°C, tumor coagulation could be increased up to 1.5 cm (up to a 59% increase in spherical volume of the ablation zone).109 Improved tumor cytotoxicity is also likely to reduce the local recurrence rate at the treatment site. Although there is high temperature heating throughout the zone of RFA, heterogeneity of thermal diffusion (especially in the presence of vascularity) retards uniform and complete ablation.117 Since local control requires complete tumor destruction, ablation may be inadequate even if large

zones of ablation that encompass the entire tumor are created. By killing tumor cells at lower temperatures, this combined paradigm will not only increase necrosis volume but may also create a more complete area of tumor destruction by filling in untreated gaps within the ablation zone.118 Combined treatment also has the potential to achieve equivalent tumor destruction with a concomitant reduction of the duration or course of therapy (a process that currently takes hours to treat larger tumors, with many protocols requiring repeat sessions). A reduction in the time required to completely ablate a given tumor volume would permit patients with larger or greater numbers of tumors to be treated. Shorter heating time could also potentially improve their quality of life by reducing the number of patient visits and the substantial costs of prolonged procedures that require image guidance.

Combined Radiofrequency Ablation and Transarterial Chemoembolization Independently, RFA and transarterial chemoembolization (TACE) are both commonly used modalities for locoregional treatment of liver tumors such as hepatocellular carcinoma (HCC), but several limitations to their optimal outcomes persist. RFA is less effective in larger tumors and difficult to use in specific locations or near larger blood vessels. TACE can be used for larger tumors, but rates of tumor necrosis are lower (30%-90% compared to > 90% for RFA). Therefore, several experimental and clinical studies have investigated administering them in a combined manner. The rationale for combined RF/TACE is based upon several potential advantages. This includes combined two-hit cytotoxic effects of exposure to nonlethal low-level hyperthermia in peri-ablational tumor and adjuvant chemotherapy. Alterations in tumor perfusion can potentiate the effects of either RFA through pre-ablation embolization of tumor vasculature, or TACE with postablation peripheral hyperemia increasing blood flow for TACE. Finally, performing TACE first can improve tumor visualization (through intratumoral iodized oil deposition) for RFA. RFA and TACE can be administered in varying paradigms based upon the sequence and time between therapies. Mostafa et al. demonstrated in VX2 tumors implanted in rabbit liver that the largest treatment volumes were obtained when TACE preceded RFA, compared with RF before TACE or either therapy alone.104 Ahrar et al. investigated the effects of high-temperature heating on commonly used TACE agents (doxorubicin, mitomycin-C, and cisplatin) and found minimal change in the cytotoxic activity of the agents when exposed to clinically relevant durations of heating (>100°C for < 20 minutes).119 For HCC tumors invisible on ultrasound and unenhanced CT, Lee et al. reported that 71% of tumors could be adequately visualized for subsequent RFA.120 Based upon this, the optimal strategy for administration is likely performing TACE first, followed by RFA. This is confirmed in a recent randomized controlled study by Morimoto et al. comparing TACE-RF (on the same day) to RF alone for intermediate-sized HCC (3.1-5 cm diameter), where TACE-RF resulted in lower rates of local tumor progression at 3 years.121 Another approach in combination therapy is to primarily treat the target tumor with a single modality (either TACE

CHAPTER 140 IMAGE-GUIDED THERMAL TUMOR ABLATION: BASIC SCIENCE AND COMBINATION THERAPIES 1035

A

B

C

D

FIGURE 140-4. Combining transarterial chemoembolization (TACE) with radiofrequency ablation (RFA) in a 45-year-old man with a 4.2-cm focal hepatocellular carcinoma. Initially, Lipiodol-based TACE was performed, with good Lipiodol tumor staining on immediate fluoroscopic images (A) and follow-up noncontrast axial computed tomography (CT) images (B). C, On 1-month follow-up contrast-enhanced T1-weighted axial magnetic resonance imaging, however, a residual focus of arterial enhancement can be seen along lateral aspect of tumor. This was successfully treated with percutaneous RFA using a single 3-cm internally cooled electrode. D, Follow-up contrast-enhanced axial CT image demonstrates complete treatment (determined by lack of residual contrast enhancement in ablated tumor).

or RF) followed by additional adjuvant treatments using either one or both modalities for residual disease. For example, RFA can be performed for an initial presentation of limited disease, followed by TACE performed at a later date for residual peripheral tumor, satellite nodules, or new foci. Likewise, TACE can be performed initially to control more extensive disease, with subsequent RF directed at small foci of local recurrence or new small nodules (Fig. 140-4). Several studies using combination therapy have reported increases in ablation size and increased treatment efficacy with combination RF-TACE, particularly as the primary treatment of large (>5 cm) unresectable tumors. For example, Yang et al. treated 103 patients with recurrent unresectable HCC after hepatectomy and reported lower intrahepatic recurrence and longer 3-year survival compared to either therapy alone.122 The potential benefit of combining RF with TACE for small ( 60˚ zone Vascular delivery to tumor site

Necrosis at margins of tumor

> 43˚ zone

Normal liver tissue

Thermal ablation

Tissue

FIGURE 140-5. Method for combining thermal ablation with targeted drug delivery. Left, Drugs are brought to tumor site as part of normal circulation. Middle, Temperature elevations inside ablation zone facilitate local drug release, which then accumulates in sublethal region at periphery of ablation zone. Right, Net result is a larger zone of ablation than possible with ablation alone. (© 2008 by the Board of Regents of the University of Wisconsin System.)

A

B

C

D

FIGURE 140-6. Increased tumor destruction with combined RF and liposomal doxorubicin in an 82-year-old male with an 8.2-cm vascular hepatoma. A, Computed tomography (CT) image obtained immediately following radiofrequency ablation (RFA) shows persistent regions of residual untreated tumor (black zone = ablated region). B, Two weeks following therapy, there is interval increase in coagulation; 1.5-cm inferior region of residual tumor and 1.2-cm anteromedial portion of tumor no longer enhance. A persistent nodule of viable tumor is identified. This was successfully treated with a course of RFA. C, CT image obtained immediately following RFA demonstrates persistence of a large vessel coursing through nonenhancing ablated lesion. D, Two weeks post therapy, there is no enhancement throughout this region, and no vessel was seen on any of three phases of contrast enhancement. No evidence of local tumor recurrence was identified at 48-month follow-up. (From Goldberg SN, Kamel IR, Kruskal JB, et al. Radiofrequency ablation of hepatic tumors: Increased tumor destruction with adjuvant liposomal doxorubicin therapy. AJR Am J Roentgenol 2002;179:93–101.)

Combining Thermal Ablation with Adjuvant Radiation Several studies have reported early investigation into combination RFA and radiation therapy, with promising results.3,128-130 There are known synergistic effects of combined external beam radiation therapy and low-temperature hyperthermia.52 Experimental animal studies have demonstrated increased tumor necrosis, reduced tumor growth, and improved survival with combined external beam radiation and RFA, compared to either therapy alone.130,131 For example, in a rat breast adenocarcinoma model, Horkan et al. demonstrated significantly longer mean endpoint survival for animals treated with combination RFA and 20-Gy external beam radiation (94 days) compared to either radiation (40 days) or RFA (20 days) alone. Preliminary clinical studies in primary lung malignancies confirm the synergistic effects of these therapies.3 Potential causes for the synergy include sensitization of the tumor to subsequent radiation due to the increased oxygenation resulting from hyperthermia-induced increased blood flow to the tumor.132

Another possible mechanism, which has been seen in animal tumor models, is an inhibition of radiation-induced repair and recovery and increased free radical formation.133 Future work is needed to identify the optimal temperature for ablation and optimal radiation dose, as well as the most effective method of administering radiation therapy (external beam radiation therapy, brachytherapy, or yttrium microspheres), on an organ-by-organ basis.

IMPROVING IMAGE GUIDANCE AND TUMOR TARGETING For a discussion of some of the major challenges posed by thermal ablation and advances in overcoming them, please visit www.expertconsult.com.

CONCLUSIONS Several different percutaneous and minimally invasive therapies have been well described for use in treating focal malignancies. Investigators have characterized many of the

CHAPTER 140 IMAGE-GUIDED THERMAL TUMOR ABLATION: BASIC SCIENCE AND COMBINATION THERAPIES1036.e1 Thermal ablation is commonly performed with a percutaneous approach using imaging guidance with a single or combination of modalities (CT, ultrasound, or MRI). A successful ablation is contingent upon the operator’s ability to visualize the tumor, position the electrode within the target, and accurately evaluate the treatment zone upon ablation completion. Several significant challenges to performing a successful ablation exist at each of these steps. For example, diagnostic imaging studies are often obtained with modalities that are separate from the modality being used for treatment (e.g., diagnostic MRI and ablation being performed with CT or ultrasound), positional variations between diagnostic and treatment imaging precludes an exact overlay of different imaging studies, and target tumors are often highly variable in their imaging appearance in a modality-specific manner (e.g., a focal hepatocellular carcinoma may be clearly visible on MRI, barely visible on ultrasound, and invisible with noncontrast-enhanced CT). Electrode positioning often requires traversing a narrow course or window in a three-dimensional (3D) trajectory when only two-dimensional real-time imaging is available, often when the target is moving with respiratory motion. Finally, correlating the immediate postprocedure imaging to prior diagnostic imaging to determine adequacy of treatment, especially at the tumor margin, can often be difficult.

All these factors make treating some lesions extremely technically challenging.134 Several technologies are being developed to address some of these difficulties.134 Image fusion software is now becoming commercially available to allow image overlay of two different imaging modalities for diagnostic interpretation. Multimodality image fusion for procedural guidance takes this further by pairing an existing dataset from a prior diagnostic study to the “real-time” modality being used for the procedure. CT-ultrasound fusion systems often use a sensor in the ultrasound transducer and initial landmark localization to fuse a prior CT scan (which allows multiplanar reconstruction from the CT dataset) with real-time ultrasound images (see Fig. 140-3). Finally, needle tracking systems are also being developed using either electromagnetic (EM) or optical (infrared-based) tracking technology.135 In these systems, the needle tip is identified and localized in 3D space, and the needle trajectory overlays existing imaging. EM-based devices use a small field generator to create a rapidly changing magnetic field, in which a sensor coil in the needle tip creates an electric current, allowing localization within 3D space. Several of these devices have been tested for simple procedures such as joint injection or biopsy and can likely be applied to ablative procedures as well.136,137

CHAPTER 140 IMAGE-GUIDED THERMAL TUMOR ABLATION: BASIC SCIENCE AND COMBINATION THERAPIES 1037 basic principles underlying the tumor ablative features of these treatments. We have provided an overview of the basic principles of each of these strategies and described the recent technologic modifications that have been developed to further improve clinical success of these therapies. Future directions of research are now looking to combination therapies, such as combining chemotherapy and/or radiation therapy with thermal ablation, as the next step to better understand how thermal energy interacts with tissue to further improve predictability and clinical effectiveness of percutaneous thermal tumor ablations.

•

Successful treatment is a balance between complete tumor destruction and minimizing damage to surrounding normal parenchyma and adjacent structures.

•

Widespread adoption of radiofrequency ablation will rely upon improving and increasing tumor ablation volume. Adjuvant therapies such as antiangiogenic agents, chemotherapeutics, and radiation in this capacity hold great promise.

▶ SUGGESTED READINGS

•

Focal thermal tumor ablation is a viable alternative for the treatment of many solid focal malignancies, especially in nonsurgical candidates.

Goldberg SN, Grassi CJ, Cardella JF, et al. for the Society of Interventional Radiology Technology Assessment Committee and the International Working Group on Image-Guided Tumor Ablation. Image-guided tumor ablation: standardization of terminology and reporting criteria. J Vasc Interv Radiol 2009;20(7 Suppl):S377–90.

•

Benefits of thermal ablation include low morbidity and mortality, low cost, nonsurgical patient inclusion, and same-day discharge.

The complete reference list is available online at www.expertconsult.com.

KEY POINTS

CHAPTER 140 IMAGE-GUIDED THERMAL TUMOR ABLATION: BASIC SCIENCE AND COMBINATION THERAPIES1037.e1 REFERENCES 1. Ahmed M, Brace CL, Lee Jr FT, Goldberg SN. Principles of and advances in percutaneous ablation. Radiology 2011;258:351–69. 2. Callstrom MR, et al. Painful metastases involving bone: percutaneous image-guided cryoablation–prospective trial interim analysis. Radiology 2006;241(2):572–80. 3. Dupuy DE, et al. Radiofrequency ablation followed by conventional radiotherapy for medically inoperable stage I non-small cell lung cancer. Chest 2006;129(3):738–45. 4. Lencioni R, et al. Early stage hepatocellular carcinoma in patients with cirrhosis: long-term results of percutaneous image-guided radiofrequency ablation. Radiology 2005;234(3):961–7. 5. Goldberg SN, Gazelle GS, Mueller PR. Thermal ablation therapy for focal malignancy: a unified approach to underlying principles, techniques, and diagnostic imaging guidance. Am J Radiol 2000;174: 323–31. 6. Dodd 3rd GD, et al. Minimally invasive treatment of malignant hepatic tumors: at the threshold of a major breakthrough. Radiographics 2000;20(1):9–27. 7. Shimada K, et al. Role of the width of the surgical margin in a hepatectomy for small hepatocellular carcinomas eligible for percutaneous local ablative therapy. Am J Surg 2008;195(6):775–81. 8. Gervais DA, et al. Radiofrequency ablation of renal cell carcinoma: part 1, Indications, results, and role in patient management over a 6-year period and ablation of 100 tumors. AJR Am J Roentgenol 2005;185(1):64–71. 9. Hiraki T, et al. Percutaneous radiofrequency ablation for clinical stage I non-small cell lung cancer: results in 20 nonsurgical candidates. J Thorac Cardiovasc Surg 2007;134(5):1306–12. 10. Lencioni R, et al. Response to radiofrequency ablation of pulmonary tumours: a prospective, intention-to-treat, multicentre clinical trial (the RAPTURE study). Lancet Oncol 2008;9(7):621–8. 11. Goldberg SN, et al. Percutaneous tumor ablation: increased coagulation necrosis with combined radiofrequency and percutaneous doxorubicin injection. Radiology 2001;220:420–7. 12. Gillams A, et al. Radiofrequency ablation of neuroendocrine liver metastases: the Middlesex experience. Abdom Imaging 2005;30(4): 435–41. 13. Livraghi T, et al. Hepatocellular carcinoma: radiofrequency ablation of medium and large lesions. Radiology 2000;214:761–8. 14. Dodd 3rd GD, et al. Radiofrequency thermal ablation: computer analysis of the size of the thermal injury created by overlapping ablations. AJR Am J Roentgenol 2001;177(4):777–82. 15. Schramm W, Yang D, Haemmerich D. Contribution of direct heating, thermal conduction and perfusion during radiofrequency and microwave ablation. Conf Proc IEEE Eng Med Biol Soc 2006; 1:5013–6. 16. Ahmed M, et al. Computer modeling of the combined effects of perfusion, electrical conductivity, and thermal conductivity on tissue heating patterns in radiofrequency tumor ablation. Int J Hyperthermia 2008;24(7):577–88. 17. Thrall DE, et al. A comparison of temperatures in canine solid tumours during local and whole-body hyperthermia administered alone and simultaneously. Int J Hyperthermia 1990;6(2):305–17. 18. Seegenschmiedt M, Brady L, Sauer R. Interstitial thermoradiotherapy: review on technical and clinical aspects. Am J Clin Oncol 1990;13:352–63. 19. Trembley B, Ryan T, Strohbehn J. Interstitial hyperthermia: physics, biology, and clinical aspects. Hyperthermia and Oncology, Vol. 3, Utrecht: VSP; 1992. p. 11–98. 20. Larson T, Bostwick D, Corcia A. Temperature-correlated histopathologic changes following microwave thermoablation of obstructive

21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42.

tissues in patients with benign prostatic hyperplasia. Urology 1996;47:463–9. Zevas N, Kuwayama A. Pathologic analysis of experimental thermal lesions: comparison of induction heating and radiofrequency electrocoagulation. J Neurosurg 1972;37:418–22. Goldberg SN, et al. Treatment of intrahepatic malignancy with radiofrequency ablation: radiologic-pathologic correlation. Cancer 2000; 88:2452–63. Goldberg SN, et al. Radiofrequency tissue ablation: importance of local temperature along the electrode tip exposure in determining lesion shape and size. Acad Radiol 1996;3:212–8. Mertyna P, et al. Radiofrequency ablation: the effect of distance and baseline temperature on thermal dose required for coagulation. Int J Hyperthermia 2008;24(7):550–9. Mertyna P, et al. Radiofrequency ablation: variability in heat sensitivity in tumors and tissues. J Vasc Interv Radiol 2007;18(5):647–54. Daniels C, Rubinsky B. Electrical field and temperature model of nonthermal irreversible electroporation in heterogeneous tissues. J Biomech Eng 2009;131(7):071006. Haemmerich D, Pilcher TA. Convective cooling affects cardiac catheter cryoablation and radiofrequency ablation in opposite directions. Conf Proc IEEE Eng Med Biol Soc 2007;2007:1499–502. Ring ME, et al. Determinants of impedance rise during catheter ablation of bovine myocardium with radiofrequency energy. Pacing Clin Electrophysiol 1989;12(9):1502–13. Goldberg SN, et al. Radiofrequency tissue ablation: increased lesion diameter with a perfusion electrode. Acad Radiol 1996;3:636–44. Leveillee RJ, et al. Enhanced radiofrequency ablation of canine prostate utilizing a liquid conductor: the virtual electrode. J Endourol 1996;10(1):5–11. King RWP, Shen LC, Wu TT. Embedded insulated antenna for communication and heating. Electromagnetics 1981;1:51–72. Schepps JL, Foster KR. The UHF and microwave dielectric properties of normal and tumour tissues: variation in dielectric properties with tissue water content. Phys Med Biol 1980;25(6):1149–59. Wolf FJ, et al. Microwave ablation of lung malignancies: effectiveness, CT findings, and safety in 50 patients. Radiology 2008; 247(3):871–9. Wright AS, et al. Radiofrequency versus microwave ablation in a hepatic porcine model. Radiology 2005;236(1):132–9. Yang D, et al. Measurement and analysis of tissue temperature during microwave liver ablation. IEEE Trans Biomed Eng 2007;54(1): 150–5. Brace CL, et al. Microwave ablation with a single small-gauge triaxial antenna: in vivo porcine liver model. Radiology 2007;242(2): 435–40. Strickland AD, et al. Experimental study of large-volume microwave ablation in the liver. Br J Surg 2002;89(8):1003–7. Wang Y, et al. Internally cooled antenna for microwave ablation: results in ex vivo and in vivo porcine livers. Eur J Radiol 2008; 67(2):357–61. Ito K, et al. Coaxial-slot antenna for interstitial microwave thermal therapy and its application to clinical trial. Conf Proc IEEE Eng Med Biol Soc 2004;4:2526–9. Shibata T, et al. Small hepatocellular carcinoma: comparison of radio-frequency ablation and percutaneous microwave coagulation therapy. Radiology 2002;223(2):331–7. Sun Y, et al. Comparison of ablation zone between 915- and 2,450MHz cooled-shaft microwave antenna: results in in vivo porcine livers. AJR Am J Roentgenol 2009;192(2):511–4. Dachman AH, et al. US-guided percutaneous laser ablation of liver tissue in a chronic pig model. Radiology 1990;176:129–33.

1037.e2PART 2 NONVASCULAR INTERVENTIONS · SECTION 20 GENERAL CONSIDERATIONS IN ENERGY-BASED ABLATION 43. Jiao LR, et al. Clinical short-term results of radiofrequency ablation in primary and secondary liver tumors. Am J Surg 1999;177(4): 303–6. 44. Nolsoe CP, et al. Interstitial hyperthermia of colorectal liver metastases with a US-guided Nd-YAG laser with a diffuser tip: a pilot clinical study. Radiology 1993;187(2):333–7. 45. Diederich CJ, Nau WH, Deardorff DL. Prostate thermal therapy with interstitial and transurethral ultrasound applicators: a feasibility study. In: Thomas P. Ryan, editor. Surgical Applications of Energy, Proceedings of SPIE, Vol. 3249. 1998. p. 2–13. 46. Vogl TJ, et al. Malignant liver tumors treated with MR imagingguided laser-induced thermotherapy: technique and prospective results. Radiology 1995;196(1):257–65. 47. Vogl T, et al. Laser-induced thermotherapy of malignant liver tumors: general principles, equipments, procedure - side effects, complications and results. Eur J Ultrasound 2001;13:117–27. 48. Vogl TJ, et al. Internally cooled power laser for MR-guided interstitial laser-induced thermotherapy of liver lesions: initial clinical results. Radiology 1998;209(2):381–5. 49. Yang R, et al. Liver cancer ablation with extracorporeal high-intensity focused ultrasound. Eur Urol 1993;23(Suppl 1):17–22. 50. Jolesz FA, Hynynen K. Magnetic resonance image-guided focused ultrasound surgery. Cancer 2002;8(Suppl 1):S100–12. 51. Sanghvi NT, Hawes RH. High-intensity focused ultrasound. Gastrointest Endosc Clin N Am 1994;4(2):383–95. 52. Algan O, et al. External beam radiotherapy and hyperthermia in the treatment of patients with locally advanced prostate carcinoma. Cancer 2000;89(2):399–403. 53. Tempany CM, et al. MR imaging-guided focused ultrasound surgery of uterine leiomyomas: a feasibility study. Radiology 2003;226(3): 897–905. 54. Ficarra V, et al. Short-term outcome after high-intensity focused ultrasound in the treatment of patients with high-risk prostate cancer. BJU Int 2006;98(6):1193–8. 55. Cooper IS. Cryogenic surgery: A new method of destruction or extirpation of benign or malignant tissue. N Engl J Med 1963;268: 743–9. 56. Onik G, Kane RA, Steele G, et al. Monitoring hepatic cryosurgery with sonography. AJR Am J Roentgenol 1986;147:665–9. 57. Lee FT, et al. Hepatic cryosurgery with intraoperative US guidance. Radiology 1997;202:624–32. 58. Cozzi PJ, Stewart GJ, Morris DL. Thrombocytopenia after hepatic cryosurgery for colorectal metastases: Correlates with hepatic injury. World J Surg 1994;18:774–7. 59. Rubinsky B, et al. The process of freezing and the mechanism of damage during hepatic cryosurgery. Cryobiology 1990;27:85–97. 60. Caleffi M, et al. Cryoablation of benign breast tumors: evolution of technique and technology. Breast 2004;13(5):397–407. 61. Gulesserian T, et al. Comparison of expandable electrodes in percutaneous radiofrequency ablation of renal cell carcinoma. Eur J Radiol 2006;59(2):133–9. 62. Nakada SY, et al. Comparison of radiofrequency ablation, cryoablation, and nephrectomy in treating implanted VX-2 carcinoma in rabbit kidneys. J Endourol 2004;18(5):501–6. 63. Littrup PJ, Babkin AV, Duncan RV. Cryotherapy technology assessment: a near critical nitrogen (NCN) system can provide faster, colder, and more predictable treatments. In: Scientific Assembly and Annual Meeting of the Radiological Society of North America. Chicago, IL: Radiological Society of North America; 2006. 64. Hoffmann NE, Bischof JC. The cryobiology of cryosurgical injury. Urology 2002;60:40–9. 65. Mazur P. Freezing of living cells: mechanisms and implications. Am J Physiol 1984;143:C125–42.

66. Lovelock JE. The hemolysis of human red blood cells by freezing and thawing. Biochem Biophys Acta 1953;10:414–26. 67. Steponkus PL. Role of plasma membrane in freezing injury and cold acclimation. Annu Rev Plant Physiol 1984;35:543–84. 68. Toner M. Nucleation of ice crystals inside biological cells. In: Steponkus PL, editor. Advances in Low-Temperature Biology, Vol 2. London: JAI Press; 1993. p. 1–52. 69. Pollock GA, Pegg DE, Hardie IR. An isolated perfused rat mesentery model for direct observation of the vasculature during cryopreservation. Cryobiology 1986;23:500–11. 70. Hoffmann NE, Bischof JC. Cryosurgery of normal and tumor tissue in the dorsal skin flap chamber. II. J Biomech Eng 2001; 123:310–6. 71. Barker JH, Bartlett R, Funk W. The effect of superoxide dismutase on the skin microcirculation after ischemia and reperfusion. Prog Appl Microcirc 1987;12:276–81. 72. Zook N, Hussmann J, Brown R. Microcirculatory studies of frostbite injury. Ann Plast Surg 1998;40:246–53. 73. Kim C, et al. Finite-element analysis of ex vivo and in vivo hepatic cryoablation. IEEE Trans Biomed Eng 2007;54(7): 1177–85. 74. Rewcastle JC, et al. A model for the time dependent threedimensional thermal distribution within iceballs surrounding multiple cryoprobes. Med Phys 2001;28(6):1125–37. 75. Rubinsky B, Onik G, Mikus P. Irreversible electroporation: a new ablation modality–clinical implications. Technol Cancer Res Treat 2007;6(1):37–48. 76. Lee EW, Loh CT, Kee ST. Imaging guided percutaneous irreversible electroporation: ultrasound and immunohistological correlation. Technol Cancer Res Treat 2007;6(4):287–94. 77. Onik G, Mikus P, Rubinsky B. Irreversible electroporation: implications for prostate ablation. Technol Cancer Res Treat 2007; 6(4):295–300. 78. Rubinsky J, et al. Optimal parameters for the destruction of prostate cancer using irreversible electroporation. J Urol 2008;180(6): 2668–74. 79. Goldberg SN, et al. Percutaneous radiofrequency tissue ablation: optimization of pulsed-RF technique to increase coagulation necrosis. J Vasc Interv Radiol 1999;10:907–16. 80. Goldberg SN, et al. Large-volume tissue ablation with radiofrequency by using a clustered, internally cooled electrode technique: laboratory and clinical experience in liver metastases. Radiology 1998;209:371–9. 81. McGahan JP, Loh S, Boschini FJ, et al. Maximizing parameters for tissue ablation by using an internally cooled electrode. Radiology 2010;256:397–405. 82. Brace CL, et al. Radiofrequency ablation: simultaneous application of multiple electrodes via switching creates larger, more confluent ablations than sequential application in a large animal model. J Vasc Interv Radiol 2009;20(1):118–24. 83. Laeseke PF, et al. Multiple-electrode radiofrequency ablation creates confluent areas of necrosis: in vivo porcine liver results. Radiology 2006;241(1):116–24. 84. Brace CL, et al. Radiofrequency ablation with a high-power generator: device efficacy in an in vivo porcine liver model. Int J Hyperthermia 2007;23(4):387–94. 85. Solazzo SA, et al. High-power generator for radiofrequency ablation: larger electrodes and pulsing algorithms in bovine ex vivo and porcine in vivo settings. Radiology 2007;242(3):743–50. 86. Laeseke PF, et al. Microwave ablation versus radiofrequency ablation in the kidney: high-power triaxial antennas create larger ablation zones than similarly sized internally cooled electrodes. J Vasc Interv Radiol 2009;20(9):1224–9.

CHAPTER 140 IMAGE-GUIDED THERMAL TUMOR ABLATION: BASIC SCIENCE AND COMBINATION THERAPIES1037.e3 87. Goldberg SN, et al. Radiofrequency tissue ablation using multiprobe arrays: Greater tissue destruction than multiple probes operating alone. Acad Radiol 1995;2:670–4. 88. Bangard C, et al. Large-volume multi-tined expandable RF ablation in pig livers: comparison of 2D and volumetric measurements of the ablation zone. Eur Radiol 2010;20(5):1073–8. 89. Rossi S, Buscarini E, Garbagnati F. Percutaneous treatment of small hepatic tumors by an expandable RF needle electrode. AJR Am J Roentgenol 1998;170:1015–22. 90. Siperstein AE, et al. Laparoscopic thermal ablation of hepatic neuroendocrine tumor metastases. Surgery 1997;122:1147–55. 91. Leveen RF. Laser hyperthermia and radiofrequency ablation of hepatic lesions. Semin Interv Radiol 1997;12:313–24. 92. Appelbaum L, et al. Algorithm optimization for multitined radiofrequency ablation: comparative study in ex vivo and in vivo bovine liver. Radiology 2010;254(2):430–40. 93. McGahan JP, et al. Hepatic ablation using bipolar radiofrequency electrocautery. Acad Radiol 1996;3(5):418–22. 94. Desinger K, Stein T, Mueller GJ, et al. Interstitial bipolar RF-thermotherapy (REITT): therapy planning by computer simulation and MRI monitoring–a new concept for minimally invasive procedures. Proc SPIE 1998;3249:147–60. 95. Lee JM, et al. Bipolar radiofrequency ablation using wet-cooled electrodes: an in vitro experimental study in bovine liver. AJR Am J Roentgenol 2005;184(2):391–7. 96. Seror O, et al. Large (>or=5.0-cm) HCCs: multipolar RF ablation with three internally cooled bipolar electrodes–initial experience in 26 patients. Radiology 2008;248(1):288–96. 97. Lee JM, et al. Multiple-electrode radiofrequency ablation of in vivo porcine liver: comparative studies of consecutive monopolar, switching monopolar versus multipolar modes. Invest Radiol 2007; 42(10):676–83. 98. Hsieh CL, et al. Effectiveness of ultrasound-guided aspiration and sclerotherapy with 95% ethanol for treatment of recurrent ovarian endometriomas. Fertil Steril 2009;91(6):2709–13. 99. Hines-Peralta A, et al. Hybrid radiofrequency and cryoablation device: preliminary results in an animal model. J Vasc Interv Radiol 2004;15(10):1111–20. 100. He N, et al. Microwave ablation: An experimental comparative study on internally cooled antenna versus non-internally cooled antenna in liver models. Acad Radiol 2010;17:894–9. 101. Lin SM, et al. Radiofrequency ablation for hepatocellular carcinoma: a prospective comparison of four radiofrequency devices. J Vasc Interv Radiol 2007;18(9):1118–25. 102. Lu DS, et al. Influence of large peritumoral vessels on outcome of radiofrequency ablation of liver tumors. J Vasc Interv Radiol 2003;14(10):1267–74. 103. Patterson EJ, et al. Radiofrequency ablation of porcine liver in vivo: effects of blood flow and treatment time on lesion size. Ann Surg 1998;227(4):559–65. 104. Mostafa EM, et al. Optimal strategies for combining transcatheter arterial chemoembolization and radiofrequency ablation in rabbit VX2 hepatic tumors. J Vasc Interv Radiol 2008;19(12):1740–8. 105. Goldberg SN, et al. Radio-frequency tissue ablation: effect of pharmacologic modulation of blood flow on coagulation diameter. Radiology 1998;209(3):761–7. 106. Horkan C, et al. Radiofrequency ablation: Effect of pharmacologic modulation of hepatic and renal blood flow on coagulation diameter in a VX2 tumor model. J Vasc Interv Radiol 2004;15(3): 269–74. 107. Hakime A, et al. Combination of radiofrequency ablation with antiangiogenic therapy for tumor ablation efficacy: study in mice. Radiology 2007;244(2):464–70.

108. Lee EW, et al. Advanced hepatic ablation technique for creating complete cell death: irreversible electroporation. Radiology 2010; 255(2):426–33. 109. Liu Z, et al. Radiofrequency tumor ablation: insight into improved efficacy using computer modeling. AJR Am J Roentgenol 2005; 184(4):1347–52. 110. Aube C, et al. Influence of NaCl concentrations on coagulation, temperature, and electrical conductivity using a perfusion radiofrequency ablation system: an ex vivo experimental study. Cardiovasc Intervent Radiol 2007;30(1):92–7. 111. Gillams AR, Lees WR. CT mapping of the distribution of saline during radiofrequency ablation with perfusion electrodes. Cardiovasc Intervent Radiol 2005;28(4):476–80. 112. Laeseke PF, et al. Use of dextrose 5% in water instead of saline to protect against inadvertent radiofrequency injuries. AJR Am J Roentgenol 2005;184(3):1026–7. 113. Yang D, et al. Expanding the bioheat equation to include tissue internal water evaporation during heating. IEEE Trans Biomed Eng 2007;54(8):1382–8. 114. Brace CL, et al. Tissue contraction caused by radiofrequency and microwave ablation: a laboratory study in liver and lung. J Vasc Interv Radiol 2010;21(8):1280–6. 115. Isfort P, et al. [In vitro experiments on fluid-modulated microwave ablation]. Rofo 2010;182(6):518–24. 116. Ahmed M, Goldberg SN. Combination radiofrequency thermal ablation and adjuvant IV liposomal doxorubicin increases tissue coagulation and intratumoural drug accumulation. Int J Hyperthermia 2004;20(7):781–802. 117. Goldberg SN, et al. Percutaneous radiofrequency tissue ablation: does perfusion-mediated tissue cooling limit coagulation necrosis? J Vasc Interv Radiol 1998;9(1 Pt 1):101–11. 118. Goldberg SN, et al. Radiofrequency ablation of hepatic tumors: increased tumor destruction with adjuvant liposomal doxorubicin therapy. AJR Am J Roentgenol 2002;179(1):93–101. 119. Ahrar K, et al. 2004 Dr. Gary J. Becker Young Investigator Award: Relative thermosensitivity of cytotoxic drugs used in transcatheter arterial chemoembolization. J Vasc Interv Radiol 2004;15(9):901–5. 120. Lee MW, et al. Percutaneous radiofrequency ablation of small hepatocellular carcinoma invisible on both ultrasonography and unenhanced CT: a preliminary study of combined treatment with transarterial chemoembolisation. Br J Radiol 2009;82(983):908–15. 121. Morimoto M, et al. Midterm outcomes in patients with intermediatesized hepatocellular carcinoma: a randomized controlled trial for determining the efficacy of radiofrequency ablation combined with transcatheter arterial chemoembolization. Cancer 2010;116: 5452–60. 122. Yang W, et al. Combination therapy of radiofrequency ablation and transarterial chemoembolization in recurrent hepatocellular carcinoma after hepatectomy compared with single treatment. Hepatol Res 2009;39(3):231–40. 123. Wang W, Shi J, Xie WF. Transarterial chemoembolization in combination with percutaneous ablation therapy in unresectable hepatocellular carcinoma: a meta-analysis. Liver Int 2010;30(5):741–9. 124. Goldberg SN, et al. Radiofrequency thermal ablation with adjuvant saline injection: effect of electrical conductivity on tissue heating and coagulation. Radiology 2001;219:157–65. 125. Gaber MH, et al. Thermosensitive liposomes: extravasation and release of contents in tumor microvascular networks. Int J Radiat Oncol Biol Phys 1996;36(5):1177–87. 126. Negussie AH, et al. Formulation and characterisation of magnetic resonance imageable thermally sensitive liposomes for use with magnetic resonance-guided high intensity focused ultrasound. Int J Hyperthermia 2011;27(2):140–55.

1037.e4PART 2 NONVASCULAR INTERVENTIONS · SECTION 20 GENERAL CONSIDERATIONS IN ENERGY-BASED ABLATION 127. Gasselhuber A, et al. Mathematical spatio-temporal model of drug delivery from low temperature sensitive liposomes during radiofrequency tumour ablation. Int J Hyperthermia 2010;26(5):499–513. 128. Chan MD, et al. Combined radiofrequency ablation and high-dose rate brachytherapy for early stage non-small-cell lung cancer. Brachytherapy 129. Grieco CA, et al. Percutaneous image-guided thermal ablation and radiation therapy: outcomes of combined treatment for 41 patients with inoperable stage I/II non-small-cell lung cancer. J Vasc Interv Radiol 2006;17(7):1117–24. 130. Horkan C, et al. Reduced tumor growth with combined radiofrequency ablation and radiation therapy in a rat breast tumor model. Radiology 2005;235(1):81–8. 131. Solazzo S, et al. RF ablation with adjuvant therapy: comparison of external beam radiation and liposomal doxorubicin on ablation efficacy in an animal tumor model. Int J Hyperthermia 2008;24(7): 560–7.

132. Mayer R, et al. Hyperbaric oxygen and radiotherapy. Strahlenther Onkol 2005;181(2):113–23. 133. Solazzo S, et al. Liposomal doxorubicin increases radiofrequency ablation-induced tumor destruction by increasing cellular oxidative and nitrative stress and accelerating apoptotic pathways. Radiology 2010;255:62–74. 134. Wood BJ, et al. Navigation systems for ablation. J Vasc Interv Radiol 2010;21(8 Suppl):S257–63. 135. Krucker J, et al. Electromagnetic tracking for thermal ablation and biopsy guidance: clinical evaluation of spatial accuracy. J Vasc Interv Radiol 2007;18(9):1141–50. 136. Klauser AS, et al. Fusion of real-time US with CT images to guide sacroiliac joint injection in vitro and in vivo. Radiology 2010; 256(2):547–53. 137. Khan MF, et al. Navigation-based needle puncture of a cadaver using a hybrid tracking navigational system. Invest Radiol 2006;41(10): 713–20.

SECTION TWENTY-ONE

THE LIVER CHAPTER

141

Energy-Based Ablation of Hepatocellular Cancer Riccardo Lencioni, Elena Bozzi, Laura Crocetti, and Carlo Bartolozzi Hepatocellular carcinoma (HCC) is the fifth most common cause of cancer, and its incidence is increasing worldwide because of the dissemination of hepatitis B and C virus infection.1,2 Patients with cirrhosis are at the highest risk of developing HCC and should be monitored every 6 months to diagnose the tumor at an asymptomatic stage.2 If diagnosed at an early stage, patients should be considered for any of the available options that may provide a high rate of complete response. These include surgical resection, liver transplantation, and percutaneous techniques of tumor ablation.3 Indication for surgical resection is currently restricted to patients with single asymptomatic HCC and extremely well-preserved liver function who have neither clinically significant portal hypertension nor abnormal bilirubin.4,5 Cadaveric liver transplantation is limited by the shortage of donors, and living donor liver transplantation is still at an early stage of clinical application.4,5 As a result, percutaneous ablation plays a key role in therapeutic management of HCC.

PRETREATMENT EVALUATION Imaging Assessment In the setting of a patient with known hepatitis B or cirrhosis of other etiology, a solid nodular lesion found during surveillance has a high likelihood of being HCC. However, it has been shown by pathologic studies that many small nodules detected in cirrhotic livers do not correspond to HCC.6 Current guidelines recommend further investigation of nodules detected during surveillance with dynamic imaging techniques, including contrast-enhanced multidetector computed tomography (MDCT) and contrast-enhanced magnetic resonance imaging (MRI).4 In fact, one of the key pathologic factors for differential diagnosis that is reflected in dynamic imaging studies is the vascular supply to the lesion. Through the progression from regenerative nodule, to low-grade dysplastic nodule (DN), to high-grade DN, to frank HCC, one sees loss of visualization of portal tracts and development of new arterial vessels, termed nontriadal arteries, which become the dominant blood supply in overt HCC lesions.6 It is this neovascularity that allows HCC to be diagnosed and is the key for imaging cirrhotic patients. A rational diagnostic protocol should be structured according to the actual risk of malignancy and the possibility of achieving a reliable diagnosis. Since the prevalence of HCC among ultrasound-detected nodules is strongly related to the size of the lesion, the diagnostic workup depends on lesion size. Lesions smaller than 1 cm in 1038

diameter have a low likelihood of being HCC, but minute hepatic nodules detected by ultrasound may become malignant over time. Therefore, these nodules should be followed up to detect growth suggestive of malignant transformation. A reasonable protocol is to repeat ultrasound every 3 months until the lesion grows to more than 1 cm, at which point additional diagnostic techniques are applied.4 It has to be emphasized, however, that the absence of growth during the follow-up period does not rule out the malignant nature of the nodule, because even an early HCC may take more than 1 year to increase in size.4 When the nodule exceeds 1 cm in size, the lesion is more likely to be HCC, and diagnostic confirmation should be pursued. Current guidelines recommend that the diagnosis of HCC can be made noninvasively without biopsy in a nodule larger than 1 cm that shows the characteristic vascular features of HCC (i.e., arterial hypervascularization with washout in the portal venous or delayed phase) even in patients with normal α-fetoprotein values. Such lesions should be treated as HCC, since the positive predictive value of the clinical and radiologic findings exceeds 95%, provided that examinations are conducted by using state-of-the-art equipment and interpreted by radiologists with extensive expertise in liver imaging.7 For lesions above 1 cm without a characteristic vascular profile, a second dynamic study is required. If definitive diagnosis is not reached with these techniques, biopsy is recommended.4 It has to be pointed out that noninvasive criteria based on imaging findings can be applied only in patients with established cirrhosis.4 For nodules detected in noncirrhotic livers, as well as for those showing atypical vascular patterns, biopsy is recommended. Ultrasound is widely accepted for HCC surveillance, but multidetector CT or dynamic MRI are required for intrahepatic staging of the disease; these examinations provide a comprehensive assessment of the liver parenchyma and can identify additional tumor foci.

Clinical Staging In most solid malignancies, tumor stage at presentation determines prognosis and treatment management. Most patients with HCC, however, have two diseases—liver cirrhosis and HCC—and complex interactions between the two have major implications for prognosis and treatment choice.8 Therefore, the TNM system has limited usefulness in the clinical decision-making process because it does not take into account hepatic functional status. Several scoring systems have been developed in the past few years in attempts to stratify patients according to expected survival.

CHAPTER 141 ENERGY-BASED ABLATION OF HEPATOCELLULAR CANCER 1039 However, the only system that links staging with treatment modalities is the Barcelona Clinic Liver Cancer (BCLC) staging system.9,10 The BCLC includes variables related to tumor stage, liver functional status, physical status, and cancer-related symptoms and provides an estimation of life expectancy that is based on published response rates to the various treatments. In the BCLC system, early-stage HCC includes patients with World Health Organization (WHO) performance status of 0, preserved liver function (Child-Pugh class A or B), and solitary tumor or up to 3 nodules smaller than 3 cm each, in the absence of macroscopic vascular invasion and extrahepatic spread. If the patient has ChildPugh class A cirrhosis and a solitary tumor smaller than 2 cm, the stage may be defined as very early. Patients with multinodular HCC with neither vascular invasion nor extrahepatic spread are classified as intermediate-stage according to the BCLC staging system, provided they have a performance status of 0 and Child-Pugh class A or B cirrhosis.11 Patients with portal vein invasion or extrahepatic disease are classified as advanced stage. The terminal stage includes patients who have either severe hepatic decompensation (Child-Pugh class C) or performance status greater than 2.

TREATMENT OF EARLY-STAGE HEPATOCELLULAR CARCINOMA Patients with early-stage HCC can benefit from curative therapies including surgical resection, liver transplantation, and percutaneous ablation, and have the possibility of longterm cure, with 5-year survival figures ranging from 50% to 75%.10 However, there is no firm evidence to establish the optimal first-line treatment for early-stage HCC because of the lack of randomized controlled trials (RCTs) comparing radical therapies. Patients should be evaluated in referral centers by multidisciplinary teams involving hepatologists, oncologists, interventional radiologists, surgeons, and pathologists to guarantee careful selection of candidates for each treatment option and ensure expert application of these treatments.10

Surgical Resection Resection is the treatment of choice for HCC in noncirrhotic patients, who account for about 5% of the cases in Western countries. However, in patients with cirrhosis, candidates for resection have to be carefully selected to reduce the risk of postoperative liver failure. It has been shown that a normal bilirubin concentration and the absence of clinically significant portal hypertension are the best predictors of excellent outcomes after surgery.12 In experienced hands, such patients have treatment-related mortality of less than 1% to 3% and may achieve 5-year survival higher than 70%.12-14 In contrast, survival drops to less than 50% at 5 years in patients with significant portal hypertension, and to less than 30% at 5 years in those with both adverse factors (portal hypertension and elevated bilirubin).12 Most groups restrict the indication for resection to patients with a single tumor in a suitable location. Anatomic resections guided by intraoperative ultrasound techniques are preferred to wedge resections, because they

include any microsatellite lesions possibly located in the same hepatic segment as the main tumor. In fact, it is known that neoplastic dissemination occurs at very early stages in HCC via the invasion of small peripheral portal vein branches.6 After resection, tumor recurrence rate exceeds 70% at 5 years, including recurrence due to dissemination and de novo tumors developing in the remnant cirrhotic liver. The most powerful predictors of recurrence are the presence of microvascular invasion and/or additional tumor sites besides the primary lesion.12

Liver Transplantation Liver transplantation is the only option that provides cure of both the tumor and the underlying chronic liver disease. It is recognized as the best treatment for patients with solitary HCC smaller than 5 cm in the setting of decompensated cirrhosis and for those with early multifocal disease (up to 3 lesions, none larger than 3 cm).4 However, for patients with a solitary small tumor in well-compensated cirrhosis, the optimal treatment strategy is still under debate.15 The reported outcomes of patients who actually underwent transplantation are better than those of patients submitted to resection, especially if the substantially lower rates of tumor recurrence ( 3 cm). In a study reported by DeBaere et al., a percutaneous approach allowed local control of 90% of treated lesions after 1 year of follow-up, but the mean diameter of these lesions was significantly smaller than those in our series.47 In our experience, the time interval from treatment to recurrence was related to the size of the lesion; local relapse occurred in virtually all cases during the first 12-month period. The median time to local recurrence was 9.3 months. Repeat RFA was carried out in 37.6% of metastases with progressive local tumor growth (Fig. 142-3). Distant metachronous metastases had developed at follow-up in 36% of our patients (Fig. 142-4). The estimated

Local Control 83.1%

Survival 3 yr: 62% 5 yr: 39.5% 3 yr: 28% Median: 28.9%

3 yr: 64% 5 yr: 44%

median time until detection of new metastases was 12.3 months in our experience. Kaplan-Meier analysis of the overall survival rates at 1, 3, and 5 years in our series was 96.2%, 62.0%, and 39.5%, respectively, with an estimated median survival time of 51 months. The overlapping long-term outcome of both surgical resection and RFA will allow future randomized studies to compare RFA with surgery for resectable metastatic liver tumors and enable proper evaluation of the impact of RFA. RFA should not replace hepatic resection whenever applicable, especially for large metastases (>6 cm) because of the higher risk of local recurrence as a result of residual viable tumor tissue within the necrotic area. The precise role of RFA in currently adopted therapeutic regimens has yet to be defined.55 RFA performed after previous chemotherapy regimens is still associated with worse survival because of extensive liver disease stage, chemotherapy-refractory tumor, or diffuse extrahepatic disease. A comparison by Machi et al.56 of RFA used as first- or second-line treatment or in the salvage setting showed greater survival benefit when used as a first-line therapeutic option. Randomized controlled trials to evaluate the value of combined adjuvant systemic chemotherapy are urgently required in the short term. Not so long ago, the “test-of-time” approach was proposed by surgeons to delay resection of liver metastases so additional, still undetected lesions could develop and become identifiable, thus limiting the number of resections carried out in patients in whom more metastases may ultimately develop. On the one hand, RFA applied in this setting can significantly decrease the number of potential resections, resulting in complete tumor control in some of these patients and avoiding major surgery. Furthermore, offering cancer patients a treatment option such as RFA rather than a “wait-and-see” approach is favorable for the patient. In a recent study using RFA as a “test of time” before resection in a group of 88 potentially operable patients with 134 metastases from colorectal carcinoma monitored for a period of 18 to 75 months, RFA was successful in 60.2% of the patients, 43.4% of whom were

1052 PART 2 NONVASCULAR INTERVENTIONS · SECTION 21 THE LIVER

A

B

C

FIGURE 142-3. A 72-year-old man with history of colon carcinoma and prior radiofrequency ablation of three metachronous liver metastases. A, The 9-month follow-up computed tomography (CT) study showed a small (12 mm) local recurrence along posterior margin of treated metastasis in segment VII. Percutaneous ultrasound-guided local re-treatment of recurrence was performed; 24-hour (B) and 2-year (C) CT scans showed complete local control of metastasis.

FIGURE 142-4. This 48-year-old woman with a history of colon carcinoma underwent percutaneous ultrasound-guided radiofrequency ablation (RFA) of three metachronous liver metastases. One year after RFA, computed tomography showed complete local control of treated lesions (one necrotic area is seen in segment II in subcapsular location) but also development of multiple new metastases that cannot undergo local treatment.

disease free during the course of the study. In the remaining patients (56.6%), new untreatable intrahepatic or extrahepatic metastases developed. RFA was unsuccessful in 39.8% of the patients; 57.1% of them underwent resection, and new untreatable metastases developed in the remaining 42.9%. No patients became untreatable because of the growth of incompletely ablated lesions. In summary, 50% of patients were spared surgery that would have been noncurative, whereas an additional 26.1% of patients avoided resection because of the curative result of RFA.57

Noncolorectal Metastases Although the vast majority of RFA procedures address liver metastases of colorectal origin, a significant subset of

secondary hepatic tumors occurs in patients with breast cancer.30 Metastatic breast cancer has traditionally been considered a manifestation of systemic disease requiring chemotherapy; however, liver-only metastatic disease occurs in 5% to 12% of breast cancer patients and may be suitable for surgical metastasectomy. Early reports demonstrated improved survival rates (18%-24% at 5 years’ follow-up) in patients undergoing resection of limited hepatic metastases from breast cancer.21-23 More recent reports have yielded even superior survival rates (51% with disease-free interval and stage of the primary as prognostic factors).24 Therefore, surgery may allow long-lasting benefit and be potentially curative; resected patients with the liver as the first and sole site of relapse have a threefold greater median survival than patients treated with standard nonsurgical treatment. On this background, we used percutaneous RFA to achieve local control of liver metastases in patients with breast cancer; 24 patients with 64 liver metastases were treated, and in 92% of lesions, complete necrosis was achieved with a single treatment session (Fig. 142-5). Interestingly, the rate of complete necrosis was superior to that obtained with colorectal metastases of comparable size (96% for lesions < 3 cm and 75% for 3- to 5-cm lesions); these figures suggest that occult microangioinvasion of surrounding liver tissue may be absent or less pronounced in metastatic breast tumors.30 In more than half the patients (58%), new metastases developed during follow-up, but 63% of 16 patients with liver-only metastatic breast cancer were free of disease 4 to 44 months after the treatment. In view of the natural history of the disease and its minimal invasiveness, RFA is an effective local therapeutic alternative to surgical resection of liver metastases in breast cancer patients. Furthermore, RFA may be extremely useful for tumor recurrence after hepatectomy, because in more than half of these cases the liver is the only involved site of relapse. Combination treatment with hormonal therapy and systemic or intraarterial infusion chemotherapy is required.

CHAPTER 142 ENERGY-BASED ABLATION OF OTHER LIVER LESIONS 1053

A

B

FIGURE 142-5. A, A 1.9-cm subcapsular liver metastasis in segment V from breast carcinoma treated with percutaneous ultrasound-guided radiofrequency ablation after unsuccessful chemotherapy. B, Three-year follow-up contrast-enhanced computed tomography scan showed complete local control and marked shrinkage of necrotic area.

Liver metastasectomy from an endocrine primary tumor is followed by long survival in a substantial proportion of patients, with 56% to 74% 5-year survival rates.22,25-27 Interesting results have been reported with RFA used alone or in combination with surgery for local control of metastases from neuroendocrine tumors.28 Eradication of disease was achieved in 28.5% of patients treated with curative intent.29

COMPLICATIONS Major complications such as hemorrhage, cholecystitis, and gastric or bowel wall involvement occur in a small percentage of cases (range, 0.7% to 2.6%). Reported mortality rates are usually lower than 0.5%. In our experience, patients usually experience mild to moderate pain in the right side of the abdomen that lasts for 2 to 5 days; the pain may become severe in patients with subcapsular lesions, but pericapsular injection of a longacting anesthetic (ropivacaine) immediately after the procedure for postoperative pain control can obviate this problem. Fever may be present for the first 2 or 3 days after treatment and recedes with antipyretic drugs (acetaminophen). No impairment in liver function test results has been observed. Minimal pleural effusion may be documented by CT performed at 24 hours and usually resolves within a week.

POSTPROCEDURAL AND FOLLOW-UP CARE Antiemetic therapy and analgesia are offered on demand. Laboratory tests include CBC, coagulation parameters, and a liver function test the day after the procedure. RFA is considered successful when complete coagulative necrosis of the lesion and a safety margin are obtained.

Initial cross-sectional examination is crucial to assess the completeness of treatment, exclude complications, and provide baseline imaging for follow-up purposes. Spiral contrast-enhanced CT (or MRI) at 24 hours is the most widely used technique in follow-up schemes. Other methods, especially diffusion-weighted MRI combined with contrast-enhanced MRI, are being performed with the goal of increasing sensitivity in the evaluation of necrotic tissue. Recent studies have demonstrated diffusion-weighted imaging to be accurate in distinguishing between viable tumor tissue and necrosis.58,59 Patients continue to undergo contrast-enhanced multidetector CT or MRI on a routine basis in the long-term assessment of response to therapy to allow prompt identification of new lesions. Cross-sectional imaging studies obtained at 3- to 4-month intervals must be integrated with liver function tests and serum CEA levels to detect local or distant recurrence. In our experience, contrast-enhanced ultrasonography has also proved valuable in the follow-up setting to confirm local recurrences/new metastases and plan further treatment.48 Nevertheless, although effort is continually being expended to increase the resolution and accuracy of imaging techniques, residual peripheral microscopic foci of malignancy or foci within the treated lesion may go undetected and give rise to local recurrence in the short term. Fluorodeoxyglucose (FDG)-labeled positron emission tomography (PET) is a complementary imaging modality in the event of uncertain treatment response when applied after a reasonable time interval.60 Areas of abnormal FDG uptake after ablative procedures have been reported to represent disease relapse or residual viable tumor after ablation with a high degree of sensitivity.50,51 In some institutions, it may be proposed as a whole-body surveillance technique in treated patients. When diffusion-weighted MRI is readily available and routinely performed, it has become a valuable diagnostic option.62

1054 PART 2 NONVASCULAR INTERVENTIONS · SECTION 21 THE LIVER KEY POINTS •

Because thermal ablation is a local treatment, metastases that can undergo ablation are those originating from tumors that mostly (if not even exclusively) metastasize to the liver (e.g., colorectal malignancies) or slow-growing metastases from tumors treated by chemotherapy with incomplete local control of the disease (e.g., endocrine and breast malignancies).

•

Accurate selection of patients and lesions to be treated and use of state-of-the-art technology for guiding and performing ablation are of crucial importance for achieving good results.

•

In experienced hands, thermal ablation can achieve local control of liver metastases in up to 85% to 90% of treated lesions. Repeated thermal ablation for either local recurrence or new metastases is technically feasible and very often performed.

•

In long-term studies, survival rates of patients who have undergone thermal ablation can equal those of patients who have undergone surgical resection.

•

Adverse effects, cost, and duration of hospitalization after thermal ablation are usually lower than after surgical resection.

▶ SUGGESTED READINGS Atwell TD, Charboneau JW, Nagorney DM, Que FG. Radiofrequency ablation of neuroendocrine metastases. In: vanSonnenberg E, McMullen W, Solbiati L, editors. Tumor Ablation. Principles and Practice. New York: Springer; 2005. p. 332–41. Cheung L, Livraghi T, Solbiati L, et al. Complications of tumor ablation. In: vanSonnenberg E, McMullen W, Solbiati L, editors. Tumor Ablation. Principles and Practice. New York: Springer; 2005. p. 440–58. Fahy BN, Jarnagin WR. Evolving techniques in the treatment of liver colorectal metastases: role of laparoscopy, radiofrequency ablation, microwave coagulation, hepatic arterial chemotherapy, indications and contraindications for resection, role of transplantation, and timing of chemotherapy. Surg Clin North Am 2006;86(4):1005–22. Solbiati L, Ierace T, Tonolini M, Cova L. Ablation of liver metastases. In: vanSonnenberg E, McMullen W, Solbiati L, editors. Tumor Ablation. Principles and Practice. New York: Springer; 2005. p. 311–21. Solbiati L, Ahmed M, Cova L, et al. Small liver colorectal metastases treated with percutaneous radiofrequency ablation: local response rate and long-term survival with up to 10-year follow-up. Radiology 2012;265(3):958–68.

The complete reference list is available online at www.expertconsult.com.

CHAPTER 142 ENERGY-BASED ABLATION OF OTHER LIVER LESIONS1054.e1 REFERENCES 1. Bhattacharya R, Rao S, Kowdley KV. Liver involvement in patients with solid tumors of nonhepatic origin. Clin Liver Dis 2002;6:1033–43. 2. Nordlinger B, Rougier P. Nonsurgical methods for liver metastases including cryotherapy, radiofrequency ablation and infusional treatment: what’s new in 2001? Curr Opin Oncol 2002;14:420–3. 3. Sica GT, Ji H, Ros PR. CT and MR imaging of hepatic metastases. AJR Am J Roentgenol 2000;174:691–8. 4. Valls C, Andia E, Sanchez A, et al. Hepatic metastases from colorectal cancer: preoperative detection and assessment of resectability with helical CT. Radiology 2001;218:55–60. 5. Harvey CJ, Blomley MJK, Eckersley RJ, et al. Hepatic malignancies: improved detection with pulse-inversion US in the late phase of enhancement with SHU 508A—early experience. Radiology 2000;216: 903–8. 6. Albrecht T, Hoffmann CW, Schmitz SA, et al. Phase-inversion sonography during the liver specific late phase of contrast enhancement: improved detection of liver metastases. AJR Am J Roentgenol 2001; 176:1191–8. 7. Quaia E, Bertolotto M, Forgacs B, et al. Detection of liver metastases by pulse inversion harmonic imaging during Levovist late phase: comparison with conventional ultrasound and helical CT in 160 patients. Eur Radiol 2003;13:475–83. 8. Solbiati L, Tonolini M, Cova L, et al. The role of contrast-enhanced ultrasound in the detection of focal liver lesions. Eur Radiol 2001; 11(Suppl 3):E15–26. 9. Albrecht T, Blomley MJ, Burns PN, et al. Improved detection of hepatic metastases with pulse-inversion US during the liver-specific phase of SHU 508A: multicenter study. Radiology 2003;227:361–70. 10. Semelka RC, Helmberger TC. Contrast agents for MR imaging of the liver. Radiology 2001;218:27–38. 11. Geoghegan JG, Scheele J. Treatment of colorectal liver metastases. Br J Surg 1999;86:158–69. 12. Yoon SS, Tanabe KK. Surgical treatment and other regional treatments for colorectal cancer liver metastases. Oncologist 1999;4:197–208. 13. Taylor M, Forster J, Langer B, et al. A study of prognostic factors for hepatic resection for colorectal metastases. Am J Surg 1997;173: 467–71. 14. Nordlinger B, Guigner M, Vaillant JC, et al. Surgical resection of colorectal carcinoma metastases to the liver. A prognostic scoring system to improve case selection based on 1568 patients, Association Française de Chirurgie. Cancer 1996;77:1254–62. 15. Elias D, Cavalcanti A, Sabourin JC, et al. Resection of liver metastases from colorectal cancer: the real impact of the surgical margin. Eur J Surg Oncol 1998;24:174–9. 16. Cady B, Jenkins RL, Steelde GD, et al. Surgical margin in hepatic resection for colorectal metastasis: a critical and improvable determinant of outcome. Ann Surg 1998;227:566–71. 17. Holm A, Bradley E, Aldrete JS. Hepatic resection of metastasis from colorectal carcinoma: morbidity, mortality and patterns of recurrence. Ann Surg 1989;209:428–34. 18. Fong Y, Salo J. Surgical therapy of hepatic colorectal metastasis. Semin Oncol 1999;26:514–23. 19. Liu LX, Zhang WH, Jiang HC. Current treatment for liver metastases from colorectal cancer. World J Gastroenterol 2003;9:193–200. 20. Hugh TJ, Kinsella AR, Poston GJ. Management strategies for colorectal liver metastases—part I. Surg Oncol 1997;6:19–30. 21. Elias D, Lasser PH, Montrucolli D, et al. Hepatectomy for liver metastases from breast cancer. Eur J Surg Oncol 1995;21:510–3. 22. Elias D, Cavalcanti A, Eggenspieler P, et al. Resection of liver metastases from a noncolorectal primary: indications and results based on 147 monocentric patients. J Am Coll Surg 1998;187:487–93.

23. Raab R, Nussbaum KT, Behrend M, et al. Liver metastases of breast cancer: results of liver resection. Anticancer Res 1998;18:2231–3. 24. Maksan SM, Lehnert T, Bastert G, et al. Curative liver resection for metastatic breast cancer. Eur J Surg Oncol 2000;26:209–12. 25. Chen H, Hardacre JM, Uzar A, et al. Isolated liver metastases from neuroendocrine tumors: does resection prolong survival? J Am Coll Surg 1998;187:88–92; discussion 92–93. 26. Berney T, Mentha G, Roth AD, et al. Results of surgical resection of liver metastases from non-colorectal primaries. Br J Surg 1998;85: 1423–7. 27. Lindell G, Ohlsson B, Saarela A, et al. Liver resection of noncolorectal secondaries. J Surg Oncol 1998;69:66–70. 28. Wessels FJ, Schell SR. Radiofrequency ablation treatment of carcinoid hepatic metastases. J Surg Res 2001;95:8–12. 29. Hellman P, Ladjevardi S, Skogseid B, et al. Radiofrequency tissue ablation using cooled tip for liver metastases of neuroendocrine tumors. World J Surg 2002;26:1052–6. 30. Livraghi T, Goldberg SN, Solbiati L, et al. Percutaneous radiofrequency ablation of liver metastases from breast cancer: initial experience in 24 patients. Radiology 2001;220:145–9. 31. Goldberg SN, Gazelle GS, Mueller PR. Thermal ablation therapy for focal malignancy. A unified approach to underlying principles, techniques, and diagnostic imaging guidance. AJR Am J Roentgenol 2000;174:323–31. 32. Solbiati L, Goldberg SN, Ierace T, et al. Hepatic metastases: percutaneous radio-frequency ablation with cooled-tip electrodes. Radiology 1997;205:367–73. 33. Solbiati L, Ierace T, Goldberg SN, et al. Percutaneous US-guided radiofrequency tissue ablation of liver metastases: Treatment and follow-up in 16 patients. Radiology 1997;202:195–203. 34. Livraghi T, Goldberg SN, Lazzaroni S, et al. Small hepatocellular carcinoma: treatment with radiofrequency ablation versus ethanol injection. Radiology 1999;210:655–61. 35. Livraghi T, Goldberg SN, Lazzaroni S, et al. Hepatocellular carcinoma: radio-frequency ablation of medium and large lesions. Radiology 2000;214:761–8. 36. Solbiati L, Livraghi T, Goldberg SN, et al. Percutaneous radiofrequency ablation of hepatic metastases from colorectal cancer: long-term results in 117 patients. Radiology 2001;221:159–66. 37. Curley SA, Izzo F, Delrio P, et al. Radiofrequency ablation of unresectable primary and metastatic hepatic malignancies: results in 123 patients. Ann Surg 1999;230:1–8. 38. Jiao LR, Hansen PD, Havlik R, et al. Clinical short term results of radiofrequency ablation in primary and secondary liver tumors. Am J Surg 1999;177:303–6. 39. Lencioni R, Goletti O, Armillotta N, et al. Radiofrequency thermal ablation of liver metastasis with cooled-tip electrode needle: results of a pilot clinical trial. Eur Radiol 1998;8:1205–11. 40. Rose DM, Allegra DP, Bostick PJ, et al. Radiofrequency ablation: a novel primary and adjunctive ablative technique for hepatic malignancies. Am Surg 1999;65:1009–14. 41. Rossi S, Buscarini E, Garbagnati F, et al. Percutaneous treatment of small hepatic tumors by an expandable RF needle electrode. AJR Am J Roentgenol 1998;170:1015–22. 42. De Baere T, Elias D, Dromain C, et al. Radiofrequency ablation of 100 hepatic metastases with a mean follow-up of more than 1 year. AJR Am J Roentgenol 2000;175:1619–25. 43. Goldberg SN, Gazelle GS, Solbiati L, et al. Radiofrequency tissue ablation: increased lesion diameter with a perfusion electrode. Acad Radiol 1996;3:636–44. 44. Goldberg SN, Solbiati L, Hahn PF, et al. Large-volume tissue ablation with radiofrequency by using a clustered, internally cooled electrode

1054.e2PART 2 NONVASCULAR INTERVENTIONS · SECTION 21 THE LIVER technique: laboratory and clinical experience in liver metastases. Radiology 1998;209:371–9. 45. Goldberg SN, Stein MC, Gazelle GS, et al. Percutaneous radiofrequency tissue ablation: optimization of pulsed-radiofrequency technique to increase coagulation necrosis. J Vasc Interv Radiol 1999;10: 907–16. 46. Livraghi T, Goldberg SN, Monti F, et al. Saline-enhanced radiofrequency tissue ablation in the treatment of liver metastases. Radiology 1997;202:205–10. 47. DeBaere T, Bessoud B, Dromain C, et al. Percutaneous radiofrequency ablation of hepatic tumors during temporary venous occlusion. AJR Am J Roentgenol 2002;178:53–9. 48. Rhim H, Goldberg SN, Dodd 3rd GD, et al. Essential techniques for successful radio-frequency thermal ablation of malignant hepatic tumors. Radiographics 2001;21(Spec No):S17–35; discussion S36–S39. 49. Goldberg SN, Kamel IR, Kruskal JB, et al. Radiofrequency ablation of hepatic tumors: increased tumor destruction with adjuvant liposomal doxorubicin therapy. AJR Am J Roentgenol 2002;179:93–101. 50. Akhurst T, Larson SM. Positron emission tomography imaging of colorectal cancer. Semin Oncol 1999;26:577–83. 51. Fong Y, Fortner J, Sun RL, et al. Clinical score for predicting recurrence after hepatic resection for metastatic colorectal cancer: analysis of 1001 consecutive cases. Ann Surg 1999;230:309–21. 52. Registry of Hepatic Metastases. Resection of the liver for colorectal carcinoma metastases: a multi-institutional study of indications for resection. Surgery 1998;103:278–88. 53. Scheele J, Stang R, Altendorf-Hofmann A, et al. Resection of colorectal carcinoma metastases to the liver: a prognostic scoring system to improve case selection, based on 156 patients. Cancer 1996;77: 1254–62. 54. Elias D, De Baere T, Smayra T, et al. Percutaneous radiofrequency thermoablation as an alternative to surgery for treatment of liver tumour recurrence after hepatectomy. Br J Surg 2002;89:752–6. 55. Leen E, Horgan PG. Radiofrequency ablation of colorectal liver metastases. Surg Oncol 2007;16:47–51. 56. Machi J, Oishi AJ, Sumida K, et al. Long-term outcome of radiofrequency ablation for unresectable liver metastases from colorectal cancer: evaluation of prognostic factors and effectiveness in first-and second-line management. Cancer J 2006;12:318–26.

57. Livraghi T, Solbiati L, Meloni F, et al. Percutaneous radiofrequency ablation of liver metastases in potential candidates for resection: the “test of time” approach. Cancer 2003;97:3027–35. 58. Vossen JA, Buijs M, Kamel IR. Assessment of tumor response on MR imaging after locoregional therapy. Tech Vasc Interv Radiol 2006;9: 125–32. 59. Taouli B, Vilgrain V, Dumont E, et al. Evaluation of liver diffusion isotropy and characterization of focal hepatic lesions with two singleshot echo-planar MR imaging sequences: prospective study in 66 patients. Radiology 2003;226:71–8. 60. Anderson GS, Brinkmann F, Soulen MC, et al. FDG positron emission tomography in the surveillance of hepatic tumors treated with radiofrequency ablation. Clin Nucl Med 2003;28:192–7. 61. Solbiati L, Ahmed M, Cova L, et al. Small liver colorectal metastases treated with percutaneous radiofrequency ablation: local response rate and long-term survival with up to 10-year follow-up. Radiology 2012;265(3):958–68. 62. Koh DM, Scurr E, Collins DJ, et al. Colorectal hepatic metastases: quantitative measurements using single-shot echo-planar diffusionweighted MR imaging. Eur Radiol 2006;16:1898–905. 63. Berber E, Pelley R, Siperstein AE. Predictors of survival after radiofrequency thermal ablation of colorectal cancer metastases to the liver: a prospective study. J Clin Oncol 2005;23:1358–64. 64. Gillams AR, Lees WR. Radio-frequency ablation of colorectal liver metastases in 167 patients. Eur Radiol 2004;14:2261–7. 65. Veltri A, Moretto P, Doriguzzi A, et al. Radiofrequency thermal ablation (RFA) after transarterial chemoembolization (TACE) as a combined therapy for unresectable non-early hepatocellular carcinoma (HCC). Eur Radiol 2006;16(3):661–9. 66. Jakobs TF, Hoffmann RT, Trumm C, et al. Radiofrequency ablation of colorectal liver metastases: mid-term results in 68 patients. Anticancer Res 2006;26:671–80. 67. Lencioni R, Crocetti L, Cioni D, et al. Percutaneous radiofrequency ablation of hepatic colorectal metastases: technique, indications, results, and new promises. Invest Radiol 2004;39:689–97. 68. Sorensen SM, Mortensen FV, Nielsen DT. Radiofrequency ablation of colorectal liver metastases: long-term survival. Acta Radiol 2007;48: 253–8.

CHAPTER

143

Cryoablation of Liver Tumors

Mark D. Mamlouk, Eric vanSonnenberg, Stuart G. Silverman, Paul R. Morrison, Charles D. Crum, and Kemal Tuncali

With the development and maturity of ablation techniques, radiologists now can treat a wide variety of solid tumors. There are several options when deciding which technique to use. Radiofrequency, microwave, laser, ethanol, and cryoablation all have been shown to destroy tumor tissue effectively.1-5 Although the choice of ablation strategy is usually based on institutional expertise and availability, these techniques (and other nonradiologic methods—i.e., surgery, chemotherapy, and radiation therapy) occasionally can be combined for a given patient, depending on the location and specifics of the target. This chapter focuses on the use of cryoablation to treat neoplasms that involve the liver.

INDICATIONS Cryoablation can be used to treat primary malignant liver lesions such as hepatocellular carcinoma (HCC), as well as benign lesions such as symptomatic hemangiomas and liver adenomas.10,11 Nonetheless, the most common use of ablation in the liver in the United States is for metastatic disease from colon, breast, and neuroendocrine primaries (Fig. 143-1). Because fewer than 15% of patients with metastatic liver tumors are candidates for curative surgical resection, percutaneous tumor ablation is a viable alternative. Likewise, a minority of patients with primary HCC are candidates for curative surgical resection. Tumors often are deemed surgically unresectable because too much liver parenchyma would be sacrificed, or the tumor lies too close to major biliary or vascular structures such that adequate margins cannot be obtained. Another restriction for surgery is various comorbid conditions, such as cardiopulmonary disease or nutritional status. A percutaneous approach allows treatment of these problematic tumors for surgery because the tissue destruction can be targeted precisely at the tumor and the immediately surrounding adjacent liver parenchyma, with preservation of a greater percentage of normal parenchyma.10 Likewise, cryoablation also may be able to induce tumor necrosis when vascular or biliary structures are challenging for the surgeon because of proximity of the tumor to vital structures in the hepatic hilum (although radiologists too must exercise great caution in these same anatomic regions). Cryoablation is indicated in patients who have pathologically proven metastatic lesions in which surgical cure is not an option. Surgery may not be feasible because of bilobar disease, limited hepatic reserve, or cardiopulmonary compromise. Cryoablation generally is used for one to five lesions that are confined to the liver and measure up to 5 cm in diameter. Most institutions consider the presence of metastatic disease outside the liver to be a relative contraindication to cryoablation, with the possible exception of lung or bone metastases in patients with breast cancer. Cryoablation is indicated for the treatment of residual

tumor or local recurrence in patients who have previously undergone treatment with surgery or ablation, as well as new metachronous lesions after primary treatment. Cryoablation also has been used to treat primary liver neoplasms, most commonly HCC. Indications for cryoablation of primary HCC are different from those for metastatic disease because of the underlying pathology. HCC is caused by underlying chronic liver disease that is usually secondary to hepatitis or alcoholic liver disease. Because of the underlying disease, the only definitive treatment of HCC is liver transplantation. As a result of the limited availability of donated livers and the high cost of transplantation, only a very small percentage of patients can undergo this option. Surgery can offer complete resection of the first HCC and any satellite lesions, but a second or more lesions may have already developed or will develop in the overwhelming majority of patients. If surgical resection was chosen to treat the primary lesion, only a fraction will be able to undergo a second resection.12 Ablation affords the possibility of complete destruction of the first and subsequent lesions, without significant loss of functioning liver parenchyma. We have treated patients up to five times percutaneously for secondarily diagnosed lesions. Strategically, patients undergo surveillance, and as new lesions appear, they too can undergo ablation. Cryoablation also has been used to treat benign liver lesions, specifically hemangiomas and adenomas.10,11 Adenomas are seen predominantly in younger women taking oral contraceptives; the tumors occasionally cause abdominal pain. Cessation of oral contraceptives is first-line therapy, but with persistent problems (especially pain), more invasive treatment may be necessary. Cryoablation is a valuable alternative because of the amount of liver that can be preserved and the low risks involved in the ablation procedure. Symptomatic focal nodular hyperplasia can be treated as well.

CONTRAINDICATIONS Percutaneous cryoablation is a relatively safe procedure, so there are few absolute contraindications to its use. As with most invasive procedures, the only absolute contraindications are an uncorrectable coagulopathy or an uncooperative patient. Contraindications to general anesthesia can be considered relative because the procedure can be done under intravenous conscious sedation if needed. However, intravenous conscious sedation is suboptimal if extended breath-holds will be necessary and the patient is unable to be completely cooperative. When cryoablation is used to treat liver masses, there are several hazardous areas where special care has to be taken to avoid damage to adjacent structures. Ablating liver lesions that are adjacent to the hepatic hilum, gallbladder, 1055

CHAPTER 143 CRYOABLATION OF LIVER TUMORS1055.e1 For several decades, cryoablation has been used surgically to treat malignancies.6 It was first described in the liver in 1963.7 Because of the large size and assorted geometries of the probes, the surgeon would place them from an open or laparoscopic approach with the aid of intraoperative ultrasound. However, the development of thin needle-like probes and the use of computed tomography (CT) and magnetic resonance imaging (MRI) have allowed safe and

effective use of percutaneous cryoablation.6 Currently, percutaneous cryoablation is used to treat a variety of primary neoplasms throughout the body, including those in the liver, kidney, lung, prostate gland, and musculoskeletal systems. The mechanism of tumor kill is multifactorial, but includes predominantly cell dehydration, cell membrane rupture, and ischemia, the latter due to vascular stasis and thrombosis.8,9

1056 PART 2 NONVASCULAR INTERVENTIONS · SECTION 21 THE LIVER

A

C

B

D

FIGURE 143-1. A, Axial contrast-enhanced magnetic resonance imaging (MRI) shows a large minimally enhancing liver metastasis (arrow) from a primary breast cancer. B, T2 oblique sagittal image shows intraprocedural MRI-guided placement of multiple cryoprobes (arrows) into hyperintense tumor. C, T2 oblique sagittal image shows a hypointense iceball after 15 minutes of freezing. D, Axial contrast-enhanced MRI shows treated tumor surrounded by a nonenhancing margin of ablated normal tissue (arrowheads) 24 hours after procedure.

stomach, kidney, pancreas, and colon can be accomplished, but careful planning and execution are essential. Vascular structures are relatively resistant to cryoablation because of the warming effects of flowing blood. However, the biliary system is vulnerable, especially if harboring bacteria. If the lesion to be ablated is juxtaposed to major branches of the biliary system, warm fluid can be circulated through the system via a catheter placed percutaneously or by the endoscopic route. Care also must be taken not to cause damage to structures outside the liver, such as adjacent stomach or bowel. Sometimes patient positioning (i.e., prone versus supine) can provide adequate distance to avoid injury to critical structures. If positioning of the patient does not provide adequate safety of adjacent structures, sterile water, dextrose, or a normal saline barrier can be interposed (Fig. 143-2). This is accomplished percutaneously by injecting sterile fluid through a fine needle between the lesion and the bowel to be displaced, termed hydrodissection.13,14 Overall, when care is taken in both planning and execution of cryoablation, it is a relatively safe procedure.

EQUIPMENT During the early days of cryosurgery, the probes were cooled with liquid nitrogen. Because these LN2 probes were relatively large (3- to 8-mm in outer diameter), for years cryoablation was practical only in a surgical setting.15 Because the size of the probe was less than optimal, a complicated Seldinger technique was used that required several minutes to position and reposition the probes. Even though these probes were effective in causing tissue destruction, the probe size and handling of the liquid nitrogen were negative features in light of the now contemporary gasbased systems. The argon gas–based systems that were developed in the 1990s have virtually replaced liquid nitrogen devices. Argon systems work on the Joule-Thompson principle, in which low temperatures are achieved by the rapid expansion of a high-pressure gas through a thin aperture.15 Because of the low viscosity of argon gas, it can be circulated rapidly within the probe tip. This allows faster cooling at the probe tip, which hastens removal and repositioning (if necessary) of

CHAPTER 143 CRYOABLATION OF LIVER TUMORS 1057

L I I

L

C

A

B

FIGURE 143-2. Image-guided control demonstrated in two separate ablation procedures. A, Intraprocedural axial T1 magnetic resonance imaging (MRI) shows critical relationships several minutes into a multiprobe freeze of a liver metastasis in dome of liver. Iceball (I) within liver (L) is observed as its lateral edge approaches adjacent lung (arrows). Such clear visualization of field allows radiologist to adjust rate of freezing to protect normal tissues. B, Intraprocedural coronal T2 MRI provides guidance to protect adjacent structures by additional interventional techniques. Multiprobe freezing of a metastasis is conducted in liver (L). Colon (C) is protected from iceball (I) because it has been distanced by injection of saline through additional percutaneous needles (not seen in image plane). Saline is seen here as hyperintense region between ice (I) and colon (C).

the cryoprobes. After the freezing phase is complete, highpressure helium gas can then be circulated through the probe tip to cause rapid thawing. Helium provides an effect opposite to that of argon gas under the Joule-Thompson principle, causing a warming of the probe to allow rapid thawing for probe removal from the frozen tissue or repositioning. There are currently two major manufacturers of cryoablation equipment: Galil Medical (Yokneam, Israel) and EndoCare Inc. (Endo Pharmaceuticals, Chadds Ford, Pa.). Both companies use an argon/helium-based system to cool and thaw the probe tip, respectively, and have stand-alone consoles that are connected to gas tanks. Both systems have a similar computer-based interface that demonstrates the vital statistics of the different probes and their respective temperatures (Fig. e143-1). Currently, Galil Medical is the only company to offer a system that is MRI compatible. Galil Medical and EndoCare cryoprobes for percutaneous ablation are needle-like, 17-gauge in size. The Galil system can accommodate up to 25 probes; the EndoCare system up to 8 probes. All three main cross-sectional imaging techniques can be used to guide percutaneous placement of cryoprobes for ablation. Ultrasound, CT, and MRI each has advantages and disadvantages. Ultrasound was first used in the early 1980s; its utility has since expanded to guide percutaneous cryoablation. Ultrasound allows real-time imaging during probe placement, and therefore major structures such as the gallbladder, large bile ducts, and larger hepatic vessels are well visualized and can be avoided. The iceball created during cryoablation is intensely echogenic on ultrasound; this allows real-time monitoring as the iceball forms and propagates. Therefore, damage to the surrounding normal liver parenchyma is limited. The main drawback of ultrasound is the distal acoustic shadow produced by the iceball. Because of the echogenicity of the iceball, ultrasound cannot penetrate beyond it, and

structures distal to the iceball are not visualized because of acoustic shadowing. This can be problematic when trying to avoid critical structures that are deep to the tumor being ablated. Reverberation artifact and the critical angle effect are also troublesome when ultrasound guidance is used. The critical angle effect can cause the iceball to appear speciously larger, whereas reverberation causes apparent echoes within the iceball.16 The critical angle effect is due to variation in the speed of sound as it travels through soft tissue versus ice, and is more apparent when using a vector transducer than a linear transducer. CT has many attributes that make it an appealing imaging choice for ablation. CT is widely available and is often primarily used to detect and monitor lesions that have been ablated. CT also shows the anatomy of the liver and adjacent structures well. This allows identification of adjacent structures that may need to be avoided, such as major vessels, the gallbladder, and bowel. In soft tissue, such as liver parenchyma and solid tumors, the iceball is generally well seen as a hypodense region (Fig. 143-3). The margins of cell death and the visible iceball also correlate well when using CT.17-19 One drawback of the use of CT for imageguided ablation is exposure of the patient and personnel to ionizing radiation. In a retrospective study of 20 CT-guided percutaneous liver tumor cryoablations, the total effective dose for the procedure was 72 ± 18 mSv.20 Although that report does suggest ways to reduce such exposures, it calls attention to an added risk with CT guidance. Of course, the uncertain risk due to the radiation is weighed against the imminent health risks associated with the disease state for which the ablation is being utilized. One technological assist to identifying a tumor location in unenhanced CT images is to “register” (map; overlay) preprocedural MRI images with an intraprocedural CT scan. Contemporary efforts use a technique of nonrigid registration to elastically morph the MRI into fitting the CT scan. The resulting information-rich MRI-CT can improve

CHAPTER 143 CRYOABLATION OF LIVER TUMORS1057.e1

FIGURE e143-1. Cryoablation equipment. Contemporary cryoablation systems circulate high-pressure argon gas (and helium) to cryoprobes with capability of multiple-probe treatments. Here, several 17-gauge probes (large arrow) are used for percutaneous computed tomography–guided hepatic tumor ablation. Computer interface (monitor [arrowhead]) provides feedback on ablation temperatures and elapsed time, and allows for individual control of individual probes.

1058 PART 2 NONVASCULAR INTERVENTIONS · SECTION 21 THE LIVER

A

B

C

FIGURE 143-3. Image-guided cryoablation. A, Axial contrast-enhanced magnetic resonance imaging (MRI) identifies two enhancing liver metastases (arrows). B, Unenhanced axial computed tomography (CT) shows cryoprobes (arrow) placed under CT guidance. Iceball appears as a hypodense rounded shape (arrowheads) around probes. C, Axial contrast-enhanced postprocedure MRI at 24 hours shows corresponding nonenhancing regions of ablation covering tumors.

tumor visualization during interventional planning, targeting, and monitoring.21,22 In recent years, MRI has been used for a variety of interventional procedures and ablations.10,23 In one institution’s experience of over 300 cryoablation procedures, two thirds were MRI guided.9 MRI has a distinct advantage over CT in that the tumor is nearly always distinguishable from normal liver parenchyma, without the aid of intravenous contrast agents.24 The multiplanar capability of MRI is another advantage. The cryoprobes can be accurately placed and verified in all three planes—axial, sagittal, and coronal. Cryoablation can be performed in traditional MRI scanners with the development of MRI-compatible probes; however, with open or wide-bore systems, the radiologist has sufficient access to the patient and can guide the position of the probes during cryoablation.10,25 This allows nearly real-time imaging during ablation and greater accuracy in controlling the developing iceball.10,26 Another benefit of MRI is assessing temperature within the iceball and tissues because of T1-weighted imaging characteristics.27,28 Like ultrasound and CT, the visible iceball on MRI corresponds well to the actual ablative zone size.29,30 However, MRI demonstrates the demarcations from iceball to tumor to normal parenchyma without any interference or artifact, unlike ultrasound and CT. For information on the use of PET and PET-CT in cryoablation and radiofrequency ablation, please visit www.expertconsult.com.

TECHNIQUE Anatomy and Preprocedural Care Thorough knowledge of hepatic anatomy is essential to safely ablate primary and metastatic hepatic neoplasms. Hepatobiliary surgeons have used the Couinaud system since the 1950s.32 This system is used to divide the liver into eight segments that are supplied by the portal triad and drained by the hepatic veins. The plane that divides the inferior vena cava (IVC) and the middle hepatic vein separates the left lobe from the right lobe of the liver and is

known as Cantlie’s line. The left lobe of the liver is divided into segments II, III, and IV, with the umbilical fissure dividing segments II and III from IV. The right lobe of the liver is divided into segments V, VI, VII, and VIII. The plane formed by the IVC and the right hepatic vein separates the anterior segments V and VIII from the posterior segments VI and VII. The portal veins divide the liver into upper and lower segments. The caudate lobe is considered segment I and receives portal flow from both the right and left sides. This knowledge allows the radiologist to communicate with surgeons more appropriately about available options when considering surgery or ablation. Hepatic resection varies from single or double segmentectomies (II and III, IV and V, or VI and VII) to extended lobectomies or trisegmentectomies that include parenchyma from both lobes of the liver. One of the most important principles in all of interventional radiology is to define a safe access route before beginning the procedure. This requires the identification of all key structures before commencing liver ablation. First, care must be taken to avoid any bowel that may be interposed between the liver and the abdominal wall, the so-called Chilaiditi variant. The liver also has an intimate relationship with the diaphragm. Care must be taken to avoid the diaphragm and not cause a pneumothorax, diaphragmatic injury, or other intrathoracic complications. Planning should also include identification and avoidance of the larger vascular structures (IVC, portal and hepatic veins), porta hepatis, and central biliary ducts. With regard to antibiotic prophylaxis, we administer 1 g of cefazolin sodium (Ancef [SmithKline Beecham Pharmaceuticals, Philadelphia, Pa.]) intravenously before each procedure, and every 8 hours thereafter for a total of three doses after the procedure.33 However, there are different views regarding the need for routine antibiotic usage with ablation, and without clear consensus.

Technical Aspects Before starting an ablation procedure, appropriate imaging (CT, ultrasound, or MRI) is performed to identify the target

CHAPTER 143 CRYOABLATION OF LIVER TUMORS1058.e1 PET and PET-CT have been proposed as image guidance modalities for cryoablation and radiofrequency ablation (RFA). The role of fluorine-18 fluorodeoxyglucose (18FFDG) intraprocedurally is the subject of nascent research of the general hypothesis that nuclear medicine imaging could assist in interventions. In a recent study of 12 patients having undergone thermal ablation (3 of which were cryoablation) in the presence of 18F-FDG, no decrease in the

avidity of the 18F-FDG was observed in light of the exposure to the extreme temperatures of the treatments. Thus, although the tracer may be useful in targeting the metabolically active target, it may serve little use in monitoring the ablation or assessing acute effects.31 The utility of a postablation injection, of course, may hold some role; certainly tracers with half-lives shorter than FDG may open new paradigms for ablation.

CHAPTER 143 CRYOABLATION OF LIVER TUMORS 1059

FIGURE 143-4. Monitoring ablation over time. That cryoablation is a dynamic process is evident in coronal T2 magnetic resonance imaging (MRI) acquired in oblique sagittal plane during treatment of a liver metastasis. First image shows three cryoprobes (arrowheads) in cross-section situated within tumor. Probes are spaced so individual iceball from each probe can work synergistically to form one large, hypointense iceball (last image) to cover target. Images between show same image plane at intervals of approximately 2 to 4 minutes. Although MRI and computed tomography do not provide images in real time, it is possible to scan on a time frame that delineates rate of iceball growth.

and define the safest access route. A small, 22- to 25-gauge needle is used for biopsy if not previously done. Another needle is left in the proper location in the tumor to be ablated as a marker, and the cryoprobe is inserted into the neoplasm by using the tandem technique adjacent to the marker. Alternatively, a modified Seldinger technique can be used. Additional cryoprobes can be placed for adequate coverage of larger lesions. One advantage of multiple small cryoprobes is a more uniform and predictable freeze, with improved control of the iceball.15 After proper placement of the cryoprobes, two 15-minute freezes separated by a 10-minute thaw are completed to achieve sufficient necrosis of the lesion. The freeze cycle routinely reaches temperatures between −110°C and −120°C, whereas the thaw cycle temperature rises to approximately 40°C. Imaging can be continued throughout the procedure to monitor iceball growth (Fig. 143-4). After successful ablation of the lesion and complete thawing, the probe(s) is removed. Postprocedural imaging is performed to document the adequacy of ablation and whether additional freeze/thaw cycles are needed.

OUTCOMES Many metrics are used to measure success when evaluating an innovative approach to the care of a patient with cancer. The gold standard by which all treatments are measured is the 5-year survival rate. In a study of 420 patients with unresectable HCC, comparisons were made with a subset that had transarterial chemoembolization (TACE) plus cryoablation, while another group had cryoablation alone. The 5-year survival rates were 39% in the combined group versus 23% in the cryoablation-alone group.34 In another study of percutaneous ablation of 20 patients with HCC less than 3 cm in diameter, complete response was achieved in 65%, 20% partial response, and 15% stable disease.35 The initial measure of ablation is assessed on immediate postablation scans as the percentage of tumor destroyed. In two studies, the initial success of hepatic tumor cryoablation was 60% to 85%.34,36 Tumor size is a key factor in adequacy of therapy. In the first study, HCCs were ablated with the aid of percutaneous ethanol,34 whereas in the latter study,36 all unresectable hepatic malignancies were ablated without any adjunctive therapy. Local recurrence at follow-up ranges from 3% to 53%.37 There is evidence in the surgical literature that open or laparoscopic cryoablation does not increase the 2- or 5-year

survival rate in metastatic colon cancer over standard surgical treatment, even though local recurrence is reduced and a greater amount of normal liver parenchyma is preserved with cryoablation alone.38 Several studies also have evaluated surgical cryoablation of neuroendocrine metastases to the liver; the endpoint for success was relief of symptoms after ablation and reduction of serum markers.39,40

COMPLICATIONS The mortality rate with surgical cryoablation ranges from 1.2% to 7.5%, with the major complication rate from 5.9% to 45%.41-43 There is a strong correlation between the volume of liver ablated and major complications. Studies have demonstrated that when approximately 30% to 35% of the liver volume or more is ablated, a dramatic increase in complications occurs.41-43 Ablation of tumors larger than 5 cm, or an area of liver 30 cm2 or greater, also increases the major complication rate.44-46 Major complications include parenchymal fracture, coagulopathy, acute renal failure secondary to myoglobinuria, and cryoshock. Cryoablation hemorrhage is attributed mainly to fracture of the hepatic parenchyma and capsule, whereas coagulopathy plays a minor role. Hemorrhage results from parenchymal fracture that extends from the center of the iceball through the hepatic vessels, either arterial or venous, to the liver capsule. Limiting the torque or movement of cryoprobes during freezing and thawing cycles can minimize this complication. This is extremely important when the freeze cycle is near its end and the liver is coldest. Care also must be taken to allow complete thawing of the iceball before removal of the probe when the liver is most vulnerable to cracking, which was a risk with cryosurgery. The least understood cause of cryoablation-related mortality has been described as cryoshock. This is a constellation of acute respiratory distress syndrome, thrombocytopenia, disseminated intravascular coagulation (DIC), hypotension, and multiorgan failure.47 Cryoshock has been likened to septic shock without an identifiable infectious source. This syndrome was seen in 21 of 2173 patients following cryosurgery; 6 of these 21 died postprocedurally.43 This phenomenon appears to be mostly specific to cryoablation, but a recent study has shown that a moderate systemic inflammatory response can occur with RFA.48 Cryoshock has occurred after ablation of both primary liver tumors and metastases. Cryoshock is related to the volume of liver or

1060 PART 2 NONVASCULAR INTERVENTIONS · SECTION 21 THE LIVER tumor ablated (or both) and the number of freeze/thaw cycles. Minor complications may occur after cryoablation. Patients routinely experience mild to moderate right upper quadrant pain that is treated with narcotic analgesics. Most minor complications from ablation are similar to those that occur with biopsy of tumors: bleeding, diaphragmatic injury, pneumothorax, hepatic abscess (sometimes delayed), and subcapsular hematoma.36,42,49,50 Injury to the biliary system can occur during cryoablation. The smaller blood vessels of the liver are warmed by blood, but bile ducts do not have this protection. Therefore, specific care must be taken to avoid injury to the biliary system. Warm saline, sterile water, or 5% dextrose in water can be circulated through a catheter placed in the bile ducts percutaneously or endoscopically to protect the main bile ducts.51 Hypothermia can also be associated with cryoablation; it can be avoided with routine use of the Bair Hugger heating blanket (Arizant Healthcare, Prairie, Minn.) throughout the procedure; this problem is more likely to occur in small children.14 Of concern during cryoablation in the liver is the production of myoglobin. Myoglobinemia may develop but usually is transient, and uncommonly severe enough to require prophylactic treatment.33 Rarely, myoglobinemia can progress to acute tubular necrosis and renal failure from myoglobinuria.41,52 Ensuring proper hydration before the procedure, as well as use of mannitol diuresis and alkalinization of urine if myoglobinuria develops, can help protect the kidneys from this complication. Lastly, selflimited biochemical and hematologic changes commonly occur after percutaneous cryoablation of liver tumors and are usually milder than those reported after liver cryosurgery.33 However, as has been suggested following cryosurgery, percutaneous cryoablation can cause thrombocytopenia and even DIC. The amount of decrease in the platelet count correlates with the amount of normal parenchyma sacrificed and included in the ablation. Necrosis of normal parenchyma is associated with a rise in aspartate aminotransferase (AST) and alanine aminotransferase (ALT) and may be a harbinger of thrombocytopenia.33 Because of this phenomenon, we have often considered other ablative agents, such as RFA, when planning the ablation of multiple tumors in which large amounts of normal liver parenchyma are expected to be included in the ablation zones. Several other causes of mortality seen after cryoablation are not specific to this method: postablation myocardial infarction, pulmonary embolism, sepsis, colonic perforation, and hepatic insufficiency. There also have been reports of hepatorenal syndrome, portal vein thrombosis, strangulated bowel, variceal hemorrhage, and rupture of the ablated area into the abdominal cavity, resulting in postablation death.*

POSTPROCEDURAL AND FOLLOW-UP CARE Immediately after the cryoablation procedure, we perform contrast-enhanced CT or MRI to assess the effectiveness of ablation. Patients are subsequently sent to the postprocedure unit for 2 hours to recover from the effects of general anesthesia. They are then transferred to a regular floor for overnight observation. During this time, blood is drawn for follow-up laboratory studies, including hemoglobin and hematocrit, white blood cell count, platelet count, AST and ALT, coagulation profile, creatinine, and myoglobin. Follow-up imaging is arranged for 3 months after the ablation procedure. If the postablation course is uneventful, most patients are discharged from the hospital the day after the procedure. After discharge, contact by telephone is maintained for a few days. The patient is seen as an outpatient 1 week after the procedure. Three-month imaging follow-up—CT, MRI, PET-CT, or any combination—is performed and the results compared with preprocedure images to further document the results of the ablation and regression or progression of disease. The latter may require another ablative procedure. KEY POINTS •

Percutaneous cryoablation is an effective method for tumor destruction in the liver and has been used for both primary and metastatic lesions.

•

With proper planning and care, most tumors in the liver can safely undergo cryoablation as an alternative to liver resection.

•

Computed tomography, ultrasound, and magnetic resonance imaging all have been used to guide placement of cryoprobes to monitor intraprocedural iceball formation and for postprocedural follow-up.

▶ SUGGESTED READINGS

CRYOABLATION VERSUS RADIOFREQUENCY ABLATION

Atwell TD, Charboneau JW, Que FG, et al. Treatment of neuroendocrine cancer metastatic to the liver: the role of ablative techniques. Cardiovasc Intervent Radiol 2005;28:409–21. Jain S, Sacchi M, Vrachnos P, et al. Recent advances in the treatment of colorectal liver metastases. Hepatogastroenterology 2005;52: 1567–84. Jansen MC, van Hillegersberg R, Chamuleau RA, et al. Outcome of regional and local ablative therapies for hepatocellular carcinoma: a collective review. Eur J Surg Oncol 2005;31:331–47. vanSonnenberg E, McMullen W, Solbiati L. Tumor Ablation: Principles and Practice. New York: Springer; 2005. Weber SM, Lee FT Jr. Expanded treatment of hepatic tumors with radiofrequency ablation and cryoablation. Oncology 2005;19(11 Suppl 4):27–32.

For a discussion of this topic, please visit www. expertconsult.com.

The complete reference list is available online at www.expertconsult.com.

*References 36, 39, 40, 44, 47, 53.

CHAPTER 143 CRYOABLATION OF LIVER TUMORS1060.e1 Cryoablation and RFA each have advantages and disadvantages. One advantage of cryoablation is that multiple cryoprobes can be activated simultaneously to create a large iceball. Intraprocedural monitoring is better assessed with cryoablation than RFA. It has been suggested that there is less pain with cryoablation than with RFA.15,54 Disadvantages of liver cryoablation may include higher complication rates54 and what appears to be a greater propensity to thrombocytopenia, as described earlier.

Little data from large, randomized studies exist comparing cryoablation to RFA. With regard to efficacy of tumor kill between cryoablation and RFA, one nonrandomized study described complete devascularization in 69% of liver tumors (primary or secondary) by cryoablation and 76% by RFA, but this was not statistically significant.37 To date, there are no well-constructed trials that would favor either.55

1060.e2PART 2 NONVASCULAR INTERVENTIONS · SECTION 21 THE LIVER REFERENCES 1. Nordlinger B, Rougier P. Nonsurgical methods for liver metastases including cryotherapy, radiofrequency ablation, and infusional treatment: what’s new in 2001? Curr Opin Oncol 2002;14:420–3. 2. De Sanctis JT, Goldberg SN, Mueller PR. Percutaneous treatment of hepatic neoplasms: A review of current techniques. Cardiovasc Intervent Radiol 1998;21:273–96. 3. Seifert JK, Junginger T, Morris DL. A collective review of the world literature on hepatic cryotherapy. J R Coll Surg Edinb 1998;43: 141–54. 4. Weber SM, Lee FT Jr, Warner TF, et al. Hepatic cryoablation: US monitoring of extent of necrosis in normal pig liver. Radiology 1998;207:73–7. 5. Gage AA. History of cryosurgery. Semin Surg Oncol 1998;14: 99–109. 6. Gill IS, Novick AC, Soble JJ, et al. Laparoscopic renal cryoablation: initial clinical series. Urology 1998;52:543–51. 7. Cooper IS. Cryogenic surgery: a new method of deconstruction or extirpation of benign or malignant tissues. N Engl J Med 1963; 268:743–9. 8. Dodd GD, Soulen MC, Kane RA, et al. Minimally invasive treatment of malignant hepatic tumors: at the threshold of a major breakthrough. Radiographics 2000;20:9–27. 9. Tatli S, Acar M, Tuncali K, et al. Percutaneous cryoablation techniques and clinical applications. Diagn Interv Radiol 2010;16:90–5. 10. Silverman SG, Tuncali K, Adams DF, et al. MR imaging-guided percutaneous cryotherapy of liver tumors: initial experience. Radiology 2000;217:657–64. 11. Hedayati P, vanSonnenberg E, Shamos R, et al. Treatment of symptomatic focal nodular hyperplasia with percutaneous radiofrequency ablation. J Vasc Interv Radiol 2010;21:582–5. 12. Arii S, Teramoto K, Kawamura T, et al. Characteristics of recurrent hepatocellular carcinoma in Japan and our surgical experience. J Hepatobiliary Pancreat Surg 2001;8:397–403. 13. Farrell MA, Charboneau JW, Callstrom MR, et al. Paranephric water instillation: a technique to prevent bowel injury during percutaneous renal radiofrequency ablation. AJR Am J Roentgenol 2003;181: 1315–7. 14. Brown SD, vanSonnenberg E. Issues in imaging-guided tumor ablation in children versus adults. AJR Am J Roentgenol 2007;189:626–32. 15. Beland M, Mueller PR, Gervais DA. Thermal ablation in interventional oncology. Semin Roentgenol 2007;42:175–90. 16. Wong WS, Chinn DO, Chinn M, et al. Cryosurgery as a treatment for prostate carcinoma: results and complications. Cancer 1997;79: 963–74. 17. Saliken JC, McKinnon JG, Gray R. CT for monitoring cryotherapy. AJR Am J Roentgenol 1996;166:853–5. 18. Sandison GA, Loye MP, Rewcastle JC, et al. X-ray CT monitoring of iceball growth and thermal distribution during cryosurgery. Phys Med Biol 1998;43:3309–24. 19. Lee FT Jr, Chosy SG, Littrup PJ, et al. CT-monitored percutaneous cryoablation in a pig liver model: pilot study. Radiology 1999; 211:687–92. 20. Park BK, Morrison PR, Tatli S, et al. Estimated effective dose of CT-guided percutaneous cryoablation of liver tumors. Eur J Radiol 2011 (Epub). 21. Elhawary H, Oguro S, Tuncali K, et al. Multimodality non-rigid image registration for planning, targeting and monitoring during CT-guided percutaneous liver tumor cryoablation. Acad Radiol 2010;17:1334–44. 22. Giesel FL, Mehndiratta A, Locklin J, et al. Image fusion using CT, MRI and PET for treatment planning, navigation and follow up in percutaneous RFA. Exp Oncol 2009;31(2): 106–14.

23. Mortele KJ, Tuncali K, Cantisani V, et al. MRI-guided abdominal intervention. Abdom Imaging 2003;28:756–74. 24. Daniel BL, Butts K, Block WF. Magnetic resonance imaging of frozen tissues: temperature-dependent MR signal characteristics and relevance for MR monitoring of cryosurgery. Magn Reson Med 1999; 41:627–30. 25. Harada J, Dohi M, Mogami T, et al. Initial experience of percutaneous renal cryosurgery under the guidance of a horizontal open MRI system. Radiat Med 2001;19:291–6. 26. Morrison PR, Silverman SG, Tuncali K, Tatli S. MRI-guided cryotherapy. J Magn Reson Imaging 2008;27:410–20. 27. Matsumoto R, Oshio K, Jolesz FA. Monitoring of laser and freezinginduced ablation in the liver with T1-weighted MR imaging. J Magn Reson Imaging 1992;2:555–62. 28. Gilbert JC, Rubinsky B, Wong ST, et al. Temperature determination in the frozen region during cryosurgery of rabbit liver using MR image analysis. Magn Reson Imaging 1997;15:657–67. 29. Tacke J, Adam G, Speetzen R, et al. MR-guided interstitial cryotherapy of the liver with a novel, nitrogen-cooled cryoprobe. Magn Reson Med 1998;39:354–60. 30. Gilbert J, Kurhanewicz J. Principles of MR-guided cryoablation. In: Lufkin RB, editor. Interventional MRI. St Louis: CV Mosby; 1999, p. 215–9. 31. Sainani NI, Shyn PB, Tatli S, et al. PET/CT-guided radiofrequency and cryoablation: is tumor fluorine-18 fluorodeoxyglucose activity dissipated by thermal ablation? J Vasc Interv Radiol 2011;22:354–60. 32. Couinaud C. Le foie: etudes anatomiques et chirugicales. Paris: Masson; 1957. 33. Nair RT, Silverman SG, Tuncali K, et al. Biochemical and hematologic alterations following percutaneous cryoablation of liver tumors: experience in 48 procedures. Radiology 2008;248:303–11. 34. Xu KC, Niu LZ, He WB, et al. Percutaneous cryoablation in combination with ethanol injection for unresectable hepatocellular carcinoma. World J Gastroenterol 2003;9:2686–9. 35. Lee SM, Won JY, Lee DY, et al. Percutaneous cryoablation of small hepatocellular carcinomas using a 17-gauge ultrathin probe. Clin Radiol 2011;66:752–9. 36. Jansen MC, van Hillegersberg R, Chamuleau RA, et al. Outcome of regional and local ablative therapies for hepatocellular carcinoma: a collective review. Eur J Surg Oncol 2005;31:331–47. 37. Adam R, Hagopian EJ, Linhares M, et al. A comparison of percutaneous cryosurgery and percutaneous radiofrequency for unresectable hepatic malignancies. Arch Surg 2002;137:1332–9. 38. Steele G Jr. Cryoablation in hepatic surgery. Semin Liver Dis 1994;14:120–5. 39. Bilchik AJ, Sarantou T, Foshag LJ, et al. Cryosurgical palliation of metastatic neuroendocrine tumors resistant to conventional therapy. Surgery 1997;122:1040–7. 40. Cozzi PJ, Englund R, Morris DL. Cryotherapy treatment of patients with hepatic metastases from neuroendocrine tumors. Cancer 1995;76:501–9. 41. Sarantou T, Bilchik A, Ramming KP. Complications of hepatic cryosurgery. Semin Surg Oncol 1998;14:156–62. 42. Sohn RL, Carlin AM, Steffes C, et al. The extent of cryosurgery increases the complication rate after hepatic cryoablation. Am Surg 2003;69:317–22. 43. Seifert JK, Morris DL. World survey on the complications of hepatic and prostate cryotherapy. World J Surg 1999;23:109–13. 44. Crews KA, Kuhn JA, McCarty TM, et al. Cryosurgical ablation of hepatic tumors. Am J Surg 1997;174:614–7. 45. Cha C, Lee FT Jr, Rikkers LF, et al. Rationale for the combination of cryoablation with surgical resection of hepatic tumors. J Gastrointest Surg 2001;5:206–13.

CHAPTER 143 CRYOABLATION OF LIVER TUMORS1060.e3 46. Onik GM, Atkinson D, Zemel R, Weaver ML. Cryosurgery of liver cancer. Semin Surg Oncol 1993;9:309–17. 47. Bagia JS, Perera DS, Morris DL. Renal impairment in hepatic cryotherapy. Cryobiology 1998;36:263–7. 48. Jansen MC, van Wanrooy S, van Hillegersberg R, et al. Assessment of systemic inflammatory response (SIR) in patients undergoing radiofrequency ablation or partial liver resection for liver tumors. Eur J Surg Oncol 2008;34:662–7. 49. Pearson AS, Izzo F, Fleming RY, et al. Intraoperative radiofrequency ablation or cryoablation for hepatic malignancies. Am J Surg 1999; 178:592–9. 50. Seifert JK, Stewart GJ, Hewitt PM, et al. Interleukin-6 and tumor necrosis factor-alpha levels following hepatic cryotherapy: association with volume and duration of freezing. World J Surg 1999;23: 1019–26.

51. Silverstein JC, Staren E, Velasco J. Thermal bile duct protection during liver cryoablation. J Surg Oncol 1997;64:163–4. 52. Weaver ML, Atkinson D, Zemel R. Hepatic cryosurgery in treating colorectal metastases. Cancer 1995;76:210–4. 53. Yeh KA, Fortunato L, Hoffman JP, Eisenberg BL. Cryosurgical ablation of hepatic metastases from colorectal carcinomas. Am Surg 1997; 63:63–8. 54. McWilliams JP, Yamamoto S, Raman SS, et al. Percutaneous ablation of hepatocellular carcinoma: current status. J Vasc Interv Radiol 2010;21:S204–213. 55. Martin RC. Hepatic tumor ablation: cryo versus radiofrequency, which is better? Am Surg 2006;72:391–2.

CHAPTER

144

Chemical Ablation of Liver Lesions

Joshua L. Weintraub, Thomas J. Ward, and John H. Rundback

The presentation of a patient with a liver lesion ranges from the patient with an incidentally discovered and asymptomatic simple liver cyst to the patient with multifocal primary or metastatic liver tumors. The interventions used to treat this wide range of lesions is equally as varied, with percutaneous chemical ablation having a continually evolving role.

CHEMICAL ABLATION OF BENIGN LIVER LESIONS Simple Liver Cysts Liver cysts are common, with a prevalence of approximately 2.5% in the general population and a female predominance.1 A majority of these cysts are incidentally discovered by ultrasound or computed tomography (CT) and are clinically insignificant, so no further evaluation or treatment is necessary. For a minority of patients, larger cysts can produce symptoms that range from abdominal discomfort and nausea to compression of the biliary tree or inferior vena cava, intraperitoneal rupture, hemorrhage, torsion, or infection.2-6 The majority of symptomatic cysts occur in females older than 50, with a 9 : 1 female-to-male ratio and a predisposition for the right hepatic lobe.7 Previous attempts at percutaneous treatment of the symptomatic cyst focused on cyst aspiration. This strategy produced suboptimal results due to high rates of cyst recurrence, with reported recurrence rates between 78% and 100%.8,9 Advances in percutaneous treatment of liver cysts occurred with the introduction of ablative agents that treat the underlying pathology of the process. True simple hepatic cysts have an epithelial lining and are the result of congenitally aberrant bile ducts that are separated from the remainder of the intrahepatic biliary tract. The columnar or cuboidal epithelial cells that line the cyst produce fluid that reaccumulates after simple aspiration. With newer techniques, after cyst aspiration a variety of chemical agents (e.g., tetracycline hydrochloride, doxycycline, minocycline hydrochloride, hypertonic saline) have been injected into the cyst in an attempt to decrease previously observed recurrence rates by damaging the fluidproducing cells.10-13 The most studied chemical agent administered for this purpose is ethanol, which produces epithelial cell damage that prevents fluid secretion and results in subsequent cyst destruction. The topics Percutaneous Ethanol Sclerotherapy (PES) and Puncture, Aspiration, Injection, and Reaspiration of Cystic Echinococcosis (PAIR of CE) are discussed in full at www.expertconsult.com.

CHEMICAL ABLATION OF MALIGNANT LIVER LESIONS Unresectable hepatic malignancies, whether primary or metastatic, represent one of the more difficult challenges in the care of patients with cancer. Hepatocellular carcinoma (HCC) is the most common primary liver cancer worldwide, with over 500,000 new cases diagnosed annually. HCC is an extremely malignant tumor that generally occurs in the setting of chronic hepatic cirrhosis. Its association with hepatitis B and C infection is expected to increase its incidence. Treatment options for HCC are limited, with less than half of all patients being candidates for surgical resection. In these patients, median survival from time of diagnosis is approximately 6 to 20 months, despite aggressive therapy. Colorectal carcinoma accounts for 148,000 new cases of cancer a year and is the second most common cause of cancer death in the United States. Of the 55,000 deaths annually, over 50% are due to liver metastases after the primary lesion has been resected. Surgical resection of the hepatic metastases has been demonstrated to improve survival. In patients who are not resection candidates, regional therapy may show a benefit. Some other rare tumors (e.g., neuroendocrine, carcinoid, and islet cell tumors; ocular melanoma) have a unique propensity to metastasize exclusively to the liver. Systemic chemotherapies have been shown to be largely ineffective. Because of this, regional cancer therapy and hepatic artery embolization have been employed to control symptoms and attempt to improve overall survival. Percutaneous chemical ablation is an established technique of regional therapy for small ( 400 mL). The catheter is then removed and the skin bandaged. A study that compared single-session ethanol sclerotherapy to prolonged catheter drainage with negative pressure demonstrated similar results between the two methods, with statistically insignificant differences in regard to average cyst volume reduction, final cyst volume, or cyst disappearance rates.18 Should the patient experience significant pain, the ethanol is immediately aspirated off, generally with subsequent resolution of the patient’s acute symptomatology. Lidocaine may be injected into the cyst cavity via the drainage catheter prior to ethanol administration to minimize the occurrence of this event. Controversies The principle controversies surrounding hepatic cyst PES focus on the appropriateness of the treatment and whether laparoscopic treatment is more appropriate as first-line treatment. There is little debate that hepatic cysts that produce significant mass effect on adjacent structures should be treated. Treatment performed for abdominal discomfort or pain is more controversial. Care must be taken to exclude other etiologies for the patient’s symptoms prior to intervention. Performing interventionalists may prefer to do a trial cyst aspiration without sclerotherapy. If symptoms do not abate, the cyst in question is excluded as the cause of pain. Should pain resolve after aspiration and recur with cyst recurrence, PES may be advised. A variety of options are available for surgical management. Prior to widespread use of laparoscopic surgeries, open cystectomy, hepatic wedge resections, and lobectomies were performed, with significant rates of morbidity and mortality. In comparison, laparoscopic “de-roofing” has significantly lowered adverse events, cost, and length of stay, with low rates of recurrence.19-21 No prospective randomized control studies have been performed that compare percutaneous to laparoscopic treatment. As such, the expertise of the physicians at the performing institution generally guides treatment decisions. Outcomes PES is effective at reducing cyst volume and relieving the symptoms for which the procedure is performed. A review of the literature performed by Drenth et al. of 34 articles that involved 292 patients with solitary or multiple hepatic

1061.e2PART 2 NONVASCULAR INTERVENTIONS · SECTION 21 THE LIVER cysts demonstrated an at-least-partial recurrence rate of 21% during follow-up, with the majority being asymptomatic and reintervention rarely performed.22 The largest study examined cyst size reduction and symptom relief in 25 patients with 59 symptomatic hepatic cysts, with 11 patients having a single cyst. Overall, 8 cysts resolved completely, and mean diameter of remaining cysts decreased from 9 to 3 cm. Symptoms completely resolved or improved in all patients with solitary cysts; symptoms recurred in 7 patients with polycystic liver disease, owing to growth of other untreated cysts.23 Complications Reported complications after PES are generally minor and transient and include mild pain, nausea, vomiting, and systemic ethanol absorption. More severe pain can occur if injected ethanol leaks around the drainage catheter. One of the more serious reported complications is irreversible sclerosing cholangitis, which may be seen in the setting of ethanol injection into a cyst that communicates with the biliary tree. Postprocedural and Follow-up Care If tolerated without immediate complication, PES can be performed as an outpatient procedure. Short-term follow-up at 1 to 2 months with ultrasound is generally performed, with further follow-up performed only for symptom recurrence. A small sample of 39 benign intraabdominal cysts, 21 hepatic, that were treated with PES followed by frequent ultrasound during the first year demonstrated that of those cysts that recurred, all recurred at a smaller volume and subsequently decreased spontaneously in size without intervention. The authors concluded that initial recurrence after percutaneous ablation is usually a transient reactive inflammatory reaction and not true treatment failure.24 Should symptoms recur in the setting of cyst recurrence, repeat PES or laparoscopic de-roofing may be considered.

Puncture, Aspiration, Injection, and Reaspiration of Cystic Echinococcosis Cystic echinococcosis (CE) is a zoonotic infection of the Echinococcus tapeworm, with Echinococcus granulosus and Echinococcus multilocularis being the species that most commonly cause human disease. In endemic areas and Middle Eastern, South American, and Mediterranean countries, the annual incidence of CE ranges from 1 to 200 per 100,000, with a mortality rate of 2% to 4% and higher mortality rates seen in the setting of inadequate medical care.25 CE can affect multiple organ systems, although 80% of patients present with a solitary cyst, most commonly hepatic.26 Disease presentation and progression is markedly variable. Cysts may remain stable for years, regress or resolve spontaneously, or grow rapidly with symptom onset generally secondary to mass effect on adjacent structures, although intraperitoneal rupture and subsequent anaphylaxis can occur. Traditionally, surgical management has been the standard of care for treating active or symptomatic CE because of the fear that percutaneous treatment could result in spillage of cyst contents and peritoneal seeding or a fatal anaphylactic reaction. Use of the anthelmintic chemotherapy

drugs mebendazole and albendazole prior to percutaneous intervention decreases this risk and has resulted in increased use of percutaneous treatment strategies at many highvolume centers (Fig. e144-2).27 Indications The appropriateness for a puncture, aspiration, injection, and reaspiration (PAIR) treatment strategy depends on the complexity and activity of the cyst being treated. In 1996, the World Health Organization Informal Working Group on Echinococcosis (WHO-IWGE) produced a standardized classification system to replace the previous classification schemes.28 The most popular scheme at the time was proposed by Gharbi et al. in 1981 and is still regularly referenced.29 PAIR is first-line treatment for WHO-IWGE CE1 and CE3a cysts (Gharbi types 1 and 2).30 These cysts contain clear fluid; a detached membrane and daughter cysts may or may not be present. It is also indicated for patients who are not surgical candidates, patients who refuse surgery, and for cases that failed surgery or medical management. Contraindications Complex or inactive cysts (WHO-IWGE CE2, CE3b, CE4, and CE5 [Gharbi types 3, 4, and 5]) are not treated with PAIR. Complex CE is treated surgically, whereas inactive CE is managed with a watch-and-wait approach. Superficially located CE is generally not treated with PAIR because of the risk of cyst content leakage into the peritoneum. Cyst communication with the biliary tree precludes PAIR with 95% ethanol owing to the risk of sclerosing cholangitis. As previously mentioned, an INR less than 1.5 and an absolute platelet count over 50,000 are preferred because they are associated with decreased complication rates. Equipment The equipment necessary for PAIR includes an ultrasound machine with a 3- to 5-MHz curved-array transducer, an 18- to 22-gauge needle, a 0.035 Amplatz wire, appropriate fascial dilators and a 6F to 12F multipurpose drainage catheter. The most commonly administered protoscolicides are 20% sodium chloride and 95% ethanol. The procedure is performed with conscious and local sedation in a fluoroscopy suite or on a CT scanner with continuous hemodynamic monitoring, with appropriate resuscitation material available. Other required equipment includes a sterile tray, sterile sheets, and local anesthetic. Technique PAIR of CE is performed with real-time ultrasound or CT guidance after the diagnosis of CE is confirmed with an indirect hemagglutination test (IHA), or enzyme-linked immunosorbent assay (ELISA) confirms the presence of antibodies against Echinococcosis, and ultrasound or CT has identified a characteristic cyst. Albendazole or mebendazole is administered at least 4 hours before the procedure and continued for 1 month after.31 The procedure is performed with conscious sedation as well as continuous cardiac and hemodynamic monitoring, with resuscitation equipment close at hand. Prophylactic antibiotics are generally not administered. Local 1% lidocaine is administered at the determined puncture site prior to gaining access to the cyst via an 18- to 22-gauge needle.

CHAPTER 144 CHEMICAL ABLATION OF LIVER LESIONS1061.e3 When possible, an intercostal approach is taken, with the puncture trajectory selected to traverse at least 1 cm of normal liver to prevent spillage of cyst contents. Access to and characterization of the cyst is confirmed with fluid aspiration. Cyst fluid is then aspirated, with fluid sent for parasitologic evaluation. Evaluation for biliary communication can be performed with fluoroscopy or CT after injection of dilute watersoluble contrast or by testing the aspirated fluid for bilirubin. After biliary communication is excluded, a scolicidal agent—sterile 20% hypertonic saline or 95% ethanol—is injected. The volume administered is approximately 50% of the volume of cystic fluid aspirated. The agent is left in place for 20 minutes and the patient rolled in different positions to ensure uniform mixing of the cyst contents with the hypertonic saline or ethanol. The fluid is then completely removed. In the case of large cysts (>10 cm in diameter), a PAIR procedure may be performed with drainage (PAIR-D). Following a traditional PAIR procedure, a 0.035-inch Amplatz wire is placed through the needle into the cyst, with the needle then withdrawn. The Seldinger technique is used to advance an 8F multipurpose drainage catheter with side holes into the cyst, with care taken to only dilate the tract through the subcutaneous tissues and most superficial liver to reduce the risk of cyst content spillage. The drainage catheter is left to drainage and irrigated with isotonic fluid 4 or 5 times daily until drain output is less than 10 mL/day.32 Controversies Controversy initially surrounded the safety of percutaneous treatment of CE with PAIR, owing to the fear that this strategy would result in increased rates of peritoneal seeding or anaphylactic reactions. For decades, CE was treated with an open surgical approach, but after the introduction of chemotherapeutic drugs to treat CE, there has been increased interest in percutaneous management. Large series have since demonstrated PAIR to be a safe and efficacious treatment strategy that is supported by long-term data when appropriate patient selection criteria are used.30,33 The optimal protoscolicidal agent is also a point of controversy. No study has demonstrated superiority of ethanol over hypertonic saline, although some believe ethanol is more sclerosing and results in quicker cyst involution.33,34 In the setting of cystobiliary communication, however,

the risk of sclerosing cholangitis precludes administering ethanol. Outcomes Smego et al. performed a large meta-analysis of 769 patients with CE who were treated with PAIR and albendazole or mebendazole, compared to 952 era-matched historical control subjects treated surgically. PAIR produced clinical and parasitologic cure in 95.8% of patients, compared to 89.8% in the surgical group (P < .0001).27 Rates of cyst disappearance or solidification observed with follow-up ultrasound after PAIR generally range from 80% to 100%.30,32,35 Postprocedure seroconversion rates, as determined by serial IHAs or ELISAs, range from 50% to 100% in smaller studies.30,34,36,37 A statistically significant lower rate of disease recurrence was also observed with PAIR compared to surgery: 1.6% to 6.3%, resectively.27 Complications In the previously cited meta-analysis, major reactions (i.e., anaphylaxis, abscess, cyst infection, biliary fistula, sepsis) were observed in 7.9% of PAIR patients and 25.1% of surgical controls (P < .0001). An anaphylactic reaction that resulted in patient death occurred in 1 PAIR patient and 7 surgical patients. Minor reactions were also less frequently observed with PAIR: 13.1% compared to 33% in the surgical cohort (P < .0001). Although fever and allergic reaction were observed more frequently with PAIR (5.5%-2.5% and 4.8%-0.1%, respectively), management of these events was generally successful with antihistamines and antipyretics. Postprocedural and Follow-up Care Patients are monitored in the hospital for adverse reaction, with 2 or 3 days being a typical hospital stay, and then continued on albendazole or mebendazole for 28 days post procedure. Follow-up imaging and blood tests are performed to assess for treatment response. Treatment cure is generally defined as disappearance of symptoms and either (1) real-time observation of endocyst separation from the pericyst or subsequent cyst solidification, (2) decreased cyst size with increased density on CT scan, or (3) seroconversion on IHA or ELISA ( 7 mEq/L) should be corrected with dialysis before the procedure,9 and systolic blood pressure less than 180 mmHg is desirable. If the PCN is elective, antiplatelet agents or anticoagulants should be withheld for 5 days before the procedure when feasible.10 Patients who are allergic to contrast media should receive premedication with either oral or

CHAPTER 146 PERCUTANEOUS NEPHROSTOMY, CYSTOSTOMY, AND NEPHROURETERAL STENTING 1077 TABLE 146-1. Indications for Percutaneous Nephrostomy

TABLE 146-2. Causes of Urinary Obstruction Benign

Malignant

Intrinsic

Stone Clot Sequestered papilla Fungus ball

Urothelial tumor

Extrinsic

Postoperative scar tissue Surgical ligature Crossing vessel Retroperitoneal fibrosis

Pelvic tumor Adenopathy

Relief of Urinary Obstruction Urosepsis or suspected infection Acute renal failure Intractable pain Urinary Diversion Hemorrhagic cystitis Traumatic or iatrogenic ureteral injury Inflammatory or malignant urinary fistula Diagnostic Testing Antegrade pyelography Ureteral perfusion (i.e., Whitaker test) Biopsy of a urothelial lesion

TABLE 146-3. Whitaker Test

Access for Endourologic Procedures Stone removal Dilation or stenting of a ureteral stricture Endopyelotomy Foreign body retrieval Ureteral occlusion for urinary fistula Fungus ball removal or direct infusion of antifungal agents

A

Pressure Gradient

Result

22 cm H2O

Positive for upper urinary tract obstruction

B

FIGURE 146-1. Whitaker test in a patient after pyeloplasty of left ureteropelvic junction. A, Curved maximum intensity projection of abdomen demonstrates a horseshoe kidney with hydronephrosis and cortical thinning of left moiety. Furosemide (Lasix) renogram was consistent with obstruction on left, and a Whitaker test was requested. B, Despite marked dilation of left renal collecting system, there was no pressure gradient between left renal pelvis and bladder.

Access sheath Heavy-duty wire

A

B

C

FIGURE 146-2. Access for percutaneous nephrolithotomy (PCNL) in a patient with a large obstructing stone in a horseshoe kidney. A, Axial image from a non–contrast-enhanced computed tomography scan demonstrates stone. B, Sonographic and fluoroscopic guidance was used to access collecting system from a left posterior approach. Contrast material is seen surrounding stone (outline) lodged at ureteropelvic junction. C, A 6F coaxial access sheath is in renal pelvis, and a heavy-duty wire has been advanced through sheath into bladder. A 5F Kumpe catheter was advanced over the wire into bladder and left in place to provide access for PCNL, which followed.

1078 PART 2 NONVASCULAR INTERVENTIONS · SECTION 22 THE GENITOURINARY TRACT TABLE 146-4. Contraindications to Percutaneous Nephrostomy Absolute Severe uncorrectable coagulopathy Percutaneous approach to kidney that crosses colon, spleen, or liver Relative Severe hyperkalemia or other metabolic imbalance Uncontrolled hypertension Ongoing use of antiplatelet agents or anticoagulants Extremely short life expectancy because of underlying terminal illness

TABLE 146-5. Key Equipment for Percutaneous Nephrostomy Basic Equipment 21- or 22-gauge needle with stylet 0.018-inch mandril wire 6F coaxial introducer 0.038-inch heavy-duty J-tipped wire Fascial dilators Locking pigtail drain Additional Equipment 0.018-inch nitinol wire 0.035-inch stiff glidewire 0.035-inch stiff Amplatz wire 5F Kumpe catheter Peel-away sheath

intravenous (IV) steroids. Rarely, patients will have an anatomic variant such as a retrorenal colon that precludes safe percutaneous access to the kidney (Table 146-4).

Equipment Equipment for PCN is listed in Table 146-5. Image guidance for PCN may be accomplished with ultrasound, CT, fluoroscopy, or some combination of these modalities, most commonly ultrasound and fluoroscopy. For ultrasoundguided access to the renal collecting system, a low-frequency transducer, usually 3.5 MHz, is necessary for sufficient penetration of the retroperitoneal soft tissues to allow good visualization of the kidney. Any multislice CT scanner should provide adequate imaging to identify an appropriate access site and guide needle placement, and combined CT-fluoroscopy can provide real-time guidance for placement of the PCN. If CT-fluoroscopy is not available, it may be necessary to establish needle access to the kidney and then transfer the patient to a fluoroscopy unit for the remainder of the procedure. In the fluoroscopy suite, a rotating image intensifier is extremely helpful because oblique positioning of the tube allows real-time visualization of wire and catheter manipulations while keeping the operator’s hands outside the field of view. Ultrasound-guided access is often performed with a freehand technique, but needle guides are available and may be useful, especially for less experienced operators. A sterile cover is necessary for the transducer and cord. It may also

be helpful to place a clear plastic drape over the ultrasound control panel to allow the operator to make adjustments to the settings. If fluoroscopic guidance alone is used, IV administration of contrast material may be necessary to aid in visualization of the collecting system, especially if it is not dilated. Depending on the patient’s size, an initial injection of 75 to 100 mL of 68% ioversol (Optiray 320) is typically given, and an additional bolus can be administered if needed. Several different sets are available for percutaneous access to the kidney. The Neff Percutaneous Access Set (Cook Medical, Bloomington, Ind.) and the AccuStick II Introducer System (Boston Scientific, Natick, Mass.) include a 21- or 22-gauge needle with a stylet, a 6F coaxial introducer with a locking cannula, a 0.018-inch mandril wire, and a heavy-duty 0.038-inch J-tipped guidewire (150 cm). A few other tools may be helpful in certain situations. An 80-cm, 0.018-inch nitinol wire (ev3/Covidien, Plymouth, Minn.) is easier to advance through the access needle into the collecting system than the standard 0.018-inch mandril wire and may be extremely useful when the collecting system is not very dilated. If the patient is obese or has scar tissue in the retroperitoneum, an 18-gauge needle with a stylet is sturdier and less likely to be deflected during initial access, and a stiffer wire like a 0.035-inch Amplatz wire (Boston Scientific) may greatly facilitate dilation of the tract for placement of the drainage catheter. If placement of a nephroureteral stent or internal double-J ureteral stent is planned, a 5F Kumpe catheter (Cook Medical) and 0.035-inch stiff or regular glidewire (Boston Scientific) will facilitate access into the ureter and bladder. In the presence of a tight ureteral stricture, a 4F Glide catheter (Boston Scientific) or 6.3F or 8F van Andel catheters (Cook Medical), which taper to 3F or 5F, respectively, may be useful to cross the stricture. Occasionally, balloon dilation of a stricture is necessary before placement of a ureteral stent, in which case a noncompliant balloon should be used. Side-port sheaths may also aid in crossing a stricture, and peel-away sheaths are useful for placement of internal ureteral stents. Nephrostomy, nephroureteral, or double-J ureteral drains are available in many different varieties. Most of the nephrostomy tubes placed by interventional radiologists are 8F, 10F, or 12F all-purpose locking pigtail drains, although mini-pigtail drains may be placed when the renal pelvis is not capacious enough to allow a standard pigtail to form. Nephroureteral drains have a loop in the bladder and a locking pigtail in the renal pelvis and are available in different diameters, usually 8F and 10F, and lengths, typically 20 to 28 cm. Internal double-J ureteral drains are available in essentially the same sizes as nephroureteral drains.

Technique Anatomy and Approach Planning the approach for access is the most crucial step in performing successful uncomplicated PCN placement. Before the procedure, any prior studies or cross-sectional imaging should be reviewed to evaluate the anatomy, location, and orientation of the target kidney. Malrotation or malposition of the kidney, duplication of the collecting system, and any cysts, diverticula, tumor, or stones should be noted.11 The location of the lung, liver, spleen, and colon

CHAPTER 146 PERCUTANEOUS NEPHROSTOMY, CYSTOSTOMY, AND NEPHROURETERAL STENTING 1079 relative to the kidney should be considered, as well as body habitus, such as morbid obesity or spina bifida. For routine PCN, the patient should be placed in a prone or prone oblique position with the ipsilateral side elevated 20 to 30 degrees. If ultrasound guidance is planned for access, it is often helpful to scan the patient before sterile preparation to optimize positioning. In some patients, placement of a roll or pillow beneath the lower abdomen will reverse the normal lordotic curvature of the lumber spine and improve sonographic visualization of the approach. The ideal entry site to the kidney is through a relatively avascular plane known as the Brödel line that lies at an angle of about 30 to 45 degrees from midline when the patient is prone. The collecting system should be accessed peripherally via the tip of the calyx to decrease the likelihood of significant bleeding, and a posterior calyx should be chosen to facilitate placement of a nephrostomy tube or any additional endourologic procedures. The ideal skin entry site is approximately 10 cm lateral to the midline but not beyond the posterior axillary line. If the entry site is too medial, the PCN will cross the paraspinal musculature and make dilation of the tract more difficult and the tube more painful for the patient, and an entry site that is too lateral increases the risk of colonic perforation. The tract should avoid the inferior margin of the rib to decrease the risk of injury to an intercostal artery. If a supracostal approach is necessary for treatment of an upper pole stone, the risk of injury to lung, liver, or spleen is significantly less in the T11-T12 interspace than in the T10-T11 interspace.12 Technical Aspects Numerous techniques for image-guided PCN placement have been described. The renal collecting system can be accessed using a single-stick or double-stick method with ultrasound, fluoroscopy, or CT, alone or in combination. The single-stick method is more likely to be possible in patients with moderate to severe hydronephrosis, because it facilitates visualization of a calyx for direct access. A double-stick method is useful when the collecting system is not dilated, making it initially more difficult to access a calyx. In this technique, the first needle is used to access and opacify the collecting system so a more appropriate site for definitive access can be targeted (Fig. 146-3). There is reportedly no difference in technical success and complication rates for a single- versus double-stick technique.3,13 If ultrasound is used for guidance, the probe should be oriented along the long axis of the kidney, and the needle tip should be visualized from the skin entry site to the point of entry in the calyx. If fluoroscopic guidance alone is used, the site for access can be chosen on the basis of anatomic landmarks, by targeting a stone, or by opacification of the collecting system with IV contrast material. The C-arm should be rotated initially so that the needle is viewed end on as it is advanced toward the appropriate calyx, then the C-arm can be rotated to the orthogonal view to determine the depth. CT guidance for PCN may be required when the patient has variant anatomy. In addition, CT guidance can allow the procedure to be performed with the patient in a supine oblique position when the patient is unable to lie prone. In CT-guided cases, it is important to select the calyx that can be approached most easily in the horizontal plane.

Hemostat Access needle

FIGURE 146-3. Double-stick technique for percutaneous nephrostomy. After direct access of left renal pelvis with a 21-gauge needle, contrast material and air were injected to identify appropriate site for placement of nephrostomy drain. A posterior calyx is seen outlined by air and is marked with hemostat for access with a second 21-gauge needle.

Routine Percutaneous Nephrostomy Placement Once the access site has been chosen, the skin and subcutaneous tissues along the expected path for the nephrostomy are infiltrated with a local anesthetic, and a small dermatotomy is made with a scalpel. After needle access is achieved, a small amount of urine should be aspirated for culture, then a small amount of contrast material should be injected to determine the precise point of entry. If the needle entry site is not appropriate, the first needle can be used to further opacify the collecting system with contrast material and air if necessary so a more appropriate site can be targeted for entry with a second needle. When opacified, the posterior calyces are typically seen end on, whereas the anterior calyces project more laterally in the anteroposterior projection; however, injection of a small amount of air or CO2, which will fill the nondependent posterior calyces in a prone patient, may be necessary for confirmation. Care should be taken to aspirate a larger volume of urine than the amount of contrast material injected to avoid overdistention and possible sepsis. If the needle entry site is appropriate, the 0.018-inch wire is advanced through the needle and coiled in the renal collecting system. If the system is not very dilated, a 0.018-inch nitinol wire may be much easier to manipulate into the renal pelvis. More stable access is obtained when the wire is advanced into the ureter. Fluoroscopic guidance is very helpful to ensure that the wire is in good position and does not become kinked during the subsequent steps of the procedure. Once the 0.018-inch wire is in appropriate position, the access needle is exchanged over the wire for the 6F coaxial sheath, and the 0.018-inch wire and inner stiffeners are removed. A heavy-duty J-tipped 0.038-inch wire is advanced into the renal collecting system, and serial dilation of the tract is performed over the wire. The percutaneous drainage catheter, usually a locking pigtail drain, is advanced over the wire, and the wire and inner stiffener are removed. The pigtail is formed and locked. All the urine should be aspirated from the collecting system if there is any suspicion of

1080 PART 2 NONVASCULAR INTERVENTIONS · SECTION 22 THE GENITOURINARY TRACT infection, and a small amount of contrast material can then be injected to confirm appropriate position of the drain. The drain should be flushed and secured to the skin with either suture or an adhesive retention device. Percutaneous Nephrostomy Placement in a Nondilated Collecting System Occasionally, PCN will be necessary in a patient who has a nondilated system, which can be seen in the setting of a ureteral leak or fistula, hemorrhagic cystitis, or nondilated obstruction. The pathophysiology of obstruction without hydronephrosis is not well understood, but reversal of acute renal failure after PCN in a nondilated system has been well documented.14 A double-stick technique for fluoroscopically guided access to a nondilated system has been described.15 Intravenous contrast material is administered to opacify the collecting system, and a 22-gauge needle is quickly advanced directly into the renal pelvis. This needle can be used to distend the collecting system with a small amount of contrast material and air, then PCN can be performed as described previously. If the patient is in renal failure and IV contrast material cannot be given, ultrasound guidance can be used to access the renal pelvis as well. Nephroureteral Drain Placement Additional endourologic manipulations should not be performed at the time of the initial PCN if there is any possibility of infection or if there is significant bleeding after PCN. If neither of these conditions is present and the patient would benefit from a ureteral stent, placement of an internal-external nephroureteral stent or an internal double-J ureteral stent can be attempted. Placement of a double-J internal ureteral stent is the ultimate goal in patients who are candidates for this type of stent because it is much more comfortable and convenient for them. Cystoscopic retrograde exchange of the double-J stent must be possible, or an internal stent should not be placed. Other contraindications to placement of either type of ureteral stent include bladder outlet obstruction, irritable or neurogenic bladder, incontinence, and bladder tumors.3 A 5F Kumpe catheter and a 0.035-inch stiff glidewire is usually a good combination for traversal of the ureter into the bladder or an ileal conduit. If a stricture that cannot be crossed easily is present, several options are available. If the wire crosses the lesion but the catheter does not, a 4F Glide catheter or a van Andel catheter that tapers from 6.3F to 3F may work. If these catheters do not work alone, placement of a long side-arm 7F sheath that extends into the ureter will provide extra support while pushing the catheter over the wire.16 Successful use of a microwire and microcatheter to cross a tight ureteral stricture has also been described.17 If the lesion is difficult to cross with the catheter, balloon dilation of the stricture will probably be necessary before placement of the nephroureteral stent. Once the catheter is advanced into the bladder or conduit, the guidewire should be exchanged for a stiffer wire (e.g., 0.035-inch Amplatz wire) to facilitate advancement of the balloon across the lesion. Dilation with a 4-mm balloon should allow placement of an 8F nephroureteral stent. If these steps fail initially, the collecting system should be allowed to decompress via a PCN for 24 to 48 hours and then a repeat attempt made. The second attempt is often successful because the edema surrounding the stricture has resolved. If the

TABLE 146-6. Ureteral Stent Length Based on Patient Height Patient Height

Ureteral Stent Length

6′4˝

26 cm

ureteral stricture still proves impassible, sharp recanalization with a transjugular intrahepatic portosystemic shunt or transseptal needle has been described, or a rendezvous procedure with both antegrade and retrograde access to the stricture may be helpful.3,18,19 Selection of the length of the nephroureteral stent can be based on the height of the patient, which is a very reliable method, or on a measurement made with the wire, which tends to overestimate the length of the ureter (Table 146-6). When the stent is placed into the bladder, the distal pigtail should extend far enough beyond the ureterovesical junction (UVJ) that it will not retract into the ureter, but not so far that an excess amount of catheter is present in the bladder. The nephroureteral stent is advanced over the wire until the distal pigtail is beyond the ureteral orifice and the side holes for the proximal pigtail are just beyond the UPJ. The wire and inner stiffener are pulled back to allow first the distal loop and then the proximal pigtail to form. If the proximal pigtail is going to be in appropriate position in the renal pelvis, the wire and stiffener are removed and the catheter is locked. If the proximal pigtail is not in good position, the wire and stiffener can be used to reposition the stent before they are removed. Double-J Internal Ureteral Stent Placement Placement of a double-J internal ureteral stent is the same as placement of a nephroureteral stent up until the point at which a stiff guidewire is coiled in the bladder, but it may be a bit trickier thereafter. A 9F peel-away sheath is advanced over the wire until the tip of the sheath is at the UPJ or just in the proximal part of the ureter. The double-J stent system (Boston Scientific, Cook Medical) consists of an inner plastic cannula, a pusher, and the stent. A nylon suture is attached to the proximal end of the stent, and there is a radiopaque marker on the distal end of the pusher. The pusher typically comes loaded on the inner plastic cannula. The stent is loaded over the plastic cannula while making sure the suture does not become tangled and that the proximal end of the stent does not overlap the pusher; the whole assembly is advanced over the wire until the marker on the pusher is in the renal pelvis. The wire and inner plastic cannula are removed, leaving the pusher and the stent through the peel-away sheath. The distal loop forms automatically when the wire and stiffener are removed, and the proximal loop is formed by pulling on the nylon suture while applying pressure with the pusher. Once the proximal loop is formed, the suture is cut and the pusher is removed, leaving the peel-away sheath. A safety PCN can be advanced through the peel-away before its removal (Fig. 146-4). A safety PCN that is capped is typically left in the renal pelvis for 24 to 48 hours after placement of an internal ureteral stent to ensure the patient will tolerate internal

CHAPTER 146 PERCUTANEOUS NEPHROSTOMY, CYSTOSTOMY, AND NEPHROURETERAL STENTING 1081

A

B

C

FIGURE 146-4. Conversion of bilateral nephroureteral stents to double-J internal ureteral stents in a patient with extrinsic malignant ureteral obstruction. A, Right nephroureteral stent has been removed over a 0.035-inch stiff guidewire, and 8F 26-cm double-J stent has been advanced over wire through a 9F peel-away sheath. Radiopaque tip of pusher is visible in renal pelvis. B, Wire and plastic inner stiffener have been removed, and pusher has been used to help form proximal pigtail. C, Final fluoroscopic image demonstrates both stents in good position. A safety percutaneous nephrostomy was not placed on either side because exchanges were atraumatic and patient had tolerated capping of both nephroureteral stents for several weeks before procedure.

drainage. An antegrade nephrostogram can be performed via the PCN to confirm that the internal stent is functioning before removal. If a locking pigtail drain is used, fluoroscopic visualization during removal is recommended to avoid dislodging the ureteral stent with the pigtail, but this potential problem can be avoided by using a nonlocking pigtail drain as the safety PCN. The need for a safety PCN at all may be debatable because primary placement of an internal ureteral stent alone was successful in 83% of carefully selected patients in a recent study.20 Retrograde Nephroureteral Stent Placement Placement of a ureteral stent across a ureteroenteric anastomosis into an ileostomy may be accomplished via either an antegrade or retrograde approach. If an internal ureteral stent was placed at the time of surgical anastomosis, the stent may not have an end hole for over-the-wire exchange or repositioning, but it is usually possible to advance a 0.035-inch glidewire alongside the existing stent. After wire access into the renal pelvis is achieved, the existing stent can be removed, and a 7F 70-cm pigtail drain, called a urinary diversion soft stent (Boston Scientific), can be placed over the wire. If a ureteroenteric stent is not present, the anastomosis may be difficult to locate from a retrograde approach through the ostomy, and antegrade access via a PCN may become necessary. A routine PCN can be performed with the patient in the prone position, but after the guidewire is manipulated across the anastomosis and into the ostomy, it is helpful to have the patient in a lateral decubitus position so that both the flank and ostomy can be accessed. Once in the ostomy, the wire can usually be retrieved with hemostats, although snare retrieval may occasionally be required to establish through-and-through access (Fig. 146-5). After through-and-through access is achieved, the 7F 70-cm pigtail drain can be advanced retrogradely over the wire and formed in the renal pelvis. The drain can be shortened as needed but should be left long enough to extend into the ostomy bag to facilitate exchange of the drain.

Tip-deflector wire and pigtail Distal end of internal double-j

FIGURE 146-5. Retrograde exchange of bilateral double-J internal ureteral stents placed across ileoureteral anastomosis in a patient after ileal urinary diversion for severe radiation-induced cystitis. Distal end of double-J stent had retracted into ileal conduit and could not be grasped through stoma. A pigtail catheter and tip deflector wire were used to pull stent out through ostomy so it could be exchanged.

Removal of an Encrusted Percutaneous Nephrostomy Encrustation of the tube can make it extremely difficult to remove, and it may be impossible to exchange the tube over a guidewire. A hydrophilic wire can be advanced in the existing tract alongside the PCN to maintain access to the collecting system when the tube is removed. Removal of a pigtail that does not completely release when the catheter is cut may result in laceration of the renal parenchyma, and it may be necessary to advance a peel-away sheath over the tube to straighten the pigtail for safe removal. Removal of Internal Double-J Stents Antegrade access may be used for percutaneous retrieval of double-J ureteral stents that have fractured or become

1082 PART 2 NONVASCULAR INTERVENTIONS · SECTION 22 THE GENITOURINARY TRACT

A

B

C

FIGURE 146-6. Percutaneous nephrostomy in a patient with hydronephrosis and elevated creatinine after cadaveric renal transplantation. A, An antegrade nephrostogram via a 5F Kumpe catheter demonstrates a rounded filling defect at ureterovesical junction, consistent with a stone. B, Axial image from a non– contrast-enhanced CT scan confirms presence of stone (arrow). C, Patient’s nephrostomy drain was inadvertently dislodged, and a 5F Kumpe catheter was manipulated along existing tract to reestablish access into collecting system. A repeat nephrostogram demonstrated a residual stricture in distal part of ureter after passage of stone. Patient later underwent antegrade ureteroscopy with holmium laser ablation of a calcified stricture in distal end of ureter.

dislodged or can no longer be accessed from a retrograde approach. Use of snare catheters or forceps to grab the stent and use of balloon catheters inflated alongside to pull or push a stent into better position for snaring have been described, with technical success rates of 95% and 100% in two series.21,22 Replacement of a Dislodged Percutaneous Nephrostomy Catheter dislodgement within the first week of placement usually requires creation of a new tract, but mature tracts can frequently be reaccessed under fluoroscopic guidance. Gentle injection of a small amount of contrast material at the exit site will often help delineate the tract, which can then be traversed with a 5F Kumpe catheter and a 0.035inch stiff glidewire. Ureteral Occlusion Ureteral occlusion may be necessary in addition to urinary diversion in patients with a urine leak or fistula, urinary incontinence, or intractable hemorrhagic cystitis. Numerous methods for ureteral occlusion have been described, including antegrade placement of coils, Gelfoam, tissue adhesive, and detachable balloons, as well as retrograde injection of glutaraldehyde cross-linked collagen in the submucosa at the ureteral orifice.4,5 One technique with durable success was described by Farrell et al. and involves the use of Gianturco steel coils (Cook Medical) with or without Gelfoam. Gianturco coils ranging in size from 5 mm × 5 cm to 12 mm × 5 cm were packed in the distal part of the ureter. Placement of larger coils at both the proximal and distal ends of the nest was suggested to help avoid migration of smaller coils into the renal pelvis, and occlusion of the distal ureter was recommended to reduce the risk for a uretero–iliac artery fistula or infection related to a sort of “blind-loop” syndrome.5 Percutaneous Nephrostomy Placement in a Transplant Kidney In patients with posttransplant complications, an antegrade nephrostogram is performed first to establish the diagnosis and is followed by placement of a PCN or nephroureteral drain, with dilation of strictures if indicated. A severely

narrowed, tapered, aperistaltic ureter should be considered to be partially obstructed in these patients.23 An anterolateral calyx should be targeted, and a transperitoneal puncture, which increases the risk of complications, should be avoided by access of a mid- to upper-pole calyx from a lateral approach (Fig. 146-6). The remainder of the procedure is the same as in the native kidney, although short nephroureteral stents with 8 or 10 cm between the pigtails (Cook Medical) are available for use in renal transplant patients (Fig. 146-7). A biliary drain (Cook Medical) may also be the right length for use as an internal-external nephroureteral drain in a transplant kidney. Percutaneous Nephrostomy Placement in Pregnancy Relief of urinary obstruction may be accomplished by either a retrograde or antegrade approach in pregnant patients. The advantages of PCN are that it provides more rapid decompression of the obstructed kidney, avoids the potential for bladder irritability, facilitates routine drain exchanges, and provides access for definitive stone management after delivery. The challenges of PCN include difficulty positioning a patient who is often unable to lie prone, administering sedation that is safe for the fetus as well as the mother, and keeping the radiation dose to the absolute minimum required for accurate PCN placement. Although it may be possible to perform the entire procedure with ultrasound guidance, the potential need for fluoroscopy and the risk of radiation exposure to the fetus should be discussed with the patient when informed consent is obtained. Percutaneous Nephrostomy Placement in Pediatric Patients PCN in infants and newborns is often performed for temporary urinary drainage until definitive management is possible for conditions such as bilateral UPJ obstruction, UPJ obstruction with a single functional kidney, obstruction after pyeloplasty, UVJ obstruction, posterior urethral valves, and primary obstructing megaureter. In addition to the issues of appropriate sedation and monitoring for these patients, the procedure may be technically challenging even when the collecting system is extremely dilated. In infants, the kidney is so small that even with severe hydronephrosis,

CHAPTER 146 PERCUTANEOUS NEPHROSTOMY, CYSTOSTOMY, AND NEPHROURETERAL STENTING 1083

A

B

FIGURE 146-7. Percutaneous nephrostomy in a patient with hydronephrosis and elevated creatinine after revision of a right renal transplant ureteral anastomosis with a ureteroureterostomy. A, Sagittal ultrasound image obtained during access to right pelvic transplant kidney with a 21-gauge micropuncture needle, which is easily visible entering an anterior calyx. B, Antegrade nephrostogram after placement of an 8F locking pigtail drain demonstrates distal ureteral obstruction that is probably ischemic in nature.

TABLE 146-7. Technical Success Rates for Percutaneous Nephrostomy Obstructed dilated system without stones

98%

Obstructed system in renal transplant

98%

Nondilated system (with or without stones)

85%

Complex stone disease, staghorn calculi

85%

Modified from Ramchandani P, Cardella JF, Grassi CJ, et al. Quality improvement guidelines for percutaneous nephrostomy. J Vasc Interv Radiol 2003;14:S277–81.

the total volume of urine is actually minimal, and it may be difficult to coil a guidewire within the collecting system, a problem that is magnified when UPJ obstruction prevents passage of the wire down the ureter. In addition, the renal parenchyma is often thin and offers little resistance to leakage of urine, and there is less support from the retroperitoneal soft tissues, so the kidney is quite mobile and readily displaced during attempts at dilation of the tract. Difficulty with PCN placement via a standard micropuncture technique in patients younger than 14 weeks with severe dilation led Koral et al. to describe a modified technique. Access to the collecting system with a 0.018-inch system was replaced by access with a 19-gauge needle followed by a 0.035-inch wire and a 6F Navarre drain (C.R. Bard Inc., Covington, Ga.) to minimize the need for exchanges that allow urine to leak out of the collecting system. The Navarre drain was favored because its taper allows primary placement without dilation, and it coils easily in a small renal pelvis. The modified technique was thought to provide a greater level of success in infants and newborns, particularly those with UPJ obstruction.24

Outcomes Outcomes for PCN are shown in Table 146-7. A PCN can be successfully placed in 98% to 99% of obstructed dilated kidneys. Success rates drop to 85% in patients with nondilated collecting systems, complex stone disease, or staghorn calculi.9 Success rates are also lower (≈91%-92%) when ultrasound guidance alone is used for the procedure.25 One

analysis of emergency PCN outcomes reported that fluoroscopy and overall procedure times were significantly less for experienced operators than for inexperienced operators, and that 20% to 33% of procedures performed by inexperienced operators were repeated the next day because of catheter dislodgement or malposition.26 A simple technique for producing a gelatin-based phantom for use as a training simulator for ultrasound-guided PCN has been described and may improve outcomes for inexperienced operators.27 Antegrade ureteral stenting is technically successful in 88% to 96% of cases, and 80% of obstructed ureters can be stented primarily.20,28 If stenting is not possible initially, a repeat attempt after decompression is often successful. The clinical success of ureteral stenting depends on the cause of the ureteral stricture, with higher long-term success rates being reported in patients with intrinsic disease rather than extrinsic obstruction.29 Several studies have evaluated the results of balloon dilation of both benign and malignant strictures and consistently report the most durable success when treating benign strictures that are short and not associated with ischemia.30,31 In one study comparing the results of a single dilation followed by 3 weeks of internal ureteral stenting, benign strictures showed a patency rate of 67% and 57% at 12- and 36-month follow-up, versus 18% and 14% for malignant strictures.30 Another study found a 90% patency rate at 2 years after balloon dilation of benign strictures with no evidence of vascular compromise, but the study also came to the conclusion that endopyelotomy or endoureterotomy is indicated for treatment of all other benign strictures.31 Finally, a recent report on the use of a cutting balloon for treatment of benign ureteroenteric anastomotic strictures described a 3-year patency rate of 62% regardless of whether the stricture was ischemic in nature.32 In patients with malignant obstruction, plastic internal stents have a major complication rate of 34% and a failure rate of 36% over a mean indwelling time of 95 days.33 Unfortunately, metallic stents are little if any better. One review of standard vascular metal stents used to treat 90 patients with malignant ureteral obstruction reported primary patency in 51% but failed to discuss the outcomes in detail.34 Use of a 6F metallic double-J stent (Resonance [Cook Medical]) has been reported in a series of 40 patients treated

1084 PART 2 NONVASCULAR INTERVENTIONS · SECTION 22 THE GENITOURINARY TRACT with 76 stents, with somewhat positive results. At follow-up, 6 of the 40 patients had been able to avoid nephrostomy because of the metallic stent.33

Septic shock with major increase in level of care

1%-3%

dislodgement and blockage. Catheter dislodgement occurs in 1% to 2% of patients within the first 30 days of placement but may occur in up to 11% to 30% of patients who are monitored long term.37 Reported success rates for reinsertion were significantly higher for catheters with a longer indwelling time (93% when in place > 3 months) and for those with a shorter interval between dislodgement and reinsertion (91% within 24 hours).36 PCN blockage is reported to occur in approximately 1% of patients and is more likely in those with very crystalline urine and in patients who are noncompliant.37 The most common problem associated with nephroureteral and double-J ureteral stents is obstruction related to encrustation of the stent (Fig. 146-9).20 Other common complications include bladder irritative symptoms and flank pain that may occasionally be severe enough to warrant removal of the stent, mild hematuria, ascending urinary tract infection (UTI), and malpositioning.3 Finally, a rare complication of ureteral stents is erosion of the stent, either through the renal pelvis, with resultant urinoma, bleeding from vessels in the renal hilum, or retroperitoneal abscess; or into the iliac artery, with resultant intermittent or massive hematuria. Extensive pelvic surgery and pelvic irradiation are risk factors for the formation of a fistula between the ureter and iliac artery, which can be difficult to diagnose.20

Septic shock in setting of pyonephrosis

7%-9%

Patient Care

Hemorrhage requiring transfusion

1%-4%

Complications The overall complication rate for PCN is approximately 10% (Table 146-8).9 The most commonly occurring complications are sepsis and bleeding, both of which have a wide range of severity. The reported rate of sepsis requiring a major increase in the level of care is 1% to 3%, whereas the reported rate of hemorrhage requiring transfusion is 1% to 4%. Other complications that are reported to occur in less than 1% of patients include vascular injury requiring embolization or nephrectomy, bowel perforation, pleural complications, and injury to the renal pelvis (Fig. 146-8).9,35 As would be expected, there is evidence that complication rates are higher for inexperienced operators than for experienced ones.26 Other problems with PCN tubes that are frequent but not universally considered complications include catheter TABLE 146-8. Complication Rates for Percutaneous Nephrostomy

Vascular injury requiring embolization or nephrectomy

0.1%-1%

Bowel transgression

0.2%

Pleural complications

0.1%-0.2%

Catheter dislodgement within 30 days

1%-2%

Modified from Ramchandani P, Cardella JF, Grassi CJ, et al. Quality improvement guidelines for percutaneous nephrostomy. J Vasc Interv Radiol 2003;14:S277–81; and Millward SF. Percutaneous nephrostomy: a practical approach. J Vasc Interv Radiol 2000;11:955–64.

H

Patient preparation for PCN placement includes correction of any significant coagulopathy, premedication for contrast allergy, and treatment of hyperkalemia. Patients should be given prophylactic antibiotics, which have been shown to be beneficial in decreasing the incidence of sepsis both in high-risk patients, including those with stones, bacteriuria, ureteroenteric conduits, indwelling catheters, and diabetes, and in those who are at low risk.38 If the patient is considered to be at low risk for a septic complication, IV administration of a single dose of a first-generation cephalosporin

Staghorn stone

Renal calices Pigtail drain L

PCN

R

Retroperitoneal abscess drain Bladder

A

F

B

C

FIGURE 146-8. Young patient with spina bifida, renal and bladder stones, and fever. A, Posterior view from coronal maximum intensity projection reconstruction demonstrates spina bifida and a right staghorn stone. B, After percutaneous access of right renal collecting system, pigtail drain was very difficult to position because of size of staghorn stone. An attempt to place a nephroureteral drain instead was complicated by perforation of renal pelvis by wire. Volume-rendered image obtained 1 day after percutaneous nephrostomy (PCN) demonstrates pigtail outside renal pelvis. C, Retroperitoneal abscess developed and required placement of a drainage catheter. Nephrostomy drain had been repositioned into renal pelvis, but pigtail remained poorly formed. PCN was ultimately converted to a more stable nephroureteral stent, and retroperitoneal abscess resolved with percutaneous drainage.

CHAPTER 146 PERCUTANEOUS NEPHROSTOMY, CYSTOSTOMY, AND NEPHROURETERAL STENTING 1085

A

B

FIGURE 146-9. Noncompliant patient with a right renal stone, fever, and flank pain during pregnancy. A percutaneous nephrostomy drain was placed initially but was later converted to a nephroureteral stent because nephrostomy drain kept coming out. Patient was lost to follow-up and returned several months after delivery with fever and flank pain. A, Distal pigtail of nephroureteral stent is completely encrusted, and a wire could not be advanced through stent. A 0.035-inch stiff guidewire was advanced along stent into bladder. B, A locking pigtail drain was placed over guidewire and positioned in renal pelvis for decompression until patient could undergo cystoscopic lithotripsy.

within 2 hours of the start of the procedure should be sufficient. If the patient is considered to be at high risk, broad antibiotic coverage against the organisms that typically infect the genitourinary tract, including gram-negative rods such as Escherichia coli, Proteus, Klebsiella, and Enterococcus, should be chosen.38 UTIs in the urology patient population demonstrate higher resistance rates than community UTIs, and significant increases in resistance to ampicillin and trimethoprim have been shown39; however, antibiotic susceptibilities vary by region and patient population, and the patient’s medical record should be reviewed for any prior urine culture sensitivities to aid in selection of the most appropriate antibiotic. Antibiotics should be continued for 48 to 72 hours or longer if necessary in high-risk groups. Conscious sedation is often administered during PCN, typically as a combination of short-acting agents such as fentanyl and midazolam. Regardless of whether conscious sedation is used, continuous physiologic monitoring of the heart rate, blood pressure, and oxygen saturation should be performed throughout the procedure by dedicated nursing personnel. The patient should have stable IV access, both for administering periprocedural medications and for resuscitation in the event of sepsis, and should receive oxygen via nasal cannula. Immediately after the procedure, the patient should be transferred to a recovery area where frequent monitoring of vital signs can be continued at intervals of 15 or 30 minutes for at least 4 hours. Output from the nephrostomy should be recorded to ensure that it is draining well and that hematuria, if present, is clearing. Antibiotic therapy should be continued as appropriate. Because patients with urosepsis before the procedure are at increased risk for septic complications, it may be prudent to monitor these patients in an intensive care setting where intubation and aggressive fluid resuscitation can be performed rapidly if needed.37 PCN may be performed safely as an outpatient procedure in a carefully selected group of patients.40 Outpatient PCN should not be considered for patients who are elderly,

debilitated, or live alone or for those who are at increased risk for complications because of conditions such as staghorn calculi, severe hypertension, diabetes, or ongoing use of antiplatelet or anticoagulant agents. After initial placement, PCN drains should be left to external gravity drainage, which should be continued until the underlying problem has been treated or has resolved. Some patients, especially those with malignancies, may require long-term drainage, and patient education is the key to successful management. Patients should be given written instructions that describe how to care for the drain and what to do if the tube becomes blocked or displaced. Patients should understand when and how to contact the interventional radiology department in the event of a problem. Drains should be flushed twice daily, or patients instructed to remain well hydrated, or both, to prevent encrustation. Finally, routine PCN exchanges should be scheduled at 2- to 3-month intervals. If a nephroureteral drain is placed, external drainage should continue until any hematuria has resolved, then the drain can be capped to promote internal drainage. If the patient tolerates capping of a nephroureteral drain, the drain can be exchanged for either a PCN or an internal double-J with a safety PCN, depending on the clinical situation. If the patient continues to drain well internally, the PCN can be removed. The capping trial should last at least 24 to 48 hours to ensure that the internal drainage is adequate. The drain should be returned to external gravity drainage if the patient experiences fever, chills, flank pain, or leakage around the drain at any time while it is capped.

SUPRAPUBIC CYSTOSTOMY Indications Percutaneous suprapubic cystostomy is indicated for the treatment of acute urinary retention when transurethral catheterization of the bladder is not possible or is contraindicated. The most common underlying problems in

1086 PART 2 NONVASCULAR INTERVENTIONS · SECTION 22 THE GENITOURINARY TRACT patients requiring suprapubic cystostomy are bladder outlet obstruction secondary to benign prostatic hypertrophy or prostatic cancer and neurogenic bladder dysfunction, but the procedure may also be beneficial in patients with urethral trauma, radiation-induced cystitis, vesicocolonic or vesicovaginal fistulas, and urinary incontinence.41

Contraindications Relative contraindications to percutaneous cystostomy include coagulopathy, multiple previous abdominal and pelvic surgeries with extensive scar tissue, bladder tumor, morbid obesity, and inability to adequately distend the bladder to displace overlying small-bowel loops.

Equipment The sonographic and fluoroscopic equipment used for placement of a suprapubic cystostomy is the same as that for PCN. Access into the bladder can be achieved with a 19-gauge needle, which is exchanged over a 0.035-inch stiff Amplatz wire (Boston Scientific) for sequential fascial dilators. If a small-bore drainage catheter is desired, an 8F or 10F locking pigtail drain should be placed. If a large-bore catheter is required for long-term drainage, a 16F, 18F, or 20F Foley catheter should be placed through a peel-away sheath that is 2 French sizes larger than the Foley catheter (Table 146-9). A Stamey percutaneous suprapubic catheter

set (Cook Medical) is also available in 8F to 16F for placement of a suprapubic cystostomy via a trocar technique (Fig. 146-10).

Technique The preprocedural workup should include correction of any coagulation abnormalities, administration of prophylactic antibiotics, and placement of a transurethral Foley catheter if possible. The procedure is usually performed with the patient under conscious sedation, so appropriately trained personnel must be present to monitor the patient during and after the procedure. Percutaneous suprapubic cystostomy may be performed on an outpatient basis in selected patients. The bladder should be accessed approximately 2 to 3 cm above the pubic symphysis or at the junction of the

TABLE 146-9. Key Equipment for Percutaneous Suprapubic Cystostomy 18- or 19-gauge needle 0.035-inch stiff Amplatz wire Fascial dilators or balloon catheter Drainage catheter: 10F or 12F locking pigtail drain 16F to 20F Foley catheter (requires a peel-away sheath)

Guide wire

A

C

B

D

Urethal injury

E

FIGURE 146-10. Elderly patient with acute urinary retention after iatrogenic urethral injury. A, Sagittal ultrasound demonstrates a 0.035-inch stiff guidewire that was advanced through a 19-gauge access needle. B, After fascial dilation, a 10F locking pigtail drain was advanced over the wire. C, Appearance of pigtail drain after aspiration of 1000 mL of urine. D, Three days after placement, suprapubic drain became dislodged and tract could not be identified again. Bladder was not distended, and a decision was made to attempt retrograde placement of a transurethral Foley catheter. A 5F Kumpe catheter and a 0.035-inch stiff guidewire were manipulated across urethral injury into bladder, and a 16F Foley catheter was placed over the wire. E, Injection of contrast material demonstrated appropriate position of Foley catheter.

CHAPTER 146 PERCUTANEOUS NEPHROSTOMY, CYSTOSTOMY, AND NEPHROURETERAL STENTING 1087 mid- and lower thirds of the anterior bladder wall. The area of the trigone should be spared to avoid bladder spasm.42 A vertical midline or slight paramedian approach will avoid injury to the inferior epigastric artery, which lies along the lateral edge of the rectus sheath. If the bladder is fully distended or can be adequately distended with a transurethral Foley catheter, percutaneous suprapubic cystostomy may be performed under ultrasound guidance alone via a trocar technique.43 Otherwise, a combination of sonographic and fluoroscopic guidance can be used for placement of a suprapubic cystostomy via a modified Seldinger technique.41 If the bladder is not well distended initially, a 19-gauge needle should be advanced into the bladder under direct sonographic guidance to ensure that no bowel loops are traversed, and this needle can then be used to distend the bladder to capacity, usually with 150 to 500 mL of dilute contrast medium, before catheter placement. Once the bladder is fully distended, a 0.035-inch stiff Amplatz wire is advanced through the needle and coiled within the urinary bladder under fluoroscopic guidance. The tract can be dilated with sequential fascial dilators or a balloon catheter. A peel-away sheath that is 2 French sizes larger than the Foley catheter to be placed is advanced over the wire. An end hole is created in the tip of the Foley, and it is advanced over the wire through the peel-away sheath. The wire and peel-away sheath are removed, the Foley balloon is inflated with sterile saline and pulled up against the anterior wall of the bladder, and the Foley catheter is connected to a gravity drainage bag. Retention sutures are avoided because they can cause local skin irritation and typically are not necessary. Whereas placement of a small-bore catheter (e.g., 10F or 12F locking pigtail drain) is usually adequate for urgent decompression of the bladder, placement of a Foley catheter that is at least 16F in diameter is required for effective longterm bladder drainage. Placement of a large catheter may be difficult in patients with extensive scar tissue related to previous surgeries or in patients who are obese, in which case a two-step procedure may be necessary, with initial placement of a small pigtail drain that can be upsized after the tract has formed.41 Creation of a suprapubic cystostomy may enable placement of a transurethral Foley catheter in patients who have suffered a traumatic or iatrogenic injury to the urethra. After bladder access is achieved, a 5F Kumpe catheter and a 0.035-inch stiff guidewire can be used to probe for the urethra. It may be possible to advance the catheter and wire in antegrade fashion across the urethra to establish throughand-through access over which a transurethral Foley catheter can be placed from a retrograde approach.

Outcomes The reported success rates for placement of a percutaneous suprapubic cystostomy are essentially 100%.41,43 Although initial placement of a large-bore drainage catheter may not be possible in patients who are obese or have extensive scarring that prevents adequate dilation of the tract, it is almost always successful with a two-step procedure. In two series with follow-up of up to 18 and 36 months, all catheters were well tolerated and remained functional until they were no longer needed.41,42

Complications Potential complications associated with the procedure include hematuria (which is usually transient), sepsis, and injury to the bowel. Chronically indwelling catheters are associated with an increased risk for infection and stone formation, and as with other types of drainage catheters, suprapubic cystostomy tubes can become blocked or dislodged.

Postprocedural and Follow-up Care All patients should be monitored by checking vital signs at 15- to 30-minute intervals for 4 hours after the procedure. Patients or caregivers should be educated about routine maintenance of the drainage catheter and should be given detailed information about whom to contact in the event the drain stops working, starts leaking, or becomes dislodged. Because long-term bladder catheters are associated with an increased risk for stone formation and infection, patients should be maintained on low-dose trimethoprimsulfamethoxazole, and routine catheter exchanges should be scheduled every 2 to 3 months. Once the tract has matured, the catheter can be safely exchanged at the bedside.

KEY POINTS •

Review of the patient’s medical history and any prior imaging studies should always be performed prior to percutaneous nephrostomy (PCN) or suprapubic cystostomy.

•

Emergency PCN can be life saving in patients with urinary obstruction and urosepsis or acute renal failure.

•

Treatment of renal stone disease is the most common indication for PCN, but PCN can also be extremely valuable in the management of ureteral injuries or strictures, urine leaks, and hemorrhagic cystitis.

•

When PCN is performed for access prior to a urologic procedure such as percutaneous nephrostolithotomy, discussion with the referring urologist to plan the approach is essential to a successful outcome.

•

Technical success rates for PCN range from 85% in patients with nondilated collecting systems or complex stone disease to 99% in patients with obstructed dilated kidneys.

•

Percutaneous suprapubic cystostomy can provide rapid relief of acute urinary retention when transurethral catheterization of the bladder is not possible.

▶ SUGGESTED READINGS Adamo R, Saad WEA, Brown DB. Percutaneous ureteral interventions. Tech Vasc Interv Radiol 2009;3(12):205–15. American College of Radiology (ACR) practice guideline for the performance of percutaneous nephrostomy. ACR Practice Guideline; Amended 2004 (Res.25), p. 463–71. Barnacle AM, Wilkinson AG, Roebuck DJ. Paediatric interventional uroradiology. Cardiovasc Intervent Radiol 2011;2(34):227–40. Funaki B, Tepper JA. Percutaneous nephrostomy. Semin Intervent Radiol 2006;23(2):205–8.

1088 PART 2 NONVASCULAR INTERVENTIONS · SECTION 22 THE GENITOURINARY TRACT Hausegger KA. Percutaneous nephrostomy and antegrade ureteral stenting: technique-indications-complications. Eur Radiol 2006;16: 2016–30. Millward SF. Percutaneous nephrostomy: a practical approach. J Vasc Interv Radiol 2000;11:955–64. Ramchandani P, Cardella JF, Grassi CJ, et al. Quality improvement guidelines for percutaneous nephrostomy. J Vasc Interv Radiol 2003;14: S277–81.

Saad WEA, Moorthy M, Ginat D. Percutaneous nephrostomy: native and transplanted kidneys. Tech Vasc Interv Radiol 2009;3(12):172–92. Uppot RN. Emergent nephrostomy tube placement for acute urinary obstruction. Tech Vasc Interv Radiol 2009;2(12):154–61.

The complete reference list is available online at www.expertconsult.com.

CHAPTER 146 PERCUTANEOUS NEPHROSTOMY, CYSTOSTOMY, AND NEPHROURETERAL STENTING1088.e1 REFERENCES 1. Goodwin WE, Casey WC, Woolf W. Percutaneous trocar (needle) nephrostomy in hydronephrosis. JAMA 1955;157:891–4. 2. Sandhu C, Anson KM, Patel U. Urinary tract stones. Part II: current status of treatment. Clin Radiol 2003;58:422–33. 3. Hausegger KA. Percutaneous nephrostomy and antegrade ureteral stenting: technique-indications-complications. Eur Radiol 2006;16: 2016–30. 4. Gonzalez CM, Case JR, Nadler RB. Glutaraldehyde cross-linked collagen occlusion of the ureteral orifices with percutaneous nephrostomy: a minimally invasive option for treatment of refractory hemorrhagic cystitis. J Urol 2001;166:977–8. 5. Farrell TA, Wallace M, Hicks ME. Long-term results of transrenal ureteral occlusion with use of Gianturco coils and gelatin sponge pledgets. J Vasc Interv Radiol 1997;8:449–52. 6. Shokeir AA, El-Diasty T, Essa W, et al. Diagnosis of ureteral obstruction in patients with compromised renal function: the role of noninvasive imaging modalities. J Urol 2004;171:2303–6. 7. Jaffe RB, Middleton Jr AW. Whitaker test: differentiation of obstructive from nonobstructive uropathy. AJR Am J Roentgenol 1980;134:9–15. 8. Farrell TA, Hicks ME. A review of radiologically guided percutaneous nephrostomies in 303 patients. J Vasc Interv Radiol 1997;8:769–74. 9. Ramchandani P, Cardella JF, Grassi CJ, et al. Quality improvement guidelines for percutaneous nephrostomy. J Vasc Interv Radiol 2003;14:S277–81. 10. Payne CS. A primer on patient management problems in interventional radiology. AJR Am J Roentgenol 1998;170:1169–76. 11. Saad WEA, Moorthy M, Ginat D. Percutaneous nephrostomy: native and transplanted kidneys. Tech Vasc Interv Radiol 2009;3(12): 172–92. 12. Hopper KD, Yakes WF. The posterior intercostals approach for percutaneous renal procedures: risk of puncturing the lung, spleen and liver as determined by CT. AJR Am J Roentgenol 1990;154:115–7. 13. Funaki B, Vatakencherry G. Comparison of single-stick and doublestick techniques for percutaneous nephrostomy. Cardiovasc Intervent Radiol 2004;27(1):35–7. 14. Naidich JB, Rackson ME, Mossey RT, Stein HL. Nondilated obstructive uropathy: percutaneous nephrostomy performed to reverse renal failure. Radiology 1986;160:653–7. 15. Patel U, Hussain FF. Percutaneous nephrostomy of nondilated renal collecting systems with fluoroscopic guidance: technique and results. Radiology 2004;233:226–33. 16. Koukouras D, Petsas T, Liatsikos E, et al. Percutaneous minimally invasive management of iatrogenic ureteral injuries. J Endourol 2010;12(24):1921–27. 17. Keeling AN, Lee MJ. Crossing ureteric strictures: microcatheters to the rescue when conventional methods fail. Cardiovasc Intervent Radiol 2007;6(30):1234–7. 18. Yates DR, Mehta SS, Spencer PA, Parys BT. Combined antegrade and retrograde endoscopic retroperitoneal bypass of ureteric strictures. BJU Int 2010;7(105):992–7. 19. Lang EK. Percutaneous ureterocystostomy and ureteroneocystostomy. AJR Am J Roentgenol 1988;150:1065–8. 20. Patel U, Abubacker MZ. Ureteral stent placement without postprocedural nephrostomy tube: Experience in 41 patients. Radiology 2004;230:435–42. 21. Shin JH, Yoon HK, Ko GY, et al. Percutaneous antegrade removal of double J ureteral stents via a 9-F nephrostomy route. J Vasc Interv Radiol 2007;9(18):1156–61. 22. Liang HL, Yang TL, Huang JS, et al. Antegrade retrieval of ureteral stents through an 8-French percutaneous nephrostomy route. AJR Am J Roentgenol 2008;5(191):1530–5.

23. Bhayani SB, Landman J, Slotoroff C, et al. Transplant ureter stricture: Acucise endoureterotomy and balloon dilation are effective. J Endourol 2003;17:19–22. 24. Koral K, Saker MC, Morello FP, et al. Conventional versus modified technique for percutaneous nephrostomy in newborns and young infants. J Vasc Interv Radiol 2003;14:113–6. 25. Gupta S, Gulati M, Shankar K, et al. Percutaneous nephrostomy with real-time sonographic guidance. Acta Radiol 1997;38:454–7. 26. Lee WJ, Mond DJ, Patel M, et al. Emergency percutaneous nephrostomy: technical success based on level of operator experience. J Vasc Interv Radiol 1994;5:327–30. 27. Rock BG, Leonard AP, Freeman SJ. A training simulator for ultrasoundguided percutaneous nephrostomy insertion. Br J Radiol 2010;991(83): 612–4. 28. Watson RA, Gillian MT, Patel U. Primary antegrade ureteric stenting: prospective experience and cost-effectiveness analysis in 50 ureters. Clin Radiol 2001;56:568–74. 29. Chung SY, Stein RJ, Landsittel D, et al. 15-year experience with the management of extrinsic ureteral obstruction with indwelling ureteral stents. J Urol 2004;172:592–5. 30. Byun SS, Kim JH, Oh SJ, Kim HH. Simple retrograde balloon dilation for treatment of ureteral strictures: etiology-based analysis. Yonsei Med J 2003;44:273–8. 31. Richter F, Irwin RJ, Watson RA, Lang EK. Endourologic management of benign ureteral strictures with and without compromised vascular supply. Urology 2000;55:652–7. 32. Cornud F, Chretien Y, Helenon O, et al. Percutaneous incision of stenotic uroenteric anastomoses with a cutting balloon catheter: longterm results. Radiology 2000;214:358–62. 33. Modi AP, Ritch CR, Arend D, et al. Multicenter experience with metallic ureteral stents for malignant and chronic benign ureteral obstruction. J Endourol 2010;7(24):1189–93. 34. Liatsikos EN, Kagadis GC, Barbalias GA, Siablis D. Ureteral metal stents: A tale or a tool? J Endourol 2005;19:934–9. 35. Lewis S, Patel U. Major complications after percutaneous nephrostomy—lessons from a department audit. Clin Radiol 2004; 59:171–9. 36. Collares FB, Faintuch S, Kim SK, Rabkin DJ. Reinsertion of accidentally dislodged catheters through the original track: what is the likelihood of success? J Vasc Interv Radiol 2010;6(21):861–4. 37. Millward SF. Percutaneous nephrostomy: a practical approach. J Vasc Interv Radiol 2000;11:955–64. 38. Ryan MJ, Ryan BM, Smith TP. Antibiotic prophylaxis in interventional radiology. J Vasc Interv Radiol 2004;15:547–56. 39. Cullen IM, Manecksha RP, McCullagh E, et al. The changing pattern of antimicrobial resistance within 42,033 Escherichia coli isolates from nosocomial, community and urology patient-specific urinary tract infections, Dublin, 1999–2009. BJU Int 2012;109:1198–206. 40. Gray RR, McLoughlin RF, Pugash RA, et al. Outpatient percutaneous nephrostomy. Radiology 1996;198:85–8. 41. Lee MJ, Papanicolaou N, Nocks BN, et al. Fluoroscopically guided percutaneous suprapubic cystostomy for long-term bladder drainage: an alternative to surgical cystostomy. Radiology 1993;188:787–9. 42. Papanicolaou N, Pfister RC, Nocks B. Percutaneous, large-bore suprapubic cystostomy: technique and results. AJR Am J Roentgenol 1989;152:303–6. 43. Aguilera PA, Choi T, Durham BA. Selected topics: emergency radiology. J Emerg Med 2004;26:319–21.

CHAPTER

147

Renal and Perirenal Fluid Collection Drainage Justin J. Campbell and Debra A. Gervais

Percutaneous image-guided drainage is a useful technique for managing many fluid collections involving the upper genitourinary tract. For infected collections in the peritoneal cavity, percutaneous abscess drainage (PAD) has become the primary drainage procedure since its introduction in the early 1980s. This procedure’s safety and efficacy has been demonstrated in numerous large clinical trials.1-5 The logical extension of the technique to collections in the kidney and perirenal space has gained acceptance and in many cases obviated the need for urologic surgery. Benefits of percutaneous drainage of fluid collections in most cases include avoiding general anesthesia, laparotomy, and prolonged postoperative hospitalization, with resultant reduction in morbidity, mortality, and hospital costs. In this chapter, we address techniques for percutaneously draining renal and perirenal abscesses as well as other fluid collections such as hematomas and urinomas. Percutaneous nephrostomy techniques and ureteral stenting are addressed in other chapters.

INDICATIONS Percutaneous fluid drainage is a therapeutic option that lies between medical and surgical management. In general, the prerequisites for percutaneous renal and perirenal fluid collection drainage are identical to the indications for fluid collection drainage elsewhere in the abdomen—presence of a fluid collection and one or more of the following: suspicion the collection is infected, need for fluid characterization, or suspicion the collection is producing symptoms that warrant drainage.6 A few caveats should be added in the setting of suspected urinomas. The majority of small urinomas will resolve spontaneously, but if a small urinoma fails to resolve in several days or if a large urinoma is discovered, percutaneous drainage should be considered. Regardless of the collection size, if there is any suspicion a urinoma is infected, drainage should be pursued. The goal of drainage can be complete cure of a collection, or drainage may be employed as a temporizing measure before definitive surgical treatment. For example, in a critically ill patient with a suspected abscess in a nonfunctioning kidney, PAD can be performed to stabilize the patient before performing a definitive urologic procedure like nephrectomy. Similarly, treatment of urinomas caused by obstruction mandate not only drainage of the urinoma but also correction of the underlying leak and/or obstruction and reestablishment of urine flow to the bladder or to the outside for successful treatment of the urinoma.

CONTRAINDICATIONS Contraindications to percutaneous fluid collection drainage are relative and should be weighed carefully against the

suitable medical and surgical alternatives. According to the American College of Radiology (ACR) guidelines, the relative contraindications include coagulopathy that cannot be adequately corrected, hemodynamic instability that precludes completion of the procedure, inability of the patient to be positioned or cooperate adequately for the procedure, known adverse reaction to contrast medium when its administration is deemed necessary for success of the procedure, lack of a safe access route for drainage of a collection, and severely compromised cardiopulmonary status in patients undergoing procedures where risk or further cardiopulmonary compromise are inherent to the procedure.6 In each case, the risks of the procedure must be carefully assessed. The decision regarding management of renal and perirenal fluid collections is best made in concert with the surgical and medical teams caring for the patient. Lastly, every attempt is made to identify pregnant patients before their exposure to ionizing radiation. Pregnant patients undergoing procedures requiring use of ionizing radiation are counseled regarding the risk to the fetus of radiation exposure, as well as the clinical benefit of the procedure. Every attempt is made to avoid exposing the fetus to radiation in accordance with ALARA (as low as reasonably achievable) principles. Efforts to reduce radiation exposure for these patients include ultrasound guidance whenever possible. If computed tomography (CT) is required for a procedure, imaging is limited to only those areas of the body required to perform the procedure.

EQUIPMENT As with abscess drainage in the peritoneal cavity, the primary tool necessary for abscess drainage is the drainage catheter. A large number of catheters are designed for fluid collection drainage. The majority are constructed from a kink-resistant polyurethane material. Catheters can be placed by either the Seldinger or trocar method. The trocar method consists of placing a flexible polyurethane catheter loaded with a metal stiffener and pointed metal trocar directly into a collection. Once the tip is placed into the collection, the catheter is fed off the trocar and stiffener into the collection. The trocar and stiffener are then removed. Alternatively, the catheter can be placed by the Seldinger method, which consists of placing the catheter into a collection over a wire. After placing a wire into a collection, the tract is dilated using serial dilators over the wire. The catheter containing the metal stiffener is placed over the wire and fed off the stiffener once it has passed through any tissues likely to prevent the catheter from easily following the wire (e.g., fascial planes, muscle). Many catheters are available with a hydrophilic coating that facilitates placement and deployment. Catheters are available in sizes ranging from 8F to 16F. Numerous large, “oblong-shaped” 1089

1090 PART 2 NONVASCULAR INTERVENTIONS · SECTION 22 THE GENITOURINARY TRACT holes at the catheter tip facilitate drainage. The majority of catheter tips have a string-locked pigtail configuration for retention, but other internal retention devices such as balloons and mushroom types are also available. Needle access is generally performed before catheter placement. A variety of needles can be used. Superficial collections are generally amenable to access with spinal needles. Deeper collections can be sampled using 10- to 25-cm Chiba needles. Needle diameters range from 18 to 22 gauge. Sampling of fluid is usually possible with 20-gauge needles, but if there is suspicion that the collection is viscous, larger-gauge needles can be employed. This needle can then be used as a guide for the tandem trocar technique. If the Seldinger technique is employed, the access needle can be used to place the wire over which the drainage catheter will be placed. Several needles designed to accept a 0.038-inch wire can be used to facilitate this technique. At our institution, Ring (Cook Medical, Bloomington, Ind.) and AccuStick (MediTech/Boston Scientific, Natick, Mass.) are used in this setting. When the Seldinger technique is used, one may have to redirect a working wire to a specific location for ultimate catheter deployment. This is usually achieved with a directional catheter to guide the wire to a specific target location under fluoroscopic guidance. A number of such catheters have been developed for vascular radiology. A 5F Kumpe catheter or Cobra catheter are useful for this purpose. These catheters accept 0.038-inch wires. In general, directional catheters measuring 40 to 65 cm in length are used in the nonvascular interventional radiology setting. When using the Seldinger technique, tract dilation is often necessary to allow catheter placement. This can be achieved using dilators, stiff devices with tapered ends that are placed over the working wire. When dilating the tract, care must be taken to not kink the wire. After performing serial dilation, the catheter can then be placed over the wire. Alternatively, the catheter can be placed after dilation with a Peel-Away introducer (Cook Medical).

TECHNIQUE Anatomy and Approaches Pus, urine, blood, and even pancreatic fluid can accumulate in the renal parenchyma and surrounding tissues. An understanding of distribution of the fluid in the renal and perirenal spaces must be founded on an understanding of the fascial planes that compose this area. The fascia that surrounds the kidneys envelops the perirenal spaces that contain the adrenal glands, proximal ureters, and fat that surrounds the kidneys. This perirenal fascia forms a long, tapered cone that reflects the embryologic ascent of the kidneys from the pelvis into their adult position in the retroperitoneum. The eponymous names of the anterior and posterior layers of the renal fascia are the Gerota fascia and Zuckerkandl fascia, respectively, but the name Gerota fascia is frequently used in the medical literature to describe both the anterior and posterior layers of renal fascia.7 The anterior and posterior layers fuse laterally to join the lateral conal fascia. Although there is no consensus on the strength or completeness of the inferior fascia, published reports suggest that inferior patency exists in some patients, allowing bidirectional flow of fluid between the perirenal and

pelvic compartments.8-10 Similarly, although there is some controversy about the medial boundaries of the perirenal spaces, there is published evidence that patency between the right and left perirenal spaces across the midline exists at and below the level of the lower poles of the kidneys.9 The lower abdominal aorta and inferior vena cava are within this midline communication between the two perirenal spaces. Posterior to the perirenal space is the posterior pararenal space, which contains only a variable amount of properitoneal fat. This space is bounded posteriorly by the transversalis fascia and anteriorly by the perirenal fascia. Because this space contains only fat, it is rarely the primary site of a pathologic process. The perirenal fascia fuses at its posterior and lateral aspects to the surface of the pro-peritoneal fat. The fused surfaces of the posterior perirenal fascia and anterior surface of the posterior pararenal space create a potential space, the posterior interfascial or retrorenal plane, which may be filled from fluid arising in either of the compartments it bounds.11 Anterior to the perirenal space lies the anterior pararenal space that contains the ascending and descending colon, pancreas, and portions of the duodenum. As with the fusion of the posterior perirenal fascia and anterior surface of the posterior pararenal space, the fusion of the anterior perirenal fascia creates an additional potential space, the anterior interfascial plane or retromesenteric fusion plane, which can also be filled with fluid arising from the compartment it bounds.11 On the left, lateral in relation to the anterior renal fascia, the dorsal mesocolon fuses with the surface of the posterior pararenal fat to form the lateral conal fascia. Similarly, on the right, lateral in relation to the anterior pararenal fascia, the mesentery of the ascending colon fuses with the posterior renal fat to form the lateral conal fascia. Potential spaces are created with the formation of these interfascial planes as well. The retromesenteric anterior interfascial space, retrorenal posterior interfascial space, and lateral conal plane communicate at the fascial trifurcation.11 Within the perirenal space, the perinephric fat is divided by numerous thin fibrous lamellae into multiple smaller compartments that may or may not communicate. These fibrous structures are continuous with and interconnect the renal capsule and anterior and posterior renal fascia planes. These connections allow for bidirectional spread of blood, fluid, or edema from the perirenal fascial planes into the perinephric space.11 The majority of abscesses in the kidney and perirenal space are the result of pyelonephritis. Other processes that can result in renal and perirenal abscesses include direct extension of infection from an adjacent compartment, superinfection of collections in the perirenal space, and rare processes such as a dropped gallstone. Before discussing drainage of these collections, it is important to clarify terminology. The literature regarding imaging and treatment of acute bacterial renal infections is plagued with inconsistency and ambiguity. The terms acute pyelonephritis, acute bacterial nephritis, and lobar nephronia are frequently used, but there is no consensus among radiologists, urologists, pathologists, nephrologists, and infectious disease experts on the meaning of these terms. The terms bacterial nephritis12 and lobar nephronia13 were initially introduced to identify a subset of patients with acute pyelonephritis who had severe regional or generalized

CHAPTER 147 RENAL AND PERIRENAL FLUID COLLECTION DRAINAGE 1091 parenchymal abnormalities without abscess, a protracted clinical course, and eventual atrophy of the affected parenchyma. The majority of these patients were diabetic or immunocompromised. The goal of identifying this group of patients was to distinguish them from patients with uncomplicated acute pyelonephritis who responded rapidly to antibiotics, as well as from patients with renal and/or extrarenal abscesses. The term lobar nephronia was used to describe a lobar distribution of pathology. For the most part, infections of the renal parenchyma are the result of ascending spread of pathogens from the lower tract or hematogenous seeding of the renal parenchyma.14 In experimental models of acute ascending nonobstructive renal infection, vesicoureteral and intrarenal reflux have been demonstrated to carry bacteria to the pyelocaliceal system and papillary ducts and tubules, where an inflammatory response is initiated.15 Inflammation then radiates centrifugally from the papilla to the cortex along medullary rays in a lobar or sublobar distribution. The configuration of the duct openings at the papilla in part determines which papilla will allow reflux of urine and which will not.16 It was this proposed lobar distribution that gave rise to the term lobar nephronia.13 Bacterial nephritis was an additional term used for those patients who did not have lobar distribution of infection but who were likely to have a severe disease process with a protracted clinical course. Both of these terms have shortcomings. First, the literature now clearly demonstrates that otherwise healthy children and adults can have severe infections of the renal parenchyma that prolong treatment and result in parenchymal loss.17,18 Second, extensive experience with CT has now demonstrated that renal infections frequently do not correspond to a renal lobe. Although most renal infections are thought to be the result of an ascending infection, the extent to which reflux of urine is responsible for inoculation is unclear; acute pyelonephritis occurs in both adults and children, without evidence of reflux. One model used to explain this phenomenon is the ability of certain strains of virulent Escherichia coli to ascend the ureter against the flow of urine by the expression of surface adhesins. The bacteria then colonize the upper tract and penetrate parenchyma via collecting ducts and tubules.14 Furthermore, although less common, hematogenous spread of infection also accounts for many cases of renal infection. In animal models, bacteria reaching the renal parenchyma by this route begin in the cortex and involve the medulla within 24 to 48 hours. These lesions are typically multiple and rounded and have a nonlobar distribution throughout the periphery of the kidney. After 48 hours, however, distinguishing the two patterns is often not possible because spreading inflammation obscures the initial underlying pattern.19 To escape this nosologic quagmire in a way that reflects the underlying pathophysiologic process, takes into account the continuous nature of the spectrum of focal or diffuse renal infections, and allows the radiologist to convey an accurate, easily understood report of the process involving the kidney that will help guide therapeutic decisions, the Society of Uroradiology proposes a simplified nomenclature based on the term acute pyelonephritis.19 For CT imaging of patients clinically diagnosed with acute pyelonephritis, the society recommends that all regions of hypoattenuation visualized in the kidney be considered

FIGURE 147-1. Axial computed tomography image with intravenous contrast enhancement demonstrates slight enlargement of left kidney relative to right. Note areas of heterogeneous enhancement (arrowheads) without a discrete area of hypoattenuation with a discrete wall to suggest abscess formation. These findings are most consistent with acute pyelonephritis in this patient with fever, left flank pain, and leukocytosis.

evidence of acute pyelonephritis (Fig. 147-1). The report should then further characterize the process as (1) focal (uni- or multi-) or diffuse, (2) unilateral or bilateral, (3) with or without focal or diffuse enlargement of the kidney, and (4) complicated (renal or extrarenal abscess, obstruction, gas formation) or not.19 If treated too late or inadequately, severe bacterial renal infections may progress to form an abscess. Initially, microabscesses will form that can then coalesce to form macroabscesses. Historically, pathologists and surgeons have used the term renal carbuncle to describe multiple coalescent hematogenously seeded renal abscesses. Over time, the term came to be used to describe any renal abscess. Although there is a distinctive macroscopic appearance to this entity, no corresponding imaging features have been described, so the term should be avoided by radiologists. A mature abscess will have a well-defined wall and on CT imaging will appear as an area of nonenhancement within renal parenchyma, with a peripheral rim of enhancement (Fig. 147-2). Before forming a mature abscess, however, there is an initial liquefaction of renal parenchyma in regions of inflammation. On CT, this can appear as an area of low attenuation measuring near water density that does not enhance. Importantly, this lesion may not yet be a drainable collection. When CT findings are indeterminate, needle aspiration can be performed. If less than 1 mL of fluid is aspirated, it is likely this lesion is not drainable and will heal with long-term antibiotic therapy.19 Abscesses in the perirenal space most commonly result from rupture of a renal abscess into the perinephric fat. Additional causes of perirenal abscesses include extension of abscesses from adjacent compartments into the perirenal space and superinfection of collections in the perirenal space (e.g., urinomas, hematomas). Continuous leak of urine into the perinephric fat will result in a collection of urine. These collections are most commonly referred to as urinomas, but other names have been used in the medical literature, including uriniferous perirenal pseudocyst.20 Urinomas are most commonly the result of forniceal rupture in the setting of obstructive uropathy. Other causes include both iatrogenic and noniatrogenic trauma (Fig. 147-3).

1092 PART 2 NONVASCULAR INTERVENTIONS · SECTION 22 THE GENITOURINARY TRACT

A

C

B

FIGURE 147-2. A, Axial computed tomography (CT) image with intravenous contrast enhancement demonstrates slight enlargement of left kidney relative to right. Note areas of heterogeneous enhancement with a discrete area of hypoattenuation with a perceptible wall in interpolar region of left kidney (arrow). These findings are most consistent with acute pyelonephritis and abscess in this patient with fever, left flank pain, and leukocytosis. B, Under CT guidance, a 20-gauge Chiba needle (arrow) was placed into area of discrete hypoattenuation; 2 mL of purulent material was aspirated. C, A 12F locking pigtail catheter (arrow) was then placed into collection using tandem trocar method; 10 mL of purulent material was aspirated.

A

C

B

FIGURE 147-3. A, Axial computed tomography (CT) image with intravenous contrast enhancement in pyelographic phase in this patient after blunt abdominal trauma demonstrates fractured right kidney. A large low-attenuation collection is visualized adjacent to lower pole of kidney (arrowheads). A small amount of high-density excreted contrast is visualized within this low-attenuation collection. B, Delayed image demonstrates filling of this large low-attenuation collection with excreted contrast (arrowheads) consistent with a large urinoma. C, A 10F pigtail catheter (arrow) was placed into this collection by trocar method under CT guidance. Urinoma resolved without need for additional urinary diversion.

Hematomas in the perirenal space have numerous causes. Abdominal trauma with resultant renal contusion, laceration, or fracture can cause a perinephric hematoma. Additionally, rupture of an aortic aneurysm can lead to a hematoma confined to the perirenal space.21 Spontaneous perirenal hematomas raise concern for an underlying neoplasm such as renal cell carcinoma or angiomyelolipoma.22 The underlying lesion may be small and can be obscured by the hematoma in the acute setting. Therefore, management of a completely spontaneous perirenal hematoma or a perirenal hematoma arising in the setting of minimal trauma always includes follow-up imaging until the hematoma has resorbed or another cause for the hemorrhage has been identified. Collections in the anterior pararenal space are more common on the left than on the right. Collections in the left anterior pararenal space are most frequently associated with inflammatory disease involving the tail of the pancreas (Fig. 147-4). Collections in the right anterior pararenal space can also be associated with inflammatory disease of the pancreas, but duodenal injury or perforation can also cause collections in this compartment (Fig. 147-5). Because the posterior pararenal space contains only fat, collections rarely originate from this compartment and are usually the

result of extension of fluid from an adjacent compartment (Fig. 147-6). The most common cause of a collection arising from the posterior pararenal space is a hematoma. These hematomas can occur in the setting of anticoagulation or a bleeding diathesis. Because of the intimate association of the anterior interfascial and lateral conal planes to the retroperitoneal ascending colon on the right and descending colon on the left, edema and inflammatory change associated with diverticulitis, ischemia, infectious colitis, and retrocecal appendicitis can cause fluid collections that spread into these planes.11

Technical Aspects Patient Preparation Before draining a renal or perirenal fluid collection, several issues regarding patient preparation must be addressed. Coagulation factors including prothrombin time (PT), partial thromboplastin time (PTT), and platelet count are usually checked before intervention. Any coagulopathy is then corrected to the degree possible before performing the procedure. Second, adequate antibiotic therapy is initiated if there is clinical evidence of infection. In the majority of cases, the team caring for the patient has begun antibiotic

CHAPTER 147 RENAL AND PERIRENAL FLUID COLLECTION DRAINAGE 1093

A

B

C

FIGURE 147-4. A, Axial computed tomography (CT) image with intravenous contrast enhancement demonstrates a gas- and fluid-containing collection in left anterior pararenal space (arrow) most consistent with an abscess in this patient with pancreatitis. B, Under CT guidance, a 20-gauge Chiba needle (arrow) was placed into collection. C, A 10F locking pigtail catheter was then placed into collection using tandem trocar method; 15 mL of purulent material was aspirated.

FIGURE 147-5. Axial computed tomography (CT) image with intravenous contrast enhancement demonstrates gas-containing periduodenal collection (arrow) with gas tracking into right anterior pararenal space (arrowhead) in patient who developed fever, abdominal pain, and leukocytosis after duodenal perforation from endoscopic retrograde cholangiopancreatography. Patient ultimately underwent successful transhepatic percutaneous drainage of collection.

therapy before consulting radiology for potential PAD. In the rare instance in which antibiotic therapy has not been initiated, broad-spectrum antibiotics are given before drainage. Antimicrobial therapy is not withheld for fear of sterilizing cultures; because the interventional manipulation can seed the bloodstream with bacteria, the risk of sepsis is high.23 Because most patients will experience some intraprocedural pain despite local anesthesia, intravenous (IV) sedation is almost always employed; a benzodiazepine and narcotic combination works well. At our institution, we use intravenously administered fentanyl and midazolam (Versed). Typically, fentanyl (25 μg IV) and midazolam (0.25-0.5 mg IV) are administered at 5- to 10-minute intervals until adequate sedation is achieved. IV sedation is administered according to institutional guidelines, and interventional radiologists should be familiar with their own institution’s guidelines.

Placing the Catheter There are several important considerations when planning the route for abscess drainage. The primary consideration is identifying the most direct route from skin to collection that is free of intervening organs and vital structures such as bowel, solid viscera, vessels, and nerves. Additionally, every attempt is made to avoid contaminating sterile areas. In the case of renal and perirenal fluid collection drainage, particular attention is given to the possibility of pleural transgression, but avoidance of the pleural space is not possible in all cases. Lastly, the catheter is generally placed in the most dependent position possible to facilitate drainage. For example, if the patient will be in the supine position after the catheter is placed, a posterior or lateral approach may optimize drainage. When the route for drainage has been selected, the patient can be positioned in the most appropriate position for catheter placement. When the patient is properly positioned, the skin is prepped and draped in the usual sterile fashion, and sedation is initiated. Whether employing the Seldinger or trocar technique, initial access to the abscess cavity is obtained under CT or ultrasound guidance. Fluoroscopy can be used in conjunction with either of these two modalities to confirm catheter position or assist with optimal catheter positioning in difficult cases. The decision regarding which imaging modality to employ is made on the basis of several factors, including position of the collection, presence of overlying structures, and visibility of the collection and adjacent structures with each modality. Availability and user preference also play a role in this decision. Ultrasonography has a number of advantages. The first is speed. Because ultrasonography is a real-time imaging modality, it is generally much faster than CT. Additionally, ultrasonography is also generally more widely available and less expensive than CT. The modality is portable, which allows for use at the bedside in critically ill patients whose condition is not stable enough to allow travel to the radiology suite. Lastly, ultrasound does not expose the patient to ionizing radiation. Unlike CT, however, intervening fat, gas, and bone significantly degrade visualization of collections. Abscesses deep within the abdomen or pelvis can be difficult to visualize. For the same reason, ultrasonography can be difficult to use in obese

1094 PART 2 NONVASCULAR INTERVENTIONS · SECTION 22 THE GENITOURINARY TRACT

A

B

C

FIGURE 147-6. A, Axial computed tomography (CT) image with intravenous contrast enhancement demonstrates gas- and fluid-containing collection in left posterior pararenal space (arrowheads) in patient with severe necrotizing pancreatitis. B, Under CT guidance, a 20-gauge Chiba needle (arrow) was placed into collection. C, A 12F locking pigtail catheter (arrow) was then placed into collection using tandem trocar method; 130 mL of purulent material was aspirated.

patients with large amounts of subcutaneous fat. Similarly, some collections in the retroperitoneum can be difficult to visualize. Limitations of ultrasonography also include the need for significant operator experience. CT is not as operator dependent as ultrasonography. Furthermore, deep collections and collections with overlying bone, gas, or fat are easily visualized. Provided the patient’s weight does not exceed table limits, and patient size allows placement of both the patient and a guiding needle into the gantry, CT-guided drainage is not limited by obesity to the same degree as ultrasound. Finally, both final catheter position and degree of evacuation of abscess contents are better visualized with CT than ultrasound. Disadvantages of CT include the use of ionizing radiation, increased time required for procedures, and lack of portability. Both CT and ultrasound have limitations in their ability to distinguish solid tissue from drainable fluid collections. CT relies on attenuation measurements to distinguish tissues. The majority of fluid collections are low attenuation, but some drainable fluid collections can contain high-attenuation materials like iodinated contrast medium or blood products. Likewise, some nondrainable tissues (e.g., necrotic tissues, certain neoplasms) can be very low in attenuation. Unlike CT, however, ultrasound has particular difficulty visualizing collections that contain gas, secondary to limited sound transmission. Additionally, some hypoechoic through-transmitting tissues (e.g., lymphoma) can be mistaken for fluid. Regardless of which imaging modality is ultimately chosen to guide drainage, a combination of pre-drainage diagnostic imaging modalities can be used to characterize collections and assess feasibility of drainage before attempting percutaneous drainage. The catheter may be placed using either a trocar or Seldinger technique. Both techniques are viable options and offer advantages and disadvantages in different clinical settings. Although each technique has its proponents and detractors, no randomized trial exists to compare them, and the choice of which to employ comes down to operator preference, which itself is usually a function of experience. The trocar technique is generally performed in tandem to a guiding needle. As discussed earlier, the guiding needle is usually 18 to 22 gauge and variable in length, depending on the depth of the collection. The guiding needle is placed

into the collection under CT guidance. Once needle placement within the collection has been confirmed with imaging, the needle acts as a guide for catheter placement. Before placing the catheter, a sample of fluid can be aspirated and sent for laboratory analysis. This initial aspiration can be helpful in characterizing the fluid. If no fluid is aspirated, the collection may be very viscous, suggesting the need for a larger catheter. Alternatively, if the fluid is not grossly purulent and there is a suspicion the collection is not infected, a Gram stain can be performed if the decision of whether to place a catheter will be influenced by these results. Once the decision to place the catheter has been made, a small nick is made in the skin adjacent to the guiding needle entrance site, and the underlying subcutaneous tissue is bluntly dissected. The catheter containing the hollow metal stiffener and inner trocar needle is then advanced into the cavity adjacent to the guiding needle to the appropriate predetermined depth. When the catheter tip is within the cavity, the catheter is fed off the trocar needle and metal stiffener. Syringe aspiration can then be performed to help confirm catheter position. Minimal fluid is aspirated at this time because decreasing the size of the target cavity makes subsequent repositioning technically more difficult if repositioning is required. The catheter position is then confirmed with imaging before securing the device. The guiding needle is removed after satisfactory catheter position is confirmed. The advantage of the trocar method is the speed with which the catheter can be deployed. This can be very helpful when the patient is critically ill or lightly sedated. The main disadvantage of this technique is limited ability to reposition the catheter without removing and reinserting it if initial positioning is suboptimal. Using the Seldinger technique, the initial access can be performed with a needle under ultrasound or CT guidance. When the needle position within the cavity is confirmed with imaging, a wire is placed through the needle into the cavity. The needle is then removed over the wire, and the tract is serially dilated to the appropriate size to allow catheter placement. Two different types of needles can be used for initial access. “One-stick” systems use a 21- to 22-gauge needle that accepts a 0.018-inch wire. A three-component device containing a sheath, dilator, and stiffener is placed

CHAPTER 147 RENAL AND PERIRENAL FLUID COLLECTION DRAINAGE 1095 over this wire. With removal of the stiffener and dilator, a 0.035- or 0.038-inch working wire can be placed through the sheath. Further dilation can then be performed to place the catheter over this wire into the collection. Alternatively, a larger needle such as a Ring needle (Cook Medical) can be placed that accepts a 0.035- or 0.038-inch wire. The Seldinger technique is helpful when the window to the collection is narrow. Furthermore, this technique offers the advantage of guiding the catheter into optimal position with the use of directional catheters under fluoroscopic guidance. The disadvantage of this technique is primarily that it can be more time consuming than the trocar technique, which can be a particularly important consideration in critically ill patients. Additionally, serial dilation can be painful even in well-sedated patients. After Placing the Catheter Immediately after the procedure, imaging is performed to assess adequacy of cavity drainage and final catheter location. This is generally evaluated with the modality used to place the catheter. The catheter can be hard to visualize with ultrasound, but decrease in cavity size can usually be assessed that suggests the correct catheter position. CT offers excellent visualization of both the cavity and catheter. Injection of contrast medium under fluoroscopy can also be used to confirm catheter position. At the time of initial drainage, care is exercised to be sure to inject less volume than the volume of fluid removed from the cavity so as to minimize the risk of bacteremia and acute sepsis. Because of the viscosity of iodinated contrast, the injected contrast medium is then aspirated and the catheter flushed with saline to minimize catheter clogging. Additionally, catheter position can be independently confirmed by flow of appropriate fluid (e.g., pus from an abscess). When satisfactory catheter position is confirmed, the collection is then decompressed with a large aspiration syringe until return of fluid or debris ceases. For large cavities, irrigation of the cavity with small aliquots of sterile saline can be performed until the return of the irrigant turns clear. Irrigation will often result in higher volumes of return fluid than the irrigant facilitating drainage. Irrigation must cease immediately if the irrigant volume is not completely returned with aspiration. Failure to achieve complete return indicates that the irrigant is collecting in a location not amenable to aspiration by the catheter, and further irrigant will simply expand this collection. Irrigation of the cavity must be distinguished from flushing the catheter. Irrigation involves instillation of fluid into the cavity for lavage and more complete drainage of cavity contents. Flushing is simply to clear the catheter and any associated drainage tubing to prevent occlusion of the catheter by particulate debris and necrotic tissue.24,25 Importantly, overzealous irrigation can cause bacteremia and paradoxical worsening of the patient’s clinical status. When decompression of the cavity and catheter position have been confirmed, the pigtail is formed (when using pigtail catheter), and the catheter is secured to the skin. The catheter can be sutured to the skin, but this can lead to irritation. For this reason, a number of retention devices are available that do not require a skin suture. The catheter is then connected to a drainage bag. A three-way stopcock can be placed between the catheter and drainage bag. This device has the advantage of facilitating subsequent catheter

flushes, but has the disadvantage of a narrow lumen that can impede drainage with viscous fluid or collections containing debris. In particular, for catheters larger than 14F, the standard stopcock lumen is narrower than the catheter lumen, although larger-bore stopcocks are available. The catheter is flushed with 10 mL of saline every 8 to 12 hours to prevent drained material from obstructing the catheter lumen. Additionally, the connection tubing to the bag is flushed because this tubing can also become obstructed.

CONTROVERSIES Areas of variation in the practice of percutaneous drainage of renal and perirenal collections include the use of intracavitary fibrinolytics, drainage in the setting of coagulopathies or anticoagulation therapy, and use of the Seldinger or trocar technique. The presence of infected hematomas and viscous purulent material have been associated with poor drainage despite proper catheter position. There is some evidence that instillation of fibrinolytics can shorten hospital stays and lower treatment costs, but to date no large clinical trials have been performed to establish definitive guidelines concerning the use of intracavitary fibrinolytics for the practitioner for renal and perirenal fluid collection drainage. The decision regarding intracavitary fibrinolytic use is made by the radiologist together with the other members of the team caring for the patient. Similarly, there is variation in the practice of percutaneous renal and perirenal fluid collection drainage in the setting of coagulopathies. The risk of a bleeding complication increases with the presence of a coagulopathy, but the cutoff values for PT and platelet count vary from institution to institution. Although most patients will have recent laboratory values available before percutaneous drainage, not all practitioners will check values before the procedure. Most institutions will use a cutoff for PT of 15 to 18 seconds, corresponding to an international normalized ratio of 1.5 to 2.2. Cutoff values for platelets vary from 50,000 to 70,000/ mm3. At this time, no scientific trial has established guidelines for practitioners of percutaneous drainage concerning either the necessity of checking laboratory values before percutaneous drainage or for cutoff values of PT or platelets. Practice also varies concerning percutaneous drainage in the setting of patients receiving anticoagulation therapies such as aspirin and clopidogrel (Plavix). Multiple new anticoagulation agents give rise to new challenges concerning if and when to withhold anticoagulation therapy when performing percutaneous drainage. In each case, the benefits of percutaneous drainage must be weighed carefully along with the risks of a bleeding complication, as well as the risks of withholding anticoagulation therapy. These decisions are always made in concert with the other clinicians involved in the patient’s care. Lastly, the area of greatest practice variation in percutaneous drainage concerns use of the trocar or Seldinger technique. Proponents of each may closely adhere to one and eschew the other, but at this time, no prospective clinical trial has established the supremacy of either technique. As discussed previously, there are situations in which each technique offers advantages and disadvantages. Given the extensive experience with both, at this time the decision of which technique to employ comes down to the preference of a given practitioner in a given situation.

1096 PART 2 NONVASCULAR INTERVENTIONS · SECTION 22 THE GENITOURINARY TRACT

OUTCOMES The efficacy of percutaneous renal and perirenal abscess drainage with the use of antibiotics has been well documented. Rives et al. found that treatment of renal and perirenal abscesses with antibiotics alone is not effective in the majority of cases.26 In 1981, Elyaderani reported successful treatment of a renal abscess with percutaneous drainage in a patient too ill to undergo surgery.27 Several small series subsequently demonstrated variable success with percutaneous drainage of renal abscesses.28-31 In 1988, Sacks et al. reported a series of 18 patients who underwent percutaneous treatment of infected renal and perirenal collections; 61% required no surgical intervention, and the remaining 39% underwent subsequent surgery after being “cooled off ” with percutaneous drainage. These patients experienced a reduced rate of surgical morbidity.32 The results were echoed in the 1990 report of a series of 30 patients with renal or perirenal abscesses reported by Deyoe et al.33 In this series, patients who did not improve clinically with antibiotics alone underwent percutaneous drainage; the treatment was curative in 67% of cases. The remaining patients required additional surgical intervention. In the same year, Lang reported a series of 33 patients with percutaneously drained renal and perirenal abscesses. In this series, 90% of the abscesses were eradicated without surgical intervention. All patients required some repositioning of catheter subsequent to the initial drainage, and 11 patients required additional catheter placement to break up loculations.34 Causes of failure in these series include complex collections with multiple loculations, renal stones, osteomyelitis, necrotic renal tissue, and tumor. As with percutaneous drainage of collections elsewhere in the abdomen, there is an inverse relationship between the complexity of a renal or perirenal collection and the success of percutaneous drainage. As complexity increases, efficacy of percutaneous drainage generally decreases. As demonstrated by Lang, this can be offset by aggressive catheter repositioning, as well as placement of additional catheters to break up loculations. In the setting of an infected renal cyst, renal stones can serve as a nidus of infection and therefore decrease the success of percutaneous drainage. When possible, an attempt is made to remove such stones percutaneously at the time of drainage. Similarly, if bone adjacent to a perirenal infection has become colonized, the success of percutaneous abscess drainage decreases. Definitive surgical treatment of infected bone is usually required to eradicate a perirenal abscess. Lastly, colonization of necrotic tissue or tumor by bacteria can give rise to abscesses. These collections are almost never successfully eradicated with percutaneous drainage and require surgical intervention. Percutaneous drainage can be used as a temporizing measure in this setting before definitive surgical treatment. In rare cases, a drain may be placed in a tumor abscess as a palliative measure in a patient who is not a surgical candidate. Before placement, the patient is made aware that the drain will have to remain in place for the life of the patient.

COMPLICATIONS Major complications resulting from percutaneous renal and perirenal collections are rare. Published complications

include sepsis requiring intervention, hemorrhage requiring transfusion, superinfection of sterile fluid collections, bowel transgression requiring intervention, and pleural transgression requiring intervention.32-34 Hemorrhage requiring transfusion usually is the result of laceration of visceral arteries, spleen, or liver. Hemorrhage is more likely in patients with uncorrected coagulopathies and patients with difficult access routes for drainage. Every attempt should be made to correct any underlying coagulopathy to minimize bleeding risks. Additionally, the safest access route for collection drainage is selected to avoid organs and vessels. Patients commonly become bacteremic after the procedure and develop a fever.34 This is not considered a major complication and occurs with relative frequency. Four of 33 patients in the series published by Lang experienced transient febrile episodes.34 The patient should defervesce by postprocedure day 2.35 Septic shock and bacteremia requiring new intervention are considered major complications. Superinfection of sterile cavities is also considered a major complication. For this reason, sterile technique is employed when performing drainage of any fluid collection.35 Enteric complications include bowel perforation with resultant fistula formation and/or peritonitis. These complications can occur if bowel is transgressed en route to abscess drainage or when a catheter erodes into adjacent bowel.32 In either case, if the catheter tip is placed within the bowel, the catheter is left in place, and a second catheter is placed into the abscess cavity. The catheter is then left within the bowel to allow an epithelialized tract to form, which takes between 10 and 15 days. The catheter can then be safely removed, and the tract will close spontaneously. Transgression of the pleura with catheter placement can result in pleural fluid collections and/or pneumothorax.34,36 If clinically significant, these collections and pneumothoraces can be treated with chest tube placement.37

POSTPROCEDURAL AND FOLLOW-UP CARE After the catheter has been placed in a renal or perirenal collection, the interventional radiology service should play an active role in catheter management. A collaborative approach with surgical and medical consultants helps maintain a uniform treatment plan. Moreover, because of the wide array of catheters available to different medical specialties, nursing, medical, and surgical staff may not be familiar with the types of catheters employed by interventional radiologists. Therefore, staying involved to evaluate the need for repositioning, exchange, placement of additional catheters, and removal is an important part of interventional radiology follow-up. On daily rounds, the radiologist can inspect the catheter, assess its connections, and determine the quantity and character of the drainage output as well as condition of the patient.38,39 The major criteria for catheter removal include clinical improvement of the patient, improvement in the patient’s laboratory tests, decreased drainage from the catheter, resolution of the cavity, and absence of a fistula. As already noted, it is not unusual for patients to become febrile in the first 24 hours with transient bacteremia, but they should defervesce by postprocedural day 2. Additionally, a patient’s laboratory tests should improve over a similar time course

CHAPTER 147 RENAL AND PERIRENAL FLUID COLLECTION DRAINAGE 1097 and normalize within 48 to 72 hours. If clinical status and laboratory tests are not improving, the possibility of undrained collections should be considered. Catheter outputs are another useful indicator of when a catheter should be removed. Catheter outputs must be followed closely. The nondraining catheter (outputs of 0 to 20 mL/day) is one of the most common clinical scenarios to face the team caring for the patient after catheter placement. Diagnostic possibilities include obstruction of the drainage catheter, presence of a loculated/fibrinous collection, catheter migration out of the abscess, or complete drainage of the collection. When drainage decreases, catheter patency is evaluated by a simple flush with 3 to 5 mL of sterile saline. Causes of obstruction include stopcock malfunction, internal or external kinking of the catheter, or obstruction of the lumen by debris or blood. Evaluation of the catheter at the bedside includes inspection of the stopcock and external portion of the catheter as well as gentle flushing of the catheter with saline. If the problem cannot be resolved at the bedside, additional imaging is indicated to evaluate catheter position and collection size. If the catheter is in the correct position and the collection remains, it is likely the material within the collection is too particulate or fibrinous to drain. Solutions to this problem include exchanging the catheter for a larger catheter or instilling thrombolytic therapy. If the catheter has moved out of the collection, the catheter can be repositioned or replaced. In the setting of an infected collection, if fever persists or drainage increases or remains high (>100 mL/day), it is necessary to re-image with CT or an abscessogram to evaluate for a residual abscess or development of new abscesses or communication with the bowel. Up to 40% of catheters require repositioning or placement of additional drainage catheters.40 In the series published by Lang et al., which reported the highest success curative rate for percutaneous renal/perirenal abscess drainage, catheter repositioning was performed in all 33 cases, and additional catheters were added in 11 patients. Fistulas can be to the bladder, pancreatic duct, or bile duct, but the most common fistula to occur after abscess drainage is to the bowel.41 The presence of a fistula can be confirmed with a fluoroscopically guided catheter injection using iodinated dye. The catheter should remain in place long enough for enteric communication to close. Many of these communications close within 2 weeks,39 but it is not unusual for fistulas to close over weeks to months. In the case of a urinoma, although there is no absolute quantity of output from a drainage catheter that indicates the presence of a continuous urine leak, outputs are expected to decrease daily. Similarly, on follow-up imaging, the urinoma should decrease in size. If outputs from the catheter do not decrease, or if the urinoma does not demonstrate evidence of resolution on follow-up imaging despite an appropriately placed urinoma drainage catheter, diversion of urine to allow healing of the defect in the

collecting system should be considered. This can be achieved with a percutaneous nephrostomy catheter in addition to the percutaneous urinoma drainage catheter to decompress the collecting system and facilitate urinary drainage. In some cases, however, a nephrostomy catheter alone may not be sufficient to divert enough urine to allow for healing of the defect in the collecting system. In these cases, it is necessary to place a nephrostomy catheter in combination with placement of a ureteral stent or nephro-ureteral catheter. The combination of a percutaneous drainage catheter with either a nephrostomy catheter and a ureteral stent or with a nephroureteral catheter can allow for greater diversion of urine away from the area of the leak and promote healing of the collecting system.42-44 KEY POINTS •

Percutaneous drainage is an effective technique for treating many infected renal and perirenal collections.

•

The technique can be employed as a curative procedure or as a temporizing measure before definitive surgical treatment.

•

Urinomas are frequently amenable to treatment with percutaneous drainage alone or percutaneous drainage in combination with urinary diversion techniques.

▶ SUGGESTED READINGS Bernardino ME, Baumgartner BR. Abscess drainage in the genitourinary tract. Radiol Clin North Am 1986;24:539–49. Deyoe LA, Cronan JJ, Lambiase RE, Dorfman GS. Percutaneous drainage of renal and perirenal abscesses: Results in 30 patients. AJR Am J Roentgenol 1990;155:81–3. Gore RM, Balfe DM, Aizenstein RI, Silverman PM. The great escape: interfascial decompression planes of the retroperitoneum. AJR Am J Roentgenol 2000;175:363–70. Lambiase RE, Deyoe L, Cronan JJ, Dorfman GS. Percutaneous drainage of 335 consecutive abscesses: results of primary drainage with 1-year follow-up. Radiology 1992;184:167–79. Lang EK. Renal, perirenal, and pararenal abscesses: percutaneous drainage. Radiology 1990;174:109–13. Lang EK, Glorioso 3rd L. Management of urinomas by percutaneous drainage procedures. Radiol Clin North Am 1986;24:551–9. Sacks D, Banner MP, Meranze SG, et al. Renal and related retroperitoneal abscesses: Percutaneous drainage. Radiology 1988;167:447–51. Talner LB, Davidson AJ, Lebowitz RL, et al. Acute pyelonephritis: Can we agree on terminology? Radiology 1994;192:297–305. Titton RL, Gervais DA, Hahn PF, et al. Urine leaks and urinomas: diagnosis and imaging-guided intervention. Radiographics 2003;23:1133–47. vanSonnenberg E, Wittich GR, Goodacre BW, et al. Percutaneous abscess drainage: update. World J Surg 2001;25:362–9; discussion 370–72.

The complete reference list is available online at www.expertconsult.com.

CHAPTER 147 RENAL AND PERIRENAL FLUID COLLECTION DRAINAGE1097.e1 REFERENCES 1. Gerzof SG, Gale ME. Computed tomography and ultrasonography for diagnosis and treatment of renal and retroperitoneal abscesses. Urol Clin North Am 1982;9:185–93. 2. Johnson WC, Gerzof SG, Robbins AH, Nabseth DC. Treatment of abdominal abscesses: comparative evaluation of operative drainage versus percutaneous catheter drainage guided by computed tomography or ultrasound. Ann Surg 1981;194:510–20. 3. vanSonnenberg E, Ferrucci Jr JT, Mueller PR, et al. Percutaneous drainage of abscesses and fluid collections: technique, results, and applications. Radiology 1982;142:1–10. 4. vanSonnenberg E, Wing VW, Casola G, et al. Temporizing effect of percutaneous drainage of complicated abscesses in critically ill patients. AJR Am J Roentgenol 1984;142:821–6. 5. Lambiase RE, Deyoe L, Cronan JJ, Dorfman GS. Percutaneous drainage of 335 consecutive abscesses: results of primary drainage with 1-year follow-up. Radiology 1992;184:167–79. 6. American College of Radiology (ACR). Percutaneous catheter drainage of infected intra-abdominal fluid collections. ACR Appropriateness Criteria. Radiology 2000;215(Suppl). Accessed from www.acr.org on February 15, 2006. 7. Chesbrough RM, Burkhard TK, Martinez AJ, Burks DD. Gerota versus Zuckerkandl: the renal fascia revisited. Radiology 1989;173:845–6. 8. Korobkin M, Silverman PM, Quint LE, Francis IR. CT of the extraperitoneal space: normal anatomy and fluid collections. AJR Am J Roentgenol 1992;159:933–42. 9. Kneeland JB, Auh YH, Rubenstein WA, et al. Perirenal spaces: CT evidence for communication across the midline. Radiology 1987;164: 657–64. 10. Auh YH, Rubenstein WA, Schneider M, et al. Extraperitoneal paravesical spaces: CT delineation with US correlation. Radiology 1986;159: 319–28. 11. Gore RM, Balfe DM, Aizenstein RI, Silverman PM. The great escape: interfascial decompression planes of the retroperitoneum. AJR Am J Roentgenol 2000;175:363–70. 12. Davidson AJ, Talner LB. Urographic and angiographic abnormalities in adult-onset acute bacterial nephritis. Radiology 1973;106:249–56. 13. Rosenfield AT, Glickman MG, Taylor KJ, et al. Acute focal bacterial nephritis (acute lobar nephronia). Radiology 1979;132:553–61. 14. Roberts JA. Etiology and pathophysiology of pyelonephritis. Am J Kidney Dis 1991;17:1–9. 15. Rushton HG, Majd M. Dimercaptosuccinic acid renal scintigraphy for the evaluation of pyelonephritis and scarring: a review of experimental and clinical studies. J Urol 1992;148(5 pt 2):1726–32. 16. Torres VE, Kramer SA, Holley KE, et al. Interaction of multiple risk factors in the pathogenesis of experimental reflux nephropathy in the pig. J Urol 1985;133:131–5. 17. Rathore MH, Barton LL, Luisiri A. Acute lobar nephronia: a review. Pediatrics 1991;87:728–34. 18. Rauschkolb EN, Sandler CM, Patel S, Childs TL. Computed tomography of renal inflammatory disease. J Comput Assist Tomogr 1982;6: 502–6. 19. Talner LB, Davidson AJ, Lebowitz RL, et al. Acute pyelonephritis: can we agree on terminology? Radiology 1994;192:297–305. 20. Healy ME, Teng SS, Moss AA. Uriniferous pseudocyst: computed tomographic findings. Radiology 1984;153:757–62. 21. Rosen A, Korobkin M, Silverman PM, et al. CT diagnosis of ruptured abdominal aortic aneurysm. AJR Am J Roentgenol 1984;143: 265–8. 22. Yip KH, Peh WC, Tam PC. Spontaneous rupture of renal tumours: the role of imaging in diagnosis and management. Br J Radiol 1998;71: 146–54.

23. Hovsepian DM. Transrectal and transvaginal abscess drainage. J Vasc Interv Radiol 1997;8:501–15. 24. vanSonnenberg E, Ferrucci Jr JT, Mueller PR, et al. Percutaneous radiographically guided catheter drainage of abdominal abscesses. JAMA 1982;247:190–2. 25. Mueller PR, vanSonnenberg E, Ferrucci Jr JT. Percutaneous drainage of 250 abdominal abscesses and fluid collections: II. Current procedural concepts. Radiology 1984;151:343–7. 26. Rives RK, Harty JI, Amin M. Renal abscess: emerging concepts of diagnosis and treatment. J Urol 1980;124:446–7. 27. Elyaderani MK, Subramanian VP, Burgess JE. Diagnosis and percutaneous drainage of a perinephric abscess by ultrasound and fluoroscopy. J Urol 1981;125:405–7. 28. Finn DJ, Palestrant AM, DeWolf WC. Successful percutaneous management of renal abscess. J Urol 1982;127:425–6. 29. Kuligowska E, Newman B, White SJ, Caldarone A. Interventional ultrasound in detection and treatment of renal inflammatory disease. Radiology 1983;147:521–6. 30. Cronan JJ, Amis Jr ES, Dorfman GS. Percutaneous drainage of renal abscesses. AJR Am J Roentgenol 1984;142:351–4. 31. Fernandez JA, Miles BJ, Buck AS, Gibbons RP. Renal carbuncle: comparison between surgical open drainage and closed percutaneous drainage. Urology 1985;25:142–4. 32. Sacks D, Banner MP, Meranze SG, et al. Renal and related retroperitoneal abscesses: Percutaneous drainage. Radiology 1988;167: 447–51. 33. Deyoe LA, Cronan JJ, Lambiase RE, Dorfman GS. Percutaneous drainage of renal and perirenal abscesses: Results in 30 patients. AJR Am J Roentgenol 1990;155:81–3. 34. Lang EK. Renal, perirenal, and pararenal abscesses: percutaneous drainage. Radiology 1990;174:109–13. 35. vanSonnenberg E, Mueller PR, Ferrucci Jr JT. Percutaneous drainage of 250 abdominal abscesses and fluid collections: I. Results, failures, and complications. Radiology 1984;151:337–41. 36. Neff CC, Mueller PR, Ferrucci Jr JT, et al. Serious complications following transgression of the pleural space in drainage procedures. Radiology 1984;152:335–41. 37. McNicholas MM, Mueller PR, Lee MJ, et al. Percutaneous drainage of subphrenic fluid collections that occur after splenectomy: efficacy and safety of transpleural versus extrapleural approach. AJR Am J Roentgenol 1995;165:355–9. 38. Goldberg MA, Mueller PR, Saini S, et al. Importance of daily rounds by the radiologist after interventional procedures of the abdomen and chest. Radiology 1991;180:767–70. 39. vanSonnenberg E, Wittich GR, Goodacre BW, et al. Percutaneous abscess drainage: update. World J Surg 2001;25:362–9; discussion 370–72. 40. Stabile BE, Puccio E, vanSonnenberg E, Neff CC. Preoperative percutaneous drainage of diverticular abscesses. Am J Surg 1990;159:99– 104; discussion. 41. Papanicolaou N, Mueller PR, Ferrucci Jr JT, et al. Abscess-fistula association: radiologic recognition and percutaneous management. AJR Am J Roentgenol 1984;143:811–5. 42. Matthews LA, Spirnak JP. The nonoperative approach to major blunt renal trauma. Semin Urol 1995;13:77–82. 43. Kantor A, Sclafani SJ, Scalea T, et al. The role of interventional radiology in the management of genitourinary trauma. Urol Clin North Am 1989;16:255–65. 44. Titton RL, Gervais DA, Hahn PF, et al. Urine leaks and urinomas: diagnosis and imaging-guided intervention. Radiographics 2003;23: 1133–47.

CHAPTER

148

Thermal Ablation of Renal Cell Carcinoma Kamran Ahrar, Surena F. Matin, and Michael J. Wallace

In 2013, over 65,150 new cases of kidney cancer were diagnosed in the United States, resulting in over 13,680 deaths.1 Renal parenchymal tumors (i.e., renal cell carcinoma [RCC]) account for the majority (85%) of kidney cancers.2 Urothelial cancers of the renal pelvis (i.e., transitional cell carcinoma [TCC]) account for most of the remaining cases. The incidence of RCC has increased on average 2% to 3% a year for the past 3 decades.3 At present, over 50% of RCCs are detected incidentally during imaging studies prompted by nonurologic complaints.4,5 Tumors detected incidentally are smaller (stage I [Table 148-1]) at diagnosis and are often asymptomatic.3 Partial nephrectomy is the standard of care for small RCCs and provides excellent results. Recurrencefree survival, cancer-specific survival, and overall survival rates for open or laparoscopic partial nephrectomy have been reported as better than 95%, 97%, and 89%, respectively.6 Other alternative treatment options advocated by the American Urological Association (AUA) include active surveillance and thermal ablative therapies.6 Thermal ablative therapies destroy tumors by using thermal energy—either heat (e.g., radiofrequency [RF], laser, microwave, or focused ultrasound) or cold (e.g., cryoablation).7 The technologies most widely used in clinical practice for ablation of renal tumors are RF ablation (RFA) (Figs. 148-1 to 148-4) and cryoablation (Figs 148-5 and 148-6). Laser, microwave, and high-intensity focused ultrasound technologies are in various stages of experimental or early clinical application but have not been utilized to any substantial degree in clinical practice for treating renal tumors. With either cryoablation or RFA, once the tumor is ablated, it is left in situ. This is a departure from the basic oncologic surgical principle of excision with clear margins. For ablation technology to be considered a clinically acceptable alternative to extirpation, it must be reliable and completely destroy all viable tumor. There must also be a means of imaging and monitoring the area of ablation to preserve and protect surrounding vital structures while ensuring complete tumor destruction. Thermal ablation of renal tumors can be performed either intraoperatively or percutaneously. Historically, RFA has been used most commonly for percutaneous image-guided ablation, whereas cryoablation has been used most commonly in open or laparoscopic approaches.8 More recently, small-diameter cryoprobes have become available for percutaneous use, resulting in an increasing number of percutaneous cryoablation procedures. At the same time, more urologic surgeons have adopted intraoperative RFA over cryoablation, resulting in an increasing number of laparoscopic RFA procedures. The basic principles of thermal ablation are discussed elsewhere in this volume. In this chapter, we briefly discuss specific ablation devices that are commonly used to treat RCC. Our main focus is a review of indications, contraindications, technique, complications, follow-up care, and outcomes of percutaneous thermal ablation of RCC. 1098

ABLATION DEVICES Radiofrequency Ablation Currently, three RFA devices have been approved by the U.S. Food and Drug Administration (FDA) and are commercially available in the United States. Each of these uses a generator to deliver alternating electric current via an electrode, causing ionic agitation and frictional heating within the target tissue. Each of these ablation systems uses a different strategy to obtain the largest possible zone of ablation. The maximum attainable in vivo ablation zone varies by applicator size and design and is consistently less than the ex vivo maximum,9 in part owing to the presence of adjacent flowing blood or large fluid-filled structures that produce a cooling or heat sink effect. RF applicators range in size from 17 to 14 gauge and vary in design. The RITA StarBurst probe (AngioDynamics, Latham, N.Y.) and the LeVeen electrode (Boston Scientific, Natick, Mass.) are both multi-tined applicators with an expandable array design that allows for a scalable teardropshaped or spherical ablation zone. The Covidien system (Covidien, Mansfield, Mass.) uses a straight probe design with a central channel that allows for circulation of chilled fluid as the basis for an internally cooled electrode. The cooling of the electrode tip and the pulsed delivery of energy help prevent tissue charring or excess gas formation. This strategy is expected to maximize the potential zone of ablation. The Cool-tip electrodes (Covidien) are available as either a single configuration or a cluster configuration in which three closely spaced (4 cm) and central tumors, but more than one ablation session may be necessary, and hemorrhagic complications may be more common in these cases.19,21,23 Close proximity of the tumor to other vital structures such as bowel or pancreas should be considered. Adjuvant techniques may have to be used to isolate the kidney and separate the tumor from other structures that may be damaged by thermal energy.

CONTRAINDICATIONS The only absolute contraindication to ablation therapy is uncorrectable coagulopathy. The majority of other “relative” contraindications, such as systemic illness or sepsis, usually can be treated before the ablation procedure. An additional contraindication to percutaneous ablation is when adjacent vital structures such as bowel cannot be separated from the zone of ablation by accepted techniques and thus cannot be protected from the effects of thermal energy.

1100 PART 2 NONVASCULAR INTERVENTIONS · SECTION 22 THE GENITOURINARY TRACT

FIGURE 148-1. Top row, This 70-year-old woman with multiple medical comorbid conditions was found to have a 3-cm solid, enhancing mass involving posterior aspect of right kidney. Middle row, She was treated with computed tomography (CT)-guided percutaneous radiofrequency ablation. Three overlapping ablations were performed. Bottom row, Immediately after ablation, axial contrast-enhanced CT images demonstrate high-density changes within tumor, lack of enhancement indicating margin of ablation zone, and a general reduction in size of exophytic tumor. (Copyright Kamran Arhar, MD.)

TECHNIQUE Patient Assessment All patients should undergo a preprocedural assessment that includes serum creatinine level, estimated glomerular filtration rate (GFR), platelet count, and a coagulation profile. Patients with proven RCC should have chest imaging

(e.g., chest radiograph) and a metabolic profile. These are standard basic tests for surveillance of metastatic disease, the incidence of which is very low but still possible for small tumors. Patients also should be assessed for their ability to tolerate intravenous sedation and should meet institutional criteria. Local institutional factors and physician preferences determine whether intravenous sedation19 or general anesthesia24,25 is used during ablation therapy. General

CHAPTER 148 THERMAL ABLATION OF RENAL CELL CARCINOMA 1101

FIGURE 148-2. Follow-up unenhanced (left) and contrast-enhanced (right) axial computed tomography images of patient in Figure 148-1 at 1, 6, 12, and 24 months demonstrate evolution of postablation changes: persistent high-density material in ablation zone, lack of enhancement, and a thin halo of soft tissue density in perinephric fat. (Copyright Kamran Arhar, MD.)

1102 PART 2 NONVASCULAR INTERVENTIONS · SECTION 22 THE GENITOURINARY TRACT

FIGURE 148-3. Top, This 42-year-old woman was diagnosed with bilateral solid, enhancing renal tumors. She also had bilateral renal cysts and multiple pancreatic cysts. A genetic workup confirmed diagnosis of von Hippel-Lindau disease. She required right nephrectomy for a large renal cell carcinoma that had replaced most of right kidney. Two separate tumors in left kidney were treated with percutaneous radiofrequency ablation. Bottom, Axial computed tomography images at 2 years demonstrate no enhancement in ablation zones in left kidney. (Copyright Kamran Arhar, MD.)

anesthesia optimizes patient tolerance, allows greater control of respiratory motion when the applicators are being placed, and may facilitate accurate tumor targeting. A percutaneous biopsy is recommended prior to ablation of renal tumors.26 The differential diagnosis of a small enhancing renal mass includes benign entities such as lipidpoor angiomyolipoma, oncocytoma, papillary adenoma, and metanephric adenoma. As the size of a renal mass decreases, the likelihood of a benign diagnosis increases. Approximately 25% of renal tumors smaller than 4 cm are benign, but when tumors smaller than 1 cm were resected, over 40% were benign.27 About 5% of angiomyolipomas are indistinguishable from small RCCs on cross-sectional imaging.28 Tuncali et al. reviewed biopsy and imaging data of 27 patients referred for cryoablation of a small renal mass.29 Ten lesions (37%) smaller than 2 cm were deemed benign. The “diagnostic accuracy” of renal biopsy has improved over the years, particularly with the addition of routine core biopsies, and is better than 95% in most contemporary series. Sensitivity for detection of malignancy ranges from 84% to 100% in studies of renal masses published after 2006.30 When percutaneous biopsy establishes a diagnosis (e.g., lymphoma or angiomyolipoma) other than RCC, ablation is not indicated. A positive result with diagnosis of RCC provides details about tumor subtype and grade, information that may become relevant during follow-up visits, insurance claims, or should the patient ever develop metastatic disease requiring systemic therapy. A positive result is also important for the validation of ablation therapy and defining the standard of care for small renal masses in the future. On the other hand, concerns over falsenegative biopsy results remain; some clinicians recommend

proceeding with therapy when a negative histologic result conflicts with imaging findings.31,32 Ideally, the biopsy should be performed during a separate encounter so sufficient time is given for a complete histologic evaluation. For large tumors (>4 cm), consideration may be given to selective embolization prior to ablation. Embolization may help improve the effectiveness of ablation by reducing perfusion-mediated cooling (the “heat sink effect” during RFA) or heating (during cryoablation) and may help reduce the risk of hemorrhagic complications.33-35

Imaging Modalities Ultrasonography, computed tomography (CT), and magnetic resonance imaging (MRI) have been used to guide renal ablation proceures.25,36-38 Ultrasonography provides valuable real-time multiplanar imaging for initial tumor targeting and placement of the applicators. Some investigators have reported favorable outcomes for ultrasoundguided RFA compared with other imaging modalities.39 However, inability to visualize the tumor, gas bubbles during RFA, and large tumor size limit the effectiveness of ultrasonography.40 During cryoablation procedures, the edge of the ice ball creates intense acoustic shadowing that precludes assessment of the leading edge of the ablation zone.41-43 Ultrasonography is not optimal for assessment of nearby loops of bowel, and images may become degraded by perinephric hematoma that may develop during the procedure. Currently, CT is the imaging modality most widely used for percutaneous image-guided renal ablation.19,24,25 It provides guidance for precise probe placement and is not affected by gas formation during RFA. CT is ideal for

CHAPTER 148 THERMAL ABLATION OF RENAL CELL CARCINOMA 1103

A

B

C

D

E

F

G

H

I

FIGURE 148-4. A, This 70-year-old man with multiple medical comorbid conditions was found to have a 4-cm mass involving upper pole of left kidney. B-C, He was referred for percutaneous radiofrequency ablation. Axial computed tomography (CT) images obtained with patient prone demonstrate interposition of lung parenchyma between chest wall and renal tumor. D, A blunt-tipped needle was placed in pleural space, and air was injected to separate parietal and visceral pleura, creating a controlled pneumothorax. E-F, Axial CT images demonstrate transpleural course of radiofrequency ablation electrode without puncturing visceral pleura. A pneumothorax evacuation tube was placed under CT guidance before termination of procedure. Tube was removed the next morning. G-I, Follow-up axial CT images at 6 months taken before contrast, in arterial phase, and in late phase demonstrate stable high-density changes with no evidence of enhancement (CT Hounsfield units measured 64, 60, and 59, respectively). (Copyright Kamran Arhar, MD.)

monitoring other vital structures (e.g., bowel) that may come in contact with the tumor. During cryoablation, the location and size of the ice ball can be visualized rather clearly on CT (see Fig. 148-6),44 whereas in RFA the ablation zone is less clearly visualized, and the boundary between treated and untreated tissue is not precisely demarcated. To overcome this problem, a contrast-enhanced postablation CT examination is typically performed just before completing the RFA (see Fig. 148-1).24,25 CT fluoroscopy enables real-time visualization of the applicator tip as it is being placed and facilitates precise targeting of the tumor. The main disadvantage of CT or CT fluoroscopy is patient and operator exposure to ionizing radiation. For this reason,

some researchers advocate using a combination of ultrasonography and CT. Ultrasonography can be used for initial placement of the applicators, and CT can be used for any additional imaging during the procedure, including repositioning the electrodes, monitoring adjacent loops of bowel, assessing the size and margins of the ice ball, and performing a contrast-enhanced study to confirm complete tumor ablation. MRI is an ideal imaging modality for ablation guidance and monitoring because it provides excellent soft-tissue contrast and spatial resolution compared with CT and ultrasound.37,38,45 In addition to providing multiplanar and near real-time imaging, MRI also allows for monitoring of

1104 PART 2 NONVASCULAR INTERVENTIONS · SECTION 22 THE GENITOURINARY TRACT

A

B

C

D

FIGURE 148-5. A, This 53-year-old man with history of esophageal cancer was found to have a 3.6 cm left renal tumor. Biopsy showed renal cell carcinoma, papillary type 1. B, Axial computed tomography (CT) image of kidney during cryoablation shows three of the four cryoprobes engulfed in a low-density ice ball. C, Immediately after two freeze/thaw cycles, contrast-enhanced CT showed active extravasation and associated retroperitoneal hematoma. Selective embolization (not shown) was promptly performed because of dropping blood pressure. D, Contrast-enhanced follow-up CT image at 12 months shows zone of ablation with no evidence of residual disease. (Copyright Kamran Arhar, MD.)

*

A

B

FIGURE 148-6. A, Axial computed tomography image of kidney during cryoablation of a left upper pole tumor shows close proximity of growing ice ball (arrows) to colon (star). B, Hydrodissection was performed by injection of 180 mL of sterile fluid with a small amount of iodinated contrast to separate colon from ice ball. (Copyright Kamran Arhar, MD.)

CHAPTER 148 THERMAL ABLATION OF RENAL CELL CARCINOMA 1105 tissue coagulation and treatment results,37 which CT does not. Also, the lack of ionizing radiation is a significant advantage over CT. Unfortunately, current challenges in producing special MRI-compatible equipment and the lack of availability of interventional MRI units have limited its use in daily clinical practice.

Ablation Procedure Optimal patient positioning is determined during the planning phase when cross-sectional images are reviewed to assess the feasibility and safety of access to the target tumor. In a typical ablation procedure, the patient is placed in the prone position. During CT-guided procedures, it is helpful to move the arms away from the body in an abducted and extended position and place them next to the patient’s head (“the superman position”). This reduces streak artifact and facilitates tumor visualization. When patients are under general anesthesia, care must be taken to not overextend the arms and to provide adequate padding to avoid brachial plexus injury.46 An ipsilateral decubitus position can help reduce the risk of pneumothorax, although it may limit access to the tumor for insertion of the applicators. A contralateral decubitus position can help separate loops of bowel from a tumor in the anterior or lateral aspect of the kidney. In rare instances, anteriorly or laterally located tumors may be treated with the patient in the supine position. An ablation margin of 5 to 10 mm around the tumor is desirable.12 The size of the ablation zone depends on the size and vascularity of the lesion, the lesion’s proximity to vascular structures, the ablation modality used, and the number, size, and configuration of the applicators. Care must be taken to avoid adjacent vital structures such as bowel and large vessels, and consideration must be given to both the potential applicator path and the projected zone of ablation. When vital structures are detected in the path of the applicator or are contiguous with the expected ablation zone, simple and noninvasive measures may suffice to render the procedure safe. These maneuvers include changing the patient’s position or levering the applicator against the skin to lift the tumor off the bowel or vascular structure. Torquing the applicator can potentially increase the distance from the tumor to the bowel by 3 or 4 mm.47 Ideally, a safe margin between the probe tines and the nearest adjacent bowel is 1 to 2 cm.48 To protect nearby structures such as bowel, nerves, or muscles, gas insufflation or hydrodissection (see Fig. 148-6) can be used to create a barrier for heat transmission.19,49,50 During RFA procedures, nonionic sterile fluid may be instilled using a small-gauge needle placed between the tumor and the bowel under CT or MRI guidance. With injection of 135 to 150 mL of fluid, one can displace the adjacent bowel loop by 2.1 to 2.5 cm.50 Often, additional hydrodissection attempts are necessary because the fluid spills into the paracolic gutters or Morrison pouch.50 Alternatively, gas can be insufflated into the peritoneal cavity or perinephric space to prevent thermal damage to adjacent vital structures. Gas has a tendency to dissipate throughout the peritoneal space; thus, larger volumes of gas are required compared with fluid. When injected into the peritoneum, a gas volume of 1200 mL yields 1.5 cm of bowel displacement.51 When injected into the perirenal space, 15 to 20 mL of gas may be sufficient to achieve a similar degree of bowel

separation.52 Adequacy of insufflation is best monitored with CT because gas can obscure the view of the tumor when MRI or ultrasonography is used.51 If hydrodissection does not create the desired effect, large angioplasty or esophageal dilation balloons may be inserted between the tumor and the structure at risk.48 More than one balloon may be necessary to create an adequate separation between the ablation zone and adjacent structures. Each balloon may be placed in coaxial fashion through an introducer sheath. Initially, a needle is inserted into the desired space. Over a guidewire, an introducer sheath is advanced into the desired position. The balloon is then placed into the sheath but not beyond it. With the balloon in position, the sheath is withdrawn. Balloon expansion is completed once the optimal position has been obtained. Peripheral thermosensors can be placed to ensure adequate ablation and to prevent thermal injury to normal renal parenchyma and adjacent structures (e.g., periureteric tissue). Carey et al. reported 100% primary effectiveness for RFA of 37 tumors that were 3-5 cm in diameter in which real-time temperature feedback of the ablation zone was used to determine the appropriate treatment endpoint.53 Ablation was continued until both deep and peripheral thermosensors recorded temperatures of at least 60°C. These independent real-time thermosensors can also be used to determine whether and where an electrode has to be redeployed. Thermal damage to the ureter or ureteropelvic junction (UPJ) can lead to stricture formation. To reduce this risk during RFA of an adjacent renal mass, retrograde pyeloperfusion with a cooled nonionic solution can be performed.54-56 In this technique, a 5F to 6F ureteral catheter is placed in retrograde fashion for infusion of cooled fluid into the renal pelvis. A 14F to 16F Foley catheter is placed in the bladder for drainage. Cantwell et al. described infusion of 1.5 to 2.0 L of 5% dextrose in water cooled overnight to 2° to 6°C at a pressure of 80 cm H2O.56 The ureteral catheter and the Foley catheter are removed at the end of the procedure. Froemming et al. described a probe retraction technique used to protect the ureter during cryoablation.57 After the cryoprobe is positioned, its proximity to important structures is assessed using CT. Activation of the cryoprobe creates an initial small ice ball that fixes the cryoprobe in relation to the tumor and also acts as a point of fixation for manipulation. By manipulating the applicators, one can retract the tumor and kidney away from structures to be avoided (e.g., the ureter). Cryoablation can then be resumed with standard freeze/thaw cycles. Renal tumors involving the upper pole of the kidney pose a risk of pneumothorax. To minimize this risk, triangulation and oblique trajectories may be considered, avoiding the lung base. Placing the patient in the ipsilateral decubitus position elevates the lung base on that side and thus reduces the plane of contact between the tumor and overlying lung. However, access to the tumor may be more limited in this position. In order to access tumors in the upper pole of the kidney, an iatrogenic pneumothorax may be created such that the electrodes would not traverse the visceral pleura or lung parenchyma (see Fig. 148-4).58 This procedure involves placing an 18- or 20-gauge needle and injecting gas (e.g., air) into the pleural space. After ablation is completed, the pneumothorax is treated with simple aspiration or

1106 PART 2 NONVASCULAR INTERVENTIONS · SECTION 22 THE GENITOURINARY TRACT placement of a small-bore (8F-10F) chest catheter under CT guidance. Alternatively, one can create an iatrogenic pleural effusion by injecting nonionic fluid instead of gas. This technique allows for precise placement and repositioning of the RFA electrodes under CT guidance, without repeated puncture of the visceral pleura. For RFA with unipolar devices, two to four dispersive electrodes are applied to the upper thighs. During prolonged ablations, the thighs should be checked periodically to assess for potential skin burns around the dispersive electrodes. Multi-tined RFA electrodes can be used for ablation of renal tumors, but care must be taken to avoid puncturing the collecting system, renal vasculature, or adjacent structures with the tines. Cool-tip RFA electrodes are needle-like in design and easier to place than multi-tined electrodes. Using a single electrode, overlapping ablations are often required to create a zone of ablation large enough to cover the entire tumor and a tumor-free margin. As many as three Cool-tip RFA electrodes can be inserted simultaneously. A switch box allows delivery of power to each electrode sequentially to achieve large areas of ablation. Cryoprobes used for percutaneous ablation are small in diameter (17 gauge to 2.4 mm) and create ice balls with cytotoxic isotherms that are often too small to completely ablate most renal tumors.59 For this reason, multiple cryoprobes are often placed at the onset to take advantage of the synergy between the probes. Cryoprobes are placed 1 to 2 cm apart and no more than 1 cm from the tumor margin.60 High technical success rates can be achieved even in large tumors (>4 cm), but hemorrhagic complication rates may be higher with insertion of multiple cryoprobes.23 Immediately after ablation, contrast-enhanced CT or MRI should be performed to assess the ablation zone, confirm complete ablation of the tumor, and rule out any complications. This is particularly relevant to RFA, during which treatment efficacy is difficult to assess. Immediately after RFA, contrast-enhanced CT shows a mild diffuse enhancement within the ablation zone. This should not be mistaken for residual tumor. The ablation zone can be better appreciated by identifying the relatively low-density, sharply demarcated margins and comparing these with the preablation imaging findings.61 Additional imaging findings include perinephric fat stranding, thickening of the perirenal fascia, locules of gas in the surrounding tissue, perinephric or subcapsular hemorrhage, and fluid in the adjacent tissues or paracolic gutters that may relate to hydrodissection.62 At some institutions, renal ablation procedures are performed on an outpatient basis,20 whereas other institutions routinely admit patients24 for overnight observation.

COMPLICATIONS The incidence of complications after thermal ablation of RCC has been reported as 3% to 12%.19,20,63,64 In a multicenter study of 271 ablation procedures,63 a total of 30 complications (11.1%) were reported. Procedures included both RFA and cryoablation that were performed either percutaneously or intraoperatively. There were 5 major complications (1.8%), 25 minor complications (9.2%), and 1 death (0.4%). Major and minor complication rates were 1.4% and 12.2% for cryoablation and 2.2% and 6% for RFA,

respectively. Minor complications associated with the percutaneous approach included probe site pain or paresthesia, wound infection, hematuria, and (rarely) infarction. Complications may arise from probe insertion or may be due to application of thermal energy. In most series, hemorrhage accounts for most of the major complications; it usually arises from direct mechanical injury to a vessel by the applicator. Tumors that are centrally located have an increased risk of hemorrhage, which may lead to hematuria and/or perinephric hematoma. In most cases, hemorrhage is self-limited and does not require treatment. Up to 2% of cases treated with RFA may require blood transfusion.19 The risk of hemorrhage may be higher with cryoablation, particularly when large and central tumors are ablated (see Fig. 148-5). In a retrospective review of 108 lesions that were larger than 3 cm and treated with percutaneous cryoablation, Schmit et al. reported an 8% major complication rate.65 Significant hemorrhage following removal of the cryoprobes from the ablated tumor occurred in four of the six patients who sustained a major complication. Fracture of the ice ball with associated parenchymal injury is a recognized complication of cryoablation that can result in significant hemorrhage.66 Potential risk factors include use of larger-diameter or multiple cryoablation probes, initiation of a second adjacent ice ball after the primary ice ball has already been formed, and removal of the cryoablation probes before the ice ball has completely thawed.66,67 If bleeding persists despite conservative measures and blood transfusion, transarterial embolization may be required. Massive hemorrhage due to an arteriovenous fistula is rare but has been described.68 When tumors are located in the upper pole of the kidney, the applicators may traverse the parietal pleura. In most cases, the visceral pleura remains intact and a pneumothorax is not encountered. The incidence of pneumothorax following ablation has been reported in up to 2% of cases.20 The majority of cases can be managed conservatively. Moderate to severe pneumothorax or pneumothorax associated with new respiratory symptoms may require aspiration or chest tube placement. Seeding of the needle tract is another potential but rare complication of ablation therapy.69 Structures that are at greatest risk of thermal injury are the ureter, nerves, and adjacent bowel. During RFA of central tumors, the UPJ is at risk of thermal injury resulting in stricture and UPJ obstruction.70 Some investigators think that compared to RFA, cryoablation has a higher safety profile with respect to the urothelium.71 In a review of 129 CT-guided percutaneous cryoablation cases, the radiographic ice ball involved the sinus, with 41 ice balls overlapping the renal sinus by 6 mm or more. There were no cases of collecting system injuries in this series.71 More inferiorly, thermal injury to the ureter has been reported at a rate of 1% to 2%,19,20 and tumors in the medial aspect of the lower pole are at greatest risk of injury because of their close proximity to the ureter. The risk of ureteral stricture is higher when the distance between the tumor and ureter is less than 2 cm.21 Retrograde pyeloperfusion can reduce the risk of thermal injury to the UPJ and ureter during ablation.54-56 These injuries can manifest radiologically as ureteral wall thickening, periureteral fat stranding, hydronephrosis, or urinoma. If not promptly identified and treated, acute renal failure can ensue, leading to loss of the renal unit.70 When the ablation zone involves the psoas

CHAPTER 148 THERMAL ABLATION OF RENAL CELL CARCINOMA 1107 muscle or posterior abdominal wall, nerve injuries may cause pain, paresthesia, or abdominal wall laxity.72,73 A distance of less than 5 mm between the tumor and bowel is associated with increased risk of thermal injury to the bowel; this is most commonly encountered with lower-pole and posterior tumors.48 Shortly after ablation, CT may demonstrate bowel wall thickening. In the weeks after the procedure, the bowel may adhere to the kidney, which may lead to a fistula.74 Long-term serious sequelae include stricture, obstruction, and perforation. As discussed earlier, hydrodissection, injection of gas, and probe manipulation render ablation safer when bowel loops are in close proximity to the ablation zone.

FOLLOW-UP CARE The technical success for ablation of renal tumors is determined by follow-up imaging consisting of CT or MRI studies. Detection of subtle abnormalities, which may be the only sign of residual or recurrent tumor, heavily depends on the quality of imaging studies. Renal protocol CT algorithms at most institutions consist of images that are obtained prior to contrast administration, in the corticomedullary phase of contrast enhancement, and in the excretory phase. Similarly, MRI studies are obtained before and after contrast administration. Persistent enhancing nodules in the ablation zone up to 3 months post treatment can indicate residual disease.62 Recurrent disease is suspected if the ablation zone is enlarging on serial scans or if nodular contrast enhancement that was not present on the initial postablation study is identified.62 In a multicenter study by Matin et al., 70% of all incomplete treatments were detected in the first 3 months after ablation, and over 90% were detected within 12 months of ablation.75 Based on these observations, three to four imaging studies are recommended in the first year after ablative therapy, at 1, 3, 6 (optional), and 12 months (see Fig. 148-2).75 In addition to evaluation of the ablation zone, the renal vein and inferior vena cava are evaluated for evidence of enlargement or abnormal enhancement. A search for a new primary tumor and metastatic disease is performed. The RFA zone on follow-up CT studies appears as an area of nonenhancing soft tissue surrounded by normal enhancing renal parenchyma, giving it a “bull’s-eye” appearance.62 Over time, a thin band of soft tissue density forms, which engulfs the ablated tumor and the perinephric fat that was involved in the ablation zone. This “halo sign” develops in 75% of the cases.76 On MRI studies, the RFA zone is hypointense on T2-weighted sequences compared with normal renal parenchyma and can have variable intensity on T1-weighted sequences.77,78 Subtraction of postgadolinium and noncontrast T1-weighted data may enhance detection of subtle foci of residual or recurrent disease.77 Hemorrhage can artificially increase the size of the ablation zone on the immediate postprocedure scan, but the lesion should slowly involute to preablation size on serial scans.79 After cryotherapy, the zone of ablation is typically nonenhancing on CT and MRI surveillance studies, but residual contrast enhancement has been reported despite complete tumor ablation.80-82 In a review of 32 lesions treated with laparoscopic cryoablation, Stein et al. identified persistent ablation zone enhancement in 15.6% (5/32) at 3 months; 3

of these lesions displayed enhancement at 6 months, and 1 displayed enhancement at 9 months.80 The latter underwent partial nephrectomy that did not confirm recurrent cancer.80 The ablation zone is frequently isointense on T1-weighted sequences and hypointense on T2-weighted sequences relative to the renal parenchyma. Involution of the tumor mass on surveillance studies is more prominent following cryoablation than RFA because of tissue resorption. After RFA, the lesion is replaced by scar tissue.62 Gill et al. reported that tumor size decreased by an average of 75% 3 years after ablation, and an additional 38% of cryoablated tumors were not detectable by MRI at 3 years.83 With an increasing number of patients undergoing ablation, management and outcome of tumor recurrence after ablative therapy has received some attention in recent years. For recurrent tumors, further management options include active surveillance, repeat ablation, and surgical extirpation. Given that the median annual growth rate of small renal masses (25%), it should be drained either with an intravenous cannula and a syringe or with a catheter and a one-way valve (Heimlich) or underwater drain (Fig. 152-5). Once the pneumothorax disappears and the lung has reexpanded, the drain can be removed. • If hemoptysis does occur, it is important to reassure the patient that it will stop. If there is significant bleeding, an urgent angiogram may help but in most circumstances is unnecessary. • Ensure the sample is labeled correctly and the patient’s identification is correct. The laboratory may refuse to process an unidentified sample. KEY POINTS •

Percutaneous biopsy is a rapid and efficient method of obtaining a cytologic and histologic diagnosis.

•

Diagnostic results are highest for malignant lesions.

•

Benign lesions usually require core biopsy rather than aspiration biopsy.

•

Computed tomography (CT)-guided biopsy is now the preferred method of percutaneous lung and pleural biopsy because of its higher yield. While CT-guided biopsy is favored for most mediastinal adenopathy, subcarinal nodes may be more amenable to biopsy under endobronchial ultrasound guidance.

•

Preferably, a cytopathologist should be present at the time of biopsy to indicate whether more material is required.

•

Use of a coaxial needle reduces the number of pleural punctures and decreases the risk for pneumothorax.

•

Pneumothorax resulting from lung or pleural biopsy can usually be managed by the interventional radiologist.

•

Turning the patient onto the side of the biopsy afterward will reduce the incidence of pneumothorax.

POSTPROCEDURAL AND FOLLOW-UP CARE • Always image the chest at the end of the biopsy procedure. Most significant pneumothoraces develop within 1 hour or less after the biopsy. • Ask the patient to breathe quietly, and keep the patient resting for 2 to 3 hours. A chest radiograph should then be performed to exclude pneumothorax. An outpatient may be discharged if there is no pneumothorax.14 • In cases of tension pneumothorax, an intravenous cannula (Introcan Safety [B. Braun, Bethlehem, Pa.]) can be used to reduce the pressure, and the pneumothorax can be aspirated completely.

1136 PART 2 NONVASCULAR INTERVENTIONS · SECTION 23 THE THORAX ▶ SUGGESTED READINGS Anderson JM, Murchison J, Patel D. CT-guided lung biopsy: factors influencing diagnostic yield and complication rate. Clin Radiol 2003; 58:791–7. Heck SL, Blom P, Berstad A. Accuracy and complications in computed tomography fluoroscopy-guided needle biopsies of lung masses. Eur Radiol 2006;16:1387–92.

House AJS. Biopsy techniques in the investigation of diseases of the lung, mediastinum and chest wall. Radiol Clin North Am 1979;17: 393–412.

The complete reference list is available online at www.expertconsult.com.

CHAPTER 152 PERCUTANEOUS BIOPSY OF THE LUNG, MEDIASTINUM, AND PLEURA1136.e1 REFERENCES 1. Salazar AM, Westcott JL. The role of transthoracic needle biopsy for the diagnosis and staging of lung cancer. Clin Chest Med 1993;14: 99–110. 2. House AJS, Thomson KR. Evaluation of a new transthoracic needle for biopsy of benign and malignant lung lesions. AJR Am J Roentgenol 1979;129:215–20. 3. Heck SL, Blom P, Berstad A. Accuracy and complications in computed tomography fluoroscopy-guided needle biopsies of lung masses. Eur Radiol 2006;16:1387–92. 4. Laurent F, Latrabe V, Vergier B, et al. CT-guided transthoracic needle biopsy of pulmonary nodules smaller than 20 mm: results with an automated 20-gauge coaxial cutting needle. Clin Radiol 2000;55: 281–7. 5. Yeow KM, Tsay PK, Cheung YC, et al. Factors affecting diagnostic accuracy of CT-guided coaxial cutting needle lung biopsy: retrospective analysis of 631 procedures. J Vasc Interv Radiol 2003;14:581–8. 6. Geraghty PR, Kee ST, McFarlane G, et al. CT-guided transthoracic needle aspiration biopsy of pulmonary nodules: needle size and pneumothorax rate. Radiology 2003;229:475–81. 7. Kinoshita F, Kato T, Sugiura K, et al. CT-guided transthoracic needle biopsy using a puncture site-down positioning technique. AJR Am J Roentgenol 2006;187:926–32. 8. Manhire A, Charig M, Clelland C, et al. Guidelines for radiologically guided lung biopsy. Thorax 2003;58:920–36. 9. Adams K, Shah PL, Edmonds L, Lim E. Test performance of endobronchial ultrasound and transbronchial needle aspiration biopsy for mediastinal staging in patients with lung cancer: systematic review and meta-analysis. Thorax 2009;64:757–62.

10. Kirchner J, Kickuth R, Laufer U, et al. CT fluoroscopy-assisted puncture of thoracic and abdominal masses: a randomized trial. Clin Radiol 2002;57:188–92. 11. Anderson JM, Murchison J, Patel D. CT-guided lung biopsy: factors influencing diagnostic yield and complication rate. Clin Radiol 2003; 58:791–7. 12. Milner LB, Ryan K, Gullo J. Fatal intrathoracic hemorrhage after percutaneous aspiration lung biopsy. AJR Am J Roentgenol 1979;132: 280–1. 13. Charig MJ, Phillips AJ. CT-guided cutting needle biopsy of lung lesions—safety and efficacy of an out-patient service. Clin Radiol 2000;55:964–9. 14. Yamagami T, Kato T, Iida S, et al. Efficacy of manual aspiration immediately after complicated pneumothorax in CT-guided lung biopsy. J Vasc Interv Radiol 2005;16:477–83. 15. Barnes D, Souza C, Entwisle J. The role of percutaneous biopsy in the evolving diagnosis and treatment of lung cancer. Clin Radiol 2010; 65:951–2. 16. Aviram G, Greif J, Man Y, Schwartz Y et al. Diagnosis of intrathoracic lesions: are sequential fine-needle aspiration (FNA) and core biopsy (CNB) combined better than either investigation alone? Clin Radiol 2007;62:221–6. 17. Yeow KM, See LC, Lui KW, et al. Risk factors for pneumothorax and bleeding after CT-guided percutaneous coaxial cutting needle biopsy of lung lesions. J Vasc Interv Radiol 2001;12:1305–12. 18. Cox JE, Chiles C, McManus CM, et al. Transthoracic needle aspiration biopsy: variables that affect risk of pneumothorax. Radiology 1999; 212:165–8.

CHAPTER

153

Treatment of Effusions and Abscesses

Robert J. Lewandowski, Sudhen B. Desai, and Albert A. Nemcek, Jr.

Percutaneous management of thoracic and abdominal fluid collections, including pleural effusions and abscesses, has become standard of care in many clinical scenarios.1 This shift from operative management has been facilitated by advances in quality and rapidity of cross-sectional imaging and interventional radiology technology. Percutaneous management of effusions and abscesses is cost-effective and decreases both morbidity and mortality compared with alternative treatments.2-8 To be truly effective, the interventional radiologist must participate in the clinical care of the patient and coordinate care with referring medical and surgical colleagues.

INDICATIONS Most patients undergoing percutaneous management of effusions or abscesses are symptomatic and have had recent imaging studies performed. In fact, these symptoms have typically led to the imaging examination (i.e., most effusions or abscesses are not found incidentally). If the patient is asymptomatic, percutaneous drainage may not be needed as an adjunct to antibiotic or other medical therapy, although diagnostic sampling may be necessary. The patient’s symptoms help categorize the urgency of draining an effusion or abscess. Patients with symptoms of pain, low-grade fever, and a mildly elevated white blood cell (WBC) count do not require urgent attention. However, patients with hypotension, tachycardia, high-grade fever or fever spikes, or a markedly elevated WBC count require decompression as soon as possible. When percutaneous management of an effusion or abscess is clearly indicated, a decision regarding fluid aspiration versus catheter placement should be made. In general, drainage catheters are placed unless a fluid collection is small, not walled off, or known to be sterile. If aspirated fluid appears purulent, a drainage catheter is generally placed.

CONTRAINDICATIONS There are few contraindications to percutaneous management of effusions or abscesses. Two are lack of a safe route for catheter insertion and uncorrectable coagulopathy, issues that often become evident when reviewing the patient’s imaging studies and laboratory results. Any apparent contraindication should be discussed with the referring service to ascertain the risks versus potential benefits. To allow percutaneous therapy, the abnormal fluid collection must be clearly visualized on imaging, and there must be an acceptable path from a skin entry site to the abnormality. Although it is often necessary and acceptable to place a needle or catheter through normal organ

parenchyma when managing an intraparenchymal abscess, it is often unacceptable to traverse an uninvolved organ or normal bowel. However, some have argued for the benefits of transgastric drainage, and normal hepatic parenchyma may be transgressed if it is the only route to the fluid collection.9 It is sometimes possible to create a more direct and safer access route to the target fluid by injecting saline, lidocaine, or CO2 into the intervening tissues.10,11 This may displace potential obstacles out of the way, but it also may make the target less conspicuous on imaging. Manual compression of tissues with the ultrasound probe may help displace intervening bowel loops, and alternative patient positioning may help clear a potential passageway. It should be noted that a referring surgeon may accept traversal of diseased bowel if an operation is impending. All patients being considered for percutaneous intervention should have their coagulation profile evaluated, including platelet counts, prothrombin time, international normalized ratio (INR), and partial thromboplastin time. Although it is difficult to give strict coagulation parameters, the platelet count should be greater than 50,000 and the INR should be 1.5 or lower.12 All abnormalities are best corrected in conjunction with the referring service. Some patients will need platelet transfusion, fresh frozen plasma, vitamin K, protamine, or other strategies to make the procedure safer.13 The medication administration record should also be reviewed. Specifically, drugs such as aspirin, heparin, subcutaneous heparin, and clopidogrel (Plavix) should be sought because they may increase the risk for a hemorrhagic complication without significantly altering the patient’s coagulation profile. Ideally these medications are stopped for a defined period (aspirin and Plavix, 1 week; subcutaneous heparin, 12 hours; intravenous heparin, 2 hours) before percutaneous intervention,14 but each case must be considered individually.

EQUIPMENT Image-guided procedures are facilitated by ultrasound, fluoroscopy, computed tomography (CT), or a combination of these modalities. Factors that influence which modality is most appropriate include target location, target size, and operator preference. At our institution, percutaneous drainage procedures are typically performed with a combination of ultrasound and fluoroscopy. Ultrasound is used to place a needle into the effusion or abscess, whereas fluoroscopy is used to visualize placement of a drainage catheter. Ultrasound is frequently our preferred modality to access an abnormal fluid collection because it is fast and readily available, produces no ionizing radiation, and allows real-time visualization of needle placement. However, it is operator dependent, and smaller, deeper fluid collections may not be 1137

1138 PART 2 NONVASCULAR INTERVENTIONS · SECTION 23 THE THORAX readily seen. Patient body habitus and intervening bowel gas are further detriments to optimal visualization with ultrasonography. Fluoroscopy may be the sole imaging modality if a few criteria are met: the effusion or abscess must be diagnosed on previous cross-sectional imaging, and there must be a previously determined safe passage to this abnormal fluid collection. A recognizable collection of air (air/fluid level) is helpful in ensuring proper needle placement. In our practice, CT is usually reserved for lesions not visible sonographically or fluoroscopically. Abscesses that are deeply located, retroperitoneal, or small in size frequently require CT localization. Once an imaging modality is selected, the interventionalist must decide on an access needle, guidewire, dilators, and drainage catheter. The preferred access needle is at least a few centimeters longer than needed to reach the fluid collection and allows passage of a 0.035- to 0.038-inch guidewire. The guidewire enables progressive dilation and subsequent catheter placement. Dilators should be chosen to at least the size of the drainage catheter. Some advocate dilating to 2 French sizes larger than the catheter. Various types of drainage catheters are available. A locking 10F pigtail catheter is a reasonable standard catheter. If the fluid collection is small, an 8F drain can be used, and if the aspirated fluid appears viscous, a 12F or larger catheter may be placed. Large-bore tunneled catheters are being placed with greater frequency for longer-term home drainage.

TECHNIQUE Anatomy and Approach The most crucial aspect of percutaneous drainage is determining an access route. The safest and shortest distance between the skin access site and target fluid collection should be sought. Review of available imaging studies and knowledge of anatomic landmarks and the expected location of blood vessels, nerves, and other tissues facilitate avoidance of these structures. Once chosen, the access site is marked, prepared, and draped in sterile fashion. The skin access site is then infiltrated with a local anesthetic, a small incision is made with a scalpel, and the tissues are locally dissected to aid subsequent catheter placement. At this point, either the Seldinger or trocar technique is used to place the drainage catheter. With the Seldinger technique, an access needle is placed into the fluid collection. A wire is passed through the access needle and coiled in the collection. Serial dilation over the wire precedes placement of the catheter over the wire. The catheter is then advanced without a stylet (a metal or plastic stiffener is used instead). The wire is removed, and the catheter is secured in place. The trocar technique does not make use of an access wire. Instead, a sharp stylet passed through the catheter aids its advancement. The tandem trocar approach makes use of a reference needle placed into the fluid collection under imaging guidance. The drainage catheter is then advanced via the trocar technique into the cavity adjacent to the reference needle in the same direction as the reference needle until the fluid collection is entered. The direct trocar approach does not make use of a reference needle. The drainage catheter is simply advanced into the fluid collection based on previous imaging studies. This technique is generally reserved for large superficial collections.

Technical Aspects Thoracic Pleural Effusions Pleural effusions have many causes and can usually be aspirated (thoracentesis) for symptom relief. Malignant pleural effusions tend to recur rapidly, however, and some special strategies are used to deal with these. A reasonable management plan is to place a tunneled drainage catheter (e.g., Denver Pleurx catheter [Denver Biomaterials, Golden, Colo.]) and monitor the patient on an outpatient basis. Autopleurodesis has been shown to occur in 40% to 50% of these patients.15 If pleurodesis does not occur, a sclerosing agent may be used to scar the parietal and visceral pleural layers together. Such agents include talc, bleomycin, and tetracycline/doxycycline.16,17 Empyema Infected pleural fluid collections typically result from pneumonia or are posttraumatic in nature (Fig. 153-1). They can often be successfully managed percutaneously with drains. Infusion of thrombolytic agents into these drains has been used to eradicate the fibrinous components of the inflammatory process.18,19 Lung Abscess Lung abscesses are typically secondary to aspiration, hematogenous spread of infection, or spread from an adjacent inflammatory process. More than 80% of lung abscesses respond to the traditional therapy of antibiotics and bronchial lavage. In patients who fail such therapy, percutaneous management may be an alternative to thoracotomy.7,20 A risk of placing a percutaneous drain in the lung is creation of a bronchopleural fistula. Minimizing drainage catheter size and avoiding traversal of normal lung help minimize this risk. Pericardial Effusions Interventional radiologists may be asked to drain pericardial fluid. In our experience, this is best approached with ultrasound guidance. Percutaneous creation of a pericardial window with an angioplasty balloon is a safe therapeutic option, particularly in patients with symptomatic malignant effusions.21 Abdominal Hepatic Pyogenic hepatic abscesses are most frequently sequelae of previous hepatic or biliary surgery, but they may also result from trauma, an infectious process elsewhere in the abdomen (e.g., diverticulitis), or an existing hepatic malignancy (Fig. 153-2). Percutaneous management of these abscesses is quite effective for both pyogenic and amebic hepatic abscesses.22,23 A subcostal approach is preferred because it is less painful and there is less chance of crossing the pleural space and causing empyema. The intercostal approach is reserved for instances where a subcostal approach is not possible. It should be noted that although drainage of hydatid (echinococcal) hepatic cysts was once considered a contraindication because of the risk of sepsis, the literature now supports percutaneous management. Hypertonic saline can be injected into the cystic cavities as a scolicidal agent.24

CHAPTER 153 TREATMENT OF EFFUSIONS AND ABSCESSES 1139

A

B

C

D

E

F

FIGURE 153-1. A-C, Loculated left pleural fluid collection complicating a case of pneumonia. D-F, Correlative images after drain placement via interventional radiology and management with thrombolytic therapy; note marked improvement in left fluid collection.

B

A

C

D

E

FIGURE 153-2. A-B, Hepatic abscess in segments 4 and 8 post liver transplantation complicated by hepatic artery occlusion. C, Fluoroscopic spot image of percutaneously placed drain via interventional radiology. D, Tube check of that drain demonstrates small, irregular, persistent cavity without a fistula to biliary tree. E, Follow-up computed tomography image demonstrating resolution of abscess and removal of drain.

Renal Renal abscesses result from hematogenous spread of infection or from pyelonephritis (Fig. 153-3). The infectious process may extend into the perinephric space. Percutaneous management is typically reserved for large parenchymal abscesses, those with a perinephric component, or small intrarenal abscesses that do not respond to appropriate antibiotic therapy. These fluid collections often contain an

inflammatory component that may not respond to percutaneous drainage. A retroperitoneal approach through the avascular plane of Brödel is recommended.25,26 Splenic Splenic abscesses result from hematogenous spread of infection, direct extension from an adjacent infectious process, or previous trauma (Figs. 153-4 and 153-5).

1140 PART 2 NONVASCULAR INTERVENTIONS · SECTION 23 THE THORAX

A

B

C

D

FIGURE 153-3. A, Computed tomography (CT) scan demonstrating left renal abscess complicating a case of pyelonephritis that failed to resolve with intravenous antibiotic therapy. B, Fluoroscopic image after percutaneous intrarenal drain placement and manual injection of contrast material; note irregular cavity without a fistula to renal collecting system. C-D, CT images after drain removal demonstrate resolution of left renal abscess.

B

A

C

D

FIGURE 153-4. Computed tomography (CT) (A) and ultrasound (B) images demonstrating splenic abscess in human immunodeficiency virus–positive patient who complained of fevers and chills on return from trip to Thailand. C, Fluoroscopic image after interventional radiology placement of percutaneous drain. D, CT image after drain removal demonstrating resolution of abscess.

CHAPTER 153 TREATMENT OF EFFUSIONS AND ABSCESSES 1141

B

A

C

FIGURE 153-5. A, Computed tomography (CT) image demonstrating splenic abscess in patient with mycotic pseudoaneurysm of femoral artery. B, CT image during needle placement (arrow) into splenic abscess. C, CT image taken after completion of abscess management demonstrates resolution of infective process.

C

B

A

F

D

E

FIGURE 153-6. A, Computed tomography (CT) image depicting large fluid collection in lesser sac complicating a case of pancreatitis. B, Ultrasound image during drain placement. C, Fluoroscopic image taken after drain placement via interventional radiology. D-E, CT images after drain placement and removal demonstrate resolution of fluid collection.

Percutaneous management of these abscesses has historically been avoided because of the risk for significant hemorrhage, but percutaneous drainage techniques have proved to be safe and effective.27 To minimize bleeding risk, a small (8F or 10F) catheter is typically used, and the catheter is placed such that it traverses the least possible amount of normal splenic parenchyma. Transpleural catheter placement is best avoided to prevent direct spread of infection to the pleural space. Pancreatic Management of pancreatic abscesses can be medical, operative, endoscopic, or percutaneous (Fig. 153-6). The role of the interventional radiologist in percutaneous management is evolving. The presence of tissue fragments and saponified fat often makes percutaneous drainage difficult,

and surgical débridement may be needed. Fluid aspiration or sampling is often useful in differentiating sterile from infected fluid collections and helps refine antibiotic therapy. Percutaneous abscess drainage may result in cure or can make future operative management less technically demanding.28,29 Enteric Enteric abscesses typically complicate disease processes such as appendicitis, diverticulitis, or inflammatory bowel disease (Fig. 153-7). The goal of percutaneous abscess drainage in these cases is to facilitate an elective one-stage operation at a time remote from the initial evaluation. Operating on these patients often initially results in multistage procedures and can result in significant morbidity. Drainage catheters may be left in place for several weeks

1142 PART 2 NONVASCULAR INTERVENTIONS · SECTION 23 THE THORAX

B

A

C

D

FIGURE 153-7. A, Computed tomography image from woman with fever 5 days after right hemicolectomy for cecal volvulus; note large abscess in right lower quadrant adjacent to surgical anastomosis. B, Fluoroscopic image taken after percutaneous drain placement via interventional radiology. C-D, Fluoroscopic images taken during routine tube check. A fistula to colon, representing a hole at surgical anastomosis, was detected. Patient had continued to have high drain output for several months.

A

B

FIGURE 153-8. A, Computed tomography image of woman with Crohn disease; note abscess in pelvic cul-de-sac posterior to vagina. B, Fluoroscopic image taken after drain placement via interventional radiology; drain was placed from transvaginal approach.

because it is not uncommon for an abscess-fistula complex to be present.30-33 Pelvic Pelvic abscesses are often challenging to access because of the surrounding bony pelvis (Fig. 153-8). An anterior transperitoneal approach cannot be used in many circumstances because of overlying structures, most notably bowel and bladder. Alternative routes include transgluteal, transrectal, and transvaginal approaches.34-36 Although the transgluteal route has been used for deep pelvic abscesses, this method is frequently quite painful and not well tolerated by patients. The catheter should be directed near the sacrum if possible to avoid the sciatic nerve as it exits through the greater sciatic foramen. Subphrenic Subphrenic abscesses are most often postoperative complications (Fig. 153-9). They may be technically challenging to access because of their location under the diaphragm, bordered by the rib cage. Although an extrapleural route is advised because of the risk of pleural contamination, an intercostal approach may be necessary.37

OUTCOMES The efficacy of percutaneous management of abscesses and effusions has been demonstrated in multiple series.2-8

However, it should be stated that success of percutaneous drainage depends on more than just drainage technique and patient management. Comorbid conditions and complexity of fluid collections can also affect success rates. The Society of Interventional Radiology (SIR) has published a quality initiative for percutaneous drainage of abscesses.38 Successful diagnostic fluid aspiration is defined as aspiration of sufficient material for diagnosis. Per these guidelines, success should be achieved 95% of the time. Regarding drainage of infected fluid collections, results are defined in terms of success (complete or partial success) or failure (including recurrence). Complete resolution of infection implies that no further intervention is needed, whereas partial success means either that drainage of the abscess was adequate and surgery was subsequently performed to repair an underlying problem, or that the drainage was a temporizing measure before surgery. By SIR guidelines, curative drainage should occur in more than 80% of cases, whereas partial success occurs in 5% to 10%. The SIR set a threshold of 85% for successful drainage (curative and partial success). Failure and recurrence both occur in 5% to 10% of patients.

COMPLICATIONS The SIR Standards of Practice Committee has published criteria for complications by outcome. Minor complications are defined as those that require no therapy (A) and those

CHAPTER 153 TREATMENT OF EFFUSIONS AND ABSCESSES 1143

A

C

B

FIGURE 153-9. A, Computed tomography (CT) image from young man complaining of fever and left upper quadrant pain after resection of a desmoid tumor; large subphrenic abscess is present. B, Fluoroscopic image taken after percutaneous drain placement via interventional radiology. C, CT image after drain removal demonstrates resolution of subphrenic abscess.

that require nominal therapy or admission for overnight observation (B). Major complications include those that require major therapy and hospitalization for less than 48 hours (C) or those that require major therapy, an increased level of patient care, and prolonged hospitalization for more than 48 hours (D). Permanent adverse sequelae (E) and death (F) are other classifications of major complications. According to the SIR, the overall procedure threshold in adults for all major complications resulting from percutaneous drainage of abscesses and fluid is 10%.38 Major complications of percutaneous procedures may include bleeding, infection, and inadvertent injury to bowel or viscera. Hemorrhage is often avoided by correcting any coagulopathy before the procedure. Hemorrhage is suspected in the setting of increasing postprocedural pain accompanied by tachycardia or hypotension. Occasionally blood is seen draining from the catheter. If hemorrhage is suspected, the drain should be left in place and the coagulation profile repeated. A CT scan should be obtained urgently because this study can verify hemorrhage and assess its cause and extent. If clinically stable, the patient can be observed closely and serial complete blood cell counts obtained. If the patient appears unstable, angiography with potential embolization of a bleeding vessel could be performed (if CT demonstrates bleeding from an intraabdominal visceral source). Infection or sepsis is best avoided by ensuring the patient is on an appropriate antibiotic regimen before the procedure. If the patient is not taking preprocedure antibiotics, broad-spectrum coverage should be provided before catheter placement. Sepsis risk can also be decreased by minimizing catheter manipulation during the procedure.

Irrigation of the abscess cavity should be gentle. If contrast material is to be injected, a minimal amount should be used. Inadvertent injury to bowel or viscera is best avoided by optimizing the access route and directly visualizing catheter advancement under real-time imaging. If a catheter mistakenly perforates a solid organ, bleeding may result. This should be managed as described earlier. If a catheter is mistakenly placed into a loop of bowel, the catheter should remain in place until a mature tract develops. The tract will close spontaneously. If the bladder is mistakenly perforated by a drainage catheter, the catheter should be removed and a Foley catheter placed in the bladder. The Foley catheter should remain in place for approximately 1 week to allow the perforation to heal.

POSTPROCEDURAL AND FOLLOW-UP CARE After insertion of the catheter and aspiration of fluid, the catheter must be secured to the patient’s skin to prevent inadvertent removal or displacement. There are many methods of catheter fixation involving commercially available fixation devices or suturing techniques. A collection bag is connected to the catheter and placed for dependent gravity drainage. Thoracic collections or those with particulate matter or highly viscous fluid may be placed under continuous low wall suction. Drain output should be monitored and recorded at routine intervals. Abscess drains are often irrigated with 5 to 10 mL of normal saline every 8 hours to facilitate tube patency. The catheter and external fixation device must be inspected daily on ward rounds. Signs of catheter

1144 PART 2 NONVASCULAR INTERVENTIONS · SECTION 23 THE THORAX dislodgment or malfunction should be sought. The fluid draining into the collecting bag should also be evaluated; changes in character or quality provide clinical insight. For example, catheter output that does not decrease over time or increases after a few days of drainage probably indicates a fistulous communication. If a fistula is suspected, a catheter sinogram is performed. Otherwise, there is no definitive rule when it comes to repeat imaging after percutaneous drainage. Certainly, if the patient does not respond clinically to percutaneous drainage, cross-sectional imaging may divulge new or inadequately drained fluid collections. There is some controversy in the published literature regarding the best methods of draining complex fluid collections and abscesses. Specifically, the role of adjunctive fibrinolytic therapy is debated. These agents have been shown to decrease viscosity of the fluid collection, making its removal theoretically easier.39 In vitro studies with urokinase have demonstrated promising results in decreasing the draining fluid’s viscosity.40 However, a recent New England Journal of Medicine article reported no improvement in the treatment of pleural infections treated with streptokinase in nonselected patients.41 The endpoint of percutaneous drainage is multifactorial. It is crucial to discuss catheter plans with the referring medical or surgical service before catheter removal to avoid the necessity of replacing the catheter. Major criteria before catheter removal are clinical improvement (fever, pain, appetite), normalization of laboratory parameters (WBC count), decreasing catheter output (≤10 mL/day), resolution of cavity size (either repeat imaging or catheter sonogram), and absence of a fistulous communication. Some advocate exchanging the pigtail catheter for a straight

catheter and gradually withdrawing the drain over a period of a few days, a surgical approach to drain management. This theoretically prevents reaccumulation of fluid secondary to premature closure of the drainage tract.42 However, because catheter removal is typically delayed until drainage is minimal and a mature tract has formed, this process is no longer routinely advocated.4 After catheter removal, a clean dressing should be applied. It is not unusual to observe a small amount of drainage from the catheter insertion site after catheter removal. KEY POINTS •

Percutaneous management of abscesses and effusions is safe and effective.

•

The role of percutaneous management is evolving.

•

Percutaneous management of disease processes should be performed in conjunction with referring physicians.

▶ SUGGESTED READINGS Mueller PR, vanSonnenberg E, Ferrucci JT Jr. Percutaneous drainage of 250 abdominal abscesses and fluid collections. Part II: current procedural concepts. Radiology 1984;151:343–7. vanSonnenberg E, Mueller PR, Ferrucci JT Jr. Percutaneous drainage of 250 abdominal abscesses and fluid collections. Part I: results, failures, and complications. Radiology 1984;151:337–41.

The complete reference list is available online at www.expertconsult.com.

CHAPTER 153 TREATMENT OF EFFUSIONS AND ABSCESSES1144.e1 REFERENCES 1. Fulcher AS, Turner MA. Percutaneous drainage of enteric-related abscesses. Gastroenterologist 1996;4:276–85. 2. Gerzof SG, Robbins AH, Johnson WC, et al. Percutaneous catheter drainage of abdominal abscesses: a five-year experience. N Engl J Med 1981;305:653–7. 3. vanSonnenberg E, Mueller PR, Ferrucci JT Jr. Percutaneous drainage of 250 abdominal abscesses and fluid collections. Part I: results, failures, and complications. Radiology 1984;151:337–41. 4. Mueller PR, vanSonnenberg E, Ferrucci JT Jr. Percutaneous drainage of 250 abdominal abscesses and fluid collections. Part II: current procedural concepts. Radiology 1984;151:343–7. 5. Lambiase RE, Deyoe L, Cronan JJ, Dorfman GS. Percutaneous drainage of 335 consecutive abscesses: results of primary drainage with 1-year follow-up. Radiology 1992;184:167–79. 6. O’Moore PV, Mueller PR, Simeone JF, et al. Sonographic guidance in diagnostic and therapeutic interventions in the pleural space. AJR Am J Roentgenol 1987;149:1–5. 7. vanSonnenberg E, D’Agistino HB, Casola G, et al. Lung abscess: CT-guided drainage. Radiology 1991;178:347–51. 8. Deveney CW, Lurie K, Deveney KE. Improved treatment of intraabdominal abscess. A result of improved localization, drainage, and patient care, not technique. Arch Surg 1988;123:1126–30. 9. Maher MM, Gervals DA, Kalra MK, et al. The inaccessible or undrainable abscess: how to drain it. Radiographics 2004;24:717–35. 10. Langen HJ, Jochims M, Gunther RW. Artificial displacement of kidneys, spleen, and colon by injection of physiologic saline and CO2 as an aid to percutaneous procedures: experimental results. J Vasc Interv Radiol 1995;6:411–6. 11. Shah H, Harris VJ. Saline injection into the perirectal space to assist transgluteal drainage of deep pelvic abscesses. J Vasc Interv Radiol 1997;8:119–21. 12. Payne CS. A primer on patient management problems in interventional radiology. AJR Am J Roentgenol 1998;170:1169–76. 13. Practice parameter for the use of fresh-frozen plasma, cryoprecipitate, and platelets. Fresh-Frozen Plasma, Cryoprecipitate, and Platelets Administration Practice Guidelines Development Task Force of the College of American Pathologists. JAMA 1994;271:777–81. 14. Mehta RP, Johnson MS. Update on anticoagulant medications for the interventional radiologist. J Vasc Interv Radiol 2006;17:597–612. 15. Pollak JS, Burdge CM, Rosenblatt M, et al. Treatment of malignant pleural effusions with tunneled long-term drainage catheters. J Vasc Interv Radiol 2001;12:201–8. 16. Tan C, Sedrakyan A, Browne J, et al. The evidence on the effectiveness of management for malignant pleural effusion: a systematic review. Eur J Cardiothorac Surg 2006;29:829–38. 17. Kilic D, Akay H, Kavukçu S, et al. Management of recurrent malignant pleural effusion with chemical pleurodesis. Surg Today 2005;35: 634–8. 18. Moulton JS, Benkert RE, Weisiger KH, Chambers JA. Treatment of complicated pleural fluid collections with image-guided drainage and intracavitary urokinase. Chest 1995;108:1252–9. 19. Wells RG, Havens PL. Intrapleural fibrinolysis for parapneumonic effusion and empyema in children. Radiology 2003;228:370–8. 20. Parker LA, Melton JW, Delany DJ, Ynakaskas BC. Percutaneous small bore catheter drainage in the management of lung abscesses. Chest 1987;92:213–8. 21. del Barrio LG, Morales JH, Delgado C, et al. Percutaneous balloon pericardial window for patients with symptomatic pericardial effusion. Cardiovasc Intervent Radiol 2002;25:360–4.

22. Benazzouz M, Afifi R, Ibrahimi A, et al. [Liver abscess: diagnosis and treatment. Study of a series of 22 cases.] Ann Gastroenterol Hepatol (Paris) 1996;31:333–6. 23. vanSonnenberg E, Mueller PR, Schiffman HR, et al. Intrahepatic amebic abscesses: indications for and results of percutaneous catheter drainage. Radiology 1985;156:631–5. 24. Acunas B, Rozanez I, Celik L, et al. Purely cystic hydatid disease of the liver: treatment with percutaneous aspiration and injection of hypertonic saline. Radiology 1992;182:541–3. 25. Helenon O, Comud F, Di Stefano D, et al. [Percutaneous treatment of abscess of the kidney and retroperitoneum.] J Radiol 1989;70:541–8. 26. Deyoe LA, Cronan JJ, Lambiase RE, Dorfman GS. Percutaneous drainage of renal and perirenal abscesses: results in 30 patients. AJR Am J Roentgenol 1990;155:81–3. 27. Quinn SF, vanSonnenberg E, Casola G, et al. Interventional radiology in the spleen. Radiology 1986;161:289–91. 28. Lee MJ, Ruttner DW, Legemate DA, et al. Acute complicated pancreatitis: redefining the role of interventional radiology. Radiology 1992; 183:171–4. 29. Lee MJ, Wittich GR, Mueller PR. Percutaneous intervention in acute pancreatitis. Radiographics 1998;18:711–24; discussion 728. 30. Casola G, vanSonnenberg E, Neff CC, et al. Abscesses in Crohn disease: percutaneous drainage. Radiology 1987;163:19–22. 31. Mueller PR, Saini S, Wittenburg J, et al. Sigmoid diverticular abscesses: percutaneous drainage as an adjunct to surgical resection in 24 cases. Radiology 1987;164:321–5. 32. Neff CC, vanSonnenberg E, Casola G, et al. Diverticular abscesses: percutaneous drainage. Radiology 1987;163:15–8. 33. vanSonnenberg E, Wittich GR, Casola G, et al. Periappendiceal abscesses: percutaneous drainage. Radiology 1987;163:23–6. 34. Gazelle GS, Haaga JR, Stellato TA, et al. Pelvic abscesses: CT-guided transrectal drainage. Radiology 1991;181:49–51. 35. vanSonnenberg E, D’Agistino HB, Casola G, et al. US-guided transvaginal drainage of pelvic abscesses and fluid collections. Radiology 1991;181:53–6. 36. Bennett JD, Kazak RI, Taylor BM, Jory TA. Deep pelvic abscesses: transrectal drainage with radiologic guidance. Radiology 1992;185: 825–8. 37. Eisenberg PJ, Lee MJ, Bolund GW, Mueller PR. Percutaneous drainage of a subphrenic abscess with gastric fistula. AJR Am J Roentgenol 1994;162:1233–7. 38. Bakal CW, Sacks D, Burke DW, et al. Quality improvement guidelines for adult percutaneous abscess and fluid drainage. J Vasc Interv Radiol 2003;14:S223–5. 39. Tillett WS, Sherry S, Read CT. The use of streptokinase-streptodornase in the treatment of chronic empyema; with an interpretive discussion of enzymatic actions in the field of intrathoracic diseases. J Thorac Surg 1951;21:325–41. 40. Park JK, Kraus FC, Haaga JR. Fluid flow during percutaneous drainage procedures: an in vitro study of the effects of fluid viscosity, catheter size, and adjunctive urokinase. AJR Am J Roentgenol 1993;160: 165–9. 41. Maskell NA, Davies CW, Nunn AJ, et al. U.K. Controlled trial of intrapleural streptokinase for pleural infection. N Engl J Med 2005;352: 865–74. 42. vanSonnenberg E, Ferrucci JT Jr, Mueller PR, et al. Percutaneous drainage of abscesses and fluid collections: technique, results, and applications. Radiology 1982;142:1–10.

CHAPTER

154

Lung Ablation

Jeffrey P. Guenette and Damian E. Dupuy Lung cancer is the most common primary cancer, with over 220,000 new cases in 2010, and is the most common cause of cancer death, with nearly 160,000 deaths per year.1,2 Surgery is the mainstay of treatment for primary lung cancer, but it is estimated that over 15% of all patients and 30% of patients over age 75 with stage I or II non–small cell lung cancer (NSCLC) are medically inoperable.3 Moreover, local therapy with external beam radiation has shown a best 2-year overall survival of only 51%,4 with newer stereotactic body radiotherapy techniques (i.e., stereotactic body radiation therapy [SBRT]) showing 3-year survival rates of 42% to 60% for early-stage, medically inoperable NSCLC.5,6 Pulmonary metastatic disease is generally an indicator of wide disease dissemination requiring systemic therapy, but when a finite number of metastatic deposits exist in the lung, resection may improve prognosis for certain pathologies, including primary sarcoma, renal cell carcinoma, colorectal carcinoma, and breast carcinoma.7-12 In the case of colorectal carcinoma with isolated pulmonary metastases, 5-year survival rates can exceed 50%.12 Again, as with primary lung cancer, pulmonary metastases are often inoperable owing to patients’ poor cardiopulmonary function, advanced age, and other medical comorbidities. Image-guided thermal ablation is an increasingly studied and used treatment for both cure and palliation of primary and secondary lung cancer. The technique has been used for a little over a decade and has been shown to be safe and effective in treating a variety of solid tumors including many in the lung, liver, kidney, breast, bone, and adrenal gland. It may be used in conjunction with radiotherapy and systemic chemotherapy and is relatively low cost. Thermal ablation modalities include radiofrequency, microwave, laser, and cryotherapy. Randomized controlled clinical trials have not yet been conducted to establish efficacy of thermal ablation alone, in combination with, or relative to other established treatments of lung neoplasms. However, an increasing number of single and multicenter cohort studies have consistently established the safety and suggest efficacy for medically inoperable disease. The remainder of this chapter highlights the basic biophysics underlying thermal ablation modalities, specifically within the unique anatomy and physiology of the lung, the understood safety and efficacy of thermal lung ablation given our experience to date, newly introduced advances in thermal lung ablation techniques, and the current role of thermal ablation in the treatment of lung cancer.

BASIC BIOPHYSICS OF LUNG ABLATION Electromagnetic (EM) energy takes form as oscillating perpendicular waves of electrical and magnetic fields traveling at the speed of light. EM wave frequencies range from 10 Hz (waves per second) to 1024 Hz, with radio waves

having frequencies as low as 104 Hz, followed by microwaves, infrared waves, visible light, ultraviolet waves, x-rays, and gamma rays in increasing order of wave frequency.13,14 Three of the four thermal ablation modalities utilize waves along this spectrum.

Radiofrequency Ablation Radiofrequency ablation (RFA) is the most widely used ablation modality for the treatment of solid tumors; its safety has been firmly established in solid malignancies of the lung (Fig. 154-1 and Table 154-1), liver, bone, breast, kidney, and adrenals.15 With this technique, an active RF electrode (applicator) is placed in the tumor under imaging guidance, and a grounding pad (reference electrode) is applied to the chest wall opposite the applicator or on the thigh. Most clinical RFA electrodes emit radio waves in the range of 375 to 500 kHz, generating electrical field lines between the applicator and reference electrode that oscillate with the alternating current. These fields result in collision of electrons with molecules adjacent to the applicator, generating frictional heat.16 Temperatures exceeding 60°C, regardless of the heating source, induce cell cytotoxicity via thermolabile protein denaturation,17 and this temperature is generally considered the baseline for inducing immediate coagulative necrosis. Care is taken to keep the electrode tip temperature below 100°C to avoid charring and vaporization, which occur at 110°C, because these processes increase electrical impedance and inhibit heat dissipation to surrounding tissue.18 The goal is to ablate an appropriate circumferential margin: 95% of microscopic neoplastic extension is within a margin of 8 mm from the tumor border for adenocarcinoma and 6 mm for squamous cell carcinoma.19 Three RFA systems are currently commercially available in the United States. The LeVeen system (Boston Scientific/ RadioTherapeutics, Watertown, Mass.]) consists of an array of electrode tines that are deployed through a 14- to 17-gauge needle and curve backward toward the device handle. The device measures impedance. The RITA system (RITA Medical Systems Inc., Mountain View, Calif.) is also composed of an array of electrode tines that are deployed through a 14- to 17-gauge needle, but these tines course forward and laterally. The RITA system measures temperature through multiple peripheral thermocouples and includes perfusion electrodes that infuse saline into the tissue to enhance energy distribution. The Cool-tip system (Covidien, Boulder, Colo.) consists of either a single electrode or a triple “cluster” of three electrodes spaced 5 mm apart. The Cool-tip electrodes are perfused with cold saline or water to minimize charring and include a thermocouple at the tip for temperature measurement. There have been no head-to-head comparisons of safety or efficacy between these devices in the lung. 1145

1146 PART 2 NONVASCULAR INTERVENTIONS · SECTION 23 THE THORAX

B

A

D

C

E

FIGURE 154-1. Biopsy-proven right upper lobe non–small cell carcinoma in a 64-year-old man who was referred for lung ablation. A, Axial positron emission tomography (PET)/computed tomography (CT) image shows focally intense metabolic activity (arrow) in right upper lobe. B, Axial nonenhanced CT image shows initial lesion (arrow). C, Axial nonenhanced CT image shows new tumor growth (arrow) following initial ablation. D, Prone axial CT image obtained during repeat radiofrequency ablation shows electrode (arrow) appropriately placed in mass. E, Axial contrast-enhanced 6-month follow-up CT scan shows residual thermal scar (arrow) in right upper lobe, without evidence of growth or enhancement.

The lung presents unique considerations for RFA because of the large quantities of air between thin layers of parenchyma, continuous air flow, and high volume of pulmonary vascular flow. Air acts as an insulator, so RF energy is concentrated in the tumor. However, the air exchange and vascular flow dissipate heat from normal surrounding parenchyma,20,21 creating difficulty in establishing a therapeutic ablation margin. Compared to solid organs, the lung has less water per unit volume of tissue,22 so the infusion of saline may be warranted to help promote conduction of electric current.23

Microwave Ablation Microwave ablation (MWA) has been established as a treatment option for many tumors of the lung (Fig. 154-2; also see Table 154-1), liver, kidney, and bone. Most MWA antennae (applicators) available for clinical use emit microwaves in the range of 900 to 2450 MHz.24 Unlike RF waves, which generate heat via the friction induced by an electric current, microwaves generate heat by increasing the kinetic energy of water molecules: water molecules in the tissue adjacent to a microwave antenna function as electrical dipoles that repeatedly rotate to align with the applied oscillating

electromagnetic field, thus raising the heat of the tissue.25 Microwave energy is transferred to a much larger area, up to 2 cm surrounding the antenna, than RF wave energy, so MWA creates a larger zone of active heating.26 Six MWA systems are currently available in the United States. Three systems use a 915-MHz generator (Evident [Covidien, Boulder, Colo.], AveCure [MedWaves, San Diego, Calif.], MicroTherm X [BSD Medical, Salt Lake City, Utah]), and three systems use a 2450-MHz generator (Certus 140 [NeuWave Medical, Madison, Wis.], Amica [Hospital Service, Rome, Italy], Acculis MTA [Microsulis Medical Ltd., Hampshire, United Kingdom]), and all use straight antennas. Two of the systems (MicroTherm X, Certus 140) allow synchronous delivery of MW energy in three antennas simultaneously. One system (AveCure) does not require shaft cooling to protect the skin. The other five manufacturers use antennae that are perfused with roomtemperature fluid or carbon dioxide gas (Certus 140) to reduce heating of the nonactive proximal portion of the applicator and prevent damage to the skin and other proximal tissues. Microwaves behave differently than RF waves in the lung. Microwave propagation is not insulated by air and not hindered by the limited water content of lung parenchyma.25 In theory, MWA may have some advantages over RFA in

CHAPTER 154 LUNG ABLATION 1147 TABLE 154-1. Outcomes from Leading and Supportive Studies Conducted on Various Lung Ablation Modalities Author

Year

No. Patients

No. Tumors

Notable Findings

Radiofrequency Ablation Hiraki et al.49

2011

50

52 primary

Overall survival 94% at 1 year, 86% at 2 years, 74% at 3 years, 67% at 4 years, 61% at 5 years Stage IA survival 95% at 1 year, 89% at 2 years, 83% at 3 years, 73% at 4 years, 66% at 5 years Stage IB survival 92% at 1 year, 75% at 2 years, 50% at 3 years, 50% at 4 years, 50% at 5 years Cancer-specific survival 100% at 1 year, 93% at 2 years, 80% at 3 years, 80% at 4 years, 74% at 5 years Disease-free survival 82% at 1 year, 64% at 2 years, 53% at 3 years, 46% at 4 years, 46% at 5 years Median and mean disease-free survival 42 months

Palussiere et al.39

2010

127

210

Probability of survival 72% at 2 years, 60% at 3 years, 51% at 5 years

Zemlyak et al.50

2010

64 25 resection 12 RFA 27 cryo

Overall survival at 3 years for resection (wedge or segmentectomy) 87.1%, for RFA 87.5%, for cryo 77%, with no statistically significant difference Cancer-specific survival at 3 years for resection 90.6%, for RFA 87.5%, for cryo 90.2%, with no statistically significant difference Length of hospital stay significantly shorter for RFA/cryo than resection

Huang et al.35

2010

329 237 primary 92 mets

Median progression-free interval 21.6 months Overall survival 68.2% at 1 year, 35.3% at 2 years, 20.1% at 5 years NSCLC survival 80.1% at 1 year, 45.8% at 2 years, 24.3% at 5 years Pulmonary mets survival 50.6% at 1 year, 30.1% at 2 years, 17.3% at 5 years Local progression in 78 (23.7%) during follow-up, most likely related to incomplete ablation due to technical issues or tumor size >4 cm

Beland et al.51

2010

79

79 primary

Median disease-free survival 23 months Recurrence in 43% at mean follow-up of 17 months Larger tumor sizes associated with larger risk of recurrence, suggesting initial volume of ablation not adequate

Lanuti et al.52

2009

31

34 primary

Overall survival 78% at 2 years, 47% at 3 years Median overall survival 30 months, median disease-free survival 25.5 months Disease-free survival 57% at 2 years, 39% at 3 years Mean progression-free survival 33 ± 3.8 months Local progression in 31.5%, tumors >3 cm had greatest recurrence rate, tumors 3 cm Stage I NSCLC overall survival 78% at 1 year, 57% at 2 years, 36% at 3 years, 27% at 4 years, and 27% at 5 years Colorectal pulmonary mets overall survival 87% at 1 year, 78% at 2 years, 57% at 3 years, 57% at 4 years, and 57% at 5 years

de Baere et al.56

2006

60

100

Local control 93% at 18 months Overall survival 71% at 18 months Lung disease free 34% at 18 months

Dupuy et al.57

2006

24

Thanos et al.58

2006

22

22 14 primary 8 mets

Primary lung cancer overall survival 14% at 3 years Pulmonary mets overall survival 42% at 3 years

Yan et al.59

2006

55

55 mets

Overall median survival 33 months, despite 30/55 with previously resected liver mets Actuarial survival 85% at 1 year, 64% at 2 years, and 46% at 3 years Univariate analysis: lesion size, location, repeat RFA predictive of survival Multivariate analysis: only lesion size remained predictive

Local control 58% Overall survival 95% at 1 year

Local control 92% Stage IA cumulative survival 92% at 12 months, 62% at 24 months, and 46% at 56 months Stage IB cumulative survival 73% at 12 months, 42% at 24 months, and 31% at 60 months

Continued

1148 PART 2 NONVASCULAR INTERVENTIONS · SECTION 23 THE THORAX TABLE 154-1. Outcomes from Leading and Supportive Studies Conducted on Various Lung Ablation Modalities—cont’d Author

Year

No. Patients

No. Tumors

Notable Findings

Fernando et al.60

2005

18 5 primary 13 mets

33

Only 13 patients treated percutaneously Disease progression seen in 8 nodules (38%) in 6 patients (33%) Overall mean survival 20.97 months, median survival not reached Stage I mean progression-free interval 17.6 months Other stages progression-free interval 14.98 months

Akeboshi et al.61

2004

31

54 13 primary 41 mets

Complete tumor necrosis 69% in 36 lesions 3 cm

Kang et al.62

2004

50 23 primary 27 mets

120

Complete tumor necrosis in tumors >3.5 cm In tumors >3.5 cm, complete tumor necrosis area within 3.5 cm diameter PET-demonstrated tumor destruction in 70% of cases

Lee et al.63

2004

30

32 27 NSCLC 5 mets

Complete tumor necrosis in tumors 3 cm, 23% complete ablation In 20 lesions with incomplete necrosis, 50% necrosis In 8/12 cases of complete necrosis, 5-mm well-demarcated nonenhancing zone of ground-glass opacity surrounding ablated lesion No local tumor recurrence at 22.2 months

Yasui et al.64

2004

35

99 3 primary 96 mets

Complete tumor necrosis 91%, mean diameter of 1.9 cm

Herrera65

2003

18

33

Complete response 6% Partial response 44% Stable response 33% Progression 17%

Microwave Ablation Wolf et al.66

2008

50

Primary local control 74% at median follow-up of 10 months Additional 6% secondary local control for total 80% local control rate Actuarial survival 65% at 1 year, 55% at 2 years, and 45% at 3 years

Feng67

2002

20

28 8 primary 20 mets

Local response rate 57%

2009

64

108 mets

Primary local control rate 78% Secondary local control rate of 72% Survival rate 81% at 1 year, 59% at 2 years, 44% at 3 years, 44% at 4 years, 27% at 5 years

Zemlyak et al.50

2010

64 25 resection 12 RFA 27 cryo

Kawamura et al.68

2006

20

35 mets

Local control rate 80% Survival rate 90% at 1 year

Wang et al.69

2005

187

234 196 primary 38 mets

Mean Karnofsky performance score increased from 75.2 ± 1.3 before ablation to 82.6 ± 1.4 1 week post ablation

Laser Ablation Rosenberg et al.29

Cryoablation Overall survival at 3 years for resection (wedge or segmentectomy) 87.1%, for RFA 87.5%, for cryo 77%, with no statistically significant difference Cancer-specific survival at 3 years for resection 90.6%, for RFA 87.5%, for cryo 90.2%, with no statistically significant difference Length of hospital stay significantly shorter for RFA/cryo than resection

cryo, Cryoablation; mets, metastases; NSCLC, non–small cell lung cancer; PET, positron emission tomography; RFA, radiofrequency ablation.

the lung, including a greater convection profile, resistance to heat sinks, lack of charring and related impedance limitations, and the ability to more rapidly create larger ablative volumes by using multiple applicators simultaneously.25,27 Such advantages could theoretically improve ablation of the tumor margin and thus reduce recurrence rates. A literature search reveals only one study of MWA versus RFA in the lung. This study examined MWA and RFA in normal porcine lung parenchyma and revealed MWA lesion diameters were larger and more circular than those created by RFA.27

Laser Ablation Laser ablation utilizes electromagnetic waves whose frequencies fall in the infrared spectrum and are of a monochromic wavelength (generally 1064 nm) generated by an Nd:YAG source and transmitted through fiberoptic cables. Lasers generate heat via molecular photon absorption.28 Only one laser system (Biotex Inc., Houston, Tex.) is currently available commercially in the United States, and that system has not yet been applied to primary lung tumors. European studies (see Table 154-1) have shown laser

CHAPTER 154 LUNG ABLATION 1149

A

B

C

FIGURE 154-2. Biopsy-proven right lower lobe carcinoid tumor in a 79-year-old woman who was referred for lung ablation. A, Prone axial computed tomography (CT) image obtained immediately prior to microwave ablation (MWA) shows periaortic mass (arrow). B, Prone axial CT image obtained during MWA shows antenna (arrow) appropriately placed in mass. C, Axial contrast-enhanced 1-year follow-up CT scan shows residual thermal scar (arrow) in right lower lobe, without evidence of growth or enhancement.

A

B

C

FIGURE 154-3. Biopsy-proven right lower lobe osteosarcoma metastasis in a 22-year-old man who was referred for tumor ablation. A, Axial computed tomography (CT) image obtained immediately prior to cryoablation shows 6.8 cm × 5.4 cm mass (star) with internal mineralization adjacent to pericardium (arrows) and diaphragm. B, Axial CT image obtained after fourth freeze cycle shows cryoprobes (arrows) in ice ball (arrowheads), which extends to pericardium. C, Axial contrast-enhanced 2-month follow-up CT scan shows consolidated nonenhancing region (arrows) consistent with tumor necrosis.

ablation of secondary tumors in the lung to be safe and suggest that laser ablation has the potential for improving long-term survival.29

Cryoablation Cryoablation does not utilize the electromagnetic spectrum, but relies on applicators perfused with either liquid nitrogen or pressurized argon gas. Liquid nitrogen has been used in the operating room setting to treat liver tumors for over 3 decades,30 but the more recent development of pressurized argon gas systems has significantly decreased applicator diameters and made percutaneous cryoablation of other tissues, including the lung (Fig. 154-3; also see Table 154-1), more feasible. Pressurized argon gas can cool to −140°C owing to the Joule-Thomson effect. Temperatures below −40°C induce direct cell destruction primarily through ice crystal formation, which leads to osmotic shifts in intracellular and extracellular water. In addition, as the

tissue thaws, endothelial damage via the same mechanism results in increased capillary permeability, edema, aggregation of platelets, and thrombosis, eliminating the blood supply to the tumor. Necrosis is generally seen in the central part of the lesion closest to the applicator, and apoptosis is seen in the peripheral part of the lesion.31,32 A freeze/thaw/ freeze cycle is typically used as the second freeze/thaw produces faster and more extensive cooling and thereby extends certain tissue destruction closer to the outer limit of the frozen volume.32 Two systems, CryoHit (Galil Medical, Plymouth Meeting, Pa.) and Cryocare (Endocare Inc., Irvine, Calif.), are currently available for clinical use in the United States. Both systems allow simultaneous placement of and treatment with up to 15 probes of 1.5 to 2.4 mm diameter. In the lung, the airways do not appear to affect the size of the ablation zone.33 As with cryoablation in other tissues, the ice ball may be visualized under computed tomography (CT) or magnetic resonance imaging (MRI),

1150 PART 2 NONVASCULAR INTERVENTIONS · SECTION 23 THE THORAX allowing direct comparison of ablative volume borders to tumor margins, increasing operator confidence in measuring cytotoxic margins and managing treatments near critical structures.34

EXPERIENCE TO DATE The majority of lung ablation literature has focused on RFA and is in the form of single-institution case series and retrospective cohort studies. These studies have shown the safety and efficacy of thermal ablation as a palliative and potentially curative treatment option for medically inoperable NSCLC and pulmonary metastases. The outcomes from the major leading and supporting thermal lung ablation studies are summarized in Table 154-1. There have been no prospective trials directly comparing thermal ablation methods to one another, to radiation therapy, or to chemotherapy. Neither have there been significant series studying the efficacy of thermal ablation for medically operable lung disease or comparing thermal ablation to surgery. Such studies are unlikely in the near future, given the proven efficacy of surgery and associated ethical concerns. A prospective pilot trial looking at the 2-year post-RFA survival and local control in patients with stage IA medically inoperable NSCLC is currently in progress. Complications related to thermal ablation have been well documented. In a series of 329 RFA ablations, Huang et al.35 experienced pneumothorax (19.1%), pneumonia (4.5%), hemoptysis (4.2%), hemothorax (3.0%), and pericardial tamponade (0.9%). Additional documented complications of lung ablation include chronic obstructive pulmonary disease exacerbation (0%-6%), pulmonary abscess (0%-6%), pleural effusion requiring drainage (0%-4%), acute respiratory distress syndrome (0%-3%), parenchymal hemorrhage (0%-1%), phrenic nerve injury (0%-1%), death (0%-1%), and case reports of acute renal failure, vocal cord paralysis, atrial fibrillation, pulmonary embolus, third-degree skin burn, stroke, tumor tract seeding, pacemaker resetting, hypertonic saline-induced lobar necrosis, and prolonged bronchopleural fistula.36 Immunohistologic analysis by Schneider et al.37 confirmed RFA-induced cell death in a series of 32 tumors treated intraoperatively prior to curative resection. Incomplete ablation was associated with vascular structures within the tumor tissue and in the marginal zones of the tumor tissue. In addition, scattered vital tumor tissue was identified in 50% of the ablations. A similar study was performed by Jaskolka et al.38 on tumors ablated percutaneously 2 to 4 weeks prior to scheduled resection. This study confirmed coagulative necrosis by hematoxylin and eosin (H&E) staining and loss of proliferating cell nuclear antigen (PCNA) protein staining but retention of MIB-1 staining. Given the complete necrosis identified by Jaskolka, it is possible that the scattered vital tissues found by Schneider may have been due to the very short period between ablation and resection, but emphasizes the need for close periprocedural management of temperature and impedance and suggests utility in further histologic studies of ablated lung tissues. In general, ablation success is assessed on imaging follow-up. In a study of 350 tumors treated with RFA, Palussiere et al.39 found four patterns of evolution at 1 year post procedure: fibrosis (50.5%), nodules (44.8%), cavitation

(2.4%), and atelectasis (1.4%). The nodular pattern was most frequently associated with local tumor progression, but the pattern was not indicative of outcome, and thus no pattern warrants discontinuation of follow-up. Imaging features associated with local recurrence following RFA are outlined in a review by Eradat et al.40: contrast enhancement on CT at 3 months should never exceed that of the original tumor, but benign peripheral enhancement can be expected to persist for up to 6 months, while nodular or central enhancement of 10 to 15 Hounsfield units suggests incomplete ablation, progression, or recurrence.

NEWLY INTRODUCED ADVANCES IN THERMAL LUNG ABLATION Several recent studies have revealed new techniques that may improve thermal ablation efficacy and have shown potential in aggressively treating patients who have not been included in the more conservative typical cohorts studied to date. Healey et al.41 have described a technique for securing lung tumors during RF electrode insertion. In a swine study, Santos et al.42 increased MWA volume by occluding the bronchus prior to ablation to reduce heat dissipation via air exchange. In a series of six patients, de Baere et al.43 successfully performed complete ablations of metastatic tumors adjacent to large pulmonary arteries by temporarily occluding the arteries with a percutaneous balloon catheter to reduce heat dissipation via perfusion, but with complications of long-lasting vascular occlusion and atele